Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2022 Oct 5;7(41):36825–36835. doi: 10.1021/acsomega.2c05325

Proteomic Analysis of Adenovirus 5 by UHPLC-MS/MS: Development of a Robust and Reproducible Sample Preparation Workflow

Mostafa Zarei †,*, Jérôme Jonveaux †, Peng Wang ‡, Friedrich M Haller ‡, Bingnan Gu ‡, Atanas V Koulov †, Michael Jahn †
PMCID: PMC9583333  PMID: 36278084

Abstract

graphic file with name ao2c05325_0006.jpg

Adenoviruses (AdVs) have recently become widely used therapeutic vectors for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccine. AdVs are large, nonenveloped viruses with an icosahedral capsid formed from several proteins that encloses double-stranded DNA. These proteins are the main components and key players in initial stages of infection by the virus particles, so their heterogeneity and content must be evaluated to ensure product and process consistency. Peptide mapping can provide detailed information on these proteins, e.g., their amino acid sequences and post-translational modifications (PTMs), which is crucial for the development and optimization of the manufacturing processes. However, sample preparation remains the main bottleneck for successful proteomic analysis of the viral proteins (VPs) of AdVs due to their low concentrations and vast stoichiometric ranges. To address this problem, we have developed a fast and efficient protocol for preparing samples for proteomic analysis of VPs of AdV5 that requires no cleaning step prior to liquid chromatography–tandem mass spectrometry (LC-MS/MS). The approach enabled identification of 92% of amino acids in AdV5 VPs on average and quantification of 53 PTMs in a single LC-MS/MS experiment using trypsin protease. The data obtained demonstrate the method’s potential utility for supporting the development of novel AdV-based gene therapy products (GTPs).

Introduction

Adenoviruses (AdVs) have become widely employed in gene therapies for intracellular DNA delivery and as DNA-based vaccine vehicles. Recent advances in gene transfer technology have enabled the production of modified viral DNA to encode antigens of interest and led to the development of various AdV-based vaccines against HIV, influenza, and Ebola virus.1,2 In addition, several AdV-based therapeutic vector-based vaccines have been recently developed against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and given emergency approval for use by the World Health Organization (WHO) to help efforts to combat the Covid-19 pandemic.3−5

AdVs are nonenveloped viruses with diameters of 70–90 nm. Each virus particle has an icosahedral capsid formed from several proteins that encloses double-stranded DNA with a total molecular weight of about 150 MDa.6,7 AdV5 has 13 main proteins that are categorized as major proteins (Hexon, Penton base, and Fiber), cement/minor proteins (pIIIa, pVI, pVIII, and pIX), and core proteins (pV, pVII, pTP, pμ, AVP, and pIVa2).8 Major and minor proteins account for the highest and lowest percentages of the total weight of the AdV5 proteins (∼64.1 and ∼15.6%, respectively). Alteration of the type of cell line used or scale of production and purification can affect AdVs’ composition and influence the interactions of virus particles with cells and hence the products’ biological activity and potency. The VPs are major components of the virus particles, and their heterogeneity and content are routinely evaluated to ensure product and process consistency. Therefore, several assays for determining viral protein composition have been established for use in quality control processes to ensure that the proteins are expressed and viral particles are assembled correctly.9,10 The typically applied techniques are sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), capillary gel electrophoresis (CGE), and revered-phase (RP) chromatography.9−13

SDS-PAGE separates proteins into molecular-weight-based bands and quantifies them according to the bands’ intensities. This approach is simple and robust, but it cannot distinguish between different viral strains, unless there are sufficiently large differences in molecular weights of the proteins. Also, the method is labor intensive, and accordingly has low throughput. Alternatively, proteins in the bands can be separately identified by either matrix-assisted laser desorption ionization time-of-flight mass spectroscopy (MALDI-MS) or liquid chromatography–tandem MS (LC-MS/MS) after cutting them into smaller species followed by in-gel trypsin digestion. This approach is even more laborious and further limited by low recovery of long peptides from the gels.14,15

CGE has been applied for the identification and quantification of multiple VPs in influenza vaccines, but it requires extensive sample workup and method optimization. Another drawback is that it does not provide additional detailed information regarding new or unknown species.16,17

RP chromatography is one of the most widely and longest established techniques for measuring AdV particle concentrations through quantification of VPs. In the denaturing environment of an RP mobile phase, which typically includes trifluoroacetic acid (TFA) and acetonitrile (ACN), AdV particles dissociate into DNA and VPs, thereby yielding specific viral proteome fingerprints.10−13

Recently, VPs of AdV vector types 26 and 35 were separated by a fast RP method. This method has been successfully validated for monitoring the stability of the proteins and identifying serotypes, degradation products, and process intermediates. Proteins were identified in this approach by fraction collection followed by peptide mapping. The conditions were not compatible with direct LC-MS analysis, largely due to a high concentration of TFA (0.17%) and flow rate (0.6 mL/min).18

All of these techniques can potentially be used to separate different VPs and determine their purity, but unlike LC-MS/MS, they cannot provide an in-depth characterization on a molecular level. LC-MS/MS is a versatile technique that is capable of verifying amino acid sequences and monitoring post-translational modifications (PTMs) of proteins. However, the performance of this approach in the analysis of VPs can be strongly affected by their structural complexity (e.g., disparate protein abundances in complex matrices) and limited availability of samples from developmental steps. To overcome these problems, several approaches for mitigation have been developed. For example, fully automated two-dimensional LC-MS/MS has been applied for the characterization of the AdV5 proteome after tryptic digestion.19 The peptides from digested proteins were fractionated by strong cation exchange (SCX) chromatography with step gradients of ammonium chloride. Eluted samples were then directly loaded onto an RP column and subjected to LC-MS/MS analysis. Although two-dimensional chromatographic separation showed some advantages compared to one-dimensional LC-MS/MS, sequence coverages of the VPs were still low (2–33%). This might have been due to successive losses of peptides during multiple dilution steps of the digested samples.

To improve amino acid sequencing of VPs of AdV, nano-LC-MS/MS with a 3 h gradient, after digestion with three proteases (trypsin, chymotrypsin, and Lys-N), has been applied.8 This enabled the identification of 13 main structural proteins of AdV with a sequence coverage of 55–100% and a total average sequence coverage of 87.5%. However, extensive sample preparation is required, the throughput of the method is limited, and it does not meet the requirements for rapid pharmaceutical product and process development.

In another study, a combination of stable isotope labeling of amino acids (SILAC) in cell culture followed by high-throughput quantitative MS was used. This enabled identification of 27 VPs, covered by two or more distinct peptides, and detection of all structural VPs (major, core, and cement/minor proteins) and several nonstructural VPs.20

Here, we present the development of a new approach for fast and reproducible VP sample preparation that enables the generation of low-volume trypsin digests for single LC-MS/MS analysis. It involves precipitation of VPs followed by solubilization of the protein pellets in sodium deoxycholate (SDC)/N-dodecyl-β-d-maltoside (DDM) for denaturation and digestion, thus minimizing sample handling and losses. Direct LC-MS/MS analysis of the digested samples with no further pretreatment enables identification of all main structural proteins of AdV5 with high amino acid sequence coverage and quantification of associated PTMs. Our results strongly indicate that the method has sufficient reliability to provide analytical support for the development of AdV-based gene therapy products (GTPs).

Experimental Section

Materials and Methods

Chemicals

Tris 2-carboxyethyl phosphine (TCEP), ammonium bicarbonate (ABC), ultrapure formic acid, acetic acid, guanidine-HCl, chloroform, methanol, acetonitrile (ACN), water, trifluoroacetic acid (TFA), and sodium deoxycholate (SDC) were purchased from Sigma-Aldrich (St. Louis, MO). Vivaspin 500 (10 kDa MWCO) was purchased from Cytiva (Marlborough, MA). Sequencing-grade trypsin and N-dodecyl-β-d-maltoside (DDM) were purchased from Promega (Milwaukee, WI) and Thermo Fisher Scientific (Waltham, MA), respectively.

Vector Production and Purification

One vial of HEK293 RCB was thawed and expanded in various sizes of shake flasks every 3 or 4 days. When enough cell mass was reached, the culture was used to inoculate a WAVE20 perfusion bioreactor at a target seeding density. The culture was medium-exchanged and infected with wild-type AdV5. The culture then was lysed and clarified with filters and stored as column load. The two batches of column loads were thawed and purified using Source 15Q resin-packed HR16 column. The purified AdV5 were pooled and sterile-filtered using a 0.2 μm filter. The filtered samples were aliquoted for LC-MS analysis.

Sample Preparation

Samples of recombinant AdV5 were prepared as described above. We also used a recombinant human monoclonal IgG4 antibody (designated mAb A) produced and purified in-house using standard manufacturing procedures.

Peptide mapping was performed after reducing the volume of 300 μL of AdV5 samples (containing roughly 40 μg of VPs) to 200 μL before precipitation, as described below. The volumes were reduced by centrifuging the samples in Vivaspin filters with 10 kDa cutoff membranes at 10,000g and 20 °C for approximately 20 min.

Protein Precipitation by Chloroform/Methanol/Water

VPs were precipitated with chloroform/methanol/water as previously described.21,22 Briefly, 800 μL of methanol, 200 μL of chloroform, and 600 μL of water were sequentially added to 200 μL of concentrated sample with short vortex-mixing and fast centrifugation steps (10 s at 14,000g) following each addition. The protein precipitate appeared at the interface as a white layer between upper and lower phases. The upper phase was carefully removed and discarded, then 200 μL of cold methanol was added to the remaining mixture. After centrifugation at 14,000g and 4 °C for 10 min, the supernatant was removed. Finally, the pellet was dried by vacuum centrifugation.

Enzymatic Digestion of mAb A

Portions (60 μg) of mAb A were dissolved in 5 μL of a mixture of SDC (0.5, 0.75, 1.0, 1.25, 1.5, 1.75, and 2.0% w/v) and DDM (0.5% w/v) in 50 mM ammonium bicarbonate (pH 8.0). Then, following vortex-mixing at RT, the samples were diluted to 0.1, 0.15, 0.2, 0.25 0.3, 0.35, and 0.4% w/v of SDC and 0.1% w/v DDM concentrations by adding 20 μL of 50 mM ammonium bicarbonate (pH 8.0). Proteins were reduced by adding 1 μL of 500 mM TCEP and incubating the resulting mixtures for 30 min at 50 °C. The samples were subjected to trypsin digestion (with a 30:1 w/w protein-to-protease ratio) for 3 h at 37 °C, then the digests were transferred directly to an HPLC vial for LC-MS/MS analysis, as described below.

Enzymatic Digestion of Viral Proteins

Protein pellets prepared as described above were dissolved in 5 μL of a mixture of SDC (1.75% w/v) and DDM (0.5% w/v) in 50 mM ammonium bicarbonate (pH 8.0). Then, following vortex-mixing at RT, the samples were diluted to SDC and DDM concentrations of 0.35 and 0.1% w/v, respectively, by adding 20 μL of 50 mM ammonium bicarbonate (pH 8.0). Proteins in the samples were reduced and digested as described above. The digests were then transferred directly to an HPLC vial for LC-MS/MS analysis, as described below.

Liquid Chromatography–Mass Spectrometry

A system consisting of a Vanquish UPLC instrument coupled to an Orbitrap Fusion Lumos mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) was used for all analyses. For peptide mapping analysis, an Acquity UPLC Peptide CSH C18 column (130 Å, 1.7 μm, 2.1 mm × 150 mm, Waters Corporation) was used to separate peptides in the digested samples, with a mobile phase consisting of 0.1% formic acid in water (A) and acetonitrile (B). The peptides were separated and eluted at a flow rate of 0.25 mL/min with a linear gradient from 1% B to 30% B over 140 min, followed by 30% B to 40% B over 15 min. The column was then washed with 98% B for 10 min and conditioned with 1% B for 8 min before the next injection.

The mass spectrometer was operated in data-dependent mode in the 200–2000 m/z range with a spray voltage of 3.5 kV and a heated capillary temperature of 320 °C. Full-scan spectra were recorded with a resolution of 120,000 (full width at half-maximum resolution at 400 m/z) using an automatic gain control (AGC) target value of 2.0e5 with a maximum injection time of 100 ms. Up to 20 of the most intense ions with 2–8 charge states were selected for higher-energy c-trap dissociation (HCD) with a normalized collision energy of 35%. Fragment spectra were recorded at an isolation width of 2.5 Da and resolution of 15,000 using an AGC target value of 5.0e4 and maximum injection time of 200 ms. Dynamic exclusion was activated for 5 s within a 10 ppm window for precursor selection. Fragment ions were recorded by the orbitrap analyzer.

Data Analysis

For peptide mapping, tolerances of 6 and 20 ppm were applied in database searching against AdV5 viral capsid protein sequences for peptides detected in MS and MS/MS analyses, respectively. For this, FASTA protein sequence files were downloaded from the UniProt database (www.uniprot.org, downloaded 30.11.2021) including the complete reviewed entries (31) of “human adenovirus C serotype 5”. N-terminal acetylation, methionine and tryptophan oxidation, phosphorylation, and asparagine deamidation were included as variable modifications in the searches.

Results and Discussion

Development of a Robust and Reproducible Sample Preparation Workflow

There have been strenuous efforts to develop a universal, robust, and reproducible sample processing strategy for proteomic analysis.23−27 Important factors to consider when choosing procedures to minimize sample losses and maximize detection of peptides by MS include sample concentration and matrix type. However, there is little consensus regarding optimal sample processing protocols, and this step remains the main bottleneck for successful proteomic analysis. Sample losses during the processing steps are inevitable, which does not critically affect the performance of most methods if sufficiently large quantities of the starting materials (at least 50–150 μg of protein) are available. However, sample losses can pose major problems in proteomic analysis of GTPs, as typically only limited amounts of materials are available during developmental stages (e.g., early clinical phases). The complex sample matrices and wide molecular-weight ranges of VPs in GTPs compared to those in recombinant protein therapeutics also complicate their proteomic analysis. Thus, it is challenging to apply general classical sample processing workflows for proteomic analysis of GTPs and a new approach is needed. Such an approach must be able to overcome these problems, efficiently remove interferences in sample matrices, and disassemble capsid proteins with no need for any further treatment. It should also involve minimum liquid–liquid handling steps, with no protein- or peptide-level clean-up step, and cause minimal artificial modification in sample processing to enable accurate quantification of PTMs. Moreover, it should require the use of a single protease that can digest and provide high amino acid sequence coverage of all of the VPs in the low amounts of AdV available in samples from early developmental steps.

We previously described a sample preparation method for proteomic analysis of adeno-associated viruses (AAVs) that circumvents many of the challenges associated with the common classical approach. The structural composition (e.g., capsid proteins and double-stranded DNA) and matrix complexity of AdVs are similar to those of AAVs, but the throughput of this method could be limited by the low relative abundance of several components of the AdV5 proteome (0.1–0.3% of the total protein mass). Therefore, the AAV method required further optimization to reduce potential sample losses. Due to the limited available amounts of AdV5 materials, a monoclonal antibody (mAb A) was used initially for proof of concept in all optimization steps of the workflow, as described in the Materials and Methods section.

The method essentially involves concentration of virus particles using a centrifugal filter with an appropriate cutoff membrane (10 kDa). This is followed by the use of organic solvents to disassemble the particles into their structural proteins and DNA, with precipitation of the VPs at the interface between organic and aqueous phases.21,22 Sample matrix components and DNA partition into either the aqueous or organic phase and are removed prior to dryness of the protein precipitate. In this new approach, tertiary structures of the proteins are disrupted after precipitation and a further denaturation step with common chaotropic reagents (e.g., urea or guanidine hydrochloride) is not necessary.

To minimize sample processing steps, precipitated proteins are redissolved in a low volume of aqueous SDC/DDM solution, which is known to be an excellent denaturation and digestion solution for proteins. We and other authors have previously shown that dissolving proteins in SDC or DDM solution can increase potential cleavage sites by trypsin and enhance hydrophobic peptides’ solubility, thereby improving amino acid sequence coverage.28−30

After digestion of the proteins, it can be quite problematic to introduce a sample directly into an MS system, as a high concentration of SDC (1%) severely suppresses electrospray ionization and interferes with chromatographic separation of the peptides. Therefore, in previously published protocols, SDC is degraded after the addition of TFA and precipitates are removed by centrifugation.22,28 This step is a major bottleneck when low volumes of material are available. Under these conditions, the supernatant cannot be fully separated from the pellet, resulting in significant sample loss (up to 21%), as shown by the results of precipitating SDC in mAb A solution, washing the pellet with water, then subjecting both the supernatant and washed pellet to LC-MS/MS analysis (Supporting Figure 1). Washing the pellet with water or organic solvent and combining the wash solutions with the previously removed supernatant can improve the peptide recovery rate, but this requires an additional sample processing step to decrease the large sample volume by vacuum centrifugation prior to LC-MS/MS.31 In the study presented here, we further optimized the workflow by decreasing the percentage of SDC to enable omission of the SDC removal step and proceed directly to LC-MS/MS analysis.

To evaluate the possibility of direct LC-MS/MS analysis of the digested VPs, the following variables had to be assessed: solubility of SDC in an acidic mobile phase (precipitation risk), peak shapes and signal intensities of the peptides in the presence of SDC during chromatographic separation and MS detection, respectively, and the tryptic digestion of proteins in low concentrations of SDC.

Precipitation of SDC is very likely after injection of a digested sample into an acidic mobile phase, which might cause column blockage. However, there was no evidence of its precipitation during incubation of SDC/DDM and mobile phase A mixtures (with ratios varying from 1:1 to 1:10) at room temperature for 30 min. In addition, we observed no increase in column pressure during several successive LC-MS/MS analyses of digested samples of mAb A (with and without SDC removal).

Chromatographic separation and MS detection were further evaluated by digesting mAb A in the presence of SDC at various concentrations (0.1, 0.15, 0.2, 0.25, 0.3, 0.35, and 0.4% w/v) then introducing the resulting peptides directly to the LC-MS/MS system, as described in the Materials and Methods section. Signal intensities and peak shapes of eight known peptides (here named P1–P8) were compared to those obtained with the classical approach (digestion in 1% SDC followed by SDC precipitation and removal after addition of 2 μL of TFA). These peptides were selected as they had proper MS signal intensities, different retention times, no miss-cleaved, and no propensity to post-translational modification. The obtained data showed no differences in the chromatographic separation and peak shapes of the selected peptides across the entire gradient (Supporting Figures 2 and 3). Signal intensities of the selected peptides tended to increase with increasing amounts of SDC up to 0.4% w/v (Supporting Table 1).

Clearly, higher amounts of SDC can enhance the denaturation/solubilization of proteins (here mAb A) and enhance the protease performance. In addition, signal intensities of the selected peptides in the digested samples were significantly higher than those obtained with the approach involving SDC removal when the SDC percentage was ≥0.35% w/v. For example, signal intensities of P7 and P8 in samples with SDC ≥ 0.25% w/v were higher than in samples obtained after SDC removal (Supporting Table 1). This could be related to the hydrophobic nature of these peptides and partial precipitation that may occur during the SDC removal step. The signal intensities of the selected peptides with 0.35 and 0.4% w/v of SDC were comparable, but 0.35% w/v SDC was selected as the optimal concentration for further experiments to avoid potential contamination in the source of the mass spectrometer in long-term use. It should also be noted that the chosen model compound mAb A has a more rigid structure than VPs due to several disulfide bridges, so even lower concentrations of SDC could be sufficient for the effective denaturation and solubilization of VPs.

In our previous study, DDM was used as a combinatorial detergent to increase the solubility of the hydrophobic peptides and avoid their precipitation after SDC removal. The SDC removal step was omitted in the workflow presented here, but due to the lower SDC proportion (0.35% w/v) than in the standard workflow (1% w/v), DDM was added to enhance the proteins’ solubility and hence increase the performance of the protease. DDM has hydrophobic and hydrophilic properties, as its hydrophobic chain strongly interacts with the RP column and it only elutes at the end of the gradient (at ≥75% ACN). In addition, to avoid mass overloading of the RP column after injection of all digested materials, the effect of lower levels of DDM (0.1, 0.2, 0.3, 0.4, and 0.5% w/v) on signal intensities of the selected peptides was evaluated. The results showed that similar signal intensities were obtained with all of these DDM concentrations (data not shown), so 0.1% w/v DDM was utilized in our further experiments. In summary, a combination of SDC (0.35% w/v) and DDM (0.1% w/v) in 50 mM ammonium bicarbonate (pH 8.0) solution was chosen as the optimal buffer to dissolve and digest VPs of AdV5 for our characterization study.

UHPLC-MS/MS Analysis of the AdV5 Proteome

Experimental Design

Our workflow involves the concentration of AdV5 samples with an appropriate cutoff filter (10 kDa) to reduce their volume, followed by precipitation of VPs, their solubilization in SDC/DDM solution, reduction, and digestion with trypsin. Finally, generated peptides are analyzed by LC-MS/MS (Figure 1). The method has two purposes: to confirm the identities of the main VPs of AdV5 (high amino acid sequence coverage) and quantify their PTMs (Table 2).

Figure 1.

Figure 1

Open in a new tab

Experimental workflow. AdV5 proteins are concentrated using a 10 kDa cutoff filter. The capsid is denatured and viral proteins are precipitated. The protein pellet is dissolved in SDC/DDM solution, reduced, and digested by trypsin. Peptides are analyzed by liquid chromatography–tandem mass spectrometry (LC-MS/MS). The raw MS files are searched against spectra of AdV5 proteins in an appropriate database to determine amino acid sequence coverage and identify PTMs.

Table 2. Summary of PTMs of AdV5 Proteins with ≥0.5% Relative Abundance.
protein name mod. names amino acid sequence mod. AA mass error (ppm) % PTM (with ≥0.5% relative abundance)
hexon acetylation ATPSMMPQWSYMHISGQDASEYLSPGLVQFAR A2 0.4 97.4
acetylation/oxidation ATPSMMPQWSYMHISGQDASEYLSPGLVQFAR A2, M6 0.2 0.8
ATPSMMPQWSYMHISGQDASEYLSPGLVQFAR A2, W10 0.4 0.5
oxidation VVLYSEDVDIETPDTHISYMPTIK M303 0.6 0.7
VGNNFAMEINLNANLWR M460 0.4 0.6
deamidation ATETYFSLNNK N42 0.4 1.9
TTPMKPCYGSYAKPTNENGGQGILVK N244 0.5 19.4
IIENHGTEDELPNYCFPLGGVINTETLTK N403 0.8 0.9
TGQENGWEK N437 0.4 20.2
VGNNFAMEINLNANLWR N456 0.5 2.1
WSLDYMDNVNPFNHHR N536 0.4 0.5
SMLLGNGR N552 0.1 15.5
YKDYQQVGILHQHNNSGFVGYLAPTMR N825 0.9 0.6
phosphorylation ATPSMMPQWSYMHISGQDASEYLSPGLVQFAR S5 1.9 1.3
penton deamidation QNGVLESDIGVK N199 0.2 7.7
SFYNDQAVYSQLIR N477 0.8 0.9
NSIGGVQR N533 0.4 1.9
fiber deamidation ARPSEDTFNPVYPYDTETGPPTVPFLTPPFVSPNGFQESPPGVLSLR N37 0.7 13.8
LSEPLVTSNGMLALK N59 0.4 14.3
MGNGLSLDEAGNLTSQNVTTVSPPLKK N68 0.5 17.3
EPIYTQNGK N200 0.6 17.4
YGAPLHVTDDLNTLTVATGPGVTINNTSLQTK N231 0.4 1.7
NGDLTEGTAYTNAVGFMPNLSAYPK N482 0.5 11.3
SNIVSQVYLNGDK N523 0.3 15.9
phosphorylation TKSNINLEISAPLTVTSEALTVAAAAPLMVAGNTLTMQSQAPLTVHDSK S102 0.3 22.1
pIIIa acetylation MMQDATDPAVR M1 0.2 100.0
MQDATDPAVR M2 0.1 78.0
oxidation LMVTETPQSEVYQSGPDYFFQTSR M171 0.4 0.5
deamidation AALQSQPSGLNSTDDWR N22 0.1 2.3
LLGEEEYLNNSLLQPQR N493 1.1 0.8
NLPPAFPNNGIESLVDK N510 0.6 4.1
NLPPAFPNNGIESLVDK N511 0.4 17.5
phosphorylation RPSSLSDLGAAAPR S450 0.2 3.2
pVI acetylation MEDINFASLAPR M1 0.2 98.3
acetylation/oxidation MEDINFASLAPR M1, M1 0.3 1.7
oxidation AWNSSTGQMLR M66 0.2 0.8
deamidation AWNSSTGQMLR N60 0.7 2.3
pVIII acetylation SKEIPTPYMWSYQPQMGLAAGAAQDYSTR S2 0.6 100.0
oxidation EIPTPYMWSYQPQMGLAAGAAQDYSTR W11 0.6 1.2
INYMSAGPHMISR M34 0.4 0.5
deamidation NNLNPR N67 0.1 2.8
phosphorylation VRSPGQGITHLTIR S118 0.4 19.9
pIX acetylation STNSFDGSIVSSYLTTR S2 0.5 98.2
oxidation QNVMGSSIDGRPVLPANSTTLTYETVSGTPLETAASAAASAAAATAR M30 0.5 1.0
deamidation QNVMGSSIDGRPVLPANSTTLTYETVSGTPLETAASAAASAAAATAR N43 0.7 1.7
phosphorylation STNSFDGSIVSSYLTTR S2 1.4 1.8
pV oxidation HKDMLALPLDEGNPTPSLKPVTLQQVLPALAPSEEKR M126 0.4 0.7
ESGDLAPTVQLMVPK M175 0.4 0.7
QVAPGLGVQTVDVQIPTTSSTSIATATEGMETQTSPVASAVADAAVQAVAAAASK M235 0.8 0.8
phosphorylation QVAPGLGVQTVDVQIPTTSSTSIATATEGMETQTSPVASAVADAAVQAVAAAASK S224 0.7 1.6
pVII acetylation SILISPSNNTGWGLR S2 0.5 100.0
deamidation NYTPTPPPVSTVDAAIQTVVR N70 0.5 0.5
phosphorylation NYTPTPPPVSTVDAAIQTVVR S79 0.4 0.9
pμ oxidation MRGGILPLLIPLIAAAIGAVPGIASVALQAQRH M48 0.4 5.9
Open in a new tab

Amino Acid Sequence Coverage of AdV5 Capsid Proteins

Digested VPs of AdV5 were analyzed according to the developed method, as illustrated in Figure 1 and briefly described in the Material and Methods section. Due to the vast differences in relative abundances of AdV5 structural proteins (e.g., Hexon and pTP account for 59.5 and 0.1% of the total protein mass, respectively), several modifications of our previously reported LC-MS/MS setup were required.22 First, in silico trypsin digestion of VPs of AdV5 resulted in the generation of more than 350 peptides with four or more amino acids, so a longer gradient was needed to separate and identify the peptides. The highest sequence coverage was obtained when we used a longer gradient (150 or 180 min) rather than 100 min gradients (Figure 2). Additionally, to enhance retention and detection of small peptides, the column with 300 Å pores used in our previous study was replaced with one with smaller pores (130 Å).

Figure 2.

Figure 2

Open in a new tab

Total ion current (TIC) chromatogram obtained in an AdV peptide mapping experiment with the workflow described in Figure 1. Approximately 40 μg of VPs of AdV5 was loaded on the column.

We obtained average amino acid sequence coverages (two technical replicates) for the major, cement, and core proteins of 93.2, 97.7, and 85.0%, respectively (Table 1). The highest sequence coverages (≥99%) were obtained for Penton, pIIIa, and pIX proteins, and the lowest (55%) for pTP. The sequence coverage of pTP was also lowest in a previous study (36% with trypsin digestion and 55% when three different proteases were used).8 These results show that obtaining high sequence coverage of pTP is very challenging. There are several reasons for this. First, it accounts for the lowest percentage (0.1%) of the total mass of VPs of AdV5, so it is close to the detection limit of the applied UHPLC-MS/MS system. Second, covalent attachment of pTP to DNA could complicate its isolation by the presented approach.32,33 However, precipitation by chloroform/methanol/water can generally remove contaminating DNA, so we hypothesize that either pTP is not fully precipitated or precipitation of pTP with DNA could reduce trypsin’s ability to digest it properly. In addition, the use of a nano-LC-MS/MS system in combination with DNAase in the sample processing step could further enhance its sequence coverage. Despite these complications, we obtained 55% sequence coverage of pTP, clearly showing that the presented approach is a convenient method for the characterization of VPs with low abundance in complex matrices.

Table 1. Sequence Coverage of the Main Proteins of AdV5 Obtained Using the Presented Approach.
    UniProt ID no. residues (MW) #copies/virion sequence coveragea (%)
major proteins hexon P04133 952 (108.0 kDa) 720 (240 trimers) 91.4
penton base P12538 571 (63.3 kDa) 60 (12 pentamers) 98.9
fiber P11818 581 (61.6 kDa) 36 (12 trimers) 89.2
cement proteins pIIIa P12537 585 (65.2 kDa) 68 ± 2 99.1
pVI P24937 250 (27.0 kDa) 342 ± 4 98.6
pVIII P24936 227 (24.7 kDa) 127 ± 3 93.8
pIX P03281 140 (14.5 kDa) 247 ± 2 99.3
core proteins pV P24938 368 (41.4 kDa) 157 ± 1 94.3
pVII P68951 198 (22.0 kDa) 833 ± 19 91.9
pTP P04499 671 (76.5 kDa) 2 55.1
pμ Q2KS10 80 (8.8 kDa) ∼104 92.5
AVP P03253 204 (23.1 kDa) 15 ± 5 91.7
pIVa2 P03271 449 (50.9 kDa) 7 ± 1 84.2
Open in a new tab
a

Average of two replicates.

Lower sequence coverages for some proteins of AdV5 (e.g., pV and pVII) are related to the generation of small peptides by trypsin digestion that cannot be retained on the RP column. These proteins have high frequencies of arginine and lysine in their amino acid sequences and the resulting tryptic peptides (1–3 amino acids) cannot be retained even by a column with small (130 Å) pores. However, in this study, we obtained 94 and 92% amino acid sequence coverage for pV and pVII, respectively (Supporting Figure 4).

A recent LC-MS/MS-based study confirmed the presence of additional VPs, classified as “nonstructural proteins,” that might be present in AdV5 virus particles.20 To further investigate the VPs, peptides identified from the MS/MS data were used in searches against a list of human adenovirus C serotype 5 proteins in the UniProt database (31 reviewed proteins) and we found an additional 14 nonstructural viral proteins using the approach presented here. These proteins with their sequence coverages are listed in Supporting Table 2. They are expected to have low abundance compared to structural viral proteins of AdV5, and their presence could be related to the type of purification steps used. To our knowledge, there is insufficient information regarding the details and roles of these proteins in virus infections. However, the use of an LC-MS/MS system with a lower flow rate and higher sensitivity (e.g., a micro- or nano-LC-MS/MS system) could enhance their detection and identification if necessary.

Amino acid sequences of some main structural proteins can differ among AdV serotypes, and peptide mapping analysis can be used for their identification. For instance, hexon and fiber proteins of AdV5 and AdV2 significantly vary and can be used as markers for serotype identification. In contrast, other VPs have much more constant amino acid sequences (e.g., pVII and pμ in AdV5 and AdV2).

The obtained data show that this method greatly increased the amino acid sequence coverage of the main structural proteins and can discriminate between different adenovirus serotypes, so it can be used to confirm the identities of viral vectors.

Quantitative Analysis of PTMs of AdV5

Quantifying PTMs of the VPs at the site-specific level during the progression of an infection can provide important new information on the biological and pathogenetic mechanisms of viral infections. So far, limited information on these mechanisms has been obtained from quantitative proteomic approaches, and there is a need for further characterization of proteins in model viral systems like AdV5.15,34 In this study, we did not assess the functional roles of the identified PTMs in infection. However, our results highlight the sensitivity of the presented approach for identifying multiple PTMs in a single LC-MS/MS run, and hence its potential application for correlating PTMs with biological functions. Our analysis revealed a total of 53 PTMs in the main structural VPs of AdV5 (with at least 0.5% relative abundance), including phosphorylations, acetylations, deamidations, and oxidations (Table 2 and Supporting Figure 5). These modifications are known to affect the properties of GTPs (e.g., stability) and are influenced by multiple variables of the manufacturing process, including the cell line and purification techniques.15,35−37

These modifications are discussed below with an example of an MS/MS spectrum. PTMs of the viral proteins are known from intracellular maturation of capsid proteins, but some of these modifications (e.g., deamidations and oxidations) can also be induced during purification, storage, or sample processing in the proteomic workflow. Analysis of several therapeutic proteins by the presented approach confirmed that the applied workflow is unlikely to increase deamidation and oxidation levels (data not shown). However, further investigation is needed to clarify the actual rates of deamidation and oxidation induced during the manufacturing processes of GTPs.

Protein acetylation is recognized as an important regulatory event during diverse infections with human viruses.38 Acetylation of the N-terminal amino acid is the most frequently detected type of modification of a given amino acid of the VPs. It has been shown that N-terminal acetylation of VPs can play a significant role in their intracellular trafficking and entry into the nuclei.39 Therefore, it is important to confirm their N-terminal sequences and quantify their acetylation in GTPs. This modification is confirmed by the associated mass shift (+42.01 Da) in total peptide mass, and its localization is confirmed by data on the b-fragment ions generated in the MS/MS analysis. Acetylation mainly occurs when methionine of the N-terminus of VPs is cleaved off by methionine aminopeptidase, then the resulting N-terminal amino acid residue is acetylated.40 For instance, we found that ≥98.0% of N-terminals of pIX could be acetylated after methionine removal (Figure 3A). The MS/MS spectrum of the corresponding peptide showed a series of b-fragment ions (b1–b8) with +42.01 Da mass shifts on the serine residue, confirming acetylation at the N-terminus. However, the retained N-terminal methionine could also be acetylated, as demonstrated by a series of b-fragment ions (b1–b7) for the N-terminal of pVI with 42.01 Da mass shifts (Figure 3B). Some of the amino acids can undergo additional modifications, for example, the N-terminal acetylated methionine of pVI could also be oxidized. This modification was confirmed by MS/MS analysis of the b1-fragment ions, as shown in the enlarged mass spectra in Figure 3B,C.

Figure 3.

Figure 3

Open in a new tab

MS/MS spectra of N-terminal amino acid sequences of pIX and pVI. (A) Acetylated peptides of the N-terminus of pIX without methionine and with acetylation of the serine. (B) Acetylated methionine and (C) acetylated and oxidized methionine from the N-terminal peptides of pVI. Enlarged mass spectra of the b1 ions are displayed in (B, C), with the observed mass shift (+16 Da) in the corresponding insets.

Acetylation at N-terminals of some VPs (e.g., pentone base and fiber) cannot be quantified by the approach presented here, as these proteins have high arginine and lysine frequencies in their N-terminal amino acid sequences, so these peptides cannot be retained and detected, as discussed above. However, the use of a second protease (e.g., Asp-N) might provide the necessary information, as we previously described.22

Phosphorylation of serine, threonine, and tyrosine is another important type of PTM, which is involved in the stability of virus capsids, and thus likely affects the infection process.41 Phosphorylation is identified by neutral loss of H3PO4 (97.98 Da) from proteolytic peptide molecular ions in MS/MS spectra (Figure 4A). Relative abundances of phosphotyrosine (pY), phosphothreonine (pT), and phosphoserine (pS) in normally growing cells are approximately 2, 12, and 86%, respectively.42 The phosphorylated peptides have low intensities, and for comprehensive analysis, an additional chromatographic fractionation and enrichment step of the phosphopeptides is needed before LC-MS/MS analysis.28,42 Therefore, due to the absence of this step in our workflow, we identified a lower number of phosphorylation sites than in previous studies.15,34 In total, we identified seven serine phosphorylation sites, and no tyrosine or threonine phosphorylation sites, in accordance with their expected stoichiometric ratios.

Figure 4.

Figure 4

Open in a new tab

Typical MS/MS spectra of phosphorylated (A) and deamidated (B) peptides of VPs of AdV5.

Oxidation and deamidation are common PTMs in biopharmaceuticals, and they affect both biological activity and efficacy.43 These modifications are expected to be critical quality attributes (CQA) and therefore need to be evaluated throughout the development of viral vectors as pharmaceutical products. Nevertheless, the effects of these protein modifications on the efficacy and safety of viral vectors have been poorly studied, despite their potential importance. For example, deamidation of amino acids on the surface of AAV capsids reportedly leads to charge heterogeneity and changes in vector functions.35 Effects of oxidation and deamidation on the stability and properties of virus particles can be simulated by appropriate exposure to hydrogen peroxide and solutions with high pH, respectively. Chemical modification in virus particles can be monitored by several methods, e.g., capillary zone electrophoresis (CZE), dynamic light scattering (DLS), and electrophoretic light scattering (ELS).17 However, these methods only provide holistic information on virus particle levels, and cannot quantify or localize these modifications on protein or amino acid level.

RP chromatography is the only currently available method that can provide information on levels of oxidation in different VPs, through changes in protein retention times or appearance of new peaks, but it cannot localize these modifications in specific methionine’s or tryptophan’s of the VPs.

Using the novel approach presented here, we identified 12 oxidation sites in total in the main VPs of AdV5. Levels of oxidation were less than 1% at all of these sites except M48 in pμ (5.9%).

Deamidation of asparagine residues (resulting in a +0.98 Da mass shift) is a common, irreversible modification, and its mechanism has been studied by LC-MS/MS in great detail.44 It has been shown that deamidation rates depend on primary and higher-order structures of the proteins, pH, and temperature. In addition, asparagine in SNG, ENN, LNG, and LNN amino acid motifs is reportedly most prone to deamidation.45 An example of an MS/MS spectrum of the corresponding deamidated peptides showed a series of y10–y15 fragment ions with a +0.98 Da mass shift on the corresponding asparagine’s (Figure 4B). As sample processing and digestion times are short in the presented workflow, the detected deamidation level (Table 2) is highly unlikely to be artifactual. This was confirmed by the detection of low levels of deamidation in mAb A during the method optimization steps, as described above. In total, we detected 25 deamidation sites with mainly NG, NN, and NS motifs in the main VPs of AdV5 (Table 2). The highest level of deamidation in the VPs was associated with the NG motif. A similar study found that the main deamidations of AAV8 are in hypervariable regions (HVRs) with an NG motif. HVRs are largely responsible for interactions with target cells and the immune system, so they play a significant role in decreasing transduction by altering receptor binding.35

As discussed above, the method presented here enables simultaneous quantification of multiple PTMs. Its applicability could be extended by changing the search parameters as appropriate for the quantification of other PTMs (e.g., N- and O-glycosylation) if necessary. However, even without such enhancement, it can provide more information than previous approaches about amino acid sequences and associated PTMs, thereby improving the manufacturing processes of GTPs.

Conclusions

The novel workflow for AdV VP analysis presented here affords substantially lower analysis times than traditional peptide mapping methods, which require the use of several proteases and long sample processing for sufficient in-depth characterization. We provide here the first demonstration that a solution containing SDC and DDM can be used for simultaneous VP denaturation and digestion prior to direct LC-MS/MS analysis without any further clean-up steps (e.g., SDC removal, buffer exchange, and desalting by C18 cartridges). Analysis of the main VPs by the developed approach enabled substantially higher than previous average sequence coverage (up to 92%) as well as the quantification of 53 PTMs with single LC-MS/MS runs.

Along with increased analytical performance, the minimization of sample preparation steps reduces the risk of protein or peptide losses and enables high-throughput analysis when numerous samples must be rapidly characterized to support the development, stability testing, and characterization of GTPs.

The comprehensive information provided by our approach can provide highly valuable mechanistic insights into viral infection that are difficult to obtain by other approaches. In addition, we envisage that our approach will motivate future research to monitor and control the host cell proteins of GTPs produced using different vectors and manufacturing processes.

Acknowledgments

The authors acknowledge Dr. Joachim Weber and the Forensic Chemistry team of Drug Product Services at Lonza for their support and helpful discussions.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.2c05325.

  • TIC and XIC of the optimization steps for selection of proper amount of SDC/DDM; amino acid sequence coverages of the main viral proteins of AdV5; MS/MS spectra of the detected PTMs; and average amino acid sequence coverage of nonstructural proteins of AdV5 (PDF)

The authors declare no competing financial interest.

Supplementary Material

ao2c05325_si_001.pdf (2.6MB, pdf)

References

  1. Tatsis N.; Ertl H. C. Adenoviruses as Vaccine Vectors. Mol. Ther. 2004, 10, 616–629. 10.1016/j.ymthe.2004.07.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Milligan I. D.; Gibani M. M.; Sewell R.; Clutterbuck E. A.; Campbell D.; Plested E.; Nuthall E.; Voysey M.; Silva-Reyes L.; McElrath M. J.; et al. Safety and Immunogenicity of Novel Adenovirus Type 26- and Modified Vaccinia Ankara-Vectored Ebola Vaccines: A Randomized Clinical Trial. JAMA 2016, 315, 1610–1623. 10.1001/jama.2016.4218. [DOI] [PubMed] [Google Scholar]
  3. Elkashif A.; Alhashimi M.; Sayedahmed E. E.; Sambhara S.; Mittal S. K. Adenoviral vector-based platforms for developing effective vaccines to combat respiratory viral infections. Clin. Transl. Immunol. 2021, 10, e1345 10.1002/cti2.1345. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Kremer E. J. Pros and Cons of Adenovirus-Based SARS-CoV-2 Vaccines. Mol. Ther. 2020, 28, 2303–2304. 10.1016/j.ymthe.2020.10.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Hasanpourghadi M.; Novikov M.; Ertl H. C. J. COVID-19 Vaccines Based on Adenovirus Vectors. Trends Biochem. Sci. 2021, 46, 429–430. 10.1016/j.tibs.2021.03.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Gallardo J.; Perez-Illana M.; Martin-Gonzalez N.; San Martin C. Adenovirus Structure: What Is New?. Int. J. Mol. Sci. 2021, 22, 5240 10.3390/ijms22105240. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Ahi Y. S.; Mittal S. K. Components of Adenovirus Genome Packaging. Front. Microbiol. 2016, 7, 1503 10.3389/fmicb.2016.01503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Benevento M.; Di Palma S.; Snijder J.; Moyer C. L.; Reddy V. S.; Nemerow G. R.; Heck A. J. R. Adenovirus composition, proteolysis, and disassembly studied by in-depth qualitative and quantitative proteomics. J. Biol. Chem. 2014, 289, 11421–11430. 10.1074/jbc.M113.537498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Roitsch C.; Achstetter T.; Benchaibi M.; Bonfils E.; Cauet G.; Gloeckler R.; L’Hôte H.; Keppi E.; Nguyen M.; Spehner D.; et al. Characterization and quality control of recombinant adenovirus vectors for gene therapy. J. Chromatogr. B: Biomed. Sci. Appl. 2001, 752, 263–280. 10.1016/S0378-4347(00)00557-0. [DOI] [PubMed] [Google Scholar]
  10. Takahashi E.; Cohen S. L.; Tsai P. K.; Sweeney J. A. Quantitation of adenovirus type 5 empty capsids. Anal. Biochem. 2006, 349, 208–217. 10.1016/j.ab.2005.11.014. [DOI] [PubMed] [Google Scholar]
  11. Siegel S. A.; Hutchins J. E.; Witt D. J. Purification of adenovirus hexon by high performance liquid chromatography. J. Virol. Methods 1987, 17, 211–217. 10.1016/0166-0934(87)90131-5. [DOI] [PubMed] [Google Scholar]
  12. Lehmberg E.; Traina J. A.; Chakel J. A.; Chang R. J.; Parkman M.; McCaman M. T.; Murakami P. K.; Lahidji V.; Nelson J. W.; Hancock W. S.; et al. Reversed-phase high-performance liquid chromatographic assay for the adenovirus type 5 proteome. J. Chromatogr. B: Biomed. Sci. Appl. 1999, 732, 411–423. 10.1016/S0378-4347(99)00316-3. [DOI] [PubMed] [Google Scholar]
  13. McVey D.; Zuber M.; Ettyreddy D.; Reiter C. D.; Brough D. E.; Nabel G. J.; King C. R.; Gall J. G. Characterization of human adenovirus 35 and derivation of complex vectors. Virol. J. 2010, 7, 276 10.1186/1743-422X-7-276. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Liu Y-H.; Vellekamp G.; Chen G.; Mirza U. A.; Wylie D.; Twarowska B.; Tang J. T.; Porter F. W.; Wang S.; Nagabhushan T. L.; Pramanik B. N. Proteomic study of recombinant adenovirus 5 encoding human p53 by matrix-assisted laser desorption/ionization mass spectrometry in combination with database search. Int. J. Mass Spectrom. 2003, 226, 55–69. 10.1016/S1387-3806(02)00975-2. [DOI] [Google Scholar]
  15. Bergström Lind S.; Artemenko K. A.; Elfineh L.; Zhao Y.; Bergquist J.; Pettersson U. Post translational modifications in adenovirus type 2. Virology 2013, 447, 104–111. 10.1016/j.virol.2013.08.033. [DOI] [PubMed] [Google Scholar]
  16. van Tricht E.; Geurink L.; Pajic B.; Nijenhuis J.; Backus H.; Germano M.; Somsen G. W.; Sänger-van de Griend C. E. New capillary gel electrophoresis method for fast and accurate identification and quantification of multiple viral proteins in influenza vaccines. Talanta 2015, 144, 1030–1035. 10.1016/j.talanta.2015.07.047. [DOI] [PubMed] [Google Scholar]
  17. Geurink L.; van Tricht E.; van der Burg D.; Scheppink G.; Pajic B.; Dudink J.; Sänger-van de Griend C. Sixteen capillary electrophoresis applications for viral vaccine analysis. Electrophoresis 2022, 43, 1068–1090. 10.1002/elps.202100269. [DOI] [PubMed] [Google Scholar]
  18. van Tricht E.; de Raadt P.; Verwilligen A.; Schenning M.; Backus H.; Germano M.; Somsen G. W.; Sänger-van de Griend C. E. Fast, selective and quantitative protein profiling of adenovirus-vector based vaccines by ultra-performance liquid chromatography. J. Chromatogr. A 2018, 1581–1582, 25–32. 10.1016/j.chroma.2018.10.045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Chelius D.; Huhmer A. F.; Shieh C. H.; Lehmberg E.; Traina J. A.; Slattery T. K.; Pungor E. Jr. Analysis of the Adenovirus Type 5 Proteome by Liquid Chromatography and Tandem Mass Spectrometry Methods. J. Proteome Res. 2002, 1, 501–513. 10.1021/pr025528c. [DOI] [PubMed] [Google Scholar]
  20. Alqahtani A.; Heesom K.; Bramson J. L.; Curiel D.; Ugai H.; Matthews D. A. Analysis of purified wild type and mutant adenovirus particles by SILAC based quantitative proteomics. J. Gen. Virol. 2014, 95, 2504–2511. 10.1099/vir.0.068221-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Wessel D.; Flügge U. I. A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Anal. Biochem. 1984, 138, 141–143. 10.1016/0003-2697(84)90782-6. [DOI] [PubMed] [Google Scholar]
  22. Zarei M.; Wang P.; Jonveaux J.; Haller F. M.; Gu B.; Koulov A. V.; Jahn M. A novel protocol for in-depth analysis of recombinant adeno-associated virus capsid proteins by UHPLC-MS/MS. Rapid Commun. Mass Spectrom. 2022, 36, e9247 10.1002/rcm.9247. [DOI] [PubMed] [Google Scholar]
  23. Kanu A. B. Recent developments in sample preparation techniques combined with high-performance liquid chromatography: A critical review. J. Chromatogr. A 2021, 1654, 462444 10.1016/j.chroma.2021.462444. [DOI] [PubMed] [Google Scholar]
  24. Wojtkiewicz M.; Berg Luecke L.; Kelly M. I.; Gundry R. L. Facile Preparation of Peptides for Mass Spectrometry Analysis in Bottom-Up Proteomics Workflows. Curr. Protoc. 2021, 1, e85 10.1002/cpz1.85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Burns A. P.; Zhang Y. Q.; Xu T.; Wei Z.; Yao Q.; Fang Y.; Cebotaru V.; Xia M.; Hall M. D.; Huang R. A.; et al. Universal and High-Throughput Proteomics Sample Preparation Platform. Anal. Chem. 2021, 93, 8423–8431. 10.1021/acs.analchem.1c00265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hughes C. S.; Sorensen P. H.; Morin G. B. A Standardized and Reproducible Proteomics Protocol for Bottom-Up Quantitative Analysis of Protein Samples Using SP3 and Mass Spectrometry. Methods Mol. Biol. 2019, 1959, 65–87. 10.1007/978-1-4939-9164-8_5. [DOI] [PubMed] [Google Scholar]
  27. Rogers J. C.; Bomgarden R. D. Sample Preparation for Mass Spectrometry-Based Proteomics; from Proteomes to Peptides. Adv. Exp. Med. Biol. 2016, 919, 43–62. 10.1007/978-3-319-41448-5_3. [DOI] [PubMed] [Google Scholar]
  28. Zarei M.; Sprenger A.; Rackiewicz M.; Dengjel J. Fast and easy phosphopeptide fractionation by combinatorial ERLIC-SCX solid-phase extraction for in-depth phosphoproteome analysis. Nat. Protoc. 2016, 11, 37–45. 10.1038/nprot.2015.134. [DOI] [PubMed] [Google Scholar]
  29. Zhang X. Less is More: Membrane Protein Digestion Beyond Urea-Trypsin Solution for Next-level Proteomics. Mol. Cell. Proteomics 2015, 14, 2441–2453. 10.1074/mcp.R114.042572. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Masuda T.; Tomita M.; Ishihama Y. Phase transfer surfactant-aided trypsin digestion for membrane proteome analysis. J. Proteome Res. 2008, 7, 731–740. 10.1021/pr700658q. [DOI] [PubMed] [Google Scholar]
  31. Lin Y.; Liu Y.; Li J.; Zhao Y.; He Q.; Han W.; Chen P.; Wang X.; Liang S. Evaluation and optimization of removal of an acid-insoluble surfactant for shotgun analysis of membrane proteome. Electrophoresis 2010, 31, 2705–2713. 10.1002/elps.201000161. [DOI] [PubMed] [Google Scholar]
  32. Shenk T.Adenoviridae: The Viruses and Their Replication. In Fundamental Virology, 3rd ed.; 1996; pp 979–1016. [Google Scholar]
  33. Yao X.; Freas A.; Ramirez J.; Demirev P. A.; Fenselau C. Proteolytic 18O Labeling for Comparative Proteomics: Model Studies with Two Serotypes of Adenovirus. Anal. Chem. 2001, 73, 2836–2842. 10.1021/ac001404c. [DOI] [PubMed] [Google Scholar]
  34. Källsten M.; Gromova A.; Zhao H.; Valdes A.; Konzer A.; Pettersson U.; Bergström Lind S. Temporal characterization of the non-structural Adenovirus type 2 proteome and phosphoproteome using high-resolving mass spectrometry. Virology 2017, 511, 240–248. 10.1016/j.virol.2017.08.032. [DOI] [PubMed] [Google Scholar]
  35. Giles A. R.; Sims J. J.; Turner K. B.; Govindasamy L.; Alvira M. R.; Lock M.; Wilson J. M. Deamidation of Amino Acids on the Surface of Adeno-Associated Virus Capsids Leads to Charge Heterogeneity and Altered Vector Function. Mol. Ther. 2018, 26, 2848–2862. 10.1016/j.ymthe.2018.09.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Rumachik N. G.; Malaker S. A.; Poweleit N.; Maynard L. H.; Adams C. M.; Leib R. D.; Cirolia G.; Thomas D.; Stamnes S.; Holt K.; et al. Methods Matter: Standard Production Platforms for Recombinant AAV Produce Chemically and Functionally Distinct Vectors. Mol. Ther.--Methods Clin. Dev. 2020, 18, 98–118. 10.1016/j.omtm.2020.05.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Osamor V. C.; Chinedu S. N.; Azuh D. E.; Iweala E. J.; Ogunlana O. O. The interplay of post-translational modification and gene therapy. Drug Des., Dev. Ther. 2016, 10, 861–871. 10.2147/DDDT.S80496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Murray L. A.; Combs A. N.; Rekapalli P.; Cristea I. M. Methods for characterizing protein acetylation during viral infection. Methods Enzymol. 2019, 626, 587–620. 10.1016/bs.mie.2019.06.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Popa-Wagner R.; Porwal M.; Kann M.; Reuss M.; Weimer M.; Florin L.; Kleinschmidt J. A. Impact of VP1-specific protein sequence motifs on adeno-associated virus type 2 intracellular trafficking and nuclear entry. J. Virol. 2012, 86, 9163–9174. 10.1128/JVI.00282-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Hwang C. S.; Shemorry A.; Varshavsky A. N-terminal acetylation of cellular proteins creates specific degradation signals. Science 2010, 327, 973–977. 10.1126/science.1183147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Selzer L.; Kant R.; Wang J. C.; Bothner B.; Zlotnick A. Hepatitis B Virus Core Protein Phosphorylation Sites Affect Capsid Stability and Transient Exposure of the C-terminal Domain. J. Biol. Chem. 2015, 290, 28584–28593. 10.1074/jbc.M115.678441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Olsen J. V.; Blagoev B.; Gnad F.; Macek B.; Kumar C.; Mortensen P.; Mann M. Global, in vivo, and site-specific phosphorylation dynamics in signaling networks. Cell 2006, 127, 635–648. 10.1016/j.cell.2006.09.026. [DOI] [PubMed] [Google Scholar]
  43. Gupta S.; Jiskoot W.; Schoneich C.; Rathore A. S. Oxidation and Deamidation of Monoclonal Antibody Products: Potential Impact on Stability, Biological Activity, and Efficacy. J. Pharm. Sci. 2022, 111, 903–918. 10.1016/j.xphs.2021.11.024. [DOI] [PubMed] [Google Scholar]
  44. Gaza-Bulseco G.; Li B.; Bulseco A.; Liu H. C. Method to differentiate asn deamidation that occurred prior to and during sample preparation of a monoclonal antibody. Anal. Chem. 2008, 80, 9491–9498. 10.1021/ac801617u. [DOI] [PubMed] [Google Scholar]
  45. Chelius D.; Rehder D. S.; Bondarenko P. V. Identification and characterization of deamidation sites in the conserved regions of human immunoglobulin gamma antibodies. Anal. Chem. 2005, 77, 6004–6011. 10.1021/ac050672d. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao2c05325_si_001.pdf (2.6MB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES