1. References
[1] Sbrana, T. & Ahluwalia, A. in New Technologies for Toxicity Testing 138–153 (2011).
[2] Mattei, G., Giusti, S. & Ahluwalia, A. Design Criteria for Generating Physiologically Relevant In Vitro Models in Bioreactors.
Processes 2, 548–569 (2014).
[3] Miranda-Azpiazu P., Panagiotou S., et al. A novel in vitro 3D blood brain barrier model for comprehensive drug permeability and
toxicity testing. Paper presented at: Advances in Cell and Tissue Culture; 15/06/15; Pisa, Italy.
[4] Alias E., Hernández-Sánchez, F., et al. Development of an in-vitro tri-culture of human cardiac microvascular toxicological model
for cardiovascular sensor. Poser presented at: Advances in Cell and Tissue Culture; 15-17/06/15; Pisa, Italy.
[5] Vinci, B. et al. Modular bioreactor for primary human hepatocyte culture: Medium flow stimulates expression and activity of
detoxification genes. Biotechnol. J. 6, 554–564 (2011).
[6] InLiveTox Final Report. Development and evaluation of a novel tool for physiologically accurate data generation. 27/09/2012
Emma Surplice, Jodie Barber, Matthew Walters, Stella Homer,
J. Malcolm Wilkinson and Kelly S. Davidge
Kirkstall Ltd. 3 Aspen Way, Centurion Business Park, Rotherham S60 1FB
The Solution
• Quasi Vivo® perfusion systems provide
nutrient flow which allows for co-culture
and development of 3D structures [1]
• Cell viability and function are improved
• Modelling of the QV500 (MCmB) shows
improved oxygen availability in
comparison with other milli- and micro-
fluidic systems (Fig.1 and [2])
Figure 1. Oxygen concentration profiles for
micro- and milli-fluidic systems and the
MCmB. Hypoxic regions are represented in
white. Taken from [2].
Acknowledgements
Kirkstall would like to acknowledge Prof Arti Ahluwalia and her team at the University of Pisa for
collaboration and their expertise. We would also like to thank Profs Sheila MacNeil and John
Haycock (University of Sheffield) for supervision and provision of laboratory facilities for the neurite
and skin work; Profs Claus-Michael Lehr (Saarland University) and Vicki Stone (Heriot-Watt
University) for collaboration during the InLiveTox project; and our Innovate UK partners for
NeuroTox: Dr Sikha Saha (University of Leeds), and Prof Pankaj Vagdama (Queen Mary
University London) and CVTox: Drs May Azzawi and Araida Hidalgo-Bastida (Manchester
Metropolitan University) and Dr Tim Gibson (ELISHA).
The Problem
• Current in vivo animal models do not accurately represent human
physiology
• 9 out of 10 drugs that pass animal testing fail during clinical testing,
costing the industry $ millions per failed drug candidate
• Traditional in vitro techniques can provide valuable insights
• There are limitations to these systems: they only partially replicate
normal biological processes, which impacts on the utility and reliability
of the resultant data
Quasi Vivo® Chambers
• QV500 (left) allows cell culture in a submerged chamber and is
compatible with scaffolds, gels and coverslips
• QV600 (middle) is designed for growth at an air-liquid or liquid-
liquid interface
• QV900 (right) provides a range of configuration options within
the footprint of a standard well-plate, plus a clear base for
imaging
LIVER 1
Primary hepatocytes grown in the QV500 under flow show
enhanced CYP34A gene expression when compared to growth
under static conditions (Fig. 10) [5] and improved IC50 values,
where the response to the drug diclofenac more closely
represents that found in human clinical trials (Fig. 11).
Figure 11. IC50 plot comparing the response
of primary human hepatocytes cultured under
static (red) with flow (blue) culture conditions
to the drug diclofenac.
Diclofenac IC50 assay
Concentration ( M)
0 200 400 600 800 1000 1200
Fractionalvitality
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Flow
Static
Flow fit
Static fit
IC50
Clinical
data
Figure 10. Effect of flow rate on CYP34A gene
expression. Primary human hepatocytes were
cultured in MCmB (QV500) and mRNA expression
measured using qRT-PCR. From [5].
Static
Flow
LIVER 2
Hepatocytes grown on a variety of scaffolds show enhanced proliferation and
penetration when under flow conditions (Fig. 13). In addition, the architecture
of primary human hepatocytes (Upcytes®) grown under flow closely
resembles that of fresh, in vivo hepatocytes (Fig. 14). Data from
collaborations with Reinnervate and Medicyte.
Figure 14. Primary human
hepatocytes under flow (left)
compared to human hepatocytes in
vivo (right).
Figure 13. HepG2 cells proliferate and penetrate deeper into scaffolds when in
flow conditions over a 6 day culture period compared to static conditions.
RESPIRATORY
Calu human airway epithelial cells show greater proliferation under flow than
in static conditions (Fig. 8). They also create and maintain more effective
tight junctions (Fig. 9). Data from collaboration with the University of
Nottingham.
Day 1 Day 3 Day 5
StaticFlow
Figure 8. Calu human airway epithelial cells grown under both static and flow
conditions.
Figure 9. Formation of the tight
junctions between Calu human
airway epithelial cells grown under
flow conditions, stained with E-
cadherin.
NEUROTOX- Innovate UK project number 131730
This project aims to utilise the QV600 and flow to create a
physiologically relevant Blood Brain Barrier (BBB) model using human
blood endothelial cells (HBECs), astrocytes (HAs) and blood vascular
pericytes (HBVPs). We have shown that all three cell types survive
under 50 µL flow (Fig. 2) and that flow plus conditioned medium
increases the viability of HBVPs and HBECs (Fig. 3) [3].
HBECs HAs HBVPs
BeforeFlowFlow(50uL/min)
0 .0
0 .2
0 .4
0 .6
0 .8
570nmabsorbance
D ifferen t m ed ia
C on d ition ed m ed ia
M od el 40 l/m in
*
* *
H B V P s H B E C s
Figure 2. Comparison between Human Blood Endothelial Cells,
Human Astrocytes and Human Blood Vascular Pericytes before flow
and after 50uL/min of flow. Taken from [3].
Figure 3. Viability of HBVPs and HBECs
in standard medium under static
conditions (black bar), in conditioned
medium optimised for these experiments
in a static environment (white bar) and
conditioned medium under flow
conditions (striped bar). Taken from [3].
GUT- EU FP7 project InLiveTox
The aim of this project was to develop a
physiologically relevant gut model using cells
from the GI tract (Caco2), vascular
endothelium and the liver (Fig. 12) [6]. The
gut model would then be used to monitor the
passage of nanoparticles across the
membrane .Two loops were setup: a
prototype QV600 in one, with the basal side
connected to the second loop containing
endothelial cells and hepatocytes in the
QV500.We showed that a gut model under
flow works as well as current in vivo and in
vitro models.
Figure 12. The system conformation used to study drug toxicity
of the gastrointestinal tract. The intestinal epithelium (Caco-2) was
seeded on a membrane in a prototype QV600 chamber allowing
flow either side of the cells. The endothelial cells and hepatocytes
were seeded on in separate QV500 chambers. The conformation
allowed fluid to flow from the intestinal epithelium to the
endothelial cells and on to the hepatocytes.
NERVE
Neurites were grown in the QV500 with 50 µl/min flow. Compared with static
conditions, there was increased neurite formation (Fig. 5) and a significant
increase in the number of neurites over 6 days; cells grown in static
conditions were dead after 3 days (Fig 6). In addition, there was an increase
in necrosis of the neurites in the static cell culture whereas there was
significant proliferation of the cells in flow conditions.
Figure 5. Neurites before (left) and after (right) 50
µl/min flow was applied.
Figure 6. Culture of neurites in static
conditions and with 50 µl/min flow.
SKIN
Human de-epithelialised dermis 3D
cultures were grown with perfused
flow in prototype QV600 chambers.
After 15 days, skin grown under flow
conditions had a more organised
structure than that of static
submerged samples and was of
comparable quality to that grown
under static air liquid interface
conditions (Fig. 4).
Static -
Submerged
Perfused
Flow - ALI
Static - ALI
40X400X
Figure 4. Human de-epithelialised dermis 3D cultures
under submerged static, perfused flow ALI and static
ALI.
CVTOX- Innovate UK project number 131728
Figure 7. 7 day Static co-
culture of HCMs (blue),
HECs (green) and SMCs
(red) marked using
fluorescent stains. Taken
from [4].
This project is investigating the
creation of a co-culture representing
cardiac tissue: human cardiomyocytes
(HCMs), smooth muscle cells
(HSMCs) and endothelial cells
(HECs). We have demonstrated that
all three cell types can survive in
single culture under flow (data not
shown) and can be cultured together
in a single QV500 chamber in static
conditions (Fig. 7) [4].
Kirkstall is collaborating with many world-class researchers to develop advanced cell culture applications, described below.