Novel analogs of antitumor agent calixarene 0118: Synthesis, cytotoxicity, click labeling with 2-[18F]fl uoroethylazide, and in vivo evaluation
Tilman L€appchen a, *, Ruud P.M. Dings b, c, Raffaella Rossin a, Justine F. Simon a,
Ton J. Visser d, Martine Bakker a, Priya Walhe a, Tiemen van Mourik a, Katia Donato a, Judy R. van Beijnum e, Arjan W. Griffi oen e, Johan Lub a, *, Marc S. Robillard a,
Kevin H. Mayo b, Holger Grüll a
aDepartment Minimally Invasive Healthcare, Philips Research, High Tech Campus 11, 5656 AE Eindhoven, The Netherlands
bDepartment of Biochemistry, Molecular Biology & Biophysics, University of Minnesota Health Sciences Center, 6-155 Jackson Hall, 321 Church Street, Minneapolis, MN 55455, USA
cDepartment of Radiation Oncology, Winthrop P. Rockefeller Cancer Institute, University of Arkansas for Medical Sciences, 4301 W. Markham St., Slot 771, Little Rock, AR 72205-7199, USA
dSyncom BV, Kadijk 3, 9747 AT Groningen, The Netherlands
eAngiogenesis Laboratory, Department of Medical Oncology, VU University Medical Center, De Boelelaan 1118, 1081 HV Amsterdam, The Netherlands
a r t i c l e i n f o
Article history:
Received 13 September 2014 Received in revised form
15 October 2014
Accepted 16 October 2014 Available online 18 October 2014
Keywords: Compound 0118 Calixarene
PET imaging Click chemistry Fluorine-18
Anti-angiogenic therapy
a b s t r a c t
Calixarene 0118 is a potent anti-angiogenic agent that effectively inhibited tumor growth in preclinical studies, and is currently being evaluated in a phase I clinical trial. We have designed two close mimetics of calixarene 0118 containing a terminal alkynyl-functional group, and developed an optimized semi- automated procedure for radiolabeling with 2-[18F]fl uoroethylazide using click chemistry. Following
18
semi-preparative HPLC purifi cation and formulation, the lower-rim modifi ed analog [ F]6 and the equatorially labeled [18F]13 were obtained in >97% radiochemical purity and overall decay-corrected
isolated radiochemical yields of 18.7 ± 2.7% (n ¼ 4) and 10.2 ± 5.0% (n ¼ 4), respectively, in a total synthesis time of about 2 h. Preliminary in vivo studies in nude mice bearing human tumor xenografts revealed highest accumulation of both tracers in the liver, followed by spleen, kidney, lung and bone, with no substantial uptake in the tumor. Still, these fi rst-in-class radiotracers are a valuable tool for pharmacokinetic profi ling and improvement of calixarene-based anti-angiogenic therapeutics in the future, as similar radiolabeling strategies may be applied to other compounds in the calixarene series. The cold reference compounds of the radiotracers were characterized in terms of cytotoxicity and anti- proliferative effects on HUVEC cells and on MA148 human ovarian carcinoma cells, along with the respective precursors, a small series of 0118 analogs modifi ed with short-chain linear alkyl substituents, and a PEG3-spaced calixarene dimer. While all of the new analogs proved at least equipotent to parent 0118, some of them inhibited HUVEC and MA148 cell growth almost 4- and 10-fold more effectively, rendering these analogs promising candidates for further evaluation in anti-angiogenic cancer therapy.
© 2014 Elsevier Masson SAS. All rights reserved.
1.Introduction
Abbreviations: n-BuLi, n-butyl lithium; DMF, N,N-dimethylformamide; DMSO, dimethyl sulfoxide; HSQC, heteronuclear single quantum coherence; HUVEC, hu- man umbilical vein endothelial cells; %ID/g, percentage of injected dose per gram of tissue; KOtBu, potassium tert-butoxide; NaOMe, sodium methoxide; PBS, phos- phate buffered saline; PEG, polyethylene glycol; PET-CT, positron emission to- mography e computed tomography; SAR, structureeactivity relationship; TFA, trifl uoroacetic acid; VOI, volume of interest.
* Corresponding authors.
E-mail addresses: [email protected], [email protected] (T. L€appchen), [email protected] (J. Lub).
In the past decades, drug-based cancer therapy has evolved from general cytotoxic chemotherapy to increasingly complex personalized treatment regimens using drugs matching the specifi c targets of a particular cancer. Beside new generations of anti- proliferative cytotoxic drugs and anti-hormone therapies, an increasing number of anti-angiogenic agents e small molecules, peptides and therapeutic antibodies e have been developed and
http://dx.doi.org/10.1016/j.ejmech.2014.10.048
0223-5234/© 2014 Elsevier Masson SAS. All rights reserved.
many of those have progressed to the clinic [1]. Angiogenesis, the formation of new blood vessels from pre-existing vasculature, is a crucial prerequisite for primary tumor growth above a size of about 1e2 mm3, and has also been found essential for development of metastases [2e4].
In the early years, most anti-angiogenic agents discovered were endogenous proteins that inhibit EC growth, e.g. platelet factor-4 (PF4), plasmin fragment angiostatin, collagen XVIII fragment endostatin, and bactericidal-permeability-increasing protein (BPI). Containing numerous hydrophobic and cationic residues, and a common antiparallel b-sheet motif, these proteins are character- ized by a high level of compositional and structural similarity [5]. Based on these fi ndings, a new generation of synthetic b-sheet forming peptide 33-mers have been designed, one of which, named anginex, showed particular potent anti-angiogenic activity in vitro and in vivo [6]. Anginex was found to inhibit tumor growth and reduce microvessel density both in murine melanoma and breast sarcoma models, and in athymic mice bearing different types of human tumor xenografts, among others LS174T colon adenocarci- noma, and MA148 and SKOV-3 ovarian carcinoma [7e10]. Exten- sive structureeactivity studies established five hydrophobic residues in b-strands 1 and 2 on the same face of the amphipathic anti-parallel b-sheet structure of anginex as being essential for anti-proliferative activity, and formed the foundation for the design of a series of partial peptide mimetics containing a b-sheet- inducing dibenzofuran (DBF) turn, some of which more effectively inhibited tumor growth in mice than anginex [11].
Although anginex and its partial peptide mimetic 6DBF7 have shown promising anti-tumor effects in vivo, non-peptidic com- pounds are generally known to be superior drugs, mainly because they potentially allow oral administration, typically lack an im- mune response, and can be optimized in terms of chemical and metabolic stability, resulting in a better pharmacokinetic profi le. With the overall backbone dimensions of key residues in the two- stranded beta-sheet of anginex roughly matching those of the calix [4]arene scaffold, calix[4]arene derivatives containing basic and hydrophobic substituents on the upper and lower rim, respectively, were envisioned as potential non-peptidic analogs of anginex.
Indeed, from a library of 23 substituted calix[4]arene analogs, two candidates, termed compound 0118 and 1097, proved to be potent inhibitors of angiogenesis in vitro, and were found to reduce tumor growth and microvessel density in B16 murine melanoma and MA148 xenografts in athymic mice [12]. While since its discovery, calixarene 0118 (Fig. 1) was known to display multimodal activities similar to anginex, such as inhibition of endothelial cell prolifera- tion, migration, and induction of apoptosis [12], it was only recently that both calixarene 0118 [13] and anginex [14] were discovered to also target the same receptor, galectin-1 (gal-1). Interestingly, SAR- data from anginex and HSQC mapping studies with calixarene 0118 indicate that both ligands interact with gal-1 via their hydrophobic surfaces [11,13]. This is very different from the majority of galectin-
1antagonists presently available, which are b-galactoside analogs and glycomimetics targeting the canonical b-galactoside carbohy- drate binding site [15e17]. In fact, calixarene 0118 was established to act as a noncompetitive, allosteric inhibitor of gal-1, with its binding site being located at the back face of gal-1 opposite to the b- galactoside binding site [13]. In the meantime, after successful completion of extensive preclinical studies [12,18], calixarene 0118 (also known as PTX008, OTX008) is currently being evaluated by OncoEthix in a phase I clinical trial in patients with malignant advanced solid tumors (ClinicalTrials.gov: NCT01724320) [17], and ongoing early discovery research has led to the identification of an even more potent analog, PTX013 [19].
Despite the promising in vitro and in vivo data collected for lead compound 0118 and others in the series, radiolabeled analogs have not been reported so far, although they may prove highly valuable, both as a research tool and companion diagnostic to study target engagement in vivo, and for pharmacokinetic profiling of this class of compounds. Designing radiolabeled 0118 analogs without introduction of major structural changes is an intrinsically diffi cult task, particularly in view of the stringent structural constraints established in previous SAR-studies [12,13] indicating that only minor modifi cations are tolerated at the hydrophobic upper rim of the calixarene, and the symmetric nature of the lower rim sub- stituents, which is introduced already at a very early stage in the previously reported chemical syntheses of these compounds. We
Fig. 1. Chemical structure of compound 0118, novel 0118 analogs 11aed, and radiotracers [18F]6 and [18F]13.
have addressed these challenges by devising two alternative click chemistry based approaches building on the design of non- symmetrical 0118 analogs containing a terminal alkynyl func- tional group, either as an equatorial substituent at a methylene bridge position or as a replacement of one of the four lower rim substituents. Copper-catalyzed azide-alkyne cycloaddition (CuAAC) employing the well-established 2-[18F]fl uoroethylazide synthon then gave access to the corresponding radiolabeled triazole ana- logs, which structurally closely resemble the parent calixarene 0118 (Fig. 1).
Here, we report on the radiosynthesis of F-18 labeled calixarene 0118 analogs, including extensive optimization of CuAAC reaction conditions for this type of compounds, and the chemical synthesis of the precursors and reference compounds. In addition, we employed our alkynyl-precursors to prepare a 0118 dimer inter- connected by a PEG3-chain, and created a series of 0118 derivatives monosubstituted at an equatorial methylene bridge position with alkyl substituents of different chain length (Fig. 1). All compounds were characterized in terms of anti-proliferative activity in vitro, and the study was complemented by a preliminary in vivo inves- tigation of both radiotracers in tumor-bearing mice.
2Results and discussion
21.Chemistry
21.1.Synthesis of compound 0118 analogs with modified lower-rim substituents
Compared to compound 0118, synthesis of unsymmetrically substituted analogs is far more challenging, and a number of alternative routes towards click precursor 4 have been explored. The fi rst pathway towards key intermediate 3 involves direct condensation of tetrahydroxycalix[4]arene 1 with ethyl bro- moacetate to yield tri-functionalized compound 2a (Scheme 1, Route A). Although selective trialkylation of tetrahydroxycalix[4]
arenes had been reported in the presence Ba(OH)2$8H2O and BaO [20], attempts using similar conditions to convert 1 with ethyl bromoacetate into 2a resulted in incomplete conversion and concomitant hydrolysis of the ester groups, which has also been observed by Shimizu et al. [21]. To avoid ester hydrolysis, the
condensation was repeated with BaO only without the addition of Ba(OH)2$8H2O, which gave 2a, but only in poor yield (12%) due to complications in the extraction process possibly arising from complex-formation with BaO. When CaH2 was used as the base, no ester cleavage was observed but the conversion of the reac- tion was incomplete and mainly bis-alkylated material was iso- lated. To check the principal feasibility of the fi rst approach, intermediate 2a was reacted with 4-bromobut-1-yne in the presence of K2CO3 and NaI in acetonitrile under refl ux. Disap- pointingly, however, the reaction did not proceed and only starting material was recovered.
Since preparation of the alkyne-functionalized key intermediate
3via the originally devised route was not successful due to insuf- ficient reactivity of 4-bromobut-1-yne in the fi nal alkylation step, it was decided to switch to an alternative approach based on selective monoalkylation with 4-bromobut-1-yne in the first step, and sub- sequent alkylation of the remaining hydroxyl-groups with ethyl bromoacetate (Scheme 1, Route B). Employing published pro- cedures for monoalkylation of tetrahydroxycalix[4]arene 1 with benzyl bromide [22] and various alkyl halides [23], reaction with 4- bromobut-1-yne proved to be very slow. To increase the yield and prevent formation of bis-alkylated material, 4-iodobut-1-yne was used instead of 4-bromobut-1-yne, and a number of different bases (K2CO3, KOtBu, NaOMe) and solvents (acetonitrile, DMF) were tested. Finally, an optimized procedure was developed comprising addition of multiple portions of NaOMe, which led to a conversion of 49% and allowed isolation of a 3:1 mixture of target compound 2b (26% isolated yield) and starting material 1. Since removal of the remaining starting material proved difficult, this mixture was directly used in the next step without further purifi cation. Treat- ment with excess ethyl bromoacetate in the presence of K2CO3 gave compound 3 in 67% yield after repeated purifi cation by silica col- umn chromatography, although the material still contained impu- rities. The partly purifi ed material was then reacted with N,N- dimethylethylenediamine similar to published methods [24], and the crude material purified by preparative HPLC to obtain target compound 4 in a purity of >99%. Copper-catalyzed azide-alkyne cycloaddition (CuAAC) of crude 4 with freshly prepared 2- fluoroethylazide 5 [25] followed by preparative HPLC yielded reference compound 6 in a purity of >99%.
Scheme 1. Synthesis of click precursor 4 and reference compound 6.
Scheme 2. Synthetic strategy towards equatorially substituted calixarene 0118 analogs.
2.1.2.Synthesis of equatorially substituted compound 0118 analogs While focus of earlier work in calixarene chemistry was pri-
marily on modifi cation of lower and upper rim positions, fewer reports address the preparation of calixarenes functionalized at the methylene bridge, and synthesis of these compounds was tradi- tionally achieved by fragment condensation instead of introduction of substituents on the fully formed calixarene. The discovery of a new method for monosubstitution of calix[4]arenes at the methy- lene bridge position via lithiation followed by reaction with elec- trophiles [26] has made these class of compounds more readily accessible, and triggered us to evaluate its applicability for the preparation of a series of equatorially monosubstituted compound 0118 analogs. Starting directly from tetramethoxycalix[4]arene 7 instead of its upper-rim p-tert-butyl analog employed in the orig- inal procedure, the desired series of 2-alkyl substituted tetrame- thoxycalix[4]arenes 8aed were obtained in good yields from the corresponding alkyl iodides and propargyl bromide, respectively (Scheme 2). Interestingly, a lower excess of n-BuLi (1.15 equiv) as employed in a more recent report [27] resulted in signifi cantly lower yields for the present series of compounds, and effi cient re- action apparently did only occur when using a large excess of n- BuLi (4.5 equiv.) and alkyl halide (9 equiv.) as described in the original work [26]. Removal of the methoxy-protecting groups with BBr3 yielded 2-alkyl-25,26,27,28-tetrahydroxycalix[4]arenes 9aed, which were reacted with excess ethyl bromoacetate in the presence of Na2CO3 to the corresponding tetraacetates 10aed. While in previously reported procedures [22,28,29], K2CO3 was used as the base, we found that Na2CO3 favors formation of the desired cone conformer in higher yields. Final isolated yields, however, were on average only about 30% due to substantial losses during purifi cation by silica column chromatography, which was required to obtain the
compounds in high purity. Reaction with N,N-dimethylethylene- diamine [24] afforded the corresponding compound 0118 analogs 11aed in 80e99% yield and >90% purity. In summary, starting from tetrahydroxycalix[4]arene 1, the 5-step synthetic pathway pro- vided the equatorially monoalkylated 0118 analogs 11aed in overall yields ranging from 11% to 24%. To ensure high purity (>99%) of the test compounds for the biological assays, part of the material was further purified by preparative HPLC.
2.1.3.Synthesis of a compound 0118 analog with a single equatorial triazole moiety and a triazole-PEG-tethered 0118 dimer
Copper-catalyzed azide-alkyne cycloaddition of propargyl- modifi ed 0118 derivative 11d with 2-fluoroethylazide 5 cleanly provided the corresponding triazole analog 13 in 72% yield, and an analogous procedure using di-azide 12 (0.5 equiv with respect to alkyne 11d) furnished the PEG3-tethered 0118 dimer 14, which was obtained in 47% yield after direct purifi cation of the reaction mixture by preparative HPLC (Scheme 3). This result is in agree- ment with an earlier report employing Sharpless’ conditions [30]
with a lower-rim monopropargyl calix[4]arene [31], while diffi – culties were encountered when using these conditions for lower- rim tetrapropargyl calix[4]arenes [29].
2.2.Cytotoxicity studies (cell viability studies)
Functional efficacy of the new analogs in terms of cytotoxicity and inhibition of cell proliferation was assessed by employing a colorimetric cell viability assay (for details, see Section 4.2). Dose- response curves obtained after incubation of human MA148 ovarian carcinoma cells and HUVEC with increasing concentrations of the analogs are expressed as the percentage of cell survival
Scheme 3. CuAAC-reaction for preparation of reference compound 13 and calixarene 0118 dimer 14.
Fig. 2. Effect of new compound 0118 analogs on cell viability of MA148 human ovarian carcinoma cells (A, B) and HUVEC (C, D) as assessed by a colorimetric assay (see Section 4.2 in the experimental section). Data are reported as mean ± SEM relative to untreated control (100%).
Table 1
IC50 values (mM) of new 0118 analogs on cell viability.
0118 11a 11b 11c 11d 13 14 4 6
HUVEC 30 25 20 15 30 30 15 8 30
MA148 2 1.5 0.2 1 0.2 0.2 0.8 2.5 0.5
growth were comparable to parent compound 0118, while both analogs showed improved potency towards MA148 cell growth. Furthermore, for both cell lines, the PEG3-linked calixarene dimer 14 only results in slightly more than twice the inhibitory effect of the monomeric 0118.
2.3.Radiochemistry
relative to control with vehicle only (Fig. 2). The inhibitory potency
of the analogs may be compared based on their respective IC50
Radiolabeled compound 0118 derivatives [
18
F]6 and [18F]13
values extrapolated from these curves, which are summarized in Table 1. Results indicate that while all of the new 0118 calixarene analogs are at least equipotent to parent 0118, some of them do inhibit HUVEC and MA148 cell survival almost 4- and 10-fold more potently. All of the equatorially mono-alkylated calixarenes decrease cell viability more effectively than does the parent com- pound, with potency in the HUVEC-series increasing with increasing alkyl chain length from the methyl-substituted deriva- tive 11a to the pentyl-substituted derivative 11c. Calixarenes 6 and 13, the cold analogs of radiotracers [18F]6 and [18F]13, were found to share a similar inhibition pattern, i.e., IC50-values for HUVEC cell
(Fig. 1) were prepared in a two-step reaction sequence (Scheme 4) involving radiosynthesis of the intermediate 2-[18F]fl uo- roethylazide [18F]5 followed by CuAAC click coupling to the corre- sponding alkyne-modifi ed calix[4]arenes 4 (Scheme 1) and 11d (Schemes 2 and 3), respectively. The labeling agent 2-[18F]fl uo- roethylazide was prepared by nucleophilic fl uorination of 2- azidoethyl-4-toluenesulfonate (15) [32] in an automated system using a slightly modifi ed procedure based on previously published methods [25,33]. After purifi cation by co-distillation with aceto- nitrile at 90 ti C under a fl ow of argon into a receiver vial pre-cooled to ti 40 ti C, 2-[18F]fl uoroethylazide [18F]5 was obtained in 53 ± 5%
Scheme 4. Radiosynthesis of F-18 labeled calixarenes [18F]6 and [18F]13, and chemical structure of ligands employed in the CuAAC-reaction.
(n ¼ 10) decay-corrected isolated radiochemical yield starting from [18F]fl uoride in a total synthesis time of 60 min.
Conditions for the click reaction with propargyl-calix[4]arene 11d were optimized in terms of reaction time, temperature, amount of CuSO4/Na-ascorbate, and addition of the Cu(I)- stabilizing ligands BPDS, TBTA, and Monophos (Table 2). Initial test reactions based on the procedure of Kobus et al. [33] using 0.5 equiv. of CuSO4 and 5 equiv. of Na-ascorbate with respect to alkyne 11d showed maximum conversions after 15 min of less than 10% at 23 ti C and about 40% at 80 ti C (Table 2, entries 1 and 9). Intrigued by the dramatic rate accelerations reported for CuAAC- reactions in the presence of BPDS [34e36], TBTA [34,36], and
MonoPhos [37], we set out to investigate the effect of these ligands (Scheme 4) on the click reaction of [18F]5 with propargyl-calix[4]
arene 11d. Using conditions identical to the initial test reactions without ligand (Table 2, entries 1 and 9), at 23 ti C, only addition of TBTA led to a noteworthy increase in reaction rate (Table 2, entry 4 versus entry 1), while at 80 ti C both TBTA and BPDS resulted in similar improvement of analytical yield from about 40% without ligand to >60% with addition of ligand after 15 min (Table 2, entries 12 and 10 versus entry 9). Still, for 18F-radiolabeling using click chemistry, preferably >90% analytical yields should be obtained within reaction times up to 15 min. As an alternative option, we therefore investigated whether higher rate accelerations could be obtained by increasing the copper concentration. Indeed, using a 3- fold molar excess of CuSO4 with respect to alkyne precursor 11d, almost 5 times higher analytical yields were obtained at room
Table 2
Optimization of CuAAC conditions e effect of temperature, ligands, catalyst load, and reaction time.a
temperature (Table 2, entry 6 versus entry 1), and essentially quantitative product formation was observed at 80 ti C, both with alkyne precursor 11d (Table 2, entry 14 and 15) and with alkyne
Entry T CuSO4/ Ligand (equiv.) Buffer Analytical yield (%) at: precursor 4 (Table 2, entry 16).
(ti C)
Na-asc. (equiv.)
(pH)
15 min
30 min
Interestingly, when using an aqueous solution of alkyne pre- cursor 11d instead of the DMSO stock solution, the room temper-
123 0.5/5 e
223 0.5/5 BPDS (0.55)
Phosphate (6.0) Phosphate (6.0)
8.3
7.7
16.1
10.7
ature reaction apparently did not proceed at all (Table 2, entry 7). Although water is generally known to accelerate the CuAAC reac- tion [30,37,38], in this particular case, DMSO apparently plays an
323 0.5/5 MonoPhos (0.55) Phosphate (6.0)
0.2
0.4
important role as a co-solvent. While large rate accelerations in the presence of BPDS have also been reported at pH 5 [35], in a sys-
423 0.5/5 TBTA (0.55) Phosphate 19.4 35.2 ± 2.4b tematic study comparing performance of various CuAAC-ligands at
523 0.5/5 BPDS (0.55)
(6.0) Tris/HCl (8.5)
26.4 ± 3.2b 43.0 ± 0.3b
different pH-values, Lewis et al. [36] have observed fastest rates with BPDS at pH 8.5. Intrigued by their fi ndings, we investigated
623 3/30 e
Phosphate (6.0)
38.4
67.5
whether quantitative yields could be obtained for our system using BPDS in the same buffer (Tris/HCl) at pH 8.5. Albeit at room tem-
7f 23 3/30 e
Phosphate (6.0)
1.1± 0.4b 1.8 ± 0.2b
perature, yields for the reaction with 0.5 equiv. CuSO4 triplicated when employing Tris/HCl-buffer pH 8.5 instead of phosphate buffer
823 3/30 BPDS (3.30)
Tris/HCl (8.5)
98.7
98.5
pH 6.0 (Table 2, entry 5 versus entry 2), just the opposite trend was observed at 80 ti C, where yields dropped down to about 10%, far
980 0.5/5 e
1080 0.5/5 BPDS (0.55)
Phosphate (6.0) Phosphate (6.0)
42.2 ± 2.8c 52.1 ± 8.5c
63.7 ± 1.3d 71.8 ± 7.6d
below the 64% yield obtained with phosphate buffer (Table 2, entry 13 versus entry 10), presumably due to competing side reactions with Tris base at the higher temperature. Although quantitative
1180 0.5/5 MonoPhos (0.55) Phosphate (6.0)
45.2 ± 9.0d 73.2 ± 10.3d
yields were obtained with the Tris-based buffer system at ambient temperature when increasing the CuSO4/alkyne and BPDS/alkyne
1280 0.5/5 TBTA (0.55) Phosphate 65.4 78.9 molar ratio to 3 and 3.3, respectively (Table 2, entry 8), analytical
1380 0.5/5 BPDS (0.55)
1480 3/30 e
15g 80 3/30 e
16g,h 80 3/30 e
(6.0) Tris/HCl (8.5)
Phosphate (6.0)
Phosphate (6.0) Phosphate (6.0)
9.6 ± 1.5b 12.2 ± 2.0b
98.9 ± 0.2b e
88.1 ± 10.2c e 87.5 ± 11.6e e
HPLC-data of the crude reaction mixtures showed a cleaner reac- tion profi le of the ligand-free reaction at 80 ti C (Table 2, entry 14) compared to the ligand supported reaction at room temperature. We therefore decided to employ the ligand-free reaction for routine production of the tracers for animal studies on our automated system.
In the semi-automated procedure, 2-[18F]fl uoroethylazide
18
[ F]5 was directly distilled into the pre-cooled receiver vial
aUnless noted otherwise, all reactions were carried out by adding 2-[18F]fl uo- roethylazide [18F]5 in acetonitrile (150 mL) to a solution of 2-propargyl-calix[4]arene 11d (1.5 mmol) and the indicated ligand in either DMSO or DMSO-H2O, followed by addition of an aqueous CuSO4/Na-ascorbate solution, which was prepared imme- diately before use by pre-mixing the appropriate stock solutions of CuSO4 in H2O and Na-ascorbate in either 0.5 M sodium phosphate buffer pH 6.0 or 0.5 M Tris/HCl- buffer pH 8.5. Total volume of DMSO in the reaction mixtures was 250 mL, total volume of water 125 mL, and buffer concentration in the fi nal mixture was ca. 47 mM. Reaction mixtures were stirred with a magnetic stirring bar.
bn ¼ 2.
cn ¼ 4.
dn ¼ 3.
en ¼ 5.
fReaction mixture contained water (250 mL) instead of DMSO (250 mL).
gResults obtained using the semi-automated radiosynthesis procedure for routine production of the tracers (3 mmol of 4 or 11d in 500 mL of DMSO, 250 mL of buffered aqueous CuSO4/Na-ascorbate solution, [18F]5 in acetonitrile, ca. 300e400 mL, directly distilled onto the precursor solution in DMSO).
hReaction with 28-(30 -butynyloxy)calix[4]arene 4 instead of 11d.
already containing the DMSO-solution of alkyne precursors 11d and 4, respectively. As soon as distillation was complete, the receiver vial was quickly warmed to room temperature, and the freshly made aqueous CuSO4/Na-ascorbate solution was manu- ally added immediately after its preparation. Following reaction for 15 min at 80 ti C while stirring with a magnetic stirring bar, the mixture was diluted with water acidifi ed with 0.1% TFA, and injected into the semi-preparative HPLC system. In a number of experiments, radiochemical conversion was determined by analytical HPLC of the diluted crude reaction mixture. Different from the quantitative conversions observed in the test reactions (Table 2, entry 14), yields were sometimes lower and more var- iable (Table 2, entry 15), which is possibly related to differences in the acetonitrile concentration of the click reaction mixture arising from subtle inter-experimental variations in the distilla- tion step, confi rming earlier reports on the negative effect of
higher acetonitrile concentrations on radiochemical conversions in the CuAAC reaction [35]. Finally, the product fraction from the semi-preparative HPLC-run was collected, trapped on a C8- SepPak, and formulated in ethanol to obtain radiotracers [18F]6 and [18F]13 in overall decay-corrected isolated radiochemical yields of 18.7 ± 2.7% (n ¼ 4) and 10.2 ± 5.0% (n ¼ 4), respectively. Radiochemical purity of both tracers was always >97% and spe- cifi c activity generally >5 GBq/mmol. Typically, 400e700 MBq of formulated radiotracers were obtained from 6 to 8 GBq of [18F]Fti eluted from the QMA-cartridge in a total synthesis time of less than 2 h.
2.4.Distribution coefficient and radiotracer stability
Log D-values in 1-octanol/PBS pH 7.4 as determined by the shake flask method were þ0.8 and ti0.1 for the lower-rim and upper-rim [18F]fl uoroethyltriazole-substituted compound 0118
mechanism supported by the predominant and fast hepatobiliary clearance of [18F]6 (Fig. 4A). Interestingly, biodistribution data (3 h post injection) revealed that the higher dose of cold compound administered for the blood kinetics study (1 mg versus 0.1 mg), resulted in a 25% reduced liver uptake of [18F]13, while at the same time kidney uptake was doubled (Supplementary Table 1).
To assess whether higher tumor uptake and tumor-to-organ ratios could be obtained at earlier time points, we conducted dy- namic PET-CT studies. PET-images were reconstructed starting from 5 min post tracer injection in timeframes of 5 ti 1 min, 4 ti 5 min, 3 ti 10 min, and 3 ti 20 min. Following co-registration of PET- and CT-scans and identification of VOIs in tumor and muscle based on manual segmentation of the CT-images, tracer levels in both tissues were quantifi ed for the different timeframes. Calculated tumor-to- muscle ratios based on these values were below 1 for [18F]6 and between 1.5 and 2 in case of [18F]13 for all timeframes investigated. Notably, the more rapid hepatobiliary clearance of [18F]6 compared
analogs [
18
F]6 and [18F]13, respectively. Evidently, at pH 7.4, the
to [18F]13 noted in the biodistribution studies was confi rmed in the
presence of an additional lower-rim substituent containing a pro- tonatable trialkylamino-group results in an approximate 10-fold reduced lipophilicity of [18F]13 compared to [18F]6.
Both radiotracers proved highly stable in PBS pH 7.4 and mouse serum at 37 ti C. After incubation of either [18F]6 or [18F]13 for 3 h, more than 98% of intact tracer was found both in PBS and mouse serum. Tracer binding to serum proteins after acetonitrile precipi- tation was negligible, as over 94% of total radioactivity was recov- ered in the supernatant of the mouse serum samples after centrifugation.
2.5.Biodistribution and dynamic PET studies
For an initial assessment of tumor targeting potential and their pharmacokinetic profi le, preliminary in vivo studies with radio-
18
labeled 0118 analogs [ F]6 and [18F]13 were conducted in mice bearing galectin-1 expressing MDA-MB-231-LITG tumors. Bio- distribution results obtained 3 h after intravenous injection of [18F]
6 (0.2e0.3 MBq, corresponding to about 60 ng (63 pmol) of 6) and [18F]13 (0.3e0.4 MBq, corresponding to about 100 ng (94 pmol) of 13) indicate highest uptake of both tracers in the liver, followed by spleen, kidneys, lung, and bone, with only low retention in other organs, including the tumor (Fig. 3A). Evidently, the hepatobiliary route plays an important role for the clearance of both tracers, with the higher accumulation of [18F]6 in the intestine suggesting faster hepatobiliary excretion of this radiotracer compared to [18F]13 (Fig. 3B and Supplementary Table 2). Conversely, as a result of the slower excretion of [18F]13 from the liver and the faster clearance from blood (after 3 h, blood levels of [18F]6 were 0.76 ± 0.09 %ID/g versus 0.05 ± 0.01 %ID/g for [18F]13), radioactivity levels in the liver are almost threefold higher for [18F]13 compared to [18F]6 (Fig. 3A). Further differences between the tracers include the almost twofold
dynamic PET-scans, which showed immediate uptake in the liver, followed by redistribution to the gallbladder and substantial accumulation of [18F]6 in the intestines already after 30 min (Fig. 4A). In contrast, intestinal uptake of the more hydrophilic [18F]
13 was slower, while at the same time excretion via the bladder was more pronounced (Fig. 4B).
3.Conclusion
We have designed two close mimetics of galectin-1 targeting antitumor agent calixarene 0118 amenable to radiolabeling with 2- [18F]fluoroethylazide via straightforward click chemistry. Following optimization of reaction conditions and translation to a semi- automated procedure, the lower-rim radiolabeled [18F]6 and the equatorially modifi ed [18F]13 were obtained in 18.7 ± 2.7% (n ¼ 4)
and 10.2 ± 5.0% (n ¼ 4) decay-corrected isolated radiochemical yield and >97% radiochemical purity after semi-preparative HPLC purifi cation and formulation. In vitro stability studies in mouse serum indicated only negligible binding to serum proteins and excellent stability. Preliminary in vivo studies in mice bearing galectin-1 expressing MDA-tumor xenografts showed highest up- take of both tracers in the liver, followed by spleen, kidneys, lung and bone, with only low retention in other organs, including the tumor. Nevertheless, these new radiotracers may become a valu- able tool for pharmacokinetic profi ling and further development of this class of calixarene-based therapeutics in the future.
Building on the synthetic methodology developed for prepara- tion of the cold reference compounds of the radiotracers (6, 13) and their precursors (4, 11d), we also created a small series of 0118 analogs (11aec) containing a single alkyl substituent at an equa- torial position on one of the methylene groups, and a PEG3-spaced calixarene dimer (14). Characterization of these eight 0118 de-
18
higher uptake of [
F]6 in the lung, while [18F]13 leads to about
rivatives in terms of cytotoxicity and anti-proliferative effects on
twofold higher bone uptake, suggesting faster radiodefl uorination of the tracer in vivo (Fig. 3A). Blood clearance of [18F]6 (~3.5 MBq, ca. 1 mg of 6) and [18F]13 (~8.5 MBq, ca. 1 mg of 13) followed bi-phasic exponential decay kinetics, with short initial half-lives (t1/2,a) of 1.2 min and 1.6 min, and a subsequent slower elimination phase with t1/2,b of 74 min and 12 min, respectively (Fig. 3C and Supplementary Table 3). For both tracers, the major part was cleared from the blood by distribution to tissues and extravascular spaces in the a-phase (63% for [18F]6 versus 79% for [18F]13). The slower blood clearance of [18F]6 in the elimination phase is paral- leled by the higher lipophilicity of this tracer, presumably giving rise to increased plasma protein binding [39]. Alternatively, the slower clearance may partly be due to enterohepatic recycling, i.e., intestinal reabsorption of the tracer after biliary excretion [40], a
HUVEC cells and MA148 human ovarian carcinoma cells indicated that all analogs were at least equipotent to lead compound 0118, while some of them were almost 4- to 10-times more potent in- hibitors of HUVEC- and MA148 cell growth, respectively. As such, the new analogs may hold potential as a new generation anti- angiogenic agents in cancer therapy with improved potency.
4.Experimental
4.1.Chemistry
4.1.1.General methods and materials
All reagents and solvents were obtained from commercial sources (SigmaeAldrich, Acros, Biosolve, Merck, and Cambridge
Fig. 3. Ex vivo biodistribution and blood clearance: Uptake of radiotracers [18F]6 (0.2e0.3 MBq, ca. 60 ng of 6) and [18F]13 (0.3e0.4 MBq, ca. 100 ng of 13) in selected organs and tissues of mice bearing MDA-MB-231-LITG tumor xenografts 3 h after intravenous injection, expressed as percentage of injected dose per gram of tissue (% ID/g) (A), and percentage of injected dose per organ (%ID/organ) (B); stomach, small and large intestines were measured including content. Blood clearance of [18F]6 (~3.5 MBq, ca. 1 mg) and [18F]13 (~8.5 MBq, ca. 1 mg) fitted to a two-phase exponential decay function (C). Data are mean ± SD (n ¼ 3; *P < 0.05; **P < 0.005; ***P < 0.0005).
Isotope Laboratories for deuterated solvents) and were used without further purification unless stated otherwise. 25,26,27,28- Tetrahydroxy-calix[4]arene 1 was from Carbosynth Limited (UK), 1,11-diazido-3,6,9-trioxaundecane 12 from Toronto Research
Fig. 4. Dynamic PET-scans: Whole-body PET/CT-images of mice bearing MDA-MB- 231-LITG tumor xenografts in the left hind leg. Mice were injected with 12e13 MBq of [18F]6 (A) or [18F]13 (B), and list-mode PET data was reconstructed for the indicated timeframes. PET-data in the images are maximum intensity projections.
Chemicals Inc. (Canada), and 2-fl uoroethyl 4- methylbenzenesulfonate (precursor of 5) was from Molekula (UK). Unless stated otherwise, all reactions were performed in dried glassware under nitrogen atmosphere. NMR spectra were recorded on a Bruker DPX300, a Bruker Avance600, a Varian VNMRS spectrometer, or a Varian MP300 spectrometer. Chemical shift values are reported in d (ppm) relative to tetramethylsilane as
the internal standard (TMS: d ¼ 0 ppm for 1H and d ¼ 0 ppm for 13C), or referenced to the residual solvent peaks (CDCl3: d 7.26 for 1H and d 77.0 for 13C). 1H NMR multiplicities are abbreviated as follows: s ¼ singlet, d ¼ doublet, t ¼ triplet, q ¼ quartet, dd ¼ doublet of doublets, dt ¼ doublet of triplets, bs ¼ broad singlet. 13C NMR multiplicities (q ¼ quaternary, t ¼ tertiary, s ¼ secondary, and p ¼ primary) were distinguished using a DEPT pulse sequence or an attached proton test (APT). Preparative column chromatography was performed on a Combifl ash Companion apparatus (Teledyne Isco) employing pre-packed silica cartridges from Grace (USA). High-resolution ESI mass spectra (HRMS) were recorded on an Agilent ESI-TOF mass spectrometer.
Preparative HPLC was performed on two different systems. System A: Agilent 1200 apparatus, equipped with a C18 Zorbax column (21.2 ti 150 mm, 5 mm) applying a linear gradient of acetonitrile (B) in water (A), both containing 0.1% TFA. Gradient details: 30% B (0 min / 5 min), 30% B / 50% B (5 min / 12 min), 50% B (12 min / 15 min), 50% B / 95% B (15 min / 16 min), 95% B (16 min / 18 min), 95% B / 30% B (18 min / 19 min), 30% B
(19 min / 22 min). Flow: 10 mL/min. Injection volume: 0.5 mL. UV detection: 215 nm, 254 nm. System B: Instrument: Agilent 1100 series, equipped with a Gemini NX C18 100A Axia column (30 ti 100 mm, 5 mm) employing a linear gradient of acetonitrile (B) in aqueous 20 mM ammoniumbicarbonate (A). Gradient details: 50%B (0 min / 3 min), 50% B / 95% B (3 min / 9 min), 95% B
(9min / 10 min). Flow: 40 mL/min. Injection volume: 25 mL. UV detection: 215 nm, 254 nm.
Analytical HPLCeMS was carried out on System C: Agilent 1100 series with UV detector and HP 1100 mass TOF detector, equipped with a Kinetex C18 (50 ti 2.10 mm; 2.6 mm) column, variable wavelength UV-detector, and API ES TOF positive and negative mass detection. Column temperature: 35 ti C. Flow: 0.60 mL/min. Injection volume: 1 mL. Mobile phase: 9.65 g ammonium acetate, 2250 mL H2O, 150 mL methanol, 100 mL acetonitrile (eluent A); 9.65 g ammonium acetate, 250 mL H2O, 1350 mL methanol, 900 mL acetonitrile (eluent B).
4.1.2.25,26,27-Trihydroxy-28-(30 -butynyloxy)calix[4]arene (2b)
To 25,26,27,28-tetrahydroxycalix[4]-arene 1 (4 g, 9.42 mmol) in acetonitrile (220 mL) was added freshly prepared NaOMe (600 mg, 11.12 mmol). The mixture was refl uxed for 30 min, cooled and 4- iodobut-1-yne (3.29 g, 24 mmol) in acetonitrile (40 mL) was added (4-iodobut-1-yne was freshly prepared from NaI (7.2 g, 48 mmol) and 4-bromobut-1-yne (3.2 g, 24 mmol) in acetonitrile (40 mL) at refl ux for 1 h). The mixture was refl uxed overnight and the conversion was checked by 1H NMR (24%). Additional NaOMe (400 mg, 7.41 mmol) was added and the mixture was refluxed over the weekend (conversion 32%). Additional NaOMe (400 mg, 7.41 mmol) was added and the mixture was stirred at refl ux for another 48 h. NMR analysis showed a conversion of 39%. Then additional freshly prepared NaOMe (400 mg, 7.41 mmol) was added and the mixture was refluxed for another 48 h. Conversion (49%) and also 5% bis-alkylated material was formed. Then the mixture was worked up by evaporation of the solvent. Dichloromethane (100 mL) was added to the residue and the mixture was washed with water (3 ti 50 mL). After evaporation of the organic solvents an off-white solid was isolated. The title compound was isolated as a mixture with unreacted 25,26,27,28-tetrahydroxycalix[4]arene. The material was stirred in ethyl acetate (15 mL) and the solid was fi ltered. The mother liquor was evaporated and 1.5 g of crude product was isolated. According to 1H NMR this sample contains approx. 75% of the title compound (corresponding to 1.16 g, 2.43 mmol, 26% yield of 2b) and 25% of starting material. The solid was used in the next step without further purification. 1H NMR (300 MHz, CDCl3) d 9.7 (s, 1H), 9.17 (s, 2H), 7.09 (m, 8H), 6.90 (m, 1H), 6.69 (m, 3H), 4.43 (d, 2H, 2J ¼ 13.0 Hz), 4.29 (m, 4H), 3.49 (d, 4H, 2J ¼ 12.9 Hz), 3.06 (dt, 2H, 4J ¼ 2.7 Hz, 3J ¼ 6.6 Hz), 2.23 (t, 1H, 4J ¼ 2.7 Hz). 13C NMR (75 MHz, CDCl3) d 151.32 (q),151.05 (q),149.45 (q), 134.34 (q), 129.69 (q), 129.23 (q), 129.11 (q), 129.02 (q), 128.66 (q), 128.59 (t), 128.49 (t), 126.58 (q), 122.20 (t) 121.15 (t), 80.42 (q), 74.60 (t), 71.28 (s), 32.15 (s), 31.66 (s), 20.35 (s).
4.1.3.25,26,27-Tri[(ethoxycarbonyl)methoxy]-28-(30 -butynyloxy) calix[4]arene (3)
To a solution of 2b (1.5 g with purity 75%, corresponding to 2.43 mmol of 2b) in acetonitrile (20 mL) was added K2CO3 (964 mg, 6.98 mmol). The mixture was stirred for 30 min and then an excess of ethyl bromoacetate (2.63 g, 15.75 mmol) was added. The mixture was heated to 70 ti C for 96 h. After cooling, the acetonitrile was evaporated. The residue was taken up in dichloromethane (100 mL). The organic layer was washed with water (2 ti 50 mL). After separation the organic layer was evaporated. The impure compound was purifi ed by column chromatography (silicagel, dichloromethane). Several impure fractions were isolated. The fi rst
combined fractions (900 mg) contained product and by-products. The second batch of combined fractions was enriched in product (1 g) and the third batch (500 mg) was a combination of fractions which contained mainly tetraethyl ester. The fi rst batch was puri- fied by column chromatography again on silicagel with dichloro- methane to remove by-products and then ethyl acetate:heptanes ¼ 1:2 to elute the product. Approximately 450 mg of a mixture was isolated which was enriched in product. This material was combined with the 1 g batch. A total of 1.45 g of material was purified again by silica column chromatography using a mixture of ethyl acetate:heptanes ¼ 6:1 followed by ethyl acetate:heptanes ¼ 4:1 to obtain the title compound (1.2 g, 1.63 mmol, 67%). Unfortunately, 1H NMR analysis still indicated presence of some impurities. A 300 mg batch was used in the next step without further purification. The remainder (900 mg) was again purified by repeated silica column chromatography, fi rst us- ing dichloromethane followed by ethyl acetate:heptanes ¼ 1:4, then toluene:ethyl acetate ¼ 95:5 to obtain the pure title com- pound (30 mg, 0.04 mmol) and a slightly less pure fraction (260 mg, 0.35 mmol). 1H NMR (300 MHz, CDCl3) d 6.79e6.68 (m, 6H), 6.57e6.53 (m, 6H), 4.85 (d, 2H, 2J ¼ 13.6 Hz), 4.80e4.56 (m, 8H),
4.28 (q, 4H, 3J ¼ 7.2 Hz), 4.24 (q, 2H, 3J ¼ 7.2 Hz), 4.10 (t, 2H,
J ¼ 7.6 Hz), 3.28 (d, 2H, 3J ¼ 13.3 Hz), 3.22 (d, 2H, 3J ¼ 13.3 Hz), 2.92 (dt, 2H, 4J ¼ 2.6 Hz, 3J ¼ 7.6 Hz), 1.97 (t, 1H, 4J ¼ 2.6 Hz), 1.27e1.36 (m, 9H). 13C NMR (75 MHz, CDCl3) d 170.41 (q),170.11 (q),156.59 (q), 156.34 (q), 155.53 (q), 135.62 (q), 135.10 (q), 134.61 (q), 134.25 (q), 128.93 (t), 128.77 (t), 128.68 (t), 128.57 (t), 123.07 (t), 122.93 (t), 82.19 (q), 72.74 (s), 71.79 (s), 71.70 (s), 69.36 (t), 60.96 (s), 60.79 (s), 31.48 (s), 31.29 (s), 20.03 (s), 14.50 (p), 14.44 (p). Analytical HPLCeMS (system C), gradient: 20% B / 95% B (0 min / 1.5 min), 95% B (1.5 min / 4.0 min). Retention time: 3.65 min. Purity
>99.99% (UV, 215, 254 nm). MS calculated for C44H46O10: 734.31; MS (API-ES TOF Pos): m/z 752 [MþNH4]þ, 757 [MþNa]þ.
4.1.4.25,26,27-Tris-N-(N,N-dimethyl-2-aminoethyl) carbamoylmethoxy-28-(30 -butynyloxy)calix[4]arene (4)
To powdered 3 (300 mg, 0.41 mmol) was added under nitrogen N,N-dimethylethylenediamine (5 mL). The mixture was stirred at room temperature for 1 h and was then stirred for 48 h at 50 ti C. The excess of N,N-dimethylethylenediamine was removed by evapora- tion under reduced pressure. A sample of 220 mg was dissolved in tetrahydrofuran (unstabilized, concentration of crude 70 mg/mL) and purified by preparative HPLC on system B (see general
methods). Fractions containing the product (tR ¼ 6.4 min; broad peak) were pooled and evaporated at the rotary evaporator to remove the acetonitrile. The water was removed by freeze-drying to obtain the pure target compound as an off-white foam (110 mg, 128 mmol, c.y. 38%).
1H NMR (300 MHz, CDCl3) d 6.74e6.67 (m, 6H), 6.60e6.47 (m, 6H), 4.57 (d, 2H, 2J ¼ 13.9 Hz), 4.56e4.33 (m, 8H), 4.14 (t, 2H, 3
J ¼ 7.3 Hz), 3.56e3.40 (m, 6H), 3.28 (d, 2H, 2J ¼ 14.2 Hz), 3.26 (d, 2H, 2J ¼ 13.8 Hz), 2.75 (dt, 2H, 4J ¼ 2.7 Hz, 3J ¼ 7.3 Hz), 2.53 (t, 4H, 3
J ¼ 6.6 Hz), 2.45 (t, 2H, 3J ¼ 6.5 Hz), 2.26 (s, 12H), 2.20 (s, 6H), 2.09 (t,1H, 4J ¼ 2.6 Hz). 13C NMR (75 MHz, CDCl3) d 169.58 (q),169.46 (q), 156.11 (q), 155.60 (q), 155.47 (q), 135.21 (q), 134.92 (q), 133.94 (q), 133.80 (q), 129.18 (t), 129.00 (t), 128.85 (t), 128.71 (t), 123.29 (t), 123.22 (t), 81.81 (q), 74.31 (s), 74.15 (s), 72.61 (s), 70.47 (t), 58.31 (s), 58.18 (s), 57.98 (s), 45.51 (p), 45.44 (p), 45.29 (p), 37.26 (s), 37.17 (s), 35.65 (s), 31.32 (s), 31.07 (s), 19.94 (s). Analytical HPLCeMS (system C), gradient: 20% B / 90% B (0 min / 1.0 min), 90% B / 100% B (1.0 min / 3.5 min), 100% B (3.5 min / 4.0 min). Retention time: 2.03 min. Purity 99.64% (UV, 218 nm). MS calculated for C50H64N6O7: 860.48; MS (API-ES TOF Pos): m/z 431 [Mþ2H]2þ, 861 [MþH]þ, 883 [MþNa]þ; MS (API-ES TOF Neg): m/z 859 [Mti H]ti, 919 [MtiCH3COO]ti .
4.1.5.25,26,27-Tris-N-(N,N-dimethyl-2-aminoethyl) carbamoylmethoxy-28-{20 -[1-(2-fluorethyl)-1H-[1,2,3]triazo-4-yl]
ethyloxy}calix[4]arene (6)
To a stirred solution of CuSO4 ti 5H2O (21.3 mg, 86 mmol) and L- ascorbic acid (30.3 mg, 172 mmol) in water (1 mL) under an N2 at- mosphere was added a solution of crude 4 (50 mg, 57 mmol) in N,N- dimethylformamide (0.7 mL). After addition of 2-fluoroethylazide 5 (5.09 mg, 57 mmol in 1 mL of N,N-dimethylformamide; solution freshly prepared following published procedures [25]), the mixture was stirred at room temperature overnight, and then evaporated to dryness. The solid residue was treated with dichloromethane (5 mL) and water (5 mL). A few drops of NaOH (1 M) were added. The layers were separated and the aqueous layer was again extracted with dichloromethane (5 mL). The combined organic layers were washed with water (5 mL) and evaporated in vacuo. The crude material (40 mg) was taken up in tetrahydrofuran (unstabi- lized, 60 mg/mL, injection volume 25 mL) and purifi ed by prepar- ative HPLC on system B (see general methods). Fractions containing
the product (tR ¼ 6.8 min; broad peak) were pooled and evaporated at the rotary evaporator to remove the acetonitrile. The water was removed by freeze-drying. The target compound was isolated as a white solid 15 mg (15.7 mmol, c.y. 27.7%). A larger batch (50 mg) has been synthesized in the same way. 1H NMR (300 MHz, CDCl3) d 7.96 (m, 1H), 7.64 (m, 2H), 7.60 (s, 1H), 6.67 (s, 6H), 6.59e6.41 (m, 6H),
4.79 (dt, 2JHF ¼ 46.9 Hz, 3JHH ¼ 4.6 Hz, 2H), 4.74e4.48 (m, 8H), 4.27 (d, 2J ¼ 14.2 Hz, 4H), 4.26e4.18 (m, 2H), 3.55e3.33 (m, 6H), 3.28 (t, 3
J ¼ 6.9 Hz, 2H), 3.23 (d, 2J ¼ 13.9 Hz, 4H), 2.51e2.40 (m, 6H), 2.22 (s, 6H), 2.20 (s, 12H). 13C NMR (75 MHz, CDCl3) d 169.55 (q), 169.16 (q), 156.33 (q), 145.02 (q), 134.52 (q), 134.27 (q), 134.17 (q), 134.08 (q), 129.30 (t), 128.79 (t) 128.76 (t), 128.49 (t), 123.23 (t), 123.03 (t),
122.92 (t), 122.76 (t), 122.74 (t), 81.54 (s, doublet, 1JCF ¼ 171 Hz), 74.49 (s), 73.87 (s), 73.77 (s), 58.27 (s), 58.11 (s), 50.50 (s, doublet, 2
JCF ¼ 20 Hz), 45.41 (p), 37.14 (s), 31.33 (s), 30.98 (s), 26.76 (s). 19F NMR (282 MHz, CDCl3) d ti150.75 to ti151.23 (m, 1F). Analytical HPLCeMS (system C), gradient: 20% B / 95% B (0 min / 1.5 min), 95% B (1.5 min / 4.0 min). Retention time: 2.41 min. Purity 99.82% (UV, 224 nm). MS calculated for C52H68FN9O7: 949.52; MS (API-ES TOF Pos): m/z 475 [Mþ2H]2þ, 950 [MþH]þ.
4.1.6.25,26,27,28-Tetramethoxycalix[4]arene (7)
Compound 7 was prepared as described [41], with minor modifi cations. Briefl y, 25,26,27,28-Tetrahydroxycalix[4]-arene 1 (7.5 g, 17.7 mmol) was dissolved in a 10:1 mixture of anhydrous tetrahydrofuran:N,N-dimethylformamide (165 mL). NaH (12 g of a 60% dispersion in mineral oil, 300 mmol) was triturated with n- pentane to remove the mineral oil. After drying under a gentle stream of nitrogen, the NaH was added to the precursor solution in tetrahydrofuran/N,N-dimethylformamide, followed by MeI (33 mL, 531 mmol). The reaction mixture was refl uxed for 2 h, treated with methanol (10 mL) to decompose the excess of NaH, and evaporated in vacuo. The solid was partitioned between water (300 mL) and dichloromethane (300 mL). The layers were separated and the water layer was again extracted with dichloromethane (300 mL). The combined organic layers were back-extracted with water (150 mL), dried on MgSO4, and evaporated in vacuo. Traces of MeI and N,N-dimethylformamide were effi ciently removed by coeva- poration with dichloromethane. Yield: 8.3 g (17.3 mmol, 98%) of a white to slightly yellow solid. 1H NMR (300 MHz, CD3CN sat. with NaI) d 7.34 (d, 3J ¼ 7.7 Hz, 8H), 6.92 (t, 3J ¼ 7.7 Hz, 4H), 4.28 (d, 2
J ¼ 12.5 Hz, 4H), 4.14 (s, 12H), 3.60 (d, 2J ¼ 12.5 Hz, 4H). 13C NMR (75 MHz, CD3CN sat. with NaI) d 153.80(q), 136.33(q) 130.26(t), 126.77(t), 65.58(p), 29.75(s). HRMS (ESI, m/z): Calculated for C32H32O4Hþ ([MþH]þ): 481.2379, Found: 481.2382.
4.1.7.General procedure for preparation of 25,26,27,28- tetramethoxy-2-alkylcalix[4]arenes (8aed)
25,26,27,28-Tetramethoxycalix[4]arene 7 (2.00 g, 4.16 mmol) was dissolved in anhydrous THF (100 mL). The clear yellow solution was cooled to ti20 ti C, and n-BuLi (11.7 mL of 1.6 M n-BuLi in hex- anes, 18.7 mmol) was added dropwise in the course of 30 min. The resulting blood-red solution was allowed to stir for 45 min. Alkyl iodide (37.4 mmol) was added, and the solution, which has turned orangeebrown, was allowed to stir for 1 h while warming up to room temperature. After quenching with sat. aq. KHSO4 (20 mL) and evaporation of most of the THF, water (80 mL) was added, and the product was extracted with dichloromethane (2 ti 80 mL). The combined organic extracts were washed with brine (80 mL), dried on MgSO4, and filtered over a short column of silica to remove highly polar impurities (baseline spot on TLC), followed by passing an additional 80 mL of dichloromethane, and evaporated in vacuo.
4.1.7.1.25,26,27,28-Tetramethoxy-2-methylcalix[4]arene (8a). Alkylation with methyl iodide (2.4 mL) yielded the title compound as an off-white solid (1.96 g, 3.96 mmol, 95%). 1H NMR (300 MHz, CD3CN sat. with NaI) d 7.41 (dd, 3J ¼ 7.9 Hz, 4J ¼ 1.5 Hz, 2H), 7.34 (d, 3
3J ¼ 7.7 Hz, 4H), 7.31 (dd, 3J ¼ 7.7 Hz, 4J ¼ 1.5 Hz, 2H), 6.96 (t, J ¼ 7.7 Hz, 2H), 6.92 (t, 3J ¼ 7.7 Hz, 2H), 4.88 (q, 3J ¼ 7.5 Hz,1H), 4.28
(d, 2J ¼ 12.5 Hz, 2H), 4.27 (d, 2J ¼ 12.5 Hz, 2H), 4.14 (s, 6H), 4.12 (s, 6H), 3.60 (d, 2J ¼ 12.5 Hz, 3H),1.71 (d, 3J ¼ 7.5 Hz, 3H). HRMS (ESI, m/
z): Calculated for C33H34O4Hþ ([MþH]þ): 495.2530, Found: 495.2553.
4.1.7.2.25,26,27,28-Tetramethoxy-2-propylcalix[4]arene (8b). Alkylation with 1-iodopropane (3.7 mL) yielded the title compound as pale yellow solid (1.82 g, 3.48 mmol, 84%). 1H NMR (300 MHz, CD3CN sat. with NaI) d 7.40e7.28 (m, 8H), 6.99e6.88 (m, 4H), 4.69 (t, 3J ¼ 8.1 Hz,1H), 4.27 (d, 3J ¼ 12.5 Hz, 3H), 4.14 (s, 6H), 4.12 (s, 6H), 3.60 (d, 2J ¼ 12.5 Hz, 2H), 3.59 (d, 2J ¼ 12.5 Hz, 1H), 2.20e2.02 (m, 2H), 1.44e1.24 (m, 2H), 0.99 (t, 3J ¼ 7.2 Hz, 3H). 13C NMR (75 MHz, CD3CN sat. with NaI) d 153.89 (q), 153.79 (q), 140.00 (q), 136.41 (q), 136.37 (q), 136.16 (q), 130.29 (t), 129.89 (t), 126.99 (t), 126.79 (t), 65.68 (p), 65.56 (p), 36.88 (s), 36.04 (t), 29.92 (s), 29.77 (s), 22.60 (s), 14.60 (p). HRMS (ESI, m/z): Calculated for C35H38O4Hþ ([MþH]þ): 523.2843, Found: 523.2818.
4.1.7.3.25,26,27,28-Tetramethoxy-2-pentylcalix[4]arene (8c). Alkylation with 1-iodopentane (4.9 mL) yielded the title compound as a white solid (1.91 g, 3.47 mmol, 83%). 1H NMR (300 MHz, CD3CN sat. with NaI) d 7.39e7.28 (m, 8H), 6.99e6.88 (m, 4H), 4.67 (t, 3
J ¼ 8.1 Hz, 1H), 4.27 (d, 2J ¼ 12.5 Hz, 3H), 4.14 (s, 6H), 4.11 (s, 6H), 3.60 (d, 2J ¼ 12.5 Hz, 2H), 3.59 (d, 2J ¼ 12.5 Hz, 1H), 2.20e2.02 (m, 2H), 1.43e1.23 (m, 6H), 0.86 (t, 3J ¼ 7.2, Hz, 3H). 13C NMR (75 MHz, CD3CN sat. with NaI) d 153.87 (q), 153.79 (q), 140.00 (q), 136.40 (q), 136.35 (q), 136.15 (q), 130.28 (t), 129.89 (t), 126.97 (t), 126.78 (t), 65.66 (p), 65.55 (p), 36.20 (t), 34.51 (s), 32.46 (s), 29.91 (s), 29.75 (s), 29.08 (s), 22.97 (s), 14.26 (p). HRMS (ESI, m/z): Calculated for C37H42O4Hþ ([MþH]þ): 551.3156, Found: 551.3176.
4.1.7.4.25,26,27,28-Tetramethoxy-2-(prop-20 -yn-10 -yl)calix[4]arene (8d). Preparation of 8d was performed using a slightly modifi ed procedure. Briefl y, the lithiated intermediate obtained by dropwise addition of n-BuLi (7.8 mL of 1.6 M n-BuLi in hexanes, 12.5 mmol) to an ice-cooled solution of tetramethoxycalix[4]arene 7 (1 g, 2.08 mmol) in anhydrous tetrahydrofuran (30 mL) was added (almost) at once to a pre-cooled solution (ice bath) of propargyl bromide (2.78 mL of an 80 wt. % solution in toluene, 25.0 mmol) in anhydrous tetrahydrofuran (15 mL). The reaction mixture was stirred for 1 h in an ice-bath, allowed to warm to room temperature (ca. 20 min), and then quenched with saturated aqueous KHSO4
(10mL) and worked up as described in the general procedure. 1H NMR of the crude product (950 mg) indicated a conversion of ca. 80% (corresponding to 771 mg, 1.49 mmol, 72% yield of pure 8d). The crude product was purifi ed by column chromatography on silicagel employing a gradient of 0e10% of ethyl acetate in hep- tanes. Fractions containing the product were pooled, evaporated in vacuo, and co-evaporated with dichloromethane to remove traces of heptanes. Yield: 402 mg (0.78 mmol, 37%) of a white solid. 1H NMR (300 MHz, CD3CN sat. with NaI) d 7.43 (dd, 3J ¼ 7.9 Hz, 4J ¼ 1.4 Hz, 2H), 7.35 (d, 3J ¼ 7.7 Hz, 6H), 6.97 (t, 3J ¼ 7.7 Hz, 2H), 6.92 (t, 3J ¼ 7.7 Hz, 2H), 4.95 (t, 3J ¼ 8.6 Hz, 1H), 4.28 (d, 2J ¼ 12.5 Hz, 2H), 4.27 (d, 2J ¼ 12.5 Hz, 1H), 4.18 (s, 6H), 4.14 (s, 6H), 3.62 (d, 2J ¼ 12.5 Hz, 2H), 3.60 (d, 2J ¼ 12.5 Hz, 1H), 3.15 (dd, 3J ¼ 8.4 Hz, 4J ¼ 2.5 Hz, 2H), 2.33 (t, 4J ¼ 2.5 Hz, 1H). 13C NMR (75 MHz, CD3CN sat. with NaI) d 153.86 (q), 153.63 (q), 138.38 (q), 136.36 (q), 136.34 (q), 136.30 (q), 130.44 (t), 130.33 (t), 130.29 (t), 127.06 (t), 126.90 (t), 126.83 (t), 83.98 (q), 71.42 (t), 65.91 (p), 65.54 (p), 35.88 (t), 29.90 (s), 29.73 (s), 23.90 (s). HRMS (ESI, m/z): Calculated for C35H34O4Hþ ([MþH]þ): 519.2535, Found: 519.2561.
4.1.8.General procedure for preparation of 25,26,27,28- tetrahydroxy-2-alkylcalix[4]arenes (9aed)
25,26,27,28-Tetramethoxy-2-alkylcalix[4]arene 8aed (0.7e2.9 mmol) was dissolved in anhydrous dichloromethane (50 mL for 8aec, 30 mL for 8d) and cooled to ti 78 ti C. A 1.0 M solution of BBr3 in dichloromethane (4.8e18.7 mL, 6.5 equiv.) was added dropwise via a syringe, and the resulting mixture was stirred for 1 h at ti 78 ti C, then allowed to warm to room temperature. After 30 min at room temperature, the reaction mixture was quenched by addition of saturated aq. NaHCO3 (50 mL). Layers were separated, and the organic layer was washed with water (50 mL). After drying on MgSO4, the ex- tracts were passed over a thin column of silica, followed by rinsing with dichloromethane (200 mL), and evaporated in vacuo. For 9d, a slightly different workup was employed (see below).
4.1.8.1.25,26,27,28-Tetrahydroxy-2-methylcalix[4]arene (9a). Demethylation of 8a (1.16 g, 2.35 mmol) with 1.0 M BBr3 in dichloromethane (15.3 mL, 15.3 mmol) yielded the title compound as a white foam (0.776 g, 1.77 mmol, 75%). 1H NMR (300 MHz,
CDCl3) d ¼ 10.15 (s, 4H), 7.18e6.98 (m, 8H), 6.77 (t, 3J ¼ 7.5 Hz, 2H), 6.72 (t, 3J ¼ 7.7 Hz, 2H), 4.74 (q, 3J ¼ 7.2 Hz, 1H), 4.26 (d, 2J ¼ 13.9 Hz,
1H), 4.25 (d, 2J ¼ 13.9 Hz, 2H), 3.53 (d, 2J ¼ 13.9 Hz, 3H), 1.71 (d, 3J ¼ 7.2 Hz, 3H). HRMS (ESI, m/z): Calculated for C29H26O4Hþ ([MþH]þ): 439.1904, Found: 439.1895.
4.1.8.2.25,26,27,28-Tetrahydroxy-2-propylcalix[4]arene (9b). Demethylation of 8b (1.50 g, 2.87 mmol) with 1.0 M BBr3 in dichloromethane (18.7 mL, 18.7 mmol) yielded the title compound as a white foam (1.28 g, 2.74 mmol, 95%). 1H NMR (300 MHz, CDCl3)
d ¼ 10.12 (s, 4H), 7.19e6.88 (m, 8H), 6.80e6.64 (m, 4H), 4.53 (t,
J ¼ 7.8 Hz, 1H), 4.26 (d, 2J ¼ 13.9 Hz, 1H), 4.25 (d, 2J ¼ 13.8 Hz, 2H), 3.53 (d, 2J ¼ 13.9 Hz, 3H), 2.22e2.05 (m, 2H),1.42e1.23 (m, 2H), 0.95
(t, 3J ¼ 7.3 Hz, 3H). HRMS (ESI, m/z): Calculated for C31H30O4Hþ ([MþH]þ): 467.2217, Found: 467.2214.
4.1.8.3.25,26,27,28-Tetrahydroxy-2-pentylcalix[4]arene (9c). Demethylation of 8c (720 mg, 1.31 mmol) with 1.0 M BBr3 in dichloromethane (8.5 mL, 8.5 mmol) yielded the title compound as a white foam (594 mg, 1.20 mmol, 92%). 1H NMR (300 MHz, CDCl3)
3d ¼ 10.11 (s, 4H), 7.15e6.95 (m, 8H), 6.80e6.65 (m, 4H), 4.49 (t,
J ¼ 7.7 Hz, 1H), 4.26 (d, 2J ¼ 13.9 Hz, 1H), 4.25 (d, 2J ¼ 13.9 Hz, 2H), 3.53 (d, 2J ¼ 13.9 Hz, 3H), 2.22e2.08 (m, 2H), 1.40e1.25 (m, 6H),
0.92e0.83 (m, 3H). HRMS (ESI, m/z): Calculated for C33H34O4Hþ ([MþH]þ): 495.2530, Found: 495.2508.
4.1.8.4.25,26,27,28-Tetrahydroxy-2-(prop-20 -yn-10 -yl)calix[4]arene (9d). Demethylation of 8d (379 mg, 0.73 mmol) in anhydrous dichloromethane (30 mL) with 1.0 M BBr3 in the same solvent (4.75 mL, 4.75 mmol), followed by a slightly modifi ed workup (partitioning between 120 mL of sat. aq. NaHCO3 and 120 mL of dichloromethane, washing of the organic layer with 120 mL of water, drying over MgSO4, and evaporation in vacuo) yielded the crude title compound in quantitative yield. After purifi cation by silica column chromatography employing a gradient of 5e30% of ethyl acetate in heptanes, the pure target compound was obtained as a white solid (36 mg, 0.078 mmol, 11%). 1H NMR (300 MHz, CDCl3) d 10.10 (s, 4H), 7.08e7.02 (m, 8H), 6.78 (t, 3J ¼ 7.7 Hz, 2H), 6.73 (t, 3J ¼ 7.7 Hz, 2H), 4.85 (t, 3J ¼ 7.5 Hz, 1H), 4.26 (d, 2J ¼ 13.9 Hz,
3H), 3.54 (d, 2J ¼ 13.9 Hz, 3H), 3.07 (d, 3J ¼ 7.5 Hz, 2H), 1.90 (t,
J ¼ 2.4 Hz, 1H). 13C NMR (75 MHz, CDCl3) d 149.98 (q), 148.96 (q), 130.29 (q), 129.17 (t), 129.09 (t), 129.03 (t), 128.44 (q), 128.33 (q), 128.28 (q),124.54 (t),122.61 (t),122.41 (t), 82.45 (q), 69.56 (t), 35.47 (t), 31.94 (s), 31.84 (s), 22.32 (s). HRMS (ESI, m/z): Calculated for C31H26O4Hþ ([MþH]þ): 463.1909, Found: 463.1932.
4.1.9.General procedure for preparation of 25,26,27,28-tetra [(ethoxycarbonyl)methoxy]-2-alkylcalix[4]arenes (10aed)
25,26,27,28-Tetrahydroxy-2-alk(yn)ylcalix[4]arene 9aed (0.49e1.61 mmol) was dissolved in dry acetonitrile (15 mL) followed by addition of fi nely powdered anhydrous Na2CO3 (4.9e16.1 mmol, ca. 10 equiv.). The resulting suspension was stirred for 1 h at 30 ti C, ethyl bromoacetate (4.9e16.1 mmol, 10 equiv.) was added, and the reaction mixture was refl uxed for 20 h. The solvent was evaporated in vacuo and the residue partitioned between dichloromethane (30 mL) and water (30 mL). The water layer was extracted with dichloromethane
(2 ti 15 mL). The combined organic extracts were dried on MgSO4, fi ltered, and evaporated in vacuo to yield the crude product, which was purifi ed by silica column chromatog- raphy employing a gradient of 15e30% of ethyl acetate in heptane to obtain the pure target compound. For 10d, a slightly different workup was employed (see below).
4.1.9.1.25,26,27,28-Tetra[(ethoxycarbonyl)methoxy]-2-methylcalix [4]arene (10a). Precursor and reagents: 9a (367 mg, 0.84 mmol); Na2CO3 (887 mg, 8.4 mmol); ethyl bromoacetate (0.93 mL, 8.4 mmol). Crude yield: yellow oil (536 mg, 0.68 mmol, 81%). Yield after silica column chromatography: white solid (155 mg, 0.20 mmol, 24%). 1H NMR (300 MHz, CDCl3) d 6.68e6.57 (m, 12H), 5.25 (q, 3J ¼ 7.2 Hz, 1H), 4.91 (d, 2J ¼ 13.5 Hz, 1H), 4.83 (d, 2
J ¼ 13.7 Hz, 2H), 4.76e4.72 (m, 8H), 4.22 (q, 3J ¼ 7.2 Hz, 4H), 4.21 (q,
3
J ¼ 7.2 Hz, 4H), 3.23 (d, 2J ¼ 13.7 Hz, 3H), 1.54 (d, 3J ¼ 7.2 Hz, 3H), 1.30 (t, 3J ¼ 7.2 Hz, 6H), 1.29 (t, 3J ¼ 7.2 Hz, 6H). 13C NMR (75 MHz, CDCl3) d 170.19 (q), 170.07 (q), 155.86 (q), 155.53 (q), 139.57 (q), 134.85 (q), 134.56 (q), 134.23 (q), 128.49 (t), 128.43 (t), 125.03 (t), 122.95 (t), 122.82 (t), 71.52 (s), 71.31 (s), 60.54 (s), 60.48 (s), 31.51 (s), 31.44 (s), 30.97 (t), 20.45 (p), 14.21 (p). HRMS (ESI, m/z): Calculated for C45H50O12Hþ ([MþH]þ): 783.3375, Found: 783.3378.
4.1.9.2.25,26,27,28-Tetra[(ethoxycarbonyl)methoxy]-2-propylcalix [4]arene (10b). Precursor and reagents: 9b (228 mg, 0.49 mmol); Na2CO3 (518 mg, 4.9 mmol); ethyl bromoacetate (0.54 mL, 4.9 mmol). Crude yield: yellow oil (373 mg, 0.46 mmol, 94%). Yield after silica column chromatography: white solid (122 mg, 0.15 mmol, 31%). 1H NMR (300 MHz, CDCl3) d 6.67e6.55 (m, 12H), 5.02 (t, 3J ¼ 7.7 Hz, 1H), 4.89 (d, 2J ¼ 13.5 Hz, 2H), 4.86 (d, 2
J ¼ 13.7 Hz,1H), 4.85e4.68 (m, 8H), 4.23 (q, 3J ¼ 7.2 Hz, 4H), 4.20 (q,
3
J ¼ 7.2 Hz, 4H), 3.24 (d, 2J ¼ 13.7 Hz, 1H), 3.22 (d, 2J ¼ 13.5 Hz, 2H), 1.88 (q, 3J ¼ 7.3 Hz, 2H), 1.48 (sext, 3J ¼ 7.3 Hz, 2H), 1.30 (t,
J ¼ 7.2 Hz, 6H), 1.28 (t, 3J ¼ 7.2 Hz, 6H), 0.98 (t, 3J ¼ 7.3 Hz, 3H). 13C NMR (75 MHz, CDCl3) d 170.24 (q), 169.97 (q), 155.93 (q), 155.73 (q), 138.22 (q), 134.75 (q), 134.61 (q), 134.50 (q), 128.48 (t), 128.43 (t), 128.31 (t), 125.29 (t), 122.94 (t), 122.77 (t), 71.50 (s), 71.15 (s), 60.49 (s), 60.44 (s), 36.63 (s), 36.39 (t), 31.69 (s), 31.39 (s), 21.47 (s), 14.48 (p),14.24 (p),14.19 (p). HRMS (ESI, m/z): Calculated for C47H54O12Hþ ([MþH]þ): 811.3688, Found: 811.3675.
4.1.9.3.25,26,27,28-Tetra[(ethoxycarbonyl)methoxy]-2-pentylcalix[4]
arene (10c). Precursor and reagents: 9c (500 mg, 1.01 mmol); Na2CO3 (1.07 g, 10.1 mmol); ethyl bromoacetate (1.12 mL, 10.1 mmol). Crude yield: yellow oil (472 mg, 0.56 mmol, 56%). Yield after silica column chromatography: white solid (138 mg, 0.16 mmol, 16%). 1H NMR (300 MHz, CDCl3) d 6.70e6.55 (m, 12H), 4.99 (t, 3J ¼ 7.7 Hz, 1H), 4.90 (d, 2J ¼ 13.5 Hz, 2H), 4.85 (d, 2
J ¼ 13.8 Hz,1H), 4.85e4.68 (m, 8H), 4.23 (q, 3J ¼ 7.2 Hz, 4H), 4.19 (q,
3
J ¼ 7.2 Hz, 4H), 3.24 (d, 2J ¼ 13.8 Hz, 1H), 3.22 (d, 2J ¼ 13.5 Hz, 2H), 1.88 (q, 3J ¼ 7.3 Hz, 2H),1.51e1.38 (m, 2H),1.38e1.26 (m, 4H),1.30 (t, 3
J ¼ 7.2 Hz, 6H), 1.28 (t, 3J ¼ 7.2 Hz, 6H), 0.88 (t, 3J ¼ 7.1 Hz, 3H). 13C NMR (75 MHz, CDCl3) d 170.27 (q), 169.98 (q), 155.96 (q), 155.73 (q), 138.20 (q), 134.72 (q), 134.63 (q), 134.54 (q), 128.56 (t), 128.48 (t), 128.43 (t), 128.29 (t), 125.29 (t), 122.95 (t), 122.77 (t), 71.52 (s), 71.14 (s), 60.50 (s), 60.45 (s), 36.66 (t), 34.33 (s), 32.29 (s), 31.69 (s), 31.39 (s), 28.04 (s), 22.63 (s), 14.24 (p), 14.20 (p), 14.10 (p). HRMS (ESI, m/
z): Calculated for C49H58O12Naþ ([MþNa]þ): 861.3820, Found: 861.3856.
4.1.9.4.25,26,27,28-Tetra[(ethoxycarbonyl)methoxy]-2-(prop-20 -yn- 10 -yl)calix[4]arene (10d). Precursor, solvents and reagents: 9d (660 mg, ca. 1.4 mmol of crude 9d obtained after deprotection of pure 8d); anhydrous acetonitrile (25 mL); Na2CO3 (1.45 g, 13.7 mmol); ethyl bromoacetate (1.52 mL, 13.7 mmol). A slightly different workup was employed: after cooling down to room temperature, salts were removed by fi ltration, the fi ltrate concen- trated in vacuo, and the residue was taken up in dichloromethane (100 mL) and water (100 mL). Layers were separated, and the water layer was washed with dichloromethane (2 ti 50 mL). The com- bined organic extracts were dried over MgSO4, filtered, evaporated in vacuo, and purifi ed by silica column chromatography using a gradient of 30e60% of ethyl acetate in heptanes to obtain the pure target compound as a pale yellow glassy solid (401 mg, 0.50 mmol, 36%). 1H NMR (300 MHz, CDCl3) d 6.70e6.58 (m, 12H), 5.36 (t, 3
J ¼ 7.9 Hz,1H), 4.95e4.72 (m, 11H), 4.25 (q, 3J ¼ 7.1 Hz, 4H), 4.20 (q,
3
J ¼ 7.1 Hz, 4H), 3.25 (d, 2J ¼ 13.7 Hz, 1H), 3.23 (d, 2J ¼ 13.7 Hz, 2H), 2.82 (dd, 3J ¼ 7.9 Hz, 4J ¼ 2.6 Hz, 2H), 1.93 (t, 4J ¼ 2.6 Hz, 1H), 1.30 (t, 3
J ¼ 7.1 Hz, 6H), 1.29 (t, 3J ¼ 7.1 Hz, 6H). 13C NMR (75 MHz, CDCl3) d 170.20 (q), 170.05 (q), 155.93 (q), 155.79 (q), 136.66 (q), 134.65 (q), 134.64 (q), 134.59 (q), 128.98 (t), 128.55 (t), 128.41 (t), 124.88 (t), 122.98 (t), 122.84 (t), 83.60 (q), 71.73 (s), 71.26 (s), 69.72 (t), 60.56 (s), 60.50 (s), 36.04 (t), 31.64 (s), 31.38 (s), 23.67 (s), 14.24 (p), 14.19 (p). HRMS (ESI, m/z): Calculated for C47H50O12Hþ ([MþH]þ): 807.3381, Found: 807.3388.
4.1.10.General procedure for preparation of 25,26,27,28-tetrakis-N- (N,N-dimethyl-2-aminoethyl)carbamoylmethoxy-2-alkylcalix[4]
arenes (11aed)
A solution of 25,26,27,28-tetra[(ethoxycarbonyl)methoxy]-2- alk(yn)ylcalix[4]arene 10aed (0.22e0.60 mmol) in N,N-dimethy- lethylenediamine (10 mL) was stirred at 50 ti C for 48 h. The excess of N,N-dimethylethylenediamine was evaporated under reduced pressure and the resulting crystals were triturated twice with diethyl ether (fi rst 10 mL, then 5 mL) and dried in vacuo. The crude product was then dissolved in 0.2 M aq. HCl, and purifi ed by
preparative HPLC on system A (see general methods). Fractions containing the product were pooled, evaporated at the rotary evaporator to remove the acetonitrile, and lyophilized to yield the TFA-salt of the pure product. To obtain the free tetra-amine, the salt was taken up in saturated aq. NaHCO3 (30 mL) and extracted with dichloromethane (2 ti 30 mL). The combined organic extracts were dried on MgSO4, fi ltered, and evaporated in vacuo.
4.1.10.1.25,26,27,28-Tetrakis-N-(N,N-dimethyl-2-aminoethyl)carba- moylmethoxy-2-methylcalix[4]arene (11a). Precursor: 10a (176 mg, 224 mmol). Crude yield: off-white solid (170 mg, 179 mmol, 80%). Preparative HPLC on system A (see general methods): tR (product) ¼ 8.3e10.8 min (broad peak). Yield after preparative HPLC (TFA-salt): white solid (107 mg, 76 mmol, 34%). Yield after extraction (free tetra-amine): white solid (53 mg, 56 mmol, 25%). 1H NMR (300 MHz, CDCl3) d 7.75 (bs, 2H), 7.61 (bs, 2H), 6.70e6.50 (m, 12H), 4.99 (q, 3J ¼ 7.2 Hz, 1H), 4.63 (d, 3J ¼ 14.1 Hz, 2H), 4.53e4.32 (m, 9H), 3.53e3.35 (m, 8H), 3.27 (d, 2J ¼ 14.1 Hz, 2H), 3.23 (d, 2
J ¼ 12.8 Hz, 1H), 2.50e2.41 (m, 8H), 2.20 (bs, 24H), 1.49 (d,
3
J ¼ 7.2 Hz, 3H). 13C NMR (75 MHz, CDCl3) d ¼ 169.55 (q), 169.53 (q), 155.75 (q), 155.66 (q), 139.48 (q), 134.43 (q), 134.15 (q), 133.83 (q),
128.88 (t), 128.82 (t), 128.76 (t), 125.37 (t), 123.29 (t), 123.22 (t), 74.36 (s), 74.09 (s), 58.15 (s), 58.13 (s), 45.36 (p), 45.34 (p), 37.14 (s), 37.09 (s), 31.45 (t), 31.21 (s), 31.09 (s), 20.31 (p). HRMS (ESI, m/z): Calculated for C53H74N8O8Hþ ([MþH]þ): 951.5702, Found: 951.5727.
4.1.10.2.25,26,27,28-Tetrakis-N-(N,N-dimethyl-2-aminoethyl)carba- moylmethoxy-2-propylcalix[4]arene (11b). Precursor: 10b (486 mg, 599 mmol). Crude yield: off-white solid (584 mg, 596 mmol, 99%). Preparative HPLC on system A: tR (product) ¼ 10.8e12.3 min. Yield after preparative HPLC (TFA-salt): white solid (343 mg, 239 mmol, 40%). Yield after extraction (free tetra-amine): white solid (158 mg, 161 mmol, 27%). 1H NMR (300 MHz, CDCl3) d 7.77 (bs, 2H), 7.64 (bs,
3
2H), 7.71e6.50 (m, 12H), 4.88 (t, J ¼ 7.7 Hz, 1H), 4.64 (d,
2
J ¼ 14.1 Hz, 2H), 4.57e4.32 (m, 9H), 3.52e3.36 (m, 8H), 3.28 (d,
2
J ¼ 14.1 Hz, 2H), 3.22 (d, 2J ¼ 14.1 Hz, 1H), 2.50e2.41 (m, 8H), 2.21 (s, 12H), 2.20 (s, 12H), 1.91e1.77 (m, 2H), 1.46 (sext, 3J ¼ 7.2 Hz, 2H),
0.97 (t, 3J ¼ 7.2 Hz, 3H). 13C NMR (75 MHz, CDCl3) d ¼ 169.55 (q), 169.52 (q), 155.95 (q), 155.62 (q), 138.27 (q), 134.51 (q), 133.95 (q),
133.80 (q), 128.89 (t), 128.79 (t), 128.68 (t), 125.74 (t), 123.19 (t), 74.22 (s), 74.01 (s), 58.13 (s), 45.33 (p), 37.12 (s), 37.06 (s), 36.90 (t), 36.45 (s), 31.33 (s), 31.15 (s), 21.60 (s), 14.46 (p). HRMS (ESI, m/z): Calculated for C55H78N8O8Hþ ([MþH]þ): 979.6015, Found: 979.6026.
4.1.10.3.25,26,27,28-Tetrakis-N-(N,N-dimethyl-2-aminoethyl)carba- moylmethoxy-2-pentylcalix[4]arene (11c). Precursor: 10c (268 mg, 319 mmol). Crude yield: off-white solid (313 mg, 311 mmol, 97%). Preparative HPLC on system A: tR (product) ¼ 12.8e14.0 min. Yield after preparative HPLC (TFA-salt): white solid (177 mg, 121 mmol, 38%). Yield after extraction (free tetra-amine): white solid (110 mg, 109 mmol, 34%). 1H NMR (300 MHz, CDCl3) d 7.77 (bs, 2H), 7.63 (bs, 2H), 6.70e6.48 (m, 12H), 4.87 (t, 3J ¼ 7.5 Hz, 1H), 4.65 (d, 2
J ¼ 14.1 Hz, 2H), 4.56e4.30 (m, 9H), 3.52e3.37 (m, 8H), 3.28 (d,
2
J ¼ 14.1 Hz, 2H), 3.22 (d, 2J ¼ 14.1 Hz, 1H), 2.52e2.40 (m, 8H), 2.22 (s, 12H), 2.21 (s, 12H), 1.91e1.77 (m, 2H), 1.51e1.36 (m, 2H), 1.36e1.23 (m, 4H), 0.88 (t, 3J ¼ 7.1 Hz, 3H). 13C NMR (75 MHz, CDCl3) d 169.55 (q), 169.50 (q), 155.97 (q), 155.61 (q), 138.30 (q), 134.51 (q), 133.94 (q), 133.80 (q), 128.88 (t), 128.79 (t), 128.68 (t), 125.72 (t), 123.21 (t), 123.18 (t), 74.24 (s), 74.00 (s), 58.14 (s), 45.33 (p), 45.31 (p), 37.24 (t), 37.14 (s), 37.05 (s), 34.26 (s), 32.30 (s), 31.34 (s), 31.14 (s), 28.35 (s), 22.73 (s), 14.15 (p). HRMS (ESI, m/z): Calculated for C57H82N8O8Hþ ([MþH]þ): 1007.6328, Found: 1007.6350.
4.1.10.4.25,26,27,28-Tetrakis-N-(N,N-dimethyl-2-aminoethyl) carbamoylmethoxy-2-(prop-20 -yn-10 -yl)calix[4]arene (11d). Precursor: 10d (351 mg, 435 mmol). Instead of 50 ti C for 48 h, the precursor was treated with N,N-dimethylethylenediamine (7.5 mL) at room temperature for 5 days. Crude yield: off-white solid (369 mg, 378 mmol, 87%). According to 1H NMR, purity of this material was >90%. Part of the material (ca. 200 mg) was purifi ed by preparative HPLC on system A: tR (product) ¼ 8.0e10.5 min. Yield after preparative HPLC (TFA-salt): pale yellow solid (170 mg, 119 mmol, 58%). Yield after extraction (free tetra-amine): white solid (68 mg, 70 mmol, 34%). 1H NMR (300 MHz, CDCl3) d 7.86 (bs, 2H), 7.64 (bs, 2H), 6.70e6.51 (m, 12H), 5.29 (t, 3J ¼ 7.6 Hz, 1H), 4.75 (d, 2J ¼ 14.3 Hz, 2H), 4.52 (d, 2J ¼ 14.2 Hz, 3H), 4.36 (d, 2J ¼ 14.3 Hz, 6H), 3.58e3.38 (m, 8H), 3.31 (d, 2J ¼ 14.3 Hz, 2H), 3.22 (d, 2
J ¼ 13.9 Hz, 1H), 2.73 (dd, 3J ¼ 7.8 Hz, 4J ¼ 2.4 Hz, 2H), 2.55e2.46 (m, 8H), 2.23 (s, 12H), 2.22 (s, 12H), 1.96 (t, 4J ¼ 2.4 Hz, 1H). 13C NMR (75 MHz, CDCl3) d 169.60 (q), 169.53 (q), 156.07 (q), 155.75 (q), 136.84 (q), 134.54 (q), 133.85 (q), 133.81 (q), 129.35 (t), 128.89 (t), 128.84 (t), 125.51 (t), 123.27 (t), 123.23 (t), 83.19 (q), 74.37 (s), 74.13 (s), 70.05 (t), 58.12 (s), 58.08 (s), 45.31 (p), 37.09 (s), 37.06 (s), 36.45 (t), 31.41 (s), 31.03 (s), 23.56 (s). HRMS (ESI, m/z): Calculated for C55H74N8O8Hþ ([MþH]þ): 975.5708, Found: 975.5723.
4.1.11.25,26,27,28-Tetrakis-N-(N,N-dimethyl-2-aminoethyl) carbamoylmethoxy-2-{[1-(2-fluoroethyl)-1H-[1,2,3]triazo-4-yl]
methyl}calix[4]arene (13)
To a stirred solution of CuSO4 ti 5H2O (67 mg, 268 mmol) and L- ascorbic acid (92 mg, 522 mmol) in water (3 mL) under an N2 at- mosphere was added a solution of crude 11d (169 mg, 173 mmol) in N,N-dimethylformamide (2.5 mL) and 2-fl uoroethylazide 5 (212 mmol; 3.6 mL of a 0.0588 M solution in N,N-dimethylforma- mide freshly prepared following published procedures [25]). The mixture was stirred at room temperature overnight, and then evaporated to dryness. The residue was treated with dichloro- methane (15 mL) and aqueous Na2CO3 (15 mL). The layers were separated and the aqueous layer was extracted again with dichloromethane (2 ti 15 mL). The combined organic layers were dried over MgSO4, filtered, and evaporated in vacuo to yield the target compound (133 mg, 125 mmol, 72%). The crude material was taken up in a mixture of acetonitrile and water (ca. 2.5 mL) and purifi ed by preparative HPLC on system A. Fractions containing the
product (broad peak; tR ¼ 7.0e9.5 min) were pooled, partly evap- orated at the rotary evaporator to remove the acetonitrile, and lyophilized to obtain the TFA-salt of 13 as a white solid (116 mg, 76 mmol, 44%). The solid was taken up in dichloromethane (30 mL) and thoroughly extracted with saturated aqueous NaHCO3 (30 mL). The layers were separated and the aqueous layer was again extracted with dichloromethane (30 mL). The combined organic extracts were dried over MgSO4, fi ltered, and evaporated to yield target compound 13 as the free tetra-amine. Yield: 48 mg (45 mmol, 26%) of a white solid. 1H NMR (600 MHz, CDCl3) d 7.88 (bs, 2H), 7.78 (bs, 2H), 7.67 (s, 1H), 6.70e6.52 (m, 12H), 5.51 (t, 3J ¼ 8.0 Hz, 1H),
4.77 (dt, 2JHF ¼ 46.8 Hz, 3JHH ¼ 4.6 Hz, 2H), 4.64e4.37 (m, 11H), 4.24e4.13 (m, 2H), 3.51e3.38 (m, 8H), 3.32 (d, 3J ¼ 8.2 Hz, 2H), 3.26 (d, 2J ¼ 14.3 Hz, 2H), 3.23 (d, 2J ¼ 14.1 Hz, 1H), 2.53e2.40 (m, 8H), 2.23 (s, 24H). 13C NMR (75 MHz, CDCl3) d 169.60 (q), 169.56 (q), 155.95 (q), 155.74 (q), 146.81 (q), 137.16 (q), 134.45 (q), 133.90 (q), 129.03 (t), 128.96 (t), 128.76 (t), 125.89 (t), 123.27 (t), 123.21 (t),
81.55 (s, doublet, 1JCF ¼ 171 Hz), 74.13 (s), 73.95 (s), 58.16 (s), 58.00 (s), 50.43 (s, doublet, 2JCF ¼ 21 Hz), 45.34 (p), 45.23 (p), 37.05 (s), 36.97 (s), 36.33 (t), 31.31 (s), 31.09 (s), 30.18 (s). HRMS (ESI, m/z): Calculated for C57H78FN11O8Hþ ([MþH]þ): 1064.6097, Found: 1064.6121.
4.1.12.1,10 -(((Oxybis(methylene))bis(oxy))bis(ethane-2,1-diyl)) bis(4-(25,26,27,28-(tetra-N-(N,N-dimethyl-2-aminoethyl) carbamoylmethoxy)calix[4]arene-2-yl)-1H-1,2,3-triazole) (14)
Crude 11d (100 mg, 102 mmol) was dissolved in N,N-dime- thylformamide (2.4 mL) by slightly warming with a heat gun. Under N2 atmosphere and continued stirring were added solutions of
CuSO4 ti 5H2O (38 mg, 152 mmol in 1 mL of water) and (L)-ascorbic acid (60 mg, 303 mmol in 1 mL of water), followed by a freshly prepared solution of 1,11-diazido-3,6,9-trioxaundecane 12 (12.4 mg, 50.8 mmol in 1 mL of N,N-dimethylformamide), and the reaction mixture was stirred at 50 ti C overnight. For isolation of the product, aliquots of the reaction mixture were 1:1 diluted with
2M aq. HCl, and purifi ed by preparative HPLC on system A. Frac- tions containing the product (peak; tR ¼ 16.0e17.5 min) were
pooled, evaporated at the rotary evaporator to remove the aceto- nitrile, and lyophilized to yield the TFA-salt of the pure product as a white solid (75 mg, 24 mmol, 47%). The solid was taken up in dichloromethane (20 mL), and thoroughly extracted with 1 M aqueous NaOH (20 mL). Layers were separated, and the aqueous layer was again extracted with dichloromethane (3 ti 20 mL). The combined organic layers were dried on MgSO4, fi ltered, and evap- orated in vacuo to yield target compound 14 as the free tetra-amine. Yield: 44 mg (20 mmol, 39%) of a white fluffy solid. 1H NMR (600 MHz, CDCl3) d 7.89 (bs, 4H), 7.85 (bs, 4H), 7.55 (s, 2H), 6.65e6.51 (m, 24H), 5.42 (t, 3J ¼ 7.9 Hz, 2H), 4.55e4.39 (m, 22H), 4.22 (d, 2J ¼ 13.5 Hz, 4H), 3.81 (t, 3J ¼ 5.3 Hz, 4H), 3.52e3.36 (m, 24H), 3.29 (d, 3J ¼ 7.8 Hz, 4H), 3.23 (d, 2J ¼ 14.2 Hz, 2H), 3.22 (d, 2
J ¼ 14.2 Hz, 4H), 2.47e2.35 (m, 16H), 2.21 (s, 24H), 2.19 (s, 24H). 13C NMR (150 MHz, CDCl3) d 169.57 (q), 169.42 (q), 155.83 (q), 155.63 (q), 145.95 (q), 137.04 (q), 134.44 (q), 134.08 (q), 133.91 (q), 128.94 (t), 128.85 (t), 128.71 (t), 125.76 (t), 123.20 (t), 123.12 (t), 74.00 (s), 70.42 (s), 70.33 (s), 69.43 (s), 58.15 (s), 58.00 (s), 50.08 (s), 45.36 (p), 45.33 (p), 37.09 (s), 37.06 (s), 36.26 (t), 31.17 (s), 31.04 (s), 30.02 (s). HRMS (ESI, m/z): Calculated for C118H164N22O19H22þ ([Mþ2H]2þ): 1097.6344, Found: 1097.6384.
4.2.Cell viability assay
MA148 cells, a human epithelial ovarian carcinoma cell line, were seeded in non-coated 96-well plates (Corning; Lowell, MA), whereas human umbilical vein derived EC (HUVEC) were seeded in a 96-well culture plate coated with 0.2% gelatin for 2 h at 20 ti C (SigmaeAldrich; St. Louis, MO). Both cell types were seeded at a concentration of 3000 cells per well and allowed to adhere for at least 3 h at 37 ti C in 5% CO2/95% air before treatments were initiated. The cells were then exposed to complete medium, with or without various concentrations of 0118 for 72 h or as indicated otherwise. Cell counting kit (CCK-8; Dojindo; Gaithersburg, MD) was used to assess cell viability rates relative to untreated cells, as described earlier [42]. All measurements were done in triplicate, and the experiments were done at least three times. The IC50 was deter- mined as half the maximal inhibitory drug concentration for each cell line. HUVEC and MA148 cells were kindly provided by Prof. Dr. Ramakrishnan (University of Minnesota) [43] and propagated as described earlier [44].
4.3.Radiochemistry
4.3.1.General materials and methods
Radiosynthesis was performed on a custom-designed ‘Modular- Lab’ system. The individual modules, which make up this ‘Modular- Lab’ system are commercially available from Eckert & Ziegler Eurotope GmbH (Berlin, Germany). The core of the system is made up of two Peltier reactor modules (PRM), which allow temperature control from ti 40 ti C to þ150 ti C. Both are equipped with
temperature and radioactivity sensors, magnetic stirrers, reactor cameras and pneumatic reactor lifts. The reactions were performed in 3 mL borosilicate glass V-vials from Alltech (Grace Discovery Sciences, Deerfi eld, IL) equipped with reactor heads from Eckert &
Ziegler and EPDM flat seals. Connections for liquid transfer were from PTFE and FEP tubing. Solvents were evaporated employing a constant stream of argon regulated by a fl ow controller module. Radioactive exhaust vapors were condensed in a vacuum trap cooled with liquid nitrogen, and subsequently passed through an activated carbon filter. Liquid transfers were performed using either vacuum or a positive pressure of argon (1.5 bar), or both. The valve modules used for assembling the system were 3/2-way and 2/
2-way solenoid valve modules (SMC-valves type LVM), a single stopcock module, and stopcock manifold modules for gas transfer and the fi nal radiopharmaceutical formulation step. Semi- preparative HPLC purifi cation was performed on a SymmetryPrep C18 column (100 Å, 7 mm, 7.8 ti 300 mm, Waters Corporation, Milford, MA) integrated into an HPLC module equipped with an electrically driven 6-port-multi-channel valve, an Omron E3X-DA-S fl uid sensor, a preparative sample loop, a radioactivity detector in series with a WellChrom fi xed wavelength detector K-200
(l ¼ 254 nm; Knauer GmbH, Berlin, Germany), and two WellChrom HPLC pumps K120 (Knauer GmbH). The complete system was
placed in a hot-cell, whereas the PLC (programmable logic controller), the cooling unit, the cold trap and the shielded vacuum pump were located in the service corridor behind the hot-cell. All processes were remotely controlled on a PC employing the dedi- cated Modular-Lab software interface from Eckert & Ziegler, which allows straightforward set-up of the interactive process panel, fl exible programming, and provides GMP-compliant batch records including temperature, activity, and UV traces.
Analytical HPLC-analysis for monitoring reaction progress and composition of crude products and for quality control of the fi nal tracer product was carried out on an Agilent 1100 Series system with a binary pump and a variable wavelength UV-detector (preset to 271 nm, which is lmax of 6) in series with a Gabi-Star radioac- tivity detector (Raytest GmbH, Straubenhardt, Germany). The samples (20 mL) were injected onto a Symmetry C18 column (100 Å, 5 mm, 3.9 ti 150 mm, Waters Corporation, Milford, USA), which was eluted at 1 mL/min with a linear gradient of acetonitrile (B) in water (A), both containing 0.1% TFA. Three different gradients were used
Radioactivity of the [18F]-charged QMA-cartridges, [18F]-in- termediates, and the fi nal product was measured in a calibrated digital ionization chamber (model VIK-202, Veenstra Instruments, Joure, The Netherlands). Water was purifi ed and de-ionized (18 MU cm) by means of a Milli-Q water filtration system (Milli- pore). The Sep-Pak® Plus Light C8 cartridges for solid phase extraction were purchased from Waters (USA). The syringe filters (GD/X syringe filter, PTFE, 0.45 mm, with borosilicate prefi lter) for filtration of the crude click reaction mixture were from Whatman (UK). Anhydrous acetonitrile and Kryptofi x 222 were from Merck, tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA) and (R)-
(ti)-(3,5-dioxa-4-phosphacyclohepta[2,1-a:3,4-a0 ]dinaphthalen-4- yl)dimethylamine (MonoPhos) from SigmaeAldrich, and 4,7- diphenyl-1,10-phenanthroline disulfonic acid (BPDS) from Pfaltz
& Bauer, Inc. (USA). Purity of all ligands was >96% (HPLC). All other reagents, catalysts, and solvents were obtained from Acros (Belgium) or SigmaeAldrich and were used without further purifi cation.
4.3.2.Radiosynthesis of 2-[18F]fluoroethylazide [18F]5
[18F]Fti was purchased from BV Cyclotron VU (Amsterdam, The Netherlands). It was produced by the 18O(p,n)18F nuclear reaction in an IBA 18/9 cyclotron and subsequently trapped on a QMA- cartridge (Waters Sep-Pak® Plus Light QMA; carbonate form) for shipment. After arrival at our facilities (about 1e2 half-lives after EOB), [18F]Fti was eluted from the anion exchange column into a
3mL V-vial using 1 mL of acetonitrile/water (9/1, v/v), which con- tained Kryptofi x 222 (13 mg, 34 mmol) and K2CO3 (2 mg, 14 mmol). The solution was dried under an argon fl ow (~70 mL/min) and reduced pressure at 70 ti C for 5 min, 100 ti C for 2 min, and 110 ti C for
6min (4 min lift-position ‘up’ and 2 min lift-position ‘down’). To remove residual water, anhydrous acetonitrile (0.9 mL) was added, and the solution was dried again at 70 ti C for 2 min and 110 ti C for 3 min (2 min lift-position ‘up’ and 1 min lift-position ‘down’). This co-evaporation cycle was repeated once.
After cooling to 40 ti C, a solution of 2-azidoethyl-4- methylbenzenesulfonate 15 (5 mL; prepared according to pub- lished procedures [32,33] with minor modifi cations) in anhydrous acetonitrile (0.7 mL) was added to the K[18F]FeK222 residue, allowed to react for 15 min at 80 ti C, and the intermediate 2-[18F]
fluoroethylazide [18F]5 was co-distilled at 90 ti C under a constant
18
for analysis of the intermediate 2-[18F]fl uoroethylazide [
F]5
stream of argon (20 mL/min) into a 3 mL septum-capped V-vial,
(Gradient I: 3 min at 50% B followed by a linear gradient to 75% B in
7min, a linear gradient to 95% B in 1 min, and subsequent isocratic elution for 2 min) and for analysis of click reaction mixtures and formulated radiotracers [18F]6 (Gradient II: 3 min at 30% B, followed by a linear gradient to 40% B in 5 min, isocratic elution for 2 min, a linear gradient to 95% B in 1 min, then isocratic elution at 95%B, return to 30% B, and equilibrating for at least 5 min before injection of the next sample) and [18F]13 (Gradient III: 3 min at 27% B, fol- lowed by a linear gradient to 29.5% B in 10 min, isocratic elution for
2min, a linear gradient to 95% B in 1 min, then isocratic elution at 95%B, return to 27% B, and equilibrating for at least 5 min before injection of the next sample), respectively. Retention times (tR) for the starting materials, intermediates, reference compounds and radiotracers (the latter based on the radio-trace) were as follows: 2- azidoethyl-4-methylbenzenesulfonate 15 (Gradient I: 5.5 min, Gradient II: 12.4 min); [18F]5 (Gradient I: 2.9 min; Gradient II: 4.6 min; Gradient III: 5.2 min); click precursor 4 (Gradient II: 7.6 min); reference compound 6 (Gradient II: 6.9 min), radiotracer [18F]6 (Gradient II: 7.3 min); click precursor 11d (Gradient III: 11.3 min); reference compound 13 (Gradient III: 8.7 min), radio- tracer [18F]13 (Gradient III: 9.1 min). UV and radioactivity detectors were arranged in series, leading to a delay of about 0.4 min in the radio-trace.
which was pre-cooled to ti 40 ti C. After 10 min, the integrated radioactivity detectors in the reactors indicated that distillation was complete, and the radioactivity collected in the receiver vial was measured in a dose calibrator. Decay-corrected isolated radio- chemical yield of 2-[18F]fl uoroethylazide was 53 ± 5% (n ¼ 10). HPLC-analysis of the distillate indicated absence of precursor 15 in the UV-trace and a radiochemical purity of >99%. The volume of the 2-[18F]fluoroethylazide distillate in acetonitrile in these experi- ments was about 300e400 mL.
4.3.3. Optimization of CuAAC conditions for reaction of [18F]5 with precursor 11d
Stock solutions of precursor 11d (1.46 mg, 1.5 mmol) were pre- pared in: A) 250 mL of DMSO (Table 2, entries 1, 2, 5, 6, 8, 9, 10, 13, and 14); B) 200 mL of DMSO (Table 2, entries 3, 4, 11, and 12); C) 50 mL of sodium phosphate buffer pH 6.0 and 200 mL of water (Table 2, entry 7).
Ligand stock solutions: TBTA and MonoPhos were dissolved in DMSO at a concentration of 16.5 mM (Table 2, entries 3, 4, 11 and 12), and BPDS was dissolved in water at the same concentration (Table 2, entry 2, 5, 10, and 13), or at 99.0 mM (Table 2, entry 8).
Catalyst stock solutions: Solutions of CuSO4 ti 5H2O and sodium L-ascorbate were freshly prepared before each experiment.
CuSO4 ti 5H2O was dissolved in water at the following concentra- tions: A) 10 mM (Table 2, entries 1, 3, 4, 9,11,12); B) 30 mM (Table 2, entries 2, 5, 10, 13); C) 60 mM (Table 2, entries 6, 7, 14); D) 180 mM (Table 2, entry 8). Sodium L-ascorbate stock solutions were as fol- lows: A) 150 mM in 0.5 M sodium phosphate buffer pH 6.0 (Table 2, entries 1, 2, 3, 4, 9,10,11,12); B) 150 mM in 0.5 M Tris/HCl-buffer pH 8.5 (Table 2, entries 5 and 13); C) 900 mM in 0.5 M sodium phos- phate buffer pH 6.0 (Table 2, entries 6 and 14); C) 900 mM in 0.5 M Tris/HCl-buffer pH 8.5 (Table 2, entry 8); D) 900 mM in water (Table 2, entry 7).
Experimental procedure: When applicable, ligand stock solu- tions (50 mL) were first added to the solution of precursor 11d in a
3mL V-vial containing a magnetic stirring bar. Immediately before addition to the precursor solution, the catalyst mixture was pre- pared as follows: Aqueous CuSO4-solutions, 150 mL (Table 2, entries 1, 3, 4, 6, 7, 9, 11, 12, and 14; concentration depending on experi- ment, see above) or 50 mL (Table 2, entries 2, 5, 8, 10, 13; concen- tration depending on experiment, see above) were mixed with aqueous Na-ascorbate solutions (100 mL, concentration and buffer depending on experiment, see above), and then half of the total volume was added to the corresponding precursor solution. A so- lution of 2-[18F]fluoroethylazide in acetonitrile (150 mL, 50e150 MBq) was added, and the solution was thoroughly mixed by pipetting up and down. The reaction vial was then sealed with a septum screw cap, and the mixture was incubated at the indicated temperature (23 ti C or 80 ti C), while stirring with a magnetic stirring bar. After 15 min and 30 min, samples were retrieved using a sy- ringe, and frozen in liquid nitrogen. Immediately before HPLC analysis, samples were thawed by diluting to an appropriate ac- tivity concentration with water containing 0.1% TFA, and directly injected. Analytical yield (conversion) was calculated based on the peak areas of unreacted 2-[18F]fl uoroethylazide and click product [18F]13 in the radio-trace.
4.3.4. Radiosynthesis, purification, and formulation of [18F]6 and [18F]13
water (9 mL), eluted with ethanol (1 mL) into a 3 mL V-vial, evaporated to dryness at 80 ti C under a stream of argon, and redissolved in DMSO to the desired target concentration for sub- sequent in vitro and in vivo studies.
4.4.In vitro characterization of radiotracers
4.4.1.Distribution coefficient Log D7.4 in 1-octanol/PBS pH 7.4
Log D7.4 values were measured using the shake fl ask method. To 1-octanol (0.5 mL) and PBS pH 7.4 (0.5 mL), which had been pre- saturated with each other, was added 1 mL of the fi nal purified radiotracer stock solution in DMSO (ca.1 MBq, radiochemical purity as determined by analytical HPLC was >98%), and the solvents were thoroughly mixed in an eppendorf mixer for 30 min (in addition, the tubes were manually mixed in 5e10 min intervals using a vortex). Following phase separation by centrifugation in an eppendorf centrifuge at 16,100 rcf for 5 min, aliquots of both phases (3 ti 25 mL for each phase) were counted in a g-counter (Wizard 1480; PerkinElmer). Log D was calculated as log (cpm in octanol/
cpm in PBS).
4.4.2.Radiotracer stability in PBS pH 7.4 and in mouse serum
The radiotracers were incubated at 37 ti C under gentle agitation (350 rpm) in either PBS pH 7.4 or mouse serum (Balb/C mouse serum, purchased from Innovative Research, and fi ltered through a 0.45 mm PTFE syringe fi lter before use) at activity concentrations of 10e15 MBq/mL, and 25e50 MBq/mL, respectively. Maximum DMSO concentration in the incubation mixtures (due to addition of the tracer stock solution in DMSO) was below 2%. For the PBS- incubation, aliquots were removed after 0, 1, 2, and 3 h, and directly analyzed by radio-HPLC. For the serum samples, aliquots (150e250 mL) were retrieved (same time points), thoroughly mixed with an equal volume of ice-cold acetonitrile (ti 20 ti C), and proteins separated by centrifugation in an Eppendorf centrifuge (16,100 rcf, 5 min). An aliquot of the supernatant was removed for radio-HPLC, and another aliquot (50 mL) for measuring of radioactivity in the
18
The intermediate 2-[18F]fl uoroethylazide [
F]5 e prepared as
supernatant. To quantify probe binding to serum proteins, the
described in Section 4.3.2 e was co-distilled at 90 ti C under a constant stream of argon (20 mL/min) into a 3 mL septum-capped
V-vial pre-cooled to ti 40 ti C, which contained a solution of pre- cursor 11d (2.9 mg, 3 mmol) or 4 (2.6 mg, 3 mmol) in DMSO (500 mL).
In the meantime, stock solutions of 60 mM CuSO4 ti 5H2O in water and 900 mM sodium L-ascorbate in 0.5 M sodium phosphate buffer pH 6.0 were prepared. When co-distillation of [18F]5 was complete (ca. 10 min), the collector vial was quickly warmed up to room temperature, and 250 mL of the aqueous catalyst mixture (prepared immediately before addition to the reaction vessel by mixing
300 mL of the CuSO4 ti 5H2O-stock solution with 200 mL of the Na- ascorbate stock solution) was manually added via a syringe, which corresponds to 3 equiv. of CuSO4 (9 mmol) and 30 equiv. of ascorbate (90 mmol) in the fi nal reaction mixture. Following reaction at 80 ti C for 15 min, the reaction mixture was diluted with water acidifi ed with 0.1% TFA (1.5 mL) and loaded onto the C18 semi-preparative HPLC-column pre-conditioned with the initial eluent mixture (for [18F]6: 30% acetonitrile in water e both eluents containing 0.1% TFA; for [18F]13: 27% acetonitrile in water e both eluents containing 0.1% TFA) via a preparative sample loop. The flow was increased step- wise from 2 mL/min to 7 mL/min within 1 min, and elution with the initial eluent mixture was continued for 3.5 min, followed by a linear gradient (for [18F]6: 30%e40% in 5 min followed by isocratic elution for 5 min; for [18F]13: 27%e29.5% in 10 min). Retention time
remaining supernatant was discarded, and radioactivity of the pellet and supernatant-aliquot determined in the gamma counter. Importantly, before injection into the HPLC, supernatants were fi rst 10-fold diluted with eluent A (0.1% TFA in water). To eliminate any effects due to sample processing, stability values (% intact radio-
tracer) were reported relative to the control sample (t ¼ 0), for which the amount of intact probe was set to 100%.
4.5.Animal studies
Mice were housed and handled according to institutional guidelines complying with European legislation and the Dutch national law “Wet op de Dierproeven” (Staatsblad 1985, 336). Approval of the local IACUC was obtained prior to the commence- ment of the studies. Female nude BALB/c mice (19e24 g body weight; Charles River Laboratories, USA) were inoculated subcu- taneously with 3 ti 106 MDA-MB-231-LITG cells in 100 mL of sterile phosphate-buffered saline and were used when the tumors had reached a size of 70e350 mm3 (typically after 17e23 days); tumor weights as determined after dissection were 0.141 ± 0.077 g (n ¼ 12). Animals were kept in a temperature-controlled environ- ment with a 12-h light/12-h dark cycle. They received a standard diet (Sniff R/M-H, Sniff Spezialdi€aten GmbH, Soest, The Netherlands) and acidifi ed water ad libitum.
18
for [
F]6 and [18F]13 on these systems was about 9 min. The
Four groups of mice (n ¼ 3e4) were intravenously injected with
product fraction (total volume ca. 1.5e2 mL) was diverted into a septum-capped bottle containing water (35 mL). The purified ra- diotracers were trapped on a C8 Sep-Pak® cartridge, rinsed with
radiotracers [18F]6 (0.2e0.3 MBq, ca. 60 ng of 6 for the low-dose biodistribution; ~3.5 MBq, ca. 1 mg of 6 for the blood kinetics) and [18F]13 (0.3e0.4 MBq, ca. 100 ng of 13 for the low-dose
biodistribution; ~8.5 MBq, ca. 1 mg of 13 for the blood kinetics), administered as solutions in 10 mM sodium phosphate-buffered 0.9% saline (pH 5.5) containing 10% vol. DMSO. To assess blood clearance, blood samples (20e30 mL) were drawn from the vena saphena at 2e5e10e30e60e120 min post radiotracer injection. After 3 h, mice were euthanized by cervical dislocation, blood was obtained by cardiac puncture, and selected tissues and organs were harvested, blotted dry and weighed. Phosphate-buffered saline (1 mL) was then added to the samples, and sample radioactivity was measured in a g-counter (Wizard 1480 from PerkinElmer; energy window 400e600 keV) along with standards to determine the percentage of injected dose per gram of tissue (%ID/g) and the percentage of injected dose per organ (%ID/organ). For blood, muscle, and bone, the uptake of tracer as %ID/organ was calculated from the %ID/g values assuming that these organs constitute 6, 41, and 11% of the body weight. The %ID/organ values for stomach, small and large intestines relate to organ including content. Data analysis was performed with GraphPad Prism version 6.04. All data are presented as mean %ID/g or %ID/organ ± one standard deviation (SD). P-values were calculated using a 2-tailed unpaired Student’s t- test. Differences were considered statistically signifi cant when the P-value was less than 0.05. For calculation of blood half-lives, blood clearance data was fi tted using two-phase exponential decay
2 2
functions ([18F]6: R ¼ 0.987; [18F]13: R ¼ 0.998).
For PET-CT imaging, mice were anesthetized using medical air containing 2e3% isofl urane (0.4 L/min), and anesthesia was main- tained throughout the whole experimental procedure with medical air containing 1e2% isofl urane (0.4 L/min). Body temperature was stabilized by heating the animal bed via a continuous flow of warm air. After sedation, the tail veinwas cannulated and radiotracers [18F]
6 (12 MBq, 4.8 nmol) and [18F]13 (13 MBq, 3.7 nmol) were admin- istered as solutions in 10 mM sodium phosphate-buffered 0.9% sa- line (pH 5.5) containing 10% vol. DMSO. Immediately after injection of the radiotracers, a 2 h whole body PET list-mode scan was ac- quired on a MOSAIC Animal PET scanner (Philips Medical Systems, Cleveland, OH) with an axial resolution of 2.2 mm, and a trans- mission scan using a Cs-137 external source was performed for attenuation correction. After the PET-scan, a whole-body CT scan with a resolution of 0.1 mm was acquired on a NanoSPECT-CT scanner (Bioscan Inc., Washington, DC). The animal positioning was kept constant during both scans. To enhance soft tissue contrast, the acquisition parameters were set to 45 kV peak tube voltage, 80 mm fi eld of view, and 360 projections at 2000 ms per projection. After the scan, blood was retrieved by cardiac puncture, and animals were sacrifi ced by cervical dislocation and dissected. Selected tis- sues were counted in the g-counter as described above. %ID/g and % ID/organ values were similar to the biodistribution results obtained for the two groups of mice employed for the blood clearance study (results not shown here). PET images for the timeframes of interest were reconstructed with a 3DRAMLA algorithm employing atten- uation correction, and CT images with an Exact Cone Beam Filtered Back Projection (Logan fi lter) to a resolution of 0.1 mm ti 0.1 mm ti 0.1 mm. Co-registration of PET and CT images was performed with a Philips IMALYTICS Research Workstation (Version 3.0, Philips Research, Aachen, Germany) based on anatomical landmarks and a rigid registration algorithm. Volumes of interest (VOI) for analysis of tracer levels in muscle and tumor were defined by manual segmentation of the reconstructed CT images.
Conflict of interest
Co-authors A.W. Griffi oen and K.H. Mayo have a fi nancial in- terest in PepTx, a pharmaceutical company holding license on PTX compounds. T. L€appchen, R. Rossin, K. Donato, J. Lub, and H. Grüll are employees of Philips Electronics Nederland B.V.
Acknowledgments
We thank Mari€elle van Egmond for synthesis of tetramethox- ycalix[4]arene 7, and we thank Jeroen Pikkemaat, Jeroen P.W. van den Berg, and Hugo Knobel for technical support with NMR- and LC-MS analyses. Dr. Iris Verel, Caren van Kammen, Carlijn van Helvert, and Monique Berben are gratefully acknowledged for assistance in planning and executing of the in vivo studies. This work was fi nancially supported by the Center for Translational Molecular Medicine e Mammary Carcinoma Molecular Imaging for Diagnosis and Therapeutics (CTMM-MAMMOTH), contract 2010249, project number 03O-201.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.10.048. These data include MOL files and InChiKeys of the most important compounds described in this article.
References
[1]S. Chakrabarti, C.J. Barrow, R.K. Kanwar, V. Ramana, J.R. Kanwar, Current protein-based anti-angiogenic therapeutics, Mini Rev. Med. Chem. 14 (2014) 291e312.
[2]J. Folkman, Tumor angiogenesis: therapeutic implications, N. Engl. J. Med. 285 (1971) 1182e1186.
[3]M.S. O’Reilly, L. Holmgren, Y. Shing, C. Chen, R.A. Rosenthal, M. Moses, W.S. Lane, Y. Cao, E.H. Sage, J. Folkman, Angiostatin: a novel angiogenesis inhibitor that mediates the suppression of metastases by a Lewis lung carci- noma, Cell 79 (1994) 315e328.
[4]Y. Cao, R. Langer, A review of Judah Folkman’s remarkable achievements in biomedicine, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 13203e13205.
[5]R.P. Dings, I. Nesmelova, A.W. Griffi oen, K.H. Mayo, Discovery and develop- ment of anti-angiogenic peptides: a structural link, Angiogenesis 6 (2003) 83e91.
[6]A.W. Griffi oen, D.W.J. van der Schaft, A.F. Barendsz-Janson, A. Cox, H.A.J.S. Boudier, H.F.P. Hillen, K.H. Mayo, Anginex, a designed peptide that inhibits angiogenesis, Biochem. J. 354 (2001) 233e242.
[7]R.P.M. Dings, M. Loren, H. Heun, E. McNiel, A.W. Griffi oen, K.H. Mayo, R.J. Griffi n, Scheduling of radiation with angiogenesis inhibitors anginex and avastin improves therapeutic outcome via vessel normalization, Clin. Cancer Res. 13 (2007) 3395e3402.
[8]D.W.J. van der Schaft, R.P.M. Dings, Q.G. de Lussanet, L.I. van Eijk, A.W. Nap, R.G.H. Beets-Tan, J.C.A.B.T. Steege, J. Wagstaff, K.H. Mayo, A.W. Griffi oen, The designer antiangiogenic peptide anginex targets tumor endothelial cells and inhibits tumor growth in animal models, FASEB J. 16 (2002) 1991e1993.
[9]R.P.M. Dings, D.W.J. van der Schaft, B. Hargittai, J. Haseman, A.W. Griffi oen, K.H. Mayo, Anti-tumor activity of the novel angiogenesis inhibitor anginex, Cancer Lett. 194 (2003) 55e66.
[10]A.E.M. Dirkx, M.G.A.O. Egbrink, K. Castermans, D.W.J. van der Schaft, V.L.J.L. Thijssen, R.P.M. Dings, L. Kwee, K.H. Mayo, J. Wagstaff, J.C.A.B.T. Steege, A.W. Griffi oen, Anti-angiogenesis therapy can overcome endothelial cell anergy and promote leukocyte-endothelium interactions and infiltration in tumors, FASEB J. 20 (2006) 621e630.
[11]K.H. Mayo, R.P.M. Dings, C. Flader, I. Nesmelova, B. Hargittai, D.W.J. van der Schaft, L.I. van Eijk, D. Walek, J. Haseman, T.R. Hoye, A.W. Griffi oen, Design of a partial peptide mimetic of anginex with antiangiogenic and anticancer ac- tivity, J. Biol. Chem. 278 (2003) 45746e45752.
[12]R.P.M. Dings, X.M. Chen, D.M.E.I. Hellebrekers, L.I. van Eijk, Y. Zhang, T.R. Hoye, A.W. Griffi oen, K.H. Mayo, Design of nonpeptidic topomimetics of anti- angiogenic proteins with antitumor activities, J. Natl. Cancer Inst. 98 (2006) 932e936.
[13]R.P.M. Dings, M.C. Miller, I. Nesmelova, L. Astorgues-Xerri, N. Kumar, M. Serova, X.M. Chen, E. Raymond, T.R. Hoye, K.H. Mayo, Antitumor agent calixarene 0118 targets human galectin-1 as an allosteric inhibitor of carbo- hydrate binding, J. Med. Chem. 55 (2012) 5121e5129.
[14]V.L.J.L. Thijssen, R. Postel, R.J.M.G.E. Brandwijk, R.P.M. Dings, I. Nesmelova, S. Satijn, N. Verhofstad, Y. Nakabeppu, L.G. Baum, J. Bakkers, K.H. Mayo, F. Poirier, A.W. Griffioen, Galectin-1 is essential in tumor angiogenesis and is a target for antiangiogenesis therapy, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 15975e15980.
[15]H. Leffl er, U.J. Nilsson, Low-molecular weight inhibitors of galectins, in: A.A. Klyososv, P.G. Traber (Eds.), Galectins and Disease e Implications for Targeted Therapeutics, ACS Symposium Series, vol. 1115, American Chemical Society, Washington DC, 2012, pp. 47e59.
[16]P.M. Collins, C.T. Oberg, H. Leffler, U.J. Nilsson, H. Blanchard, Taloside in- hibitors of galectin-1 and galectin-3, Chem. Biol. Drug Des. 79 (2012) 339e346.
[17]L. Astorgues-Xerri, M.E. Riveiro, A. Tijeras-Raballand, M. Serova, C. Neuzillet, S. Albert, E. Raymond, S. Faivre, Unraveling galectin-1 as a novel therapeutic target for cancer, Cancer Treat. Rev. 40 (2014) 307e319.
[18]L. Astorgues-Xerri, M.E. Riveiro, A. Tijeras-Raballand, M. Serova, G.A. Rabinovich, I. Bieche, M. Vidaud, A. de Gramont, M. Martinet, E. Cvitkovic, S. Faivre, E. Raymond, OTX008, a selective small-molecule inhibitor of galectin-1, downregulates cancer cell proliferation, invasion and tumour angiogenesis, Eur. J. Cancer 50 (2014) 2463e2477.
[19]R.P.M. Dings, J.I. Levine, S.G. Brown, L. Astorgues-Xerri, J.R. MacDonald, T.R. Hoye, E. Raymond, K.H. Mayo, Polycationic calixarene PTX013, a potent cytotoxic agent against tumors and drug resistant cancer, Invest. New Drugs 31 (2013) 1142e1150.
[20]K. Iwamoto, K. Araki, S. Shinkai, Conformations and structures of tetra-O- alkyl-p-tert-butylcalix[4]arenes. How is the conformation of calix[4]arenes immobilized? J. Org. Chem. 56 (1992) 4955e4962.
[21]H. Shimizu, K. Iwamoto, K. Fujimoto, S. Shinkai, Chromogenic calix[4]arene, Chem. Lett. 20 (1991) 2147e2150.
[22]J. Cooper, M. Drew, P. Beer, Alkali metal cation cooperative anion recognition by heteroditopic bis(calix[4]arene) rhenium(I) bipyridyl and ferrocene re- ceptor molecules, J. Chem. Soc. Dalton Trans. 16 (2000) 2721e2728.
[23]C.M. Shu, W.S. Chung, S.H. Wu, Z.C. Ho, L.G. Lin, Synthesis of calix[4]arenes with four different “lower rim” substituents, J. Org. Chem. 64 (1999) 2673e2679.
[24]L.H. Bryant Jr., A.T. Yordanov, J.J. Linnoila, M.W. Brechbiel, J.A. Frank, First noncovalently bound calix[4]arene-Gd-III-albumin complex, Angew. Chem. Int. Ed. 39 (2000) 1641e1643.
[25]M. Glaser, E. Arstad, “Click labeling” with 2-[18F]fluoroethylazide for positron emission tomography, Bioconjugate Chem. 18 (2007) 989e993.
[26]P.A. Scully, T.M. Hamilton, J.L. Bennett, Synthesis of 2-alkyl- and 2-carboxy-p- tert-butylcalix[4]arenes via the lithiation of tetramethoxy-p-tert-butylcalix[4]
arene, Org. Lett. 3 (2001) 2741e2744.
[27]M.P. Hertel, A.C. Behrle, S.A. Williams, J.A.R. Schmidt, J.L. Fantini, Synthesis of amine, halide, and pyridinium terminated 2-alkyl-p-tert-butylcalix[4]arenes, Tetrahedron 65 (2009) 8657e8667.
[28]F. Arnaudneu, E.M. Collins, M. Deasy, G. Ferguson, S.J. Harris, B. Kaitner, A.J. Lough, M.A. Mckervey, E. Marques, B.L. Ruhl, M.J. Schwingweill, E.M. Seward, Synthesis, X-ray crystal-structures, and cation-binding proper- ties of alkyl calixaryl esters and ketones, a new family of macrocyclic mo- lecular receptors, J. Am. Chem. Soc. 111 (1989) 8681e8691.
[29]E.H. Ryu, Y. Zhao, Effi cient synthesis of water-soluble calixarenes using click chemistry, Org. Lett. 7 (2005) 1035e1037.
[30]V.V. Rostovtsev, L.G. Green, V.V. Fokin, K.B. Sharpless, A stepwise Huisgen cycloaddition process: copper(I)-catalyzed regioselective “ligation” of azides and terminal alkynes, Angew. Chem. Int. Ed. 41 (2002) 2596e2599.
[31]S. Cecioni, R. Lalor, B. Blanchard, J.P. Praly, A. Imberty, S.E. Matthews, S. Vidal, Achieving high affinity towards a bacterial lectin through multivalent topo- logical isomers of calix[4]arene glycoconjugates, Chem. Eur. J. 15 (2009) 13232e13240.
[32]Z.P. Demko, K.B. Sharpless, An intramolecular [2þ3] cycloaddition route to fused 5-heterosubstituted tetrazoles, Org. Lett. 3 (2001) 4091e4094.
[33]D. Kobus, Y. Giesen, R. Ullrich, H. Backes, B. Neumaier, A fully automated two- step synthesis of an 18F-labelled tyrosine kinase inhibitor for EGFR kinase activity imaging in tumors, Appl. Radiat. Isot. 67 (2009) 1977e1984.
[34]H.S. Gill, J.N. Tinianow, A. Ogasawara, J.E. Flores, A.N. Vanderbilt, H. Raab, J.M. Scheer, R. Vandlen, S.P. Williains, J. Marik, A modular platform for the rapid site-specific radiolabeling of proteins with F-18 exemplifi ed by quan- titative positron emission tomography of human epidermal growth factor receptor 2, J. Med. Chem. 52 (2009) 5816e5825.
[35]L. Iddon, J. Leyton, B. Indrevoll, M. Glaser, E.G. Robins, A.J.T. George, A. Cuthbertson, S.K. Luthra, E.O. Aboagye, Synthesis and in vitro evaluation of [18F]fluoroethyl triazole labelled [Tyr3]octreotate analogues using click chemistry, Bioorg. Med. Chem. Lett. 21 (2011) 3122e3127.
[36]W.G. Lewis, F.G. Magallon, V.V. Fokin, M.G. Finn, Discovery and character- ization of catalysts for azide-alkyne cycloaddition by fl uorescence quenching, J. Am. Chem. Soc. 126 (2004) 9152e9153.
[37]L.S. Campbell-Verduyn, L. Mirfeizi, R.A. Dierckx, P.H. Elsinga, B.L. Feringa, Phosphoramidite accelerated copper(I)-catalyzed [3þ2] cycloadditions of
azides and alkynes, Chem. Commun. (2009) 2139e2141.
[38]H.C. Kolb, M.G. Finn, K.B. Sharpless, Click chemistry: diverse chemical function from a few good reactions, Angew. Chem. Int. Ed. 40 (2001) 2004e2021.
[39]F.L. Zhang, J.P. Xue, J.W. Shao, L. Jia, Compilation of 222 drugs’ plasma protein binding data and guidance for study designs, Drug Discov. Today 17 (2012) 475e485.
[40]R. Baynes, J.E. Riviere, Hepatic biotransformation and biliary excretion, in: Comparative Pharmacokinetics e Principles, Techniques, and Applications, Wiley-Blackwell, Oxford, 2011, pp. 113e141.
[41]C.D. Gutsche, B. Dhawan, J.A. Levine, K.H. No, L.J. Bauer, Calixarenes 9- Conformational isomers of the ethers and esters of calix[4]arenes, Tetrahe- dron 39 (1983) 409e426.
[42]R.P. Dings, M.M. Arroyo, N.A. Lockwood, L.I. van Eijk, J.R. Haseman, A.W. Griffi oen, K.H. Mayo, Beta-sheet is the bioactive conformation of the anti-angiogenic anginex peptide, Biochem. J. 373 (2003) 281e288.
[43]R.P. Dings, Y. Yokoyama, S. Ramakrishnan, A.W. Griffi oen, K.H. Mayo, The designed angiostatic peptide anginex synergistically improves chemotherapy and antiangiogenesis therapy with angiostatin, Cancer Res. 63 (2003) 382e385.
[44]R.P.M. Dings, B.W. Williams, C.W. Song, A.W. Griffioen, K.H. Mayo, R.J. Griffi n, Anginex synergizes with radiation therapy to inhibit tumor growth by radi- osensitizing endothelial cells, Int. J. Cancer 115 (2005) 312e319.