How we grow CHO cells to 20+ million cells per mL and achieve titers of 5–10 g/L over a 14–21 day fed‑batch campaign.
Welcome back to our deep‑dive series on monoclonal antibody manufacturing. In Step 2, we expanded the Working Cell Bank from a single vial through shake flasks and seed bioreactors to produce enough cells to inoculate the production vessel.
Now we arrive at the core of upstream processing: the production bioreactor. This is where the magic happens – where our engineered CHO cells grow to high density, consume nutrients, and continuously secrete the monoclonal antibody into the culture medium.
For a product like Humira (adalimumab), a typical fed‑batch production run lasts 14–21 days and yields 5–10 g/L of antibody in a 10,000–25,000 L bioreactor. But achieving that titer consistently requires meticulous control of environment, feeding strategy, and cell physiology.
Let’s break down Step 3 in detail, including bioreactor setup, inoculation, fed‑batch operation, process control, and harvest criteria.
Overview: What Is a Fed‑Batch Culture?
In a batch culture, all nutrients are added at the start, and cells grow until something becomes limiting (usually glucose or glutamine). In a fed‑batch culture, we start with a basal medium and then add concentrated nutrient feeds periodically or continuously to extend the productive phase and increase cell density and titer.
Why fed‑batch?
Achieves much higher cell densities (20–30 × 10⁶ cells/mL vs. 5–10 × 10⁶ for batch)
Extends culture duration (14–21 days vs. 7–10 days)
Increases volumetric productivity (g/L/day)
Allows control of metabolic byproducts (lactate, ammonia)
Fed‑batch is the industry workhorse for mAb manufacturing. Perfusion is an alternative for unstable products or very high cell densities, but fed‑batch remains dominant for legacy products like Humira.
Step 3.1: Production Bioreactor Setup and Preparation
Before inoculation, the production bioreactor must be sterile, calibrated, and filled with sterile basal medium.
Process:
Clean and sterilize:
For stainless steel bioreactors: CIP (clean‑in‑place) with hot water, caustic (NaOH), and acid, followed by SIP (sterilize‑in‑place) with saturated steam (121°C, 30–60 minutes).
For single‑use bioreactors (SUBs): The pre‑sterilized bag (gamma‑irradiated) is installed in the holder; no CIP/SIP required.
Calibrate sensors: pH probe (calibrated with pH 4.0 and 7.0 buffers), dissolved oxygen (dO₂) probe (calibrated to zero with nitrogen and 100% with air), temperature probe.
Fill with basal medium: Transfer sterile production medium (e.g., chemically defined, animal component‑free) into the bioreactor. Typical working volume: 8,000–20,000 L for a 10,000–25,000 L vessel.
Equilibrate conditions: Set temperature to 36.5–37.0°C, pH to 7.0–7.2 (using CO₂ sparge and base addition), dO₂ to 30–60% (using air/O₂ sparge and agitation), and agitation to 50–150 rpm (depending on vessel size and impeller design).
Check sterility: Take a sample of the medium before inoculation and test for microbial contamination (typically 24‑hour incubation).
Equipment:
Stainless steel bioreactor (e.g., ABEC, Zeta, Sartorius) or single‑use bioreactor (e.g., Thermo HyPerforma SUB, Cytiva Xcellerex)
CIP/SIP skid (for stainless steel)
pH, dO₂, temperature sensors (e.g., Mettler Toledo, Hamilton)
SCADA/DCS control system (e.g., DeltaV, Rockwell PlantPAx, Emerson)
Step 3.2: Inoculation – Transferring the N‑1 Culture
The N‑1 seed bioreactor (from Step 2) contains a healthy, log‑phase culture at 2–5 × 10⁶ cells/mL with >90% viability.
Inoculation process:
Transfer the entire N‑1 culture (e.g., 1,000–2,000 L) into the production bioreactor using a sterile transfer line (steam‑sterilized or single‑use aseptic connector).
Target inoculation density: 0.3–1.0 × 10⁶ viable cells/mL. This low starting density allows several doublings before reaching peak density.
Rinse the N‑1 vessel with fresh production medium to recover residual cells.
Record time zero (T0) – the official start of the production run.
Critical parameters at inoculation:
Viability > 90%
No signs of contamination (turbidity, abnormal pH drift)
Consistent cell size and morphology (checked microscopically)
Step 3.3: The Growth Phases in a Fed‑Batch Culture
A typical fed‑batch culture progresses through four phases:
| Phase | Days | Viable Cell Density (×10⁶/mL) | Characteristics |
|---|---|---|---|
| Lag | 0–1 | 0.3–1.5 | Cells adapt to new environment; minimal growth |
| Exponential (log) | 1–7 | 1.5–20 | Rapid division; high nutrient consumption; lactate production |
| Stationary | 7–14 | 15–25 | Growth slows; viability remains high; maximum titer accumulation |
| Decline | 14–21 | decreasing | Viability drops below 70%; cell lysis releases intracellular impurities |
Goal: Extend the stationary phase as long as possible while maintaining viability >70%.
Step 3.4: Feeding Strategies – The Key to High Titer
Without feeding, glucose and glutamine would be depleted by day 5–7, and lactate/ammonia accumulation would poison the culture. We add concentrated feed media to replenish nutrients and control byproducts.
Typical Feed Components
Glucose (100–500 g/L stock) – primary carbon source
Glutamine (20–50 mM stock) or dipeptide (e.g., GlutaMAX) – nitrogen source
Amino acids (concentrated 10–20×)
Vitamins, trace elements, lipids
Plant‑derived hydrolysates (e.g., soy or wheat) – often avoided for chemically defined processes
Common Feeding Regimens
| Regimen | Description | Pros | Cons |
|---|---|---|---|
| Daily bolus | Single large addition once per day (e.g., 3–5% of working volume) | Simple; easy to sample | Glucose spikes; potential osmotic shock |
| Multiple bolus | 2–3 smaller additions per day | Smoother nutrient profile | More operator time |
| Continuous (exponential) | Pump feed continuously at an exponentially increasing rate matched to cell growth | Optimal control; minimizes byproducts | Requires modeling and automated control |
Example daily bolus schedule for a 15‑day run:
Day 2: 3% feed
Day 4: 4% feed
Day 6: 5% feed
Day 8: 5% feed
Day 10: 4% feed
Day 12: 3% feed
Control target: Maintain glucose between 2–6 g/L. Below 1 g/L → risk of starvation. Above 10 g/L → osmotic stress and lactate overflow.
Equipment for feeding:
Peristaltic pumps (e.g., Watson‑Marlow, Masterflex) with sterile tubing
Feed bags (single‑use, 50–1000 L) on load cells or scales
In‑line glucose/lactate analyzers (e.g., Roche Cedex Bio HT, Nova BioProfile FLEX2)
Step 3.5: Process Control – Keeping Cells Happy
CHO cells are demanding. We must maintain tight control of several parameters.
| Parameter | Typical Setpoint | Control Method |
|---|---|---|
| Temperature | 36.5–37.0°C | Jacketed vessel with hot water/cooling water; optional temperature shift (e.g., to 33°C late in run) |
| pH | 7.0–7.2 | CO₂ sparge (to lower pH) + base addition (0.5–1 M Na₂CO₃ or NaOH) |
| Dissolved oxygen (dO₂) | 30–60% of air saturation | Air/O₂ sparge through microsparger; agitation cascade |
| Agitation | 50–150 rpm (tip speed 1–3 m/s) | Marine or Rushton impellers; variable frequency drive |
| Pressure | 0.2–0.5 bar (positive) | Backpressure regulator; prevents contamination |
| CO₂ (partial pressure) | <150 mmHg | Air sparge to strip CO₂ (for large vessels) |
Advanced control strategies:
Temperature shift: Reduce to 33°C after peak cell density to slow metabolism, reduce lactate, and extend viability.
pH‑stat feeding: Feed addition triggered by pH rise (indicating glucose depletion).
DO‑stat feeding: Feed addition triggered by dO₂ rise (also indicating nutrient depletion).
Equipment:
Biocontroller (e.g., Sartorius DCU, Applikon ADI 1035, Thermo HyPerforma G3Lab)
Mass flow controllers (MFCs) for CO₂, air, O₂, N₂
Acid/base pumps
In‑situ sterilizable pH and dO₂ probes
Step 3.6: Monitoring and Sampling
Throughout the 14–21 day run, we take daily samples (or use in‑line sensors) to track:
| Measurement | Method | Target / Action Limit |
|---|---|---|
| Viable cell density | Vi‑Cell, Cedex, or in‑line capacitance probe | Report daily; feed adjustment |
| Viability | Trypan blue exclusion | Maintain >70%; harvest if <50% |
| Glucose | BioProfile, Nova, or Raman | 2–6 g/L; feed if <3 g/L |
| Lactate | Same | Keep <3–5 g/L; high lactate indicates hypoxia or overfeeding |
| Ammonia | Same | Keep <5 mM; reduce glutamine feed if high |
| Osmolality | Freezing point osmometer | 300–400 mOsm/kg; >450 mOsm damages cells |
| Titer (mAb) | Protein A HPLC or Octet | Track productivity; determines harvest timing |
| Product quality | SEC (aggregates), icIEF (charge variants) | Weekly; ensure within specifications |
In‑line / at‑line options:
Raman spectroscopy (e.g., Kaiser Raman RXN2) for real‑time glucose, lactate, VCD, and titer prediction.
Capacitance probes (e.g., Aber Futura) for real‑time viable biomass.
Step 3.7: Harvest Decision – When to End the Run
The production run does not continue indefinitely. We harvest when the benefit of additional titer is outweighed by risks:
Harvest criteria (typical):
Viability drops below 70% – cells begin to lyse, releasing host cell proteins (HCPs) and DNA, complicating downstream purification.
Titer plateau – no significant increase over 24–48 hours.
Product quality drift – aggregates or charge variants exceed acceptable limits.
Metabolic exhaustion – glucose consumption stops despite feeding.
At harvest, the culture is called Harvested Cell Culture Fluid (HCCF) – a cloudy mixture of cells, debris, and secreted mAb. The next step (Step 4) is clarification (depth filtration and/or centrifugation) to remove cells before purification.
Typical harvest day: Day 14 for fast‑growing clones; Day 18–21 for high‑titer, slow‑metabolism clones.
What Can Go Wrong? (Troubleshooting Step 3)
| Problem | Possible Cause | Remedy |
|---|---|---|
| Slow growth after inoculation | Low inoculation density; cold medium; pH out of range | Verify sensor calibration; pre‑equilibrate medium; increase seed volume |
| Premature viability drop | Shear from sparging; nutrient limitation; toxic metabolite accumulation | Check sparger design (use larger bubbles); increase feeding; perfusion or bleed |
| High lactate (>5 g/L) | Oxygen limitation; overfeeding glucose | Increase dO₂ setpoint; reduce glucose in feed; use lower glucose concentration |
| Foaming | Protein accumulation; cell lysis; inadequate antifoam | Add antifoam (e.g., Pluronic F‑68, antifoam C); reduce agitation; check for contamination |
| Low titer (<2 g/L) | Poor clone productivity; feeding too conservative; temperature too low | Re‑evaluate clone; increase feed volume; check temperature setpoint |
| Contamination (turbidity, pH drop) | Bacterial or yeast ingress | Discard batch; investigate sterilization and aseptic technique |
What Comes Next?
Once the production bioreactor is harvested, the HCCF moves to downstream processing. The next post will cover Step 4: Harvest and Clarification – how we remove cells and debris using depth filtration and/or centrifugation, preparing the clarified supernatant for Protein A capture.
Questions for our readers:
What feeding strategy do you use: daily bolus, multiple bolus, or continuous?
Have you implemented Raman spectroscopy for real‑time control? What challenges did you face?
The text above is owned by the site above referred.
Here is only a small part of the article, for more please follow the link
Also see:
Manostaxx – Industrial Management Consulting
