This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Process design and optimisation serves as an important aspect of the initialisation of any biologically suited commercial manufacturing process meant for the large scale production of vital and economically important products like hormones, proteins, complex macromolecules serving as fuel alternates. The sustainability and the efficiency of such commercial viable processes are largely influenced by the periodic monitoring and optimisation of the parameters that are involved and directly affect the behaviour of the living cell systems. Thus for such process monitoring model design and calibrating of data, scale-down models are adopted which focus on replicating the parametric conditions and media heterogeneity present in the commercial process on a relatively smaller scale in order to ease the study pertaining to the effect of the variation in those parameters and factors that might assist in making the system more productive. This report is a critical review of such designing methodologies and the techniques that serve well for optimising the parameter that not only help the cell cultivation and productivity of the process but also makes it effective with respect to capital cost involvement and environmental regulations.
The concept of recombinant DNA technology for commercial production of proteins led to the onset of a new generation of manufacturing sector. The medically and economically useful protein molecules were no longer in the classification of scarce products but were being manufactured on a larger scale whilst maintaining their chemical complexity and purity. Hence the need for the optimisation for such biological processes became a vital aspect of the emerging industries. A bioprocess optimisation is a technique of monitoring the experimental setup under a range of conditions in order to enhance the productivity and sustainability of the system. Thus the factors under consideration are genetic (includes the expression levels of certain enzymes that regulate the entire pathway), physiological (involves the factors like carbon fluxes, and energetic states that regulate the intrinsic homeostasis of the biological organisms), and environmental factors in respect to the system-surrounding coexistence. (e.g., O2, CO2, temperature and pH). The monitoring of each of these factors at various levels of a cell growth basically constitutes as the optimisation of that process. With scale down concept for the bioprocess optimisation, it is possible to easily assess a large set of experimental results pertaining to the testing of media composition, cell lines, and the physical and environmental conditions so that optimal operation parameters for the production could be determined.
Experimentation time period
Figure 1. Conceptual representation of bioprocess optimisation
2. Bioprocess Optimisation
The optimisation of a biological procedure is based on the fact that in the early stages of the process design it is relatively easier to measure as many as parameters required for the optimisation but as the process reaches the manufacturing stages, the optimisation is merely carried out by controlling the critical parameters derived from the early experimental deductions. However, till date the research and development in the field has led us to the ability to monitor and control the parameters outside the system (biological cell) such as the oxygen density and uptake rate; cell density; nutrient availability etc. that serve sufficient to deduce the metabolic and physiological state of the cell system. Thus with an aim to characterise and optimise the process completely, there was a need for the deduction of the parameters within the system and hence certain refined mathematical models were introduced to deduce the properties like cellular growth, yield, media composition and other physical operating parameters like pH, temperature.
Figure 2. Illustration of the comparison between the information output and experimental throughput existing for the current cell cultivation monitoring systems. (Adapted from Doig et al., 2006)
Methods for such monitoring generally focus on increasing the experimental throughput for the process and decrease the complexity of the process data by reducing the size and volume, more commonly conceptualised as scale down model. As depicted in the figure 2, it is evident that the monitoring capabilities of a process are inversely related to the extent of experimental throughput for the cell cultivation method available. It is observed that the process complexity for the cellular cultivation optimisation is relatively lower at the scale down model of micro plates and shake flasks cultures as compared to the pilot scale bioreactor. Hence, it is deemed suited to scale down the reaction model in order to monitor the high-throughput experimentation results with a benefit of less complicated process data involved. Such monitoring capabilities for the scaled down model is primarily based on the sensitivity and specificity of the suited sensors that shall provide the required information about the extrinsic properties of the system such as pH, dissolved oxygen, cell mass, the levels of CO2 dissolved in the surroundings etc.
Bioprocess optimisation is a collection of four vital steps that help define the boundaries for the experimental model. They are:
Optimal Experimental design that is basically the definition towards the problem under consideration and the extent of the requirement of experimental data. This stage could be subcategorised into three different approaches which include (1) O.F.A.T Method that is the consideration of one factor (parameter) at a time for data collection, (2) Factorial or Composite Approach that compares the system response to two different parameters and thereby provide a polynomial data model for the optimal statistical results, and (3) Optimal design Approach in which the model is predesigned in order to minimise statistical variances with respect to the parameters and the prediction data. This serves to avoid the correlation among parameters and the variation of the variance tables prepared for parameters under consideration.
Identification of the parameters that shall serve to the collection of data for the optimisation. This is carried out by sequentially experimenting with factors (F) in the designed model to collect the results (R) in accordance. Thus by the mathematical modelling of the factors against their corresponding results provide the extent of influence (E) of the parameter.
Optimisation of the system performance that shall include the operation of the designed model under the optimal parameters. It concentrates over three aspects for a process design that are minimising cost of operation and maximising productivity and quality of the products.
Control of the optimal parameters to ensure the productivity. The main aim of controlling the parameters lies with the reduction of the result disturbance in respect to the observed data and the predicted data. It serves to ensure the coherence of the experimental and theoretical results and thus helps in making an efficient optimised model.
Optimisation of a process serves to the design of an efficient model for production of various recombinant and industrial products. Thus this approach serves to provide a well suited design for various processes like fermentation, production of enzymes and proteins, cell culture, separation of various bio-molecules and commercial production of products like malt, cheese etc.
3. Scale down of bioreactors
A conceptual approach for bioprocess optimisation that serves to provide an experimental design at a smaller scale that mimics the system-surrounding heterogeneity as it is observed in the larger scale model. This is carried out in order to ease the process complexity and thereby provide the designer a suited model to estimate the system behaviour under the influence of various parameters. The scaled down model ensures the cost-effective and time saving testing of different parameters along with the provision of evaluating the effects of process alterations in existing operating processes. In a critical review it is often noticed that scaling down of various existing processes in a well-defined environment at a smaller scale (say 10 to 100 mL) helps in monitoring and controlling the process in order to obtain data suited to characterize the microbial strains and the process conditions that effect their life cycle, thus providing the user an opportunity for carrying out high-throughput experimentation. Contrary to the previous belief regarding the optimisation of the biological process being dependent on the reactor volume, recent scale down methods including microtiter plates and other sophisticated sensory probes have provided a breakthrough in the earlier arguments of the volume and probe relationships. Hence, the development of miniature sensors for measurements has been an important step in closing the information gap between the data provided by the lab-scale and small-scale bioreactors.
Figure 3. Illustration of a scale down model.
As observed from the figure, it is evident that a scale down model works by mimicking the system and surrounding conditions present in an operating manufacturing process at a smaller scale in order to facilitate the monitoring of the parameters and thereby allowing a user to optimise and improvise the existing conditions in order to enhance the efficiency or the productivity of an operating process.
3.1 Methodology for Scale Down
The scale down of an optimal process serves as an important aspect in providing a platform essential for the optimisation of a bioprocess. The scale down concept greatly depends upon the optimisation techniques and requirements and could be subcategorized into different stages when it comes to defining the methods.
Identification of process that is existent in terms of a manufacturing scale procedure so as to replicate the process parameters to a smaller scale.
Identification of the system and the surrounding environment conditions.
Identification and precise replication of the process parameters that were sustained in the large scale bioreactor
Monitoring of the parameters in the small scale model so as to carefully study and record the interactions between the cells and their surroundings.
Collecting and optimising the data in order to provide a complete record for the process under consideration and measurements required for efficient operation.
This section broadly focuses on the nature and features of the apparatus and the sensory probes that are employed during developing a scale down model for the process optimisation.
Miniature bioreactor systems
It emerged as a novel approach towards scaling down of bioreactors for the purpose of process optimisation that has been found to be a faster and an efficient technique owing to its inherent high-throughput experimentation capabilities by simultaneously performing multiple cell cultivations in parallel.
Microscale/ Microwell systems
With the knowledge of the type of system to be employed for a particular process scale down, it is also essential to know the nature of probes that shall be utilised to measure and monitor the factors that drive the cell viability and cultivation capabilities. These factors are categorised as pH, dissolved oxygen, cell mass, the levels of CO2 dissolved in the surroundings etc. These measurements could be classified into:
3.3.1 Online in situ measurements
It is an approach in parameter effect measurement that focuses on the real time data mining and thus is suited for the parameters that are used in closed control loops like temperature, pH, dissolved oxygen and carbon dioxide. Thus sensors measuring these factors and monitoring them are placed locally in the reactor system at specific locations (specially designed ports) in order to ensure a uniform and homogeneous observation for the data. It is often noted that as the size (volume) of the bioreactor is scaled up, the overall productivity of the process is affected adversely by the increase in heterogeneity of the environment. Hence, it is often assured that the probes are strategically placed in order to accurately profile the reactor.
pH measurement is the most common parametric data as it is often noted that the careful monitor and control of a specific pH range is essential for an optimal cell growth. The general measuring probes for pH include steam sterilised glass electrodes but their low mechanical stability led to the use of optical sensors that follow the principles of absorbance or fluorescence of light from a range of pH-sensitive dyes. Though the use of optical sensors is limited by the narrow operating range, they have an advantage of sensitive measurements in values close to the pKa of the dye. Recent developments are focussed upon the broadening of the measurement ranges for the device and use of polymeric casts for the device in order to make it more cost-effective.
Dissolved Oxygen is often the limiting substrate for the cell cultivations due to low solubility in water and thus is essential to be measured and controlled in order to prevent the growth inhibition. The sensors generally used are steam sterilizable Clark-type electrodes that work on the principle of measuring the oxygen in the exhaust and thus monitor the amount of oxygen dissolved. Another measurement technique involves the sensors that could quantify the dissolved oxygen on the basis of the amount of fluorescence quenched. In comparison it is noted that the electrochemical sensors perform well under high oxygen concentrations while the optical sensors are employed for solutions having lower air saturation.
Cell Mass is an important factor that plays a vital role in the induction of product formation. Often it becomes essential for cultured cells to reach a specific mass before product formation or cell harvesting could be induced for commercial production. Scale down methods provide a suitable platform for the monitoring of cell mass and thus inferring the limitations offered in the environment for a sustainable growth. The measurement techniques include primitive methods of estimating the dry cell mass at regular intervals and also the modern breakthroughs of applying the principles of capacitance or permittivity within polarised membranes or measuring cell density over time with the use of optical density and absorbance results. The measurements for dry cell mass are carried out at specific time intervals by obtaining defined culture volume and oven heating it at a temperature of 105 till the water evaporates but the heat fails to damage the wall integrity of the cells, this dried cell mass is then weighed and tabulated against time intervals. The optical methods include the spectroscopic measurement of the absorbance of light at a particular wavelength and thus determining the variations in the optical density of the medium with respect to time intervals and thereby plotting a graphical representation in order to correlate the cell cultivation. Another approach developed for the purpose of the measurement of cell mass is the use of the phenomenon of capacitance produced by the cells under the effect of electric current within a polarized membrane. The capacitance directly corresponds to the cell density and thus the cell mass.
Dissolved Carbon dioxide plays an influential role in the cell cultivation as it can be exchanged by the cellular membranes and thus can affect the intrinsic pH values of the cells. Also the accumulation of the gas produced by the cells while respiring could eventually lead to the disruption in the steady state of the model design and might end up in reducing the efficiency of the system. Monitoring of the gas is carried out by various techniques but so far none of them has proved to be an economically and ethically (in respect to the system surrounding co-existence) suited one. General methods used were the use of the Severinghaus-type electrodes that consist of bicarbonate buffer separated from the process medium by a carbon dioxide permeable membrane and measures the pH variation caused in the buffer by the diffusion of the gas. But this technique faces a drawback in terms of the safety of the process medium as it is often observed that in case of a membrane failure, there develops a risk of the contamination of the media and thereby provides a negative result impact over the scale down model which was designed for monitoring and optimising. The second method involves the use of fluorescence based pH sensors that work in almost the similar fashion as they were employed for the monitoring of dissolved oxygen. A further advancement was made with the introduction of a gas permeable polymer that was meant to contain the fluorescence-based sensor in an immobilised state so as to prevent any media contamination risks but this technique was not suited for design and monitoring owing to its high capital requirement and offset costs.
3.3.2 Flow Injection Analysis (FIA)
Flow injection analysis is an alternative approach for the measurement of the parameters like pH, partial pressures exerted by oxygen and carbon dioxide in the medium, molar ratio of various components like ionic salts ( sodium, calcium, ammonium), proteins (galactose, lactose, glutamine), complex sugars (sucrose, starch, malt, glucose), by products (hydrogen peroxide, ethanol) etc. FIA serves as an insitu measurement technique that provides an efficient, cost effective and sustainable result in reasonable time duration. The recent developments in FIA have been the use of fluorescent labels on a range of analyte-sensitive proteins and thus providing a suitable opportunity to monitor them online and thereby providing a real time record on the optimisation properties.
3.3.3 Virtual Sensor Probes
Virtual measurements are performed when estimation towards the properties computed from sensors that hypothetically exist and/or from a sensor that provides an unsatisfactory result. Thus these mathematical models are suited to provide accurate approximation in results and also in a way they help in identifying and mitigating the faulty readings recorded from pre-installed sensors. Even though this approach towards parameter monitoring and optimisation serves in a way to minimise the variance in readings owing to the biological complexity of the organisms, designing of mathematical and neural networks that shall to certain extent mimic the complexity of the system, this approach still seems hypothetical and fails to correlate the theoretical and experimental findings and thus is limited to certain process optimising systems.
3.3.4 High-throughput Bioprocessing
High-throughput bioprocessing is a technique with a similar functionality as compared to the high-throughput screening for drug discovery. The basic principle in carrying out the high-throughput bioprocessing for a process lies with the scaling down of the entire process (that minimises the involvement of cost) and then treating each individual bioreactor in parallel to each other in order to reduce setup and breakdown durations for individual parameters. In this kind of optimisation method, optical sensors are found more suited for process monitoring over electrochemical sensors as they can assure real-time and accurate data collection that shall therefore support the process optimisation and efficiency.
Thus it could be inferred from this critical review that scale down procedures employed for optimisation of a biological processes primarily work with the concept of mimicking the parametric conditions and media heterogeneity as observed in a large scale bioreactor within a smaller scale or miniature bioreactor. This replication of process conditions to a miniature bioreactor ensures a comparative modelling and an efficient monitoring of parameters like genetic (includes the expression levels of certain enzymes that regulate the entire pathway), physiological (involves the factors like carbon fluxes, and energetic states that regulate the intrinsic homeostasis of the biological organisms), and environmental factors in respect to the system and their surroundings (e.g., O2, CO2, temperature and pH). The technological advancements noticed in the scaling down concepts limits itself to the nature and design of the scaled bioreactor configurations and probe design, sensitivity and their specificity. Additional areas of research might include the development of artificial or neural networks that could mimic the complexity and functionality of an entire biological system so that the parameter monitoring and optimisation could be limited to single stage scale down experimentation and thus an efficient and precise method could be utilised for increasing the sustainability and productivity of a biological process being used commercially.