Lessons from the Hamster: Cricetulus griseus ... - ACS Publications

Sep 6, 2017 - Brandeis University, Waltham, Massachusetts 02453, United States. §. Acibadem University, Medical Biochemistry, Istanbul, Maltepe, Turk...
0 downloads 0 Views 7MB Size
Subscriber access provided by - Access paid by the | UCSB Libraries

Article

Lessons from the hamster: Cricetulus griseus tissue and CHO cell line proteome comparison Kelley M. Heffner, Deniz Baycin Hizal, George S. Yerganian, Amit Kumar, Özge Can, Robert N. O'Meally, Robert Cole, Raghothama Chaerkady, Herren Wu, Michael A. Bowen, and Michael J. Betenbaugh J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.7b00382 • Publication Date (Web): 06 Sep 2017 Downloaded from http://pubs.acs.org on September 7, 2017

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.

Journal of Proteome Research 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 38

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

Journal of Proteome Research

Lessons from the hamster: Cricetulus griseus tissue and CHO cell line proteome comparison Kelley M. Heffner1*, Deniz Baycin Hizal1, George S. Yerganian2, Amit Kumar1, Özge Can3, Robert O’Meally4, Robert Cole4, Raghothama Chaerkady5, Herren Wu5, Michael A. Bowen5, Michael J. Betenbaugh1 1

Johns Hopkins University, Baltimore, MD 21218 USA Brandeis University, Waltham, MA 02453 USA 3 Acibadem University, Medical Biochemistry, Istanbul, Maltepe, TR 4 Johns Hopkins Medical Institute, Baltimore, MD 21205 USA 5 MedImmune, Gaithersburg, MD 20878 USA * Corresponding Author, email: [email protected] 2

Abstract Chinese hamster ovary cells represent the dominant host for therapeutic recombinant protein production. However, few large-scale datasets have been generated to characterize this host organism and derived CHO cell lines at the proteomics level. Consequently, an extensive label-free quantitative proteomics analysis of two cell lines (CHO-S and CHO DG44) and two Chinese hamster tissues (liver and ovary) was used to identify a total of 11801 unique proteins containing at least two unique peptides. 9359 unique proteins were identified specifically in the cell lines, representing a 56% increase over previous work. Additionally, 6663 unique proteins were identified across liver and ovary tissues providing the first Chinese hamster tissue proteome. Protein expression was more conserved within cell lines during both growth phases than across cell lines, suggesting large genetic differences across cell lines. Overall, both gene ontology and KEGG pathway analysis revealed enrichment of cell cycle activity in cells. In contrast, upregulated molecular functions in tissue include glycosylation and lipid transporter activity. Furthermore, cellular components including Golgi apparatus are upregulated in both tissues. In conclusion, this large-scale proteomics analysis enables us to delineate specific changes between tissues and cells derived from these tissues, which can help explain specific tissue function and the adaptations cells incur for applications in biopharmaceutical productions. Keywords CHO, cell culture, hamster tissue, proteomics, mass spectrometry, pathway analysis, KEGG, gene ontology, ovary, liver Introduction In 1957, Theodore Puck and colleagues established the Chinese hamster ovary (CHO) cell line that is used today across the biotechnology industry to generate therapeutic monoclonal antibodies (mAbs) and recombinant proteins that generate billions of dollars in sales 1. Biologics increasingly constitute a large percentage of top drugs and play important roles in research and development pipelines. Over the past three years there have been 11, 12, and 13 biological license application approvals in 2014, 2015, and 2016 respectively 2-4. While there are many possible production hosts including

1

ACS Paragon Plus Environment

Journal of Proteome Research

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

human embryonic kidney 5, mouse myeloma NS0, Saccharomyces cerevisiae, and Escherichia coli, CHO represents the most widely used biotechnology host for commercial proteins due to its wide regulatory acceptance, scalability and adaptability to manufacturing, and glycosylation capabilities among other characteristics 6. CHO cell lines exhibit relatively high growth rates, are easy to genetically engineer, and exhibit post-translational modifications that are compatible with humans. Increasing our knowledge about CHO cellular capabilities as well as the original Chinese hamster can provide insights that biotechnologists can use to improve production of therapeutic proteins through cell line engineering and bioprocessing modifications. Originally, ovarian tissues were isolated and used to establish a fibroblast-like cell type with nearly diploid chromosomes that became spontaneously immortal after months in culture 7. Although commonly considered ovary cells, CHO actually grew from the surrounding ovarian connective tissue 8. Over time, CHO use for industrial applications grew, as the cells were adapted to low-serum and eventually serum-free suspension culture. The first product made by CHO cells used the CHO-DXB11 cell line established at Columbia University to study dihydrofolate reductase (DHFR) activity 9. Chasin and Urlaub ultimately established the CHO DG44 cell line following deletion of both loci for DHFR on chromosome 2 10. The origin of CHO-S from the Puck CHO ancestor is less clear and may have resulted from adaptations to grow the cells in suspension simultaneously in academia and industry 8, 11. As a result, CHO cell lines represent a ‘quasi-species’ due to the high variability across parental and clonal cell lines 8. Today, derivatives of the original CHO cells exist across the biopharmaceutical industry and academia, having undergone numerous genetic modifications and adaptations, yet they serve as the dominant production platform cell line for biotherapeutics. Given the wide application and variability of the CHO cells in biotechnology as well as its origin from Chinese hamsters, burgeoning efforts are under way to begin to characterize the production host from different sources and in comparison with the Chinese hamster host organism. Recently, a range of ‘omics efforts have been initiated to characterize and differentiate CHO hosts from each other and the Chinese hamster. 12. A draft genome of CHO-K1 was completed through a collaborative effort in 2011 13 followed by the Cricetulus griseus genome in 2013 11 and several other cell lines. More recently, the CHO-DXB11 cell line was sequenced in order to aid efforts at understanding the DHFR negative phenotype and drift from other CHO cell lines 14. In addition, genomic BAC libraries were constructed for CHO-K1 and CHO-DG44 cells to study the distributions and rearrangements of hamster chromosomes 15. These initial efforts have paved the way for characterizing the diversity of the Chinese hamster against CHO parental and clonal cell lines at least at the genome level. Equally important will be efforts to decipher the characteristics of these cell lines and the host organism at the RNA, lipid, and protein level based on other ‘omics tools including transcriptomics, lipidomics, and proteomics among others. Recently, diverse transcriptomics data sets for CHO cell lines under various culture conditions were compiled into an online transcript database 16. The approach enabled the assembly of over 65000 CHO cell line transcripts 16. In another study, transcriptomics was used to compare CHO cell lines and hamster tissue by RNA-seq 17. Variability in gene expression was attributed to differences in glycolysis and glycosylation thus suggesting the existence of metabolic engineering targets that could be exploited 17. Recently, a comparison of cell

2

ACS Paragon Plus Environment

Page 2 of 38

Page 3 of 38

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

Journal of Proteome Research

lines by lipidomics revealed that phosphatidylethanolamine and phosphatidylcholine represent the major lipid classes in CHO, SP2/0, and HEK cells 18. HEK cells show increased lipid content, suggesting that differences in cell lipids can contribute to cell secretory capacity 18. These studies highlight how increased ‘omics data sets for CHO can be used for improved understanding of the production host. Proteomics can now identify and quantify thousands of cellular proteins for understanding cell physiology. These efforts, like those with transcriptomics and lipidomics, can help assist biotechnologists in their efforts to characterize distinct cell lines in developing stable host cell lines with desirable phenotypes. An initial proteomic analysis of the CHO-K1 cell line by our group identified CHO host cell proteins and pathways that are enriched and depleted compared to the CHO genome and transcriptome 19 . This large-scale profiling identified approximately 6000 cellular proteins and, together with functional analysis, revealed enrichment in genes related to protein processing and apoptosis at the proteomic level 19. In contrast, steroid hormone and glycosphingolipid metabolism were depleted 19. Additionally, bioinformatics tools were developed to enable the identification of CHO secreted proteins by detecting the presence of signal peptides, transmembrane domains, and cellular localization features 20. Over 1000 proteins were determined to represent the CHO supernatant or superome, which enables further research into high abundance host cell proteins (HCPs) for bioprocess development applications 20. These HCPs can accumulate in the production matrix and represent a challenge to bioprocessing development, especially for biotherapeutic purification steps. In another study, proteomics was used to identify host cell proteins (HCPs) that accumulate over cell culture duration and can negatively affect the process by interacting with recombinant protein 21. Proteomics was also used to evaluate changes in HCPs over 500 days in culture; results showed 92 HCPs with significant changes in expression with numerous HCPs that are known as difficult to purify 22. While many cell lines and different bioprocess conditions have been studied via transcriptomics and proteomics, comparative studies are still in their infancy particularly in relation to the original hamster host. As the relationship between derived cell lines and tissues of the original Cricetulus griseus is not well-documented, this study was undertaken to characterize the proteome of multiple CHO cell lines in the context of the Chinese hamster organism from which these lines were derived. Liver and ovarian tissues were chosen to represent biologically relevant tissues for this comparison. CHO cell lines have been created from isolated ovarian connective tissue, and the liver has been used for genome sequencing 8, 11 . Furthermore, one of the principal functions of the liver is to detoxify and metabolize chemicals; thus, the organ is highly metabolically active. Specifically, the liver plays a role in bile production, cholesterol production, protein synthesis, glucose storage and release, and substance clearance. In contrast, the ovary serves the female reproductive system by producing eggs and reproductive hormones. The organ contains extensive loose and dense connective tissue that acts as a covering and bed for vasculature, respectively. As CHO was established, it was sequentially adapted to grow efficiently in suspension and adherent using different culture media formulations. This research aims to begin to elucidate the different cellular characteristics at the proteomic level for representative CHO cell lines and hamster tissues. Consequently, two cell lines, CHO-S and CHO DG44, during both the exponential and stationary phases, were subjected to

3

ACS Paragon Plus Environment

Journal of Proteome Research

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

mass spectrometry proteomics analysis along with liver and ovary tissues. Peptide identification, protein quantification, and functional analysis were performed to characterize similarities and differences at the protein level between cell lines and tissues in order to explain physiological changes from tissue to cell line. While similar protein expression patterns were observed between CHO-S and CHO DG44 in both growth phases, these expression characteristics were distinct from those observed in liver and ovary tissues. These differences likely reflect the diverse physiological adaptations that these specific cell lines have undergone as well as the complex cellular makeup of tissues or organs. In particular, differences were detected between cell lines during processes including the cell cycle, DNA replication, and membrane transport. Between exponential and stationary phases, both CHO-S and CHO DG44 also shift physiological activities from replication activities to metabolic processes. In addition, liver and ovary exhibit enhanced metabolism compared to cell lines as determined by an examination of enriched pathways, including secretion, and in the case of ovary, extracellular matrix and adhesion. Overall, the results indicate distinct variability across physiological pathways of interest that are either upregulated or downregulated in tissues as compared to cells. These differences highlight unique aspects of cellular performance in organisms as compared to immortal cell lines in a tissue culture environment. Our findings will help us to better understand the changes that are manifested in adapting cells from tissue for different requirements, such as growth in suspension cultures and production of biotherapeutics for pharmaceutical applications. Experimental methods Briefly, the experiment section is divided into three parts: sample preparation, MS peptide identification, and bioinformatics analysis. Tissue isolation Chinese hamsters were generously provided by the lab of Dr. George Yerganian (Cytogen Research, Roxbury, MA). Euthanization was performed by CO2 and verified by puncture. Harvested tissues, including liver and ovary, were immediately frozen on dry ice and stored at -80 oC until analysis. Cell culture CHO-S and CHO DG44 were the two suspension CHO cell lines grown in shakeflask batch culture to collect samples at both exponential and stationary phases. CHO-S cells were cultured in CD-CHO medium supplemented with 8 mM glutamine (ThermoFisher Scientific, Waltham, MA). After achieving the growth curves (data not shown), samples were collected on day 2 for exponential phase and day 4 or 5 for stationary phase. CHO DG44 cells were culture in DG44 medium supplemented with 2 mM glutamine (ThermoFisher Scientific, Waltham, MA). Cells were maintained at 37 o C, 8 % CO2, and 120 RPM; viable cell density was determined by hemocytometer and trypan blue. For sample collection, approximately 3 million cells were spun down, washed with PBS on ice, frozen rapidly on dry ice, and stored at -80 oC until analysis. Sample preparation

4

ACS Paragon Plus Environment

Page 4 of 38

Page 5 of 38

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

Journal of Proteome Research

Samples for proteomics were thawed on ice and suspended in 2 % sodium dodecyl sulfate (SDS) supplemented with 0.1 mM phenylmethane sulfonyl fluoride (PMSF) and 1 mM ethylenediaminetetraacetic acid (EDTA), pH 7-8. Samples were lysed by sonicating thrice for 60 seconds at 20% amplitude followed by 90 seconds pause. Protein concentration was measured by bicinchoninic acid (BCA) protein assay after briefly spinning to remove cell debris. Three hundred micrograms of each sample was reduced in 10 mM tris(2-carboxyethyl)phosphine (TCEP), pH 7-8, at 60 oC for 1hr on a shaking platform. After bringing the sample to room temperature, approximately 17 mM iodacetamide was added to alkylate the sample to final concentration for 30 minutes. Next, samples were cleaned using 10 kDa filters to reduce the SDS concentration as suggested by the filter aided sample preparation (FASP) protocol 23. The samples were finally digested using trypsin/LysC enzyme mix at an enzyme to substrate ratio of 1:10 (Promega V507A), overnight at 37 oC on a shaking platform. 2D LC-MS Digested peptides (100 µg from each protein digest) were fractionated on the XBridge C18 Guard Column, 5 µm, 2.1 x 10 mm (Waters, Milton, MA). Fractions were concatenated into 48 samples prior to second dimension LC and MS analysis. The use of fractionation with equal peptides in each was designed to mimic biological replicates for each sample. Tandem mass spectrometry analysis of the peptides was carried out on the LTQ Orbitrap Velos (ThermoFisher Scientific, Waltham, MA) MS interfaced to the Eksigent nanoflow liquid chromatography system (Eksigent, Dublin, CA) with the Agilent 1100 auto sampler (Agilent Technologies, Santa Clara, CA). Peptides were enriched on a 2 cm trap column (YMC, Kyoto, Japan), fractionated on Magic C18 AQ, 5 µm, 100 Å, 75 µm x 15 cm column (Bruker, Billerica, MA), and electrosprayed through a 15 µm emitter (SIS, Ringoes, NY). Reversed phase solvent gradient consisted of solvent A (0.1 % formic acid) with increasing levels of solvent B to a final of 60 % (0.1 % formic acid, 90 % acetonitrile) over a period of 90 minutes. LTQ Orbitrap Velos parameters included 2.0 kV spray voltage, full MS survey scan range of 350-1800 m/z, data dependent HCD MS/MS analysis of top 10 precursors with minimum signal of 2000, isolation width of 1.9, 30 seconds dynamic exclusion limit and normalized collision energy of 35. Precursor and fragment ions were analyzed at 60000 and 7500 resolutions, respectively. Bioinformatics Peptide sequences were identified from isotopically resolved masses in MS and MS/MS spectra extracted with and without deconvolution using Thermo Scientific MS2 processor and Xtract software nodes. Data was searched against all entries in Cricetulus griseus database for CHO cell lines within Proteome Discoverer (Version 1.4, http://portal.thermo-brims.com). Oxidation on methionine, deamidation NQ and phosphoSTY were set as variable modifications, and carbamidomethyl on cysteine was set as fixed modifications in Mascot search mode used in Proteome Discoverer workflow. Mass tolerances on precursor and fragment masses were 15 ppm and 0.03 Da, respectively. Protein identifications were made using Proteome Discoverer software (Thermo) with stringent high confidence cutoff ( 0.05 Depletion

Figure 5. K-means clustering of molecular functions. Following gene ontology analysis, the resulting p-values were used to determine molecular functions that were similar or different in CHO cells vs. tissues. The color scale ranges from low p-value in green (significant enrichment) to high p-value in red (not enriched). A: Selected molecular functions enriched in cells (p < 0.05) but not tissues. B: Selected molecular functions enriched in tissues (p < 0.05) but not cells.

GO terms were clustered into categories using k-means in order to identify the terms enriched in cells only or enriched in tissues only (p < 0.05). The resulting heat 17

ACS Paragon Plus Environment

Journal of Proteome Research

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

maps are shown for different molecular functions in Figure 5. Among the molecular functions enriched in cells (Figure 5A) are DNA replication origin binding, DNA helicase activity, and RNA polymerase activity. Minichromosome maintenance deficient 5 (MCM5) is one gene involved in the molecular functions for DNA helicase activity (Figure 5A) and DNA replication initiation that was identified in this study as well as in a proteomics comparison of CHO cells undergoing treatment with sodium butyrate 49. Previous research found MCM5 was the most strongly downregulated protein as a result of sodium butyrate treatment, which results in a large reduction in growth rate and significant increases in recombinant protein productivity 49. MCM5 was also identified as a growth marker with significant differences in protein expression associated with the transition from exponential to stationary phase in CHO cells 45. Finally, recent efforts to differentiate high-producing CHO cell lines revealed significant downregulation of DNA replication initiation (MCM2, MCM5, MCM6) in the high-producers corresponding to decreased cell growth but increased production 50. Thus, efforts to knockdown genes involved in DNA replication may enable growth rate reductions that shift cell metabolism to an emphasis on protein production. Among the molecular functions enriched in tissues (Figure 5B) are AMP binding, glucose-6-phosphate dehydrogenase activity, glycoprotein binding, lipid transporter activity, and mannose binding. These activities are expected to be important for protein processing and secretion. AMP binding, for example, has been studied in agricultural biotechnology applications for its relevance in synthesis and modification of lipids 51, and lipids comprise essential organelle membranes involved in the secretory pathway. In addition, lipid transporter activity was enriched only in tissues; this GO term groups genes responsible for transporting lipids within and between cells. Example genes include glycolipid transfer protein and apolipoprotein. In rat liver, glycolipid transfer protein was found to control translocation of phospholipids across intracellular membranes 52. In addition, apolipoproteins are known to be synthesized in the liver, including apoB, aspoA-1, and apoE 53. Their regulation is dictated post-translationally depending on the lipid status within the cells. Overall changes in metabolism occur as immortal cell lines are derived from tissue; similarly, we observe differences between CHO cell lines and hamster tissues. Adhesion is an important characteristic of tissues and generally comprises various cell adhesion molecules and receptors, extracellular matrix proteins, and peripheral membrane proteins. Figure 5B highlights tissue-enriched functions such as cell adhesion molecule binding, glycoprotein binding, integrin binding, laminin receptor activity, polysaccharide binding, and receptor binding that are not enriched in cells. This emphasizes the adaption of cell lines to suspension culture occurred with a loss of adhesion activity. This has been observed in a previous study that adapted adherent Madin Darby canine kidney cells to suspension culture 54. Using proteomics, the study found that many changes in protein expression related to cytoskeletal architecture; specifically, myosin proteins were differentially expressed and played a role in cellular detachment 54. In our results, myosin family proteins were expressed only in tissues, verifying the inhibition of myosin occurs as cells lose adhesion properties in suspension culture. Laminin is another extracellular matrix component that is highly expressed in the outer layer of ovarian tissue and plays a role in follicle development 55. Recently, laminin subunit α2 was shown to control bacterial invasion when transfected in CHO cells 56.

18

ACS Paragon Plus Environment

Page 18 of 38

Page 19 of 38

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

Journal of Proteome Research

Overall, different laminin family members were upregulated in ovary tissue in our results including laminin subunit α3, laminin subunit α2 isoform X3, laminin subunit α5 isoform X1, and laminin subunit γ1 isoform X1. Finally, integrins play a role in cell attachment and show enriched activity in tissues compared to cells.

Figure 6. Cellular components enriched in tissues. The final result of gene ontology analysis was to identify cellular compartments enriched in tissues vs. CHO cells. The components are labeled and corresponding gene symbols are shown that correlated to the cellular component. Green and orange: enrichment in both liver and ovary tissues, but not in CHO-S or CHO DG44 cells. Red and blue: enrichment in ovary tissue, but not in CHO-S or CHO DG44 cells or liver tissue.

To illustrate our findings, Figure 6 highlights the cellular components enriched in tissues, including secretory machinery, such as Golgi cisternae and trans Golgi network transport vesicles, and extracellular matrix proteins. For example, Golgi phosphoprotein 3 (GOLPH3) plays an important role in cancer metabolism, as its depletion results in increased apoptosis following DNA damage 57. Subsequent overexpression of GOLPH3 can enhance cell survival as part of the DNA damage response 57. GOLPH3 also functions to control the localization of core 2 N-acetylglucosamine-transferase 1 and α-

19

ACS Paragon Plus Environment

Journal of Proteome Research

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

2,6-sialyltransferase 1 to transport vesicles 58. This can affect the utilization of these enzymes for glycosylation activities. In another experiment, GOLPH3 knockdown suppressed cell migration and decreased N-glycan sialylation as measured by HPLC and LC/MS 59. When α-2,6-sialyltransferase 1 was overexpressed in the knockdown cell line, cell migration and signaling were restored 59. GOLPH3 isoform X2 was identified in both liver and ovary tissue, but not in cell cultures. Another gene enriched in tissues is insulinlike growth factor 1 (IGF1), which is important for promotion of growth and prevention of apoptosis. In our results, IGF1 isoform X2 was identified in liver, which is the organ primarily responsible for IGF1 production. While IGF1 expression was confirmed in CHO-K1 cells through kinetic and growth curves 60, it also serves as an important component of serum for dependent cell lines and can be important for media development in serum-free conditions 61. Finally, extracellular matrix proteins were enriched specifically in ovary tissue. This highlights that a difference between the structure of liver and ovary contributes to different organ functions. The extensive ovary matrix helps to stabilize tissue structure and cell-cell communication. Collagen and laminin comprise the basal lamina, whereas integrins span the membrane to transmit signals within the cell. Collagens are highly abundant in ovary and their rearrangement is a major event in ovulation and follicle rupture 62. In addition, zona pellucida proteins, such as ZP1 and ZP2 identified in our results, are unique to ovary tissue. The zona pellucida is a layer of glycoproteins that encases oocytes. Similar to the human zona pellucida, mass spectrometry analysis of the golden hamster demonstrates the expression of all four of the ZP glycoproteins (1-4) in a targeted glycoproteomics study 63, The ZP proteins were identified in ovary tissue but not cells. Overall, these clustering results demostrate significant differences in the proteome between cell lines and tissues, especially involving proteins that are components of the extracellular matrix and secretory apparatus. KEGG functional analysis The GO functional analysis identifies molecular functions, biological processes, and cellular components that are enriched in samples at the proteomics level. Another functional analysis is KEGG, which also uses the hypergeometric distribution to identify enrichment and depletion values. In this case, the values represent pathways that are significantly upregulated or downregulated at the proteome level (p < 0.05). Analyses revealed between 117 and 125 enriched pathways in CHO cell cultures and about 140 enriched pathways in tissues. Depletion analyses revealed between 32 and 46 depleted pathways in CHO cells as compared to 30 depleted pathways in tissues to confirm more pathways are depleted in cell cultures as compared to tissues. These finding suggest that tissues are more metabolically active as measured by an increase in the number of enriched pathways and a decrease in the number of depleted pathways. While tissues show more enriched pathways than cultured cells, cells show slightly more enrichment during the stationary phases versus exponential phases Consequently, there are more pathways depleted during exponential phases than stationary phases. One example is glycosphingolipid metabolism; this pathway is depleted in CHO-S cells during the exponential phase (p = 0.038) but not at stationary phase (p > 0.05). However, it was not found as a depleted pathway in CHO DG44 cells.

20

ACS Paragon Plus Environment

Page 20 of 38

Page 21 of 38

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

Journal of Proteome Research

As another example, the MAPK signaling pathway is not enriched in CHO DG44 cells at exponential phase (p > 0.05) but is enriched at stationary phase (p = 0.045). While there are more enriched pathways in tissues than cultured cells, some of the most enriched pathways are similar between samples, including carbon metabolism, biosynthesis of amino acids, protein processing in the endoplasmic reticulum, and aminoacyl-tRNA biosynthesis. This suggests common metabolic activities that play a significant role in cells and tissues.

Figure 7. Pathway map of cell cycle. The Kyoto Encyclopedia of Genes and Genomes (KEGG) search and color tool was used to highlight genes that were identified at the proteome level for CHO cells and tissues. Blue represents genes identified at the proteome level in all samples, magenta represents genes identified at the proteome level in cultured cells only (present in CHOS exponential and CHO-S stationary and CHO DG44 exponential and CHO DG44 stationary), and orange represents genes identified at the proteome level in tissues only (present in liver and ovary but not in any CHO cells).

Shown in Figure 7 is the KEGG cell cycle pathway with colors representing genes that were identified at the proteome level. The cell cycle pathway is enriched for CHO-S and CHO DG44 at both exponential and stationary phases (p < 0.05) but it is not an enriched pathway for both liver and ovary tissues. Many cell cycle genes are observed in all samples; however, some are unique to cells or tissues only. In agreement with the GO analysis, many cell cycle functions are increased in CHO cell lines but not in tissues (genes colored magenta in Figure 7). In our results, CDK4,6 is highlighted in blue (Figure 7) which means it was identified in both CHO cell lines and tissues. One effort to

21

ACS Paragon Plus Environment

Journal of Proteome Research

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

shift cell lines from growth to production targeted the cell cycle G1-checkpoint through inhibition of cyclin-dependent kinase 4/6 (CDK4,6) 64. The inhibition precisely controlled cell growth while increasing productivity to 110 pg/cell/day and decreasing the percentage of high-mannose glycoforms 64. Another class of DNA replication molecules included the mini-chromosome maintenance 65 complex, most of which were observed in both cells and tissues. The complex unwinds double stranded DNA, recruits DNA polymerase, and initiates DNA synthesis, thus important to the survival of both cells and tissues. Both cells and tissues show MCM2, MCM3, MCM5, MCM6, and MCM7 at the proteome level. Interestingly, MCM4 was identified in all the cells but not in either tissue. Identification may be masked by the high abundance of other metabolic proteins in tissue that decrease the significance of DNA replication. Another important class of cell cycle proteins includes cyclins, such as CycA, CycB, and CycD, which were identified in the cell proteomes but not in liver or ovary tissue. Specifically, CycA concentration peaks during the G2 phase, CycB concentration peaks during the initiation of mitosis, and CycD maintains a high concentration throughout the cell cycle. CycA is involved in both the S phase and the G2 to M transition and may play a role in cancer progression 66. This is in agreement with the immortalized CHO cell lines. A study of DNA synthesis inhibition showed that CHO cells could accumulate protein and continue to grow as CycB levels accumulate 67. This aberrant growth and protein accumulation are determinants of cytotoxicity and cell death 67 . In addition to cyclins, there are many proteins essential for regulating the cell cycle that are upregulated in cells rather than tissues. These highlight key differences between immortalized CHO cell lines and hamster tissues. The origin recognition complex 68 represents another class of proteins identified in the proteome of cells but not tissues. The first step in DNA replication initiation is the assembly of the ORC; the ORC subunits then undergo changes in affinity for chromatin as the cell cycle progresses. Orc1 and Orc2 concentrations are constant throughout, but Orc1 can selectively detach from chromatin during M and S phases 69. This could explain the identification of detached Orc1 in mitotic CHO cells. Other detached ORC subunits exist during M phase and the complete ORC loses its function. Unlike in yeast, the Orc1 and Orc2 subunits are not tightly bound to the ORC in hamster 69. In another study, Orc1, Orc2, and Orc4 were identified as metabolically stable proteins in CHO cell lines 70. Orc6 also plays a role in cancer and its downregulation can lead to cell cycle arrest at G1 71. Thus, the ORC is important in actively growing cells. Interestingly, when comparing primary and transformed cell lines, it was observed that protein levels of several ORC subunits are upregulated in transformed cell lines 72. This is in agreement with our results that show enrichment of cell cycle proteins in CHO cell lines but not in liver or ovary tissue. Another cell cycle protein, p107, was identified in CHO cells only. p107 regulates the cell cycle through its phosphorylation at the S to M transition. This is in agreement with a proteome comparison between primary liver cells and immortalized tumor cells 35. Growth-promoting activities were upregulated in the Hepa1-6 tumor cells, including increased expression of p107 35. Related to p107, E2F4/5 was identified at the proteome level in cells only; it interacts with p107 to transduce TGFβ receptor signals upstream of CDK 73. E2F4/5 and other E2F homologs have been used as cell line engineering targets for prolonging culture due to the anti-apoptotic effect 74-75. This highlights how an ‘omics

22

ACS Paragon Plus Environment

Page 22 of 38

Page 23 of 38

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

Journal of Proteome Research

approach can identify worthwhile candidates for controlling the cell cycle and cell death in culture. There are few cell cycle genes identified at the proteome level in liver and ovary tissues but not in cell lines. One important example is cell division cycle 25 homolog A (Cdc25A), which regulates genomic stability by controlling cell cycle progression. Normal expression controls the G1 to S transition; overexpression leads to tumor growth 76 77 . Interestingly, Cdc25a has been targeted in CHO cell line engineering efforts. Lee evaluated wild-type and overexpressing Cdc25A cell lines by transgene copy number, recombinant protein productivity, and cell cycle progression, and found that overexpression of Cdc25A increased the specific productivity up to 2.9-fold and increased the G2 to M phase transition by 1.5-fold 78. Our inability to identify Cdc25A in any of the cell samples suggests that mRNA is not efficiently translated. Another example of a protein identified in tissues is Abl, which has various functions related to cell growth, differentiation, and migration. Interestingly, the attachment of fibroblasts to extracellular matrix directs the nuclear export of Abl, which may explain why it is identified only in the tissue proteomes 79. Abl kinases also play a role in cancer development, as increased expression is known to activate solid tumors, alter cell polarity, and induce invasion 80. Finally, stratifin (14-3-3 σ) was identified only in the tissue proteomes. The protein is a downstream target of p53 that enables tumor cell proliferation. Its methylation status and expression can be modified in cancers of the liver, breast, stomach, and colon, for example 81. In the ovary, the methylation of the stratifin gene reduces mRNA and protein expression. In turn, reduction or elimination of the stratifin methylation leads to changes in the expression levels in ovarian cancer tissue 82 . Thus, these genes can be intimately related with tissue cell cycle and cancer metabolism. Conclusions This study considerably expands on our current proteomics catalog for CHO cells by identifying proteins in two cell lines (CHO-S and CHO DG44) as well as hamster liver and ovary tissues. The total number of unique proteins with at least two unique peptides for CHO-S and CHO DG44 is 9359, representing a 56% improvement over previous work 19. Additionally, we established the first tissue proteome with 6663 unique proteins. Combining the results yielded a CHO cell line and tissue proteome of 11801 proteins at a 1% false discovery rate (FDR) with at least two unique peptides per protein. Protein intensities were compared between cell lines in exponential and stationary phase and tissues using NSAF, and functional analysis was performed using GO and KEGG hypergeometric distributions. Similarities and differences were observed at protein intensity level and in the analysis of genes and pathways. Interestingly, for both CHO-S and CHO DG44 cell lines, greater linearity in protein expression was observed within cell lines even in the two different growth stages than across cell lines, suggesting large genetic differences across cell lines. Also, the cell lines exhibited greater similarities in expression to the ovary tissues than to the liver. The GO and KEGG analyses also demonstrated significant differences in cell function and pathways between cell lines and tissues. There is a shift in activities from replication and transcription events, such as RNA splicing, to small molecule metabolic processes when comparing cells in the exponential phase to cells in the stationary phases, and to

23

ACS Paragon Plus Environment

Journal of Proteome Research

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

tissues. Furthermore, cellular components such as Golgi cisternae and the trans Golgi network were enriched in both liver and ovary tissues, but not in cells, which may suggest an increased secretory capacity in tissues that could be harnessed for CHO cell line engineering approaches. Additionally, cellular components such as extracellular matrix and endomembrane system are enriched specifically in the ovary. For example zona pellucida proteins 1 and 2 are membranous proteins upregulated specifically in the ovary. KEGG analyses indicated upregulation of the cell cycle pathway proteins in cells as compared to tissues. Thus, the functional analysis complements protein expression analyses to identify metabolic changes across samples. In summary, our results elucidate differences between cell lines and the Cricetulus griseus host from which these cell lines were originally derived. Overall, this proteomics analysis provides new insights on a greater number of protein and the overall protein intensity differences, as well as functional physiological differences related to genes and pathways, between CHO cells at different growth stages and the tissues of the Chinese hamster host. This information will enable us both to better understand this important production host at the level of proteins and pathways and to harness this knowledge in order to make more effective bio-production platforms in the future. Acknowledgements K.M.H. and M.J.B. acknowledge the financial support of MedImmune, LLC, for funding this project. In addition, this material is based upon work supported by the National Science Foundation Graduate Research Fellowship Program under Grant No. DGE1232825. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. References 1. Puck, T. T.; Sanders, P.; Petersen, D., Life Cycle Analysis of Mammalian Cells: II. Cells from the Chinese Hamster Ovary Grown in Suspension Culture. Biophysical Journal 1964, 4 (6), 441-450. 2. 2014 Biological Approvals. http://www.fda.gov/BiologicsBloodVaccines/DevelopmentApprovalProcess/Biolog icalApprovalsbyYear/ucm385842.htm. 3. 2015 Biological Approvals. http://www.fda.gov/BiologicsBloodVaccines/DevelopmentApprovalProcess/Biolog icalApprovalsbyYear/ucm434959.htm. 4. 2016 Biological Approvals. http://www.fda.gov/BiologicsBloodVaccines/DevelopmentApprovalProcess/Biolog icalApprovalsbyYear/ucm482392.htm. 5. Wang, J.; Svendsen, A.; Kmiecik, J.; Immervoll, H.; Skaftnesmo, K. O.; Planagumà, J.; Reed, R. K.; Bjerkvig, R.; Miletic, H.; Enger, P. Ø.; Rygh, C. B.; Chekenya, M., Targeting the NG2/CSPG4 Proteoglycan Retards Tumour Growth and Angiogenesis in Preclinical Models of GBM and Melanoma. PLOS ONE 2011, 6 (7), e23062.

24

ACS Paragon Plus Environment

Page 24 of 38

Page 25 of 38

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

Journal of Proteome Research

6. Kildegaard, H. F.; Baycin-Hizal, D.; Lewis, N. E.; Betenbaugh, M. J., The emerging CHO systems biology era: harnessing the ‘omics revolution for biotechnology. Current Opinion in Biotechnology 2013, 24 (6), 1102-1107. 7. Tjio, J. H.; Puck, T. T., GENETICS OF SOMATIC MAMMALIAN CELLS : II. CHROMOSOMAL CONSTITUTION OF CELLS IN TISSUE CULTURE. The Journal of Experimental Medicine 1958, 108 (2), 259-268. 8. Wurm, M. F., CHO Quasispecies—Implications for Manufacturing Processes. Processes 2013, 1 (3). 9. Kaufman, R. J.; Schimke, R. T. Amplification and loss of dihydrofolate reductase genes in a Chinese hamster ovary cell line. Mol Cell Biol 1981, 12, 1069-1076. 10. Urlaub, G.; Kas, E.; Carothers, A. M.; Chasin, L. A., Deletion of the diploid dihydrofolate reductase locus from cultured mammalian cells. Cell 1983, 33 (2), 405-12. 11. Lewis, N. E.; Liu, X.; Li, Y.; Nagarajan, H.; Yerganian, G.; O'Brien, E.; Bordbar, A.; Roth, A. M.; Rosenbloom, J.; Bian, C.; Xie, M.; Chen, W.; Li, N.; Baycin-Hizal, D.; Latif, H.; Forster, J.; Betenbaugh, M. J.; Famili, I.; Xu, X.; Wang, J.; Palsson, B. O., Genomic landscapes of Chinese hamster ovary cell lines as revealed by the Cricetulus griseus draft genome. Nat Biotechnol 2013, 31 (8), 759-65. 12. Kildegaard, H. F.; Baycin-Hizal, D.; Lewis, N. E.; Betenbaugh, M. J., The emerging CHO systems biology era: harnessing the 'omics revolution for biotechnology. Curr Opin Biotechnol 2013, 24 (6), 1102-7. 13. Xu, X.; Nagarajan, H.; Lewis, N. E.; Pan, S.; Cai, Z.; Liu, X.; Chen, W.; Xie, M.; Wang, W.; Hammond, S.; Andersen, M. R.; Neff, N.; Passarelli, B.; Koh, W.; Fan, H. C.; Wang, J.; Gui, Y.; Lee, K. H.; Betenbaugh, M. J.; Quake, S. R.; Famili, I.; Palsson, B. O.; Wang, J., The genomic sequence of the Chinese hamster ovary (CHO)-K1 cell line. Nat Biotech 2011, 29 (8), 735-741. 14. Kaas, C. S.; Kristensen, C.; Betenbaugh, M. J.; Andersen, M. R., Sequencing the CHO DXB11 genome reveals regional variations in genomic stability and haploidy. BMC Genomics 2015, 16 (1), 1-9. 15. Cao, Y.; Kimura, S.; Itoi, T.; Honda, K.; Ohtake, H.; Omasa, T., Construction of BAC-based physical map and analysis of chromosome rearrangement in Chinese hamster ovary cell lines. Biotechnol Bioeng 2012, 109 (6), 1357-67. 16. Rupp, O.; Becker, J.; Brinkrolf, K.; Timmermann, C.; Borth, N.; P¸hler, A.; Noll, T.; Goesmann, A., Construction of a Public CHO Cell Line Transcript Database Using Versatile Bioinformatics Analysis Pipelines. PLoS ONE 2014, 9 (1), e85568. 17. Vishwanathan, N.; Yongky, A.; Johnson, K. C.; Fu, H.-Y.; Jacob, N. M.; Le, H.; Yusufi, F. N. K.; Lee, D. Y.; Hu, W.-S., Global insights into the Chinese hamster and CHO cell transcriptomes. Biotechnology and Bioengineering 2015, 112 (5), 965-976. 18. Zhang, Y.; Baycin-Hizal, D.; Kumar, A.; Priola, J.; Bahri, M.; Heffner, K. M.; Wang, M.; Han, X.; Bowen, M. A.; Betenbaugh, M. J., High-Throughput Lipidomic and Transcriptomic Analysis To Compare SP2/0, CHO, and HEK-293 Mammalian Cell Lines. Analytical Chemistry 2016. 19. Baycin-Hizal, D.; Tabb, D. L.; Chaerkady, R.; Chen, L.; Lewis, N. E.; Nagarajan, H.; Sarkaria, V.; Kumar, A.; Wolozny, D.; Colao, J.; Jacobson, E.; Tian, Y.; O'Meally, R. N.; Krag, S. S.; Cole, R. N.; Palsson, B. O.; Zhang, H.; Betenbaugh, M., Proteomic analysis of Chinese hamster ovary cells. J Proteome Res 2012, 11 (11), 5265-76.

25

ACS Paragon Plus Environment

Journal of Proteome Research

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

20. Kumar, A.; Baycin-Hizal, D.; Wolozny, D.; Pedersen, L. E.; Lewis, N. E.; Heffner, K.; Chaerkady, R.; Cole, R. N.; Shiloach, J.; Zhang, H.; Bowen, M. A.; Betenbaugh, M. J., Elucidation of the CHO Super-Ome (CHO-SO) by Proteoinformatics. Journal of proteome research 2015, 14 (11), 4687-4703. 21. Levy, N. E.; Valente, K. N.; Choe, L. H.; Lee, K. H.; Lenhoff, A. M., Identification and characterization of host cell protein product-associated impurities in monoclonal antibody bioprocessing. Biotechnology and Bioengineering 2014, 111 (5), 904-912. 22. Valente, K. N.; Lenhoff, A. M.; Lee, K. H., Expression of difficult-to-remove host cell protein impurities during extended Chinese hamster ovary cell culture and their impact on continuous bioprocessing. Biotechnology and Bioengineering 2015, 112 (6), 1232-1242. 23. Wisniewski, J. R.; Zougman, A.; Nagaraj, N.; Mann, M., Universal sample preparation method for proteome analysis. Nat Methods 2009, 6 (5), 359-62. 24. McIlwain, S.; Mathews, M.; Bereman, M. S.; Rubel, E. W.; MacCoss, M. J.; Noble, W. S., Estimating relative abundances of proteins from shotgun proteomics data. BMC Bioinformatics 2012, 13, 308. 25. Sturn, A.; Quackenbush, J.; Trajanoski, Z., Genesis: cluster analysis of microarray data. Bioinformatics 2002, 18 (1), 207-8. 26. Le, H.; Chen, C.; Goudar, C. T., An evaluation of public genomic references for mapping RNA-Seq data from Chinese hamster ovary cells. Biotechnol Bioeng 2015, 112 (11), 2412-6. 27. Mudunuri, U.; Che, A.; Yi, M.; Stephens, R. M., bioDBnet: the biological database network. Bioinformatics 2009, 25 (4), 555-6. 28. Ashburner, M.; Ball, C. A.; Blake, J. A.; Botstein, D.; Butler, H.; Cherry, J. M.; Davis, A. P.; Dolinski, K.; Dwight, S. S.; Eppig, J. T.; Harris, M. A.; Hill, D. P.; IsselTarver, L.; Kasarskis, A.; Lewis, S.; Matese, J. C.; Richardson, J. E.; Ringwald, M.; Rubin, G. M.; Sherlock, G., Gene Ontology: tool for the unification of biology. Nat Genet 2000, 25 (1), 25-9. 29. Ogata, H.; Goto, S.; Sato, K.; Fujibuchi, W.; Bono, H.; Kanehisa, M., KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Research 1999, 27 (1), 29-34. 30. Latosinska, A.; Vougas, K.; Makridakis, M.; Klein, J.; Mullen, W.; Abbas, M.; Stravodimos, K.; Katafigiotis, I.; Merseburger, A. S.; Zoidakis, J.; Mischak, H.; Vlahou, A.; Jankowski, V., Comparative Analysis of Label-Free and 8-Plex iTRAQ Approach for Quantitative Tissue Proteomic Analysis. PLoS ONE 2015, 10 (9), e0137048. 31. Trinh, H. V.; Grossmann, J.; Gehrig, P.; Roschitzki, B.; Schlapbach, R.; Greber, U. F.; Hemmi, S., iTRAQ-Based and Label-Free Proteomics Approaches for Studies of Human Adenovirus Infections. International Journal of Proteomics 2013, 2013, 16. 32. Lauc, G.; Essafi, A.; Huffman, J. E.; Hayward, C.; Knežević, A.; Kattla, J. J.; Polašek, O.; Gornik, O.; Vitart, V.; Abrahams, J. L.; Pučić, M.; Novokmet, M.; Redžić, I.; Campbell, S.; Wild, S. H.; Borovečki, F.; Wang, W.; Kolčić, I.; Zgaga, L.; Gyllensten, U.; Wilson, J. F.; Wright, A. F.; Hastie, N. D.; Campbell, H.; Rudd, P. M.; Rudan, I., Genomics Meets Glycomics—The First GWAS Study of Human N-Glycome Identifies HNF1α as a Master Regulator of Plasma Protein Fucosylation. PLoS Genetics 2010, 6 (12), e1001256.

26

ACS Paragon Plus Environment

Page 26 of 38

Page 27 of 38

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

Journal of Proteome Research

33. Ning, K.; Fermin, D.; Nesvizhskii, A. I., Comparative Analysis of Different Label-Free Mass Spectrometry Based Protein Abundance Estimates and Their Correlation with RNA-Seq Gene Expression Data. Journal of Proteome Research 2012, 11 (4), 2261-2271. 34. Alonso-Gutierrez, J.; Kim, E.-M.; Batth, T. S.; Cho, N.; Hu, Q.; Chan, L. J. G.; Petzold, C. J.; Hillson, N. J.; Adams, P. D.; Keasling, J. D.; Garcia Martin, H.; Lee, T. S., Principal component analysis of proteomics (PCAP) as a tool to direct metabolic engineering. Metabolic Engineering 2015, 28, 123-133. 35. Pan, C.; Kumar, C.; Bohl, S.; Klingmueller, U.; Mann, M., Comparative Proteomic Phenotyping of Cell Lines and Primary Cells to Assess Preservation of Cell Type-specific Functions. Molecular & Cellular Proteomics : MCP 2009, 8 (3), 443-450. 36. Uhlén, M.; Fagerberg, L.; Hallström, B. M.; Lindskog, C.; Oksvold, P.; Mardinoglu, A.; Sivertsson, Å.; Kampf, C.; Sjöstedt, E.; Asplund, A.; Olsson, I.; Edlund, K.; Lundberg, E.; Navani, S.; Szigyarto, C. A.-K.; Odeberg, J.; Djureinovic, D.; Takanen, J. O.; Hober, S.; Alm, T.; Edqvist, P.-H.; Berling, H.; Tegel, H.; Mulder, J.; Rockberg, J.; Nilsson, P.; Schwenk, J. M.; Hamsten, M.; von Feilitzen, K.; Forsberg, M.; Persson, L.; Johansson, F.; Zwahlen, M.; von Heijne, G.; Nielsen, J.; Pontén, F., Tissue-based map of the human proteome. Science 2015, 347 (6220). 37. Tuffy, K. M.; Planey, S. L., Cytoskeleton-Associated Protein 4: Functions Beyond the Endoplasmic Reticulum in Physiology and Disease. ISRN Cell Biology 2012, 2012, 11. 38. Mastrangelo, A. J.; Hardwick, J. M.; Bex, F.; Betenbaugh, M. J., Part I. Bcl-2 and bcl-xL limit apoptosis upon infection with alphavirus vectors. Biotechnology and Bioengineering 2000, 67 (5), 544-554. 39. Mastrangelo, A. J.; Hardwick, J. M.; Zou, S.; Betenbaugh, M. J., Part II. Overexpression of bcl-2 family members enhances survival of mammalian cells in response to various culture insults. Biotechnology and Bioengineering 2000, 67 (5), 555564. 40. Templeton, N.; Lewis, A.; Dorai, H.; Qian, E. A.; Campbell, M. P.; Smith, K. D.; Lang, S. E.; Betenbaugh, M. J.; Young, J. D., The impact of anti-apoptotic gene Bcl-2 expression on CHO central metabolism. Metab Eng 2014, 25, 92-102. 41. Hossler, P.; Khattak, S. F.; Li, Z. J., Optimal and consistent protein glycosylation in mammalian cell culture. Glycobiology 2009, 19 (9), 936-49. 42. Yin, B.; Gao, Y.; Chung, C.-y.; Yang, S.; Blake, E.; Stuczynski, M. C.; Tang, J.; Kildegaard, H. F.; Andersen, M. R.; Zhang, H.; Betenbaugh, M. J., Glycoengineering of Chinese hamster ovary cells for enhanced erythropoietin N-glycan branching and sialylation. Biotechnology and Bioengineering 2015, 112 (11), 2343-2351. 43. Eser, P.; Demel, C.; Maier, K. C.; Schwalb, B.; Pirkl, N.; Martin, D. E.; Cramer, P.; Tresch, A., Periodic mRNA synthesis and degradation co‐operate during cell cycle gene expression. Molecular Systems Biology 2014, 10 (1), 717. 44. Edros, R. Z.; McDonnell, S.; Al-Rubeai, M., Using Molecular Markers to Characterize Productivity in Chinese Hamster Ovary Cell Lines. PLOS ONE 2013, 8 (10), e75935. 45. Carlage, T.; Kshirsagar, R.; Zang, L.; Janakiraman, V.; Hincapie, M.; Lyubarskaya, Y.; Weiskopf, A.; Hancock, W. S., Analysis of dynamic changes in the

27

ACS Paragon Plus Environment

Journal of Proteome Research

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

proteome of a Bcl-XL overexpressing Chinese hamster ovary cell culture during exponential and stationary phases. Biotechnol Prog 2012, 28 (3), 814-23. 46. Koltai, T., Clusterin: a key player in cancer chemoresistance and its inhibition. Onco Targets Ther 2014, 7, 447-56. 47. Gundemir, S.; Colak, G.; Tucholski, J.; Johnson, G. V. W., Transglutaminase 2: A molecular Swiss army knife. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 2012, 1823 (2), 406-419. 48. Singh, K. K.; Wachsmuth, L.; Kulozik, A. E.; Gehring, N. H., Two mammalian MAGOH genes contribute to exon junction complex composition and nonsense-mediated decay. RNA Biology 2013, 10 (8), 1291-1298. 49. Kantardjieff, A.; Jacob, N. M.; Yee, J. C.; Epstein, E.; Kok, Y.-J.; Philp, R.; Betenbaugh, M.; Hu, W.-S., Transcriptome and proteome analysis of Chinese hamster ovary cells under low temperature and butyrate treatment. Journal of Biotechnology 2010, 145 (2), 143-159. 50. Orellana, C. A.; Marcellin, E.; Schulz, B. L.; Nouwens, A. S.; Gray, P. P.; Nielsen, L. K., High-Antibody-Producing Chinese Hamster Ovary Cells Up-Regulate Intracellular Protein Transport and Glutathione Synthesis. Journal of Proteome Research 2015, 14 (2), 609-618. 51. Shockey, J.; Schnurr, J.; Browse, J., Characterization of the AMP-binding protein gene family in Arabidopsis thaliana: will the real acyl-CoA synthetases please stand up? Biochem Soc Trans 2000, 28 (6), 955-7. 52. Yamada, K.; Sasaki, T., Rat Liver Glycolipid Transfer Protein. A Protein Which Facilitates the Translocation of Mono- and Dihexosylcera-mides from Donor to Acceptor Liposomes. The Journal of Biochemistry 1982, 92 (2), 457-464. 53. Dixon, J. L.; Ginsberg, H. N., Hepatic synthesis of lipoproteins and apolipoproteins. Semin Liver Dis 1992, 12 (4), 364-72. 54. Chan, C. J.; Ekpenyong, A. E.; Golfier, S.; Li, W.; Chalut, K. J.; Otto, O.; Elgeti, J.; Guck, J.; Lautenschlager, F., Myosin II Activity Softens Cells in Suspension. Biophys J 2015, 108 (8), 1856-69. 55. Lee, V. H.; Britt, J. H.; Dunbar, B. S., Localization of laminin proteins during early follicular development in pig and rabbit ovaries. J Reprod Fertil 1996, 108 (1), 11522. 56. van Wijk, X. M.; Döhrmann, S.; Hallström, B. M.; Li, S.; Voldborg, B. G.; Meng, B. X.; McKee, K. K.; van Kuppevelt, T. H.; Yurchenco, P. D.; Palsson, B. O.; Lewis, N. E.; Nizet, V.; Esko, J. D., Whole-Genome Sequencing of Invasion-Resistant Cells Identifies Laminin α2 as a Host Factor for Bacterial Invasion. mBio 2017, 8 (1). 57. Buschman, M. D.; Rahajeng, J.; Field, S. J., GOLPH3 Links the Golgi, DNA Damage, and Cancer. Cancer Research 2015, 75 (4), 624-627. 58. Eckert, E. S. P.; Reckmann, I.; Hellwig, A.; Röhling, S.; El-Battari, A.; Wieland, F. T.; Popoff, V., Golgi Phosphoprotein 3 Triggers Signal-mediated Incorporation of Glycosyltransferases into Coatomer-coated (COPI) Vesicles. Journal of Biological Chemistry 2014, 289 (45), 31319-31329. 59. Isaji, T.; Im, S.; Gu, W.; Wang, Y.; Hang, Q.; Lu, J.; Fukuda, T.; Hashii, N.; Takakura, D.; Kawasaki, N.; Miyoshi, H.; Gu, J., An Oncogenic Protein Golgi Phosphoprotein 3 Up-regulates Cell Migration via Sialylation. Journal of Biological Chemistry 2014, 289 (30), 20694-20705.

28

ACS Paragon Plus Environment

Page 28 of 38

Page 29 of 38

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

Journal of Proteome Research

60. Mohamed, V. P.; Hashum, Y. Z. H.-Y.; Amid, A.; Mel, M.; Kamarulzaman, A. R.; Wahab, M. A. A.; Saberi, S. E. M., Chinese hamster ovary (CHO-K1) cells expressed native insulin-like growth factor-1 (IGF-1) gene towards efficient mammalian cell culture host system. African Journal of Biotechnology 2011, 10 (81), 18716-18721. 61. Mohamed, V. P.; Hashim, Y. Z. H.-Y.; Saberi, S. E. M.; Ibrahim, S. N.; Amid, A.; Mel, M., Effects of glutamine and serum on IGF-dependent CHO-K1 cell growth in Tflask. Advances in Environmental Biology 2014, 831. 62. Lind, A. K.; Weijdegard, B.; Dahm-Kahler, P.; Molne, J.; Sundfeldt, K.; Brannstrom, M., Collagens in the human ovary and their changes in the perifollicular stroma during ovulation. Acta Obstet Gynecol Scand 2006, 85 (12), 1476-84. 63. Izquierdo-Rico, M. J.; Jimenez-Movilla, M.; Llop, E.; Perez-Oliva, A. B.; Ballesta, J.; Gutierrez-Gallego, R.; Jimenez-Cervantes, C.; Aviles, M., Hamster zona pellucida is formed by four glycoproteins: ZP1, ZP2, ZP3, and ZP4. J Proteome Res 2009, 8 (2), 926-41. 64. Du, Z.; Treiber, D.; McCarter, J. D.; Fomina-Yadlin, D.; Saleem, R. A.; McCoy, R. E.; Zhang, Y.; Tharmalingam, T.; Leith, M.; Follstad, B. D.; Dell, B.; Grisim, B.; Zupke, C.; Heath, C.; Morris, A. E.; Reddy, P., Use of a small molecule cell cycle inhibitor to control cell growth and improve specific productivity and product quality of recombinant proteins in CHO cell cultures. Biotechnology and Bioengineering 2015, 112 (1), 141-155. 65. Kilcoyne, M.; Sharma, S.; McDevitt, N.; O’Leary, C.; Joshi, L.; McMahon, S. S., Neuronal glycosylation differentials in normal, injured and chondroitinase-treated environments. Biochemical and Biophysical Research Communications 2012, 420 (3), 616-622. 66. Desdouets, C.; Sobczak-Thepot, J.; Murphy, M.; Brechot, C., Cyclin A: function and expression during cell proliferation. Prog Cell Cycle Res 1995, 1, 115-23. 67. Kung, A. L.; Sherwood, S. W.; Schimke, R. T., Differences in the regulation of protein synthesis, cyclin B accumulation, and cellular growth in response to the inhibition of DNA synthesis in Chinese hamster ovary and HeLa S3 cells. Journal of Biological Chemistry 1993, 268 (31), 23072-80. 68. Borchi, E.; Bargelli, V.; Stillitano, F.; Giordano, C.; Sebastiani, M.; Nassi, P. A.; d'Amati, G.; Cerbai, E.; Nediani, C., Enhanced ROS production by NADPH oxidase is correlated to changes in antioxidant enzyme activity in human heart failure. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 2010, 1802 (3), 331-338. 69. Li, C. J.; DePamphilis, M. L., Mammalian Orc1 protein is selectively released from chromatin and ubiquitinated during the S-to-M transition in the cell division cycle. Mol Cell Biol 2002, 22 (1), 105-16. 70. McNairn, A. J.; Okuno, Y.; Misteli, T.; Gilbert, D. M., Chinese hamster ORC subunits dynamically associate with chromatin throughout the cell-cycle. Experimental cell research 2005, 308 (2), 345-356. 71. Gavin, E. J.; Song, B.; Wang, Y.; Xi, Y.; Ju, J., Reduction of Orc6 Expression Sensitizes Human Colon Cancer Cells to 5-Fluorouracil and Cisplatin. PLOS ONE 2009, 3 (12), e4054. 72. McNairn, A. J.; Gilbert, D. M., Overexpression of ORC subunits and increased ORC-chromatin association in transformed mammalian cells. J Cell Biochem 2005, 96 (5), 879-87.

29

ACS Paragon Plus Environment

Journal of Proteome Research

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

73. Chen, C.-R.; Kang, Y.; Siegel, P. M.; Massagué, J., E2F4/5 and p107 as Smad Cofactors Linking the TGFβ Receptor to c-myc Repression. Cell 2002, 110 (1), 19-32. 74. Krampe, B.; Al-Rubeai, M., Cell death in mammalian cell culture: molecular mechanisms and cell line engineering strategies. Cytotechnology 2010, 62 (3), 175-188. 75. Majors, B. S.; Arden, N.; Oyler, G. A.; Chiang, G. G.; Pederson, N. E.; Betenbaugh, M. J., E2F-1 overexpression increases viable cell density in batch cultures of Chinese hamster ovary cells. J Biotechnol 2008, 138 (3-4), 103-6. 76. Blomberg, I.; Hoffmann, I., Ectopic expression of Cdc25A accelerates the G(1)/S transition and leads to premature activation of cyclin E- and cyclin A-dependent kinases. Mol Cell Biol 1999, 19 (9), 6183-94. 77. Ray, D.; Kiyokawa, H., CDC25A phosphatase: a rate-limiting oncogene that determines genomic stability. Cancer Res 2008, 68 (5), 1251-3. 78. Lee, K. H.; Tsutsui, T.; Honda, K.; Asano, R.; Kumagai, I.; Ohtake, H.; Omasa, T., Generation of high-producing cell lines by overexpression of cell division cycle 25 homolog A in Chinese hamster ovary cells. J Biosci Bioeng 2013, 116 (6), 754-60. 79. Lewis, J. M.; Baskaran, R.; Taagepera, S.; Schwartz, M. A.; Wang, J. Y. J., Integrin regulation of c-Abl tyrosine kinase activity and cytoplasmic–nuclear transport. Proceedings of the National Academy of Sciences 1996, 93 (26), 15174-15179. 80. Greuber, E. K.; Smith-Pearson, P.; Wang, J.; Pendergast, A. M., Role of ABL Family Kinases in Cancer: from Leukemia to Solid Tumors. Nature reviews. Cancer 2013, 13 (8), 559-571. 81. Shiba-Ishii, A.; Noguchi, M., Aberrant Stratifin Overexpression Is Regulated by Tumor-Associated CpG Demethylation in Lung Adenocarcinoma. The American Journal of Pathology 2012, 180 (4), 1653-1662. 82. Sarno, J.; Welch, W.; Mok, S.; Garner, E., Stratifin expression in epithelial ovarian cancer: the role of methylation in vitro and in vivo. Cancer Research 2007, 67 (9 Supplement), 1057-1057. Supporting Information Excel table showing protein accession numbers, peptide, and spectra for each sample as well as the results from GO and KEGG (p-values) For TOC only

30

ACS Paragon Plus Environment

Page 30 of 38

Page 31 of 38

Protein Extraction

Sample Preparation Protein Extraction

DG44 cells

Alkylation SDS

CHO-S cells

FASP

Reduction

Ovary

Liver MS/MS

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

Journal of Proteome Research

m/z

Mass Spectrometry CHO Database Search

Fractionation Bioinformatics Analysis

Digestion Applications

Improve bioprocess development

31

ACS Paragon Plus Environment

Journal of Proteome Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

ACS Paragon Plus Environment

Page 32 of 38

Page 33 of 38

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

Journal of Proteome Research

ACS Paragon Plus Environment

A

1 2 3 4 D 5 6 7 8 9 10 11 12 13 G 14 15 16 17 18 19 20 21 J22 23 24 25 26 27 28 29 30 M 31 32 33 34 35 36 37 38

B C Journal of Proteome Research

E

F

H

I

K

L

N

O

ACS Paragon Plus Environment

Page 34 of 38

Page 35 of 38

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

Journal of Proteome Research

ACS Paragon Plus Environment

Journal of Proteome Research

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

p< 0.05 Enrichment

ACS Paragon Plus Environment

Page 36 of 38

p> 0.05 Depletion

Page 37 of 38

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

Journal of Proteome Research

ACS Paragon Plus Environment

Journal of Proteome Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

ACS Paragon Plus Environment

Page 38 of 38