Characterization of Cysteinylation and Trisulfide Bonds in a

Apr 26, 2016 - Adriana Kita, Gomathinayagam Ponniah, Christine Nowak, and Hongcheng Liu*. Product Characterization, Alexion Pharmaceuticals Inc., 352 ...
0 downloads 0 Views 1MB Size
Subscriber access provided by Brown University Library

Article

Characterization of cysteinylation and trisulfide bonds in a recombinant monoclonal antibody Adriana Kita, Gomathinayagam Ponniah, Christine Nowak, and Hongcheng Liu Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b00822 • Publication Date (Web): 26 Apr 2016 Downloaded from http://pubs.acs.org on May 2, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

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

Analytical Chemistry

Characterization of cysteinylation and trisulfide bonds in a recombinant monoclonal antibody

Adriana Kita, Gomathinayagam Ponniah, Christine Nowak, and Hongcheng Liu*

Product Characterization, Alexion Pharmaceuticals Inc 352 Knotter Drive, Cheshire, CT06410



Corresponding author



Email: [email protected]



Phone: 203-271-8354

ACS Paragon Plus Environment

Analytical Chemistry

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

A recombinant monoclonal antibody with trisulfide bonds and cysteinylation was thoroughly characterized in the current study. Trisulfide bonds and cysteinylation were first detected when the recombinant monoclonal antibody was analyzed by LC-MS to determine the molecular weights of the intact antibody and its F(ab’)2 fragment generated from IdeS digestion. LC-MS analysis of non-reduced tryptic peptides indicated trisulfide bonds are associated with the inter chain disulfide bonds of both A isoform and A/B isoform and cysteinylation is associated only with the A isoform. A low percentage of trisulfide bonds were detected in between the light chain and heavy chain disulfide bond of the A form. While the majority of trisulfide bonds and cysteinylation are associated with the hinge region peptide that involves the four closely spaced cysteine residues of the heavy chain. The locations of trisulfide bond and cysteinylation were determined using a combination of Edman sequencing and LC-MS. In the A isoform, the major site of the trisulfide bond and cysteinylation is between the first disulfide bond in the hinge region. In the A/B isoform, the trisulfide was also located in between the disulfide bond that is formed by the second pair of cysteine residues.

ACS Paragon Plus Environment

Page 2 of 28

Page 3 of 28

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

Analytical Chemistry

Recombinant monoclonal antibodies are heterogeneous due to various modifications 1-6. Several modifications are related to cysteine residues. Theoretically, all the cysteine residues are involved in the formation of disulfide bonds and each subclass of human IgG molecules has a welldefined homogeneous disulfide bond structure. Those disulfide bond structures are critical for IgG structures, stability and biological functions. However, alternative disulfide bond linkage, free cysteine residues, trisulfide bonds, thioethers and cysteinylation have been widely reported. Alternative disulfide bond linkages were first observed in IgG4 and later in IgG2. For IgG4, the two inter heavy chain disulfide bonds are in equilibrium with the formation of two intra chain disulfide bonds 7-9. The consequence of such equilibrium is the formation of half-molecules 7-10and hybrid IgG4 molecules 11,12. For IgG2, in addition to the classical disulfide bond linkage, two forms with alternative disulfide bonds were observed 13,14. The classical linkage was termed A isoform, and the A/B and B isoforms are the identified alternative disulfide bond structures (Figure 1). Isoform A is synthesized and then converted into isoform B through the intermediate isoform A/B in cell culture and human serum15. A low percentage of free cysteine has been commonly observed for recombinant and human IgG molecules 16-28. Free cysteine is associated with every single cysteine residue in recombinant IgG molecules 24,25. There are also cases that higher percentages of free cysteine residues are predominately localized in the heavy chain variable domains19,20,23. One of these studies demonstrated that the unpaired cysteine residues in the variable domain form a disulfide bond rapidly in serum23. Higher levels of free cysteine result in decreased thermal stability 22. The presence of free cysteine may also cause the formation of covalent aggregates 27. The formation of thioether linkage 29-31 and racemization of cysteine residues 32,33 are both accelerated by basic pH. A general base-catalyzed mechanism lining thioether formation and

ACS Paragon Plus Environment

Analytical Chemistry

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

racemization was proposed 33. In this mechanism, dehydroalanine is formed due to dehydrogenation of one cysteine residue, which can either form a thioether bond or reform a disulfide bond with a mixture of D and L-cysteine. Trisulfide bonds were first reported for a recombinant monoclonal IgG2 antibody localized to the inter heavy chain disulfide bonds 34, although the specific location cannot be determined. Later, trisulfide bonds were detected in all four subclasses of IgG molecules and human endogenous IgG molecules 35. Trisulfide bonds are formed due to the presence of hydrogen sulfide in cell culture 34,35. So far, cysteinylation has only been observed in IgG molecules with extra cysteine residues in addition to the conserved cysteine residues36-39. Cysteinylation of a cysteine residue in the heavy chain complementary-determining region three (CDR3) of a recombinant monoclonal antibody resulted in a conformational change, decreased thermal stability, increased propensity of aggregation and decreased activity 38. The current study reports a thorough characterization of a recombinant IgG2 monoclonal antibody with trisulfide bonds and cysteinylation. A low percentage of trisulfide bonds were found between the light chain and heavy chain disulfide bond of the A and the A/B isoforms. Relatively higher percentage of trisulfide bonds were within the hinge region between the two heavy chains. The trisulfide bond in the A isoform is formed with the first two cysteine residues in the hinge region of the two heavy chains. The trisulfide bond in the A/B isoform is formed with the second two cysteine residues in the hinge region of the two heavy chains. Interestingly, concurrent cysteinylation of two cysteine residues was observed for this recombinant monoclonal antibody, which has no extra cysteine residues. Cysteinylation was also localized to be within the hinge region associated with the first two cysteine residues in the hinge region of the two heavy chains.

ACS Paragon Plus Environment

Page 4 of 28

Page 5 of 28

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

Analytical Chemistry

Materials and methods Materials The recombinant monoclonal antibody was expressed in Chinese hamster ovary (CHO) cell line and purified at Alexion Pharmaceuticals (Cheshire, CT). Acetonitrile, ethanol, guanidine hydrochloride, iodoacetamide and trifluoroacetic acid (TFA) were purchased from Sigma (St, Louis, MO ). IdeS enzyme (FabRICATORTM) was purchased from Genovis (Cambridge, MA). Trypsin was purchased from Worthington (Lakewood, NJ). Lys-C, 12% trimethylamine, 5% phenyl isothiocyanate (PITC) n-heptane were purchased from Wako (Osaka, Japan).

FabRICATOR digestion and protein A purification. The recombinant monoclonal antibody was digested using IdeS protease to generate F(ab’)2 fragment. Each vial of IdeS enzyme was reconstituted in 500 µL water. The sample (10 mg/mL) in 20 mM sodium phosphate buffer, pH 6.8, was digested using IdeS using a ratio of 1 µL reconstituted enzyme for each 100 µg protein, at 37 °C, for 2 hours. MabSelect SuRe TM resin (GE Health Care, Pittsburgh, PA) was used to remove Fc portion of the antibody from the digested sample. MabSelect SuReTM was washed three times by suspending in 20 mM sodium phosphate buffer, pH 6.8, centrifuged briefly and resuspended in the same buffer. The digested sample was mixed with the Mabselect SuRe TM

. Supernatant was collected after centrifugation and stored at -80 °C for further analysis.

Non-reducing Peptide mapping using Lys-C and Trypsin The sample (1mg/mL) was denatured using 6 M guanidine hydrochloride in 50 mM Tris, pH 7.8 with 5 mM iodoacetamide at 37 °C for 1 hour. Iodoacetamide was used to modify free cysteine residues in the samples to prevent disulfide bond scrambling. The concentration of guanidine

ACS Paragon Plus Environment

Analytical Chemistry

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

hydrochloride was diluted to 0.5 M after denaturation using 10 mM Tris, pH 7.8. The sample was then digested using Lys-C and Trypsin at a ratio of 1:10 (w: w) enzyme to protein ratio for each enzyme at 37 °C for approximately 15 hours. A second aliquot of Lys-C and trypsin at the same 1:10 ratio was added. The samples were further digested at 37 °C for 4 hours. The digested samples were either analyzed directly by LC-MS or by RP-HPLC with UV detection for fraction collection.

LC-MS analysis of the molecular weights of antibody and its fragment An UPLC system (Waters, Bedford, MA) and a Maxis 4 G mass spectrometer (Bruker, Billerica, MA) with a Vydac C4 (1.0x 150 mm) column were used to measure the molecular weights of the antibody and its F(ab’)2 fragment. Approximately 10 µg of protein was loaded onto the column using 95% mobile phase A (0.1% TFA in water) and 5% mobile phase B (0.1% TFA in acetonitrile). After 5 minutes, proteins were eluted off the column by increasing mobile phase B to 95% within 15 minutes. The column was washed using 95% mobile phase B for 5 minutes and then equilibrated using 5% mobile phase B for 10 minutes. The flow-rate was set to 50 µL/min. The column temperature was set at 60 °C. The mass spectrometer was operated in the positive mode with a scan range of m/z 900-5500. Other settings include capillary voltage at 4500 V, nebulizer at 0.5 Bar, Dry gas at 2.0 L/min, dry temperature at 220 °C.

LC-MS analysis of peptides The same UPLC and mass spectrometer used for molecular weight analysis were used for LC-MS peptide mapping. A Proto 200 C18 column (1.0 x 150 mm, Higgins Analytical) was used to separate and introduce peptides into the mass spectrometer. Using the same mobile phases, peptide samples were loaded onto the column using 95% mobile phase A and 5% mobile phase B. After 5 minutes, the peptides were eluted off the column a long linear gradient of increasing mobile phase B to 35% within

ACS Paragon Plus Environment

Page 6 of 28

Page 7 of 28

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

Analytical Chemistry

135 minutes. The column was then washed using 95% mobile phase B and equilibrated using 5% mobile phase B. The column was set at 60 °C and the flow-rate was set at 50µL/min. A short gradient was used to analyze the collected fractions and the samples after Edman degradation. The samples were loaded at 95% mobile phase A and 5% mobile phase B for 10 minutes. Peptides were eluted off the column by increasing mobile phase B to 50% within 10 minutes. The flowrate and oven temperature were the same as used previously. The mass spectrometer was operated in the positive mode with a scan range of m/z 150-3000. Other settings include capillary voltage at 4500 V, nebulizer at 2.0 Bar, Dry gas at 10.0 L/min, dry temperature at 220 °C.

RP-HPLC and fraction collections A Waters Alliance HPLC with a C18 reversed-phase column (4.6 x 250 mm, Grace Discovery Sciences, Columbia, MD) was used for fraction collection. The same mobile phases, long gradient, and column temperature used for LC-MS peptide analysis are used for the larger column. The flow-rate was set at 1 mL/min and the column was kept at 60 °C. Peptides eluted off the column were monitored by UV214 nm. Fractions were collected following the guidance of UV absorption. The collected fractions were concentrated using a speed-vacuum and then analyzed by LC-MS using the procedure described previously.

Edman sequencing Edman sequencing was performed by following a published procedure with modifications40. In brief, fractions containing the peptides of interest were dried using a speed vacuum. Then 50 µL of ethanol, acetonitrile, trimethylamine and 5% PITC at a 1:1:1:1 ratio (v:v:v:v) was added to each fraction.

ACS Paragon Plus Environment

Analytical Chemistry

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

The samples were incubated at 37 °C for 10 minutes for coupling reaction. The modified samples were dried using a speed-vacuum and then incubated with 20 µL TFA at 50 °C for 5 minutes for cleavage. For samples that need to go through the second Edman degradation cycle, residual reagents from the first cycle were removed using RP-HPLC. The samples were dried and then subjected to the second cycle of reaction. The samples after cleavage were dried and then reconstituted using 25 µL of a solvent containing 10% acetonitrile and 0.1% TFA in water and analyzed by LC-MS.

Results and discussion Analysis of intact and F(ab’)2 by reversed-phase chromatography The recombinant monoclonal antibody was first analyzed by molecular weight measurement. As shown in Figure 2A, the main peak molecular weight of this antibody is 148,610 Da. This molecular weight corresponds to this antibody with N-terminal heavy chain pyroglutamate, no C-terminal lysine and with G0F oligosaccharides on both heavy chains. In addition, several small peaks are also observed. The peak with the molecular weight of the 148,407 Da has the same modifications as the main peak, but with G0F-GlcNAc on one heavy chain. The peak with the molecular weight of 148,737 Da is approximately 127 Da higher than the main peak and thus corresponds to the antibody with one Cterminal Lys. The peak with the molecular weight of 148,772Da is 162 Da higher than the main peak and it corresponds to the antibody with G1F on one heavy chain. The peak with the molecular weight of 148,850Da is 240 Da higher than the main peak, which corresponds to the antibody with cysteinylation of two cysteine residues. This peak disappeared after a brief incubation with free cysteine under slightly basic condition, supporting that it was formed due to cysteinylation(data not shown). The peak with the molecular weight of 149,431 Da corresponds to the antibody with a partial leader sequence on one heavy chain, as reported previously 41. The molecular weight of the main peak shoulder cannot be

ACS Paragon Plus Environment

Page 8 of 28

Page 9 of 28

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

Analytical Chemistry

accurately determined, but it is approximately 28-34 Da higher than the main peak. It was hypothesized that this shoulder was due to the presence of the trisulfide bond that increases the molecular weight by 32 Da. This shoulder disappeared after a brief reduction with free cysteine, further supporting that it is formed due to trisulfide bond (Data not shown). The molecular weight of the F(ab’)2 fragment obtained from IdeS digestion was further analyzed to localize the above-mentioned modifications and the result is shown in Figure 2B. The main peak molecular weight of 98,214 Da is in good agreement with the theoretical molecular weight of 98213 Da. The peak with the molecular weight of 98,246 is 32 Da higher than the main peak molecular weight, which corresponds to the fragment with one trisulfide bond. The peak with the molecular weight of 98,275 Da is approximately 61 Da higher than the main peak, which corresponds to the fragment with two trisulfide bonds. The peak with the molecular weight of 98,376Da is 162 Da higher than the main peak molecular weight, which corresponds to one glycation. The peak with the molecular weight of 98,454 Da corresponds to the fragment with cysteinylation of two cysteine residues. The peak with the molecular weight of 98,584 Da is approximately 370 Da higher than the main peak, which likely corresponds to the fragment with a combination of various modifications such as trisulfide bonds and more than one glycation. The peak with the molecular weight of 99,037 Da corresponds to the fragment with partial leader sequence. Analysis of the molecular weights of the intact antibody and its F(ab’)2 fragment help localize the modifications to either the F(ab’)2 or Fc regions. For this monoclonal antibody, trisulfide bond, cysteinylation, glycan and partial leader sequence are localized to the F(ab’)2 fragment. C-terminal lysine and various glycoforms are localized to the Fc portion. Trisulfide bond and cysteinylation are the focus of the current study.

ACS Paragon Plus Environment

Analytical Chemistry

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

Non-reducing peptide mapping F(ab’)2 fragment was digested under non-reducing condition and then analyzed directly by LCMS or LC with UV detection for fraction collection. The major advantage of working with F(ab’)2 is the generation of smaller tryptic peptides that contain the unique hinge disulfide bond structures of the three different isoforms. The data was analyzed by searching disulfide bond containing peptides with and without a trisulfide bond or cysteinylation. All the intra chain and inter chain disulfide bond structures in the F(ab’)2 portion were confirmed. In agreement with previous reports 35, a trisulfide bond was detected in peptides containing inter chain disulfide bonds, but not intra chain disulfide bonds. Peptides with inter chain disulfide bonds are shown in Figure 1 to aid discussion. The level of trisulfide bond was determined by the relative percentage of the extracted ion chromatogram (EIC) peak areas of the trisulfide bond containing peptide over the total EIC of the same peptide with or without trisulfide bonds. Approximately 1% trisulfide bond was found to be associated with the inter light chain and heavy chain disulfide bond formed between the light chain fifth cysteine residue and the heavy chain third cysteine residue. This disulfide bond linkage is present in the A and A/B isoforms. Approximately 8% trisulfide bond was associated with the A isoform hinge peptide with four disulfide bonds. Approximately the same 8% trisulfide bond was associated with the A/B hinge region peptides. On the other hand, only extremely low level of trisulfide bond was associated with the B form hinge peptide. Cysteinylation was only associated with the A isoform hinge region peptides. Peptides from digestion of the F(ab’)2 under non-reducing conditions were also separated using a column with relatively larger diameter to collect fractions for detailed analysis. The collected fractions were analyzed by LC-MS using a short gradient to search for fractions that contain the peptides with trisulfide bonds and cysteinylation. Those fractions were used to determine the sites of trisulfide bonds and cysteinylation.

ACS Paragon Plus Environment

Page 10 of 28

Page 11 of 28

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

Analytical Chemistry

Localization of the sites of trisulfide bonds As discussed earlier, three major trisulfide bond containing peptides were identified for this recombinant monoclonal antibody. The first peptide containing a trisulfide bond formed between the light chain and heavy chain in the A and A/B isoforms. Further analysis was not carried out because the only possible location of the trisulfide bond is between the two cysteine residues. The second peptide containing a trisulfide bond is in the A form hinge region peptide containing the four closely spaced cysteine residues in the heavy chain. The location of the trisulfide bond in this peptide has not been determined because it is challenging to obtain informative fragment ions by CID fragmentation 34. A similar observation was obtained in the current study. Multiple fragment ions were observed when this peptide was fragmented using CID, but none of the fragment ions can help localize the site of the trisulfide bond. Therefore, an alternative approach based on Edman chemistry was used to localize the site of the trisulfide bond based on a method that was used to determine the linkage of closely spaced disulfide bonds40. The workflow and data are shown in Figure 3. It was assumed that the trisulfide bond is either between the N-terminal end disulfide bond or the C-terminal end disulfide bond considering the higher steric hindrance of the two middle disulfide bonds. The hinge region peptide monoisotopic molecular weight is 2761.1 Da, and the molecular weight of the same peptide containing a trisulfide bond is 2793.1 Da. The observed monoisotopic molecular weight of 2761.1 Da (Figure 3A) and 2793.0 Da (Figure 3B) calculated from the triply charged ions are in good agreement with the theoretical molecular weights. If the trisulfide bond is located between the N-terminal end disulfide bond, a peptide with a monoisotopic molecular weight of 2557.1 Da should be observed after one cycle of Edman sequencing because of the loss of two cysteine residues and one sulfur atom. On the other hand, if the trisufide bond is between the C-terminal end disulfide bond, a peptide with a monoisotopic molecular weight of 2589.2 Da should be observed because of the loss of two cysteine residues. Only a

ACS Paragon Plus Environment

Analytical Chemistry

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

peptide with a monoisotopic molecular weight of 2557.1 Da (Figure 3C) was observed after one cycle of Edman sequencing. Therefore, the site of the trisulfide bond was localized to the first two cysteine residues. The third peptide containing the trisulfide bond is the hinge region peptide of the A/B form, where this peptide was observed with and without a trisulfide bond. The theoretical monoisotopic molecular weight of the peptide without trisulfide bond is 4295.9 Da. The monoisotopic molecular weight of this peptide calculated from the quadruple charged ion (Figure 4A) is 4295.8 Da and is in good agreement with the theoretical molecular weight. The theoretical monoisotopic molecular weight of this peptide containing a trisulfide bond is 4328.0 Da. The observed monoisotopic molecular weight is 4327.8 Da (Figure 4B), which is in good agreement with the theoretical molecular weight. The trisulfide bond can be located either between the N-terminal disulfide bond or the C-terminal end disulfide bond. After one cycle of Edman sequencing, a peptide with a monoisotopic molecular weight of 2589.1 Da was observed (Figure 4C). The observation of this molecular weight confirmed that the trisulfide bond is indeed between the two heavy chains in the hinge region. However, observation of this trisulfide bond does not exclude the possibility that a lower percentage of trisulfide bond could be between one of the other two inter chain disulfide bonds, one between the light chain and the heavy chain hinge and the other one between the heavy chain hinge and the third cysteine residue of the other heavy chain. The peptide with the monoisotopic molecular weight of 2589.1 Da was subject to the second cycle of Edman degradation. A peptide with a monoisotopic molecular weight of 2353.1 Da was observed (Figure 4D). This molecular weight can only be generated if the trisulfide bond is located in between the N-terminal end disulfide bond. In addition, peptides with monoisotopic molecular weight corresponding to 2385.2 Da, predicted assuming the trisulfide bond is between the C-terminal end disulfide bond, was not observed. Taking all the information together, the trisulfide bond is between the N-terminal end disulfide bond of the hinge region of the A/B isoform.

ACS Paragon Plus Environment

Page 12 of 28

Page 13 of 28

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

Analytical Chemistry

Localizing the site of cysteinylation Cysteinylation was only detected to associate with the hinge region peptide of the A isoform. Interestingly, the major form of this peptide contains cysteinylation at two sites. The observed monoisotopic molecular weights of 2761.1 Da (Figure 5A) and 3001.1 Da (Figure 5B) are in good agreement with the theoretical hinge region peptide monoisotopic molecular weights of 2761.1 Da for the peptide without cysteinylation and 3001.4 Da for the peptide with cysteinylation. The same approach used for localizing trisulfide bond was used to determine the location of cysteinylation. Because of the observation that the major cysteinylated form contains two cysteine residues, it was assumed that one disulfide bond was reduced and the resulting cysteine residues were cysteinylated. Cysteinylation can be either located in between the N-terminal end or the C-terminal end disulfide bond. If cysteinylation is located in between the N-terminal end disulfide bond, a peptide with a monoisotopic molecular weight of 2557.1 Da is expected. On the other hand, if cysteinylation is located in between the C-terminal end disulfide bond, a peptide with a monoisotopic molecular weight of 2797.4 Da is expected after one cycle of Edman degradation. The fact that only a peptide with the monoisotopic molecular weight of 2557.1 Da was detected indicated that the cysteinylation was located in between the N-terminal end disulfide bond. It is worthwhile to mention that the peptide with cysteinylation was stable during coupling and cleavage reaction. A peptide with the additional molecular weight corresponding to modifications with four PITC molecules was observed after the coupling reaction, two modifications on the peptide N-terminal amine groups and two on the two cysteine amino groups. A peptide with the same molecular weight of the cysteinylated peptide was observed after cleavage reaction, indicating cysteinylation was stable during the cleavage reaction.

Summary

ACS Paragon Plus Environment

Analytical Chemistry

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

In summary, trisulfide bonds and cysteinylation are localized to the hinge region peptide containing the four closely spaced cysteine residues, as shown in Figure 6. A trisulfide bond was only present to substantial levels in the A and A/B isoforms in the recombinant monoclonal antibody. The major site of the trisulfide bond is between the first disulfide bond of N-terminal end in the A and the A/B isoforms. The N-terminal first disulfide bond in A isoform is located between the first pair of cysteine residues and is located between the second pair of cysteine residues in the A/B isoform. Cysteinylation is located at the N-terminal end first pair cysteines that normally form the N-terminal end first disulfide bond. A trisulfide bond is formed by a reaction with hydrogen sulfide 35,42. The major source of hydrogen sulfide is degradation of cysteine in the feed or cell culture media 42. The current study demonstrated that free cysteine can reduce the native disulfide bond. The newly generated free thiol groups can be further modified by cysteinylation. Trisulfide bond and cysteinylation are both localized to the first N-terminal end pair of disulfide bond in the A form, indicating that this disulfide bond is highly susceptible to modifications. Trisulfide bond in the A/B form is located to the disulfide bond formed between the two cysteine residues that are adjacent to the first pair of cysteine residues that form the N-terminal end first disulfide bond in the A/B form. The location of the trisulfide bond in the A/B form indicates that this disulfide bond formed between the second pair of cysteine residues is highly susceptible in the A/B isoform. Only trace level of trisulfide bond was observed in the B isoform. The lack of substantial level of trisulfide bond in the B isoform is likely due to the fact that the next disulfide bond in the hinge region is further separated from the two susceptible disulfide bonds by two amino acids. Cysteinylation is present to a much lower percentage compared to the trisulfide bond. It is localized to the N-terminal end first disulfide bond. The location of cysteinylation also supports the notion that the first disulfide bond of the N-terminal end is the most susceptible disulfide bond. The level of trisulfide bonds and cysteinylation could also be related to the fact that A isoform has the

ACS Paragon Plus Environment

Page 14 of 28

Page 15 of 28

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

Analytical Chemistry

longest exposure time to the medium because it is synthesized first and being slowly converted into A/B and B forms. A/B isoform has a longer exposure time compared to B, but shorter exposure time that A, while B has the shortest exposure time than A isoform.

Conclusions Low levels of trisulfide bonds and cysteinylation were detected in a recombinant monoclonal IgG2 antibody. The trisulfide bond is mainly associated with the A and A/B isoforms. Cysteinylation is associated with the A isoform. The trisulfide bond was localized to be between the first disulfide bond of the four inter heavy chain disulfide bonds within the hinge region of the A isoform and between the second disulfide bond in the A/B isoform. Cysteinylation is mainly associated with the two cysteine residues that typically form the first disulfide bond within the hinge region of the A isoform. Trisulfide bonds and cysteinylation are likely formed due to the presence of hydrogen sulfide and cysteine in the cell culture medium. The locations of these two modifications are likely correlated with the levels of exposure of the modified disulfide bonds, and the exposure time in the cell culture medium

ACS Paragon Plus Environment

Analytical Chemistry

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

Reference List

1. Liu, H.; Gaza-Bulseco, G.; Faldu, D.; Chumsae, C.; Sun, J. J. Pharm. Sci. 2008, 97 (7), 2426-2447. 2. Vlasak, J.; Ionescu, R. Curr. Pharm. Biotechnol. 2008, 9 (6), 468-481. 3. Du, Y.; Walsh, A.; Ehrick, R.; Xu, W.; May, K.; Liu, H. MAbs. 2012, 4 (5), 578-585. 4. Beck, A.; Bussat, M. C.; Zorn, N.; Robillard, V.; Klinguer-Hamour, C.; Chenu, S.; Goetsch, L.; Corvaia, N.; Van, D. A.; Haeuw, J. F. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2005, 819 (2), 203-218. 5. Beck, A.; Sanglier-Cianferani, S.; Van, D. A. Anal. Chem. 2012, 84 (11), 4637-4646. 6. Beck, A.; Wagner-Rousset, E.; Ayoub, D.; Van, D. A.; Sanglier-Cianferani, S. Anal. Chem. 2013, 85 (2), 715-736. 7. Angal, S.; King, D. J.; Bodmer, M. W.; Turner, A.; Lawson, A. D.; Roberts, G.; Pedley, B.; Adair, J. R. Mol. Immunol. 1993, 30 (1), 105-108. 8. Bloom, J. W.; Madanat, M. S.; Marriott, D.; Wong, T.; Chan, S. Y. Protein Sci. 1997, 6 (2), 407-415. 9. Schuurman, J.; Perdok, G. J.; Gorter, A. D.; Aalberse, R. C. Mol. Immunol. 2001, 38 (1), 1-8. 10. King, D. J.; Adair, J. R.; Angal, S.; Low, D. C.; Proudfoot, K. A.; Lloyd, J. C.; Bodmer, M. W.; Yarranton, G. T. Biochem. J. 1992, 281 ( Pt 2), 317-323. 11. Schuurman, J.; Van, R. R.; Perdok, G. J.; Van Doorn, H. R.; Tan, K. Y.; Aalberse, R. C. Immunology 1999, 97 (4), 693-698. 12. van der Neut, K. M.; Schuurman, J.; Losen, M.; Bleeker, W. K.; Martinez-Martinez, P.; Vermeulen, E.; den Bleker, T. H.; Wiegman, L.; Vink, T.; Aarden, L. A.; De Baets, M. H.; van de Winkel, J. G.; Aalberse, R. C.; Parren, P. W. Science 2007, 317 (5844), 1554-1557. 13. Wypych, J.; Li, M.; Guo, A.; Zhang, Z.; Martinez, T.; Allen, M. J.; Fodor, S.; Kelner, D. N.; Flynn, G. C.; Liu, Y. D.; Bondarenko, P. V.; Ricci, M. S.; Dillon, T. M.; Balland, A. J. Biol. Chem. 2008, 283 (23), 16194-16205. 14. Dillon, T. M.; Ricci, M. S.; Vezina, C.; Flynn, G. C.; Liu, Y. D.; Rehder, D. S.; Plant, M.; Henkle, B.; Li, Y.; Deechongkit, S.; Varnum, B.; Wypych, J.; Balland, A.; Bondarenko, P. V. J. Biol. Chem. 2008, 283 (23), 16206-16215. 15. Liu, Y. D.; Chen, X.; Enk, J. Z.; Plant, M.; Dillon, T. M.; Flynn, G. C. J. Biol. Chem. 2008, 283 (43), 29266-29272.

ACS Paragon Plus Environment

Page 16 of 28

Page 17 of 28

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

Analytical Chemistry

16. Schauenstein, E.; Dachs, F.; Reiter, M.; Gombotz, H.; List, W. Int. Arch. Allergy Appl. Immunol. 1986, 80 (2), 174-179. 17. Gevondyan, N. M.; Volynskaia, A. M.; Gevondyan, V. S. Biochemistry (Mosc. ) 2006, 71 (3), 279284. 18. Zhang, W.; Czupryn, M. J. Biotechnol. Prog. 2002, 18 (3), 509-513. 19. Harris, R. J. Dev. Biol. (Basel) 2005, 122, 117-127. 20. Chaderjian, W. B.; Chin, E. T.; Harris, R. J.; Etcheverry, T. M. Biotechnol. Prog. 2005, 21 (2), 550553. 21. Liu, H.; Gaza-Bulseco, G.; Chumsae, C.; Newby-Kew, A. Biotechnol. Lett. 2007, 29 (11), 1611-1622. 22. Lacy, E. R.; Baker, M.; Brigham-Burke, M. Anal. Biochem. 2008, 382 (1), 66-68. 23. Ouellette, D.; Alessandri, L.; Chin, A.; Grinnell, C.; Tarcsa, E.; Radziejewski, C.; Correia, I. Anal. Biochem. 2010, 397 (1), 37-47. 24. Chumsae, C.; Gaza-Bulseco, G.; Liu, H. Anal. Chem. 2009, 81 (15), 6449-6457. 25. Xiang, T.; Chumsae, C.; Liu, H. Anal. Chem. 2009, 81 (19), 8101-8108. 26. Zhang, T.; Zhang, J.; Hewitt, D.; Tran, B.; Gao, X.; Qiu, Z. J.; Tejada, M.; Gazzano-Santoro, H.; Kao, Y. H. Anal. Chem. 2012, 84 (16), 7112-7123. 27. Huh, J. H.; White, A. J.; Brych, S. R.; Franey, H.; Matsumura, M. J. Pharm. Sci. 2013, 102 (6), 17011711. 28. Zhang, Y.; Bailey, R.; Nightlinger, N.; Gillespie, A.; Balland, A.; Rogers, R. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2015, 997, 30-37. 29. Tous, G. I.; Wei, Z.; Feng, J.; Bilbulian, S.; Bowen, S.; Smith, J.; Strouse, R.; McGeehan, P.; CasasFinet, J.; Schenerman, M. A. Anal. Chem. 2005, 77 (9), 2675-2682. 30. Cohen, S. L.; Price, C.; Vlasak, J. J. Am. Chem. Soc. 2007, 129 (22), 6976-6977. 31. Zhang, Q.; Schenauer, M. R.; McCarter, J. D.; Flynn, G. C. J. Biol. Chem. 2013, 288 (23), 1637116382. 32. Amano, M.; Hasegawa, J.; Kobayashi, N.; Kishi, N.; Nakazawa, T.; Uchiyama, S.; Fukui, K. Anal. Chem. 2011, 83 (10), 3857-3864. 33. Zhang, Q.; Flynn, G. C. J. Biol. Chem. 2013, 288 (48), 34325-34335. 34. Pristatsky, P.; Cohen, S. L.; Krantz, D.; Acevedo, J.; Ionescu, R.; Vlasak, J. Anal. Chem. 2009, 81 (15), 6148-6155.

ACS Paragon Plus Environment

Analytical Chemistry

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

35. Gu, S.; Wen, D.; Weinreb, P. H.; Sun, Y.; Zhang, L.; Foley, S. F.; Kshirsagar, R.; Evans, D.; Mi, S.; Meier, W.; Pepinsky, R. B. Anal. Biochem. 2010, 400 (1), 89-98. 36. Gadgil, H. S.; Bondarenko, P. V.; Pipes, G. D.; Dillon, T. M.; Banks, D.; Abel, J.; Kleemann, G. R.; Treuheit, M. J. Anal. Biochem. 2006, 355 (2), 165-174. 37. Ren, D.; Pipes, G.; Xiao, G.; Kleemann, G. R.; Bondarenko, P. V.; Treuheit, M. J.; Gadgil, H. S. J. Chromatogr. A 2008, 1179 (2), 198-204. 38. Banks, D. D.; Gadgil, H. S.; Pipes, G. D.; Bondarenko, P. V.; Hobbs, V.; Scavezze, J. L.; Kim, J.; Jiang, X. R.; Mukku, V.; Dillon, T. M. J. Pharm. Sci. 2008, 97 (2), 775-790. 39. Buchanan, A.; Clementel, V.; Woods, R.; Harn, N.; Bowen, M. A.; Mo, W.; Popovic, B.; Bishop, S. M.; Dall'Acqua, W.; Minter, R.; Jermutus, L.; Bedian, V. MAbs. 2013, 5 (2), 255-262. 40. Zhang, B.; Cockrill, S. L. Anal. Chem. 2009, 81 (17), 7314-7320. 41. Neill, A.; Nowak, C.; Patel, R.; Ponniah, G.; Gonzalez, N.; Miano, D.; Liu, H. Anal. Chem. 2015, 87 (12), 6204-6211. 42. Kshirsagar, R.; McElearney, K.; Gilbert, A.; Sinacore, M.; Ryll, T. Biotechnol. Bioeng. 2012, 109 (10), 2523-2532.

ACS Paragon Plus Environment

Page 18 of 28

Page 19 of 28

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

Analytical Chemistry

Figure legend Figure 1. The disulfide bond linkage of the three IgG2 isoforms. Unique disulfide bond structures of A, A/B and B isoforms from Lys-C/Trypsin digestion of the F(ab’)2 fragment. The A isoform contains two unique disulfide linked peptides with monoisotopic molecular weights of 1534.7 Da and 2761.1 Da respectively. The peptide A/B form contains a unique peptide with a monoisotopic molecular weight of 4295.9 Da and also the same 1534.7 Da. B form contains a peptide with a monoisotopic molecular weight of 5830.6 Da. The amino acid sequence of GEC contains the fifth cysteine residue of the light chain. The amino acid sequence of GPSVFPLAPCSR contains the third cysteine residue of the heavy chain. The heavy chain hinge region peptide contains the four closely spaced cysteine residues. Figure 2. Deconvoluted mass spectra of the recombinant monoclonal antibody (A) and its F(ab)2 fragment (B). The shoulder with the question mark has a molecular weight increase of approximately 28-34 Da. The FabRICATOR cleavage site is shown as inset in panel B.

ACS Paragon Plus Environment

Analytical Chemistry

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

Figure 3. The work flow for determination the location of the trisulfide bond in the A form hinge region peptide (Left) and the corresponding mass spectra (Right panel). The mass spectra correspond to the triply charged hinge region peptide without a trisulfide bond (A), with a trisulfide bond (B) and the reaction product of the trisulfide bond containing peptide after one cycle of Edman degradation (C). Figure 4. The work flow for determination the location of the trisulfide bond in the A/B form hinge region peptide (Left) and the corresponding mass spectra (Right panel). The mass spectra correspond to the quadruple charged hinge region peptide without a trisulfide bond (A), with a trisulfide bond (B) and the reaction products of the trisulfide bond containing peptide after one cycle (C) and two cycles (D) of Edman degradation. Figure 5. The work flow for determination of the location of cysteinylation in the A form hinge region peptide (Left) and the corresponding mass spectra (Right panel). The mass spectra correspond to the triply charged hinge region peptide without cysteinylation (A), with cysteinylation (B) and the reaction product of the cysteinylation containing peptide after one cycle (C) of Edman degradation. Figure 6. Locations of trisulfide bonds and cysteinylation in different isoforms. A isoform contains both trisulfide bonds and cysteinylation. Trisulfide bonds are localized to either between the inter light chain and heavy chain disulfide bond or between the first disulfide bond in the hinge region. Cysteinylation is located to the cysteine residues that form the first hinge region disulfide bond. A/B isoform contains trisulfide bonds. Trisulfide bonds are localized to either between the inter light chain and heavy chain disulfide bond or between the first disulfide bond in the hinge region. A/B isoform does not contain detectable level of cysteinylation. B isoform does not contain detectable levels of trisulfide bonds and cysteinylation.

ACS Paragon Plus Environment

Page 20 of 28

Page 21 of 28

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

Analytical Chemistry

Figure 1

ACS Paragon Plus Environment

Analytical Chemistry

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

Figure 2

ACS Paragon Plus Environment

Page 22 of 28

Page 23 of 28

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

Analytical Chemistry

Figure 3

ACS Paragon Plus Environment

Analytical Chemistry

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

Figure 4

ACS Paragon Plus Environment

Page 24 of 28

Page 25 of 28

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

Analytical Chemistry

Figure 5

ACS Paragon Plus Environment

Analytical Chemistry

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

Figure 6

ACS Paragon Plus Environment

Page 26 of 28

Page 27 of 28

Analytical Chemistry

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 ACS Paragon Plus Environment

Analytical Chemistry

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 ACS Paragon Plus Environment

Page 28 of 28