David J. Burkhart, Brian T. Kalet,Michael P. Coleman, Glen C. Post,and Tad H. Koch
Department of Chemistry and Biochemistry, University ofColorado, Boulder, Colorado
AbstractWe have reported the synthesis and biological evaluationof a prodrug to a doxorubicin active metabolite. Underphysiologic conditions, release of the active metabolite,a conjugate of doxorubicin with formaldehyde, occurswith a half-life of 1 hour. To direct this prodrug totumor, we designed two conjugates of the prodrug,doxsaliform, with the AvB3-targeting peptides,CDCRGDCFC (RGD-4C) and cyclic-(N-Me-VRGDf) (Cilen-gitide). We now report the synthesis of these doxsali-form-peptide conjugates and their evaluation using MDA-MB-435 cancer cells. A hydroxylamine ether tether wasused to attach 5VV-formyldoxsaliform to RGD-4C in itsacyclic form via an oxime functional group. Theconstruct acyclic-RGD-4C-doxsaliform showed goodbinding affinity for AvB3 in the vitronection cell adhesionassay (IC50 = 10 nmol/L) and good growth inhibition ofMDA-MB-435 breast cancer cells (IC50 = 50 nmol/L).In its bicyclic forms, RGD-4C showed less affinity forAvB3 and significantly less water solubility. Cyclic-(N-Me-VRGDf) was modified by substitution of D-4-amino-phenylalanine for D-phenylalanine to provide a novelattachment point for doxsaliform. The conjugate, cyclic-(N-Me-VRGDf-NH)-doxsaliform, maintained a high affin-ity for AvB3 (IC50 = 5 nmol/L) in the vitronectin celladhesion assay relative to the peptide bearing only thetether (0.5 nmol/L). The IC50 for growth inhibition ofMDA-MB-435 cells was 90 nmol/L. Flow cytometry andgrowth inhibition experiments suggest that the completedrug construct does not penetrate through the plasmamembrane, but the active metabolite does on release
from the targeting group. These drug conjugates couldhave significantly reduced side effects and are promisingcandidates for in vivo evaluation in tumor-bearing mice.[Mol Cancer Ther 2004;3(12):1593–604]
IntroductionDoxorubicin and its congener, epidoxorubicin, are amongthe most effective chemotherapeutics for the treatment ofa variety of solid tumors including breast tumors (1). Themechanism of action of doxorubicin has been debated sinceits discovery in the late 1960s, but its interaction with DNAis undeniable. Evidence has accumulated over the past 10years that suggests doxorubicin combines with formalde-hyde to form covalent bonds with DNA (2–5). The sourceof formaldehyde in cells is unclear, but analysis of cancercells treated with doxorubicin reveals elevated levels offormaldehyde (6). Doxorubicin induction of oxidativestress is a possible pathway to the formaldehyde necessaryfor drug-DNA virtual cross-linking (7). Synthetic deriva-tives of doxorubicin, which contain formaldehyde in theform of an N-Mannich base, are significantly morecytotoxic to both sensitive and resistant cancer cells thandoxorubicin itself (8, 9).
Two resistance mechanisms to doxorubicin are theoverexpression of the drug efflux pump P-170 glycopro-tein and suppression of oxidative stress mechanisms thatproduce formaldehyde (7, 10, 11). Using mass spectrome-try, our group has observed doxorubicin inducingformaldehyde production in MCF-7 breast cancer cellsbut not in doxorubicin-resistant MCF-7/Adr cells (6). Tocombat both resistance mechanisms, we developed aconjugate formed from doxorubicin, formaldehyde, andsalicylamide, known as doxsaliform (9). Doxsaliformcontains an N-Mannich base that, on time, hydrolyzes torelease the doxorubicin active metabolite with formalde-hyde already incorporated (Fig. 1). Doxsaliform exhibitsgreater toxicity to both sensitive and resistant cancer cells(MCF-7, MCF-7/Adr, Rtx-6, MDA-MB-231, MDA-MB-435,and PC-3; refs. 9, 12, 13).
Significant limitations for doxorubicin treatment ofcancer are drug resistance and chronic cardiotoxicity(14). One of the most promising methods to reduce the sideeffects of a cytotoxin like doxorubicin is selective deliveryto cancer cells and/or their associated angiogenesis. Aprotein complex that may be a good target for drugdelivery is the avh3 integrin. avh3 is involved in many cell-matrix recognition and cell adhesion phenomena, giving itan important role in angiogenesis and tumor metastasis.The avh3 integrin is overexpressed on the surface of tumorand endothelial cells responsible for angiogenesis (15), andits expression correlates with tumor progression in glioma,melanoma, breast cancer, and ovarian cancer (16–21). avh3
exists in discrete activation states, and activation can be
Received 7/29/04; revised 9/4/04; accepted 9/29/04.
Grant support: U.S. Army Prostate Cancer Research Program grantDAMD17-01-1-0046; National Cancer Institute of the NIH grantCA-92107; and University of Colorado Council for Research andCreative Work faculty fellowship (T.H. Koch). The National ScienceFoundation helped with the purchase of NMR equipment(grant CHE-0131003).
The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.
Requests for reprints: Tad H. Koch, Department of Chemistry andBiochemistry, University of Colorado, Boulder, CO 80309-0215.Phone: 303-492-6193; Fax: 303-492-5894.E-mail: [email protected]
Copyright C 2004 American Association for Cancer Research.
induced with manganous ion (22). Activated avh3 supportsbreast cancer cell arrest during blood flow and stronglypromotes breast cancer metastasis (23, 24). In tumor-induced angiogenesis, invasive endothelial cells bind viathis integrin to extracellular matrix components. Theinhibition of this interaction induces apoptosis of theproliferative angiogenic vascular cells (25). These factorscombined make avh3 an attractive target for antiangiogenicand antimetastatic therapies. Several RGD peptide andpeptide mimetics developed over the last decade exhibitexcellent binding affinity and selectivity for avh3 (26). Thepeptide cyclic-(N-Me-VRGDf) known as Cilengitide hasproceeded as far as phase II clinical trials as a potentantagonist of avh3 (27). Small RGD-containing peptideshave successfully been used to deliver cytotoxins, magneticresonance imaging contrast agents, radionuclides, lipo-somes, and fluorescent agents to tumors that express avh3
(28–30).Arap and coworkersV (28) report that a doxorubicin-
CDCRGDCFC (RGD-4C) conjugate that targets avh3 sub-stantially inhibited tumor growth in mice relative todoxorubicin with fewer side effects prompted furtherexploration. de Groot and coworkers (31) reported thatdoxorubicin conjugated with RGD-4C via a plasmin-cleavable tether inhibited human umbilical vascular endo-thelial cell binding to plates coated with vitronectin with anIC50 of f150 nmol/L and exhibited a cytotoxicity IC50 of750 nmol/L against the same cell line. The plasmin-activated prodrug failed to inhibit tumor growth in vivobetter than doxorubicin alone but did exhibit less toxicitybased on weight loss in a tumor-bearing mouse model (32).Here we report the synthesis and biological evaluation ofdoxsaliform conjugated to two different RGD-containingpeptides, RGD-4C and cyclic-(N-Me-VRGDf).
The conjugation of doxsaliform to avh3-targetingpeptides serves several purposes. The drug conjugate isa prodrug with little or no activity until the trigger (N-Mannich base hydrolysis) releases the cytotoxin from thepeptide. RGD-4C and cyclic-(N-Me-VRGDf) have bothbeen shown to accumulate in tumor relative to other
tissue, with a peak accumulation point of f40 to 60minutes (33). Based on this delivery schedule, a triggeredrelease of doxorubicin active metabolite with a half-life of60 minutes should localize a good portion of the drug intumor relative to other tissue. We hypothesize that thisdesign will reduce side effects such as cardiotoxicity andincrease the amount of active drug in and around thetumor.
Materials andMethodsSynthesisDesign. Doxsaliform was conjugated to the RGD-
containing peptides RGD-4C and cyclic-(N-Me-VRGDf) viaa short hydroxylamine ether tether that forms an oximebond with a formyl group added at the 500 position of thesalicylamide of doxsaliform. This oxime was found to bequite stable under a variety of aqueous conditions. The N-Mannich base that contains the formaldehyde equivalentnecessary to produce the doxorubicin active metabolitehydrolyzes with a half-life of 60 minutes at physiologictemperature and pH (9). Hydrolysis of the N-Mannich baseis also the trigger that releases the doxorubicin activemetabolite from the targeting peptide. The synthesis ofacyclic-RGD-4C-doxsaliform and cyclic-(N-Me-VRGDf-NH)-doxsaliform are detailed below and their structuresare shown in Fig. 2. See Fig. 3 and Fig. 4, respectively, forsynthetic schemes.
Materials and Instruments. All reactions were doneunder inert atmosphere. Fmoc amino acids and otherpeptide synthesis reagents were purchased from Nova-biochem (San Diego, CA) and used without furtherpurification. For amino acids with sensitive side chainsthe following were used: Fmoc-Asp-tBu, Fmoc-Cys-triphe-nylmethyl, Fmoc-Arg-2,2,4,6,7-pentamethyldihydrobenzo-furan-5-sulfonyl, Fmoc-D-4-aminoPhe[t-butoxycarbonyl(Boc)]. Fmoc-2-(2-aminoethoxy)-ethylamine hydrochloridewas obtained from Neosystem (Strasbourg, France), (Boc-aminoxy)-acetic acid from Fluka (Milwaukee, WI), and
Figure 1. Synthesis of doxsaliform (DOXSF ) and the mechanism by which it releases the proposed doxorubicin (DOX ) active metabolite.
Oregon Green 488 N-hydroxysuccinimide from MolecularProbes (Eugene, OR). 500-Formyldoxsaliform was synthe-sized as previously described (13). Melting points wereuncorrected. The 1H, COSY, HSQC, HMBC, and 13Cnuclear magnetic resonance (NMR) high-resolution spectrawere obtained with a Varian Inova 500 spectrometer (PaloAlto, CA). Electrospray mass spectra were measured with aPerkin-Elmer Sciex API III (Norwalk, CT), equipped withan ion-spray source, at atmospheric pressure. Analytichigh-performance liquid chromatography (HPLC) wascarried out on a Hewlett-Packard/Agilent 1050/1100system (Palo Alto, CA) consisting of a Hewlett-Packard1050 series auto injector and pumping system, Hewlett-Packard 1046A fluorescence detector, Agilent 1100diode array UV-Vis detector, and Agilent ChemStationdata system (Palo Alto, CA). A Vydac (Hesperia, CA)protein C-4 column (4.6 � 250 mm) was used for analyticHPLC with a flow rate of 0.5 mL/min and a gradientsolvent system of 0.1% or trifluoroacetic acid (TFA)/acetonitrile: 0 to 15 minutes, 98% to 40% aqueous; 15 to20 minutes, 40% to 15% aqueous; 20 to 25 minutes, 15% to98% aqueous; detection at 220, 254, 280, and 480 nm. Forpreparative HPLC, a Vydac 214TP1022 C-4 column (22 �250 mm) was used with the same solvent system on aVarian/Ranin (Palo Alto, CA) semipreparative HPLCconsisting of Varian Pro Star 210 pumping system, RaninDynamax UV-1 detector, and Ranin/Varian Macintoshdata system eluting at 15 mL/min. The gradient used forcyclic-(N-Me-VRGDf-NH)-tether was 0 to 11.5 minutes,98% to 80% aqueous; 11.5 to 15 minutes, 80% to 30%
aqueous; 15 to 16 minutes, 30% to 55% aqueous; and 16 to20 minutes, 55% to 98% aqueous (detection at 254 nm). Thegradient used for complete drug conjugates was 0 to 30minutes, 90% to 60% aqueous, and 30 to 50 minutes, 60% to90% aqueous (detection at 470 nm).
Synthesis of the Hydroxylamine Ether Tether, (2-{2-[2-(2,2-Dimethyl-Propionylaminooxy)-Acetylamino]-Eth-oxy}-Ethyl)-Carbamic Acid 9H-Fluoren-9-ylmethyl Ester 1.Fmoc-2-(2-aminoethoxy)-ethylamine hydrochloride 1.30 gwas weighed and placed in a dry 250-mL round-bottomed flask under an argon atmosphere. Anhydrousdimethylformamide (DMF, 10 mL) was added by syringefollowed by 2.0 mL of pyridine with stirring. (Boc-aminoxy)-acetic acid [1.04 g, 2 equivalents (eq)] andwater-soluble carbodiimide (0.69 g, 2 eq) were measuredout and added in one portion to the solution of amine.The reaction was monitored by analytic HPLC and 0.33eq of (Boc-aminoxy)-acetic acid and water-soluble carbo-diimide were added after 1 hour to drive the reaction tocompletion. The reaction was then diluted with ethylacetate (100 mL) and washed with dilute acetic acid (3 �50 mL) followed by sodium bicarbonate (pH 8.5). Theorganic phase was dried with sodium sulfate andconcentrated under vacuum to yield 1.78 g (99%) of clearsolid product 1. 1H NMR in chloroform-d : 1.42 (s, 9H), 3.20(m, 2H), 3.53 (m, 6H), 4.20 (t, J = 6.8 Hz, 1H), 4.27 (s, 2H), 4.43(d, J = 6.8 Hz, 2H), 5.75 (br, 1H), 7.26 (t, J = 7.6 Hz, 2H), 7.43(dd, J = 4.8, 7.2 Hz, 2H), 7.62 (d, J = 7.2 Hz, 2H), 7.76 (d, J = 7.6Hz, 2H); electrospray-mass spectrometry (ESI-MS) relativeintensity (m/z): 500, calculated for (M + H+) m/z 500.23.
Figure 2. The doxsaliform-peptide conjugates cyclic-(N-Me-VRGDf-NH)-doxsaliform and acyclic-RGD-4C-doxsaliform exhibit good binding affinity foravh3 and toxicity against MDA-MB-435 cancer cells. Cyclic-(N-Me-VRGDf-NH)-doxsaliform is labeled for clarification.
Partial Deprotection of 1 to {2-[2-(2-Aminooxy-Acetyl-amino)-Ethoxy]-Ethyl}-Carbamic Acid 9H -Fluoren-9-ylmethyl Ester 2 and Loading of 2 onto Resin. TheFmoc-2-(2-aminoethoxy)-ethyl-Boc-aminoxy-amide (1,1.7 g) was dissolved in a solution containing 10 mL ofTFA and 1.1 mL of thioanisole at 0jC. The solution wasallowed to stir for 1 hour at room temperature and thenconcentrated (<5 mL) and the product precipitated intocold diethyl ether (100 mL). The precipitate was thencollected as the TFA salt by filtration and washed withether (3 � 20 mL). 1H NMR in methanol-d4 showedcomplete removal of the Boc protecting group, so thecompound 2 was then loaded on trityl chloride resin asfollows. To a dry 250-mL round-bottomed flask wasadded 50 mL of dry methylene chloride, 2.2 mLanhydrous pyridine, and 2. After the amine went intosolution with stirring, 1.1 g of trityl chloride resin wasadded in one portion and the mixture allowed to stir for22 hours. The resin was then collected by filtration,washed with 17:2:1 [volume for volume (v/v)] methylenechloride:methanol:diisopropylethylamine (2 � 25 mL),and with methylene chloride (2 � 30 mL) followed bymethanol (3 � 50 mL). The resin was then dried undervacuum and the loading was determined by treatment of
an aliquot (5 mg) with 0.5 mL of 20% piperidine/DMFfor 15 minutes and dilution to 50 mL with DMF followedby UV absorbance measurement at 301 nm. Resin loadingranged from 0.5 to 0.84 mmol/g.
General Procedure for the Synthesis of Linear Peptides.The linear peptides were synthesized by the solid-phasemethod using Fmoc strategy (for details see AppliedBiosystems peptide synthesizer user’s manual), startingwith the preloaded Fmoc tether from above. The peptideswere prepared on a 0.25-mmol scale by single amino acidcouplings using a 4-fold excess of Fmoc amino acids and2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluroniumhexafluorophosphate (TBTU)/N -hydroxybenzotriazole(HOBT) activation on an Applied Biosystems 433Apeptide synthesizer. Fmoc groups were removed bysequential treatment (3�) with 20% piperidine/DMF.Acyclic-RGD-4C was synthesized in the order Cys-Phe-Cys-Asp-Gly-Arg-Cys-Asp-Cys and final Fmoc deprotec-tion of the peptide was done while still on the resin.The linear peptide was cleaved from the resin anddeprotected by a 3-hour treatment with degassedreagent K. The resin was then filtered and the motherliquor concentrated under vacuum (<5 mL) and theproduct precipitated dropwise into cold ether (60 mL).
Figure 3. Synthesis of acyclic- and cyclic-RGD-4C-tether and subsequent conjugation of acyclic-RGD-4C-tether to doxsaliform. DOXSF-CHO,5VV-formyldoxsaliform.
The peptide was collected by filtration (#1 filter paper)and washed with ether (3 � 20 mL). The crude peptidewas dried under vacuum overnight and analyzed byanalytic HPLC and mass spectrometry. The analytic HPLCtrace showed a single peak (retention time, 22.22 minutes),ESI-MS, m/z 1,180.6, calculated for (M+H+) 1,179.39.
Synthesis of the Acyclic-RGD-4C-Doxsaliform. To a50-mL pear-shaped flask containing 3 mg of 500-formyl-doxsaliform was added 2 mL of 0.1% TFA and thesolution was degassed with argon by bubbling for 5minutes. Acyclic-RDG-4C-tether (3, 12 mg, 3 eq) wasdissolved in 1.5 mL of degassed methanol and thenadded in one portion to the 500-formyldoxsaliformsolution by syringe. The reaction was allowed to stir atroom temperature and was monitored by HPLC. Afterf3 hours, the reaction was found to be complete byHPLC analysis (new peak found retention time = 17.14minutes, 480 nm). The reaction was purified directly bypreparative HPLC and all major peaks analyzed by massspectrometry. The product showed a mass spectral ion atm/z 1,883.2 (M + H+; calculated 1,883.6) and a base peak
at m/z 942.2 [(M + 2H+)/2]. The yield of acyclic-RGD-4C-doxsaliform was 1.2 mg of compound, 98% pure byanalytic HPLC. Drug was then formulated with 3 eq ofcitric acid and 6 eq of lactose and stored at �80jC.
Synthesis of Bicyclic-RGD-4C-Tether 4. Acyclic-RGD-4C-tether 3 (30 mg) was dissolved in a solution of 50 mLof TFA and 2.5 mL of DMSO. Anisole (0.5 mL) was thenadded by syringe with stirring and the solution stirredfor 1 hour. The reaction was monitored by HPLC andstopped when complete (usually 1 hour). The solutionwas then concentrated under high vacuum to yield amixture (f50:50) of bicyclic isomers. HPLC gave twopeaks at retention time, 27.05 and 27.31 minutes for thetwo isomers; ESI-MS, m/z 1,175.6, calculated for (M +H+) 1,175.4 for both isomers.
Synthesis of Protected Acyclic-N-Me-VRGDf-NH2 5.General Fmoc synthesis was done as for acyclic-RGD-4C-tether, but TBTU/HOBT hydrate coupling was found to beinefficient for coupling to N-methyl valine. Peptide still onthe resin was treated with 2 eq bromo-tris-pyrrolidino-phosphonium hexafluorophosphate, Fmoc-D-4-aminophe
Figure 4. Synthesis of cyclic-(N-Me-VRGDf-NH)-tether and subsequent conjugation to doxsaliform. Peptide synthesis and derivatization proceeded inhigh yield requiring no chromatography. SPPS, solid-phase peptide synthesis; WSCI, water-soluble carbodiimide; EtOAc, ethyl acetate; BocAA, Boc-aminooxyacetic acid; NHS, N -hydroxysuccinimide.
(Boc) and 4 eq diisopropylethylamine in dry dichloro-methane (5 mL per gram resin). The mixture was placedon a shaker for 16 hours, washed with 3 � 10 mL ofdichloromethane, and checked by the chloranil test forcoupling completion. If not complete, coupling was repeatedfor 3 hours. When coupling was finished, the resin was thentreated with 3� 10 mL of 20% piperidine in DMF for a periodof 10 minutes to complete deprotection. Resin was thenreturned to the ABI synthesizer to complete the peptidesynthesis using standard Fmoc synthesis protocol. Cleavageof the linear peptide was effected with 1% TFA in dichloro-methane (3 � 10 mL) with shaking for 5 minutes each time.The solution was concentrated under high vacuum to givethe linear peptide with protecting groups intact in 98% yieldas determined by analytic HPLC (one peak with retentiontime 17.9 minutes); ESI-MS, m/z 1,029.6, calculated for (M +H+) 1,029.51.
Cyclization of Protected N-Me-VRGDf-NH2 5 to yield6. Linear peptide (386 mg) with all protecting groups intactwas dissolved in 50 mL of ethanol and 1.2 eq of 10%aqueous HCl (v/v) was added to displace the TFA salt.When this step was omitted, trifluoroacetylation of thepeptide occurred during the cyclization reaction. Thesolution was concentrated under vacuum. Linear peptidewas then dissolved in anhydrous DMF (125 mL) and 2 eq ofwater-soluble carbodiimide were added in one portion.Reaction was monitored by HPLC and typically completewithin 3 hours. The solution was concentrated undervacuum and the residue was dissolved in 50 mL of ethylacetate and washed with 10% HCl (v/v; 2 � 50 mL). Theorganic phase was dried with sodium sulfate and concen-trated under vacuum to give pure cyclic-(N-Me-VRGDf-NH2) with protecting groups intact 6. Analytic HPLCshowed one peak (retention time, 17.2 minutes); ESI-MSm/z 1,012.5, calculated for (M + H+) 1,012.51.
Selective Removal of Boc Protecting Group from D-4-Amino-Phe of Fully Protected Cyclic-(N-Me-VRGDf-NH2) 6 to Yield 7. Fully protected cyclic-(N-Me-VRGDf-NH2; 6, 319 mg) was dissolved in 10 mL of dry ethylacetate, and 1 mol/L anhydrous HCl in ethyl acetate(1.5 mL) was added while the mixture was maintained at0jC with an ice bath. The mixture was allowed to stir for3 hours at 0jC and then concentrated under vacuum. Theproduct was then lyophilized from water to give a clearsolid. This method completely removed the Boc groupfrom the D-4-amino-Phe, but a small amount of peptidealso experienced hydrolysis of the Asp tert-butyl protect-ing group to release the acid. This mixture was carriedforward because the deprotected Asp was not deemedproblematic. HPLC analysis shows two peaks (retentiontime, 12.2 and 13.9 minutes); ESI-MS for these two peaks,m/z 856.4 and 912.6, respectively. Calculated for depro-tection of both D-4-amino-Phe(Boc) and Asp(tBu) (M +H+) 856.39; calculated for deprotection of only D-4-amino-Phe(Boc) (M + H+) 912.46.
Addition of Boc-Aminoxyacetic Acid Tether to PartiallyProtected Cyclic-(N-Me-VRGDf-NH2) 7. Clear solid (288mg) from the above reaction was dissolved in 50 mL
anhydrous DMF and 3 eq of Boc-aminooxyacetic acid wereadded followed by 1.5 eq of water-soluble carbodiimide.After stirring for 1.5 hours, the reaction was complete basedon analytic HPLC. The mixture was concentrated undervacuum, the residue dissolved in ethyl acetate (50 mL), andwashed with water (2 � 20 mL) and then 10% HCl (v/v;2 � 50 mL). The organic phase was separated, dried withsodium sulfate, and concentrated under vacuum. Twopeaks were observed by HPLC (retention time, 12.7 and14.4 minutes); ESI-MS for the two products, m/z 1,028.8and 1,084.5, respectively; calculated for (M + H+) 1,028.47and 1,084.53.
Removal of All Protecting Groups from Cyclic-(N-Me-VRGDf-NH)-Tether to Yield 9. Peptide from the abovereaction was added to a dry 50 mL round-bottomed flaskand cooled to 0jC under an argon atmosphere. Reagent K(5 mL) was added and the solution allowed to stir for3 hours at room temperature. Solution was then addeddropwise with vigorous stirring to 100 mL of anhydrousether that had been cooled with an ice bath. The whiteprecipitate was collected by filtration and washed thricewith ether (15 mL) and dried under vacuum. Pure productcyclic-(N-Me-VRGDf-NH)-tether (9, 171 mg) was obtainedas determined by HPLC (retention time, 7.2 minutes); ESI-MS, m/z 677.2, calculated for (M + H+) 677.33. To assignthe 1H NMR spectrum unequivocally, the following spectrawere run: 1H NMR, COSY, HSQC, and HMBC all in D2O.1H NMR: 0.47 (3H, d, J = 6 Hz, CH3, Val), 0.80 (3H, d, J = 6Hz, CH3, Val), 1.48 (2H, m, CH2, Arg), 1.83 (2H, m, CH2,Arg), 1.85 (1H, m, CH, Val), 2.62 (1H, dd, J = 17 and 6 Hz,CH2, D-Phe) 2.80 (3H, s, CH3, N-Me Val), 2.6-2.9 (3H, m,CH2, D-Phe, and Asp), 3.07-3.13 (2H, m, CH2, Arg), 3.45(1H, d, J = 14 Hz, Gly), 3.83 (1H, m, CH, Arg), 4.04 (1H, d,J = 14 Hz, Gly), 4.23 (1H, d, J = 11 Hz, CH, Val), 4.47 (1H, t,J = 6 Hz, CH, D-Phe), 4.6-4.8 (under HOD peak, CH2,hydroxylamine ether tether), 5.09 (1H, t, J = 7Hz, CH,Asp), 7.18 (2H, d, J = 8 Hz, CH, Phe), 7.29 (1H, d, J = 8 Hz,CH, Phe).
Conjugation of Cyclic-(N-Me-VRGDf-NH)-Tether 9 toDoxsaliform to Yield Cyclic-(N-Me-VRGDf-NH)-Doxsali-form 10. 500-Formyldoxsaliform (4 mg) was dissolved in a3:1 mixture (2 mL) of water:ethanol (v/v, pH 2.0 TFA), and8 mg cyclic-(N-Me-VRGDf-NH)-tether 9 was added. Thesolution was stirred at room temperature for 5 hours untilone peak with absorbance at 480 nm was observed byanalytic HPLC. The product was purified by preparativeHPLC to yield 4.2 mg of pure cyclic-(N-Me-VRGDf-NH)-doxsaliform 10. The conjugate was concentrated undervacuum at room temperature and stored as a red solid at�80jC. HPLC analysis showed one predominate peak(retention time, 12.7 minutes, 96%); ESI-MS, m/z 1,379.6,calculated for (M + H+) 1,379.54. To assign the 1H NMRspectrum the following spectra were obtained in DMF-d7;1H NMR, COSY, HSQC, HMBC, ROESY. 1H NMR: 0.41(3H, d, J = 6 Hz, CH3, Val), 0.77 (3H, d, J = 6 Hz, CH3, Val),1.14 (3H, d, J = 7 Hz, CH3, 5V), 1.45–1.51 (2H, m, CH2, Arg),1.87–1.93 (2H, m, CH2, Arg), 1.93–1.95 (2H, m, CH2, 2V),2.03–2.05 (1H, m, CH, Val), 2.12 (1H, dd, J = 6 and 15 Hz,
Conjugation of Cyclic-(Me-VRGDf-NH)-Tether 9 toOregon Green to Yield Cyclic-(N-Me-VRGDf-NH)-Ore-gon Green 11. To a dry-25 mL round-bottomed flask wasadded 2 mg of 5V-Oregon Green 488 N-hydroxysuccini-mide, 1 eq of cyclic-(Me-VRGDf-NH)-tether (2.7 mg) and 5mL of anhydrous DMF. The solution was allowed to stir for5.5 hours while monitored by HPLC. Solution was thenconcentrated under vacuum and the residue resuspendedin methanol for purification by preparative HPLC, whichyielded 1.1 mg (26%) of a solid yellow product. AnalyticHPLC showed a single peak (retention time, 16.9 minutes);ESI-MS, m/z 1,071.5, calculated for (M + H+) 1,071.36.
Biological EvaluationCell Culture. Human breast carcinoma cell line MDA-
MB-435 (34) was maintained in DMEM medium supple-mented with 10% fetal bovine serum, penicillin (100 units/mL), streptomycin (0.1 mg/mL), L-glutamine (2 mmol/L),sodium pyruvate (1 mmol/L), nonessential amino acids,and vitamins.
Purified Proteins. Human vitronectin and bovine serumalbumin (BSA; A-7030) were purchased from Sigma (St.Louis, MO). avh3-Specific monoclonal antibody LM609 waspurchased from Chemicon (Temecula, CA).
Cell Adhesion Assay. Cell adhesion was determined bycoating wells of 96-well plates (Corning, New York, NY)with 100 AL of 5 Ag vitronectin/mL in Dulbecco’s PBS from2.0 hours to overnight at room temperature. Wells werewashed twice with deionized water and nonspecificbinding sites were blocked with 200 AL heat-inactivated(20 minutes at 60jC) 1.0% BSA in Dulbecco’s PBS from 2.0hours to overnight at 37jC. Wells were washed five timeswith deionized water and allowed to dry for 30 minutes atroom temperature or stored at 4jC for extended periods.Cells were harvested from a subconfluent T-175 tissueculture flask by rinsing with 35 mL Dulbecco’s PBS andincubating with 2 mL of 4 mmol/L EDTA for 3 minutes at37jC. The EDTA solution was neutralized by adding 48 mLof DMEM containing penicillin-streptomycin. Cells werewashed once with 50 mL DMEM + penicillin-streptomycinand resuspended in DMEM + penicillin-streptomycin ata final concentration of 8.5 � 105 cells/mL. MnCl2 was
added, resulting in a final concentration of 500 AM.Peptides or antibody was added prior to adding cells (100AL) to 96-well plates. Cells were allowed to adhere for 90 to100 minutes at 37jC. Nonadherent cells were removed byaspiration and washing twice with Dulbecco’s PBS con-taining 900 AM Ca2+ and 500 AM Mg2+. Adherent cells werequantified via measuring cellular metabolism of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide(MTT, Promega, Madison, WI) at 37jC. For every experi-ment, each condition was done in triplicate; experimentswere done at least twice.
AvB3 Binding Assessed by Flow Cytometry with Cyclic-(N-Me-VRGDf-NH)-Oregon Green. Cells were harvestedfrom five subconfluent T-175 cell culture flasks by rinsingwith 35 mL Dulbecco’s PBS and incubating with 2 mL of 4mmol/L EDTA for 3 minutes at 37jC. The EDTA solutionwas neutralized by adding 40 mL Dulbecco’s PBS. Cellswere washed once with 9 mL Dulbecco’s PBS andresuspended in Dulbecco’s PBS + 0.5% BSA or Dulbecco’sPBS + 0.5% BSA, 500 AM MnCl2, and 500 AM MgCl2.Various concentrations of cyclic-(N-Me-VRGDf-NH)-Ore-gon Green were added and allowed to incubate for 90minutes at 37jC. Cells were washed twice, resuspended in500 AL Dulbecco’s PBS + 0.5% BSA and analyzed by flowcytometry on a MoFlo (Ft. Collins, CO) flow cytometer.Cells were analyzed with excitation at 488 nm (Ar ionlaser), with emission monitored between 510 and 550 nm.Ten thousand cells were analyzed per condition. Thedata are presented as the mean fluorescence for each con-dition with the background, drug-free cell fluorescencesubtracted.
Uptake of Doxorubicin, Doxsaliform, and Acyclic-RGD-4C-Doxsaliform. A flow cytometry method of measuringuptake of doxorubicin, doxsaliform, and acyclic-RGD-4C-doxsaliform in breast cancer cells was done as previ-ously described, with modifications (35, 36). MDA-MB-435 breast cancer cells in log phase growth were dissociatedwith trypsin-EDTA, counted, resuspended in medium at2 � 105 cells/mL, and plated into six-well plates (5 � 105
cells per well) and allowed to adhere overnight. Drugsolutions of doxorubicin, doxsaliform, and ayclic-RGD-4C-doxsaliform were prepared in DMSO with 1% acetic acid at50 AM. Before treatment with drug, cell medium wasremoved, cells were washed with HBSS (0.5 mL), and thenfresh cell medium, with or without 500 AM Mn2+, wasplaced into the wells (2 mL). Drug treatments of 0.5 AMwere accomplished by the addition of 20 AL from the drugsolutions to the desired wells and incubation for variousamounts of time (20, 40, and 60 minutes). For each timepoint, the cell medium was removed, cells were washedwith HBSS and trypsinized, and trypsinization wasquenched with 5 mL of cell medium at 4jC. Cells werepelleted by centrifugation at 200 � g for 5 minutes at 10jC.The supernatant was decanted, and the cells wereresuspended in 5 mL of Dulbecco’s PBS at 4jC, repelleted,resuspended in 2 mL of Dulbecco’s PBS at 4jC, and placedon ice. Drug treatments were done in such a manner that allcell treatment times would end at approximately the same
time to ensure comparable measurements with theFACScan instrument. The amount of drug uptake wasmeasured by flow cytometry on a Becton DickinsonBiosciences FACScan (San Jose, CA) flow cytometer usingBecton Dickinson Biosciences CellQuest software (version3.3). Cells were analyzed with excitation at 488 nm (15 mWAr ion laser), with emission monitored between 570 and600 nm. Instrument settings were optimized for the cell lineand held constant for all experiments; for anthracyclinefluorescence analysis, 10,000 cells were analyzed for eachsample. The data are presented as the mean fluorescencefor each condition divided by the background, drug-freemean fluorescence.
Growth Inhibition Assay. Growth inhibition was deter-mined as previously described (9) with minor modifica-tions. Cells were treated for 4 hours, then allowed to growuntil control wells reached f80% confluence (4–5 days).Cells were quantified by measuring crystal violet stainingor cellular metabolism of MTT. For every experiment, eachcondition was done in hexaplicate; experiments were doneat least twice.
ResultsSynthesisWe began our search for an avh3-targeting group for
doxsaliform with RGD-4C because of the success of Arapand coworkers with bicyclic-RGD-4C conjugated todoxorubicin (28). The two predominant bicyclic structuresare formed by oxidation of the four thiols to two disulfidebridges (37). We observed that on formation of the disulfide
bridges, RGD-4C became poorly water soluble over a rangeof pH. Strong evidence for the molecular structure ofdoxorubicin-RGD-4C prepared by Arap and coworkerswas not reported (28). Based on the observed change insolubility on formation of the disulfide bridges, wehypothesized that acyclic-RGD-4C was the actual peptidethat targeted Arap and coworkers’ phage to MDA-MB-435tumors in mice. Linear RGD-containing peptides areknown to have a short circulation time in the bloodstreamdue to the activity of proteases, but because targeteddelivery to tumor is relatively rapid, we sought to test thispeptide-drug conjugate. The synthetic strategy for acyclic-RGD-4C-doxsaliform outlined in Fig. 3, used an oximationreaction of a formyl group placed at the 500 position of thesalicylamide group of doxsaliform and a hydroxylamineether tether at the carboxyl terminus of the peptide. Theoximation reaction was regioselective for the aryl aldehydeand produced a robust connection between the targetinggroup and the salicylamide trigger, time-release group.Both acyclic-RGD-4C-tether and acyclic-RGD-4C-doxsali-form have good water solubility.
An attractive alternative to RGD-4C is the cyclicpeptide, cyclic-(N-Me-VRGDf), developed by Merck(Rahway, NJ) as a selective and potent avh3 antagonist.The X-ray crystal structure of cyclic-(N-Me-VRGDf)bound to avh3 shows the D-phenylalanine and N-methylvaline directed toward solvent, making these residuesattractive attachment points for conjugation of cytotoxinor other molecular probe (see Fig. 5; ref. 38). We chose toattach a short tether to the 4 position of D-phenylalaninebecause this would take advantage of the rigid nature of
Figure 5. X-ray crystal structure ofcyclic-(N-Me-VRGDf) bound to avh3
(38) reveals the exposed D-phenylala-nine and N-methyl valine as potentialattachment points for conjugate syn-thesis. Manganous ions are shown aspurple spheres. Image was createdusing PyMOL modeling software andcoordinates obtained from the ProteinData Bank (1L5G).
the aromatic ring, essentially creating a short linear tetherand projecting the steric bulk of doxsaliform towardsolvent. Cyclic-(N-Me-VRGDf-NH)-tether with the hydrox-ylamine functional group at the terminus of the tether wassynthesized from start to finish in high yield with nochromatography as shown in Fig. 4. Again, the targetinggroup was connected to 500-formyldoxsaliform via theoximation reaction. The conjugate, cyclic-(N-Me-VRGDf-NH)-doxsaliform, obtained in pure form after preparativeHPLC, was stable for months while stored at �80jC asthe TFA salt of its N-Mannich base.
avb3 BindingBicyclic-RGD-4C-tether (as a 50:50 mixture of the 1-4;2-3
and 1-3;2-4 isomers), acyclic-RGD-4C-tether, and cyclic-(N-Me-VRGDf-NH)-tether were assayed for their ability tobind the avh3 integrin present on viable MDA-MB-435cells using a vitronectin competition assay (39). Vitronec-tin is the endogenous ligand for the avh3 integrin.Conditions for inhibition of MDA-MB-435 cell adhesion tovitronectin, including the requirement of Mn2+, wereestablished by using an avh3-specific monoclonal antibody(LM609). Targeting compounds or targeted doxsaliformwere added to cell suspensions created by release of cellsfrom cell culture flasks with EDTA as opposed to trypsin topreserve the integrity of avh3. Drug-treated cells were thenadded to cell culture plates coated with BSA with orwithout vitronectin. Cells were allowed to adhere, wellswere washed with Dulbecco’s PBS, and cells werequantified by measuring cellular metabolism of MTT.Nonspecific binding (cells bound to BSA-coated wells)was subtracted from total binding (cells bound to BSA-andvitronectin-coated wells) to determine specific binding tovitronectin. The concentrations of compound required toinhibit binding of 50% of the cells to vitronectin (IC50
values) are shown in Table 1. The acyclic-RGD-4C isomerwas chosen over the bicyclic isomer for further experimentsdue to better water solubility and higher binding affinityfor the avh3 integrin. Next, acyclic-RGD-4C-doxsaliformand cyclic-(N-Me-VRGDf-NH)-doxsaliform compoundswere assayed for their ability to bind the avh3 integrin(Table 1). The binding affinities of both acyclic-RGD-4C-
tether and cyclic-(N-Me-VRGDf-NH)-tether decreased byonly one order of magnitude on addition of doxsaliform,indicating that the tethering system does not precludebinding. IC50 values for both acyclic-RGD-4C-doxsaliformand cyclic-(N-Me-VRGDf-NH)-doxsaliform in the vitronec-tin assay are significantly lower than those for the RGD-4C-doxorubicin conjugate with the plasmin-cleavable tether,pioneered by de Groot and coworkers (ref. 40; 10 and 5nmol/L versus 150 nmol/L).
Cyclic-(N-Me-VRGDf-NH)-tether was also analyzed forbinding to avh3 on MDA-MB-435 cells as a function of Mn2+
activation with cyclic-(N-Me-VRGDf-NH)-tether bound toOregon Green fluorescent dye (see Figs. 4 and 6). Bindingas a function of cyclic-(N-Me-VRGDf-NH)-Oregon Greenconcentration in the presence and absence of Mn2+ wasmeasured by flow cytometry, as shown in Fig. 6. Theexperiment was done with cells in suspension, releasedfrom the growth flask with EDTA. In the presence of Mn2+,binding of dye to cells increased with concentration of dyeand plateaued at about 100 nmol/L. In the absence of Mn2+,little binding of dye was observed even at 100 nmol/Lcyclic-(N-Me-VRGDf-NH)-Oregon Green, consistent withtargeted dye binding to activated avh3. As reported inTable 1, the IC50 for targeted dye binding to cells is 2 nmol/L, approximately midway between the values for cyclic-(N-Me-VRGDf-NH)-tether and cyclic-(N-Me-VRGDf-NH)-doxsaliform.
Uptake of Acyclic-RGD-4C-DoxsaliformUptake of acyclic-RGD-4C-doxsaliform by MDA-MB-435
cells was measured by flow cytometry after drug treatmentfor various periods of time in the presence and absence ofadditional Mn2+ beyond that present in fetal bovine serum.Concentration of targeted drug relative to doxorubicin anddoxsaliform in cells was determined from emission of thedoxorubicin fluorophore. After a 1-hour drug treatmenttime, doxorubicin was taken up 3-fold more than acyclic-RGD-4C-doxsaliform, as shown in Fig. 7, and uptake ofacyclic-RGD-4C-doxsaliform was independent of addition-al Mn2+. In Table 2, uptake of acyclic-RGD-4C-doxsaliformis compared with uptake of doxorubicin and doxsaliform attwo time points (30 minutes and 4 hours) and three drug
Table 1. IC50 values for inhibition of MDA-MB-435 cell binding to vitronectin and cell growth as a function of targeting group ordrug design
Compound IC50 for inhibition of cell binding(nmol/L)
IC50 for inhibition of cell growth(nmol/L), treatment time
Bicyclic-RGD-4C-tether 4 10 F 1Acyclic-RGD-4C-tether 3 1 F 0.2Cyclic-(N-Me-VRGDf-NH)-tether 9 0.5 F 0.1Cyclic-(N-Me-VRGDf-NH)-Oregon Green 11 2 F 0.4Acyclic-RGD-4C-doxsaliform 10 F 2 1,000 F 200, 20 min; 50 F 10, 4 hCyclic-(N-Me-VRGDf-NH)-doxsaliform 10 5 F 1 1,000 F 200, 20 min; 250 F 50, 1 h; 90 F 20, 4 hDoxsaliform >104 50 F 10, 4 hDoxorubicin >104 800 F 200, 20 min; 300 F 60, 1 h; 120 F 30, 4 h
NOTE: For structures, see Figs. 1, 2, 3, 4, and 6.
concentrations (100, 500, and 1,000 nmol/L) in the absenceof additional Mn2+. At the 30-minute time point, f30% ofthe time-release trigger of acyclic-RGD-4C-doxsaliform ordoxsaliform had fired, and at the 4-hour time point,more than 90% of the trigger had fired based on the knownhalf-life for the trigger (9). After treatment for 30 minuteswith 500 nmol/L drug, uptake of fluorophore from acyclic-RGD-4C-doxsaliform was 60% of doxorubicin and 20% ofdoxsaliform. However, after treatment for 4 hours with 500nmol/L drug, uptake of fluorophore from acyclic-RGD-4C-doxsaliform and doxorubicin was comparable and uptakeof fluorophore from doxsaliform was only 2-fold higher.These results suggest that acyclic-RGD-4C-doxsaliformdoes not significantly penetrate the cell membrane andthat the doxorubicin fluorophore only enters after thetrigger releases the doxorubicin-formaldehyde conjugate.
Cancer Cell Growth InhibitionAcyclic-RGD-4C-doxsaliform and cyclic-(N-Me-VRGDf-
NH)-doxsaliform were also assayed for their ability toinhibit growth of MDA-MB-435 cells relative to doxorubicinand doxsaliform. Cells treated in cell culture plates andnontreated (control) cells were allowed to grow to nearconfluency and then quantified via measuring crystal violetstaining or cellular metabolism of MTT. The concentrationsof drug required to inhibit growth of cells by 50% (IC50
values) are shown in Table 1 as a function of drug treatmenttime. The data in Table 1 were obtained in the absence ofadditional Mn2+ because control experiments showed noeffect from Mn2+ on cytotoxicity. Both acyclic RGD-4C-doxsaliform and cyclic-(N-Me-VRGDf)-doxsaliform aremore cytotoxic than clinical doxorubicin and comparablein cytotoxicity to doxsaliform with a drug treatment time of4 hours. With shorter drug treatment times, the cytotox-icities of targeted drugs and doxorubicin are comparable.The slight decrease in cytotoxicity observed for cyclic-(N-Me-VRGDf-NH)-doxsaliform relative to doxsaliform is
comparable to the loss relative to parent drug observed byde Groot and coworkers (31, 41). Earlier control exper-iments established that the miniscule amounts of form-aldehyde that would be released even from completehydrolysis of the conjugate would contribute nothing to theobserved growth inhibition (8). Conjugation of cytotoxicdrugs to triggers and targeting groups often causes a dropin cytotoxicity (42).
DiscussionDoxsaliform prodrug-RGD conjugates were synthesizedand evaluated for binding to avh3 in the vitronectin celladhesion assay and for inhibition of MDA-MB-435 cancercell growth. We hypothesized that a prodrug with thisdesign would bind avh3 and localize in/or near the tumorand vascular endothelial cells of the developing bloodsupply. Upon hydrolysis of the N-Mannich base, theconjugate would release the doxorubicin active metabolitelocally. Local delivery is required because the activemetabolite has a short lifetime with respect to furtherhydrolysis to doxorubicin (half-life, f5 minutes; ref. 8).The advantage of delivering the doxorubicin activemetabolite is that it is more cytotoxic to both sensitiveand resistant tumor cells.
RGD-4C as a targeting group was explored firstbecause of significant activity in tumor-bearing micereported for RGD-4C-doxorubicin conjugates with thepeptide in its oxidized form (28). The structures for
Figure 6. Binding of cyclic-(N-Me-VRGDf-NH)-Oregon Green to MDA-MB-435 cells in the presence (y) and absence (n) of Mn2+ as measuredby flow cytometry. Y-axis is in relative fluorescence units.
Figure 7. Relative uptake of acyclic-RGD-4C-doxsaliform by MDA-MB-435 cells as a function of time and additional Mn2+ with uptake ofdoxorubicin as a control. Cells were treated with 500 nmol/L drug anduptake was measured by flow cytometry observing fluorescence from thedoxorubicin fluorophore. Y-axis is in relative fluorescence units.
the conjugates, however, were not well defined by thesynthetic strategy or from spectroscopic data. A later reportestablished that 1-4;2-3-bicyclic-RGD-4C has an order ofmagnitude better affinity for avh3 than the other majorregioisomer, 1-3;2-4-bicyclic-RGD-4C (37). Oxidation ofacyclic-RGD-4C-tether gave roughly a 50:50 mixture ofthe 1-4;2-3- and 1,3;2,4-bicyclic isomers. We found thatacyclic-RGD-4C-tether (2) had better affinity for avh3
(Table 1) and much better aqueous solubility than a 50:50mixture of the two regioisomers of bicyclic-RGD-4C-tether.The result that acyclic-RGD-4C bound with higher affinitythan the mixture of bicyclic isomers is surprising becauseformation of the disulfide bridges makes the structuremore rigid. Based on this result, we selected acyclic-RGD-4C-tether for conjugation to doxsaliform. Acyclic-RGD-4C-doxsaliform conjugate exhibited a decrease in affinity foravh3 relative to the peptide alone (10 nmol/L versus1 nmol/L), but cytotoxicity against MDA-MB-435 cells(IC50 = 50 nmol/L) was comparable to that of doxsaliform.Comparison of uptake of doxorubicin fluorophore byMDA-MB-435 cells treated with either acyclic-RGD-4C-doxsaliform, doxorubicin, or doxsaliform as a function oftreatment time suggests that targeted drug does notpenetrate the plasma membrane. Appearance of doxorubi-cin fluorophore from targeted drug in cells requires releaseof the doxorubicin-formaldehyde conjugate by the salicy-lamide trigger.
Because acyclic-RGD-4C-doxsaliform has the potentialfor instability due to 4 sulfhydryl groups, we alsoexplored a cyclic RGD peptide with the cycle createdvia a peptide linkage between the amino and carboxyltermini, cyclic-(N-Me-VRGDf). Although a variant, cyclic-(KRGDf), has been used to attach various molecules tothe avh3-targeting peptide at the q-amino group of theLys, we sought to use the linear D-phenylalanine ofcyclic-(N-Me-VRGDf) as an attachment point guided bythe cocrystal structure of the ligand-binding domain ofavh3 bound to cyclic-(Me-VRGDf; Fig. 5). Based onmolecular modeling of our conjugate bound to avh3, wehypothesized this linear tether would permit attachment ofa large molecule without a significant decrease in bindingaffinity to avh3. Indeed, cyclic-(N-Me-VRGDf-NH)-doxsali-form exhibited an IC50 in the vitronectin binding assay of
5 nmol/L. Furthermore, a conjugate of cyclic-(N-Me-VRGDf-NH) with Oregon Green showed dose- and Mn2+-dependent binding to MDA-MB-435 cells by flow cytometry.The cancer cell growth inhibition by cyclic-(N-Me-VRGDf-NH)-doxsaliform is better than doxorubicin but reducedtwo times relative to doxsaliform. A higher IC50 relative todoxsaliform is attributed to a reduced rate of uptakebecause the targeted drugs do not seem to penetrate theplasma membrane, and for uptake, the salicylamide triggermust first release the doxorubicin-formaldehyde conjugate.
The likely scenario for prodrug activity in vivo basedon these experiments would be binding to avh3 overex-pressed by tumor and/or tumor vascular endothelialcells during circulation followed by hydrolysis of the N-Mannich base releasing the doxorubicin active metaboliteextracellularly. The active metabolite should enter the cellmore rapidly than free doxorubicin due to its lack ofcharge, then induce apoptosis via the formation ofcovalent cross-links in cellular DNA (43–45). Possibly,some conjugate could be internalized via receptor-mediat-ed or fluid-phase endocytosis and hydrolyzed to the activemetabolite intracellularly. Because the active metabolite isnot cationic, as opposed to doxorubicin, the P-170 drugefflux pump resistance mechanism would likely have lesseffect (11). Similarly, resistance mechanisms that suppressoxidative stress and the production of formaldehyde willhave little effect because the active metabolite released bythe trigger already has formaldehyde incorporated.
Both RGD-doxsaliform conjugates have good affinityfor avh3 and are more cytotoxic than clinical doxorubicin.The salicylamide N-Mannich base trigger hydrolyzeswith a half-life of 60 minutes (9), which is appropriatefor the rate of targeted drug delivery to tumor (28). BothRGD-targeted drug designs show good water solubilityand are promising candidates for in vivo testing in tumor-bearing nude mice.
AcknowledgmentsWe thank Prof. Renata Pasqualini (M.D. Anderson Cancer Center) for asample of MDA-MB-435 cells, Dr. Richard Shoemaker for help with theNMR experiments, Prof. Katheryn Resing for help with MS experiments,and Theresa Nahreini for help with the flow cytometry experiments.
Table 2. Uptake of acyclic-RGD-4C-doxsaliform by MDA-MB-435 cells as a function of dose and time in the absence of additional Mn2+
Mn2+ compared with uptake of doxorubicin and doxsaliform
References1. Sweatman TW, Israel M. Anthracyclines. In: Teicher BA, editor. Cancertherapeutics, experimental and clinical agents. Totowa (NJ): HumanaPress; 1997. p. 113–35.
2. Wang AHJ, Gao YG, Liaw YC, Li YK. Formaldehyde cross-linksdaunorubicin and DNA efficiently: HPLC and X-ray diffraction studies.Biochemistry 1991;30:3812–5.
3. Cullinane C, van Rosmalen A, Phillips DR. Does Adriamycin induceinterstrand cross-links in DNA? Biochemistry 1994;33:4632–8.
4. Taatjes DJ, Gaudiano G, Resing K, Koch TH. A redox pathway leadingto the alkylation of DNA by the anthracycline, anti-tumor drugs,Adriamycin and daunomycin. J Med Chem 1997;40:1276–86.
5. Zeman SM, Phillips DR, Crothers DM. Characterization of covalentAdriamycin-DNA adducts. Proc Natl Acad Sci U S A 1998;95:11561–5.
6. Kato S, Burke PJ, Fenick DJ, Taatjes DJ, Bierbaum VM, Koch TH. Massspectrometric measurement of formaldehyde generated in breast cancercells upon treatment with anthracycline antitumor drugs. Chem ResToxicol 2000;13:509–16.
7. Taatjes DJ, Gaudiano G, Koch TH. Production of formaldehyde andDNA-Adriamycin or -daunomycin adducts, initiated through redox chem-istry of DTT/iron, xanthine oxidase/NADH/iron, or glutathione/iron. ChemRes Toxicol 1997;10:953–61.
8. Fenick DJ, Taatjes DJ, Koch TH. Doxoform and Daunoform: anthracy-cline-formaldehyde conjugates toxic to resistant tumor cells. J Med Chem1997;40:2452–61.
9. Cogan PS, Fowler CR, Post GC, Koch TH. Doxsaliform: a novel N-Mannich base prodrug of a doxorubicin formaldehyde conjugate. Lett DrugDesign Dis 2004;1:247–55.
10. Sinha BK, Mimnaugh EG. Free radicals and anticancer drug resistance:oxygen free radicals in the mechanisms of drug cytotoxicity and resistanceby certain tumors. Free Radic Biol Med 1990;8:567–81.
11. Lampidis TJ, Kolonias D, Podona T, et al. Circumvention of P-GP MDRas a function of anthracycline lipophilicity and charge. Biochemistry1997;36:2679–85.
12. Cogan PS, Koch TH. Rational design and synthesis of androgenreceptor targeted non-steroidal anti-androgen ligands for the tumorspecific delivery of a doxorubicin-formaldehyde conjugate. J Med Chem2003;46:5258–70.
13. Burke PJ, Koch TH. Design, synthesis, and biological evaluationof doxorubicin-formaldehyde conjugates targeted to breast cancer cells.J Med Chem 2004;47:1193–206.
14. Hellmann K, Franks CR. Reduction of doxorubicin-induced cardiotox-icity by dexrazoxane. Contrib Oncol 1995;48:40–7.
15. Albelda SM, Mette SA, Elder DE, et al. Integrin distribution inmalignant melanoma: association of the h 3 subunit with tumor progres-sion. Cancer Res 1990;50:6757–64.
16. Pignatelli M, Cardillo MR, Hanby A, Stamp GW. Integrins and theiraccessory adhesion molecules in mammary carcinomas: loss ofpolarization in poorly differentiated tumors. Hum Pathol 1992;23:1159–66.
17. Gringas MC, Roussel E, Bruner JM, Branch CD, Moser RP.Comparison of cell adhesion molecule expression between glioblastomamultiforme and autologous normal brain tissue. J Neuroimmunol 1995;57:143–53.
18. Natali PG, Hamby CV, Felding-Habermann B, et al. Clinical significanceof avh3 integrin and intracellular adhesion molecule-1 expression incutaneous malignant melanoma lesions. Cancer Res 1997;57:1554–60.
20. Byzova TV, Plow EF. Activation of avh3 on vascular cells controlsrecognition of prothrombin. J Cell Biol 1998;143:2081–92.
21. Byzova TV, Kim W, Midura RJ, Plow EF. Activation of integrin avh3
regulates cell adhesion and migration to bone sialoprotein. Exp CellRes 2000;254:299–308.
22. Humphries MJ. Insights into integrin-ligand binding and activationfrom the first crystal structure. Arthritis Res 2002;4 Suppl 3:S69–78.
23. Felding-Habermann B, O’Toole TE, Smith JF, et al. Integrin activation
controls metastasis in human breast cancer. Proc Natl Acad Sci U S A2001; 98:1853–8.
24. Pilch J, Habermann R, Felding-Habermann B. Unique ability of integrinavh3 to support tumor cell arrest under dynamic flow conditions. J BiolChem 2002;277:21930–8.
25. Liekens S, De Clercq E, Neyts J. Angiogenesis regulators and clinicalapplications. Biochem Pharmacol 2001;61:253–70.
26. Dechantsreiter MA, Planker E, Matha B, et al. N-Methylated cyclicRGD peptides as highly active and selective avh3 integrin antagonists.J Med Chem 1999;42:3033–40.
27. Wermuth J, Goodman SL, Jonczyk A, Kessler H. Stereoisomerismand biological activity of the selective and superactive avh3 integrininhibitor cyclo(-RGDfV-) and its retro-inverso peptide. J Am Chem Soc1997;119:1328–35.
28. Arap W, Oasqualini R, Ruoslahti E. Cancer treatment by targeteddrug delivery to tumor vasculature in a mouse model. Science 1998;279:377–80.
29. Wonhee S, Han S-O, Yu L, Kim SW. An angiogenic, endothelial-cell-targeted polymeric gene carrier. Mol Ther 2002;6:664–72.
30. Hart SL, Knight AM, Harbottle RP, et al. Cell binding and internali-zation by filamentous phage displaying a cyclic Arg-Gly-Asp-containingpeptide. J Biol Chem 1994;269:12468–74.
31. de Groot FMH, Broxterman HJ, Adams HPHM, et al. Design,synthesis, and biological evaluation of a dual tumor-specific motivecontaining integrin-targeted plasmin-cleavable doxorubicin prodrug. MolCancer Ther 2002;1:901–11.
32. Devy L, de Groot FHM, Blacher S, et al. Plasmin-activated doxorubicinprodrug containing a spacer reduce tumor growth and angiogenesiswithout systemic toxicity. FASEB J 2004;18:565–7.
33. Janssen ML, Oyen WJ, Dijkgraaf I, et al. Tumor targeting withradiolabeled avh3 integrin binding peptides in a nude mouse model. CancerRes 2002;62:6146–51.
34. Calileau R, Olive M, Cruciger QV. Long-term human breast carcinomacell lines of metastatic origin: preliminary characterization. In Vitro1978;14:911–5.
35. Durand RE, Olive PL. Flow cytometry studies of intracellularAdriamycin in single cells in vitro . Cancer Res 1981;41:3489–94.
36. Taatjes DJ, Koch TH. Growth inhibition, nuclear uptake, and retentionof anthracycline-formaldehyde conjugates in prostate cancer cells relativeto clinical anthracyclines. Anticancer Res 1999;19:1201–8.
37. Assa-Munt N, Jia X, Laakkonen P, Ruoslahti E. Solution structuresand integrin binding activities of an RGD peptide with two isomers.Biochemistry 2001;40:2373–8.
38. Xiong J-P, Stehle T, Zhang R, et al. Crystal structure of theextracellular segment of integrin avh3 in complex with an Arg-Gly-Aspligand. Science 2002;296:151–5.
39. Bartsch JE, Staren ED, Appert HE. Adhesion and migration of extra-cellular matrix-stimulated breast cancer. J Surg Res 2003;110:287–94.
40. de Groot FMH, Broxterman HJ, Adams HPHM, et al. Design,synthesis, and biological evaluation of a dual tumor-specific motivecontaining integrin-targeted plasmin-cleavable doxorubicin prodrug. MolCancer Therap 2002;1:901–11.
41. de Groot FMH, de Bart ACW, Verheijen JH, Scheeren HW. Synthesisand biological evaluation of novel prodrugs of anthracyclines for selectiveactivation by the tumor-associated protease plasmin. J Med Chem1999;42:5277–83.
42. Czerwinski G, Tarasova NI, Michejda CJ. Cytotoxic agents directedto peptide hormone receptors: defining the requirements for a success-ful drug. Proc Natl Acad Sci U S A 1998;95:11520–5.
43. Skladanowski A, Konopa J. Adriamycin and daunomycin induceprogrammed cell death (apoptosis) in tumour cells. Biochem Pharmacol1993;46:375–82.
44. Skladanowski A, Konopa J. Relevance of interstrand DNA cross-linking induced by anthracyclines for their biological activity. BiochemPharmacol 1994;47:2279–87.
45. Burke PJ, Koch TH. Doxorubicin-formaldehyde conjugate, Doxoform:induction of apoptosis relative to doxorubicin. Anticancer Res 2001;21:2753–60.
