# Improvements

## Bioprocess

## Bioprocess diagram

In this section we are to present a whole DIY-Bioprocess consisting of a DIY-pump continuous pump, two bioreactors (green and blue in the Figure 1), one additional reactor (orange), a DIY-PLA-Extruder and a DIY-roller for final storage.

If we look at Figure 1, we can observe the green bioreactor (50 mL) is for cell maintenance and metabolite accumulation of the precursor metabolite Lactyl CoA with the LDH gene with an inducer A.

The blue bioreactor (e.g. 50 mL) receives the stream from the green one and is where the PhaC and Pct genes are expressed with an inducer B in order to transform to Lactyl-CoA and ultimately to Poly-Lactic Acid (PLA). This last stream is homogenized (cell disruption) before entering the orange bioreactor.

In the orange reactor, cells containing Low Molecular Weight PLA are releasing the polymer, together with active enzymes enabling further extension of the Polymeric chain. This orange reactor is of higher volum (e.g. 1000 mL) and is runned in a fed batch fashion, being emptied every 4 days and being refilled in between with a steady flow in order to facilitate further polymer extension by alternative methods (i.e. testing enzymatic polymerization and trying to adapt traditional chemical methods).

Finally the last stream can be collected, the PLA within the cell broth can be extracted and passed by an extruder where a melting process will occur and finally the raw material can be stored in a roller at by lowering back the temperature.

__Figure 1.__ A whole bioprocess diagram of the bioprocess developed to ease the engineering of *P. putida* KT2440 for bioproduction of PLA by optimizing each process step by step and in a continuous fashion, where downstream processing is as well considered. The green bioreactor is for cell cell growth and accumulation of metabolite with the LDH induced. PLA = Poly-Lactic Acid; LMW = Low Molecular Weight; HMW = High Molecular Weigh; LDH = Lactate dehydrogenase; PhaC = PHB&PLA polymerase; PCT: Lactyl-CoA reductase.

## Description of a whole made DIY-Bioprocess

The following Figure 2 shows the whole made Do-It-Yourself (DIY)-Bioprocess for the metabolic engineering of *Pseudomonas putida* specially adapted for the bioprocessing of PLA.

Click to enlarge

__ Figure 2.__ A whole-DIY-bioprocess for a metabolic engineered

*Pseudomonas putida*KT2440 for bioproduction of Poly-Lactic Acid.

From right to left, there is a drip chamber that provides the constant height (H), necessary to act with a constant value of pressure (P) control to finally deliver a constant flow (F).

The drip chamber ensures that for each drop that goes out within the Reactor 3, one drop will be drawn into the drip chamber from an upper reservoir (not shown, see Figure 6). This allows a constant flow (see violet liquid in the bioprocess) that is going to be regulated by means of the Flow regulators 1 and 2, without the need to calculate exact values of height (H) (see section Testing flows), just by providing more-than-enough potential energy and using flow regulators (see Figure 2 and Figure 6).

Once a steady flow is obtained at the values of interest, the cells grown in the Bioreactor 1 and 2 can be grown in steady state, thus simplifying variant parameters for the engineered strain. The outflow out from the Bioreactor 2 will go straight to the Reactor 3, where a cell lysis agent can be provided and make this Reactor 3 as a vessel for enzymatic extension, although other chemical methods for extension of the polymeric chain can be considered as well here. According to the flow of interest, this Reactor 3 is not run in continuous, but is run in a fed-batch fashion. Here, the time required to fill this vessel, according to the flows stated in the Table 3 is around 4 days. This way, reactions within this vessel can be held during this time before further clean up and processing.

Finally, if successful, the plastic product could be purified and it could be melted and extruded in the DIY-PLA-extruder designed for such purpose, with the help of a heater. This heater should be used at a specific distance from the extruder to ensure that the melting temperatures are optimal. At the end, by using the piston to push the melted material in the extruder, a filament of raw plastic should be able to be ejected from the extruder and stored in the roller (see Figure 4).

The DIY-PLA-extrusion system was made of inox steel in order to facilitate the melting of the bioprocuced PLA up to ~178º which is the typical Injection Molding Temperature of PLA, according to Table 1.

__Table 1.__ Poly-Lactic acid chart

Technical Name |
Polylactic Acid (PLA) |

Chemical Formula |
(C3H6O3)n |

Melt Temperature |
PLLA: 157 - 170 °C (315 - 338 °F) ^{1} |

Typical Injection Molding Temperature |
PLLA: 178 - 240 °C (353 - 464 °F) ^{2} |

Heat Deflection Temperature (HDT) |
49 - 52 °C at 0.46 MPa (66 PSI) ^{4} |

Tensile Strength |
PLLA: 61 - 66 MPa (8840 - 9500 PSI) ^{3} |

Flexural Strength |
PLLA: 48 - 110 MPa (6,950 - 16,000 PSI) ^{3} |

Specific Gravity |
PLLA: 1.24 ^{5} |

Shrink Rate |
PLLA: 0.37 - 0.41% (0.0037 - 0.0041 in/in) ^{6} |

- At standard state (at 25 °C (77 °F), 100 kPa)
- Source data
- Source data (Using ASTM D638 Test Method at 22.78°C)
- Source data (Using ASTM D648 Test Method at 22.78°C)
- Source data (Using ASTM D792 Test Method at 22.78°C)
- Source data (Using ASTM D955 Test Method at 22.78°C)

Once the plastic is extruded, it is possible to roll it over the DIY-roller as well to be able to solidify the melted PLA by decreasing the temperature below the melting point of PLA, providing an easy manner to store the final product, see Figure 3 and Figure 4.

__Figure 3.__ Scheme of the design of the extrusion system

__Figure 4.__ A DIY-PLA-extruder made of inox steel. The device is conformed by an extruder where heating enables the PLA to be melt and finally extruded throught a 1.5 mm whole by means of an applied pressure. Once the plastic material is extruded, it can be rolled over the roller.

## Why a chemostat?

We chose chemostats for the bioprocess parts where engineered cells are involved because of the process being easier to describe and thus, making simple the use of the platform for metabolic engineering optimization purposes.

Chemostats are a type of reactor that maintains constant volume of the tank through inlet and outlet flow equivalents. In this type of bioreactor, once steady state has been reached, the *specific growth rate* (µ) of microorganisms can be controlled, as demonstrated in the following mass balance of eq.1, by varying the speed with which substrate enters the tank.

The steady flow (F) allows a system of a fixed volume (V) to achieve an steady state where accumulation or leakage of biomass in the system over time is null, that is dM/dt = 0; thus allowing cells to grow at a fixed specific growth rate (μ) for the achievement of a fixed value of biomass concentration (x) over time (t) that will be regulated according to the dilution time (D = F/V), following the behaviour presented in the biomass balance described in eq.2.

As explained above, in a continuous system, if steady state, dM/dt= 0; then,

Together with the consideration of a sterile supply of LB media from the bag, and considering no bacterial cell growth in the tubes that go from the *chemostat* to the *cellstat*. Then no cells added, x_{i} = 0; and,

The equations (eq. 2, and eq.3) designed to explain the behaviour in the *chemostat*, suggest that the cell densities (x), that can be plotted by using OD_{600}, are not going to vary over time (i.e. due to constant *dilution time, D*, overcome by the *specific growth rate (μ)*).

The main advantage of using a this step-wised continuous system for the bioproduction of PLA is that, specially for metabolic engineering applications, this definitely makes the re-design of the engineered *Pseudomonas putida* KT2440 dependent on less parameters than if with another mode of operation the cell was tested, thus easing the tunage towards the total optimization of the cell in order to whether maximise cell growth and precursor accumulation or bioproduction and cell lysis due to the steady state conditions, and due to the separation of the bioprocesses in different steps.

Here, based on experimental data, a model for the continuous bioprocess that concerns the *chemostat*s 1 and 2 were considered.

Cells supplied from the *chemostat* 1 to the *chemostat* 2 are maintained at steady cell densities. In this *chemostat* 2, the cells and the inducer are mixed (i.e. with cyclohexanone al together with IPTG, inducer flows considered negligible) and pumped together with the supplied fresh cells growing at an that will be optimised to be the of the *Pseudomonas putida* grown in the most optimized carbon source tested (see growth rate in glucose in Figure 5).

## How do we prepare it all? Getting ready, growth rates.

The values of *maximum growth rate (µ _{max})* for each cell (Figure 2 and Table 1) where obtained from Tecan Infinite® M200 (Tecan Trading AG, Switzerland) plate reader to measure optical density (OD

_{600}), see Figure 2 and then convert to µ

_{max}values by using the following relationship (eq.13) [1]

__Figure 5.__ OD_{600} values for *Pseudomonas putida* grown with different substrates like Glucose, Glycerol and Benzoic Acid. The range period of maximum growth rate were data has been extracted (µ = µ_{max}) is marked in yellow.

Thus, considering the values above, the following µ_{max} can be plotted for each substrate, respectively, according to the following Table 2.

__Table 2.__ Maximum growth rate µ_{max} for *Pseudomonas putida* according with the each of the substrates tested successfully, Glycerol and Glucose.

Pseudomonas putida in Glycerol |
0.206 h^{-1} = 0.0034min^{-1} |

Pseudomonas putida in Glucose |
0.212 h^{-1} = 0.0035min^{-1} |

Therefore, it is expected that the maximum flow rate F in a volume V of 50 mL, the in the *chemostat*, taking into account the description contained between eq.1 and eq.3 in steady state:

The flow has to be lower than the correspondent to the maximum cell growth rate in order to avoid wash-out of the cells in the bioreactors. Optimal values for flow rates to maximise biorpoduction should still be calculated, but due to de lack of data for the bioproduction of PLA this has not yet been considered.

The following Table 3 contains the real operational values of the *chemostat* and the cellstat, respectively.

__Table 3.__ Parameters of the bioreactor

Parameter | chemostat (1 - LDH) |
chemostat (2 - PCT & PhaC) |
Fed-batch (Chemical / Enzymatic) |
---|---|---|---|

Flow (F) (ml/min) |
0.17 | 0.17 | |

Volume (V) (ml) |
50 | 50 | 1000 |

Dilution time (D) (min-1) |
0.0034 | 0.0034 | |

Max growth rate* (µ_{max})(min ^{-1}) |
0.0035 | 0.0035 | |

HRT** (1/D) (min) |
294 | 294 | |

Recovery time (days) |
Continuous | Continuous | 4 days |

* For µ_{max} will be the one according with the maximum rate obtained with the best carbon source

** The µ_{max} HRT: Hydraulic Retention Rime

Here it is included the ** HRT = 1/D**, which is a parameter that tells us how many time a particle (i.e. an enzyme) will stay overall in the reactor, and might be useful to further optimise the process, according to hypothetical enzyme turnover times (i.e. in the orange fed-batch bioreactor, and so on (not tested).

## Testing steady flows

Due to the lack of resources in the laboratory and the time necessary to get all the enquiries obtaining a proper pump was not possible, a Do-It-Yourself (DIY) special pumping system based on a IV perfusion pumping system was designed and adapted (see Figure 6) for pumping the media into the bioreactor continuously and with a steady flow.

For the support of the liquid (LB media with antibiotics), 4 Litre sterile bags (HOLTEC, France) were used and each bag was coupled to an unit of a IV perfusion system, *Elbiol – IV Perfusion set NMC3325V* (Baxter Healtcare, Switzerland) where 1 drop equal to 0.05 mL.

A “Y junction” together with two sterile pipettes commonly used for 10-100 uL micropipette, were used as a junction between the tubing and the *drip chamber* (see Figure 6 below for greater details).

__Figure 6.__ A pumping DIY device that uses gravity to deliver a steady flow rate because of the fixed height in the *drip chamber.*

Since the pumping system relies on gravity, the physics behind follows as described by Bernoulli equations for hydrodynamics:

A conventional siphon works by the difference in gravitational force between the tubes connected to the upper and lower reservoirs, where reservoir 1 is kept at a higher level than the other reservoir [2]. This siphoning can be theoretically described using Bernoulli’s equation (eq.4):

Where P is pressure, ρ is density of the fluid, g is gravitational acceleration, h is height difference between input and output, and V is velocity of the fluid.

In the drip chamber, fluid flow is basically driven by pressure difference; therefore, the surface of both the upper reservoir and lower end of the tube were considered as atmospheric pressure. Therefore, neglecting the values regarding pressure to measure the velocity of fluid it can be rearranged as follows:

Here, the fluid velocity is completely determined by the height (H), which is calculated by the difference in the height between the surface of the upper reservoir (the level of LB media liquid in the IV perfusion chambers) and the lower end of the output tube in the bioreactor. By using eq.5, and taking into account the area (A) of the tube, the flow rate (F) can be calculated by using the eq.6 below:

Therefore, is demonstrated the principle by which this flow rate was adjustable through alteration of the the height (H) between the upper chamber (LB reservoir) and the output (the bioreactor), only by the height of bioreactor respective to the fixed liquid level of the IV perfusion chambers. If not for the drip chamber the flow would be decreasing over time as soon as the volume level of the LB reservoir (the bag) drops.

Following these principles, the pump was settled to pump a the first two chemostats in series in order to operate the two first steps of the system presented in the Figure 2.

Nevertheless, following practical application of these principles, and to ease the use of the bioreactor for ensuring that the rate is steady and durable over time, a “safe and sure” approach was used, taking into account the laws described above where it is stated that only height (H) in the drip chamber is affecting flow rate (F).

This “safe and sure” approach relies in the Flow regulators (tube closures/screws) that were provided with the IV perfusion sets (see Figure 2 and Figure 6) and the chemostats/bioreactors. These flow taps/screws can regulate the flow by being lidded, and by using a value of height (H) much higher than the minimum required for the system to have enough potential energy to be pumped at the velocity of interest. This secured height threshold (H_{threshold} = H_{min} + *security threshold*) between the reservoir out of the LB media bag and the chemostat, can thus be regulated exclusively by screwing partially the taps and counting the output flow. For instance, also without the need of comparing experimental and theoretical values for the online regulation of the flow.

Since a flow of ~0.42 mL/min is the one used for the experiments, a flow rate of > 1mL / min was considered safe to ensure that potential energy necessary for pumping (by means of height H) was more than enough so that it can be regulated by screwing the taps. For testing this, with the taps completely open (i.e. without narrowing the taps) a secure height of H = 66 cm for the perfusion set respective to the output tube was ensured, and flow rates regulated by only screwing the taps, flow rates were tested by using a test tube of 100 mL (H0 = 21 cm) and counting the time it does take to fill a particular volume. Time was recorded every 10-20 mL and a mean value was taken as a flow rate.

Here a particular example:

If, H3 – H2 = 66 cm; H1 - H0 = 54 cm

- 10 mL: 4:40 min rate of 2.14 mL/min
- 30 mL: 18:00 min (∆ 20 mL= 13:20 min) 1.5 mL/min
- 40 mL: 25:35 min (∆ 10 mL = 7:35 min 1.32 mL/min
- 50 mL: 34:32 min (∆ 10 mL = 8:57 min) 1.12 mL/min
- 70 mL: 55:25 min (∆ 20 mL = 20:52 min) 0.95 mL/min
- 80 mL: 1:07:41 min (∆ 10 mL = 12:16 min) 0.81 mL/min
- 90 mL: 1:21:22 min (∆ 10 mL= 13:40 min) 0.73 mL/min
- 99 mL: 1:27:00 min (∆ 9 mL= 5:55 min) 1.52 mL/ min

Flow rate = *Average flow rate* = 1.26 mL/min

With such a flow rate, *Hydraulic Retention Time* (HRT = V/F) in the both * chemostats* is (40 mL) is: HRT = 9.52 min

From here flows (Table 3) can be safely calibrated then by ensuring more pressure than necessary to pump the flow and regulating the potential achieved by narrowing the taps or using the tube closures.

## References

- Widdel, F. Theory and Measurement of Bacterial Growth (2010). Retrieved from: http://www.mpi-bremen.de/Binaries/Binary13037/Wachstumsversuch.pdf
- Kim, L., Vahey, M. D., Lee, H. Y., & Voldman, J. Microfluidic arrays for logarithmically perfused embryonic stem cell culture. Lab on a Chip, 6(3), 394-406. (2006)

Feel free to contact us!