Nutrient Limitations and Feeding Strategies

Many literature studies have investigated glycosylation variations of recombinant proteins with respect to nutrient requirements including sugar feeding, nucleotide sugar feeding, oxygen levels, amino acid additions, and serum components (3,7278). Serum is known to contain growth factors, which improve cell growth, and lipids, which improve shear resistance, but serum also contains waste products and proteases, which can be detrimental to the cell and glycoprotein products. Gaw-litzek et al. (74) and Hayter et al. (76,77) both used glucose-limited chemostats to demonstrate that glycosylation was dependent on media components. Hayter et al. (76) examined the glycosylation pattern of interferon-g (IFN-g) produced by CHO cells at a constant dilution rate and at two different glucose concentrations. They were able to demonstrate that fully glycosylated IFN-g occurred more readily when glucose was not limited. The authors concluded that this effect was to be due to the physiological state of the cells.

Hahn and Goochee (67) observed that transferrin, a glycoprotein, secreted by confluent and subconfluent cultures, contains different proportions of biantennary oligosaccharides, where the biantennary glycoprotein is more biologically active. Confluent cells produced more active transferrin than the subconfluent cultures. Hahn and Goochee (67) concluded that oligosaccharide synthesis is growth dependent, thus in standard batch cultures where the growth rate varies throughout the fermentation, the glycosylation pattern will vary.

Hooker et al. (70) examined the N-oligosaccharides of recombinant IFN-g produced by CHO cells, where temperature, pH, and DO were held constant. The proportion of the biantennary structure decreased throughout the fermentation and the proportion of oligomannose and truncated species increased for both

N-glycosylation sites. This observation is very important for recombinant therapeutics, since the shift to oligomannose and truncated species typically increases the clearance rate, thus reducing the efficacy of the therapeutic. It was also suggested by the authors, that fed-batch fermentations might mitigate the deterioration of the glycoprotein (70). In a separate study, an MAb fed-batch fermentation, the MAb also had a decrease in the amount of complex glycoforms and an increase in the high-mannose and truncated glycoforms throughout the fermentation (69), which can detect and quantify microheterogeneity changes. The problem of high-mannose and truncated glycoforms is potentially a problem for any material produced in bioreactors by mammalian cells.

Gawlitzek et al. (74) examined the production of interleukin-2 (IL-2) under various nutrient limitations [glucose, fetal calf serum (FCS), glutamine, serine, aspartate, tryptophan, and oxygen]. They found the oligosaccharide profiles to be identical, regardless of the culture conditions. However, they also observed that the glycosylated to unglycosylated ratio varied depending on the culture conditions. Gawlitzek et al. (74) observed truncated IL-2, which is an indicator of proteolysis of the IL-2. Their experiments are critical to furthering the understanding of how glycosylation is influenced by culture conditions; however, none of the parameters tested were independent due to the culturing method. Each nutrient limitation lasted only 3 days and was immediately replaced by another nutrient limitation. Since cells need time to recover from stress, it was possible that the similar responses observed for oligosaccharide content were not steady-state responses to the nutrient limitations (stress), but that the effects of the early nutrient stress were still being detected at later time points. Gawlitzek et al. (75) investigated the role of the influence of FCS on glycosylation for suspension and microcarrier BHK-21 cultivated cells. They found that N-glycosylation and O-glycosylation of IL-2 were greater in the serumfree environment. Additionally, due to the recent outbreaks of mad cow's disease in Europe, numerous industrial cell culture-derived therapeutics, under clinical trial, are using serum- and protein-free media.

Jenkins et al. (3) varied the media lipid content for a batch CHO cell culture expressing IFN-g to investigate the importance of lipids on N-glycosylation. Lipoprotein supplementation increased the amount of fully glycosylated IFN-g compared to the control culture (3). Interestingly, if lipids are critical for N-glycosylation, then the cell culture conditions that alter the lipid content of cell membranes could also effect glycosylation, e.g., agitation, pH, and temperature. Castro et al. (78) stressed the importance of media composition after they observed that batch feeding did not prevent the decrease of fully glycosylated IFN-g; however, the initial glutamine concentration did impact the glycosylation pattern. The authors also observed that the addition of serum to the media caused proteolysis of IFN-g. Last, they found that the supplemental protein added to the media and the lipid composition of the media affected the glycosylation pattern (78).

Barnabe and Butler (79) investigated dual glucose and galactose feeding in hybridoma cells. It was observed that the corresponding intracellular nucleotide pool increased; however, no changes in glycosylation were observed. Hills et al. (63) fed NS0 cells expressing an IgG galactose, glucosamine, or CMP-sialic acid. The glucosamine feeding resulted in a 17-fold increase in the UDP-N-acetylhexose pools; however, the UDP-hexose pools decreased by 33%. The UDP-Glc to UDP-Gal ratio remained constant at 4:1. A 57% reduction in galactosylation was observed. The core Mans structure was unaffected. Galactose feeding resulted in a fivefold increase in the UDP-Gal pool, but no increase in b1,4-galactosylation. The CMP-sialic acid feeding did not result in increased sialylation, but it did result in a 44-fold increase in the CMP-sialic acid pool. Gu and Wang (80) fed N-acetylmannosamine to CHO cells expressing IFN-g; however, no change in sialylation was observed. Gawlitzek et al. (81) observed that glucosamine feeding increased the UDP-GxxNAc pool (sum of the UDP-GlcNAc and UDP-GalNAc pools) for CHO producing TNFR-IgG (an immunoadhesin tumor necrosis factor-IgG fusion). However, the glucosamine feeding did not alter the glycosylation (81).

Baker et al. (53) demonstrated that the addition of glucosamine, uridine, and N-acetylmannosamine (ManNAc) significantly increased the intracellular UDP-HexNAc and CMP-sialic acid concentrations for both NS0 and CHO cells. However, these increased nucleotide sugar pools did not increase the level of sialylation. These additions did, however, improve the NeuAc to NeuGc ratio in NS0 cells to a more favorable (human) level. Altamirano et al. (72,73,82) examined the dual feeding of sugars to CHO cells expressing tPA to control growth, substrate consumption, and waster product buildup. Fructose and glucose feeding resulted in decreased growth rates, the substrate utilization, and toxic waste buildup.

Butyrate feeding has been observed to increase type I tPA production (83), which is less effective that type II tPA. Kim and Lee noted increased glycoprotein production in the presence of sodium butyrate, but also observed cell growth inhibition and increased apoptosis. Santell et al. (84) suggested that sodium butyrate increased glycosylation by increasing glycoprotein reutilization. Wang et al. (85) examined CHO cells expressing EPO in fluidized beds. They observed that the addition of glucose and sodium butyrate doubled the final EPO concentration without changing the glycosylation pattern.

Cruz et al. (86) produced a recombinant antibody-IL-2 fusion in BHK cells using protein-free media. Two different protein-free media were compared. Chemo-stat cultures were used and consistent productivity and glycosylation patterns were obtained. They also demonstrated that long-term cultivation is not adverse to product consistency. The higher titers obtained from the chemostats also increased the yield in the purification steps (86).

How To Boost Your Metabolism

How To Boost Your Metabolism

In The Next 45 Minutes You're Going To Discover How To Boost Your Metabolism And Lose Weight. Who Else Wants To Boost Their Metabolism And Shed Pounds Fast?

Get My Free Ebook


Post a comment