Method and molecules
11759526 · 2023-09-19
Assignee
Inventors
- Ronald James Christie (Gaithersburg, MD, US)
- Changshou Gao (Gaithersburg, MD)
- Andre Henri St. Amant (Oakland, CA, US)
- Javier Read de Alaniz (Oakland, CA, US)
Cpc classification
A61K47/6889
HUMAN NECESSITIES
C07C233/52
CHEMISTRY; METALLURGY
A61P29/00
HUMAN NECESSITIES
C07C247/12
CHEMISTRY; METALLURGY
A61K47/6811
HUMAN NECESSITIES
C07D207/46
CHEMISTRY; METALLURGY
C07K1/1077
CHEMISTRY; METALLURGY
C07C271/22
CHEMISTRY; METALLURGY
C07C233/48
CHEMISTRY; METALLURGY
A61K47/6923
HUMAN NECESSITIES
International classification
A61K47/68
HUMAN NECESSITIES
A61K47/69
HUMAN NECESSITIES
C07C271/22
CHEMISTRY; METALLURGY
Abstract
The present invention provides a bioconjugation method and compounds for use therein. The bioconjugation method comprises the step of conjugating a biological molecule containing a first unsaturated functional group with a payload comprising a second unsaturated functional group, wherein the first and second unsaturated functional groups are complementary to each other such that conjugation is a reaction of said functional groups via a Diels-Alder reaction which forms a cyclohexene ring.
Claims
1. An antibody or binding fragment thereof conjugated to a payload, wherein the conjugation reaction linking the antibody or binding fragment thereof to the payload was via a Diels-Alder reaction between a diene and a dieneophile to form a cyclohexene ring, and wherein the resulting conjugate is represented by formula (IVa): ##STR00056## wherein R.sup.Y represents the payload; and R.sup.Z represents the antibody or binding fragment thereof; or wherein the resulting conjugate is represented by formula (IVb): ##STR00057## wherein R.sup.Y represents the payload; R.sup.Z represents the antibody or binding fragment thereof, and X.sup.4 represents —CH.sub.2—.
2. The antibody or binding fragment thereof conjugated to a payload according to claim 1, wherein the payload is selected from: a. an auristatin, b. a maytansinoid, c. a toxin, and d. a polymer.
3. A pharmaceutical composition comprising the antibody or binding fragment thereof conjugated to a payload according to claim 1 and a diluent, carrier and/or excipient.
4. A method of treating a patient comprising administering a therapeutically effective amount of the antibody or binding fragment thereof conjugated to a payload according to claim 1.
5. A method of treating a patient comprising administering a therapeutically effective amount of the pharmaceutical composition according to claim 3.
Description
BRIEF SUMMARY OF THE FIGURES
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
(34)
(35)
(36)
(37)
(38)
(39)
(40)
(41)
(42)
(43)
(44)
(45)
(46)
(47)
(48)
(49)
(50)
(51)
(52)
(53)
(54)
(55)
(56)
(57)
(58)
(59)
(60)
(61)
(62)
(63)
(64)
(65)
(66)
(67)
(68)
(69)
(70)
(71)
(72)
(73)
(74)
(75)
(76)
(77)
(78)
(79)
(80)
EXAMPLES
Example 1. Furan-Maleimide Reaction for Generation of ADCs
(81) The furan-maleimide reaction was evaluated as a conjugation modality to generate ADCs. Furan-NHS was provided by SynChem, Inc. at 90% purity.
(82) Introduction of furan functionality onto mAbs: Furan diene functionality was installed onto IgG1 mAbs by reaction of lysine primary amines with an NHS-ester activated furan linker. This approach resulted in randomly conjugated, amide-linked furan groups with the modified mAb termed mAb-furan linker. Note that mAb-furan-linker may be denoted as mAb-furan in certain figures, see figure caption for clarification. Mab solution was adjusted to 5 mg/mL (3 mL, 15 mg mAb, 0.1 μmol, 1 eq.) with PBS pH 7.2 followed by addition of 10% v/v 1 M NaHCO.sub.3. This solution was chilled on ice and 30 μL furan-NHS (10 mM stock in DMAc, 0.3 μmol, 3 eq.) was added. The reaction proceeded on ice for 5 minutes and then room temperature for 1 h with continuous mixing. Reaction progress was monitored by mass spectrometry and furan-NHS was added in 30 μL portions until a degree of conjugation of ˜2 furans/mAb was achieved. In total, 3 additions of furan-NHS were performed with a total volume of 90 μL (0.9 μmol, 9 eq) added. Reacted mAb was purified by dialysis (Slide-A-Lyzer, 10 kDa MWCO) against PBS, 1 mM EDTA, pH 7.4, 0° C. for 24 h.
(83) ##STR00024##
(84) Reaction of furan-modified mAb with maleimido-MMAEs: MMAE ADC payloads were installed onto mAb-furan-linker through Diels-Alder 4+2 cycloaddition coupling of furan groups to either alkyl- or phenyl-maleimide groups contained on MMAE. First, mAb-furan-linker solution (286 μL, 3.5 mg/mL, 6.7 nmol, 1 eq) was combined with 10% v/v NaH.sub.2PO.sub.4 and 20% v/v DMSO. Next, AM-MMAE or PM-MMAE solution (10 μL of a 10 mM stock solution in DMAc, 100 nmol, 15 eq.) was added to the antibody solution. The reaction mixture was capped under ambient atmosphere and the reaction proceeded at 37° C. for 20 h with mixing. After the 20 h reaction period was complete N-acetyl cysteine (8 μL of a 100 mM solution, 8 equivalents) was added and the solution was further incubated at room temperature for 15 minutes to quench maleimide groups. After quenching, conjugates were purified using PD Spintrap G-25 devices (GE Healthcare Life Sciences) prior to analysis by deglycosylated mass spectrometry as described below. Alloc-lysine was reacted with furan-linker modified mAb as described above using a 200 mM stock solution in 75 mM NaOH (10 PL, 2 μmol, 300 equiv.).
(85) ##STR00025##
(86) Mass spectrometry analysis: First, mAbs or mAb conjugates were deglycosylated with EndoS (New England BioLabs) by combining 50 μL sample (1 mg/mL mAb) with 5 μL glyco buffer 1 (New England BioLabs) and 5 μL Remove-iT EndoS (1:10 dilution in PBS, 20,000 units/mL, New England BioLabs) followed by incubation for 1 h at 37° C. Reduced samples were prepared by addition of 5 μL Bond-Breaker TCEP solution (0.5 M, Thermo Fisher Scientific) and incubation for 10 min at 37° C. Mass spectrometry analysis was performed using an Agilent 6520B Q-TOF mass spectrometer equipped with a RP-HPLC column (ZORBAX 300 Diphenyl RRHD, 1.8 micron, 2.1 mm×50 mm). High-performance liquid chromatography (HPLC) parameters were as follows: flow rate, 0.5 ml/min; mobile phase A was 0.1% (v/v) formic acid in HPLC-grade H.sub.2O, and mobile phase B was 0.1% (v/v) formic acid in acetonitrile. The column was equilibrated in 90% A/10% B, which was also used to desalt the mAb samples, followed by elution in 20% A/80% B. Mass spec data were collected for 100-3000 m/z, positive polarity, a gas temperature of 350° C., a nebulizer pressure of 48 lb/in.sup.2, and a capillary voltage of 5,000 V. Data were analyzed using vendor-supplied (Agilent v.B.04.00) MassHunter Qualitative Analysis software and peak intensities from deconvoluted spectra were used to derive the relative proportion of species in each sample.
(87)
(88)
(89) TABLE-US-00001 TABLE 1.1 Summary of mAb-furan-linker reactions at 37° C. for 20 hours total non-specific specific [mAb] furan/ conjugation conjugation conjugation to payload Equiv*. pH mg/mL mAb to mAb (%) to mAb (%) furan (%) AM- 15 5.5 3.5 2.5 2.4 1.1 1.3 MMAE PM- 15 5.5 3.5 2.5 16.6 12.0 4.0 MMAE alloc- 300 5.5 3.5 2.5 0 0 0 lysine *(rel to mAb)
(90)
(91) Introduction of furan functionality into an antibody was achieved using an amine-reactive furan-NHS molecule. The degree of furan functionality was controlled by the amount of furan-NHS used in the mAb modification reaction. More or less furan could be achieved by adjusting the molar feed ratio accordingly. Reaction of mAb-furan-linker with maleimido-MMAEs was inefficient and non-specific. Neither alkyl- or phenyl-maleimide payloads achieved over 5% specific conjugation, even after 20 h reaction time at 37° C. Non-specific reaction to mAb (presumably through Michael-addition to amines) was 12 times higher for PM-MMAE compared to AM-MMAE, indicating the higher reactivity of this maleimide group. Furthermore, non-specific reaction (presumably to amines) appeared to be higher than specific reaction to furans by ˜4-fold for PM-maleimide MMAE payload. Furan-maleimide coupling is not amenable for production of ADCs.
Example 2. Synthesis of Cyclopentadiene (CP1)-Containing Compounds
(92) Crosslinkers and non-natural amino acids (NNAAs) were prepared based on the general design shown in
(93) Materials and Methods: Unless stated otherwise, reactions were conducted under an atmosphere of N.sub.2 using reagent grade solvents. DCM, and toluene were stored over 3 Å molecular sieves. THF was passed over a column of activated alumina. All commercially obtained reagents were used as received. Thin-layer chromatography (TLC) was conducted with E. Merck silica gel 60 F254 pre-coated plates (0.25 mm) and visualized by exposure to UV light (254 nm) or stained with p-anisaldehyde, ninhydrin, or potassium permanganate. Flash column chromatography was performed using normal phase silica gel (60 Å, 0.040-0.063 mm, Geduran). .sup.1H NMR spectra were recorded on Varian spectrometers (400, 500, or 600 MHz) and are reported relative to deuterated solvent signals. Data for .sup.1H NMR spectra are reported as follows: chemical shift (6 ppm), multiplicity, coupling constant (Hz) and integration. .sup.13C NMR spectra were recorded on Varian Spectrometers (100, 125, or 150 MHz). Data for .sup.13C NMR spectra are reported in terms of chemical shift (6 ppm). Mass spectra were obtained from the UC Santa Barbara Mass Spectrometry Facility on a (Waters Corp.) GCT Premier high-resolution time-of-flight mass spectrometer with a field desorption (FD) source.
Synthesis of CP1-NNAA (4)
(94) ##STR00026##
2-(Cyclopentadienyl)ethanol Isomers (1)
(95) Methyl bromoacetate (6.0 mL, 63 mmol, 1.05 eq) was added to THF (60 mL) and cooled to −78° C. Sodium cyclopentadienide (2 M solution in THF, 30 mL, 60 mmol, 1 eq) was added dropwise over 10 min. The reaction was stirred at −78° C. for 2 hr. The reaction was quenched with H.sub.2O (6 mL) and silica gel (6 g) and allowed to warm to rt. The reaction mixture was filtered through a plug of silica then rinsed with DCM (100 mL). The organic layers were combined and the solvent removed to yield methyl 2-(cyclopentadienyl)acetate isomers (1:1) as a brown oil, which was used directly in the next step.
(96) LAH (4.55 g, 120. mmol, 2 eq) was added to THF (300 mL) and cooled to 0° C. Crude methyl 2-(cyclopentadienyl)acetate (60 mmol) dissolved in THF (10 mL) was added dropwise in 4 portions over 1 hr. The reaction was allowed to warm to rt and stirred until consumption of starting material (TLC, 2 hr). The reaction was cooled to 0° C. and slowly quenched with H.sub.2O (10 mL) dropwise then NaOH (4 M in H.sub.2O, 5 mL). H.sub.2O (20 mL) was added, the mixture filtered, and rinsed with Et.sub.2O (100 mL). The filtrates were combined and the solvent removed. The residue was suspended in brine (100 mL) and extracted with Et.sub.2O (3×100 mL). The organic layers were combined, washed with brine (100 mL), dried over MgSO.sub.4, filtered, and the solvent removed. The residue was filtered through a plug of silica (Hexane:EtOAc, 2:1) and the solvent removed to yield 1 (5.45 g, 83%) as an amber oil. To prevent dimerization, 1 should be stored frozen in a benzene matrix.
(97) Rf (Hexane:EtOAc, 4:1): 0.11; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 6.50-6.13 (m, 3H), 3.81 (td, J=6.3, 10.1 Hz, 2H), 3.01 (d, J=1.6 Hz, 1H), 2.95 (d, J=1.6 Hz, 1H), 2.70 (dt, J=1.2, 6.5 Hz, 1H), 2.66 (dt, J=1.4, 6.4 Hz, 1H), 1.52 (s, 1H) ppm.
(98) ##STR00027##
2-(Cyclopentadienyl)ethyl 4-nitrophenyl carbonate Isomers (2)
(99) 2 (2.86 g, 26.0 mmol, 1 eq) was added to DCM (100 mL) and cooled to 0° C. Pyridine (5.2 mL, 65 mmol, 2.5 eq) was added followed by 4-nitrophenyl chloroformate (5.76 g, 28.6 mmol, 1.1 eq) in 2 portions over 10 min. The ice bath was removed and the reaction was stirred until consumption of the starting material (TLC, 40 min). The reaction was poured into a separatory funnel and washed with a saturated NH4Cl in H.sub.2O (100 mL). The aqueous layer was extracted with DCM (100 mL). The organic layers were combined, washed with brine (50 mL), dried over Na.sub.2SO.sub.4, filtered, and the solvent removed. The residue was subjected to flash column chromatography (Hexane:EtOAc, 6:1) to yield 2 (5.69 g, 80%) as a yellow oil that solidifies in the freezer.
(100) Rf (Hexane:EtOAc, 4:1): 0.43; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 8.34-8.24 (m, 2H), 7.40-7.34 (m, 2H), 6.56-6.13 (m, 3H), 4.47 (td, J=6.8, 10.2 Hz, 2H), 3.02 (d, J=0.8 Hz, 1H), 2.98 (d, J=1.2 Hz, 1H), 2.88 (dtd, J=1.0, 6.9, 16.1 Hz, 2H) ppm.
(101) ##STR00028##
Fmoc-Lys(2-(cyclopentadienyl)ethyl formate)-OH Isomers (3)
(102) 2 (3.60 g, 13.1 mmol, 1 eq) was added to DMF (30 mL), followed by Fmoc-Lys-OH (5.78 g, 15.7 mmol, 1.2 eq) and DIPEA (5.4 mL, 32 mmol, 2.4 eq). The reaction was stirred until consumption of the starting material (NMR, 3.5 hr), then poured into EtOAc (100 mL) and H.sub.2O (140 mL). The aqueous layer was acidified with HCl (1 M, 60 mL), poured a separatory funnel, and the layers separated. The aqueous layer was extracted with EtOAc (2×100 mL). The organic layers were combined, washed with brine (100 mL), dried over Na.sub.2SO.sub.4, filtered, and the solvent removed. The residue was subjected to flash column chromatography (Hexane:EtOAc, 3:1 then DCM:MeOH:AcOH, 89:10:1) to yield 3 (4.73 g, 72%) as a white foam.
(103) Rf (DCM:MeOH:AcOH, 89:10:1): 0.50; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 7.76 (d, J=7.4 Hz, 2H), 7.66-7.56 (m, 2H), 7.39 (t, J=7.4 Hz, 2H), 7.31 (t, J=7.2 Hz, 2H), 6.57-5.96 (m, 3H), 5.85-5.54 (m, 1H), 4.84-4.11 (m, 7H), 3.27-2.61 (m, 6H), 1.99-1.11 (m, 6H) ppm.
(104) ##STR00029##
CP1-NNAA (4)
(105) 3 (4.61 g, 9.13 mmol, 1 eq) was added to DMF (130 mL), followed by piperidine (14.4 mL). The reaction was stirred until consumption of the starting material (TLC, 90 min), then the solvent was removed. Et.sub.2O (100 mL) was added to the residue, and the suspension was sonicated for 5 min. The suspension was filtered and rinsed with Et.sub.2O (100 mL). The solid was suspended in MeOH (10 mL), stirred for 10 min, Et.sub.2O (40 mL) was added, the suspension filtered and rinsed with Et.sub.20 (50 mL). The compound was dried under vacuum to yield 4 (2.15 g, 84%) as a tan powder.
(106) .sup.1H NMR (400 MHz, Methanol-d4+one drop TFA) δ 6.53-6.07 (m, 3H), 4.29-4.11 (m, 2H), 3.96 (t, J=6.3 Hz, 1H), 3.11 (t, J=1.0 Hz, 2H), 3.01-2.62 (m, 3H), 2.02-1.81 (m, 2H), 1.62-1.35 (m, 4H) ppm; MS (FD) Exact mass cald. for C.sub.14H.sub.22N.sub.2O.sub.4 [M+H].sup.+: 283.17, found: 283.19.
Synthesis of CP1-NHS (6)
(107) ##STR00030##
4-(2-(Cyclopentadienyl)ethoxy)-4-oxobutanoic acid Isomers (5)
(108) DCM (1.5 mL) was added to a vial containing 1 (0.33 g, 3.0 mmol, 1 eq). Et.sub.3N (0.42 mL, 3.0 mmol, 1 eq), DMAP (37 mg, 0.30 mmol, 0.1 eq) and succinic anhydride (0.33 g, 3.3 mmol, 1.1 eq) were added, the reaction capped under an atmosphere of air, and stirred at rt until consumption of the starting material (TLC, 60 min). The reaction mixture was poured into a separatory funnel with DCM (50 mL) and extracted with aqueous HCl (1 M, 50 mL) then H.sub.2O (50 mL). The organic layer was dried over MgSO.sub.4, filtered, and the solvent removed to yield 5 (0.57 g, 90%) as a tan powder.
(109) Rf (EtOAc): 0.67; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 11.49 (br. s., 1H), 6.49-6.05 (m, 3H), 4.27 (td, J=7.0, 9.0 Hz, 2H), 2.94 (d, J=17.6 Hz, 2H), 2.80-2.56 (m, 6H) ppm.
(110) ##STR00031##
CP1-NHS (6)
(111) THF (10 mL) was added to a vial containing 5 (0.42 g, 2.0 mmol, 1 eq). NHS (0.32 g, 2.8 mmol, 1.4 eq), EDC-HCl (0.46 g, 2.4 mmol, 1.2 eq) and DCM (5 mL) were added, the reaction capped under an atmosphere of air, and stirred at rt overnight. The solvent was removed and the residue was subjected to flash column chromatography (Hexane:EtOAc, 1:1) to yield 6 (0.48 g, 78%) as a clear, viscous oil. CP1-NHS is referred to as CP1-linker after conjugation to antibodies.
(112) Rf (Hexane:EtOAc, 1:1): 0.38; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 6.49-6.40 (m, 3H), 6.31 (dd, J=1.2, 5.1 Hz, 1H), 6.25 (td, J=1.5, 2.8 Hz, 1H), 6.11 (td, J=1.8, 3.0 Hz, 1H), 4.30 (td, J=7.0, 9.0 Hz, 4H), 3.00-2.90 (m, 8H), 2.85 (br. s., 8H), 2.80-2.68 (m, 8H); .sup.13C NMR (125 MHz, CDCl.sub.3) δ 170.8, 170.8, 168.9, 168.8, 167.6, 167.6, 144.3, 142.3, 134.2, 134.1, 132.3, 131.4, 128.4, 128.0, 64.5, 64.2, 43.5, 41.4, 29.7, 29.0, 28.7, 26.2, 25.5 ppm.
(113) The synthesis of cyclopentadiene (CP) functionalized non-natural amino acid (NNAA) began with the reaction of commercially available NaCp with methyl bromoacetate. The crude ester was reduced with LAH to alcohol 1, which was obtained as a mixture of isomers (˜1:1). 1 will dimerize when stored at −20° C., it should be stored frozen in a matrix of benzene or used immediately. The reaction of 1 with 4-nitrophenyl chloroformate produced activated carbamate 2, which can be stored for several weeks at −20° C. The reaction of 2 with copper lysinate was attempted, but isolation of NNAA 4 after treatment with 8-hydroxyquinoline or EDTA was not fruitful. The reaction of 2 with Boc-Lys-OH provided the Boc-protected NNAA in 71% (or directly from 1 using triphosgene in 38%) but efforts to remove the Boc group using TFA, formic acid, or Lewis acid lead to rapid decomposition of the CP ring. Reacting 2 with Fmoc-Lys-OH produces the Fmoc-protected 3, which could be deprotected using piperidine to obtain NNAA 4. Compound 4 has poor solubility in commonly used deuterated solvents. A drop of TFA can be added to increase solubility, but leads to decomposition after several hours. The CP protons in 4 exchange when dissolved in D.sub.2O with catalytic NaOH due to sequential [1,5]-hydride shifts.
(114) The synthesis of a CP1-functionalized NHS-ester began with the reaction of 1 with succinic anhydride to produce acid 5. The acid 5 was reacted with EDC-HCl and N-hydroxysuccinimide to yield NHS ester 6. At room temperature the CP ring on 6 will dimerize over several days, but it can be stored for over a month at −20° C.
Example 3. CP1 Diene-Maleimide Conjugation for Preparation of ADCs Via Crosslinker-Modified mAb
(115) Cyclopentadiene-maleimide reactions were evaluated for bioconjugation, where cyclopentadiene groups were introduced via a linker.
(116) Introduction of CP1 functionality onto mAbs: CP1 diene functionality was installed onto IgG1 mAbs by reaction of lysine primary amines with CP1-NHS (compound 6). This approach resulted in randomly conjugated, amide-linked cyclopentadiene groups, with the modified mAb termed mAb-CP1-linker. Note that mAb-CP1-linker may also be referred to as mAb-CP1 in some figures, see figure caption for clarification. A typical mAb modification reaction is described as follows. Mab solution was adjusted to 5 mg/mL (3 mL, 15 mg mAb, 100 nmol, 1 eq.) with PBS pH 7.2 followed by addition of 10% v/v 1M NaHCO.sub.3. This solution was chilled on ice and 30 μL CP1-NHS (10 mM stock in DMAc, 300 nmol, 3 equivalents) was added. The reaction proceeded on ice for 5 minutes followed by reaction at room temperature for 1 h with continuous mixing. Reacted mAb was purified by dialysis (Slide-A-Lyzer, 10 kDa MWCO) against PBS, 1 mM EDTA, pH 7.4, 0° C. for 24 h. CP1-linker introduction was quantified by intact deglycosylated mass spectrometry as described below and found to be 2.3 CP1 per mAb in this example, which corresponds to 77% conversion of CP1-NHS to antibody conjugate. A summary of results for this reaction performed under various conditions is described in appendix A3.1.
(117) ##STR00032##
(118) Reaction of CP1-modified mAb with maleimido-MMAEs: mAb-CP1-linker (2.3 CP1 diene/mAb, 1 mg, 6.7 nmol mAb, 1 equivalent) was diluted with PBS (pH 7.4) to a final concentration of 3.5 mg/mL. Next, DMSO was added to yield a 20% v/v solution followed by addition of 1 M sodium phosphate monobasic to yield a 10% v/v solution. Addition of all solution components yielded a mixture comprising 2.7 mg/mL mAb, 41.4 μM CP1 diene, 1.78 M DMSO, 110 mM sodium phosphate, 100 mM NaCl, pH 5.5. AM-MMAE or PM-MMAE (10 μL of a 10 mM stock solution in DMAc, 100 nmol, 15 equivalents) was added to the antibody solution. The reaction mixture was vortexed briefly and allowed to proceed at 22° C. or 37° C. with mixing. After 4 h reaction, N-acetyl cysteine (8 μL of a 100 mM solution, 120 equivalents) was added to quench unreacted maleimide groups. Samples were then purified using PD Spintrap G-25 devices (GE Healthcare Life Sciences) to remove small molecule components from the mixture. Samples were then analyzed by reduced deglycosylated mass spectrometry as described below.
(119) ##STR00033##
(120) Mass spectrometry analysis: Samples were analyzed as described in Example 1.
(121)
(122) TABLE-US-00002 TABLE 3.1 Summary of CP1-NHS mAb reactions equivalents CP1-NHS CP1-linker conversion (rel to mAb) [mAb] mg/nL per mAb (%) 3 5 2.3 77 5.4 3.75 3.9 74 4 5 3.7 93
(123)
(124)
(125) TABLE-US-00003 TABLE 3.2 Summary of mAb-CP1-linkermaleimido-MMAE reactions.sup.a Equivalents MMAE conjugation to Payload (rel to mAb) pH temp mAb-CP1-linker (%) AM-MMAE 15 5.5 37° C. 100 22° C. 100 PM-MMAE 15 5.5 37° C. 100 22° C. 100 alloc-lysine 300 5.5 37° C. 0 22° C. 0 .sup.aAll reactions performed at 2.7 mg/mL mAb-CP1-linker for 4 h.
(126)
(127) CP1 diene groups installed onto the surface of antibodies completely reacted with maleimido-MMAE prodrugs within 4 h at room temperature. No non-specific conjugation was observed by mass spectrometry, as all CP1-linker-MMAE conjugate peaks tracked from mAb-CP1-linker peaks, not unmodified mAb peaks (i.e. lacking CP1-linker). This is in stark contrast to reactions of mAb-furan-linker with maleimido MMAE's, where only ˜2-20% conjugation was observed including non-specific conjugation after 20 h at 37° C. Diene-maleimide conjugation with mAb-CP1-linker is an improvement over mAb-furan-linker based coupling.
Example 4. Kinetics of mAb-CP1-Linker Conjugation to Maleimido-MMAEs at 0.6 Molar Equivalents Maleimido-MMAE to Diene Groups Contained on CP1-mAb-Linker
(128) Reaction kinetics of mAb-CP1-linker with maleimido MMAEs was investigated at the stiociometry of 0.6 molar equivalents of maleimido-MMAE to diene contained on mAb-CP1-linker.
(129) Introduction of CP1 functionality onto mAbs: CP1 diene functionality was installed onto IgG1 mAbs by reaction of lysine primary amines with CP1-NHS (compound 6). This approach resulted in randomly conjugated, amide-linked cyclopentadiene groups. The resulting conjugate was termed mAb-CP1-linker and may also be referred to as mAb-CP1, see figure captions for clarification. Mab solution was adjusted to 3.7 mg/mL (3 mL, 11.1 mg mAb, 74 nmol, 1 eq.) with PBS pH 7.2 followed by addition of 10% v/v 1M NaHCO.sub.3. This solution was chilled on ice and 40 μL CP1-NHS (10 mM stock in DMAc, 400 nmol, 5.4 equivalents) was added. The reaction proceeded at room temperature for 1 h with continuous mixing. Reacted mAb was purified by dialysis (Slide-A-Lyzer, 10 kDa MWCO) against PBS, 1 mM EDTA, pH 7.4, 0° C. for 24 h. CP1 diene introduction was quantified by intact deglycosylated mass spectrometry as described below and found to be 3.99 CP1 dienes per mAb, which corresponds to 74% conversion of CP1-NHS to antibody conjugate.
(130) Reaction of mAb-CP1-linker with maleimido-MMAEs: CP1-modified mAb (3.99 CP1 diene/mAb, 3 mg, 80 nmol CP1 diene, 1 equivalent) was diluted with PBS (pH 7.4) to a final concentration of 1.7 mg/mL. Next, DMSO was added to yield a 20% v/v solution followed by addition of 1 M sodium phosphate monobasic to yield a 10% v/v solution. Addition of all solution components yielded a mixture comprising 1.3 mg/mL mAb, 34.6 μM CP1 diene, 1.78 M DMSO, 110 mM sodium phosphate, 100 mM NaCl, pH 5.5. AM-MMAE or PM-MMAE (5.2 μL of a 10 mM stock solution in DMSO, 52 nmol, 0.67 equivalents) was added to the antibody solution. The reaction mixture was vortexed briefly and allowed to proceed at 22° C. with mixing. Aliquots (180 μL) were removed at various timepoints and N-acetyl cysteine (2 μL of a 100 mM solution, 38 equivalents) was added to quench maleimide groups. Samples were then purified using PD Spintrap G-25 devices (GE Healthcare Life Sciences) to remove small molecule components from the mixture. Samples were then analyzed by reduced deglycosylated mass spectrometry as described below.
(131) Mass spectrometry analysis: Samples were analyzed as described in Example 1.
(132) Calculation of CP1 diene-maleimide reaction rate constants: Second order rate constants for reaction of maleimido-MMAEs with antibody dienes were determined from peak intensities in deglycosylated, reduced mass spectra. Reaction progress was monitored by both disappearance of mAb-CP1-linker peaks and appearance of mAb-CP1-linker-AM MMAE peaks, but only mAb-CP1-linker peak intensities on the antibody heavy chains were used to calculate relative abundance of reacted CP1 diene. The relative amount of unreacted CP1 diene groups on mAb heavy chains was calculated using the equation below:
(133)
a=peak intensity of unmodified heavy chain
b=sum of peak intensities of heavy chains with one CP1-linker group
c=sum of peak intensities of heavy chains with two CP1-linker groups
d=peak intensity of heavy chain with three CP1-linker groups
Note: maleimido-MMAE-containing heavy chains were also included in the calculation. For example, CP1-linker-2+1 maleimido-MMAE would be included as a CP1-linker-1 species.
(134) Results
(135)
(136)
(137)
(138) TABLE-US-00004 TABLE 4.1 Summary of mAb-CP1-linker conjugation results.sup.a payload feed.sup.b reacted diene.sup.c AM-MMAE 0.65 60.0% PM-MMAE 0.65 63.4% .sup.aall conjugation reactions performed at pH 5.5, 20% DMSO, 22° C. and 1.3 mg/mL mAb for 3.5 h .sup.bfeed calculated as molar equivalent maleimido-MMAE:CP1 diene .sup.ccalculated from peak intensities of reduced deglycosylated mass spectra.
(139) Reaction of CP1 dienes (contained on mAb-CP1-linker) with maleimido-MMAEs was rapid and specific under aqueous conditions, with complete reaction achieved on the order of 10's of minutes. Calculated molar feed ratios of maleimido-MMAE based match the observed degree of conjugation from intact mass spectra, as MMAE molar feed and conversion of diene to conjugate were essentially the same.
Example 5. Kinetics of mAb-CP1-Linker Conjugation to Maleimido-MMAEs at 1.0 Molar Equivalent Maleimido-MMAE to Diene Groups
(140) Reaction kinetics of CP1 dienes with maleimido MMAEs at 22° C. was evaluated.
(141) Introduction of CP1 functionality onto mAbs: CP1 diene functionality was installed onto IgG1 mAbs by reaction of lysine primary amines with CP1-NHS (compound 6). This approach resulted in randomly conjugated, amide-linked cyclopentadiene groups. Mab solution was adjusted to 5 mg/mL (3 mL, 5 mg mAb, 100 nmol, 1 eq.) with PBS pH 7.2 followed by addition of 10% v/v 1 M NaHCO.sub.3. This solution was chilled on ice and 40 μL CP1-NHS (10 mM stock in DMAc, 400 nmol, 4 equivalents) was added. The reaction mixture was vortexed briefly and incubated at room temperature for 1 h with continuous mixing. Reacted mAb was purified by dialysis (Slide-A-Lyzer, 10 kDa MWCO) against PBS, 1 mM EDTA, pH 7.4, 0° C. for 24 h. CP1 diene introduction was quantified by intact deglycosylated mass spectrometry as described below and found to be 3.7 CP1 dienes per mAb, which corresponds to 92% conversion of CP1-NHS to antibody conjugate.
(142) Reaction of mAb-CP1-linker with maleimido-MMAEs: mAb-CP1-linker (3.7 CP1 diene/mAb, 3 mg, 74 nmol CP1 diene, 1 equivalent) was diluted with PBS (pH 7.4) to a final concentration of 1.7 mg/mL. Next, DMSO was added to yield a 20% v/v solution followed by addition of 1 M sodium phosphate monobasic to yield a 10% v/v solution. Addition of all solution components yielded a mixture comprising 1.3 mg/mL mAb, 32.3 μM CP1 diene, 1.78 M DMSO, 110 mM sodium phosphate, 100 mM NaCl, pH 5.5. AM-MMAE or PM-MMAE (7.4 μL of a 10 mM stock solution in DMSO, 74 nmol, 1 equivalent) was added to the antibody solution. The reaction mixture was vortexed briefly and allowed to proceed at 22° C. with mixing. Aliquots (180 μL) were removed at various timepoints and N-acetyl cysteine (3 μL of a 100 mM solution, 51 equivalents) was added to quench unreacted maleimide groups. Samples were then purified using PD Spintrap G-25 devices (GE Healthcare Life Sciences) to remove small molecule components from the mixture. Samples were then analyzed by reduced deglycosylated mass spectrometry as described below.
(143) Mass spectrometry analysis: Samples were analyzed as described in Example 1.
(144) Calculation of CP1 diene-maleimide reaction rate constants: Second order rate constants for reaction of maleimido-MMAE with mAb dienes were determined from peak intensities in deglycosylated, reduced mass spectra. Reaction progress was monitored by both disappearance of mAb-CP1-linker peaks and appearance of mAb-CP1-linker-maleimido-MMAE peaks, but only mAb-CP1-linker peak intensities on the antibody heavy chains were used to calculate relative abundance of reacted CP1 dienes. Unreacted CP1 diene groups on mAb-CP1-linker was calculated using the equation below:
(145)
a=peak intensity of unmodified heavy chain
b=sum of peak intensities of heavy chains with one CP1-linker group
c=sum of peak intensities of heavy chains with two CP1-linker groups
d=peak intensity of heavy chain with three CP1-linker groups
e=peak intensity of unmodified light chain
f=sum of peak intensities of light chains with one CP1-linker group
g=sum of peak intensities of light chains with two CP1-linker groups
(146) Conjugation data were further analyzed in units of molar concentration to determine kinetic constants. Second order rate constants were determined from the slopes of curves generated from plotting 1/[CP1] versus time and linear regression analysis. Reaction half-lives were calculated from second-order reaction rate constants using the equation shown below:
(147)
k.sub.2=second order rate constant
[CP1].sub.0=CP1 diene concentration at time=0
(148)
(149)
(150)
(151) TABLE-US-00005 TABLE 5.1 Summary of diene-maleimide coupling kinetics for reaction of mAb-CP1-linker with maleimido MMAEs.sup.a,b,c 2.sup.nd order rate t.sub.1/2 conversation payload constant (M.sup.−1s.sup.−1) (min) (%).sup.d AM-MMAE 36 ± 1.4 13.5 90 PM-MMAE 54 ± 1.2 8.9 97 .sup.aall conjugation reactions performed at pH 5.5, 20% DMSO, 22° C. and 1.3 mg/mL CP1-modified mAb .sup.bthe molar ratio of MMAE:CP1 diene used was 1:1 .sup.ccalculated from peak intensities of reduced deglycosylated mass spectra .sup.dafter 150 minutes reaction
(152) Reaction of CP1 dienes with maleimido-MMAEs was rapid and specific, with half-live's on the order of several minutes. Reaction conversion was 90% or more for both maleimido-MMAE. Phenyl maleimide reaction rates were slightly higher than alkyl maleimide rates, however, rates and final conversion were comparable between the two types of maleimides.
Example 6. CP1-NNAA Incorporation into an Antibody
(153) CP1-NNAA was incorporated into position K274 of an antibody, the quality of expressed mAb, and reactivity of CP1 diene after antibody incorporation was assessed.
(154) Preparation of CP1-NNAA stock solution: CP1-NNAA (0.5 g, 1.77 mmol) was combined with 6.81 mL H.sub.2O and 1.38 mL 1 M NaOH. The resulting slurry was stirred at room temperature until all solids dissolved (10 minutes). After complete dissolution the light yellow solution was passed through a 0.2 μm filter, aliquoted, and stored at −80° C. until use. This procedure resulted in 8.2 mL of 216 mM CP1 and 168 mM NaOH stock solution.
(155) Antibody expression: 12G3H11 or 1C1 IgG1 antibody genes with an amber mutation at Fe position K274 or S239 were cloned into a proprietary pOE antibody expression vector. The construct was transfected into CHO-G22 by PEImax (1.5 L of G22 cells), along with a plasmid encoding PyIRS double mutant (Y306A/Y384F) or wild-type PyIRS and a plasmid containing tandem repeats of the tRNA expression cassette (pORIP 9×tRNA). Four hours post transfection, 3.3% of feed F9 (proprietary) and 0.2% of feed F10 (proprietary) were added to cells and the cells were further incubated at 34° C. CP1-NNAA was added the next day at final concentration of 0.26 mM for 1C1.K274 transfected cells. Cells were fed again on day 3 and day 7 with 6.6% of feed F9 and 0.4% of feed F10. Cells were spun down and supernatant was harvested on day 11. The supernatant was purified by IgSelect affinity column (GE Health Care Life Science). The antibody was eluted with 50 mM glycine, 30 mM NaCl, pH 3.5 elution buffer, neutralized with 1 M Tris buffer pH 7.5, and dialyzed into PBS, pH 7.2. Concentration of antibody eluted was determined by absorbance measurement at 280 nm. The back calculated titer was 47 mg/L for 1C1.K274.12G3H11 mAb was expressed in a similar manner at smaller scale, with CP1-NNAA feed concentration and feeding times varied. Recovered antibody was analyzed by SDS-PAGE using standard methods. Antibody was also analyzed by size exclusion chromatography and mass spectrometry as described below. Antibodies incorporating CP1-NNAA are denoted as mAb-CP1-NNAA to distinguish them from mAb-CP1-linker constructs, or mAb-[position]CP1-NNAA where [position] indicates the amino acid number and amino acid symbol that was mutated to CP1-NNAA.
(156) Size exclusion chromatography (SEC): SEC analysis was performed using an Agilent 1100 Capillary LC system equipped with a triple detector array (Viscotek 301, Viscotek, Houson, Tex.); the wavelength was set to 280 nm, and samples were run on a TSK-GEL G3000SWXL column (Toso Bioscience LLC, Montgomeryville, Pa.) using 100 mM sodium phosphate buffer, pH 6.8 at a flow rate of 1 mL/min.
(157) Conjugation of mAb-CP1-NNAA with maleimido MMAEs: mAb-CP1-NNAA (0.4 mg, 2.7 nmol, 1 equivalent) was adjusted to 3 mg/mL with PBS (0.133 mL). DMSO (27 μL) and 1 M sodium phosphate, monobasic (13 μL) was added to yield ˜20% and 10% v/v solution, respectively. Maleimido-MMAE (5 μL of 10 mM stock in DMSO, 13 nmol, 5 equivalents) was added to mAb-CP1-NNAA solution and the mixture was vortexed briefly. ADCs were prepared with both AM-MMAE and PM-MMAE. The reaction proceeded at room temperature for 1 h with continuous mixing. N-acetyl cysteine (1.1 μL of 100 mM, 108 nmol, 40 equivalents) was added and the solution was incubated for an additional 15 min to quench unreacted maleimide groups. Samples were then purified using PD Spintrap G-25 devices (GE Healthcare Life Sciences) to remove small molecule components from the mixture. Samples were subsequently analyzed by reduced mass spectrometry as described below.
(158) ##STR00034##
(159) Mass spectrometry analysis: For deglycosylated mAb analysis, EndoS (5 μL Remove-iT EndoS (1:10 dilution in PBS, 20,000 units/mL, New England BioLabs) was combined with 50 μL sample (1 mg/mL mAb) and 5 μL glyco buffer 1 (New England BioLabs) and followed by incubation for 1 h at 37° C. Reduced samples were prepared by addition of 5 μL Bond-Breaker TCEP solution (0.5 M, Thermo Fisher Scientific) and incubation for 10 min at 37° C. Mass spectrometry analysis was performed using an Agilent 6520B Q-TOF mass spectrometer equipped with a RP-HPLC column (ZORBAX 300 Diphenyl RRHD, 1.8 micron, 2.1 mm×50 mm). High-performance liquid chromatography (HPLC) parameters were as follows: flow rate, 0.5 ml/min; mobile phase A was 0.1% (v/v) formic acid in HPLC-grade H2O, and mobile phase B was 0.1% (v/v) formic acid in acetonitrile. The column was equilibrated in 90% A/10% B, which was also used to desalt the mAb samples, followed by elution in 20% A/80% B. Mass spec data were collected for 100-3000 m/z, positive polarity, a gas temperature of 350° C., a nebulizer pressure of 48 lb/in.sup.2, and a capillary voltage of 5,000 V. Data were analyzed using vendor-supplied (Agilent v.B.04.00) MassHunter Qualitative Analysis software and peak intensities from deconvoluted spectra were used to derive the relative proportion of species in each sample as previously described.
(160)
(161) TABLE-US-00006 TABLE 6.1 Summary of 1C1 K274CP1-NNAA mAb production CP1 NNAA feed (mM) 0.26 Volume (L) 1.7 Mass recovered (mg) 67 Titer (mg/L) 39 Monomer (%) 90.8
(162)
(163)
(164)
(165)
(166)
(167)
(168)
(169)
(170)
(171) TABLE-US-00007 TABLE 6.2 Summary of K274CP1-NNAA mAb-MMAE conjugation data.sup.a,b,c Conjugation Observed Calculated efficiency Δ mass Δ mass mAb payload (%) (AMU) (AMU) DAR.sup.d Comments 12G3H11 AM-MMAE 95.7 1315.47 1316.65 1.91 12G3H11 PM-MMAE 95.1 1335.96 1336.64 1.46 linker cleavage observed 1C1 AM-MMAE 100 1317.59 1316.65 2.0 unconjugated species not detected .sup.aall conjugation reactions performed at pH 5.5, 20% DMSO, 22° C. and 3 mg/mL CP1-NNAA mAb. CP1-NNAA was incorporated into position K274 in place of lysine .sup.bthe molar ratio of MMAE:CP1 diene used was 2.5:1 .sup.ccalculated from peak intensities of reduced mass spectra .sup.dDAR = drug to antibody ratio, linker cleaved species not included in DAR calculation
(172) Incorporation of CP1-NNAA into antibodies at position K274 was confirmed using two different antibody constructs; 12G3H1l and 1C1. Recovered antibody was of high quality, with no truncated product and very little aggregate. Titer achieved for the 1.7 L scale production of 1C1 antibody was reasonably high considering the low amount of CP1-NNAA fed to cells. CP1 diene reactivity was preserved throughout the expression and purification process as indicated by the nearly complete conversion of antibody to ADC.
Example 7. Serum Stability of CP1-NNAA mAb Maleimido MMAE Antibody Drug Conjugates
(173) Stability of the 4+2 cycloaddition product (cyclopentadiene-maleimide bond) in physiological milieu ex vivo by incubation in rat and mouse serum for 7 days at 37° C.
(174) Generation of ADCs: 12G3H11 K274CP1-NNAA bearing CP1-NNAA at position K274 was conjugated to maleimido MMAEs to produce the desired ADC. First, 12G3H11 K274CP1-NNAA mAb (0.4 mg, 2.7 nmol, 1 equivalent) was adjusted to 3 mg/mL with PBS (0.133 mL). DMSO (27 μL) and 1M sodium phosphate, monobasic (13 μL) was added to yield ˜20% and 10% v/v solution, respectively. Maleimido-MMAE (5 μL of 10 mM stock in DMSO, 13 nmol, 5 equivalents) was added to 12G3H11 K274CP1-NNAA mAb solution and the mixture was vortexed briefly. The reaction proceeded at room temperature for 1 h with continuous mixing. N-acetyl cysteine (1.1 μL of 100 mM, 108 nmol, 40 equivalents) was added and the solution was incubated for an additional 15 min to quench unreacted maleimide groups. Samples were then purified using PD Spintrap G-25 devices (GE Healthcare Life Sciences) to remove small molecule components from the mixture. Samples were subsequently analyzed by reduced mass spectrometry as described below to confirm conjugation and quantify the drug:antibody ratio.
(175) Serum stability assay: ADC samples were evaluated in whole rat serum (Jackson Immunoresearch cat:012-000-120) and whole mouse serum (Jackson Immunoresearch cat:015-000-120). Lyophilized serum product was reconstituted with sterile water according to the manufacturer's protocol. ADC sample was added to serum to result in a 0.2 mg/mL antibody solution. ADC/serum mixtures were passed through a 0.2 μm filter, capped in an air-tight vial and incubated at 37° C. An aliquot was removed and frozen to serve as a T=0 d reference. Remaining sample was incubated at 37° C. for 7 d, followed by recovery of antibody (conjugated and unconjugated) by immunocapture using Fc-specific anti-human IgG-agarose resin (Sigma-Aldrich). Resin was rinsed twice with PBS, once with IgG elution buffer, and then twice more with PBS. ADC serum samples were then combined with anti-human IgG resin (100 μL of ADC-serum mixture, 50 μL resin slurry) and gently mixed for 15 minutes at room temperature. Resin was recovered by centrifugation and then washed twice with PBS. The resin pellet was resuspended in 100 μL IgG elution buffer (Thermo Scientific) and further incubated for 5 minutes at room temperature. Resin was removed by centrifugation and then 20 μL of 10× Glyco buffer 1 (New England Biolabs) and 5 μL Endo S (Remove iT EndoS, New England Biolabs) was added to the supernatant followed by incubation for 1 h at 37° C. Deglycosylated human antibody solution was sterile filtered, reduced with TCEP (Bond Breaker 0.5 M TCEP solution, Thermo Fisher Scientific) and analyzed by LC/MS. Percent conjugated antibody and the quantification of linker cleavage products were determined from peak heights of mass spectra.
(176) Mass spectrometry analysis; Samples were analyzed as described in Example 1.
(177)
(178)
(179)
(180)
(181) TABLE-US-00008 TABLE 7.1 Summary of 12G3H1 K274CP1-NNAA MMAE ADC serum stability data.sup.a,b Deconjugation.sup.c Linker cleavage MMAE payload Species (%) (%).sup.c DAR.sup.d AM-MMAE mouse 0.52 ± 0.9 99.1 ± 0.1.sup.e 0 rat 0.07 ± 0.4 none detected 1.9 ± 0.01 PM-MMAE mouse none detected 78 ± 2.sup.e 0.07 ± 0.06 rat 1.4 ± 2.4 31 ± 1.sup.f 1.35 ± .02 .sup.aADCs prepared with 12G3H11 mAb bearing a K274CP1-NNAA mutation. .sup.bsamples were incubated for 7 days at 37° C. .sup.ccalculated from peak intensities of reduced deglycosylated mass spectra. .sup.dcleaved linker species not included in DAR calculation. Theoretical DAR = 2. .sup.eboth val-cit dipeptide cleavage and phenylacetamide cleavage contributed to overall linker cleavage and drug loss. .sup.fphenylacetamide cleavage in linker observed, but not val-cit dipeptide cleavage.
(182)
(183)
(184)
(185) The cyclopentadiene-maleimide conjugation products between mAb-CP1-NNAA and maleimido-MMAE's are stable in rat and mouse serum over a period of 7 days regardless of the type of maleimide contained on the MMAE payload. Other parts of the ADC payload were found to degrade before the maleimide-CP1 diene bond. Specifically, both phenyl maleimide- and alkyl maleimide-MMAE payloads exhibited high val-cit dipeptide cleavage in mouse serum, likely due to enzymatic accessibility to the highly exposed K274 conjugation site. Phenyl-maleimide conjugate showed an additional structural liability at the phenyl acetamide between the phenyl maleimide and val-cit dipeptide. This cleavage was more evident in rat serum than mouse serum. It is unclear at which point in the process that phenylacetamide cleavage occurred, since it did not increase from day 0 to day 7. It is possible that cleavage occurred during immunocapture, which includes a low pH rinsing step. Overall, the stability of cyclopentadiene-maleimide conjugation product in physiological mileau was demonstrated.
Example 8. Synthesis of Spirocyclopentadiene (CP2)-Containing Compounds
(186) Spirocyclopentadiene-containirnt crosslinkers and non-natural amino acids (NNAAs) were prepared with the general structure shown below:
(187)
(188) Synthesis of CP2-NNAA (10) began with the reaction of a commercially available NaCp solution with epichlorohydrin in a modified version of Carreira's reaction..sup.1 Racemic epichlorohydrin was used, but 7 can be synthesized in 91% ee using enantiopure epichlorohydrin. The reaction of 7 with 4-nitrophenyl chloroformate produced activated carbamate 8. Reacting 8 with Fmoc-Lys-OH produces the Fmoc-protected 9, which could be deprotected using piperidine to obtain NNAA 10. Compound 10 shows a higher stability to acid compared to 4, and none of the intermediates in its synthesis show dimerization or decomposition when stored at −20° C.
(189) The synthesis of CP2-functionalized NHS-ester 12 began with the reaction of 7 with succinic anhydride to produce acid 11. The acid 7 was reacted with EDC-HCl and N-hydroxysuccinimide to yield NHS ester 12. Compound 12 doesn't appear to dimerize when stored for several days at room temperature.
(190) Materials and Methods:Unless stated otherwise, reactions were conducted under an atmosphere of N.sub.2 using reagent grade solvents. DCM, and toluene were stored over 3 Å molecular sieves. THF was passed over a column of activated alumina. All commercially obtained reagents were used as received. Thin-layer chromatography (TLC) was conducted with E. Merck silica gel 60 F254 pre-coated plates (0.25 mm) and visualized by exposure to UV light (254 nm) or stained with p-anisaldehyde, ninhydrin, or potassium permanganate. Flash column chromatography was performed using normal phase silica gel (60 Å, 0.040-0.063 mm, Geduran). .sup.1H NMR spectra were recorded on Varian spectrometers (400, 500, or 600 MHz) and are reported relative to deuterated solvent signals. Data for .sup.1H NMR spectra are reported as follows: chemical shift (6 ppm), multiplicity, coupling constant (Hz) and integration. .sup.13C NMR spectra were recorded on Varian Spectrometers (100, 125, or 150 MHz). Data for .sup.13C NMR spectra are reported in terms of chemical shift (6 ppm). Mass spectra were obtained from the UC Santa Barbara Mass Spectrometry Facility on a (Waters Corp.) GCT Premier high resolution Time-of-flight mass spectrometer with a field desorption (FD) source.
Synthesis of CP2-NNAA (10)
(191) ##STR00035##
Spiro[2.4]hepta-4,6-dien-1-ylmethanol (7)
(192) Sodium cyclopentadienide (2 M solution in THF, 10 mL, 20 mmol, 4 eq) was added to THF (40 mL) and cooled to 0° C. Epichlorohydrin (0.39 mL, 5.0 mmol, 1 eq) was added dropwise and the reaction was stirred at 0° C. for 1.5 hr then a further 2 hr at rt. The reaction was quenched with H.sub.2O (40 mL) then transferred to a seperatory funnel. A saturated solution of NaHCO.sub.3 in H.sub.2O (40 mL) and ether (40 mL) were added and the layers separated. The organic layer was washed with brine (40 mL), dried over MgSO.sub.4, filtered, and then the solvent removed. The residue was subjected to flash column chromatography (Hexane:EtOAc, 2:1) to yield 7 (0.48 g, 78%) as a brown oil.
(193) Rf (Hexane:EtOAc, 2:1): 0.22; .sup.1H NMR (500 MHz, CDCl.sub.3) δ 6.64 (td, J=1.6, 5.1 Hz, 1H), 6.51 (td, J=1.7, 5.1 Hz, 1H), 6.27 (tdd, J=1.0, 2.1, 5.2 Hz, 1H), 6.12 (td, J=1.7, 5.1 Hz, 1H), 4.08-3.88 (m, 1H), 3.59 (dd, J=8.8, 11.7 Hz, 1H), 2.48-2.40 (m, 1H), 1.87 (dd, J=4.3, 8.7 Hz, 1H), 1.69 (dd, J=4.4, 7.0 Hz, 1H), 1.57 (br. s., 1H) ppm; .sup.13C NMR (125 MHz, CDCl.sub.3) δ 139.4, 133.9, 131.7, 128.6, 64.9, 41.9, 30.0, 17.6 ppm.
(194) ##STR00036##
4-Nitrophenyl spiro[2.4]hepta-4,6-dien-1-ylmethyl carbonate (8)
(195) 7 (2.80 g, 22.9 mmol, 1 eq) was added to DCM (100 mL) and cooled to 0° C. Pyridine (4.61 mL, 57.3 mmol, 2.5 eq) was added followed by 4-nitrophenyl chloroformate (5.08 g, 25.2 mmol, 1.1 eq). The reaction was stirred at 0° C. until consumption of the starting material (TLC, 30 min). The reaction was poured into a separatory funnel and washed with a saturated solution of NH.sub.4Cl in H.sub.2O (100 mL). The aqueous layer was extracted with DCM (50 mL). The organic layers were combined, washed with brine (50 mL), dried over Na.sub.2SO.sub.4, filtered, and the solvent removed. The residue was subjected to flash column chromatography (Hexane:EtOAc, 6:1 to 4:1) to yield 8 (5.17 g, 79%) as an amber oil.
(196) Rf (Hexane:EtOAc, 4:1): 0.28; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 8.28 (d, J=9.0 Hz, 2H), 7.37 (d, J=9.0 Hz, 2H), 6.62 (td, J=1.7, 5.2 Hz, 1H), 6.53 (td, J=1.7, 4.8 Hz, 1H), 6.25 (td, J=1.8, 5.5 Hz, 1H), 6.11 (td, J=1.6, 5.1 Hz, 1H), 4.53 (dd, J=7.6, 11.5 Hz, 1H), 4.40 (dd, J=7.4, 11.3 Hz, 1H), 2.52 (quin, J=7.6 Hz, 1H), 1.92 (dd, J=4.7, 8.6 Hz, 1H), 1.76 (dd, J=4.7, 6.7 Hz, 1H) ppm; .sup.13C NMR (100 MHz, CDCl.sub.3) δ 155.4, 152.3, 145.3, 138.6, 133.8, 131.7, 129.4, 125.2, 121.7, 70.9, 41.5, 24.6, 16.9 ppm.
(197) ##STR00037##
Fmoc-Lys(spiro[2.4]hepta-4,6-dien-1-ylmethyl carbonate)-OH (9)
(198) 8 (5.12 g, 17.8 mmol, 1 eq) was added to DMF (40 mL), followed by Fmoc-Lys-OH (7.87 g, 21.4 mmol, 1.2 eq) and DIPEA (7.44 mL, 42.7 mmol, 2.4 eq). The reaction was stirred until consumption of the starting material (NMR, 3.5 hr), then poured into EtOAc (100 mL) and H.sub.2O (140 mL). The aqueous layer was acidified to pH 2-3 with HCl (1 M, 100 mL), poured into a separatory funnel, and the layers separated. The aqueous layer was extracted with EtOAc (2×100 mL). The organic layers were combined, washed with brine (100 mL), dried over Na.sub.2SO.sub.4, filtered, and the solvent removed. The residue was subjected to flash column chromatography (Hexane:EtOAc, 3:1 then DCM:MeOH:AcOH, 89:10:1) and the solvent removed. Residual AcOH and DMF was removed by suspending the product in DCM, washing with brine, drying the organic layer over Na.sub.2SO.sub.4, filtering, then removing the solvent to yield 9 (7.43 g, 81%) as an eggshell foam.
(199) Rf (DCM:MeOH, 90:10): 0.39; .sup.1H NMR (500 MHz, CDCl.sub.3) δ 8.62 (br. s., 1H), 7.75 (d, J=7.3 Hz, 2H), 7.66-7.49 (m, 2H), 7.39 (t, J=7.4 Hz, 2H), 7.30 (t, J=7.3 Hz, 2H), 6.54 (br. s., 1H), 6.47 (br. s., 1H), 6.21 (br. s., 1H), 6.04 (br. s., 1H), 5.74 (d, J=7.3 Hz, 1H), 4.91 (br. s., 1H), 4.53-4.00 (m, 5H), 3.21-3.00 (m, 2H), 2.97 (s, 1H), 2.90 (d, J=0.8 Hz, 1H), 2.47-2.31 (m, 1H), 1.95-1.27 (m, 6H) ppm; .sup.13C NMR (125 MHz, CDCl.sub.3) 163.2, 156.7, 143.6, 141.2, 138.9, 134.5, 130.9, 128.9, 127.6, 127.0, 125.1, 119.9, 115.6, 67.0, 66.5, 53.5, 47.1, 41.6, 40.4, 36.8, 31.8, 29.2, 25.7, 22.2, 21.4, 17.1 δ ppm.
(200) ##STR00038##
CP2-NNAA (10)
(201) 9 (5.50 g, 10.6 mmol, 1 eq) was added to DMF (150 mL), followed by piperidine (16.8 mL). The reaction was stirred until consumption of the starting material (TLC, 90 min), then the solvent was removed. Et.sub.2O (100 mL) was added to the residue, and the suspension was sonicated for 5 min. The suspension was filtered and rinsed with H.sub.2O (2×100 mL) and Et.sub.2O (100 mL). The solid was suspended in MeOH (10 mL), stirred for 10 min with gentle warming (˜40° C.), Et.sub.2O (40 mL) was added, the suspension filtered and rinsed with Et.sub.2O (2×50 mL). The compound was dried under vacuum to yield 10 (2.24 g, 71%) as a white powder.
(202) Rf (DCM:MeOH, 85:15): 0.29; .sup.1H NMR (400 MHz, DMSO-d.sub.6+1 drop TFA) δ 8.20 (br. s., 3H), 7.16 (t, J=5.5 Hz, 1H), 6.48 (td, J=1.8, 5.1 Hz, 1H), 6.40 (d, J=5.1 Hz, 1H), 6.32 (d, J=5.1 Hz, 1H), 6.12 (td, J=1.9, 4.9 Hz, 1H), 4.24 (dd, J=6.7, 11.7 Hz, 1H), 3.99 (dd, J=7.6, 11.5 Hz, 1H), 3.88 (d, J=5.1 Hz, 1H), 2.94 (d, J=5.9 Hz, 2H), 2.37 (quin, J=7.5 Hz, 1H), 1.83-1.63 (m, 4H), 1.44-1.19 (m, 4H) ppm; .sup.13C NMR (100 MHz, DMSO-d.sub.6+1 drop TFA):
(203) 171.2, 156.2, 139.3, 135.2, 130.4, 128.3, 65.3, 51.9, 42.0, 29.7, 28.9, 25.7, 21.6, 16.4; MS (EI) Exact mass cald. for C.sub.15H.sub.22N.sub.2O.sub.4 [M].sup.+: 294.1580, found: 294.1571.
Synthesis of CP2-NHS (12)
(204) ##STR00039##
4-Oxo-4-(spiro[2.4]hepta-4,6-dien-1-ylmethoxy)butanoic acid (11)
(205) DCM (1.5 mL) was added to a vial containing 1 (0.37 g, 3.0 mmol, 1 eq). Et.sub.3N (0.42 mL, 3.0 mmol, 1 eq), DMAP (37 mg, 0.30 mmol, 0.1 eq) and succinic anhydride (0.33 g, 3.3 mmol, 1.1 eq) were added, the reaction capped under an atmosphere of air, and stirred at rt until consumption of the starting material (TLC, 1.75 hr). The reaction mixture was poured into a separatory funnel with DCM (50 mL) and washed with aqueous HCl (1 M, 50 mL). The aqueous layer was extracted with DCM (50 mL), the organic layers combined, dried over Na.sub.2SO.sub.4, filtered, and the solvent removed to yield 11 of sufficient purity for the next reaction.
(206) Rf (EtOAc): 0.56; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 10.60 (br. s., 1H), 6.57 (td, J=1.9, 5.3 Hz, 1H), 6.50 (td, J=1.8, 5.1 Hz, 1H), 6.21 (td, J=1.7, 5.2 Hz, 1H), 6.07 (td, J=1.8, 5.1 Hz, 1H), 4.37 (dd, J=7.4, 11.7 Hz, 1H), 4.20 (dd, J=7.0, 11.7 Hz, 1H), 2.74-2.57 (m, 4H), 2.42 (quin, J=7.8 Hz, 1H), 1.85 (dd, J=4.5, 8.4 Hz, 1H), 1.69 (dd, J=4.3, 7.0 Hz, 1H) ppm.
(207) ##STR00040##
CP2-NHS (12)
(208) THF (10 mL) was added to a vial containing 11 (theo 3.0 mmol, 1 eq). NHS (0.48 g, 4.2 mmol, 1.4 eq), EDC-HCl (0.69 g, 3.6 mmol, 1.2 eq) and DCM (5 mL) were added, the reaction capped under an atmosphere of air, and stirred at rt overnight. The solvent was removed and the residue was subjected to flash column chromatography (Hexane:EtOAc, 1:1) to yield 12 (0.59 g, 62% over two steps) as a colourless, viscous oil.
(209) Rf (Hexane:EtOAc, 1:1): 0.34; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 6.56 (td, J=1.8, 5.1 Hz, 1H), 6.48 (td, J=1.8, 5.1 Hz, 1H), 6.21 (td, J=1.6, 3.4 Hz, 1H), 6.06 (td, J=1.6, 3.4 Hz, 1H), 4.36 (dd, J=7.4, 11.7 Hz, 1H), 4.21 (dd, J=7.4, 11.7 Hz, 1H), 2.93 (t, J=7.0 Hz, 2H), 2.83 (s, 4H), 2.73 (t, J=7.4 Hz, 2H), 2.42 (quin, J=7.6 Hz, 1H), 1.83 (dd, J=4.3, 8.6 Hz, 1H), 1.68 (dd, J=4.5, 6.8 Hz, 1H) ppm; .sup.13C NMR (100 MHz, CDCl.sub.3) δ 170.8, 168.9, 167.6, 138.8, 134.3, 131.2, 129.0, 66.6, 41.5, 28.6, 26.2, 25.5, 25.1, 17.3 ppm. 1. Ledford, B. E.; Carreira, E. M., Total Synthesis of (+)-Trehazolin: Optically Active Spirocycloheptadienes as Useful Precursors for the Synthesis of Amino Cyclopentitols. Journal of the American Chemical Society 1995, 117, 11811-11812.
Example 9. CP2 Diene-Maleimide Conjugation for Preparation of ADCs Via Crosslinker-Modified mAb
(210) The feasibility of spirocyclopentadiene-maleimide reactions for bioconjugation was evaluated. Spirocyclopentadiene groups were introduced via an amine-reactive heterobifunctional linker with the same general strategy described in Example 3.
(211) Introduction of CP2 functionality onto mAbs: CP2 diene functionality was installed onto IgG1 mAbs by reaction of lysine primary amines with NHS-ester activated CP2 diene. This approach resulted in randomly conjugated, amide-linked cyclopentadiene groups. The resulting antibody is termed mAb-CP2-linker, but may also be denoted as mAb-CP2 in figures. See figure captions for clarification. A typical mAb modification reaction is described as follows. Mab solution was adjusted to 5 mg/mL (3 mL, 15 mg mAb, 100 nmol, 1 eq.) with PBS pH 7.2 followed by addition of 10% v/v 1 M NaHCO.sub.3. This solution was chilled on ice and 35 μL CP2-NHS (10 mM stock in DMAc, 350 nmol, 3.5 equivalents) was added. The reaction proceeded on ice for 5 minutes followed by reaction at room temperature for 1 h with continuous mixing. Reacted mAb was purified by dialysis (Slide-A-Lyzer, 10 kDa MWCO) against PBS, 1 mM EDTA, pH 7.4, 0° C. for 24 h. CP2 introduction was quantified by intact deglycosylated mass spectrometry as described below and found to be 3.29 CP2-linkers (and thus dienes) per mAb in this example, which corresponds to 94% conversion of CP2-NHS to antibody conjugate.
(212) ##STR00041##
(213) Reaction of CP2-modified mAb with maleimido-MMAEs: mAb-CP2-linker (3.29 CP2 dienes/mAb, 1 mg, 6.7 nmol mAb, 1 equivalent) was diluted with PBS (pH 7.4) to a final concentration of 3.16 mg/mL. Next, DMSO was added to yield a 20% v/v solution followed by addition of 1 M sodium phosphate monobasic to yield a 10% v/v solution. Addition of all solution components yielded a mixture comprising 2.43 mg/mL mAb, 53.3 μM CP2, 1.78 M DMSO, 110 mM sodium phosphate, 100 mM NaCl, pH 5.5. AM-MMAE or PM-MMAE (10 μL of a 10 mM stock solution in DMAc, 100 nmol, 15 equivalents) was added to the antibody solution. The reaction mixture was vortexed briefly and allowed to proceed at 22° C. or 37° C. with mixing. After 4 h reaction, N-acetyl cysteine (8 μL of a 100 mM solution, 120 equivalents) was added to quench unreacted maleimide groups. Samples were purified using PD Spintrap G-25 devices (GE Healthcare Life Sciences) to remove small molecule components from the mixture. Samples were subsequently analyzed by reduced deglycosylated mass spectrometry as described below.
(214) ##STR00042##
(215) Mass spectrometry analysis; Samples were analyzed as described in Example 1.
(216)
(217) TABLE-US-00009 TABLE 9.1 Summary of CP2-NHS mAb reaction equivalents CP1-NHS [mAb] CP2-linker (rel to mAb) mg/nL per mAb conversion (%) 3.5 5 3.29 94
(218)
(219)
(220) TABLE-US-00010 TABLE 9.2 Summary of mAb-CP2-linker maleimido-MMAE reactions.sup.a Equivalents MMAE conjugation payload (rel to mAb) pH temp (%) AM-MMAE 15 5.5 37° C. 88 22° C. 73 PM-MMAE 15 5.5 37° C. 95 22° C. 78 .sup.aAll reactions performed at 2.43 mg/mL mAb-CP2-linker for 4 h.
(221) CP2 diene groups installed onto the surface of antibodies partially reacted with maleimido-MMAE prodrugs within 4 h at room temperature. No non-specific conjugation was observed by mass spectrometry, as all conjugate peaks tracked from mAb-CP2-linker peaks and not unreacted mAb. This reaction is much more efficient than furan diene, but less efficient than CP1 diene for reaction with maleimido-MMAE payloads. This approach can be used for production of bioconjugates.
Example 10. Kinetics of mAb-CP2-Linker Conjugation to Maleimido-MMAEs at 1.0 Molar Equivalent Maleimido-MMAE to Diene Groups
(222) Reaction kinetics of CP2 dienes with maleimido MMAEs at 22° C. was evaluated.
(223) Introduction of CP2 diene functionality onto mAbs: CP2 functionality was installed onto IgG1 mAbs by reaction of lysine primary amines with NHS-ester activated CP2. This approach resulted in randomly conjugated, amide-linked cyclopentadiene groups. A typical mAb modification reaction is described as follows. Mab solution was adjusted to 5 mg/mL (3 mL, 15 mg mAb, 100 nmol, 1 eq.) with PBS pH 7.2 followed by addition of 10% v/v 1 M NaHCO.sub.3. This solution was chilled on ice and 35 μL CP2-NHS (10 mM stock in DMAc, 350 nmol, 3.5 equivalents) was added. The reaction proceeded on ice for 5 minutes followed by reaction at room temperature for 1 h with continuous mixing. Reacted mAb was purified by dialysis (Slide-A-Lyzer, 10 kDa MWCO) against PBS, 1 mM EDTA, pH 7.4, 0° C. for 24 h. CP2-linker introduction was quantified by intact deglycosylated mass spectrometry as described below and found to be 3.29 CP2-linkers (and thus dienes) per mAb in this example, which corresponds to 94% conversion of CP2-NHS to antibody conjugate.
(224) Reaction of CP2-modified mAb with maleimido-MMAEs: mAb-CP2-linker (3 mg, 3.29 CP2/mAb, 66 nmol CP2 diene, 1 equivalent) was diluted with PBS (pH 7.4) to a final concentration of 1.7 mg/mL. Next, DMSO was added to yield a 20% v/v solution followed by addition of 1 M sodium phosphate monobasic to yield a 10% v/v solution. Addition of all solution components yielded a mixture comprising 1.3 mg/mL mAb, 32.3 μM CP2 diene, 1.78 M DMSO, 110 mM sodium phosphate, 100 mM NaCl, pH 5.5. AM-MMAE or PM-MMAE (6.6 μL of a 10 mM stock solution in DMSO, 66 nmol, 1 equivalent) was added to the antibody solution. The reaction mixture was vortexed briefly and allowed to proceed at 22° C. with mixing. Aliquots (180 μL) were removed at various timepoints and N-acetyl cysteine (3 μL of a 100 mM solution, 45 equivalents) was added to quench unreacted maleimide groups. Samples were then purified using PD Spintrap G-25 devices (GE Healthcare Life Sciences) to remove small molecule components from the mixture. Samples were then analyzed by reduced deglycosylated mass spectrometry as described below.
(225) Mass spectrometry analysis: Samples were analyzed as described in Example 1.
(226) Calculation of CP2 diene-maleimide reaction rate constants: Second order rate constants for reaction of maleimido-MMAEs with CP2 dienes in mAb-CP2-linker were determined from peak intensities in deglycosylated reduced mass spectra. Reaction progress was monitored by both disappearance of mAb-CP2-linker peaks and appearance of mAb-CP2-linker-MMAE conjugate peaks, but only mAb-CP2-linker peak intensities on the antibody heavy chains were used to calculate relative abundance of reacted CP2 diene. Unreacted CP2 diene groups on mAb heavy chains was calculated using the equation below:
(227)
a=peak intensity of unmodified heavy chain
b=sum of peak intensities of heavy chains with one CP2 diene group
c=sum of peak intensities of heavy chains with two CP2 diene groups
d=peak intensity of heavy chain with three CP2 diene groups
e=peak intensity of unmodified light chain
f=sum of peak intensities of light chains with one CP2 diene group
g=sum of peak intensities of light chains with two CP2 diene groups
(228) Conjugation data were further analyzed in units of molar concentration to determine kinetic constants. Second order rate constants were determined from the slopes of curves generated from plotting 1/[CP2 diene] versus time and linear regression analysis. Reaction half-lives were calculated from Second-order reaction rate constants using the equation shown below:
(229)
k.sub.2=second order rate constant
[CP2].sub.0=CP2 diene concentration at time=0
(230)
(231)
(232)
(233) TABLE-US-00011 TABLE 10.1 Summary of kinetic data for reaction of mAb-CP2-linker diene with-maleimido-MMAE .sup.a,b,c 2.sup.nd order rate t.sub.1/2 conversion payload constant (M.sup.−1s.sup.−1) (min) (%) AM-MMAE 2.2 ± 0.1 219 73 PM-MMAE 2.1 ± 0.1 230 93 .sup.a all conjugation reactions performed at pH 5.5, 20% DMSO, 22° C. and 1.3 mg/mL CP2-modified mAb .sup.b the molar ratio of MMAE:CP2 diene used was 1:1 .sup.c calculated from peak intensities of reduced deglycosylated mass spectra of measurement at 48 h.
(234) Reaction of CP2 diene with maleimido-MMAEs was slower than CP1 dienes, with half-live's on the order of several hours. Unlike CP1 diene, no difference in reaction rate was noted for phenyl- vs. alkyl-maleimide. Reaction conversion was 90% for AM-MMAE, but only 73% for PM-MMAE after 48 h reaction. Given the similar rate constants for both substrates, it is possible that a portion of phenyl maleimides degrade after extended time under these conditions, thus limiting overall conversion.
Example 11. CP2-NNAA Incorporation into Antibodies
(235) Incorporation of CP2-NNAA into position K274 or S239 of an anti EphA2 (1C1) antibody, the quality of expressed mAb, and reactivity of CP2-NNAA diene after antibody incorporation was assessed.
(236) Preparation of CP2 NNAA stock solution: CP2 NNAA (0.5 g, 1.7 mmol) was combined with 7.8 mL 0.2 M NaOH in H.sub.2O. The resulting slurry was stirred at room temperature until all solids dissolved (10 minutes). After complete dissolution the light-yellow solution was passed through a 0.2 μm filter, aliquoted, and stored at −80° C. until use. This procedure resulted in 8.2 mL of 216 mM CP2 NNAA stock solution.
(237) ##STR00043##
(238) Antibody expression: 12G3H11 or 1C1 IgG1 antibody genes with an amber mutation at Fc position K274 or S239 were cloned into a proprietary pOE antibody expression vector. The construct was transfected into CHO-G22 by PEImax (1.5 L of G22 cells), along with a plasmid encoding PyIRS double mutant (Y306A/Y384F) or wild-type PyIRS and a plasmid containing tandem repeats of the tRNA expression cassette (pORIP 9×tRNA). Four hours post transfection, 3.3% of feed F9 (proprietary) and 0.2% of feed F10 (proprietary) were added to cells and the cells were further incubated at 34 degrees. CP2-NNAA was added the next day at final concentration of 0.26 mM for 1C1 K274 and 1C1 S239 transfected cells. Cells were fed again on day 3 and day 7 with 6.6% of feed F9 and 0.4% of feed F10. Cells were spun down and supernatant was harvested on day 11. The supernatant was purified by IgSelect affinity column (GE Health Care Life Science). The antibody was eluted with 50 mM glycine, 30 mM NaCl, pH 3.5 elution buffer, neutralized with 1 M Tris buffer pH 7.5, and dialyzed into PBS, pH 7.2. Concentration of antibody eluted was determined by absorbance measurement at 280 nm. The back calculated titer was 57 mg/L for 1C1 K274CP2-NNAA and 76 mg/L for 1C1 S239CP2-NNAA. 12G3H11 mAb was expressed in a similar manner at smaller scale, with CP2-NNAA feed concentration varied. Recovered antibody was analyzed by SDS-PAGE using standard methods. Antibody was also analyzed by size exclusion chromatography and mass spectrometry as described below. Antibodies incorporating CP2-NNAA are denoted as mAb-CP1-NNAA to distinguish them from mAb-CP2-linker constructs, or mAb-[position]CP2-NNAA where [position] indicates the amino acid number and amino acid symbol that was mutated to CP2-NNAA.
(239) Size exclusion chromatography: SEC analysis was performed using an Agilent 1100 Capillary LC system equipped with a triple detector array (Viscotek 301, Viscotek, Houson, Tex.); the wavelength was set to 280 nm, and samples were run on a TSK-GEL G3000SWXL column (Toso Bioscience LLC, Montgomeryville, Pa.) using 100 mM sodium phosphate buffer, pH 6.8 at a flow rate of 1 mL/min.
(240) Mass spectrometry analysis: For deglycosylated mAb analysis, EndoS (5 μL Remove-iT EndoS (1:10 dilution in PBS, 20,000 units/mL, New England BioLabs) was combined with 50 μL sample (1 mg/mL mAb) and 5 μL glyco buffer 1 (New England BioLabs) and followed by incubation for 1 h at 37° C. Reduced samples were prepared by addition of 5 μL Bond-Breaker TCEP solution (0.5 M, Thermo Fisher Scientific) and incubation for 10 min at 37° C. Mass spectrometry analysis was performed using an Agilent 6520B Q-TOF mass spectrometer equipped with a RP-HPLC column (ZORBAX 300 Diphenyl RRHD, 1.8 micron, 2.1 mm×50 mm). High-performance liquid chromatography (HPLC) parameters were as follows: flow rate, 0.5 ml/min; mobile phase A was 0.1% (v/v) formic acid in HPLC-grade H.sub.2O, and mobile phase B was 0.1% (v/v) formic acid in acetonitrile. The column was equilibrated in 90% A/10% B, which was also used to desalt the mAb samples, followed by elution in 20% A/80% B. Mass spec data were collected for 100-3000 m/z, positive polarity, a gas temperature of 350° C., a nebulizer pressure of 48 lb/in.sup.2, and a capillary voltage of 5,000 V. Data were analyzed using vendor-supplied (Agilent v.B.04.00) MassHunter Qualitative Analysis software and peak intensities from deconvoluted spectra were used to derive the relative proportion of species in each sample.
(241)
(242) TABLE-US-00012 TABLE 11.1 Summary of 1C1 K274CP2-NNAA and 1C1 S239CP2-NNAA mAb production K274 S239 NNAA feed (mM) 0.5 0.5 Volume (L) 2 2 Mass recovered (mg) 114 153 Titer (mg/L) 57 76 Monomer (%) 93.2 99
(243)
(244)
(245)
(246) TABLE-US-00013 TABLE 11.2 Summary of mass spectrometry data for IC1-K274CP2-NNAA and 1C1-S239CP2-NNAA mAbs K274 S239 WT Observed intact mass 147545.85 147628.1 147249.63 Observed change relative to WT +296.2 +378.4 NA Calculated change relative to WT +296.4 +378.6 NA Observed heavy chain mass 50325.22 50367.71 50177.73 Observed change relative to WT +147.5 +189.9 NA Calculated change relative to WT +148.2 +189.3 NA
(247)
(248)
(249)
(250) Incorporation of CP2-NNAA into antibodies at positions K274 and S239 was confirmed by mass spectrometry. Recovered antibody was of high quality, with no truncated product and very little aggregate. Titers achieved at 2 L production scale for 1C1 antibody were reasonably high considering the low amount of CP2-NNAA fed to cells.
Example 12. Production and Evaluation of ADCs with 1C1 K274CP2-NNAA and 1C1 S239CP2-NNAA mAbs
(251) Reactivity of CP2-NNAA after incorporation into mAbs at position K274 or S239 of an anti EphA2 (1C1) antibody was assessed by conjugation with AM-MMAE. Resulting ADCs were evaluated to determine drug:antibody ratio (DAR), serum stability, and in vitro cytotoxicity.
(252) Preparation of CP2-mAb ADCs: CP2-NNAA mAb ADCs were prepared in one step, simply by mixing antibody with alkyl-maleimide MMAE (AM-MMAE). First, 1C1-S239CP2-NNAA mAb solution (8 mg, 53 nmol, 1 equivalent) was diluted to 2 mg/mL with PBS (4 mL total volume). DMSO (813 μL) and 1 M sodium phosphate, monobasic (407 μL) was added to yield ˜20% and ˜10% v/v solution, respectively. AM-MMAE (53.3 μL of 10 mM stock in DMSO, 533 nmol, 10 equivalents) was added to 1C1-S239CP2-NNAA mAb solution and the mixture was vortexed briefly. The reaction proceeded at room temperature for 7 h with continuous mixing. N-acetyl cysteine (43 μL of 100 mM stock in water, 4.3 μmol, 80 equivalents) was added and the solution was incubated for an additional 15 min to quench unreacted maleimide groups. The reaction mixture was then diluted 3-fold with distilled water and subjected to CHT chromatography (Bio-Scale Mini Cartridge CHT Type II 40 μm media column). ADC was eluted with a gradient from buffer A (10 mM phosphate, pH 7.0) to buffer B (10 mM phosphate pH 7.0 containing 2M NaCl) over 25 minutes. After CHT chromatography the sample was buffer exchanged to PBS supplemented with 1 mM EDTA, pH 7.4 by dialysis in a slide-a-lyzer cassette at 4° C. 1C1-K274CP2-NNAA mAb was conjugated with AM-MMAE in the same manner, except the reaction proceeded for 17 h at room temperature.
(253) Preparation of site-specific cysteine ADCs: For some experiments, mAb-CP2-NNAA ADCs were compared to cysteine-conjugated ADCs. For this purpose, an antibody was generated comprising a cysteine at position 239 (termed 1C1-239C). Conjugation of AM-MMAE to 1C1-239C was conducted in three steps; i) reduction and dialysis, ii) oxidation, iii) reaction with AM-MMAE. First, antibodies were mildly reduced to generate free sulfhydryls by combining 4 mL of 2.5 mg/mL antibody solution in 10 mM PBS pH 7.4 containing 1 mM EDTA (10 mg antibody, 66.7 nM, 1 eq) with 53 μL of 50 mM TCEP solution in water (2.7 μmol, 40 eq relative to mAb) followed by gentle mixing at 37° C. for 3 h. Reduced antibody was transferred to a slide-a-lyzer dialysis cassette (10K MWCO) and dialyzed against PBS, 1 mM EDTA, pH 7.4, 4° C. for 24 h with several buffer changes. Reduced antibody was oxidized to reform internal disulfides by addition of dehydroascorbic acid (27 μL of 50 mM stock in DMSO, 1.3 μmol, 20 eq) followed by mixing for 4 h at room temperature. Oxidized antibody solution was combined with 20% v/v DMSO followed by addition of AM-MMAE (53 μL of a 10 mM stock in DMSO, 530 nmol, 8 eq). The reaction proceeded at room temperature with mixing for 1 h followed by addition of N-acetyl cysteine (43 μL of 100 mM stock in water, 4.2 μmol, 64 eq) to quench unreacted maleimides. The reaction mixture was then diluted 3-fold with distilled water and subjected to CHT chromatography and dialysis as described above.
(254)
(255) Mass spectrometry analysis: Samples were analysed as described in Example 1.
(256) HIC chromatography analysis: ADCs were analyzed by size exclusion chromatography using a Proteomics HIC Butyl NPS column (4.6×35 mm, 5 μm, Sepax) eluted with a gradient of 100% A to 100% B over 22 minutes (mobile phase A: 25 mM Tris pH 8.0, 1.5 M (NH.sub.4).sub.2SO.sub.4, mobile phase B: 25 mM Tris pH 8.0, 5% (v/v) isopropyl alcohol) at room temperature. Protein was detected by UV absorbance at 280 nm. Approximately 50-100 μg protein was injected for each analysis.
(257) rRP-HPLC analysis: For each analysis, the antibodies and ADCs were reduced at 37° C. for 20 minutes using 42 mM dithiothreitol (DTT) in PBS pH 7.2. 10 μg of reduced antibodies and ADCs was loaded onto a PLRP-S, 1000 Å column (2.1×50 mm, Agilent) and eluted at 40° C. at a flow rate of 0.5 mL/min with a gradient of 5% B to 100% B over 25 minutes (mobile phase A: 0.1% trifluoroacetic acid in water, and mobile phase B: 0.1% trifluoroacetic acid in acetonitrile). Percent conjugation was determined using integrated peak areas from the chromatogram.
(258) Serum stability of ADCs: ADCs were incubated in rat serum to challenge the stability of the Diels-Alder conjugate. ADCs were added to normal rat serum (Jackson Immunoresearch) to achieve a final concentration of 0.2 mg/mL (1.33 μM antibody), with the total volume of ADC solution added to serum less than 10%. The ADC-serum mixture was sterile filtered and an aliquot was removed from this mixture and frozen as a t=0 control. The remaining sample was then further incubated at 37° C. in a sealed container without stirring. Conjugated and unconjugated human antibody was recovered from rat serum by immunoprecipitation using Fc-specific anti-human IgG-agarose resin (Sigma-Aldrich). Resin was rinsed twice with PBS, once with IgG elution buffer, and then twice more with PBS. ADC-mouse serum samples were then combined with anti-human IgG resin (100 μL of ADC-serum mixture, 50 μL resin slurry) and mixed for 15 minutes at room temperature. Resin was recovered by centrifugation and then washed twice with PBS. Washed resin was resuspended in 100 μL IgG elution buffer (Thermo Scientific) and further incubated for 5 minutes at room temperature. Resin was removed by centrifugation and then 20 μL of 10× glycobuffer 1 (New England Biolabs) was added to the supernatant. Recovered human antibody solution was sterile filtered, and incubated with EndoS for 1 h at 37° C. Deglycosylated mAbs were then reduced with TCEP and analyzed by LC/MS as described above. Percent conjugated antibody was determined from peak heights of mass spectra.
(259) In vitro cytotoxicity analysis: Human prostate cancer cell line PC3 was obtained from American Type Culture Collection (ATCC). PC3 cells were maintained in RPMI1640 media (Life Technologies) supplemented with 10% heat-inactivated fetal bovine serum (HI-FBS) (Life Technologies) at 37° C. in 5% CO.sub.2. Cells were grown to exponential growth phase, harvested by mild trypsinization and seeded into 96-well culture plates at 1500 cells/well. Cells were allowed to adhere for 24 h and then treated with antibodies and ADCs subjected to 4-fold serial dilution at 9 concentrations in duplicate starting from 4000 ng/mL. Treated cells were cultured for 6 days and cell viability was determined using the CellTiter-Glo Luminescent Viability Assay (Promega) following the manufacturer's protocol. Cell viability was calculated as a percentage of control untreated cells. IC.sub.50 of the cytotoxicity for ADCs was determined using logistic non-linear regression analysis with Prism software (GraphPad).
(260)
(261)
(262) TABLE-US-00014 TABLE 12.1 Summary of mAb-CP2-NNAA ADC characterization by mass spectrometry.sup.a Observed Calculated Δ mass Δ mass mAb payload (AMU) (AMU) Conversion.sup.b DAR 1C1-K274CP2- AM- 1317.06 1316.65 98% 1.96 NNAA MMAE 1C1-S239CP2- AM- 1316.28 1316.65 97% 1.94 NNAA MMAE 1C1-239C AM- 1316.42 1316.65 95% 1.90 MMAE .sup.amAb heavy-chains were analyzed from glycosylated, reduced mass spectra .sup.bCalculated from relative peak heights of conjugated and unconjugated species
(263)
(264)
(265) TABLE-US-00015 TABLE 12.2 Summary of ADC characterization by chromatography methods Conjugaion Conjuation DAR efficiency DAR efficiency rRP- mAb payload HIC.sup.a HIC rRP-HPLC.sup.a HPLC 1C1-K274CP2- AM-MMAE 97% 1.94 91% 1.82 NNAA 1C1-S239CP2- AM-MMAE 95% 1.9 97% 1.94 NNAA 1C1-239C AM-MMAE 98% 1.96 95% 1.9 .sup.aCalculated from relative peak areas between conjugated and unconjugated species
(266)
(267)
(268)
(269)
(270) CP2-NNAA diene reacted with maleimide contained on AM-MMAE with similar conversions to cysteine sulfhydryl groups. The key difference in preparation of the mAb-CP2-NNAA ADC vs. the mAb-cysteine ADC was reduction in the number of steps in the conjugation process. Cysteine mAb required 3 steps and 2 days for production, whereas CP2-NNAA mAb ADCs were produced in one step in less than 24 h. Resulting mAb-CP2-NNAA ADCs are stable under physiologically relevant conditions and did not show drug loss when incubated in rat serum at 37° C. for 7 days. CP2-NNAA mAb ADCs were potent in vitro, with activities similar to an ADC prepared by site-specific cysteine conjugation.
Example 13. Synthesis of Cyclopentadiene and Furan-Containing Compounds
(271) Materials and Methods:Unless stated otherwise, reactions were conducted under an atmosphere of N.sub.2 using reagent grade solvents. DCM, and toluene were stored over 3 Å molecular sieves. THF was passed over a column of activated alumina. All commercially obtained reagents were used as received. Thin-layer chromatography (TLC) was conducted with E. Merck silica gel 60 F254 pre-coated plates (0.25 mm) and visualized by exposure to UV light (254 nm) or stained with p-anisaldehyde, ninhydrin, or potassium permanganate. Flash column chromatography was performed using normal phase silica gel (60 Å, 0.040-0.063 mm, Geduran). .sup.1H NMR spectra were recorded on Varian spectrometers (400, 500, or 600 MHz) and are reported relative to deuterated solvent signals. Data for .sup.1H NMR spectra are reported as follows: chemical shift (δ ppm), multiplicity, coupling constant (Hz) and integration. .sup.13C NMR spectra were recorded on Varian Spectrometers (100, 125, or 150 MHz). Data for .sup.13C NMR spectra are reported in terms of chemical shift (δ ppm). Mass spectra were obtained from the UC Santa Barbara Mass Spectrometry Facility on a (Waters Corp.) GCT Premier high resolution time-of-flight mass spectrometer with a field desorption (FD) source.
Synthesis of CP3-NHS (13)
(272) ##STR00044##
(273) 2,5-Dioxopyrrolidin-1-yl (1,2,3,4,5-pentamethylcyclopenta-2,4-dienyl)methyl succinate (13): DCM (8 mL) was added to a vial containing (1,2,3,4,5-pentamethylcyclopenta-2,4-dienyl)methanol.sup.1 (0.33 g, 2.0 mmol, 1 eq). Et.sub.3N (0.64 mL, 4.6 mmol, 2.3 eq), DMAP (46 mg, 0.38 mmol, 0.2 eq) and succinic anhydride (0.46 g, 4.6 mmol, 2.3 eq) were added, the reaction capped under an atmosphere of air, and stirred at rt overnight. The reaction was quenched with H.sub.2O (1 mL) then poured into a separatory funnel. HCl (1 M, 50 mL) was added and extracted with DCM (2×50 mL). The organic layers were combined, washed with brine (50 mL), dried over Na.sub.2SO.sub.4, filtered, and the solvent removed to yield 4-oxo-4-((1,2,3,4,5-pentamethylcyclopenta-2,4-dienyl)methoxy)butanoic acid which was used directly in the next reaction.
(274) Rf (EtOAc): 0.24; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 3.98 (s, 2H), 2.64-2.59 (m, 2H), 2.59-2.54 (m, 2H), 1.76 (s, 6H), 1.74 (s, 6H), 0.95 (s, 3H) ppm.
(275) THF (10 mL) was added to a vial containing 4-oxo-4-((1,2,3,4,5-pentamethylcyclopenta-2,4-dienyl)methoxy)butanoic acid (˜2 mmol). NHS (0.61 g, 5.3 mmol, 2.7 eq), EDC-HCl (0.87 g, 4.6 mmol, 2.3 eq) and DCM (6 mL) were added, the reaction capped under an atmosphere of air, and stirred at rt overnight. The solvent was removed and the residue was subjected to flash column chromatography (Hexane:EtOAc, 3:1.fwdarw.2:1) to yield 7 (0.39 g, 55% over two steps) as a white solid.
(276) Rf (Hexane:EtOAc, 7:3): 0.27; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 4.00 (s, 2H), 2.89 (t, J=6.7 Hz, 2H), 2.85 (br. s., 4H), 2.67 (t, J=7.8 Hz, 2H), 1.77 (s, 6H), 1.74 (s, 6H), 0.95 (s, 3H) ppm; .sup.13C NMR (100 MHz, CDCl.sub.3) δ 170.7, 168.9, 167.6, 138.4, 135.0, 68.2, 55.3, 28.6, 26.2, 25.5, 16.8, 11.0, 10.1 ppm; IR (ATR) 2973, 2935, 1815, 1782, 1729, 1208, 1089, 1069, 967 cm.sup.−1; HRMS (EI) Exact mass cald. for C.sub.19H.sub.25NO.sub.6 [M].sup.+: 363.1682, found: 363.1676.
Synthesis of F2-NHS (17)
(277) ##STR00045##
Methyl 9,9-diethoxy-6-hydroxynon-7-ynoate (14)
(278) 3,3-Diethoxyprop-1-yne (0.72 mL, 5.0 mmol, 1 eq) was added to THF (15 mL) then cooled to −78° C. nBuLi (2.33 M in hexanes, 2.4 mL, 5.5 mmol, 1.1 eq) was added dropwise then the reaction mixture stirred a further 30 min at −78° C. Methyl 6-oxohexanoate (0.87 g, 6.0 mmol, 1.2 eq) dissolved in THF (5 mL) was added dropwise then the reaction mixture stirred at −78° C. for 1 hr. The reaction mixture was poured into a separatory funnel containing a saturated aqueous solution of sodium bicarbonate (100 mL) then extracted with Et.sub.2O (2×50 mL). The combined organic layers were washed with brine (50 mL), dried over MgSO.sub.4, filtered, the solvent removed, and the residue subjected to flash column chromatography (Hexane:EtOAc, 2:1) to yield 14 (1.1 g, 80%) as a clear and colourless oil.
(279) Rf (Hexane:EtOAc, 6:4): 0.41; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 5.28 (d, J=1.6 Hz, 1H), 4.47-4.36 (m, J=3.5 Hz, 1H), 3.76-3.67 (m, 2H), 3.67-3.63 (m, 3H), 3.56 (qd, J=7.0, 9.4 Hz, 2H), 2.34-2.27 (m, 3H), 1.76-1.59 (m, 4H), 1.54-1.42 (m, 2H), 1.21 (t, J=7.0 Hz, 6H) ppm; .sup.13C NMR (100 MHz, CDCl.sub.3) δ 174.0, 91.2, 86.2, 80.0, 61.8, 60.8, 60.8, 51.5, 36.9, 33.8, 24.6, 24.4, 15.0 ppm; IR (ATR) 3451, 2932, 1736, 1437, 1328, 1135, 1051, 1012 cm.sup.1; HRMS (EI) Exact mass cald. for C.sub.14H.sub.23O.sub.5 [M−H].sup.+: 271.1545, found: 271.1546.
(280) ##STR00046##
Methyl 5-(3-methoxyfuran-2-yl)pentanoate (15)
(281) MeOH (3.9 mL) was added to a vial containing 14 (1.06 g, 3.89 mmol, 1 eq). PPh.sub.3AuNTf.sub.2 (29 mg, 0.039 mmol, 0.01 eq) was added, the reaction capped under an atmosphere of air, and stirred at rt overnight. The reaction mixture was poured into a separatory funnel containing brine (50 mL) then extracted with DCM (2×50 mL). The combined organic layers were washed with brine (50 mL), dried over Na.sub.2SO.sub.4, filtered, the solvent removed, and the residue subjected to flash column chromatography (Hexane:EtOAc, 15:1.fwdarw.9:1) to yield 15 (0.35 g, 43%) as a clear and colourless oil.
(282) Rf (Hexane:EtOAc, 9:1): 0.35; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 7.11 (d, J=2.0 Hz, 1H), 6.27 (d, J=2.0 Hz, 1H), 3.72 (s, 3H), 3.66 (s, 3H), 2.61 (t, J=6.8 Hz, 2H), 2.33 (t, J=7.2 Hz, 2H), 1.69-1.60 (m, 4H) ppm; .sup.13C NMR (100 MHz, CDCl.sub.3) δ 174.1, 143.3, 139.2, 138.9, 102.9, 59.4, 51.4, 33.7, 27.5, 24.5, 24.3 ppm; IR (ATR) 2950, 1734, 1662, 1600, 1230, 1179, 1111 cm.sup.1; HRMS (EL) Exact mass cald. for C.sub.11H.sub.16O.sub.4 [M].sup.+: 212.1049, found: 212.1045.
(283) ##STR00047##
5-(3-Methoxyfuran-2-yl)pentanoic acid (16)
(284) To a vial containing 15 (0.331 g, 1.56 mmol, 1 eq) dissolved in MeOH (4 mL) was added a solution of NaOH (0.125 g, 3.12 mmol, 2 eq) in H.sub.2O (4 mL). The reaction was capped under an atmosphere of air, and stirred at rt 30 min. The reaction mixture was poured into a separatory funnel containing H.sub.2O (50 mL) and HCl (1 M in H.sub.2O) was added to pH 2-3 (˜4 mL). The aqueous layer was extracted with DCM (2×50 mL). The combined organic layers were washed with brine (50 mL), dried over Na.sub.2SO.sub.4, filtered, the solvent removed to yield 16 (0.280 g, 90%) as a clear and colourless oil.
(285) Rf (Hexane:EtOAc, 1:1): 0.55; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 10.37 (br. s., 1H), 7.12 (d, J=2.0 Hz, 1H), 6.28 (d, J=2.0 Hz, 1H), 3.73 (s, 3H), 2.62 (t, J=6.5 Hz, 2H), 2.38 (t, J=6.5 Hz, 2H), 1.73-1.61 (m, 4H) ppm; .sup.13C NMR (100 MHz, CDCl.sub.3) δ 179.8, 143.4, 139.1, 139.0, 102.9, 59.4, 33.7, 27.4, 24.4, 24.0 ppm; IR (ATR) 3133, 2940, 1706, 1662, 1454, 1411, 1279, 1236, 1109 cm.sup.−1; HRMS (EI) Exact mass cald. for C.sub.10H.sub.14O.sub.4 [M].sup.+: 198.0892, found: 198.0890.
(286) ##STR00048##
2,5-Dioxopyrrolidin-1-yl 5-(3-methoxyfuran-2-yl)pentanoate (17)
(287) THF (5 mL) was added to a vial containing 16 (0.265 g, 1.34 mmol, 1 eq). NHS (0.216 g, 1.87 mmol, 1.4 eq), EDC-HCl (0.308 g, 1.61 mmol, 1.2 eq) and DCM (3 mL) were added, the reaction capped under an atmosphere of air, and stirred at rt overnight. The solvent was removed and the residue was subjected to flash column chromatography (Hexane:EtOAc, 2:1.fwdarw.1:1) to yield 17 (0.293 g, 74%) as a colourless, viscous oil.
(288) Rf (Hexane:EtOAc, 2:1): 0.33; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 7.11 (d, J=2.0 Hz, 1H), 6.27 (d, J=2.0 Hz, 1H), 3.72 (s, 3H), 2.82 (br. s., 4H), 2.69-2.54 (m, 4H), 1.81-1.60 (m, 4H) ppm; .sup.13C NMR (100 MHz, CDCl.sub.3) δ 169.1, 168.5, 143.5, 139.1, 138.7, 102.8, 59.3, 30.5, 27.0, 25.5, 24.2, 23.8 ppm; IR (ATR) 2948, 1814, 1735, 1638, 1413, 1206, 1058, 1046 cm.sup.−1; HRMS (EI) Exact mass cald. for C.sub.14H.sub.17NO.sub.6 [M].sup.+: 295.1056, found: 295.1062.
Synthesis of CP1b-NHS (19)
(289) ##STR00049##
3-(Cyclopenta-1,3-dienyl)propanoic acid & 3-(cyclopenta-1,4-dienyl)propanoic acid (12)
(290) Ethyl 3-bromopropionate (1.65 mL, 12.9 mmol, 1 eq) was added to THF (30 mL) and cooled to −78° C. Sodium cyclopentadienide (2 M solution in THF, 6.45 mL, 12.9 mmol, 1 eq) was added dropwise over 5 min and the reaction was stirred at −78° C. for 3.5 hr. The reaction was poured into DCM (20 mL) and silica gel was added (6 g). The reaction mixture was filtered through silica gel with DCM (100 mL) and the solvent removed to yield ethyl 3-(cyclopentadienyl)propanoate isomers as a yellow oil.
(291) Spectral data matched that of literature reported data..sup.3
(292) Rf (Hexane:EtOAc, 9:1): 0.45; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 6.47-6.02 (m, 3H), 4.17-4.11 (m, 2H), 2.96 (s, 0.31H), 2.91 (d, J=1.4 Hz, 1.69H), 2.78-2.68 (m, J=1.7 Hz, 2H), 2.59-2.53 (m, 2H), 1.26 (t, J=7.1 Hz, 3H).
(293) To a solution of ethyl 3-(cyclopentadienyl)propanoate isomers (˜12.9 mmol) dissolved in EtOH (20 mL) was added a solution of NaOH (2.0 g, 50 mmol, 3.9 eq) in H.sub.2O (20 mL). The reaction stirred at rt for 15 min. The reaction mixture was poured into a separatory funnel containing H.sub.2O (50 mL) and DCM (50 mL). The aqueous layer was acidified with HCl (1 M in H.sub.2O) to pH 2 (˜70 mL). The layers were separated then the aqueous layer extracted with DCM (50 mL). The combined organic layers were washed with brine (50 mL), dried over Na.sub.2SO.sub.4, filtered, the solvent removed to yield 18 (0.50 g, 28% two steps) as a brown solid.
(294) Rf (Hexane:EtOAc, 1:2): 0.69; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 10.57 (br. s., 1H), 6.49-6.02 (m, 3H), 2.97 (d, J=1.6 Hz, 1.07H), 2.92 (d, J=1.2 Hz, 0.93H), 2.82-2.68 (m, 2H), 2.68-2.58 (m, 2H) ppm; .sup.13C NMR (100 MHz, CDCl.sub.3) δ 179.7, 179.7, 147.1, 144.9, 134.2, 134.1, 132.3, 131.1, 127.0, 126.4, 43.3, 41.3, 33.9, 33.3, 25.5, 24.7 ppm; IR (ATR) 3070, 2926, 1705, 1412, 1283, 1205, 913 cm-1; HRMS (EI) Exact mass cald. for C.sub.8H.sub.10NO.sub.2 [M].sup.+: 138.0681, found: 138.0678.
(295) ##STR00050##
2,5-Dioxopyrrolidin-1-yl 3-(cyclopenta-1,3-dienyl)propanoate & 2,5-dioxopyrrolidin-1-yl 3-(cyclopenta-1,4-dienyl)propanoate (19)
(296) THF (10 mL) was added to a vial containing 18 (0.460 g, 3.33 mmol, 1 eq). NHS (0.537 g, 4.66 mmol, 1.4 eq), EDC-HCl (0.766 g, 4.00 mmol, 1.2 eq) and DCM (6 mL) were added, the reaction capped under an atmosphere of air, and stirred at rt overnight. The solvent was removed and the residue was subjected to flash column chromatography (Hexane:EtOAc, 2:1.fwdarw.1:1) to yield 19 (0.438 g, 56%) as an eggshell powder.
(297) Rf (Hexane:EtOAc, 2:1): 0.29; .sup.1H NMR (400 MHz, CDCl.sub.3) δ 6.47-6.08 (m, 3H), 2.97 (d, J=1.2 Hz, 1.2H), 2.92 (d, J=1.6 Hz, 0.8H), 2.90-2.75 (m, 8H); .sup.13C NMR (100 MHz, CDCl.sub.3) δ 169.1, 168.2, 168.1, 145.7, 143.9, 134.4, 133.8, 132.2, 131.4, 127.7, 127.1, 43.2, 41.4, 30.8, 30.2, 25.5, 25.3, 24.5; IR (ATR) 2947, 1810, 1779, 1735, 1420, 1366, 1204, 1062, 1046 cm.sup.−1; HRMS (EI) Exact mass cald. for C.sub.12H.sub.13NO.sub.4 [M].sup.+: 235.0845, found: 235.0848.
(298) An attempt to synthesize a pentamethylcyclopentadiene (CP3) NHS-derivative using Cp*Li and ethyl 3-bromopropionate failed, instead undergoing elimination to ethyl acrylate. The reaction of CP*Li with methyl bromoacetate was successful, but after ester hydrolysis and reacidification the compound underwent an unexpected cyclization. Our third strategy used Boydston's (1,2,3,4,5-pentamethylcyclopenta-2,4-dienyl)methanol,.sup.1 which was reacted with succinic anhydride to produce the intermediate acid, which was used without further purification. The reaction with EDC-HCl and N-hydroxysuccinimide yielded NHS ester 13. The furan (F2) NHS-derivative's design and synthesis were inspired by Sheppard's work on 3-alkoxyfurans..sup.2 The lithium salt of 3,3-diethoxyprop-1-yne was added to methyl 6-oxohexanoate to form alcohol 14, which was cyclized using catalytic gold (I) in methanol to yield 3-methoxyfuran 15. The ester of 15 was hydrolyzed then reacted with EDC-HCl and N-hydroxysuccinimide to yield NHS ester 17
(299) The synthesis of a CP1 NHS-derivative that doesn't contain an internal ester (CP1b) began with the reaction of NaCP with ethyl 3-bromopropionate, then ester hydrolysis to yield acid 12. The reaction with EDC-HCl and N-hydroxysuccinimide yielded NHS ester 19. Structural differences between CP1 and CP1 b are shown in
(300)
Example 14. Evaluation of ADCs Prepared with Linker-Modified Antibody
(301) The stability and potency of ADCs generated by Diels-Alder conjugation of AM-MMAE to linker-modified antibody were evaluated. Diels-Alder conjugates were compared to cysteine-conjugates.
(302) Materials. All antibodies (IgG1 format) were expressed and purified using standard molecular biology methods. All reagents were purchased from commercial vendors unless noted otherwise. Furan-2-ylmethyl Succinamic acid NHS ester (F1-NHS) and maleimidocaproyl-valine-citrulline-p-aminobenzyloxycarbonyl-monomethyl-auristatin-E (AM-MMAE) were purchased from SynChem, Inc. (Elk Grove Village, Ill.).
(303) Preparation of mAb-linker conjugates: Diene functionality was randomly incorporated into antibodies by reaction of the NHS ester-containing linkers 17 and 19 (F2 and CP1b) described above with lysine amines. Degree of mAb modification was controlled by the amount of NHS-linker used in the reaction and different linker densities were targeted depending on the experiment. A general procedure for modification of mAb with CP1 is described as follows: First, mAb solution was adjusted to 5 mg/mL (3 mL, 15 mg mAb, 100 nmol, 1 eq.) with PBS pH 7.2 followed by addition of 10% v/v 1 M NaHCO.sub.3. This solution was chilled on ice and 30 μL CP1b-NHS (10 mM stock in DMAc, 300 nmol, 3 equivalents, also termed CP1-linker) was added. The reaction proceeded on ice for 5 minutes followed by reaction at room temperature for 1 h with continuous mixing. Reacted mAb was purified by dialysis (Slide-A-Lyzer, 10 kDa MWCO) against PBS, 1 mM EDTA, pH 7.4, 4° C. for 24 h. CP1-linker introduction was quantified by intact deglycosylated mass spectrometry as described below.
(304) ##STR00051##
(305) Preparation of ADCs; Antibody drug-conjugates were prepared from Herceptin (on-target) or IgG-1 isotype control-1 (off-target) mAbs using both Diels-Alder conjugation via linkers and direct conjugation to antibody cysteine thiols. Diels-Alder ADCs were prepared from CP1b-NHS (compound 19) and F2-NHS (compound 17) linker-modified antibodies using the same general procedure described for mAb-CP1b-linker as follows: mAb-CP1b-linker (10 mg, 67 nmol, 1 equivalent) was diluted to 4.27 mg/mL with PBS, pH 7.4, followed by addition of DMSO (493 μL) and 1 M sodium phosphate monobasic (247 μL) to yield ˜20% and 10% v/v solutions respectively. AM-MMAE (53.3 μL of 10 mM stock in DMSO, 530 nmol, 8 equivalents) was added to the antibody solution and the reaction continued at room temperature with mixing for 4 h. N-acetyl cysteine (43 μL of a 100 mM solution in water, 4.3 μmol, 64 equivalents) was added to quench unreacted maleimides. ADC was purified from the reaction mixture using CHT chromatography. ADC solution was diluted 3-fold with distilled water and loaded onto a Bio-Scale Mini Cartridge CHT Type II 40 μm media column. ADC was eluted with a gradient from buffer A (10 mM phosphate, pH 7.0) to buffer B (10 mM phosphate pH 7.0 containing 2M NaCl) over 25 minutes at a flow rate of 5 mL/min. After CHT chromatography ADC sample was buffer exchanged to PBS using a slide-a-lyzer cassette at 4° C. The same procedure was followed for ADCs prepared with mAb-F2-linker constructs, with the exception that the AM-MMAE conjugation reaction continued for 24 h at room temperature. Note that diene content for each mAb prior to reaction with AM-MMAE is provided in Table 14.1. 8 equivalents of AM-MMAE relative to mAb used for the conjugation reaction corresponds to approximately 2 molar equivalents of AM-MMAE relative to diene.
(306) ADCs were also prepared by conjugation of AM-MMAE to cysteine thiols contained in the antibody hinge region. First, antibody (10 mg, 67 nmol, 1 equivalent) solution was adjusted to 2.5 mg/mL with PBS containing 1 mM EDTA. Next, TCEP (10 μL of 50 mM solution in water, 500 nmol, 7.5 equivalents) was added to reduce hinge disulfides, and the mixture was incubated at 37° C. with mixing for 1 h. Next, DMSO (410 μL, 10% v/v final concentration in reaction) was added and the reaction continued at room temperature with mixing for 1 h.
(307) N-acetyl cysteine was added to quench unreacted maleimide groups and ADC was purified by CHT chromatography and dialysis as described above. ADCs prepared by conjugation to hinge cysteines are denoted with Cys in the name, for example: Herceptin-Cys-MMAE. rRP-HPLC analysis: For each analysis, the antibodies and ADCs were reduced at 37° C. for 20 minutes using 42 mM dithiothreitol (DTT) in PBS pH 7.2. 10 μg of reduced antibodies and ADCs was loaded onto a PLRP-S, 1000 Å column (2.1×50 mm, Agilent) and eluted at 40° C. at a flow rate of 0.5 mL/min with a gradient of 5% B to 100% B over 60 minutes (mobile phase A: 0.1% trifluoroacetic acid in water, and mobile phase B: 0.1% trifluoroacetic acid in acetonitrile). Percent conjugation was determined using integrated peak areas from the chromatogram.
(308) Size exclusion chromatography analysis: SEC analysis was performed using an Agilent 1100 Capillary LC system equipped with a triple detector array (Viscotek 301, Viscotek, Houson, Tex.); the wavelength was set to 280 nm, and samples (50 μg) were run on a TSK-GEL G3000SWXL column (Toso Bioscience LLC, Montgomeryville, Pa.) using 100 mM sodium phosphate buffer, 10% isopropyl alcohol, pH 6.8 at a flow rate of 1 mL/min.
(309) Serum stability analysis: ADCs were incubated in rat serum to challenge the stability of the antibody-payload linkage. ADCs were added to normal rat serum (Jackson Immunoresearch) to achieve a final concentration of 0.2 mg/mL (1.33 μM antibody), with the total volume of ADC solution added to serum less than 10%. The ADC-serum mixture was sterile filtered and an aliquot was removed from this mixture and frozen as a t=0 control. Remaining sample was then further incubated at 37° C. in a sealed container for 7 d. Conjugated and unconjugated human antibody was recovered from rat serum by immunoprecipitation using Fc-specific anti-human IgG-agarose resin (Sigma-Aldrich). Resin was rinsed twice with PBS, once with IgG elution buffer, and then twice more with PBS. ADC-mouse serum samples were then combined with anti-human IgG resin (100 μL of ADC-serum mixture, 50 μL resin slurry) and mixed for 15 minutes at room temperature. Resin was recovered by centrifugation and then washed twice with PBS. Washed resin was resuspended in 100 μL IgG elution buffer (Thermo Scientific) and further incubated for 5 minutes at room temperature. Resin was removed by centrifugation and then 20 μL of 10×glycobuffer 1 (New England Biolabs) was added to the supernatant. Recovered human antibody solution was sterile filtered, and incubated with EndoS for 1 h at 37° C. Deglycosylated mAbs were then reduced with TCEP and analyzed by LC/MS as described above. Percent conjugated antibody was determined from peak heights of mass spectra as described in Example 12.
(310) Mass spectrometry analysis: Samples were analysed as described in Example 1.
(311) In vitro cytotoxicity analysis: SKBR3 and N87 cancer cell lines were obtained from American Type Culture Collection (ATCC). Cells were maintained in RPMI 1640 media (Life Technologies) supplemented with 10% heat-inactivated fetal bovine serum (HI-FBS) (Life Technologies) at 37° C. in 5% CO.sub.2. SKBR3 and NCI-N87 cells harvested in exponential growth phase were seeded in 96-well culture plates at 2500 and 2000 cells/well and allowed to adhere overnight. Cells were then treated on the following day with ADCs at 4-fold serial dilutions from 4000 and 64,000 ng/mL (9 concentrations) in duplicate. The treated cells were cultured for 6 days and cell viability was determined using the CellTiter-Glo Luminescent Viability Assay (Promega) following the manufacturer's protocol. Cell viability was calculated as a percentage of untreated control cells. IC.sub.50 values were determined using logistic non-linear regression analysis with Prism software (GraphPad).
(312) Tumor growth inhibition in vivo: Herceptin-MMAE ADCs prepared with F2 and CP1 b linker-modified mAbs were further evaluated for antitumor activity in vivo in a subcutaneous N87 xenograft model in mice. Tumors were prepared by inoculation of N87 cells (5 million N87 cells in 50% Matrigel) subcutaneously into 4-6 week old female athymic nude mice. When tumors reached approximately 200 mm.sup.3, mice were randomly assigned into groups, 5 mice per group. ADCs were administered IV at the indicated doses and dosed at day 5 post cell inoculation. Tumor dimensions (long axis and short axis) were measured twice weekly with calipers. Tumor volume was calculated using the equation:
V=½a×b.sup.2
Where,
a=tumor long axis in mm
b=tumor short axis in mm
(313)
(314)
(315)
(316)
(317)
(318)
(319)
(320)
(321) TABLE-US-00016 TABLE 14.1 Summary of linker-conjugated and cysteine-conjugated ADCs Linker MMAE Conjugation DAR DAR MMAE DAR Monomer mAb Linker reaction (MS) (MS) (rRP-HPLC) (%) Herceptin CP1b Diels-Alder 4.1 3.9 3.5 98.7 IgG isotype CP1b Diels-Alder 3.7 3.2 3.4 98.4 control Herceptin F2 Diels-Alder 4.0 3.5 4.5 97.7 IgG isotype F2 Diels-Alder 3.8 3.2 3.5 98.1 control Herceptin none Michael N/A 3.2 3.5 97.4 addition IgG isotype none Michael N/A 2.9 3.1 98.5 control addition .sup.aADCs prepared without linkers were conjugated to native cysteines
(322) TABLE-US-00017 TABLE 14.2 In vitro potency of linker-conjugated and cysteine-conjugated ADCs. MMAE N87 IC.sub.50 SKBR3 IC.sub.50 ADC DAR.sup.a (ng/mL) (ng/mL) Herceptin-CP1b-MMAE 3.7 2 2 IgG isotype control-CP1b-MMAE 3.3 8230 1811 Herceptin-F2-MMAE 4 3 1.6 IgG isotype control-F2-MMAE 3.4 9990 ~2000 Herceptin-Cys-MMAE 3.4 3.6 2.8 IgG isotype control-Cys-MMAE 3 8177 >2000 .sup.aaverage DAR calculated from MS and rRP-HPLC values reported in Table 14.1
(323) ADCs were prepared via Diels-Alder reaction or Michael addition of maleimido-MMAE to Herceptin and IgG isotype control mAbs. Diels-Alder reactive groups were introduced onto lysine amines via crosslinkers followed by reaction with AM-MMAE whereas Michael addition of AM-MMAE occurred with native cysteine thiols (termed Cys-ADCs). Both conjugation methods yielded heterogeneous ADCs with a drug content of 3-4 MMAE drugs per mAb.
(324) Diels-Alder addition of AM-MMAE to CP1b- and F2 linker-modified mAb was efficient, with nearly quantitative conversion confirmed by mass spectrometry. Furthermore, modification of mAbs with cyclopentadiene or methoxy-furan linkers and subsequent attachment of AM-MMAE by Diels-Alder reaction did not increased aggregate content in conjugate products as indicated by SEC analysis. Overall, ADCs produced by Diels-Alder conjugation were of high quality.
(325) Analysis of ADC stability in rat serum by mass spectrometry demonstrated that Diels-Alder constructs prepared with mAb-CP1 b-linker and mAb-F2-linker were more stable than constructs generated by Michael addition to cysteine thiols. Incubation in rat serum for 7 days at 37° C. resulted in 60% drug loss for mAb-cysteine ADC, whereas mAb-CP1b-linker and mAb-F2-linker ADCs showed less than 10% drug loss under the same conditions. Herceptin mAb-CP1b-linker and Herceptin mAb-F2-linker ADCs were potent inhibitors of cell proliferation towards Her2 positive N87 and SKBR3 cell lines, with IC.sub.50 values similar to the corresponding cysteine-linked ADC. Non-targeting isotype control ADCs were 2000-4000-fold less potent than on target Herceptin constructs, with similar in vitro potencies observed for Diels-Alder and cysteine ADC constructs. Finally, Herceptin ADCs prepared with CP1 b and F2 linker-modified mAbs were potent inhibitors of tumor growth in vivo. Complete tumor stasis for 30 days was observed in N87 subcutaneous tumor models at an ADC dose of 3 mg/kg. This result confirms that ADCs produced by Diels-Alder reaction are sufficiently stable in vivo to elicit a therapeutic effect.
Example 15. Comparison of Diels-Alder Reaction Rate Constants in Aqueous Buffer and Organic Solvent
(326) Kinetics of small molecule diene-maleimide reactions in organic conditions were determined for comparison with antibody-based reactions in aqueous conditions. Reaction rate constants between dienes on linker-modified mAbs and maleimide were determined for; mAb-CP1b-linker, mAb-CP2-linker, mAb-CP3-linker and mAb-F2-linker.
(327) Determination of diene-maleimide reaction rate in organic solvents: Diene and N-ethylmaleimide in CDCl.sub.3 were combined in an NMR tube (final concentration 0.01 M each) and monitored by .sup.1H NMR at room temperature. The concentration of starting material [A] was calculated using the integration of N-ethylmaleimide's ethyl peaks (3.59 or 1.20 ppm) and the DA-conjugate's ethyl peak(s) (typically 4.40 or 1.05 ppm).
[A]=0.01 M*(integration of starting material)/(integration of starting material+product)
(328) The inverse concentration (1/[A]) was plotted against time (s). The second order reaction rate (M.sup.−1s.sup.−1) was obtained from the best fit line. The average rate and standard deviation of three experiments was used.
(329) Preparation of mAb-linker conjugates: Diene functionality was randomly incorporated into antibodies by reaction of the NHS ester-containing linkers CP1b-linker (compound 19), CP2-linker (compound 12), CP3-linker (compound 13), and F2-linker (compound 17) described in Example 9 and Example 14. Reaction of CP3-NHS with mAb is shown in Scheme 15.1. Degree of mAb modification was controlled by the amount of NHS-linker used in the reaction and different linker densities were targeted depending on the experiment. The number of linkers (and thus dienes) per mAb were determined by intact deglycosylated mass spectrometry as described in Example 1.
(330) ##STR00052##
(331) Reaction of linker-modified mAb with maleimido-MMAEs: Dienes contained in linker-modified mAbs were reacted with 1 molar equivalent AM-MMAE (diene:maleimide) in aqueous buffer as described in Example 10.
(332) Calculation of mAb-linker diene-maleimide reaction rate constants: Second order rate constants for reaction of maleimido-MMAEs with dienes in mAb-linkers were determined from peak intensities in deglycosylated reduced mass spectra as described in Example 10. For one sample only heavy-chain peaks were analyzed as described in Example 4.
(333) TABLE-US-00018 TABLE 15.1 Summary of reaction rates of AM-MMAE with diene linker-modified mAbs in aqueous buffer and N-ethyl maleimide with diene linkers in CDCl.sub.3. Linker Diels-Alder Aqueous rate mAb-linker Diels-Alder rate in buffer.sup.a,b Rate in CDCl.sub.3.sup.c acceleration k.sub.2(H.sub.2O) Half-life k.sub.2(CDCl.sub.3) k.sub.2(H.sub.2O)/ Experiment mAb-Linker LAR (M.sup.−1s.sup.−1).sup.d (min) (1000 × M.sup.−1s.sup.−1).sup.e k.sub.2(CDCl.sub.3) 1 IgG-CP1-linker.sup.f 3.7 35.7.sup.h 14 97 ± 7.sup.f 368 2 IgG-CP1-linker.sup.f 3.74 77 7 794 3 IgG-CP1-linker.sup.f 3.21 119 4 1227 4 IgG-CP1-linker.sup.f 3.21 116 5 1196 1 IgG-CP2-linker 3.29 2.6 214 8.7 ± 0.6 299 2 IgG-CP2-linker 3.29 2.7 215 310 3 IgG-CP2-linker 2.63 3.0 192 345 4 IgG-CP2-linker 2.63 2.3 318 264 1 IgG-CP3-linker 3.04 10.2 59 12.1 ± 0.1 843 2 IgG-CP3-linker 3.04 16.1 39 1331 3 IgG-CP3-linker 2.81 24.1 24 1992 1 IgG-F2-linker 2.95 6.6 94 12.6 ± 0.7 524 2 IgG-F2-linker 2.95 2.3 210 183 3 IgG-F2-linker 3.03 5.5 117 437 1 IgG-furan 2.5 ND.sup.g >1200 ~0.1 ND .sup.a)All conjugation reactions were performed in PBS supplemented with 100 mM sodium phosphate monobasic, 20% DMSO, pH 5.5. .sup.b)All reactions were performed with 1 eq AM-MMAE relative to diene. mAb concentration was 1.3 mg/mL ± 0.35 mg/mL for all reactions. .sup.c)All reactions were performed in CDCl.sub.3 with diene-linker (0.01M) and N-ethylmaleimide (0.01M) at room temperature. .sup.d)k.sub.2(H.sub.2O) was calculated from the concentration of unreacted mAb-linker peaks (deglycosylated and reduced mass spectrometry analysis). Both heavy and light chains were analyzed. .sup.e)k.sub.2(CDCl.sub.3) was calculated from the integration of diagnostic peaks in .sup.1H NMR spectra. The values are the average of 3 runs and the standard deviation. .sup.f)CP1b linker was used. .sup.g)ND, not determined. .sup.h)Only heavy chain was analysed.
(334) Comparison of reaction rate constants between maleimide compounds and diene compounds in antibody/aqueous conditions and small molecule/organic solvent conditions demonstrates several key features of this reaction for bioconjugation applications. First, unmodified furan is insufficiently reactive for conjugation of maleimide compounds to antibodies under typical bioconjugation conditions. This is evidenced by lack of conjugation of AM-MMAE to IgG-furan, which showed minimal reaction after 20 h and also the ˜100-1000-fold decrease in rate constant for reaction with maleimide in organic conditions compared to other dienes. Second, acceleration of the Diels-Alder reaction in aqueous conditions in the context of antibody conjugation was confirmed. Reaction rate acceleration is crucial for practical application of this chemistry for production of bioconjugates, consideration of organic condition rate constants alone would make the Diels-Alder reaction unattractive for all dienes described here. For example, cyclopentadiene contained in CP1-linker exhibited a reaction half-life of approximately 10 minutes in aqueous antibody-based reaction with maleimide (AM-MMAE) whereas the corresponding organic-phase reaction would require 3.6 days. Finally, results demonstrate that diene reactivity is tunable, where modification of chemical structure can increase or decrease reaction rate with dienophile. Altogether, cyclopentadiene and modified furan functional groups are amenable to efficient bioconjugation reactions between antibodies and maleimido compounds under mild conditions at antibody concentrations in the ˜1-2 mg/mL range, whereas simple unmodified furan is not.
Example 15. ADC Production with 1C1-K274CP1 NNAA mAb and AZ1508 Drug-Linker
(335) The feasibility of preparing ADCs with CP1-NNAA incorporated into position K274 of an antibody with AZ1508 drug-linker was assessed.
(336) Antibody generation. CP1-NNAA was incorporated into 1C1 antibody using the methods described in Example 6.
(337) Conjugation of 1C1 K274CP1-mAb with AZ1508: K274CP1 NNAA-mAb (0.4 mg, 2.7 nmol, 1 equivalent) was adjusted to 3 mg/mL with PBS (0.133 mL). DMSO (27 μL) and 1 M sodium phosphate, monobasic (13 μL) was added to yield ˜20% and 10% v/v solution, respectively. AZ1508 (5 μL of 10 mM stock in DMSO, 13 nmol, 5 equivalents) was added to 1C1 K274CP1-mAb solution and the mixture was vortexed briefly. The reaction proceeded at room temperature for 17 h with continuous mixing. N-acetyl cysteine (1.1 μL of 100 mM, 108 nmol, 40 equivalents) was added and the solution was incubated for an additional 15 min to quench unreacted maleimide groups. Samples were then diluted 3-fold with water and purified using CHT chromatography. Samples were subsequently analyzed by reduced mass spectrometry and SEC as described in Examples 6 and 12.
(338) ##STR00053##
(339) Characterization of 1C1 K274CP1-NNAA ADCs: Samples were analyzed by reduced mass spectrometry and SEC as described in Examples 6 and 12. In vitro activity in PC3 cells was performed as described in Example 12.
(340)
(341)
(342)
(343) TABLE-US-00019 TABLE 15.1 Summary of 1C1 K274CP1-NNAA AZ1508 ADC properties.sup.a,b,c Conjugated Observed Calculated efficiency Δ mass Δ mass Monomer EC50.sup.e (%) (AMU) (AMU) DAR.sup.d (%) (ng/mL) 96 +1094.1 +1092.4 1.91 95 9.75 .sup.aall conjugation reactions performed at pH 5.5, 20% DMSO, 22° C. and 3 mg/mL 1C1 K274CP1-NNAA mAb. CP1-NNAA was incorporated into position K274 in place of lysine .sup.bthe molar ratio of AZ1508:CP1 diene used was 2.5:1 .sup.ccalculated from peak intensities of reduced mass spectra .sup.dDAR = drug to antibody ratio .sup.eDetermined in EphA2 receptor positive PC3 cells
(344) Reactivity of CP1-NNAA diene towards AZ1508 following incorporation at position K274 of the 1C1 antibody was confirmed. ADC product was of high quality, with >95% conjugation and very little aggregate. The resulting ADC was active towards receptor positive PC3 cells.
Example 16. Generation of ADCs with 1C1 S239CP2-NNAA, 1C1 K274CP2-NNAA and 1C1 N297CP2-NNAA Antibodies with AZ1508 Drug-Linker and Comparison with Analogous Cysteine-Linked Site-Specific AZ1508 ADCs
(345) ADCs bearing AZ1508 drug-linker were prepared with 1C1 antibodies incorporating CP2-NNAA at positions S239, K274, and N.sub.297 by mutation of the native amino acid codon to an amber stop codon in the expression plasmid. CP2-NNAA was incorporated into each position on separate antibodies.
(346) Antibody generation: CP2-NNAA was incorporated into 1C1 antibodies and expressed using the methods described in Example 6. Cysteine was incorporated into 1C1 antibodies by site-directed mutagenesis using standard molecular biology techniques.
(347) Conjugation of 1C1 CP2-NNAA mAbs with AZ1508: The same conjugation method was performed for all three CP2-NNAA antibody constructs, using the procedure described in Example 12, with the only difference being AM-MMAE was replaced with AZ1508. Note that drug-linker conjugation is achieved by simply mixing AZ1508 with CP2-NNAA mAb followed by mixing.
(348) ##STR00054##
(349) Conjugation of 1C1 cysteine-engineered mAbs with AZ1508. Site-specific cysteine-linked AZ1508 ADCs were prepared using the same method described in Example 12 with the only difference being AM-MMAE was substituted with AZ1508. Note that cysteine-mAb must be reduced, dialyzed, and oxidized prior to addition of AZ1508.
(350) Characterization of 1C1 CP2-NNAA ADCs and 1C1 cysteine-engineered ADCs. Samples were analyzed by SDS-PAGE, reduced mass spectrometry, HIC, rRP-HPLC and SEC as described in Examples 6 and 12. In vitro activity in cultured PC3 cells was performed as described in Example 12.
(351)
(352)
(353)
(354)
(355)
(356)
(357)
(358) TABLE-US-00020 TABLE 16.1 Summary of 1C1 CP2-NNAA and cysteine mAb ADCs.sup.a,b mAb ADC ADC ADC ADC ADC Titer monomer DAR DAR Δ mass monomer EC50 Position Mutation mg/L (%).sup.c MS.sup.d,e HIC.sup.d,f (AMU).sup.g (%).sup.c (ng/mL).sup.h S239 CP2-NNAA 76-82 99 1.96 ND +1089.5 97 9 Cys 792 96 1.96 ND +1093.7 97 7 K274 CP2-NNAA 54-69 92 1.96 1.87 +1089.0 >99 7 Cys 468 97 1.88 1.84 +1092.9 97 8 N297 CP2-NNAA 42 79 1.96 ND +1091.0 >99 10 Cys 594 93 1.60 ND +1092.3 86 4 .sup.aall CP2 NNAA conjugation reactions performed at pH 5.5, 20% DMSO, 22° C. and 3 mg/mL mAb. .sup.bthe molar ratio of AZ1508:CP2 used was 2.5:1 .sup.cdetermined by peak areas in SEC traces .sup.dDAR = drug to antibody ratio .sup.ecalculated from peak intensities of reduced mass spectra .sup.fdetermined by peak areas following analysis of intact ADCs .sup.gdetermined by reduced mass spectra .sup.hDetermined in EphA2 receptor positive PC3 cells.
(359) CP2-NNAA was incorporated into three positions of the 1C1 antibody and reactivity towards AZ1508 was confirmed. Note that CP2-NNAA was incorporated into each position on separate antibodies. ADCs prepared with CP2-NNAA mAbs were of high quality, with >95% conjugation and very little aggregate. Resulting CP2-NNAA AZ1508 ADCs were active towards receptor positive PC3 cells.
(360) Comparison of CP2-NNAA ADCs with corresponding cysteine-engineered antibodies revealed that position N297 is amenable to mutation with CP2-NNAA but not cysteine. Introduction of cysteine at this position results in disulfide scrambling during the conjugation procedure, as evidenced by SDS-PAGE results. No disulfide scrambling was observed for the N297CP2-NNAA ADC. This demonstrates that CP2-NNAA can be introduced into positions not suitable for cysteine incorporation, where cysteine may impact native disulfides in the antibody framework.
Example 17. Stability of CP1-NNAA and CP2-NNAA AZ1508 ADCs in Mouse and Rat Serum
(361) Serum stability of AZ1508 ADCs prepared by Diels-Alder conjugation was evaluated. Stability of ADCs was assessed relative to the chemical bond between the antibody and payload (i.e. the Diels-Alder adduct). Stability of analogous cysteine-linked (thiosuccinimide) ADCs is also presented for comparison.
(362) Method: 1C1 AZ1508 ADCs were incubated in mouse serum or rat serum, ex vivo for 7 d at 37° C., recovered by immunocapture, and analyzed by mass spectrometry as described in Example 7. Relative amounts of conjugated and unconjugated antibody were determined by peak heights in mass spectra as described in Example 7. For mouse serum-incubated samples, deacetylation of AZ1508 was observed. Deacetylated AZ1508 was considered as a conjugated species for calculation of DAR.
(363)
(364)
(365)
(366)
(367) Analysis of 1C1 CP2-NNAA AZ1508 ADCs following incubation in rat or mouse serum for 7 d at 37° C. demonstrated that the Diels-Alder adduct was stable, as no drug loss was observed. 1C1 CP1-NNAA AZ1508 prepared at position K274 also showed no drug loss in rat serum after 7 d incubation at 37° C., however, significant drug loss was observed for the analogous cysteine-linked ADCs subjected to the same conditions. Cysteine-linked ADC stability was position-dependent, with position S239 being stable and positions K274 and N297 were not. In the case of cysteine-linked ADCs, AZ1508 is coupled to the antibody via a thiosuccinimide linkage, which may undergo the retro-Michael deconjugation reaction, leading to drug loss. This process impacts exposed positions more than buried positions, thus position-dependent stability is observed for cysteine-linked ADCs. Diels-Alder adducts on the other hand were stable at positions unstable for thiosuccinimides, for both CP1 and CP2 dienes. Thus, Diels-Alder conjugation to cycopentadiene NNAAs represents an advantage over thiol-based conjugation strategies in terms of conjugate stability.
Example 18. Reaction Kinetics of 1C1 mAb Containing CP1-NNAA or CP2-NNAA with AZ1508
(368) Reactivity of diene NNAAs were evaluated following incorporation into positions S239, K274, or N297 in the 1C1 antibody framework.
(369) Methods: 1C1 CP1-NNAA or CP2-NNAA antibodies (3 mg, 1.3 mg/mL mAb, 17.4 μM diene, 1 equivalent) were reacted with AZ1508 (4 μL of 10 mM stock in DMSO, 40 nmol, 1 equivalent) in 0.1 M sodium phosphate, 0.15 M NaCl, 20% DMSO, pH 5.5, 22° C. Aliquots (100 μL) were removed from the reaction mixture at predetermined timepoints and N-acetyl cysteine (3 μL of 100 mM in water, 8 equivalents) was added followed by incubation for 15 minutes at room temperature to quench unreacted maleimides. Samples were then purified using PD Spintrap G-25 devices (GE Healthcare Life Sciences) to remove small molecule components from the mixture and subsequently analyzed by reduced glycosylated mass spectrometry. Mass spectrometry analysis procedures and kinetic constant calculations are described in Examples 5 and 10.
(370)
(371)
(372) TABLE-US-00021 TABLE 18.1 Summary of kinetic data for reaction of 1C1 CP1-NNAA and CP2-NNAA mAbs with AZ1508..sup.a,b,c k.sub.2 T.sub.1/2 Position Mutation (M.sup.−1s.sup.−1) R.sup.2d (min).sup.e K274 CP1-NNAA 73.2 ± 6.9.sup.b 0.99 12 ± 1 S239 CP2-NNAA 2.6 ± 0.5.sup.c 0.99 383 ± 84 K274 CP2-NNAA 1.8 ± 0.4.sup.c 0.99 545 ± 108 N297 CP2-NNAA 5.4 ± 1.1.sup.c 0.99 183 ± 41 .sup.aAll conjugation reactions were performed in PBS supplemented with 100 mM sodium phosphate monobasic, 20% DMSO, pH 5.5. .sup.bThe molar ratio of AZ1508:CP1-NNAA or CP2-NNAA diene used was 1:1. The mAb concentration was 1.3 mg/mL (17.3 μM diene) for all reactions. .sup.cReaction of deine was monitored by reduced glycosylated mass spectrometry. .sup.ddetermined from bestfit line of inverse diene concentration (M) vs time (s) plot. .sup.ecalculated using the half-life equation shown in Examples 5 and 10.
(373) Antibodies bearing CP1-NNAA or CP2-NNAA reacted with the maleimide containing drug-linker AZ1508 at 1:1 molar equivalent of diene:maleimide. Reaction half-lives of 12 minutes for CP1-NNAA mAb and 3-10 hours for CP2-NNAA mAb. Final conversions achieved after the 48 h measurement period ranged from 87-100%.
Example 19. Antitumor Activity of ADCs Prepared with CP2-NNAA mAb and AZ1508
(374) ADCs prepared with 1C1 CP2-NNAA and AZ1508 drug-linker were evaluated for their ability to inhibit tumor growth of PC3 xenografts in mice.
(375) ADC preparation: ADCs were prepared with 1C1 mAbs as described in Example 16. Non-EphA2 binding ADCs were prepared with isotype control antibody (termed R347). CP2-NNAA was incorporated into positions S239 and N297 for 1C1 mAbs and position S239 for R347 mAb.
(376) In vivo methods: Tumor growth inhibition studies were performed at Charles River Discovery Services North Carolina (CR Discovery Services) in accordance with the recommendations of the Guide for Care and Use of Laboratory Animals with respect to restraint, husbandry, surgical procedures, feed and fluid regulation, and veterinary care. PC3 xenograft tumor models were established in mice by inoculation of PC3 cells (10 million cells in 50% Matrigel) subcutaneously into 8-9 week old female athymic nude mice. Seventeen days later, designated as day 1 of the study, tumor volumes reached˜150-200 mm.sup.3 and mice were randomly assigned into groups, 8 mice per group. On day 1 of the study, dosing was initiated and ADC was administered at 3 mg/kg via tail vein injection. ADC was dosed once weekly at 3 mg/kg over three weeks for a total of three doses. Tumor dimensions (long axis and short axis) were subsequently measured twice weekly with calipers. Tumor volume was calculated using the equation in Example 14.
(377)
(378) 1C1 CP2-NNAA AZ1508 ADCs were effective at inhibiting tumor growth in EphA2-positive PC3 xenograft models in mice for at least 60 days in mice following the first injection of ADC. The off-target ADC prepared with non-binding R347 mAb was not as potent as on-target ADCs.
Example 20. Conjugation of CP1-NNAA Antibody to Maleimide Functionalized Nanoparticles
(379) 1C1 K274CP1-NNAA antibody was reacted with 60 nm maleimide-functionalized gold nanoparticles.
(380) Method: Maleimide-functionalized 60 nm gold nanoparticles were prepared from a commercial kit (Sigma Aldrich, catalogue #9009465) according to the manufacturer's instructions. First, the lyophilized maleimide-functionalized gold nanoparticle product was resuspended in 100 μL of reaction buffer provided in the kit. Next, solutions of wild-type (WT) or K274CP1-NNAA 1C1 antibodies were prepared at 0.5 mg/mL in PBS. Nanoparticle solution (10 μL) and antibody solution (10 μL) were combined and mixed by pipetting up and down several times. The conjugation reaction continued for 2 h at 25° C. followed by light scattering analysis using a Zetasizer-Nano ZS (Malvern Instruments, UK). Each conjugation reaction was performed in triplicate.
(381) ##STR00055##
(382)
(383) TABLE-US-00022 TABLE 1 Summary of light scattering data for nanoparticles and nanoparticle-mAb conjugates. Size Sample.sup.a (d. nm).sup.b PDI 60 nm Au nanoparticle control sample 1 61.5 0.136 60 nm Au nanoparticle control sample 2 57.7 0.146 60 nm Au nanoparticle control sample 3 60.8 0.112 1C1 WT mAb reaction sample 1 63.3 0.102 1C1 WT mAb reaction sample 2 59.5 0.115 1C1 WT mAb reaction sample 3 59.0 0.121 1C1 K274CP1-NNAA mAb reaction sample 1 71.8 0.07 1C1 K274CP1-NNAA mAb reaction sample 2 73.4 0.06 1C1 K274CP1-NNAA mAb reaction sample 3 70.8 0.08 .sup.aEach reaction sample represents an independent experiment. .sup.bSize is reported as the number average diameter in nanometers.
(384) Antibody incorporating CP1-NNAA conjugated to maleimide-functionalized nanoparticles via a Diels-Alder reaction.