Understanding Bioprocessing, Its Applications and the Role of a Bioreactor

Products originating from bioprocessing are everywhere in our daily lives: from the food on our table to the medicines we take and the fuel for our cars. The standard definition of bioprocessing is the production of a high-value-added product from a living source; but how does this process actually work? In this article, the stages of bioprocessing are described, with a focus on the central piece of this concept: the bioreactor.

What is bioprocessing?

Bioprocessing can be found in several fields; in biofuel production, for example, carbohydrate-rich biomasses can be converted to alcohols (ethanol and butanol), fatty acid esters (biodiesel) and cyclic hydrocarbons using fermentative processes carried out by yeast or bacterial cultures. Nevertheless, perhaps the most striking use of bioprocessing is its use in biopharmaceutical manufacturing. One of the most famous examples is the production of insulin, a protein that is much needed by people with diabetes.  Until 1978 insulin was produced via pancreatic extraction from cattle and pigs.¹ The yield and purity were not high, and insulin as a product had a high cost, besides often inducing allergic reactions and not being long-lasting. In 1983, Eli Lily started selling recombinant DNA human insulin that was expressed in Escherichia coli (E. coli).² A product with better quality, in higher amounts and higher effectiveness could then be produced, thus prolonging the lifetime of those who suffered from such disease.

Just like humans, microbial organisms can produce proteins. Using bioprocessing, we can specify which type of protein we want a microbe to produce, if we give the right instructions and conditions for that to happen. In practice, the “instructions” are often read in DNA inserts that are fed to these microbes, whereas the “conditions” include the correct choices of nutrients and agitation that will be introduced into a bioreactor where these microorganisms grow.

Bioprocessing can be divided into two main stages: upstream and downstream. In upstream bioprocessing, the cell type is chosen as a function of the desired product; the media components, type of bioreactor and scaling are also defined. The subsequent step is the harvesting of cells, which then ignites the second fundamental part of bioprocessing: downstream. During this stage, the cellular product is isolated, concentrated, purified and often modified.³ The product is then formulated and finalized into vials in a step commonly called fill-finish. In the following sections, upstream and downstream bioprocessing are further elucidated.

Upstream bioprocessing

 A central piece of upstream bioprocessing is the bioreactor. In this vessel, cell proliferation takes place, allowing the desired product to be produced in an optimized manner. A fine mass balance is necessary to achieve high efficiency in bioreactors, given that many variables are important at this stage, such as gas flow, substrate, liquid flow and initial cell concentration.⁴ For each biological source type and end product, these conditions need to be optimized, on top of defining the type of bioreactor, vessel volume and modes of operation. In this section, we dissect the multiple steps involved in upstream bioprocessing.

1.      Selection of biological materials

The type of cell culture or biological material that will carry out the bioprocessing in bioreactors is determined based on the intended product, which is most often a metabolite or recombinant protein. Although mammalian cells, bacteria and yeast cultures are the most well-known cell types used in bioprocessing, plant and insect cells can also be utilized for this end.⁵ Table 1 shows the pros and cons of each type of microbial cell culture.

Table 1. Advantages and disadvantages of the different types of cell culture used in bioprocessing

2.      Culture media selection and optimization

The culture medium provides the basic resources and energy source for a given microscopic life form to proliferate and produce the high value-added product. The choice of culture media depends on the organism to be cultivated and the cell line. In general, culture media consist of amino acids, peptides/proteins, carbohydrates and depending on the cell line, vitamins and fatty acids. A buffering system to regulate the pH of the suspension is also needed, as well as an influx of gases (CO₂, O₂) at times. Salts are also generally present to regulate the osmotic balance.

Given that different cell lines require different media for optimal cultivation – even within the same cell lines differences can occur in terms of proliferation and output depending on cultivation conditions – media optimization must occur to find the right balance of nutrient concentration, pH, osmotic pressure and gas levels in each case. The traditional approach for media optimization is to start with a known medium and then change one factor at a time through subsequent experiments. A step ahead of this approach, which can be very laborious, is the use of Design of Experiments, which uses a statistical algorithm to explore the impact of changing more than one factor at once on productivity and interactions. Media blending is also used for the same end. In this approach, different media are blended and tested successively, and productivity is measured each time to find the best blend. One disadvantage of media blending is that more than one factor is changed at once, thus decreasing the level of control of the bioprocessing engineer.⁶

3.      Inoculation

The starting point of a cell culture within media is the inoculum. The process of inoculation is defined as the transfer of a population of cells cultivated in a working stock (liquid or in agar plates, for instance) to a liquid medium of higher scale. The new environment should provide the cells with all the necessary conditions for optimal growth and production of recombinant proteins, metabolites or fermentation products. As a rule of thumb, the inoculum has a volume 20 to 200-fold lower than the culture medium where the cells will be propagated.

4.      Bioreactor operation

Bioreactors have multiple modes of operation, which are used according to the type of bioprocess being carried out, the biological material being used and the desired product. The typical operation mode is batch mode, in which microbial proliferation rises over time, nutrients decline and waste products are not removed from the vessel. Only pH and gas adjustments can be made until the end of the process. Fed-batch bioreactors address the issue of scarcity of nutrients over time, and additions of nutrients are added over the time course of the process to maintain the nutrient concentration. Perfusion processes involve not only the feeding of nutrients during the culture growth but also the constant removal of spent media in a continuous fashion. Cells are kept in the bioreactor via a cell retention mechanism so that the product can be continuously harvested. Finally, continuous fermentation is the fourth main process mode; it resembles the perfusion process in that there is a continuous flow of nutrients in and spent media out, with the difference that there is no mechanism to retain the biomass within the vessel during operation.⁴ Table 2 details the main benefits and drawbacks of each process mode type.

Table 2. Operation modes of a bioreactor for cell culture

 5.      Harvesting

After the cultivation of cells and the production of the product, the last step before downstream processing is harvesting. This is important as the purification and concentration steps carried out during downstream processing would be virtually impossible if cells were still present in the solution. The harvesting step removes cells, cell debris (using cell lysis), precipitants and any particulates in suspension. Several methods can be used for harvesting, but the most common include disc stack centrifuges, single-use centrifugal systems, depth filtration and tangential flow filtration (TFF).⁷

6.      Monitoring and control

The monitoring of the upstream process has an intrinsic relationship with the concept of Quality by Design, a systematic approach in which the knowledge of critical quality attributes (CQAs) of the product and risk management dictate the monitoring of critical process parameters (CPP). This philosophy stimulates the acquisition of real-time data in the upstream process to minimize waste, accelerate batch release and improve yield. This is made possible by the use of process analytical technology (PAT), which is a system that can analyze and control manufacturing quality by measuring CPP and CQAs in real-time, preferentially using online, in-line or at-line approaches ⁸. Some PAT tools are validated by regulatory frameworks, and they are currently used in the bioprocessing industry (Figure 1). Some of the common PAT tools utilized in bioprocessing are described below:

• Raman spectroscopy assesses vibrational modes of molecules and biomolecules. It is a powerful PAT tool for monitoring biological processes in which on-line and in-line measurements give real-time information on media nutrients such as lactate, ammonia, glucose, glutamine and total cell density in a non-destructive manner.¹¹
• Nuclear magnetic resonance (NMR) spectroscopy is traditionally used for compound characterization, especially in organic chemistry. Because high-resolution NMR uses cryogenic liquids and powerful magnets, only off-line measurements are possible. However, with the advent of low-field, benchtop-size cryogen-free instruments, real-time measurements are now possible and relevant to bioprocessing. Its versatility lies in the possibility of detection of several elements such as carbon, nitrogen, phosphorous, etc.
• Mass spectrometry is a high-resolution PAT tool that can be used in high-throughput assays for monitoring the identity and purity of complex biomolecules produced in bioreactors. It can be used to monitor a variety of CQAs, especially in monoclonal antibody production, as it can detect glycosylation and other molecular modifications critical for final product safety and efficacy.¹²
• Dynamic light scattering (DLS) can be used in quality control for determining protein size distribution and aggregation, although there are challenges regarding its use as a real-time PAT tool.
• Flow cytometry is also required in some cases where data on cell morphology and viability are needed to monitor the upstream process.

Figure 1. PAT tools for in-process monitoring in upstream bioprocessing. Credit: Technology Networks.

Downstream bioprocessing

As with upstream bioprocessing, the downstream process is very much tailored to the product that needs to be isolated and formulated. Several technologies for concentration and purification can be used in series to yield better results. Nevertheless, in general, downstream processing involves the following steps – here, with a focus on therapeutically relevant proteins.

1.      Capture

The capture stage serves the purpose of isolating the product from the mother solution so it can then be buffer exchanged and further purified. Precipitation induced by salts is one of the low-cost methods used for this end. Affinity chromatography is a robust technique with high selectivity that often uses protein A for monoclonal antibody capture. The same logic can be applied for adsorption capture using magnetic beads functionalized with protein A for the separation of therapeutic proteins.¹³

2.      Concentration

To continue the downstream process after capturing the protein of interest, it is important to concentrate the product to optimize the following steps and improve the cost-effectiveness of the entire process. The concentration step also acts as a pre-purification, as salts and low molecular weight species are discarded while the protein is enriched. The most common methods for concentration involve filtration, such as TFF, ultrafiltration or reverse osmosis. The cut-off molecular weight of these methods can be tailored depending on the size of the protein and the metabolites/salts that need to be washed off.¹⁴

3.      Purification

The purification step most frequently uses chromatography as the main separation method. Preparative columns for size-exclusion chromatography can separate the therapeutic protein in its monomeric form from aggregates or fragments. Ion-exchange chromatography can be used to separate proteins based on interactions with ions in the stationary and mobile phases. Affinity chromatography can also be used in purification, for example, nickel-functionalized columns can be used to purify histidine tail-containing proteins. More recently, monolith chromatography has been demonstrated as an efficient method for separating proteins, with the possibility of using 3D-printed columns.^{13, 14}

4.      Bioconjugation

Bioconjugation is a common step following the separation and purification of therapeutic proteins to introduce tags for later analytical testing ¹⁵ or to enhance the activity of these species and elevate their effectiveness as therapeutic agents. For example, glycoconjugation is a strategy that can enhance the immune recognition of antibodies. The conjugation of antibodies with pharmaceutical drugs (antibody-drug conjugates) can also elevate the potency of the drugs by improving their targeting of cancer cells.¹⁶ Modification of proteins to increase stability is also possible, for example with the introduction of polymers to improve solubility.¹⁷ To introduce foreign chemical groups or molecules to a protein, the side chain of lysine residues and cysteines can be used. The N-terminus and C-terminus can also be used due to their reactivity.

5.      Formulation

Formulation is the last step before a drug can be marketed. It involves adding excipients, adjusting buffer and salt concentration and often a drying method to yield a finalized drug product in solid dosage, which can then be resuspended before use. A large percentage of therapeutic proteins are marketed in powder form as this improves shelf-life and overall protein stability. Spray-drying and freeze-drying are techniques commonly used for this end; the prior addition of carbohydrates and, sometimes, polymers, helps in protecting the protein against abrupt changes in temperature and pressure.

The role of bioprocess engineering

Given the number of steps involved with bioprocessing and the range of parameters that can be optimized, bioprocess engineers have the complex task of choosing the best conditions that favor the production of a high-value final product. This starts with the choice of type of microorganism or cells that will be used (mammalian, bacterial, fungi, plant, etc.), the appropriate media for the biological system and the scale and type of reactor to be used. At this stage, several optimization steps might be needed to find the perfect balance between all the variables. The downstream process also requires intense study for any given system; certain recombinant proteins, for example, are very easily degraded or aggregated during the purification process and this might necessitate a long process of study to guarantee their integrity as a final product.

Bioreactor design for bioproduction

There are several types of bioreactors (Figure 2); the stirred-tank bioreactor was the first used on a large scale in 1943 for penicillin production.¹⁸  It consists of a vessel with a jacket to regulate temperature and continuous agitation in the form of stirring with one or multiple impellers. For biological systems in which the agitation produced by impellers can harm cells (for example, mammalian cells), bubble column bioreactors can be quite efficient. They consist of a vessel in which agitation and aeration happen simultaneously using bubbles that circulate vertically through the reactor.¹⁹ Similarly, airlift bioreactors rely on aeration for agitation; however, they consist of two interconnected areas in the vessel that divide the fluid volume. The zone where aeration and sparging occur is called the riser, whereas the other zone is called the downcomer.¹⁹

Another strategy to avoid the issue with the shearing of stirred-tank bioreactors is to entrap cells on surfaces or small compartments where the medium is continuously supplied. This is the functioning principle of hollow fibre bioreactors, in which cells are immobilized within the fibre, thus preserving their integrity.²⁰ For biological systems in which the cells undergo photosynthesis (such as microalgae), photobioreactors are useful for producing biomass. Essentially, a light source is used in combination with a specific bioreactor type (airlift, stirred tank, etc.) to induce the production of biomass by cells ²¹.

Figure 2. The five most common types of bioreactor used in bioprocessing. Credit: Technology Networks.

Implementation of single-use bioprocessing

Disposable consumables and equipment for bioprocessing were first developed in the '70s, with the introduction of disposable bags and tubing for upstream processing. In the '90s, the first hollow fiber bioreactors appeared, and then in the ‘00s disposable TFF and filtration systems were introduced. Nowadays, the concept of single-use bioprocessing is consolidated and presents several advantages over the traditional stainless-steel-only manufacturing plants.

Single-use bioprocessing (SUB) is present in basically all bioprocessing stages such as cell culture (single-use bioreactor), separation (TFF cassettes, filters), purification (chromatographic columns, membranes) and fill-finish. As a result, SUB gives flexibility to manufacturing, speed in achieving regulatory approvals from the point of inception and reduced capital investment. The running costs of SUB facilities are often lower than for traditional ones, given that pipework, installation, instrumentation and process controls are impacted by the concept. An inevitable drawback is the production of more plastic waste. In addition, the lower material strength compared to metallic installations means that batches need to be smaller and automation might be a challenge, especially as there is often a lack of standardization of single-use technology across suppliers.²²

Key benefits and drawbacks of bioprocess automation

Since the 1970s, with the introduction of Industry 3.0, the bioprocessing industry has faced a push towards automation and digitalization processes. The ultimate benefit of automation is the ability to maximize yield while waste and resource consumption are minimized. The biggest difficulty in automation, however, is the standardization of the reproducibility from batch to batch, given that the production of metabolites or recombinant proteins by cells is a complex process with several interconnected variables. For a bioprocess to be automated, a set of variables must be chosen to serve as an input to an automation model, which will then control the process. The most important variable to be measured is the biomass or the concentration of cells that perform the biological conversion to the desired product. Temperature, pH, agitation, viscosity, cell viability and gas concentration are also important variables that might be used for this end.²³

An effective automated bioprocess thus requires good sensors and/or biosensors that can measure these variables, preferably in real-time, to feed into automation systems for bioprocess control. This increases the cost of automated processes when compared to manual ones. The automation process itself may occur under different algorithms, for example: based on model simplification, predictive control, artificial neural networks or empirical procedures.²³

Biomanufacturing applications

Biomanufacturing has been applied by mankind, although in a rudimentary way, for thousands of years. Only in the last century or so has the sophistication of bioprocesses improved, amplifying the range of biomanufacturing applications. In general, the first wave of biomanufacturing, which was focused on the production of first-generation metabolites such as alcohols, acetone, glycerol and amino acids is classified as Biomanufacturing 1.0.²⁴ The second wave, in the 1940s, introduced the production of antibiotics and secondary metabolites by fungi and bacteria, was named Biomanufacturing 2.0. Biomanufacturing 3.0, in the 1980s, came with the production of recombinant proteins and antibodies from recombinant DNA, which revolutionized the biopharmaceutical field. Finally, Biomanufacturing 4.0 has introduced cell and gene therapies, tissue engineering and digitalization tools.²⁴ Below, some of the main applications of biomanufacturing are discussed.

Biofuels

The push for green utilization of energy and sustainable energy sources has stimulated research into the production of biofuels. Biodiesel, bioethanol and bio-oil among others, can be produced via fermentation of biomass or plant waste.²⁵ Cellulose, hemicellulose and lignin are substrates for bacterial or fungal fermentation, and these can be found in biosources such as wheat, rice, corn and nutshells. More recently, microalgae have also been used for biomass production.²⁶

(Bio)pharmaceuticals

The production of pharmaceutical products such as antibiotics is perhaps one of the biggest applications of biomanufacturing. The discovery that bacterial secondary metabolites may act as antibiotics for human use boosted the research during Biomanufacturing 2.0. In the last decade, the discovery of marine bacteria and their secondary metabolites has been increasing the number of known antibiotics.²⁷ Protein expressions by bacterial and mammalian cell systems also represent a large part of pharmaceuticals currently on the market. Starting with insulin and then going through growth factors, recombinant proteins and monoclonal antibodies, bioprocessing has been addressing diseases such as diabetes, cancer, autoimmune disorders, allergies and inflammatory diseases.²⁸

Vaccines

In the field of vaccines and immunization, various technologies involving the use of recombinant virus vectors have been reported. The viral unit can be modified to carry specific antigens to target cells and induce immunization. Adeno-associated virus is perhaps one of the most studied examples of this type of technology, with recent investigations for vaccination against SARS-CoV-2.²⁹ Other viral scaffold types such as Newcastle disease virus and vesicular stomatitis virus can also be used for this end.³⁰ The use of viral vector vaccines, which consist of recombinant virus that can replicate and amplify the concentration of an antigen of interest, has been explored in bioprocessing in more recent years as well. In this case, only some cell lines can provide suitable biological machinery for producing viral vectors, such as human embryonic kidney cells, mouse fibroblast cells and human lung cells.³¹