Sudhin Biopharma Co.
Developing the most efficient continuous bioprocessing platform technologies

Original Inclined Lamellar Settlers not best sized for smaller yeast Pichia pastoris cells

We have tested an original inclined lamellar settler, CS-10, designed for producing clarified liquid from mammalian cell culture at a harvest flow rate of 10 liters/day, on a perfusion bioreactor culture of yeast Pichia cells.

Our results shown on the right (Figure 1) indicate that there is a significant loss of yeast cells at a perfusion rate of 600 ml/day, but the high cell density perfusion bioreactor can be operated at perfusion rate of 400 ml/day.

After 425 hours, disturbances were caused by a power outage and an uncontrolled methanol accumulation for inducing heterologous protein expression. Thus the lamellar settler CS-10, designed for harvesting about 10 liters/day of clarified harvest from mammalian cells, is not effective in getting clarified harvest from yeast cells at perfusion flow rates larger than 500 ml/day.

Figure 1

Selective removal of cell debris from Inclined settlers shown also for yeast cells

Figure 2 shows a photograph of centrifuged samples from the settler top outlet and bioreactor, with a layer of cell debris floating in the settler top outlet sample, as methanol induction has resulted in some dead cells and cell debris due to excess methanol.

The bioreactor sample taken at the same time shows no such top layer of cell debris, indicating high viability inside the bioreactor, as dead cells and cell debris are selectively removed in the settler top effluent.

Thus unique feature of inclined settlers to remove any dead cells selectively from the bioreactor to maintain a high viability inside the bioreactor is now proven for the smaller yeast Pichia pastoris cells as well, just like this feature is repeated demonstrated for the larger mammalian cells.

Figure 2

High Cell Densities at increasing Perfusion rates from our novel Compact Cell Settlers

We have designed several prototypes of compact cell settler in a cylindrical and conical geometry. Figure 3 shows the high cell densities (OD600 > 800) of yeast cells in a 5-liter perfusion bioreactor attached with an early prototype of our Compact Cell Settler with a footprint of 12” diameter.

We see a slight improvement in the perfusion rate of relatively clarified culture broth in settler effluent upto 1000 mL/day, compared to 400 mL/day from the CS-10 of the same footprint.

At higher perfusion flow rates of about 2000 mL/day, we can see the bioreactor cell densities falling suggesting that this early prototype design is not sized appropriately for any faster perfusion rates for the smaller yeast cells.

Figure 3

Selective removal of smaller yeast cells in the top effluent of compact cell settler

Figure 4 shows the cell size distribution of samples from the compact cell settler’s top effluent and the bioreactor during the perfusion culture shown in Figure 3.

These results confirm again the main feature of inclined settlers to remove selectively the smaller cells in the compact cell settler’s top effluent compared to the size of cells entering the settler from the bioreactor.

There are no cell debris found in this settler effluent because the bioreactor was not induced with any toxic inducer like methanol.

If there were any cell debris, which will be smaller than dead cells, such smaller subcellular particles will be removed even more selectively from the bioreactor into the settler’s top effluent, to maintain very low cell debris accumulation in the perfusion bioreactor.

Figure 4

More inclined surfaces inside the cylindrical and conical settler

In our next prototype, we have changed the internal design and packed more inclined settlers through much of the cylindrical portion of the settler.

With this novel design, we were successful in achieving high cell densities of yeast cells at a much higher perfusion rate of 3000 mL/day, as shown in the Figure 5.

The perfusion bioreactor has achieved OD600 > 700 quickly after the perfusion began and gradually increased to about 1000 OD over a month long perfusion operation, through several process changes such as switching the carbon source from glycerol to glucose and pH from 4 to 5, and maintained very high OD at a steady state operation.

While there are some cells (typically smaller and/or dead cells, as shown above in Figures 2 and 4) leaving in the settler effluent at an OD about 200, the gradual increase in bioreactor OD to almost 1000 is due to the continuous recycle of large live yeast cells back to the bioreactor.

Figure 5

No Protein Sieving, but Additional Protein Production in the Compact Cell Settler

These Pichia pastoris cells were engineered to overexpress and secrete a heterologous protein product from a constitutive GAP promoter. For the perfusion bioreactor culture shown in Figure 5, the secreted protein concentrations shown in Figure 6 demonstrate higher titer in the settler effluent than that in the bioreactor for most of perfusion operation.

This higher protein concentration suggests that there is no protein sieving which is common with membrane-based cell retention device such as ATF and TFF.

Instead, these higher total protein titers in the settler effluent suggest additional protein production in the Compact Cell Settler. These results are highly reproduced in all our perfusion culture operations so far, including the next perfusion bioreactor operation shown in Figures 7 and 8.

Figure 6

Perfusion Bioreactor culture of yeast cells with a new Compact Cell Settler prototype

We have increased the inclined settler surfaces enormously in the latest design of our novel Compact Cell Settler with the same footprint as a CS-10.

Consequently, we are able to achieve clarified harvest at much higher perfusion flow rates exceeding 5 liters/day from a 5-liter perfusion bioreactor growing yeast P. pastoris cells, crossing the typical perfusion rate of 1 volume exchange per day in mammalian cell cultures.

Shown in Figure 7, the bioreactor OD600 remains above 700 for over 2 months. Small fluctuations in bioreactor OD from 200 hours to 1000 hours are due to our manipulations of the perfusion rate and feed glycerol concentration. Settler top effluent or harvest is very clarified with its OD600 mostly between 10 and 30, representing less than 5% of the yeast cells in the bioreactor.

This same size settler (12” diameter, shown in bottom photo of previous page) is predicted to be capable of achieving more than 250 liters/day of clarified harvest from mammalian cells, based on the larger size of CHO cells.

Figure 7

Higher Product Harvested from Perfusion Bioreactor compared to Repeated Fed-batch cultures

During the initial fed-batch operation before the perfusion is turned on, the secreted protein product accumulates in the bioreactor to a titer of >1.5 g/L. After perfusion begins, the bioreactor protein titer gets reduced due to its continuous harvest into the effluent and dilution by fresh medium.

The protein titers remain relatively constant from 200 hours to 1000 hours, while the perfusion rate is roughly around 2 liters/day. As the perfusion rate is increased gradually over the two months of perfusion bioreactor operation, particularly after 1000 hours, the protein titers decrease slightly due to the increased harvest and dilution rates.

However, the rate of protein accumulated in the harvest tank (product of perfusion rate and protein titer) increases faster at higher perfusion rates compared to its earlier values at lower perfusion rates. The accumulated protein harvest over the two-month long perfusion operation is 160 g from this 5 liter bioreactor operated at gradually increasing perfusion rates.

This amount is over 4x higher than what can be harvested in the clarified supernatant from about 8 repeated fed-batch culture operations from the same 5 liter bioreactor, during the same two month period.

Figure 8