Article pubs.acs.org/ac
Development of a Native Nanoelectrospray Mass Spectrometry Method for Determination of the Drug-to-Antibody Ratio of Antibody−Drug Conjugates Jia Chen, Sheng Yin, Yongjian Wu, and Jun Ouyang* Protein Analytical Chemistry, Genentech, Inc., South San Francisco, California 94080, United States ABSTRACT: Antibody−drug conjugates (ADCs), an increasingly important therapeutic modality for targeted cancer treatment, have been studied using many analytical methods. A class of ADCs that utilize the reduced interchain disulfide cysteine residues for drug attachment has attracted particular interest in drug development. One challenge in analytical characterization of this class of ADCs is that the intact mass information of the ADC molecule is not attainable using conventional reversed-phase liquid chromatography−mass spectrometry methods. In this paper, we report a mass spectrometry (MS) method engaging enzymatic digestion, nanoelectrospray ionization (nano-ESI), and native MS to achieve direct determination of the intact mass and, furthermore, to calculate the average drug-to-antibody ratio (DAR) of the cysteine-linked ADCs. The novel aspects of this method lie in the application of a nano-ESI technique and, more significantly, the utilization of limited enzymatic digestion with a cysteine protease as compared to the recently published method by Valliere-Douglass et al.1 In summary, this novel native nano-ESI MS method in combination with limited enzymatic digestion provides a sensitive method for direct DAR determination and possesses great potential in studying low-abundance ADC analytes such as those from animal or human in vivo investigations.
A
DAR also has a direct relationship with drugs being delivered in vivo;12 therefore, DAR is a key attribute related to ADC drug efficacy and patient safety. In lysine- or engineered cysteine-linked ADCs, the interchain disulfide bonds between the light and heavy chains of the mAb are not disrupted, and DAR can be readily characterized using liquid chromatography operated with mobile phases containing organic solvent coupled with mass spectrometry (LC− MS).13,14 However, for ADCs utilizing reduced interchain cysteine residues as the drug attachment sites, the light chain (LC) and heavy chain (HC) of ADCs are associated through noncovalent interactions (Figure 1). Consequently, the structure of this cysteine-linked ADC molecule becomes more fragile as more drugs are conjugated, and conventional reversed-phase (RP) LC−MS disrupts the noncovalent ADC complexes into antibody fragments. Currently, for such ADCs, the method of choice for DAR determination is hydrophobic interaction chromatography (HIC), which provides the drug distribution information by resolving and eluting ADCs in an order from low to high drug load.12,15 HIC is, in general, not amenable to MS analysis due to its buffer components; as a result, direct molecular weight confirmation of each peak is not feasible. In addition, HIC requires a relatively high amount of sample (10−50 μg); therefore, it is not optimal for characterization of limited samples from animal and human studies
ntibody−drug conjugates (ADCs) as potential cancer therapeutics have been progressively investigated since the concept was first introduced.2−6 In August 2011, Seattle Genetics’ ADCETRIS (brentuximab vedotin) for treatment of Hodgkin’s lymphoma became the first ADC approved by the FDA through a Biological License Application, which brought more promise and excitement to the ADC field.7,8 ADCs are complex conjugates that unite cytotoxic small-molecule drugs and large-molecule monoclonal antibodies (mAbs), exploiting the advantages from both moieties. The high binding specificities between mAbs and the targeted receptors make ADC an ideal delivery vehicle for the cytotoxic small-molecule drug. The ADC first recognizes and binds the receptor on the surface of the target cells. Then, together with the receptor, the ADC is internalized into the cancer cells and digested in the lysosome, where the cytotoxic drug is released and consequently kills the target cancer cells.9 Currently, three main conjugation platforms have been developed to link drug molecules to mAbs, namely, through acylation of lysines,10 alkylation of genetically engineered cysteines,11 or alkylation of cysteines generated from the reduction of existing interchain disulfides (Figure 1).4,8 After the conjugation, the ADCs are a mixture of mAbs linked to a varying number of drugs. The average number of drugs loaded onto the mAbs, expressed as the drug-to-antibody ratio (DAR), represents one of the most important features for an ADC. The DAR defines the strength of an ADC and is the weighted summation of the different drug-loaded antibody species. A higher DAR indicates that, on average, more drugs are loaded. © 2013 American Chemical Society
Received: October 11, 2012 Accepted: January 4, 2013 Published: January 4, 2013 1699
dx.doi.org/10.1021/ac302959p | Anal. Chem. 2013, 85, 1699−1704
Analytical Chemistry
Article
Figure 1. Schematic illustrations of three main conjugation platforms. The mAb is conjugated through acylation of lysine (a), alkylation of genetically engineered cysteines (b), or alkylation of cysteines from reduced interchain disulfides (c). Notice that the interchain disulfides (red line) are intact in (a) and (b) but partially disrupted in (c). The linker between the hydrophobic drug and mAb is shown by a green line.
achieved with a 12 min gradient of decreasing salt concentration (from 1.5 to 0 M ammonium sulfate) and increasing organic modifier (from 0% to 25% isopropyl alcohol) in a 25 mM sodium phosphate buffer at pH 6.95. Approximately 50 μg of each ADC sample was injected onto the HIC column. Cysteine Protease Digestion. A common cysteine protease was explored to cleave the cytotoxic drug from the ADC to reduce the hydrophobicity of the ADC samples. The formulated ADC was diluted with the enzyme digestion buffer (0.1 M Tris, 4 mM EDTA, pH 7.4) to 1 mg/mL. L-Cysteine was added to the diluted ADC samples to reach a final concentration of 1 mM for the purpose of protease activation. A limited amount of the cysteine protease was used to cleave the drug potion from the linker without fragmentation of the antibody. In this experiment, the enzyme-to-protein ratio (by weight) was 1:100000 and the reaction was carried out at 37 °C for 30 min. After the digestion, the sample was desalted and buffer exchanged into 200 mM ammonium acetate as described above prior to native MS analysis. Mass Spectrometry Instrument Settings. The mass spectrometer utilized was a Waters (Milford, MA) HDMS Synapt G1 equipped with a nano-ESI source (Waters). Positive ion mass spectra were acquired. The spraying needles were borosilicate glass capillaries with Au/Pt coatings obtained from Thermo Scientific (West Palm Beach, FL). The flow rate was approximately 20−50 nL/min. The backing pressure at the source was set at between 5.0 and 5.5 mbar (e.g., 5.3 mbar), the ionization voltage was between 1000 and 1200 V, the sample cone voltage was 185 V, the extraction cone voltage was 4 V, and the source temperature was 50 °C. The trap collision energy (CE) was 6 V, and the transfer CE was 4 V. The trap gas (argon) flow rate was 3 mL/min, and the scan range was between 1000 and 12000 m/z. Data were acquired when the spray signal became stable. Each MS spectrum was a result of averaging 100 scans (1 s/scan). The data were acquired by the MassLynx software, version 4.1 (Waters), and processed by MaxEnt I. The deconvolution output mass range was set to be 140−170 kDa with a resolution of 20 Da/channel. The uniform Gaussian model was with a 15 Da width at half-height. The minimum intensity ratio was 33%. The number of iterations was set to 20. The MS calibration was carried out with 1 mg/mL cesium iodide dissolved in 50% acetonitrile. The instrument was kept at identical settings as described above during calibration and data acquisition.
where a fast, accurate, and sensitive method is required for monitoring the drug metabolism. Recently, Valliere-Douglass et al.1 described a native MS method for analysis of reduced disulfide bond cysteine-linked ADCs. Here, we report another approach which, in addition to the native MS, employs limited enzymatic treatment and nanoelectrospray ionization (nano-ESI) to achieve optimized ionization, sensitivity, and determination of DAR for ADCs.
■
EXPERIMENTAL SECTION Materials. Cysteine-linked ADC molecules, each with DAR values of 3.4 and 5.3, were produced by Genentech, Inc. (South San Francisco, CA). The linker is maleimidocaproyl-valinecitrulline-((p-aminobenzyl)oxy)carbonyl (MC-VC-PABC), and the cytotoxic drug is monomethyl auristatin E (MMAE) (Figure 2).
Figure 2. Cysteine protease cleaves the amide bond between the VC and PABC portions of the linker, which releases the hydrophobic drug but keeps the rest of the linker (MC-VC) remaining on the antibody.
Sample Preparation. To remove incompatible salts for mass spectrometry analysis, the formulated ADC (50 μL of a 10 mg/mL sample) was buffer exchanged with 1 M ammonium acetate using a 30 kDa cutoff 15 mL Amicon centrifugal device (Millipore, Billerica, MA) at 4 °C and 4000 rpm. For complete removal of the formulation buffer, the buffer exchange step was repeated three times. The final ADC sample was buffer exchanged into 200 mM ammonium acetate using the same device and centrifugal parameters to a final concentration of approximately 2.5 mg/mL. Prior to MS analysis, this solution was further diluted to 0.5 mg/mL with 200 mM ammonium acetate. About 10 μL (i.e., 5 μg) was loaded into the nanospray needle. To conserve the samples, a 4 mL Amicon centrifugal device can be used with less starting material (e.g., 5 μL of a 10 mg/mL sample). HIC Assay. The HIC assay was conducted with a TOSOH (King of Prussia, PA) Butyl-NPR column (4.6 × 35 mm) at 24 °C column temperature and 0.8 mL/min flow rate. Elution was
■
RESULTS AND DISCUSSION HIC of ADCs before and after Amicon Buffer Exchange. To achieve optimal MS resolution and sensitivity, 1700
dx.doi.org/10.1021/ac302959p | Anal. Chem. 2013, 85, 1699−1704
Analytical Chemistry
Article
Nano-ESI under Native Conditions. Compared to covalent bond interactions, noncovalent interactions are difficult to maintain during the conventional electrospray ionization process as its conditionslow pH, high temperature, strong curtain gas as well as denaturing organic solvent, etc. tend to dissociate the noncovalent complexes.16,17 According to published research, replacing the organic solvent and acid with nondenaturing volatile salts such as ammonium acetate for the ionization could achieve satisfactory results in studies of the noncovalent protein complexes; in particular, 200 mM ammonium acetate (pH ≈ 6.5) was often the preferred choice.18−20 To further improve the ionization “softness” to better preserve the noncovalent interactions within the ADC complex, a nano-ESI source instead of a regular ESI source was utilized. Unlike the conventional ESI, nano-ESI employs a lower source temperature and spraying voltage and zero gas flow toward the spray, rendering it a much softer ionization method. Choosing nano-ESI over ESI not only warrants a gentler and more desirable ionization condition, but also provides much higher sensitivity than conventional ESI.21,22 The raw MS spectra of an ADC sample run by both conventional ESI and nano-ESI conditions are shown in Figure 4, which clearly illustrates the
the samples must be thoroughly desalted to remove any components in the formulation buffer that are MS incompatible. A buffer exchange procedure using a 30 kDa cutoff Amicon centrifugal device was designed for preparing ADC samples. To evaluate the potential impact of buffer exchange on the ADC drug distribution, HIC experiments were performed before and after buffer exchange. The resolved species are eluted in an order from low to high drug load. The profiles before and after buffer exchange show a consistent DAR distribution (Figure 3), indicating that the sample preparation procedure does not impact the drug distribution of the ADC, thus generating samples that are suitable for subsequent native MS analysis.
Figure 3. HIC profile of an ADC sample before and after buffer exchange (BE).
DAR Determination. For both HIC and deconvoluted mass spectrum profiles, the DAR values, as shown in Table 1, are calculated from the relative peak area (%) of each peak and the corresponding number of drugs loaded. The weighted peak percentage, which measures the contribution of individual drugloaded species to DAR, is calculated by multiplying the relative peak area (%) and the corresponding number of loaded drugs. DAR is then obtained by summing up the weighted peak percentage from all observed species and dividing the sum by 100, as follows: DAR =
Figure 4. Native MS spectra using either conventional ESI (top) or nano-ESI (bottom). S/N = signal-to-noise ratio.
∑ (relative peak area × number of loaded drugs) /100
Table 1. Comparison between HIC and Native Nano-ESI MS (before and after Limited Cysteine Protease Digestion) for the Drug Load Distribution (Peak Area %) and Calculated DAR Values native nano-ESI MS (before cysteine protease digestion) drug load 0 2 4 6 8 DAR
peak area (%)
weighted peak area (%)
11.5 37.4 46.5 3.8 0.8
0 74.8 186 22.8 6.4 2.9
native nano-ESI MS (after cysteine protease digestion) drug load peak area (%) 0 2 4 6 8 DAR
7.9 35.6 43.2 10.7 2.6
weighted peak area (%) 0 71.1 172.8 64.2 20.8 3.3
1701
HIC drug load 0 2 4 6 8 DAR
peak area (%)
weighted peak area (%)
6.2 31.0 48.9 11.7 2.1
0 62.0 195.6 70.2 16.8 3.4
dx.doi.org/10.1021/ac302959p | Anal. Chem. 2013, 85, 1699−1704
Analytical Chemistry
Article
higher signal-to-noise ratio (S/N = 228) for nano-ESI than the conventional ESI (S/N = 4.5). Due to this higher sensitivity feature of nano-ESI, the sample requirement for the native MS analysis is greatly reduced when compared with conventional ESI sample requirements. The experiments we report here consume less than 5 μg of ADC samples, which is much lower than the minimum sample requirement for the HIC method; therefore, this native nanoESI MS method can be used to determine DAR in the cases when sample quantities are limited. A high backing pressure utilized at the electrospray source can also improve trapping and transmission of the highmolecular-weight noncovalent complexes, leading to higher sensitivity;23−25 therefore, the backing pressure on the MS instrument in this study was set at 5.0−5.5 mBar. Native Nano-ESI MS and HIC. As shown in Figure 5b, the deconvoluted mass spectrum of an ADC from the native nanoESI MS analysis displays all expected drug-loaded species from zero to eight drugs per antibody, which resembles the HIC profile (Figure 5a). However, due to the hydrophobicity of the small-molecule drug, a significantly lower representation was observed for the higher drug-loaded species (six- and eight-drug loads, especially eight-drug species) in the MS analysis. This observation was attributed to the higher hydrophobicity for the high-drug-load species as these species have less affinity for protons, leading to lower ionization efficiency. Table 1 summarizes the native nano-ESI MS and HIC results of DAR determination for one of the ADC samples with DAR = 3.4. As reflected in the table, native nano-ESI MS data lead to a much lower DAR value (2.9) than data from the HIC method (DAR = 3.4) due to the reduced signal observed for the high-drugloaded species. A second ADC with a higher DAR was also tested; similarly, a smaller DAR value is obtained from the native MS method. To minimize the ionization suppression effect in the ADC samples, it is desirable to reduce the hydrophobicity of ADCs to equalize the ionization efficiency among species with different drug loads. One approach could be cleaving the hydrophobic drug moiety from the ADCs while leaving the linker attached. Limited enzymatic digestion was explored to achieve this goal. Cysteine Protease Digestion. In Cys-linked ADCs, the antibody connects to the hydrophobic drug (MMAE) through the linker MC-VC-PABC (Figure 2). Cathepsin B was reported to specifically cleave the drug portion (PABC-MMAE) from the conjugate, leaving the MC-VC linker remaining on the antibody.26 In this study, we utilized a common cysteine protease which achieved similar cleavage and specificity. When a trace amount of the cysteine protease is introduced into the ADC for a limited time (see the Experimental Section), the enzyme will only cleave the amide bond between VC and PABC (Figure 2) on the linker, releasing the hydrophobic drug in the form of PABC-MMAE without fragmentation of the antibody. The remaining portion of the linker (MC-VC, Figure 2) is still attached to the antibody. Compared to PABCMMAE, MC-VC contributes very little hydrophobicity to the whole ADC. Because the ratio between PABC-MMAE and MC-VC is always 1:1, the remaining MC-VC linker can perfectly serve as a drug-loading indicator in the MS analysis with little ionization suppression effect. To achieve optimal results, the cysteine protease digestion conditions must be carefully controlled and optimized for the complete removal of drug molecule and to avoid fragmentation
Figure 5. Comparisons of the HIC chromatogram (a), deconvoluted mass spectrum of native nano-ESI MS without cysteine protease digestion (b), and deconvoluted mass spectrum of native nano-ESI MS with cysteine protease digestion (c).
of the mAb. In this study, a 1:100000 (w:w) enzyme-to-protein ratio was used at 37 °C for 30 min to remove PABC-MMAE and maintain the linker MC-VC with mAb to provide conjugation information. Native Nano-ESI MS after Limited Cysteine Protease Digestion. The multiply charged species by the native nanoESI MS of an ADC (DAR = 3.4) after limited cysteine protease digestion are shown in Figure 6. As evident in this figure, five peaks are present in each charge state which correspond to the five drug-loaded species. For example, the charge state 23+ shows five peaks spaced evenly by a mass of two MC-VC 1702
dx.doi.org/10.1021/ac302959p | Anal. Chem. 2013, 85, 1699−1704
Analytical Chemistry
■
CONCLUSION Native nano-ESI MS is applicable to interchain cysteine-linked ADCs. Although the ADCs are held by few or no disulfides, the noncovalent structure can be preserved by nondisruptive native MS conditions. Applying the nano-ESI technology greatly improves the sensitivity of the native MS analysis and further reduces the “harshness” of the electrospray conditions. Removing the highly hydrophobic drug moiety significantly improved the ionization efficiency of the high-drug-loaded species. Altogether, this novel MS method enables direct mass measurement of the enzymatically cleaved ADC and determination of a DAR that is comparable to that from the standard HIC method. Therefore, it can serve as an orthogonal method for DAR determination which enhances the characterization capability for this type of cancer therapeutic. Furthermore, the method has been proven to be accurate and more sensitive as compared to the conventional chromatography method (HIC). Thus, it will be a valuable tool for analysis of human or animal in vivo study samples for which the quantities are limited.
■
Figure 6. Native nano-ESI MS spectrum of cysteine protease digested ADC with a DAR of 3.4. Charge states from 21+ to 26+ are shown. Each charge state consists of linker (MC-VC)-attached species from zero to eight linkers.
*Phone: (650) 467-1139. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank our colleagues at Genentech for their support and scientific discussions during this study, especially Nancy Chen, Lynn Gennaro, and John Stults. We also thank Michael Daly from Waters Corp. for his help with the experimental design and instrument optimization.
■
DAR by native nano-ESI MS with cysteine protease digestion trial 1
trial 2
trial 3
average
STDEV
RSD (%)
3.3 5.2
3.2 5.0
3.4 5.2
3.3 5.1
0.1 0.1
3.0 2.2
a
REFERENCES
(1) Valliere-Douglass, J. F.; McFee, W. A.; Salas-Solano, O. Anal. Chem. 2012, 84, 2843. (2) Farah, R. A.; Clinchy, B.; Herrera, L.; Vitetta, E. S. Crit. Rev. Eukaryotic Gene Expression 1998, 8, 321. (3) Carter, P. Nat. Rev. Cancer 2001, 1, 118. (4) Trail, P.; Willner, D.; Lasch, S.; Henderson, A.; Hofstead, S.; Casazza, A.; Firestone, R.; Hellstrom, I.; Hellstrom, K. Science 1993, 261, 212. (5) Peter, D, S. Curr. Opin. Chem. Biol. 2009, 13, 235. (6) Hughes, B. Nat. Rev. Drug Discovery 2010, 9, 665. (7) Younes, A.; Bartlett, N. L.; Leonard, J. P.; Kennedy, D. A.; Lynch, C. M.; Sievers, E. L.; Forero-Torres, A. N. Engl. J. Med. 2010, 363, 1812. (8) Francisco, J. A.; Cerveny, C. G.; Meyer, D. L.; Mixan, B. J.; Klussman, K.; Chace, D. F.; Rejniak, S. X.; Gordon, K. A.; DeBlanc, R.; Toki, B. E.; Law, C. L.; Doronina, S. O.; Siegall, C. B.; Senter, P. D.; Wahl, A. F. Blood 2003, 102, 1458. (9) Wu, A. M.; Senter, P. D. Nat. Biotechnol. 2005, 23, 1137. (10) Hamann, P. R.; Hinman, L. M.; Hollander, I.; Beyer, C. F.; Lindh, D.; Holcomb, R.; Hallett, W.; Tsou, H.-R.; Upeslacis, J.; Shochat, D.; Mountain, A.; Flowers, D. A.; Bernstein, I. Bioconjugate Chem. 2001, 13, 47. (11) Junutula, J. R.; Raab, H.; Clark, S.; Bhakta, S.; Leipold, D. D.; Weir, S.; Chen, Y.; Simpson, M.; Tsai, S. P.; Dennis, M. S.; Lu, Y.; Meng, Y. G.; Ng, C.; Yang, J.; Lee, C. C.; Duenas, E.; Gorrell, J.; Katta, V.; Kim, A.; McDorman, K.; Flagella, K.; Venook, R.; Ross, S.; Spencer, S. D.; Lee Wong, W.; Lowman, H. B.; Vandlen, R.; Sliwkowski, M. X.; Scheller, R. H.; Polakis, P.; Mallet, W. Nat. Biotechnol. 2008, 26, 925. (12) Hamblett, K. J.; Senter, P. D.; Chace, D. F.; Sun, M. M. C.; Lenox, J.; Cerveny, C. G.; Kissler, K. M.; Bernhardt, S. X.; Kopcha, A.
Table 2. Consistent Results Obtained Using the Native Nano-ESI MS Method with the Limited Cysteine Protease Digestion for DAR Determinationa
3.4 5.3
AUTHOR INFORMATION
Corresponding Author
linkers. Indeed, the deconvoluted spectrum (Figure 5c) demonstrates that all the species (from zero- to eight-drug loads, in this cysteine protease-treated sample, zero to eight MC-VC linkers) expected in the ADC are unequivocally observed. Calculated DARs based on the deconvoluted mass spectra are summarized in Table 1. In contrast to the values obtained without enzyme digestion, these numbers are much more comparable with those determined by the HIC method, validating the necessity to reduce the hydrophobicity of the analytes in the native nano-ESI MS analysis. The experiments were repeated three times for both DAR = 3.4 and DAR = 5.3 samples, and consistent results were obtained (Table 2).
DAR by HIC
Article
STDEV = standard deviation; RSD = relative standard deviation.
Our experiments showed that full ADC species with the mass of antibody−linker−drugs were not observed, indicating a complete drug removal from the ADC. Furthermore, the masses of mAb fragments were also carefully monitored but not observed, confirming the cleavage specificity under the condition of a limited amount of the cysteine protease. The slightly lower DAR obtained from the native nano-ESI MS after cysteine protease digestion in comparison to that from HIC may be attributed to the slight ionization suppression effect on the higher linker containing species. Nevertheless, this native nano-ESI MS approach combined with limited cysteine protease treatment has been demonstrated to be a suitable and effective method for direct intact mass measurement and DAR determination of Cys-linked ADCs. 1703
dx.doi.org/10.1021/ac302959p | Anal. Chem. 2013, 85, 1699−1704
Analytical Chemistry
Article
K.; Zabinski, R. F.; Meyer, D. L.; Francisco, J. A. Clin. Cancer Res. 2004, 10, 7063. (13) Xu, K.; Liu, L.; Saad, O. M.; Baudys, J.; Williams, L.; Leipold, D.; Shen, B.; Raab, H.; Junutula, J. R.; Kim, A.; Kaur, S. Anal. Biochem. 2011, 412, 56. (14) Lazar, A. C.; Wang, L.; Blättler, W. A.; Amphlett, G.; Lambert, J. M.; Zhang, W. Rapid Commun. Mass Spectrom. 2005, 19, 1806. (15) Wakankar, A.; Chen, Y.; Gokarn, Y.; Jacobson, F. S. mAbs 2011, 3, 161. (16) Wen, J.; Zhang, H.; Gross, M. L.; Blankenship, R. E. Biochemistry 2011, 50, 3502. (17) Loo, J. A. Mass Spectrom. Rev. 1997, 16, 1. (18) van den Heuvel, R. H.; Heck, A. J. Curr. Opin. Chem. Biol. 2004, 8, 519. (19) Yin, S.; Loo, J. A. Methods Mol. Biol. 2009, 492, 273. (20) Hernandez, H.; Robinson, C. V. Nat. Protoc. 2007, 2, 715. (21) Wilm, M.; Mann, M. Anal. Chem. 1996, 68, 1. (22) Andren, P. E.; Emmett, M. R.; Caprioli, R. M. J. Am. Soc. Mass Spectrom. 1994, 5, 867. (23) Sobott, F.; Hernández, H.; McCammon, M. G.; Tito, M. A.; Robinson, C. V. Anal. Chem. 2002, 74, 1402. (24) Kaddis, C. S.; Loo, J. A. Anal. Chem. 2007, 79, 1778. (25) Lorenzen, K.; Olia, A. S.; Uetrecht, C.; Cingolani, G.; Heck, A. J. R. J. Mol. Biol. 2008, 379, 385. (26) Klussman, K.; Mixan, B. J.; Cerveny, C. G.; Meyer, D. L.; Senter, P. D.; Wahl, A. F. Bioconjugate Chem. 2004, 15, 765.
1704
dx.doi.org/10.1021/ac302959p | Anal. Chem. 2013, 85, 1699−1704
