An Introduction to Perfusion Flow Cell Culture
Scope
This simple guide is designed for cell biologists who are newcomers to the use of perfusion systems for cell culture. It is assumed that the readers are already familiar with good laboratory practice and have previously been using 6 , 24 or 96 well plates for static well culture. In this document the term static culture refers to the method of seeding cells at the bottom of a well plate and then introducing nutrient media above the cells. During cell culture the well plate may be open to ambient air in a temperature-controlled incubator but no deliberate agitation of the plate is made. The guide illustrates techniques by reference to the proprietary Kirkstall Quasi Vivo System as this is a widely adopted equipment cell culture technique for introducing flow to 3D and multi-organ cell culture. However most of the techniques described can be applied quite generally.
Motivation for Perfusion
Static culture does not mimic the natural flow and pressure variations that cells will experience in the human body. Introducing media flow (perfusion) has many benefits. Fresh media flowing from a reservoir into a culture chamber not only provides nutrient to the cells but removes waste products from cell metabolism from the surface of the cells. For some cell types the flow stress produced on the surface of the cells by the flowing media can stimulate the cells and enhance growth and functionality.
A further benefit of perfusion is that it provides a mechanism for cells within the chamber to communicate and interact with distant cells of different types in connected chambers. Thus, introducing flow is a first step towards building more complex and more physiologically relevant cell culture models.
This guide is intended to support cell biologists to make the first steps towards the use of more complex models and to explain some of the physical principles that may be unfamiliar to many cell biologists.
Important Physical Principles
Cell culture systems incorporating flow rely on several physical principles:
- Flow of liquids (particularly media) depends on the creation of pressure gradients
- Liquid flowing through tubes experiences resistance to flow from viscosity
- Bubbles or obstructions to the flow may be created by surface tension effects.
Design of a Flow System
Flow systems will typically be of two types: single pass or recirculating flow.
In either case, many different components may be connected in series: reservoir bottles, connecting tubing, culture chambers, sampling ports, waste receptacles and a pump to produce the pressure to drive flow. The resistance to flow comes from all these components but tends to be dominated by the resistance in sections of narrow tubing.
The Schematic Diagram below shows a typical connection for a flow system with three cell culture chambers connected in series with a peristaltic pump head and media reservoir bottle in a recirculating loop where the pump is pulling the media from the chambers and returning the conditioned media from the cells back to the reservoir.
Schematic Diagram of Flow System
Flow and Pressure
Poiseuille’s law gives the resistance R to laminar flow of an incompressible fluid having viscosity η through a horizontal tube of uniform radius r and length l as R=8ηl /πr4
This indicates that the resistance to flow is very sensitive to the radius or diameter of the tubing used. Resistance to flow in a channel is proportional to 1/r4 ( r =radius) and so a 200um channel will have 625 times as much resistance as 1mm tube. This is one of the reasons why Kirkstall’s Quasi Vivo System uses much wider diameter tubing than many competitive organ on a chip systems.
The pressure needed to generate a flow of will also depend on the length of the tubing. In general the resistance to flow from the chambers or other components is much less than that from the tubing.
Many types of pumps have been utilised for organ on a chip systems:
- Gravity
- Syringe pump
- Peristaltic pump
- Pneumatic Complex Pumps
Kirkstall has selected Peristaltic pumps because they provide physiologically safe levels of pressure adequate to generate flows between 10 microlitre/min and 6000 microlitre/minute.
Pneumatic and syringe pumps are often chosen by companies developing microfluidic channels because they need much higher pressures (As much as 625 times as much pressure if they use 200 micron width channels).
Experimental Requirements
Important factors in the design of flow experiments will be:
- What is the total volume of the system
- How long does it take to fill the system with media
- What is the total media volume
- How fast does the media take to circulate round the system
- How can media be sampled from to measure metabolites
- How often should I change the media
- How to avoid flow stress
- What type of inserts can be used for the experiments.
Some quick guidelines might be helpful.
Volume and Fill Time:
A 1.6mm diameter tube with a length of 10cm has a volume given by πr2 xL =0.2millilitre
A QV1200 chamber has a volume of approx. 1.76 millilitre
At a flow rate of 100microlitre per minute the tubing will fill in 2 minutes and the chamber will fill in 17 minutes. Three chambers connected in series would take 51 minutes to fill at this low flow rate. Two techniques to speed up the filling are firstly to run the pump at a higher priming rate until the chambers are full or pre-fill the chambers with a syringe filled with media before fitting the lids.
Circulation Time and Time to reach homeostasis
Although the media flowing round the system is being metabolised by the cells, the system will reach equilibrium fairly quickly. Computer models using computational fluid dynamics have been used to understand the way the partial oxygen levels and flow stress are varying inside the chambers and tubing. (see reference 1 Hyndman et al) Experimental observations of cell activity have shown that flow is generally very beneficial to cell growth, expression of CYP genes and the development of tight barriers. However these physiological behaviours develop over one to two days. (see Reference 2 Buesch et al )
The consumption of nutrients and build up of waste products can take place over several days and depends on cell metabolic activity. Many long term culture experiments have used protocols where part of the media in the reservoir bottle has been changed after 3 to 6 days.
Avoiding Flow Stress
The most important factor influencing flow stress is the flow rate from the pump.
A table illustrating the typical flow stress in vivo is given below (Reference Kim et al)
| Cell Type | Shear Stress/dyn cm-² | Experimental Conditions | Biological Effect |
|---|---|---|---|
| Vascular endothelial cells | 10-100 | in vivo | Normal in vivo range |
| Hepatocytes | <2 | in vivo | Normal in vivo range |
| Mouse emryonic stem cells | 6.5 | in vitro | No significant negative effects on proliferation or self-renewal |
| Human umbilical vein endothelial cells (HUVEC) | 4 | in vitro | Same c-fos levels as in static controls |
| Human umbilical vein endothelial cells (HUVEC) | 25 | in vitro | Elevated c-fos levels compared with static controls |
| Bovine aortic endothelial cells (BAEC) | 4 | in vitro | Same c-fos levels as in static controls |
| Bovine aortic endothelial cells (BAEC) | 25 | in vitro | Elevated c-fos levels compared with static controls |
| HeLa Cells | 4 | in vitro | Reduced c-fos levels compared with static controls |
| HeLa Cells | 25 | in vitro | Elevated c-fos levels compared with static controls |
| Chinese hamster ovary | 4 | in vitro | Same c-fos levels as in static controls |
| Chinese hamster ovary | 25 | in vitro | Elevated c-fos levels compared with static controls |
Generally, some cells prefer low flow stress and others need high flow stress to function in a physiologically correct way
Techniques to reduce flow stress on sensitive cells like hepatocytes include :
- Reducing flow rate to less than 100 microlitre/minute
- Covering the hepatocytes with a protective barrier layer such as collagen
Seeding the cells at the base of a deeper chamber where the highest flow rate is across the top surface.
Most pumps have variable speed control, however the flow rate in the system will also vary depending on the number of chambers, the diameter and length of the tubing, the position of the chambers and reservoir in relation to the pump (for instance, placing on a higher shelf in the incubator) and, most importantly, on the type of pump used.
It is important to be aware of and control for these variables, which could influence your experiment, and so the system will need to be calibrated before use to ensure the flow rate is correct. Any subsequent modification to the system setup should be followed by recalibration.
Depending on the pump you choose to use, the calibration process will vary. If your pump does not provide you with a readout of the actual flow rate, you will need to do the following to translate the pump speed to flow rate.
To calibrate the system:
- Run the required configuration filled with sterile Phosphate-buffered saline (PBS) and collect the liquid output over 5 minutes from the final chamber. It is important that air is expelled from the system before starting to measure the flow rate.
- Measure the volume of liquid circulated using a graduated glass cylinder or by weighing the medium after 5 minutes – remember to weigh the collecting reservoir empty first.
- Carry out this procedure three times for each of three different settings (for example, low, medium and high speed), then plot the mean values on a calibration curve, an example of which is
shown in Figure 8.
Figure 8. Example calibration curve using QVMP1 with QV1200 chambers.
Selecting an Insert
The QV1200 System is compatible with most inserts designed for use with 24 well plates. Inserts are useful as they allow cells to be seeded in static conditions in a standard 24 well plate and then moved to a QV1200 perfusion flow chamber after they cells have become adherent to the membrane at the base of the insert.
Below is a guide to choosing a suitable insert (Reference 4 )
| Application | Cell Types | Pore size (μm) |
|---|---|---|
| Anglogenisis | Endothelial, HMVEC, HUVEC | 3.0 |
| Co-culture | Stem, neuronal and various others | 0.4, 1.0 |
| Epithelial Cell Polarity | Epithelial Cells | 0.4 |
| Migration | Endothelial, HUVEC, HMVEC, Neutrophils, PMNs, Lymphocytes, macrophages, monocytes, Neuronal cells, Dendritic cells, Neurite outgrowth, Epithelial fibroblasts, Leukocytes, Smooth muscle | 3.0 3.0 3.0, 5.0 3.0 3.0, 5.0, 8.0 1.0, 3.0 8.0 3.0, 5.0 8.0 |
| Invasion | Melanoma, Glioma, Lymphoma, Jurkat Osteoblasts, Breast cancer, Endothelial | 8.0 8.0 5.0, 8.0 8.0 5.0, 8.0 3.0, 5.0, 8.0 |
| Tissue Engineering | Human skin model | 0.4, 3.0 |
| Toxicity Testing | Mouse fibroblasts Human lung | 3.0 0.4 |
| Transport and Permeability Studies | Caco-2 MDCK | 0.4, 1.0 0.4, 1.0 |
Cell Function Monitoring and End point Analysis
With more complex cell culture protocols it is likely that a number of techniques will be used rather than a single simple end point assay.
The Quasi Vivo system has been designed to allow samples of media and cellular material to be easily extracted from the culture during the experiment. In the case that the cell culture cannot be disturbed, then optical observation through the high quality optical window on the base of the QV1200 chamber is possible. Samples can be taken of the media or the cellular material
Sampling Media :
The cell culture media can be sampled from three sources:
The reservoir bottle
Sampling ports in the tubing
The cell culture chamber itself
The reservoir bottle:
- Stop the pump, drain the tubing, and transfer the closed system into the laminar low hood.
- Open the reservoir bottle aseptically and aspirate the desired volume of media required.
- Close the lid again and transfer the system back into the incubator and start the pump.
The tubing via use of sampling ports: This method requires the use of sampling or injection ports at the required locations within the flow circuit. Usually these are connected just before the inlet and after the outlet of the chambers. The sampling ports contain a silicone septum through which a sterile needle can be inserted to withdraw media for analysis or injected into the system. The flow does not need to be stopped or reversed for this method of sampling.
- Wipe the outer surface of the sampling port with sterile wipe and insert a sterile needle into the septum to aspirate the desired volume of media. This could be carried out within an incubator or inside of the laminar flow hood.
The chamber:
- Reverse the direction of the flow and drain the pipes.
- Stop the pump and transfer the closed system into the laminar flow hood.
- Open the chamber (use of the supplied lid key is recommended) and aspirate the desired volume of media.
- Close the chamber by placing the lid back on the chamber and transfer the system back into the incubator.
- Connect the pump head back on the pump and start the pump to resume the flow. Please check that the flow is in the right direction.
Sampling Cells or Tissue:
Cells and tissues cultured inside the QV1200 chamber are easily accessible for analysis at the end of the experiment. Some examples of analysis techniques include (but not limited to) Western Blotting, RT- PCR, microarray analyses, immunohistochemistry and viability assays such as the MTT assay.
The simplest way to perform these assays is to remove your cells from the chamber and process them in the same way as you would for static culture.
Circulating cells which are in suspension and are present within the tubing and reservoir can also be harvested via the reservoir bottle or sampling ports as described above.
References
- Tomlinson L, Hyndman L, Firman JW,Bentley R, Kyffin JA, Webb SD,McGinty S and Sharma P (2019)
In vitro Liver Zonation of Primary RatHepatocytes.
Front. Bioeng. Biotechnol. 7:17.doi: 10.3389/fbioe.2019.00017
- A Novel In Vitro Liver Cell Culture Flow System Allowing
Long-Term Metabolism and Hepatotoxicity Studies
Stefanie Buesch, Jenny Schroeder, Maureen Bunger, Theresa D’Souza, and Magdalene Stosik
APPLIED IN VITRO TOXICOLOGY Volume XX, Number XX, 2018 Mary Ann Liebert, Inc.
DOI: 10.1089/aivt.2018.0009
- Biological effects of shear stress depending on cell types (L.Kim et al.)10
Kim, L., Toh, Y., Voldman, J. and Yu, H. (2007). A practical guide to microfluidic perfusion culture of adherent mammalian cells. Lab on a Chip, 7(6), p.681.
A Novel In Vitro Liver Cell Culture Flow System Allowing
Long-Term Metabolism and Hepatotoxicity Studies
- Corning Web Page