Enhanced Tumor Retention of a Radiohalogen Label for Site-Specific

Oct 16, 2013 - Genentech Research and Early Development, 1 DNA Way MS 463A, South San Francisco 94080, United States. •S Supporting Information...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/jmc

Enhanced Tumor Retention of a Radiohalogen Label for Site-Specific Modification of Antibodies C. Andrew Boswell,* Jan Marik, Michael J. Elowson, Noe A. Reyes, Sheila Ulufatu, Daniela Bumbaca, Victor Yip, Eduardo E. Mundo, Nicholas Majidy, Marjie Van Hoy, Saritha N. Goriparthi, Anthony Trias, Herman S. Gill, Simon P. Williams, Jagath R. Junutula, Paul J. Fielder, and Leslie A. Khawli Genentech Research and Early Development, 1 DNA Way MS 463A, South San Francisco 94080, United States S Supporting Information *

ABSTRACT: A known limitation of iodine radionuclides for labeling and biological tracking of receptor targeted proteins is the tendency of iodotyrosine to rapidly diffuse from cells following endocytosis and lysosomal degradation. In contrast, radiometal− chelate complexes such as indium-111−1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (In-111-DOTA) accumulate within target cells due to the residualizing properties of the polar, charged metal-chelate-amino acid adduct. Iodine radionuclides boast a diversity of nuclear properties and chemical means for incorporation, prompting efforts to covalently link radioiodine with residualizing molecules. Herein, we describe the Ugi-assisted synthesis of [I125]HIP-DOTA, a 4-hydroxy-3-iodophenyl (HIP) derivative of DOTA, and demonstration of its residualizing properties in a murine xenograft model. Overall, this study displays the power of multicomponent synthesis to yield a versatile radioactive probe for antibodies across multiple therapeutic areas with potential applications in both preclinical biodistribution studies and clinical radioimmunotherapies.

■

INTRODUCTION The antigen specificity of monoclonal antibodies is a powerful attribute that allows the specific in vivo delivery of payloads, including chemotherapeutic drugs and radionuclides.1 To date, only two radioimmunotherapeutic agents have been approved for clinical use, and both feature murine monoclonal antibodies targeting the CD20 receptor for treatment of lymphoma.2 Tositumomab is administered with the β-emitting iodine radionuclide, 131I, attached via tyrosine residues as shown in Figure 1A. Ibritumomab tiuxetan is administered with the βemitting yttrium radionuclide, 90Y, attached using tiuxetan, an isothiocyanate analogue of diethylenetriamine pentaacetic acid (DTPA), through lysine residues via thiourea bonds.3 This labeling strategy is somewhat analogous to the complexation of the indium radionuclide, 111In, by 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA), whose succinimidyl ester can form amide bonds via lysine as shown in Figure 1B. Beyond their clinical utility, radioimmunoconjugates are also useful as tools in preclinical and translational research for studying conventional, nonradioactive antibody therapeutics.7 Noninvasive small animal imaging modalities, whole-body autoradiography, and invasive biodistribution studies can assist drug development by providing valuable mechanistic information including confirmation of target localization, screening for off-target uptake, and assessing receptor occupancy. In addition to 131I, 125I is commonly used for such studies, with the latter having the advantages of a roughly 10-fold lower γ energy, the absence of a β particle emission, and a much longer decay halflife (Supporting Information, Table S1).8 Single photon © XXXX American Chemical Society

emission computed tomography (SPECT) imaging may be performed with 123I, 131I, and 125I, the latter being limited to preclinical small animal cameras. Positron emission tomography (PET) with 124I is also feasible,9 although the highly energetic emissions limit the image quality. Labeling of antibodies with radiometals results in a different cellular retention of radioactivity relative to iodotyrosine-based labeling.10 For both labeling methods, antibodies undergo receptor-mediated endocytosis and lysosomal degradation. However, a third critical step, cellular efflux of the radiolabel with its covalently associated amino acid, does not readily occur for radiometal-labeled antibodies (Figure 2).11 Unlike [125I]iodotyrosine, which diffuses out of the cell (intact or as free iodide) following proteolysis, 111In-DOTA−lysine cannot easily cross the plasma membrane due to its charge and polarity.12 Because of its tendency to become intracellularly trapped, 111InDOTA is referred to as a residualizing label. Significant effort has been made to derive strategies for labeling antibodies with radioiodine such that residualization occurs in a similar manner as for radiometals. This reflects, in part, the wide availability of iodine radionuclides with diverse nuclear properties, in terms of both decay half-lives and emission energies (Supporting Information, Table S1), and an abundant knowledge of halogen radiochemistry.8 For instance, the decay half-lives of 111In and 125I are 2.8 and 60 days, respectively, making the latter radionuclide far better suited for Received: September 6, 2013

A

dx.doi.org/10.1021/jm401365h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

emissions often associated with superior autoradiographic image quality and lower radiation exposure to workers. A residualizing iodine probe would combine the long decay halflife and low energy of 125I with the superior tumor accretion of radiometals while providing a facile translational route to clinical imaging (via 123I or 124I) or radioimmunotherapy (via 131 13 I) (Supporting Information, Table S1). To date, strategies used to derive residualizing iodine probes include the use of (i) nonmetabolizable carbohydrates, (ii) nonmetabolizable peptide adducts, and/or (iii) highly charged moieties. The carbohydrate derivative dilactitol-125I-tyramine is a member of the first class of residualizing radioiodine probes.14 However, the use of carbohydrates poses a potential risk, as pendant sugar groups are important for binding of antibodies to Fc receptors and other critical functions.15 Representing the second class is the residualizing peptide, IMP-R4 (MCCLys(MCC)-Lys(X)-D-Tyr-D-Lys(X)-OH, where MCC is 4-(Nmaleimidomethyl)-cyclohexane-1-carbonyl and X is 1-((4thiocarbonylamino)benzyl)-DTPA).16 This approach relies on a synthetic peptide that is conjugated with the chelate DTPA, whose charge imparts residualizing properties.17 Selected examples of the third class are shown in Figure 1C, many of which require the use of organostannane intermediates in their syntheses. 4−6 Overall, we sought to design a simple residualizing iodine probe that is accessible to laboratories with basic organic synthesis capabilities while avoiding the use of bulky peptides, carbohydrates, organostannanes, and lengthy synthetic routes. Herein, we report the Ugi multicomponent synthesis,18,19 radiolabeling, and biological evaluation of a residualizing radiohalogen probe, [125I]HIP-DOTA, (Figure 1D). The probe was site-specifically conjugated to an engineered THIOMAB20 (i.e., a full-length antibody having engineered cysteine residues) and demonstrated to exhibit superior tumor uptake relative to a conventional tyrosine radioiodinated control antibody. This novel method for introducing radioiodine labels into antibodies is a useful preclinical tool for studying the biodistribution, metabolism, and excretion of antibody therapeutics. Furthermore, antibodies labeled with residualizing radionuclides offer unique advantages as both diagnostic and radioimmunotherapeutic agents because they may provide a more sustained retention of radioactivity inside tumor cells.21 Assuming that efficacy is related to tumor radiation exposure, the use of residualizing radioimmunotherapeutic agents may encourage high target radiation exposure and favorable clinical responses.22

Figure 1. Selected nonresidualizing (A) and residualizing (B−D) methods of radiolabeling antibodies including (A) oxidative tyrosine radioiodination, (B) lysine modification with radiometal chelate, (C) lysine modification with charged iodinated groups,4−6 and (D) thiol modification with the novel compound 6, designated [125I]HIPDOTA, that can be attached to either the Fc region or the heavy (HC) or light (LC) chains of an antibody containing engineered cysteine residues.

■

RESULTS In choosing a synthetic strategy, we considered that our molecule should consist of three critical functional components: an iodotyrosine-like moiety, an activated group for antibody conjugation, and a residualizing anchor. The success of residualizing peptides, which contain iodotyrosine, suggested that a hydroxyphenyl residue is a sufficient scaffold on which to introduce radioiodine.23 This avoided the necessity of tin precursors, which are necessary to produce the N-succinimidyl m-iodobenzoate derivatives whose corresponding antibody conjugates are depicted in Figure 1C. The choice of the macrocyclic polyaminopolycarboxylic acid DOTA was based on both its known ability to residualize10 and on its commercial availability with a wide array of functional groups. We sought a chemical reaction with the ability to rapidly combine multiple chemical components in high yields and with

Figure 2. Schematic depicting the cellular fates of nonresidualizing and residualizing labels following antibody binding to an internalizing cellsurface antigen. Antibodies labeled by both methods undergo receptor-mediated endocytosis and lysosomal degradation. Diffusion of radioactivity out of the lysosome and/or cell occurs readily for iodotyrosine but is resisted for radiometal−chelate complexes and other residualizing labels.

long-term preclinical studies. This is especially true for antibodies, whose pharmacokinetic half-lives are often on the order of 1−2 weeks. Another consideration is that the γ energy of 125I is nearly 10-fold lower relative to 111In, with lower energy B

dx.doi.org/10.1021/jm401365h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 3. Synthetic route to the shelf compound, maleimide 4, and its radiolabeling and antibody conjugation to LC-V205C thio-trastuzumab.

ample flexibility in reactant structure. These requirements led us to the Ugi multicomponent reaction involving a ketone or aldehyde, an amine, an isocyanide, and a carboxylic acid to form a bisamide.18,19 Indeed, there is one previous report of the Ugi reaction applied to synthesis of ditopic chelating agents.24 We used formaldehyde, a primary amine-functionalized DOTA derivative (compound 1), 4-hydroxyphenylacetic acid, and ethyl isocyanoacetate as our four commercially available Ugi components (Figure 3). The ethyl ester 2 was converted to the acid 3 by saponification using lithium hydroxide. All attempts to synthesize the succinimidyl ester of this acid, our original desired molecule, were unsuccessful, with the formation of dimers (phenolic esters) evident by mass spectrometry. This problem was averted by changing from a lysine- to a cysteine-based conjugation strategy, taking advantage of recent developments in antibody engineering to introduce site-specific cysteines.20 Our previous work demonstrated that the maleimide functionality is compatible with the oxidative conditions of radioiodination.25 This guided our synthetic route to introduce a maleimide group by coupling the acid to N-(2-aminoethyl)maleimide using 1-ethyl-3-(3dimethylaminopropyl)carbodiimide (EDC), yielding the shelf compound 4 in only three steps (Figure 3) as compared to the six-step synthesis of the protected tin precursor of SIB-DOTA (SIB = N-succinimidyl-3-iodobenzoate).6 Compounds 2, 3, and 4 eluted on a RP-HPLC gradient at 14.8, 13.8, and 13.7 min, respectively (Supporting Information, Figure S1). Our radiochemical strategy initially involved use of the mild oxidant, N-chlorosuccinimide, to achieve iodination in chloroform on the hydroxyphenyl group. However, we subsequently discovered that maleimide 4 is sufficiently water-soluble to allow a more efficient labeling in aqueous 0.1% acetic acid in a test tube that is precoated with the water insoluble oxidizing agent, 1,3,4,6-tetrachloro-3α,6α-diphenylglucoluril (i.e., “Iodogen”) (Figure 3). The intermediate [125I]5 had a retention time of 15.5 min on reversed-phase HPLC (Supporting Information, Figure S2) and could be detected in terms of both [127I] and [125I] by mass spectrometry (Supporting Information, Figure

S3−S4). Compound 5 could be purified from free iodide and oxidant using reverse-phase solid-phase extraction cartridges, but this step was not necessary for the Iodogen method. Acid deprotection yielded the triacid [125I]6, with completeness of reaction monitored by reverse-phase radio-HPLC (Supporting Information, Figure S5). This 4-hydroxy-3-iodophenyl derivative of DOTA, compound [125I]6, was designated [125I]HIPDOTA. After exhaustive removal of acid by repeated addition and evaporation of toluene, the thiol-containing antibody could be introduced for thiol-maleimide coupling. Any remaining free thiols were reacted with iodoacetic acid to avoid dimerization or formation of adducts with thiol-containing endogenous proteins. All radioimmunoconjugates were analyzed for purity by size-exclusion HPLC and compared to the profile of unlabeled trastuzumab (Supporting Information, Figure S6− S7). The site-specificity of our labeling strategy was confirmed by SDS-PAGE analysis including both protein staining and phosphorimaging (Figure 4). Digestion with the endoprotease Lys-C was able to distinguish radiolabeled Fc from Fab. Similarly, reduction by dithiothreitol was able to resolve radiolabeled light chain (LC) from radiolabeled HC/Fc regions. On the basis of our previous experience with THIOMAB technology, we anticipated differences in stability between the heavy chain (HC-A114C), light chain (LC-V205C), and crystallizable fragment (Fc-S396C) radioimmunoconjugate variants (Kabat numbering is used).26 This hypothesis was confirmed using an in vitro plasma stability assay in which each of the [125I]6-labeled trastuzumab derivatives was incubated in mouse plasma at 37 °C. Consistent with the previously reported stabilities of antibody−drug conjugates,26 the rank order of plasma stability for our site-specific conjugates was LC > HC > Fc (Supporting Information, Figure S8−S10). These site specific differences in stability likely result from the local electrostatic environment (i.e., charged residues) causing variations in the rate of succinimide ring hydrolysis, which prevents further maleimide exchange with reactive thiols and C

dx.doi.org/10.1021/jm401365h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

Figure 4. SDS-PAGE analysis by protein staining (top) and phosphorimaging (bottom) of trastuzumab labeled site specifically with [125I]6 through its heavy chain (HC-A114C), light chain (LCV205C), and Fc (Fc-S396C) region. Digestion of antibodies with the endoprotease, Lys-C, results in cleavage between the Fab and Fc regions. In contrast, dithiothreitol (DTT) reduction separates the heavy (still attached to Fc) and light chains. Differential exposure of radioactivity in selected bands is evident despite equal loading of proteins, demonstrating the site-specificity of labeling.

enhances stability.26 The loss of the main (intact) peak at 18 min was accompanied by the appearance of two new peaks. An earlier peak at 16.5 min was attributed to protein aggregation and/or dimerization, while a later peak at 19 min was consistent with the retention time of albumin, an abundant plasma protein that is known to possess a reactive thiol. To evaluate the ability of compound 6, HIP-DOTA, to residualize, a biodistribution study was performed in the previously validated KPL-4 xenograft mouse model27 of human epidermal growth factor receptor 2 (HER2)-expressing breast cancer as previously reported.28 The radioimmunoconjugates [125I]6-[HC]−, [125I]6-[LC]−, and [125I]6-[Fc]−trastuzumab were directly compared with tyrosine-labeled 125I-trastuzumab prepared by the Chizzonite (i.e., indirect Iodogen) method. Trastuzumab labeled with 111In-DOTA, a known residualizing probe, was coadministered with each radioiodinated antibody to serve as an internal control. Simultaneous measurement of both 125I and 111In in a single sample is feasible due to the distinct γ energies of these two radionuclides. The plasma clearance profiles for all three radioimmunoconjugates were overlapping (Figure 5A), indicating that the tumor uptakes of radioactivity for 111In-DOTA-, [125I]6-, and 125 I-labeled trastuzumab were not influenced by systemic exposure and that modification of the antibody with [125I]6 did not deleteriously affect its pharmacokinetics. The lack of difference among [125I]6−trastuzumab variants, despite differences in plasma stability, reflects the inability of the radiometric pharmacokinetic assay to distinguish between intact antibody and [125I]6−albumin, a likely product of maleimide exchange with reactive thiols.26 At 3 days, tumor uptake for each of the three [125I]6− trastuzumab variants was more than triple that of 125I− trastuzumab with values of 24.6 ± 4.2 (HC-A114C), 27.7 ± 1.4 (LC-V205C), 21.2 ± 1.8 (Fc-S396C), and 6.1 ± 0.3 (tyrosinemodified) percentage of injected dose per gram (%ID/g), respectively (Figure 5B). By comparison, 3-day tumor uptake of 111 In-DOTA−trastuzumab was even higher at 48.7 ± 3.8%ID/ g. These data suggest that [125I]6-associated catabolites are

Figure 5. Plasma pharmacokinetics (A) and biodistribution at 3 days (B) of trastuzumab radiolabeled by various methods in KPL-4 xenograft-bearing mice. Trastuzumab was labeled with 125I by traditional tyrosine modification (black), by site-specific (HCA114C/LC-V205C/Fc-S396C) modification with [125I]6 (blue, green, and red, respectively), or by lysine modification with 111InDOTA (gray). Uptake is expressed as percentage of injected dose per g of tissue (%ID/g).

residualized to a greater extent than [125I]iodotyrosine. Apart from tumor, the radioactivity levels in tissues obtained by trastuzumab labeled by the five different methods were largely similar. A higher renal uptake of radioactivity was observed for [125I]6−trastuzumab, particularly the Fc variant, relative to 125 I−trastuzumab. Renal levels of radioactivity following injection of 125I-, 125I-6- (HC), 125I-6- (LC), 125I-6- (Fc), and 111 In-DOTA-labeled trastuzumab were roughly 3, 10, 8, 20 and 6%ID/g, respectively. In addition, both splenic and hepatic uptake of 111In were higher than for any of the [125I]-labeled molecules. At 1 week, the trends in tumor and tissue uptake were somewhat similar to those observed in the 3-day data, although the residualization of 111In-DOTA appears to be more sustained than [125I]6 (Supporting Information, Figure S11). Mean tumor uptakes were 6.9 ± 3.7 (HC-A114C), 8.2 ± 3.8 (LCV205C), 5.9 ± 0.7 (Fc-S396C), 1.7 ± 0.3 (tyrosine-modified), and 21.2 ± 5.6(111In-DOTA)%ID/g, respectively. The levels of 111 In uptake in %ID/g for liver (7.2) and spleen (13) were considerably higher than that for any of the 125I-labeled analogues. Noninvasive single photon emission computed tomography (SPECT) imaging was performed in live, anesthetized KPL-4 tumor-bearing mice in order to complement the biodistribution study arm (Figure 6, upper). X-ray computed tomography (CT) was performed prior to SPECT without movement or bed adjustment to allow anatomical coregistration of radioactivity with tissue structures. Seventy-two hour SPECT-CT D

dx.doi.org/10.1021/jm401365h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

the reconstructed field-of-view, which is smaller for SPECT than for CT on our system. In contrast, mostly blood pool uptake was visible in the midline plane for the other four variants (Supporting Information, Figure S12). The image quality and resolution for 111In-DOTA was noticeably inferior to the 125I-labeled variants, as evident from the degree of pixelation and bleed-over (Figure 6). This observation may be attributed to the roughly 10-fold γ energy of 111In relative to 125 I.

■

DISCUSSION The HC-A114C, LC-V205C, and Fc-S396C variants of [125I] 6−trastuzumab yielded more than triple the tumor radioactivity relative to traditional tyrosine radioiodination at 3 days postinjection but still fell short (roughly 50%) of 111In-DOTA (Figure 5B). The presence of the metal in the latter may be the reason for this discrepancy; however, previous studies showed that the presence of In3+ did not affect the level of residualization in antibodies labeled with DTPA-appended radioiodinated peptides.17 Furthermore, as others have noted,6 chelates have an ability to scavenge iron or other adventitious metals in tissue culture media or in vivo. Importantly, DOTA forms complexes with transition metals, alkaline earth metals (e.g., calcium), and several lanthanides.29 The stabilities (log Kd, defined as dissociation constant) for the transition metal complexes of DOTA range from 24 to 29 for Fe(III) to 19−21 for Zn(II), with a less stable value of 16 for the Ca(II) complex.29 The promiscuous complexation chemistry of DOTA makes it possible that a range of metal ions could be bound in the chelator cavity within a biological matrix, but iron is of particular relevance because of its presence in circulating plasma proteins such as transferrin. Alternatively, the level of tumor residualization of [125I]6associated catabolites may be inferior to that of 111In-DOTA− lysine, potentially placing HIP-DOTA at a disadvantage with respect to radiometal-based labels. However, size exclusion radiochromatographic analysis of kidney homogenates from a follow-up study in nontumor-bearing mice revealed higher levels of nonintact, low molecular weight radiocatabolites for [125I]6-[HC]−trastuzumab relative to 125I-trastuzumab (65 and 5% of renal radioactivity, respectively) (Supporting Information, Figure S13). Importantly, these data demonstrate that renal accumulation of [125I]6-associated radioactivity occurs independently of tumor uptake and are further supported by in vitro evidence of partial deconjugation in plasma (Supporting Information, Figure S8−S10). Another possibility is that dehalogenation, cleavage of the halogen−carbon bond, is affecting the probe’s tumor retention (see below). Enzymatic dehalogenation also contributes to the underlying discrepancy in tumor uptake between targeted molecules labeled with radiohalogens and radiometals. Although mostly concentrated in the thyroid, dehalogenases are also present elsewhere including the liver and kidneys.30,31 The use of organostannane precursors to derive charged metaiodobenzoate derivatives as residualizing iodine probes (see Figure 1C) is intended to avert this problem by eliminating the presence of a phenol group, thus escaping specific molecular recognition by these enzymes.4−6 Indeed, similar strategies have yielded nonresidualizing iodine probes that are more resistant to dehalogenation.25,32,33 However, both the present and previous work17,34 demonstrate that even 2-iodophenol derivatives may be successfully incorporated into residualizing probes. Consequently, the efflux of iodotyrosine and/or its catabolites

Figure 6. SPECT-CT imaging (upper) at 24 and 72 h and whole-body autoradiographic imaging (lower) at 72 h indicate relative degrees of tracer residualization in KPL-4 xenograft-bearing mice following intravenous administration of trastuzumab radiolabeled by five different methods. In the three-dimensional volume renderings from a coronal perspective (upper), the skeletal images are derived from the X-ray (anatomical) CT data while the relative levels of radioactivity from the SPECT data are indicated in a false-color scale. Tumor-toblood (T:B) ratios of radioactive uptake are shown. Post-mortem cryosection images from a sagittal perspective (lower) were acquired from the same mice as in the upper panel. False-colored phosphorimages (left) and digital photographs (right) are shown for each mouse in the tumoral plane. Tissues are labeled as tumor (T), liver (L), and kidney (K).

images of mice receiving a single intravenous injection of radiolabeled trastuzumab qualitatively revealed tumor-to-blood ratios in the following rank order: 111In-DOTA > 125I-6 (LCV205C) > 125I-6 (HC-A114C) > 125I-6 (Fc-S396C) > 125I (tyrosine-modified). The better agreement between 111InDOTA and [125I]6 by SPECT-CT in Figure 6, relative to Figure 5, may be explained by higher receptor occupancy and/ or altered internalization rates due to the higher doses of radiolabeled antibody necessary for image acquisition. Whole-body localization of radioactivity in KPL-4 tumorbearing mice was determined by autoradiography (Figure 6, lower). Digital photographs of the sagittal cryosections allow anatomical coregistration. The same rank order of relative tumor uptake (upper panel) was qualitatively observed: 111InDOTA > [125]I6 (LC-V205C) > [125]I6 (HC-A114C) > [125]I6 (Fc-S396C) > 125I (tyrosine-modified). Elevated renal uptake, especially for the Fc variant of 125I-6−trastuzumab, was evident by both SPECT and autoradiography and recapitulated the data obtained by organ harvest in Figure 5. In the midline plane (lower panel), a strikingly high uptake of radioactivity in the thyroid gland was the dominant feature in the 125I−trastuzumab (tyrosine-modified) autoradiograph (Supporting Information, Figure S12) despite the NaI blocking doses administered at both 24 and 1 h prior to dosing. The absence of this feature in the SPECT images could be explained by image artifacts near the top/bottom of the images or by the thyroid lying outside of E

dx.doi.org/10.1021/jm401365h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

DOTA chelate forms very kinetically stable complexes with larger +3 metal cations including the γ emitter 111In, the low energy β/negatron (β−) emitter lutetium-177 (177Lu), the high energy β-emitter yttrium-90 (90Y), the positron (β+) emitter yttrium-86 (86Y),9 and nonradioactive gadolinium (natGd) as a magnetic resonance spectroscopy contrast agent.3 We have developed a facile, tin-free, three-step synthetic route to a residualizing probe amenable to oxidative radioiodination and antibody labeling. In particular, the multicomponent Ugi reaction proved to be an efficient means for covalently linking our three desired components: a charged residualizing anchor, a phenol for iodine incorporation, and an activated linker for antibody conjugation. When conjugated to trastuzumab, a marketed anti-HER2 antibody, this novel probe demonstrated a 3−4-fold increase, relative to traditional tyrosine radioiodination, in tumor uptake at 3 days postinjection. Overall, our synthetic route to [125I]6, a novel residualizing radioiodine probe, is potentially useful for targeted radioimmunotherapy of cancer and may also benefit preclinical and translational research efforts for antibodies across multiple therapeutic areas.

following lysosomal proteolysis may be a more vexing problem than enzymatic dehalogenation. However, it is important to note that dehalogenases are specific for mono- or diiodinated Ltyrosine, therefore the D-tyrosine-containing IMP-R4 and the Ugi-derived [125I]6 might not even be recognized by these enzymes. Higher renal uptake of the HC-A114C, LC-V205C, and FcS396C analogues of [125I]6 thio-trastuzumab, relative to 125Itrastuzumab, was expected based on our previous experience with residualizing labels.28 In particular, the Fc variant demonstrated high renal and lower tumor uptakes (Figure 5B), consistent with its poor plasma stability (Supporting Information, Figure S10). Administration of a basic amino acid such as lysine is a potential future strategy to lower the renal uptake associated with [125I]6, as it has been previously demonstrated to reduce renal tubular reabsorption of radiometal-labeled antibody fragments by neutralization of negative charges within the luminal tubular cell surface.35 Despite the differences in both in vitro plasma stability (Supporting Information, Figure S8−S10) and in vivo radioactivity levels in tumor and kidneys (Figure 5B), no differences in plasma pharmacokinetics were observed among the three analogues of [125I]6−trastuzumab (Figure 5B). Since albumin is the most abundant free thiol-containing plasma protein, maleimide exchange from trastuzumab to albumin is a likely scenario that is further supported by the appearance of radiolabeled albumin in the in vitro plasma stability study (Supporting Information, Figure S10). The similar clearances of the HC, LC, and Fc analogues may reflect the fact that both immunoglobulins and albumin are protected from lysosomal degradation by the recycling receptor, the neonatal Fc receptor (FcRn), thereby promoting long serum half-lives in both cases. Although the present study utilized a THIOMAB version of trastuzumab with engineered cysteines,26 the maleimide derivative [125I]6 has the potential to be conjugated to any thio-containing protein. Although THIOMAB labeling confers advantages such as excellent batch-to-batch reproducibility, it does require the production, time, and expense of locating an appropriate site to introduce the unpaired cysteine into an antibody. For those who prefer to work with conventional antibodies, established methods allow conjugation to hingeregion thiols made available through mild reduction. Alternatively, thiols may be introduced to any antibody through the ε-amino groups of lysine residues by use of a heterobifunctional linker.25 Although the current approach offers several advantages over traditional radioiodination, further refinements could offer significant improvements. For instance, a future generation derivative of 6 allowing direct conjugation to lysines could additionally circumvent the problematic stability of thioether bonds which are highly dependent on the local electrostatic environment. This could improve label stability in vivo by eliminating losses to reverse Michael reactions and reducing overall renal uptake. The ability to conjugate a residualizing probe to thiols may be useful as a quantitative means to estimate cumulative drug delivery of antibody−drug conjugates in which chemotherapeutic drugs are conjugated using similar chemical methodologies.1 In addition, a residualizing halogen probe could benefit the development of radioimmunotherapeutic agents labeled with the β-emitter, 131I (Supporting Information, Table S1).13 Moreover, aside from its residualizing properties, the presence of DOTA in [125I]6 lends the possibility of incorporating a metal to yield a multimodal probe. The

■

EXPERIMENTAL SECTION

General Synthetic Methods. Unless otherwise noted, all reactions were run under an argon atmosphere in oven-dried glassware. Reactions were stirred using Teflon-coated magnetic stirrer bars. Reactions were monitored using thin layer silica gel chromatography (TLC) using 0.25 mm silica gel 60F plates with fluorescent indicator from EMD Chemicals. Plates were visualized under UV light. Products were purified via preparative reverse phase chromatography with UV detection at 254 nm using a gradient of 5− 50% water/ACN (0.1% formic acid) at 70 mL/min in 10 min; the column was a Phenomenex Gemini-NX 10 μ C18 110 Å, 100 mm × 30.00 mm. Acetonitrile (CH3CN) and trifluoroacetic acid (TFA) were purchased from EMD Chemicals. Acetic acid (CH3COOH), ethyl isocyanoacetate, deuterated chloroform (CDCl3), and N-chlorosuccinimide (NCS) were purchased from Alpha Aesar. Chloroform (CHCl3) was purchased from Mallinckrodt. Methanol (MeOH) was purchased from J. T. Baker. 4-Hydroxyphenylacetic acid was purchased from Sigma-Aldrich. Lithium hydroxide monohydrate (LiOH·H2O) was purchased from Acros. Paraformaldehyde was purchased from TCI America. 2-Maleimidoethylamine hydrochloride was purchased from Oakwood Products. 2-Aminoethyl-monoamide-DOTA-tris(tbutyl ester) hydrobromide was purchased from Macrocyclics, Inc. NMR spectra were acquired on either a Bruker 300 UltraShield (1H at 300 MHz, 13C at 75 MHz) or a Bruker Avance II 400 (1H at 400 MHz, 13C at 100 MHz) magnetic resonance spectrometer. 1H chemical shifts are reported relative to the residual solvent peak (chloroform = 7.26 ppm) as follows: chemical shift (δ), (multiplicity, integration). 13C chemical shifts are reported relative to the residual deuterated solvent 13C signals (CDCl3 = 77.16 ppm). High resolution LC-MS was performed using an Agilent 1200 series LC with PLRP-S, 1000 Å, column (50 mm × 2.1 mm, Varian Inc.) coupled to an Agilent 6220 Accurate-Mass TOF LC/MS mass spectrometer. Chromatography. Reversed-phase radiochromatography was performed using an Agilent 1200 HPLC system coupled with a γRAM model 4 radioactive detector (LabLogic, formerly IN/US) running Laura version 4 software. A flow rate of 1 mL/min and a 30 min gradient of 5% A, 95% B to 95% A, 5% B were utilized where A = 0.1% trifluoroacetic acid (aq) and B = acetonitrile. Size-exclusion radiochromatography was performed using an Agilent 1100 HPLC system coupled with a Raytest Gabi Star radioactive flow monitor running ChemStation software. The colum was a BioSep SEC S-3000, 300 mm × 7.8 mm, 5 μm (Phenomenex). The mobile phase was PBS, and the flow rate was 0.5 mL/min for 32 min (isocratic). The ChemStation analogue to digital converter was set F

dx.doi.org/10.1021/jm401365h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

addition of 3 μL (1 mCi) of Na125I in 0.1 N NaOH (aq) (PerkinElmer). The mixture was briefly vortexed and transferred to a glass test tube (Pierce) that is precoated with 1,3,4,6-tetrachloro-3α,6αdiphenylglucoluril (i.e., Iodogen). The radioiodination reaction was allowed to proceed without heating for 5 min with gentle intermittent shaking every 30 s on an Eppendorf thermomixer. The solution was transferred to a 5 mL glass vial and characterized by reversed-phase HPLC without purification. After solvent (i.e., water) removal by rotary evaporation, a 0.5 mL quantity of 95% trifluoroacetic acid/5% water was added followed by magnetic stirring for 2 h. Complete removal of acid by rotary evaporation was facilitated by successive additions of toluene in 250 μL aliquots followed by an overnight vacuum. To the residue, still in a 5 mL glass vial, was added 50 μL of phosphate-buffered saline, pH 7.4 (PBS). After a brief vortex, a pH strip was used to ensure that the pH was in the range of 6.5−7.5. This aqueous solution was quickly transferred to a freshly thawed 500 μg aliquot of deblocked ThioMab (thio-trastuzumab−HC-A114C (9.7 mg/mL), thio-trastuzumab−LC-V205C (15.1 mg/mL), or thiotrastuzumab−Fc-S396C (5 mg/mL) (Kabat numbering); Genentech, Inc.), followed by an additional 150 μL of PBS. These ThioMabs were engineered such that site-specific amino acid residues were mutated to cysteine.20 If necessary, the final pH was carefully adjusted to 7.5 by addition of 50 mM borate buffer pH 8.5 in 10 μL increments. The reaction mixture was constantly mixed at 350 rpm for 1 h, followed by addition of a 2-fold molar excess of iodoacetic acid to quench, for an additional 10 min, any remaining free thiols. The amount of radioactivity in these reactions ranged from 500 to 527 μCi, with the remainder of the initial 1 mCi presumably lost by sticking to vials and/or volatility during vacuum. The desired radioimmunoconjugates were purified using PBS-equilibrated NAP5 desalting columns (GE Healthcare) and analyzed by size exclusion chromatography. Final product activities were 255, 298, and 224 μCi for the HC-A114C, LCV205C, and Fc-S396C variants, respectively. These corresponded to conjugation yields of 62, 55, and 50%. Trastuzumab was radiolabeled by traditional means, through its tyrosine residues, with 125I by a modified indirect Chizzonite method as previously described (840 μCi, 88% radiochemical yield).36 Trastuzumab was conjugated to DOTA and radiolabeled with 111In as previously described (1.19 mCi, 71% radiochemical yield).37 SDS-PAGE. Samples of [125I]6-labeled thio-trastuzumab (LCV205C, HC-A114C, and Fc-S396C) in phosphate buffered saline were analyzed by sodium-dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE). For the endoproteinase lys-C digestion, 2.5 μg of each antibody (5 × 106 cpm) was combined with 1 μL of reconstituted (0.5 mg/mL) lys-C (lysyl endopeptidase, Wako, cat. 129-02541) and incubated at pH 8 for 1 h at 37 °C. Following the incubation, the antibodies were combined with NuPAGE 4× LDS sample buffer (pH 8.4) (cat. NP0007) and water. For the dithiothreitol (DTT) reductions, 2.5 μg of each antibody was combined directly with sample buffer, DTT, and water. Antibodies from both preparations were incubated at 70 °C for 10 min, then applied to a NuPAGE 4−12% Bis-Tris gel (cat. NP0321BOX) with NuPAGE 1× MOPS SDS running buffer (cat. NP0001). The gel was stained with Coomassie Blue R250 dye. All NuPAGE reagents were obtained from Invitrogen, Corp. The gel was exposed for 5 min at room temperature to phosphor-imaging plates (YBIP 2025MS; Fuji Film Medical Systems, Inc.) and scanned using a Fuji Film BAS-5000 scanner (Fuji Film Medical Systems, Inc.) to obtain digital images of the radioactivity in the gel. In Vitro Plasma Stability. The LC-V205C, HC-A114C, and FcS396C (Kabat numbering) versions of [125I]6−trastuzumab were added to CD-1 mouse plasma (Bioreclamation, LLC) at 2 × 107 counts per min (cpm)/mL. The mixture was incubated at 37 °C with gentle rotation for 0, 6, 24, and 96 h. The samples were transferred to dry ice and stored at −70 °C until analysis by size-exclusion HPLC. Biodistribution and Pharmacokinetics. All animal studies were conducted in accordance with the guidelines of the American Association for Accreditation of Laboratory Animal Care and the Genentech Institutional Animal Care and Use Committee. The HER2expressing (3+) human breast cancer cell line KPL-4, obtained in 2006

to 25000 units/mV, peak width 2 s, slit 4 nM (Agilent Technologies). Radioactivity was detected with a raytest Gabi Star radiodetector in line with a standard Agilent 1100 HPLC module system. The data were collected using ChemStation for LC 3D Systems (revision B.01.03[204]). Chemistry. Purities of nonradioactive compounds were assessed by UV monitoring during analytical RP-HPLC and found to be 100, 96.3, and 93.1% for 2, 3, and 4, respectively (see Supporting Information, Figure S1). Tri-tert-butyl-2,2′,2″-(10-(6-(2-(4-hydroxy-phenyl)acetyl)-2,8,11trioxo-12-oxa-3,6,9-triaza-tetradecyl)-1,4,7,10-tetraazacyclododecane-1,4,7-triyl)triacetate (2). To a 50 mL round-bottom flask were added the following: tri-tert-butyl-2,2′,2″-(10-(2-((2-aminoethyl)amino)-2-oxoethyl)-1,4,7,10-tetraazacyclododecane-1,4,7-triyl)triacetate (0.5025 g, 0.724 mmol), paraformaldehyde (0.022 g, 0.724 mmol, 1 equiv), 4-hydroxyphenylacetic acid (0.110 g, 0.724 mmol, 1 equiv), and ethyl isocyanoacetate (0.081 g, 0.724 mmol, 1 equiv). The mixture was refluxed in 10 mL of MeOH under argon at 70 °C for 5 h. After solvent removal in vacuo, purification was achieved by preparative reverse-phase HPLC (gradient from 100% water containing 0.1% formic acid to 50% water with 0.1% formic acid/ 50% CH3CN). Lyophilization yielded a fluffy white hygroscopic solid; 58.3%. 1H NMR (300 MHz, CDCl3) δ 8.87 (m, 1H), 8.48 (m, 1H), 7.03 (m, 2H), 6.81 (m, 2H), 6.34 (br s, 1H), 4.20−2.80 (m, 36H), 1.45 (s, 27H), 1.25 (t, J = 7.1 Hz, 3H). 13C NMR (75 MHz, CDCl3) δ 173.66, 173.23, 170.18, 169.99, 169.77, 169.65, 167.67, 156.60, 156.40, 130.36, 130.04, 125.58, 116.03, 115.81, 82.11, 82.02, 81.98, 81.93, 61.34, 61.24, 56.61, 55.81, 55.48, 52.81, 52.54, 52.06, 50.79, 50.58, 49.99, 49.74, 49.25, 48.77, 47.08, 41.43, 39.59, 37.97, 28.13, 14.16. HRMS (m/z): [M]+ for C44H73N7O12 892.5395, found 892.5368. 2-(2-(2-(4-Hydroxyphenyl)-N-(2-(2-(4,7,10-tris(2-(tert-butoxy)-2oxoethyl)-1,4,7,10-tetra-azacyclododecan-1-yl)acetamido)ethyl)acetamido) acetamido)acetic Acid (3). To a 50 mL round-bottom flask loaded with 2 (0.260 g, 0.291 mmol) was added LiOH·H2O (0.018 g, 0.437 mmol, 1.5 equiv), 6 mL of EtOH, and 2 mL of H2O. The reaction was stirred at room temperature for 12 h. After solvent removal in vacuo, purification was achieved by preparative reversephase HPLC (gradient from 100% 0.1% formic acid in water to 50% 0.1% formic acid in water/50% CH3CN). Lyophilization yielded a fluffy white hygroscopic solid; 65.5%. 1H NMR (300 MHz, CDCl3) δ 9.34 (m, 1H), 8.57 (s, 1H), 7.02 (m, 2H), 6.86 (m, 2H), 5.82 (br s, 1H), 4.10−2.80 (m, 34H), 1.43 (s, 27H). 13C NMR (75 MHz, CDCl3) δ 174.37, 173.76, 173.47, 170.14, 169.72, 169.24, 166.96, 156.90, 156.73, 130.30, 129.92, 124.58, 116.17, 115.92, 81.91, 81.86, 81.73, 56.55, 55.74, 52.98, 52.61, 52.25, 51.88, 51.84, 50.55, 49.16, 49.06, 47.69, 43.65, 43.63, 43.21, 39.51, 38.29, 37.84, 28.15. HRMS (m/z): [M]+ for C42H69N7O12 864.5082, found 864.5082. Tri-tert-butyl-2,2′,2″-(10-(14-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1yl)-6-(2-(4-hydroxyphenyl)-acetyl)-2,8,11-trioxo-3,6,9,12-tetraazatetradecyl)-1,4,7,10-tetraazacyclododecane-1,4,7-triyl)triacetate (4). To a 50 mL round-bottom flask loaded with 3 (0.113 g, 0.131 mmol) were added 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (0.038 g, 0.198 mmol, 1.5 equiv) and 2-maleimidoethylamine·HCl (0.035 g, 0.198 mmol, 1.5 equiv). The mixture was dissolved in 10 mL of CH3CN and stirred for 12 h at room temperature. After solvent removal in vacuo, purification was achieved by preparative reversephase HPLC (gradient from 100% 0.1% formic acid in water to 50% 0.1% formic acid in water/50% CH3CN). Lyophilization yielded a fluffy white hygroscopic solid; 41.4%. 1H NMR (300 MHz, CDCl3) δ 9.02 (m, 1H), 8.07 (m, 1H), 7.01 (m, 2H), 6.79 (m, 2H), 6.65 (s, 1H), 4.23−2.80 (m, 34H), 1.45 (s, 27H). 13C NMR (75 MHz, CDCl3) δ 173.67, 173.43, 171.04, 170.33, 170.26, 170.08, 169.94, 169.86, 168.64, 156.55, 156.47, 134.16, 130.33, 130.06, 125.57, 125.09, 115.95, 115.84, 81.93, 81.87, 81.81, 81.70, 56.63, 56.08, 55.90, 53.12, 52.01, 50.95, 49.79, 49.41, 49.39, 47.22, 43.44, 43.07, 39.58, 39.15, 38.25, 38.10, 37.74, 37.34, 28.17. HRMS (m/z): [M]+ for C48H75N9O13 986.5562, found 986.5579. Purity was determined to be 93.1% by RP-HPLC. Radiochemistry. A 5 μL aliquot of a 1 mg/mL solution of 4 in chloroform was evaporated to dryness by N2 blowdown. The residue was dissolved in 100 μL of 0.1% CH3COOH (aq) followed by the G

dx.doi.org/10.1021/jm401365h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

from Dr. J. Kurebayashi,27 was used in most studies. The cells were cultured in RPMI 1640 media plus 1% L-glutamine with 10% FBS. The KPL-4 cell line was authenticated by short tandem repeat profiling (AMEL, X; CSF1PO, 11,13; D13S317, 12; D16S539, 12; D5S818, 11,13; D7S820, 9,10; TH01, 9; TPOX, 8; vWA, 14,19) and by singlenucleotide polymorphism genotyping. Both methods indicate that KPL-4 cells are unique when compared with a database of more than 1000 human cancer cell lines. C.B-17 Icr SCID (severe combined immunodeficient; Inbred) female mice (Charles River Laboratories), weighing between 20 and 25 g were inoculated in the right mammary fat pad with approximately 3 million KPL-4 cells in a 50:50 suspension of Hanks’ Buffered Salt Solution (Invitrogen) and Matrigel (BD Biosciences) in at most 0.2 mL/mouse. When mean tumor volume reached at least 250 mm3, mice received a single bolus intravenous injection via the tail vein containing 111In-trastuzumab (5 μCi) together with either 125I-trastuzumab (5 μCi) or 125I-6-trastuzumab (5 μCi). The total protein doses associated with these radiolabeled trastuzumab tracers were 0.4 mg/kg for [125I]6 (HC-A114C), 0.5 mg/ kg for [125I]6 (LC-V205C), 0.3 mg/kg for [125I]6 (Fc-S396C), and 0.2 mg/kg for 125I (tyrosine-labeled). To minimize thyroid sequestration of 125I, 100 μL of 30 mg/mL of sodium iodide was intraperitoneally administered 1 and 24 h before dosing. Blood samples were collected at 5 min, 2 h, 1, 3, and 7 days postinjection via retroorbital bleed, with terminal tissue harvests performed at 3 and 7 days postinjection. Terminally collected samples included liver, spleen, kidneys, lungs, intestine (ileum), muscle (gastrocnemius), blood, and tumor. Tissues were counted for radioactivity using a 2480 Wizard2 automatic gamma counter (Perkin-Elmer). Counts per minute values were used to calculate the percent of injected dose per gram of tissue (%ID/g) as previously described.36 SPECT-CT Imaging. In vivo distribution was obtained by single photon emission computed tomography/X-ray computed tomography (SPECT-CT) using modification of previously reported methods.38 Radiolabeling procedures were identical as for the biodistribution study. Doses of radiolabeled trastuzumab were 202 μCi for [125I]6 (HC-A114C), 156 μCi for [125I]6 (LC-V205C), 164 μCi for [125I]6 (Fc-S396C), 414 μCi for 125I (tyrosine-labeled), and 514 μCi for 111InDOTA. These doses were calculated such that each mouse would receive a protein dose of 10 mg/kg without adjustment in specific activity. Immediately after CT acquisition, SPECT images were acquired in a window centered at 35.5 keV for 125I or on two 20% windows centered at the 173- and 247-keV photopeaks of 111In using a high-resolution 5-pinhole collimator and a 5.5 cm radius of rotation. Mice were subsequently euthanized under sedation and tissues collected for whole-body autoradiography. SPECT quantitation was accomplished using Amira software (TGS). Whole-Body Autoradiography. KPL-4 tumor-bearing mice were assessed by quantitative whole-body autoradiography. The mice from the SPECT-CT imaging study were sacrificed after the final noninvasive imaging scan at three days postinjection. Animals were processed for whole-body cryosectioning, exposed, and imaged as described previously.28

■

(trastuzumab) is marketed in the U.S. by Genentech and internationally by Roche.

■

ACKNOWLEDGMENTS We thank Helga Raab, Rachana Ohri, Emily Chan, Gillian Smelick, Joe Ware, Rami Hannoush, Michelle Schweiger, Jose Imperio, Kirsten Messick, Nicole Valle, Cynthia Young, Nina Ljumanovic, Bernadette Johnstone, and Shannon Stainton.

■

ABBREVIATIONS USED CD20, cluster of differentiation 20; DOTA, 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid; DTPA, diethylenetriaminepentaacetic acid; EDC, 1-ethyl-3-(3dimethylaminopropyl)carbodiimide; Fc, fragment crystallizable; HC, heavy chain; HER2, human epidermal growth factor receptor 2; HIP, 4-hydroxy-3-iodophenyl; %ID/g, percentage of injected dose per g; keV, kiloelectron volt; LC, light chain; MCC, N-maleimidomethyl)-cyclohexane-1-carbonyl; SCID, severe combined immunodeficient; SGMIB, N-succinimidyl 4guanidinomethyl-3-iodobenzoate; SIB, N-succinimidyl-3-iodobenzoate; SIPMB, N-succinimidyl-3-[131I]iodo-4-phosphonomethylbenzoate

■

ASSOCIATED CONTENT

S Supporting Information *

NMR spectra, chromatographic data, mass spectra, plasma stability profiles, renal metabolite analysis, and additional biodistribution and whole-body autoradiography data. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

REFERENCES

(1) Wu, A. M.; Senter, P. D. Arming antibodies: prospects and challenges for immunoconjugates. Nature Biotechnol. 2005, 23, 1137− 1146. (2) Davies, A. J. Radioimmunotherapy for B-cell lymphoma: Y-90 ibritumomab tiuxetan and I-131 tositumomab. Oncogene 2007, 26, 3614−3628. (3) Boswell, C. A.; Brechbiel, M. W. Development of radioimmunotherapeutic and diagnostic antibodies: an inside-out view. Nucl. Med. Biol. 2007, 34, 757−778. (4) Vaidyanathan, G.; Affleck, D. J.; Li, J.; Welsh, P.; Zalutsky, M. R. A polar substituent-containing acylation agent for the radioiodination of internalizing monoclonal antibodies: N-succinimidyl 4-guanidinomethyl-3-[131I]iodobenzoate ([131I]SGMIB). Bioconjugate Chem. 2001, 12, 428−438. (5) Shankar, S.; Vaidyanathan, G.; Affleck, D.; Welsh, P. C.; Zalutsky, M. R. N-Succinimidyl 3-[131I]iodo-4-phosphonomethylbenzoate ([131I]SIPMB), a negatively charged substituent-bearing acylation agent for the radioiodination of peptides and mAbs. Bioconjugate Chem. 2003, 14, 331−341. (6) Vaidyanathan, G.; White, B. J.; Affleck, D. J.; Zhao, X. G.; Welsh, P. C.; McDougald, D.; Choi, J.; Zalutsky, M. R. SIB-DOTA: a trifunctional prosthetic group potentially amenable for multi-modal labeling that enhances tumor uptake of internalizing monoclonal antibodies. Bioorg. Med. Chem. 2012, 20, 6929−6939. (7) Boswell, C. A.; Bumbaca, D.; Fielder, P. J.; Khawli, L. A. Compartmental tissue distribution of antibody therapeutics: experimental approaches and interpretations. AAAPS J. 2012, 14, 612−618. (8) Wilbur, D. S. Radiohalogenation of proteins: an overview of radionuclides, labeling methods, and reagents for conjugate labeling. Bioconjugate Chem 1992, 3, 433−470. (9) Nayak, T. K.; Brechbiel, M. W. Radioimmunoimaging with longer-lived positron-emitting radionuclides: potentials and challenges. Bioconjugate Chem. 2009, 20, 825−841. (10) Shih, L. B.; Thorpe, S. R.; Griffiths, G. L.; Diril, H.; Ong, G. L.; Hansen, H. J.; Goldenberg, D. M.; Mattes, M. J. The processing and fate of antibodies and their radiolabels bound to the surface of tumor cells in vitro: a comparison of nine radiolabels. J. Nucl. Med. 1994, 35, 899−908. (11) Perera, R. M.; Zoncu, R.; Johns, T. G.; Pypaert, M.; Lee, F. T.; Mellman, I.; Old, L. J.; Toomre, D. K.; Scott, A. M. Internalization, intracellular trafficking, and biodistribution of monoclonal antibody

AUTHOR INFORMATION

Corresponding Author

*Phone: 650-467-4603. E-mail: [email protected]. Notes

The authors declare the following competing financial interest(s): All authors have held financial interest as employees of Genentech, a member of the Roche Group. Herceptin H

dx.doi.org/10.1021/jm401365h | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry

Article

806: a novel anti-epidermal growth factor receptor antibody. Neoplasia 2007, 9, 1099−1110. (12) Rogers, B. E.; Franano, F. N.; Duncan, J. R.; Edwards, W. B.; Anderson, C. J.; Connett, J. M.; Welch, M. J. Identification of metabolites of 111In-diethylenetriaminepentaacetic acid-monoclonal antibodies and antibody fragments in vivo. Cancer Res. 1995, 55, 5714s−5720s. (13) Milenic, D. E.; Brady, E. D.; Brechbiel, M. W. Antibody-targeted radiation cancer therapy. Nature Rev. Drug Discovery 2004, 3, 488−499. (14) Thorpe, S. R.; Baynes, J. W.; Chroneos, Z. C. The design and application of residualizing labels for studies of protein catabolism. FASEB J. 1993, 7, 399−405. (15) Jefferis, R. Recombinant antibody therapeutics: the impact of glycosylation on mechanisms of action. Trends Pharmacol. Sci. 2009, 30, 356−362. (16) Stein, R.; Govindan, S. V.; Mattes, M. J.; Chen, S.; Reed, L.; Newsome, G.; McBride, B. J.; Griffiths, G. L.; Hansen, H. J.; Goldenberg, D. M. Improved iodine radiolabels for monoclonal antibody therapy. Cancer Res. 2003, 63, 111−118. (17) Govindan, S. V.; Mattes, M. J.; Stein, R.; McBride, B. J.; Karacay, H.; Goldenberg, D. M.; Hansen, H. J.; Griffiths, G. L. Labeling of monoclonal antibodies with diethylenetriaminepentaacetic acidappended radioiodinated peptides containing D-amino acids. Bioconjugate Chem. 1999, 10, 231−240. (18) Ugi, I.; Meyr, R.; Fetzer, U.; Steinbruckner, C. Versuche mit Isonitrilen. Angew. Chem. 1959, 71, 386. (19) Domling, A.; Ugi, I. I. Multicomponent Reactions with Isocyanides. Angew. Chem., Int. Ed. Engl. 2000, 39, 3168−3210. (20) Junutula, J. R.; et al. Site-specific conjugation of a cytotoxic drug to an antibody improves the therapeutic index. Nature Biotechnol. 2008, 26, 925−932. (21) DeNardo, G. L.; DeNardo, S. J.; O’Donnell, R. T.; Kroger, L. A.; Kukis, D. L.; Meares, C. F.; Goldstein, D. S.; Shen, S. Are radiometallabeled antibodies better than iodine-131-labeled antibodies: comparative pharmacokinetics and dosimetry of copper-67-, iodine-131-, and yttrium-90-labeled Lym-1 antibody in patients with nonHodgkin’s lymphoma. Clin. Lymphoma 2000, 1, 118−126. (22) Sharkey, R. M.; Behr, T. M.; Mattes, M. J.; Stein, R.; Griffiths, G. L.; Shih, L. B.; Hansen, H. J.; Blumenthal, R. D.; Dunn, R. M.; Juweid, M. E.; Goldenberg, D. M. Advantage of residualizing radiolabels for an internalizing antibody against the B-cell lymphoma antigen, CD22. Cancer Immunol. Immunother. 1997, 44, 179−188. (23) Li, W. P.; Lewis, J. S.; Kim, J.; Bugaj, J. E.; Johnson, M. A.; Erion, J. L.; Anderson, C. J. DOTA-D-Tyr(1)-octreotate: a somatostatin analogue for labeling with metal and halogen radionuclides for cancer imaging and therapy. Bioconjugate Chem. 2002, 13, 721−728. (24) Tei, L.; Gugliotta, G.; Avedano, S.; Giovenzana, G. B.; Botta, M. Application of the Ugi four-component reaction to the synthesis of ditopic bifunctional chelating agents. Org. Biomol. Chem. 2009, 7, 4406−4414. (25) Khawli, L. A.; van den Abbeele, A. D.; Kassis, A. I. N-(m[125I]Iodophenyl)maleimide: an agent for high yield radiolabeling of antibodies. Int. J. Radiat. Appl. Instrum. B 1992, 19, 289−295. (26) Shen, B. Q.; et al. Conjugation site modulates the in vivo stability and therapeutic activity of antibody−drug conjugates. Nature Biotechnol. 2012, 30, 184−189. (27) Kurebayashi, J.; Otsuki, T.; Tang, C. K.; Kurosumi, M.; Yamamoto, S.; Tanaka, K.; Mochizuki, M.; Nakamura, H.; Sonoo, H. Isolation and characterization of a new human breast cancer cell line, KPL-4, expressing the Erb B family receptors and interleukin-6. Br. J. Cancer 1999, 79, 707−717. (28) Pastuskovas, C. V.; et al. Effects of anti-VEGF on pharmacokinetics, biodistribution, and tumor penetration of trastuzumab in a preclinical breast cancer model. Mol. Cancer Ther. 2012, 11, 752−762. (29) Viola-Villegas, N.; Doyle, R. P. The coordination chemistry of 1,4,7,10-tetraazacyclododecane-N,N′,N″,N″-tetraacetic acid (H4DOTA): structural overview and analyses on structure−stability relationships. Coord. Chem. Rev. 2009, 253, 1906−1925.

(30) Gnidehou, S.; Caillou, B.; Talbot, M.; Ohayon, R.; Kaniewski, J.; Noel-Hudson, M. S.; Morand, S.; Agnangji, D.; Sezan, A.; Courtin, F.; Virion, A.; Dupuy, C. Iodotyrosine dehalogenase 1 (DEHAL1) is a transmembrane protein involved in the recycling of iodide close to the thyroglobulin iodination site. FASEB J. 2004, 18, 1574−1576. (31) Auf Dem Brinke, D.; Hesch, R. D.; Kohrle, J. Re-examination of the subcellular localization of thyroxine 5′-deiodination in rat liver. Biochem. J. 1979, 180, 273−279. (32) Khawli, L. A.; Kassis, A. I. Synthesis of 125I labeled Nsuccinimidyl p-iodobenzoate for use in radiolabeling antibodies. Int. J. Radiat. Appl. Instrum. B 1989, 16, 727−733. (33) Zalutsky, M. R.; Narula, A. S. Radiohalogenation of a monoclonal antibody using an N-succinimidyl 3-(tri-n-butylstannyl)benzoate intermediate. Cancer Res. 1988, 48, 1446−1450. (34) Stein, R.; Govindan, S. V.; Hayes, M.; Griffiths, G. L.; Hansen, H. J.; Horak, I. D.; Goldenberg, D. M. Advantage of a residualizing iodine radiolabel in the therapy of a colon cancer xenograft targeted with an anticarcinoembryonic antigen monoclonal antibody. Clin. Cancer Res. 2005, 11, 2727−2734. (35) Behr, T. M.; Sharkey, R. M.; Juweid, M. E.; Blumenthal, R. D.; Dunn, R. M.; Griffiths, G. L.; Bair, H.-J.; Wolf, F. G.; Becker, W. S.; Goldenberg, D. M. Reduction of the renal uptake of radiolabeled monoclonal antibody fragments by cationic amino acids and their derivatives. Cancer Res. 1995, 55, 3825−3834. (36) Boswell, C. A.; Ferl, G. Z.; Mundo, E. E.; Schweiger, M. G.; Marik, J.; Reich, M. P.; Theil, F. P.; Fielder, P. J.; Khawli, L. A. Development and evaluation of a novel method for preclinical measurement of tissue vascular volume. Mol. Pharmaceutics 2010, 7, 1848−1857. (37) Boswell, C. A.; et al. Impact of drug conjugation on pharmacokinetics and tissue distribution of anti-STEAP1 antibody− drug conjugates in rats. Bioconjugate Chem. 2011, 22, 1994−2004. (38) Boswell, C. A.; et al. An integrated approach to identify normal tissue expression of targets for antibody−drug conjugates: case study of TENB2. Br. J. Pharmacol. 2013, 168, 445−457.

I

dx.doi.org/10.1021/jm401365h | J. Med. Chem. XXXX, XXX, XXX−XXX