A Comprehensive Overview of Pichia pastoris Fermentation Methods: From Laboratory Shake Flasks to Industrial-Scale Production – Which Approach Should You Choose?

In the fields of recombinant proteins, enzyme preparations and active ingredients for medical aesthetics, Pichia pastoris is undoubtedly the leading eukaryotic expression platform. Many R&D personnel are only familiar with the standard methanol-supplemented batch process, yet remain unaware that different fermentation modes directly determine protein yield, activity, the extent of proteolytic degradation, and even the success or failure of pilot-scale scale-up. With the same strain, changing the fermentation strategy may result in a manifold increase in yield—or it may lead to complete failure of expression. Today, we review the mainstream Pichia pastoris fermentation modes, analysing their respective applications, advantages and critical shortcomings to provide guidance for selecting the appropriate approach for small-scale, pilot and industrial-scale production.

I. Batch Fermentation

Batch fermentation is the most fundamental mode, involving the addition of the entire culture medium in a single step. Once inoculated, no further substrate is added, and the fermentation is allowed to proceed until completion before the tank is discharged. Throughout the process, the carbon source is continuously consumed, and the cells progress through the lag phase, logarithmic phase, stationary phase and decline phase. Laboratory shake flask cultivation essentially constitutes a simplified form of batch fermentation.

Advantages: Simple operation, minimal equipment requirements, low risk of contamination; suitable for early-stage strain screening and small-scale trial condition optimisation.

Disadvantages: Unable to achieve high cell density; rapid cell death upon depletion of the carbon source; short induction time for the AOX1 promoter; low upper limit of protein expression; cell lysis in the late induction phase, leading to the release of large amounts of proteases and the degradation of the target protein; rarely used for large-scale industrial production.

Suitable for: High-throughput initial strain screening and optimisation of shake flask conditions; not suitable for scale-up production.

II. Fed-batch fermentation (with continuous feed addition) — the industry standard

This is currently the standard approach for the industrialisation of Pichia pastoris and is also the preferred process for the vast majority of CDMOs. The process is divided into three main stages: the glycerol batch culture stage, during which the cells proliferate rapidly; the glycerol feeding stage, which drives the cells to a high OD; and, following a carbon source starvation transition, the switch to methanol for induction via continuous feeding, with methanol supply strictly controlled throughout to sustain induced expression, culminating in product harvest at the target time.

Within the industry, further specialisation has emerged, including DO-stat dissolved oxygen feedback, methanol rate gradient feeding, and methanolsorbitol/glycerol mixed feeding. Mut⁺ strains consume methanol rapidly and are highly oxygen-demanding; Mutˢ strains have a milder metabolism, requiring a reduced methanol flow rate and an extended cycle time.

Advantages: Capable of achieving ultra-high cell densities, with wet cell mass exceeding 200 g/L and a high upper limit for protein titer; the process is mature, with extensive scaling-up experience, traceable processes, and compliance with biopharmaceutical regulatory requirements; The mixed carbon source strategy alleviates the oxygen stress caused by methanol alone, reduces cellular stress and minimises protein degradation.

Disadvantages: Methanol metabolism is extremely oxygen-demanding; dissolved oxygen very easily becomes a bottleneck upon scale-up; excessive accumulation of methanol leads to formaldehyde and formic acid toxicity, causing damage to the cells; tanks must be emptied, cleaned and sterilised at the end of each batch, resulting in limited equipment utilisation; fluctuations are common between batches, and the process is highly dependent on the operator’s experience.

Suitable for: The industrial production of the vast majority of recombinant proteins, enzymes, collagen and peptides; this model is the preferred choice for pharmaceutical-grade projects.

III. Continuous Fermentation (CSTR Continuous Culture)

Fresh culture medium is continuously fed into the fermenter, whilst the fermentation broth is discharged at an equal flow rate, thereby maintaining the cell concentration, substrate levels and product levels within the vessel in a long-term steady state. Unlike batch feeding, this process does not involve draining the vessel and restarting the culture; theoretically, it can sustain a state of high-efficiency production for extended periods.

Advantages: The cells remain in a logarithmic growth phase for an extended period, avoiding late-stage cell death and the massive release of proteases; equipment utilisation is significantly improved, and production cycles are extended; literature data indicate that, in some systems, yields can be increased several-fold compared to traditional fed-batch fermentation; it is suitable for secretory products, as the continuous removal of the product reduces product inhibition.

Disadvantages: The risk of contamination is significantly increased; should contaminating microorganisms enter the system during prolonged operation, the entire batch must be discarded; maintaining a steady state in methanol-induced Pichia pastoris is extremely challenging, as even minor fluctuations in methanol concentration can cause metabolic collapse; there is a risk of mutational drift in the strain, and the AOX1 promoter is prone to inactivation after prolonged passaging; process validation and regulatory compliance are difficult to achieve, and there are few examples of domestic pharmaceutical projects having been successfully implemented.

Suitable for: industrial enzymes and bulk chemical products; pharmaceutical proteins are currently rarely utilised in this manner and are mostly confined to laboratory and pilot-scale research.

IV. Semi-continuous/Repeated Feed-Batch (Cell-Recirculating Fermentation)

When fermentation reaches the late induction phase, the entire contents are not transferred to a tank; instead, centrifugation or membrane filtration is used to retain high-activity cells in the tank whilst discharging the supernatant containing the target protein; fresh medium is then added to restart induction, and this process is repeated over multiple cycles to harvest the product.

Advantages: Makes full use of pre-cultured high-density cells, saving time on repeated cell cultivation; reduces medium consumption, thereby lowering costs; compared to fully continuous fermentation, the operating cycle is controllable and the risk of mutational drift is lower.

Disadvantages: Membrane modules are prone to clogging by proteins and bacterial cells; repeated stress on the cells leads to a decline in cell viability after multiple rounds; the operational process becomes more complex, increasing the number of process control points; there is also a certain risk of bacterial contamination.

Suitable for: Secreted industrial enzymes and certain medical aesthetics ingredients; not suitable for proteins prone to the intracellular accumulation of toxic metabolites.

V. Dialysis Fermentation (Membrane-Isolated High-Density Fermentation)

By combining the fermentation system with a dialysis membrane, small-molecule metabolic by-products can be removed through the membrane, whilst large-molecule proteins and yeast cells are retained within the tank, thereby eliminating the inhibitory effects of metabolic by-products.

Advantages: Enables cell densities far exceeding those achievable with conventional fed-batch fermentation; metabolic waste is promptly removed, reducing cellular stress and significantly mitigating protease-mediated protein degradation; it offers distinct advantages for certain recombinant proteins that are prone to degradation.

Disadvantages: Specialised equipment is required; membranes are costly; scaling up is challenging; membranes are prone to clogging, making maintenance cumbersome; rarely implemented on an industrial scale, and is primarily used for research and small-scale process development.

VI. Methanol-Free Induction Fermentation Process (Novel Promoter System)

This process does not rely on AOX1 methanol induction; instead, it utilises constitutive or starvation-induced promoters such as GAP and P_{DH}, using only glycerol and glucose as carbon sources throughout the entire process, thereby circumventing a series of challenges including methanol explosion hazards, methanol toxicity and high oxygen consumption.

Advantages: No methanol is required, so equipment does not need to be explosion-proof; the fermentation process is simplified, and dissolved oxygen pressure is alleviated; operation is safe, and the process development cycle is shortened.

Disadvantages: The expression levels of some promoters are lower than those of AOX1; with constitutive promoters, exogenous proteins are expressed whilst the cells are growing, which may severely inhibit growth if the product is toxic to the cells; the database of large-scale industrial applications for novel promoters is not as extensive as that for AOX1.

Suitable for: Proteins sensitive to methanol; reducing capital expenditure on equipment for small and medium-sized enterprises; this represents an important future direction for development.