In the biopharmaceutical industry, essentially, there are three components that play the main role in producing biopharmaceutical products, the host cell, the expression vector and the bioreactor and/or production environment. To produce the highly valued and desired products, the choice of a suitable host is one of the most important aspects. The host required is not only required to produce the desired product, but also needs to demonstrate robustness in a bioreactor system. Constantly facing challenges in a bioreactor, cells often undergo apoptosis, a well-known limiting factor in biopharmaceutical production, which ultimately leads to low yield of valuable protein(s). We have genetically engineered a CHO-K1 cell line to constitutively express human insulin-like growth factor-1 (IGF-1) and murine polyoma large T-antigen (PyLT-Ag) to generate Super-CHO and CHO-T respectively, two cell lines that can potentially serve different niches in the biopharmaceutical industry. In the first part of the project, we hypothesised that suspension-adapted Super-CHO and CHO-T cells are both resilient cell lines relative to the suspension-adapted CHO-K1 (designated as CHO-XL-99) when facing nutrient depletion, one of the most common problems in a bioreactor. Furthermore, in the second part of this project, the suspension-adapted CHO cell lines were also tested against a cytotoxic heavy metal, cadmium. Without the protection of the metal-resistance element, metallothionein, both Super-CHO and CHO-T cells were also challenged with cadmium to demonstrate their robustness over the parental cell line, CHO-XL-99. In the subsequent study, this project also focussed on the transfection efficiency of each parental and engineered CHO cell lines. Different strategies have been employed in the past in an attempt to improve productivity in the biopharmaceutical industry, from alterations in vector construction, improved culture condition, down to enhanced product recovery. However, the transfer and expression of the gene-of-interest (GOI) has still proven to be the limiting factor for achieving increased specific productivity. In an effort to improve transfection efficiency, strategies including cell cycle synchronisation and various transfection methods to deliver the GOI into the cells have been employed. Thus, the third part of this project has used synchronising agents in conjunction with commercially available lipid- and polymer-based reagents as delivery vehicle for the model protein, EGFP. The combination of cell synchronisation and transfection vehicle on transfection efficiency is studied here, in addition to their individual or collective effect on cell growth, apoptosis and viability. In summary, this project demonstrates the incidence of apoptosis in the cell culture induced by nutrient depletion and heavy metal, and that the use of transfection reagents solely, or in combination with synchronising agents also correlates with the increase of apoptotic indices in the cell culture. The use of the robust cell lines for transfection is an important aspect, and the balance between cell viability and the effort for augmenting transfection efficiency has to be met in order to achieve the maximum biopharmaceutical yields.

Graduation date: 1997

Thesis (M.Phil.)--Hong Kong University of Science and Technology, 2007. / Includes bibliographical references (leaves 97-105). Also available in electronic version.

Proper progression through the cell cycle is critical for cell growth and survival. Disruption of cell cycle progression can lead to cell cycle arrest and cell death. In addition, uncontrolled cell cycle progression can lead to cancer. Cells have evolved complex mechanisms to regulate each phase of the cell cycle to ensure proper cell cycle progression. In the presence of cellular stress, cells will respond promptly to arrest the cell cycle and allow repair. In order to study this complex process, it is important to identify the complete complement of proteins involved. I took two large-scale approaches to study the cell cycle. First, I down-regulated the majority of essential genes in Saccharomyces cerevisiae, and determined how depletion of individual gene product affected progression through the cell cycle. I determined that over 65% of essential genes I tested are most important at a specific cell cycle phase. In addition, I found that two genes, Smc4 and Cse1, have novel roles in S-phase of the cell cycle. In the second approach, I discovered two anti-proliferative compounds. Both compounds caused cell cycle delay in G1 phase of the cell cycle. Chemical genetic screens in yeast allowed me to determine the pathways most sensitive to each of these two compounds. By studying the response of cells to these compounds, I confirmed that compound 13 causes mitochondrial dysfunction in cells and compound 15 causes nuclear DNA damage. Furthermore, I found that both compounds are toxic in mammalian cells and that the responses that they elicit in mammalian cells are similar to those observed in yeast cells.

The cyclin-dependent kinase (CDK) inhibitors have been demonstrated to be an important component in the regulation of plant cell cycle. Although they share a conserved CDK inhibitory region with a family of CDK inhibitors in mammals, the plant CDK inhibitors are very different from the animal and yeast CDK inhibitors. Thus studies of the plant CDK inhibitors could provide insight on the molecular mechanisms regulating the cell cycle in plants as well as the differences between plants and animals. The research described in this thesis investigated the seven Arabidopsis CDK inhibitors ICKs in terms of transgenic expression, phosphorylation, stability and interactions with other proteins. <p>ICKs were expressed in transgenic Arabidopsis plants as fusion proteins with the green fluorescent protein (GFP). Consistent with the previous studies on ICK1, ICK2 and ICK4, overexpression of all seven ICKs inhibited plant growth and resulted in plants with serrated leaves and flowers with altered morphology. A Survey based on large a number of independent transformants showed that GFP-ICK3 and GFP-ICK4 had weaker phenotypic effects compared to other GFP-ICKs. The Western blotting results showed that all GFP-ICKs were expressed at a low level in general. The levels of GFP-ICK3 and GFP-ICK4 were the lowest, suggesting that the weaker effects for ICK3 and ICK4 may partly be due to low protein levels. Treatments with MG132, an inhibitor of the proteasome, resulted in moderate but clear accumulation of fusion proteins for ICK1, ICK5, ICK6 and ICK7 in plants, suggesting that the proteasome is involved in the degradation of these proteins. <p>To study the state of protein phosphorylation, the proteins extracted from the plants were treated with calf intestinal phosphatase (CIP). The CIP treatment caused a faster migration of the GFP fusion protein for ICK1, ICK2, ICK5, ICK6 and ICK7, while the effect was not observed for control GFP and other non-specific proteins, indicating that these proteins can be phosphorylated in plants. The shift also differed among ICKs. Interestingly, dephosphorylation of ICK7 might have rendered it less stable. The protein pulldown experiments using p13Suc1-conjugated agarose beads showed that GFP-ICK4, GFP-ICK5 and GFP-ICK6 could associate with the CDK complex, similar to what has been shown for ICK1 and ICK2. CIP treatments of the p13Suc1 affinity-purified proteins also showed that ICK1, ICK2, ICK5 and ICK6 associated with the CDK complex were phosphorylated. <p>Attempts were also made to isolate peptide aptamers that are able to interact with ICKs for the purpose for expressing such an aptamer in plants. However, an aptamer that has a strong ability to interact with ICKs in two of yeast two-hybrid systems was not identified. In addition, the analysis of Arabidopsis CYCD3;1 for its interaction with ICK1 using a series of deletion mutants showed that the removal of both the N-terminal and C-terminal regions of CYCD3;1 greatly reduced or abolished the interaction with ICK1. <p> In summary, transgenic Arabidopsis plants have been obtained for expressing each of the seven Arabidopsis CDK inhibitors fused to GFP. The results confirmed and extended previous finding that overexpression of a CDK inhibitor inhibits plant growth as well as changes plant morphology. The observation that the ICK fusion proteins were generally at low and often undetectable levels, in comparison to much higher levels of the GFP protein, suggests that ICKs are unstable in the cell. Results from the MG132 experiments indicate that the 26S proteasome may play a role in the degradation of ICK1, ICK5, ICK6 and ICK7. Results from CIP treatments further show that most ICKs, particularly ICK1, ICK2, ICK5, ICK6 and ICK7, can be phosphorylated in vivo. Interestingly, ICK7 stability may depend on the status of protein phosphorylation. This study provides new understanding on how the family of proteins is regulated at the post-transcriptional level and the differences among Arabidopsis CDK inhibitors.

Proper progression through the cell cycle is critical for cell growth and survival. Disruption of cell cycle progression can lead to cell cycle arrest and cell death. In addition, uncontrolled cell cycle progression can lead to cancer. Cells have evolved complex mechanisms to regulate each phase of the cell cycle to ensure proper cell cycle progression. In the presence of cellular stress, cells will respond promptly to arrest the cell cycle and allow repair. In order to study this complex process, it is important to identify the complete complement of proteins involved. I took two large-scale approaches to study the cell cycle. First, I down-regulated the majority of essential genes in Saccharomyces cerevisiae, and determined how depletion of individual gene product affected progression through the cell cycle. I determined that over 65% of essential genes I tested are most important at a specific cell cycle phase. In addition, I found that two genes, Smc4 and Cse1, have novel roles in S-phase of the cell cycle. In the second approach, I discovered two anti-proliferative compounds. Both compounds caused cell cycle delay in G1 phase of the cell cycle. Chemical genetic screens in yeast allowed me to determine the pathways most sensitive to each of these two compounds. By studying the response of cells to these compounds, I confirmed that compound 13 causes mitochondrial dysfunction in cells and compound 15 causes nuclear DNA damage. Furthermore, I found that both compounds are toxic in mammalian cells and that the responses that they elicit in mammalian cells are similar to those observed in yeast cells.

Chk2 is a cell cycle checkpoint kinase that is essential for initiating the DNA damage response in the presence of genetic damage. Its role is highly conserved from budding yeast (where it is named Rad53) to humans. Very few cell cycle checkpoint proteins have ever been studied in fish and the role of Chk2 has never been characterized. Oncorhynchus mykiss (Rainbow trout) was chosen for this project due to its importance in the commercial aquaculture industry and the availability of rainbow trout cell cultures at the University of Waterloo. This study was the first to clone the CHK2 gene in a teleost species, verified through both genomic and cDNA cloning. A section of the CHK2 gene, specifically the forkhead associated domain (FHA), was used to express recombinant Chk2 protein and generate polyclonal anti-Chk2 antibodies. A southern blot was performed and CHK2 was found to exist as a single copy number in the rainbow trout genome. The tissue specificity of Chk2 was also examined both at the mRNA transcript and protein level. Interesting tissue specific differences were discovered with transcript levels moderately low in gill and higher in brain, while protein levels were extremely high in gill and lower in brain tissues. Protein levels were verified in both whole fish tissue samples and in cell culture suggesting that cell cultures accurately reflect the state of checkpoint proteins in vivo. These tissue specific differences suggest that in gill, Chk2 is maintained at a high protein level to combat any toxins in the water attempting to transverse this barrier tissue and gain access to the fish’s circulatory system. Meanwhile, the blood brain barrier offers protection to the highly sensitive brain tissue, suggesting that high levels of Chk2 protein are not constitutively required, but instead remain in a transcript reservoir able to be quickly translated in the event of DNA damage. To determine whether Chk2’s checkpoint role is conserved in O. mykiss, both gill and brain cell cultures were treated with low and high doses of bleocin (a commercially available form of bleomycin) known to cause high levels of double-strand breaks, the most deleterious type of DNA damage and a specific activator of the Chk2 DNA damage response (DSB). Results showed that bleocin had no effect on levels of Chk2 in gill cells, confirming that the protein is constitutively active in this tissue always on alert against potential genetic insult. In contrast, brain cells were able to upregulate Chk2 in a dose-dependent manner to bleocin induced DNA damage demonstrating that Chk2 can act as a biomarker for genetic damage in brain cells. In conclusion, the tissue specific expression of Chk2 and its ability to respond to DNA damage suggests that checkpoint proteins may serve as suitable biomarkers for DNA damage in O. mykiss and other fish species.

Chk2 is a cell cycle checkpoint kinase that is essential for initiating the DNA damage response in the presence of genetic damage. Its role is highly conserved from budding yeast (where it is named Rad53) to humans. Very few cell cycle checkpoint proteins have ever been studied in fish and the role of Chk2 has never been characterized. Oncorhynchus mykiss (Rainbow trout) was chosen for this project due to its importance in the commercial aquaculture industry and the availability of rainbow trout cell cultures at the University of Waterloo. This study was the first to clone the CHK2 gene in a teleost species, verified through both genomic and cDNA cloning. A section of the CHK2 gene, specifically the forkhead associated domain (FHA), was used to express recombinant Chk2 protein and generate polyclonal anti-Chk2 antibodies. A southern blot was performed and CHK2 was found to exist as a single copy number in the rainbow trout genome. The tissue specificity of Chk2 was also examined both at the mRNA transcript and protein level. Interesting tissue specific differences were discovered with transcript levels moderately low in gill and higher in brain, while protein levels were extremely high in gill and lower in brain tissues. Protein levels were verified in both whole fish tissue samples and in cell culture suggesting that cell cultures accurately reflect the state of checkpoint proteins in vivo. These tissue specific differences suggest that in gill, Chk2 is maintained at a high protein level to combat any toxins in the water attempting to transverse this barrier tissue and gain access to the fish’s circulatory system. Meanwhile, the blood brain barrier offers protection to the highly sensitive brain tissue, suggesting that high levels of Chk2 protein are not constitutively required, but instead remain in a transcript reservoir able to be quickly translated in the event of DNA damage. To determine whether Chk2’s checkpoint role is conserved in O. mykiss, both gill and brain cell cultures were treated with low and high doses of bleocin (a commercially available form of bleomycin) known to cause high levels of double-strand breaks, the most deleterious type of DNA damage and a specific activator of the Chk2 DNA damage response (DSB). Results showed that bleocin had no effect on levels of Chk2 in gill cells, confirming that the protein is constitutively active in this tissue always on alert against potential genetic insult. In contrast, brain cells were able to upregulate Chk2 in a dose-dependent manner to bleocin induced DNA damage demonstrating that Chk2 can act as a biomarker for genetic damage in brain cells. In conclusion, the tissue specific expression of Chk2 and its ability to respond to DNA damage suggests that checkpoint proteins may serve as suitable biomarkers for DNA damage in O. mykiss and other fish species.

The cyclin-dependent kinase (CDK) inhibitors have been demonstrated to be an important component in the regulation of plant cell cycle. Although they share a conserved CDK inhibitory region with a family of CDK inhibitors in mammals, the plant CDK inhibitors are very different from the animal and yeast CDK inhibitors. Thus studies of the plant CDK inhibitors could provide insight on the molecular mechanisms regulating the cell cycle in plants as well as the differences between plants and animals. The research described in this thesis investigated the seven Arabidopsis CDK inhibitors ICKs in terms of transgenic expression, phosphorylation, stability and interactions with other proteins. <p>ICKs were expressed in transgenic Arabidopsis plants as fusion proteins with the green fluorescent protein (GFP). Consistent with the previous studies on ICK1, ICK2 and ICK4, overexpression of all seven ICKs inhibited plant growth and resulted in plants with serrated leaves and flowers with altered morphology. A Survey based on large a number of independent transformants showed that GFP-ICK3 and GFP-ICK4 had weaker phenotypic effects compared to other GFP-ICKs. The Western blotting results showed that all GFP-ICKs were expressed at a low level in general. The levels of GFP-ICK3 and GFP-ICK4 were the lowest, suggesting that the weaker effects for ICK3 and ICK4 may partly be due to low protein levels. Treatments with MG132, an inhibitor of the proteasome, resulted in moderate but clear accumulation of fusion proteins for ICK1, ICK5, ICK6 and ICK7 in plants, suggesting that the proteasome is involved in the degradation of these proteins. <p>To study the state of protein phosphorylation, the proteins extracted from the plants were treated with calf intestinal phosphatase (CIP). The CIP treatment caused a faster migration of the GFP fusion protein for ICK1, ICK2, ICK5, ICK6 and ICK7, while the effect was not observed for control GFP and other non-specific proteins, indicating that these proteins can be phosphorylated in plants. The shift also differed among ICKs. Interestingly, dephosphorylation of ICK7 might have rendered it less stable. The protein pulldown experiments using p13Suc1-conjugated agarose beads showed that GFP-ICK4, GFP-ICK5 and GFP-ICK6 could associate with the CDK complex, similar to what has been shown for ICK1 and ICK2. CIP treatments of the p13Suc1 affinity-purified proteins also showed that ICK1, ICK2, ICK5 and ICK6 associated with the CDK complex were phosphorylated. <p>Attempts were also made to isolate peptide aptamers that are able to interact with ICKs for the purpose for expressing such an aptamer in plants. However, an aptamer that has a strong ability to interact with ICKs in two of yeast two-hybrid systems was not identified. In addition, the analysis of Arabidopsis CYCD3;1 for its interaction with ICK1 using a series of deletion mutants showed that the removal of both the N-terminal and C-terminal regions of CYCD3;1 greatly reduced or abolished the interaction with ICK1. <p> In summary, transgenic Arabidopsis plants have been obtained for expressing each of the seven Arabidopsis CDK inhibitors fused to GFP. The results confirmed and extended previous finding that overexpression of a CDK inhibitor inhibits plant growth as well as changes plant morphology. The observation that the ICK fusion proteins were generally at low and often undetectable levels, in comparison to much higher levels of the GFP protein, suggests that ICKs are unstable in the cell. Results from the MG132 experiments indicate that the 26S proteasome may play a role in the degradation of ICK1, ICK5, ICK6 and ICK7. Results from CIP treatments further show that most ICKs, particularly ICK1, ICK2, ICK5, ICK6 and ICK7, can be phosphorylated in vivo. Interestingly, ICK7 stability may depend on the status of protein phosphorylation. This study provides new understanding on how the family of proteins is regulated at the post-transcriptional level and the differences among Arabidopsis CDK inhibitors.

Hepatocellular carcinoma (HCC) or hepatoma is the top one cause of death in Taiwan based on the Cause of Death Statistics from the Department of Health, Executive Yuan, Taiwan, for many years. To identify any reliable HCC markers and further applied with the AFP measurement to improve the early diagnosis of HCCs is the most important thing. A high expression level of S-phase protein kinase associated protein 2 (SKP2) protein and its cofactor CDC28 protein kinase regulatory subunit 1B (CKS1B) involved in ubiquitination of some cyclin-dependent kinase (Cdk) inhibitors has been reported in various carcinoma. In this study, we examined the expression of CKS1B in HCC tissues and cell lines, and tested the cell cycle effect caused by specific small interference RNA (siRNA) of CKS1B in SK-hep1 cell line. Up-regulated CKS1B mRNAs in HCC cell lines and tissues were identified in our study, when comparing to the normal liver tissues. But we also found lack of up-regulated CKS1B proteins in our HCC tissues at the same time, indicated that CKS1B proteins might be unstable in HCCs. Down regulation of the Cdk inhibitors p27 was only partially associated with HCCs, and the expressions of CKS1B and p27 were not correlated to each other in HCCs, suggesting other pathway(s) might involve in the regulation(s) of CKS1B and p27 proteins in the HCCs. Down-regulation of the p21 proteins was also found to be not significantly associated with HCCs tissues, this result strongly suggested a post-translational stabilization way might regulate(s) the p21 protein levels in HCCs tissues. On the other hands, in time course experiment, disruption of CKS1B mRNA by si-CKS1B up-regulated the expressions of p27 and SKP2 protein levels and down-regulated the p21 protein level in the SK-hep1 hepatoma cell lines for 24 hrs later. But the mRNA expression level of p21 and p27 were actually both up-regulated for 48 hrs after transfected with si-CKS1B. We also tested the mRNA expression level of many cell cycle regulatory factors for 48 hrs after transfected with si-CKS1B. The results exhibited almost all of the factors (excepted p21, p27 and Cyclin D2) were down-regulated. Furthermore, we saw the apoptosis appearance of SK-hep1 cell after transfected with si-CKS1B for 48 hrs, suggesting the abnormal cell proliferation and tumorigenesis were controlled by siRNA transfection. Taken together, these results suggest that SCFSKP2-CKS1B pathway might not direct involved in ubiquitination of Cdk inhibitors. Another pathway(s), either known or novel, in addition to APC/CCHD1 (G0-G1 phase) and SCFSKP2-CKS1B (G1-S phase) regulation pathways, might regulate the tumorigenesis of HCCs.