Application Note
Connecting the gut and liver to enhance drug development
Explore how a dual-organ microphysiological system connects human gut and liver tissue to bridge gaps in predicting how drugs behave in the body.
Before an oral drug can take effect, it undergoes a complex biological pathway, from absorption in the gut to metabolism in the liver. This journey determines how much of the drug enters systemic circulation and how effective it ultimately is. Yet, current lab models often isolate these two organs, overlooking their dynamic interactions. Overcoming this requires an integrated, human-relevant system that connects the gut and liver and faithfully replicates this essential physiological axis.
Download the full application note to learn:
- How a dual-organ system using primary human tissues improves drug bioavailability predictivity
- Why this model better reflects in vivo human metabolism compared to cell or animal models
- How this approach supports smarter, earlier decisions in preclinical drug development
Top Image Credit:
iStock.com/sefa ozel
visit cn-bio.com
Connecting the gut and liver: a human relevant multi-organ microphysiological system for preclinical profiling of oral bioavailability
Yassen Abbas1, Hailey Sze1, Ashley A. Spreen2, Elizabeth M. Boazak2, William R. Thelin2 and Tomasz Kostrzewski1
1. CN Bio Innovations. 2. Altis Biosystems
cn-bio.com/208
APPLICATION NOTE
CN Bio’s organ-on-chip systems, which include the PhysioMimix® Single- and Multi-organ lab-benchtop instruments, enable researchers to model human biology in the lab through rapid and predictive human tissue-based studies.
The technology bridges the gap between traditional cell culture and human studies, advancing towards the simulation of human biological conditions to support the accelerated development of new therapeutics in application areas including oncology, infectious diseases, metabolism and inflammation.
Learn more at cn-bio.com
Introduction
Absorption, distribution, metabolism, and excretion (ADME) are four key processes that indicate the behavior of a drug following administration, and therefore play a key role in defining a compound’s pharmacokinetic (PK) properties and bioavailability. Oral bioavailability is defined as the fraction of a drug that reaches systemic circulation following absorption across the intestinal wall and first pass metabolism in the liver. ADME and bioavailability are central in determining the safety and toxicology profiles of compounds and are therefore crucial measurements at the preclinical stage of drug development.
Currently, a combination of simple in vitro assays that either model the gut (Caco-2 cell line) or the liver (liver microsomes and suspension hepatocytes), and in vivo animal models is used to profile oral bioavailability; however, significant limitations exist with both approaches. Caco-2 cells, which have been the workhorse for assessing in vitro intestinal permeability, cannot account for liver metabolism, plus the cell line has absent or low levels of enzyme and transporter expression. Liver microsomes and suspension hepatocytes are used for in vitro drug metabolism screening studies, but do not consider intestinal absorption. Collectively, these limit the accuracy of their estimations. Furthermore, in a seminal study investigating 184 compounds, animal models were found to have a weak correlation with bioavailability in humans (R2=0.34)1. A new human relevant approach, which combines oral absorption and hepatic metabolism, is therefore required to more accurately estimate drug bioavailability.
In the last decade, microphysiological systems (MPS), also known as organ-on-a-chip (OOC), have shown their potential to improve the human translatability of ADME studies. They are designed to recapitulate the structural and functional biomarkers of cells and tissues in a more physiologically relevant manner through the culture of primary human cells on perfused 3D scaffolds. Efforts to improve the in vitro to in vivo translation of drug efficacy and safety data has led to the emergence of more complex MPS where multiple organs, such as gut and liver, are fluidically linked together to simulate processes such as drug absorption and first pass metabolism2.
Aim
Here, we introduce a dual-organ MPS that links our established primary human liver MPS with a primary model of the human intestine. For the intestinal barrier, primary cells are isolated from the human jejunum and expanded on a biomimetic scaffold (RepliGut®). To link the primary gut and liver tissues together, we developed a chemically defined media that supports both organ models in the dual-organ MPS. This media enables both the maintenance of hepatic metabolic functionality and intestinal barrier integrity. Using well-studied drug compounds, we aimed to demonstrate the improved predictive capacity of this primary Gut/Liver MPS for profiling the ADME behavior of oral drugs compared to an equivalent Caco-2 Gut/Liver MPS.
Methods
Here, we describe a dual-organ MPS that links a primary human intestine (RepliGut® Planar-Jejunum, Altis Biosystems) with a primary human liver-on-a-chip (CN Bio). The dual-organ MPS is cultured using the PhysioMimix® Multi-organ System and its bespoke “Multi-chip” Dual-organ consumable plate (Figure 1A). The dual-organ plate consists of six wells, each with two compartments (i) a Transwell® compartment and (ii) a liver compartment. Fluidic flow can be independently controlled in each compartment (intestine and liver), and in the interconnecting channel between the organs (Figure 4A). The intestinal barrier was established through the expansion of human jejunum stem/progenitor cells on a Transwell® coated with a biomimetic scaffold, followed by differentiation into a polarized barrier comprised of all post-mitotic lineages found in the human intestine. Media was changed every 48 hours with either RepliGut® Growth or Maturation media. Intestinal barrier integrity was assessed with transepithelial electrical resistance (TEER) measurements every 48 hours or by determining the permeability of Lucifer yellow across the Transwell at the end of the dual-organ experiment. Expression of markers confirming intestinal origin (Villin and CDX2) and the presence of the tight-junction marker ZO-1 was confirmed by fluorescent microscopy. Mucus production was confirmed by histological staining of the RepliGut® Planar-Jejunum
cross-section with Alcian blue and by in-well immunofluorescent staining for the Muc-2 protein. Expression of metabolic and transporter genes were evaluated using RNA isolated at day 15 of Caco-2 culture or day 7 post differentiation of RepliGut® Planar-jejunum. Gene expressions of metabolic enzymes and transporters were measured by qPCR using TaqManTM Gene Expression Assays, with relative expression determined using ΔΔCT analysis.
For the liver, primary human hepatocytes (PHH) were seeded in the liver compartment of the PhysioMimix Dual-organ plate, on a porous 3D collagen-coated scaffold (Figure 1C). On day 4, following PHH seeding and the formation of microtissues, differentiated RepliGut cultures were added to the PhysioMimix Dual-organ plate to establish the Gut/Liver co-culture. A chemically defined media was used to maintain functionality of the gut and liver tissues for at least 48 hours, during which compounds were added to study their ADME profile. In this study, we compared the performance of the primary RepliGut/liver MPS versus a Caco-2 Gut/liver MPS. The Caco-2 Gut/liver MPS was established by adding Transwells with differentiated Caco-2 monolayers (at 15-17 days post seeding) to the PhysioMimix Dual-organ plate on day 4, following PHH seeding and the formation of microtissues.
We used 7-hydroxycoumarin (7-HC), a fluorescent compound that undergoes Phase II metabolism by glucuronidation, in a proof-of-concept study to demonstrate absorption through the intestinal barrier and subsequent first pass metabolism in the co-culture model. The PhysioMimix Dual-organ plate allows for flexibility in compound dosing with either Gut/Liver in co-culture or gut only (no PHH) and liver only (no gut barrier, with compounds dosed into a blank Transwell with no cells).
In this study, two compounds (Temocapril and Midazolam) who’s human ADME properties were not predicted by existing models, were investigated. Compounds were added at day 4 post seeding of PHH and addition of the gut tissues to Dual-organ plates by either oral (drug added to the apical surface of the Transwell) or intravenous (IV) dosing (liver only, drug mixed into co-culture media). Samples of media were taken at 0, 1, 4, 6, 24, and 48 hours, and analyzed by liquid chromatography–mass spectrometry (LC-MS) to determine the concentration of parent compounds in the liver compartment. An estimation of area under the curve (AUC) of both oral and IV concentration profiles was made using GraphPad Prism.
Results and Discussion
Using the PhysioMimix Multi-organ System and its Multi-chip Dual-organ plate (Figure 1), we established a primary Gut/Liver MPS to overcome the human-relevance limitations of current models in profiling oral drug bioavailability. In the RepliGut model (Figure 2A), jejunum stem and progenitor cells were expanded to confluence on a biomimetic scaffold before undergoing differentiation, which resulted in increasing barrier strength (Figure 2B). Gene expression confirmed downregulation of proliferative cell genes and upregulation of differentiated enterocyte genes relative to cells in the proliferative phase (Figure 2B). In the differentiation phase, the RepliGut model expressed tissue-specific markers confirming intestinal origin as well as markers of tight junctions and mucins, confirming multi-cellular lineages (Figure 2C). We confirmed this via histology with observations of a distinct and continuous layer of mucus at day 7 post-differentiation. By comparison, mucus production was absent in the epithelial cell line, Caco-2 model (Figure 2D). To assess the RepliGut model’s potential to improve predictivity over the Caco-2 model, we investigated the expression of major metabolic enzyme and transporter genes that are important in drug metabolism and active drug transport. Gene expression was found to be improved compared to the Caco-2 cell line (Figure 2E).
One of the challenges with the co-culture of two, or more, tissues together is establishing the conditions that maintain functionality of both tissue types. We designed a chemically defined media, which is added to the Transwell basolateral and liver compartments of the Dual-organ plate (Figure 3A). This media maintained hepatic cell health and functionality in co-culture for at least 48 hours, measured by LDH release, CYP3A4 activity and albumin production (Figure 3 B-D). Intestinal barrier integrity was also maintained for the 48 hours of co-culture, as measured by TEER (Figure 3E) and a Lucifer yellow permeability assay (Figure 3F), after the addition of the RepliGut model to the dual-organ plate. Functionality was further demonstrated by 7-HC dosing (Figure 4A) which undergoes metabolism through glucuronidation (Figure 4B), with both intestinal and hepatic tissues contributing to its clearance (Figure 4C).
To demonstrate the improved predictive capability of the primary RepliGut/Liver MPS versus an equivalent Caco-2 Gut/Liver MPS, we investigated the ADME properties of two drugs where both animal models and the Caco-2 cell line failed to accurately profile their ADME behavior and bioavailability in human. The first drug, Temocapril, is a pro-drug designed to be resistant to intestinal hydrolysis that is metabolized to Temocaprilat by carboxylesterase 1 (CES1) (Figure 5A). The isoenzyme pattern in human is well-characterized with the liver and intestine expressing CES1 and CES2 respectively3. In Caco-2 cells, there is a miss-match as CES1 is predominantly expressed, resulting in an overestimation of drug clearance (Figure 5B-C). In contrast, the more human representative ratios of both CES1 & CES2 expression by the RepliGut model make it more relevant for pro-drug studies (Figure 5G). Temocapril clearance was profiled with oral and IV dosing (Figure 5E). Rapid clearance was observed by the liver within 24 hours following IV dosing (Figure 5F). In line with intestinal CES expression, the primary RepliGut/Liver MPS correctly reported resistance to intestinal hydrolysis of Temocapril and less Temopcarilat was produced at 48 hours in the gut apical compartment, following oral dosing.
Finally, we assessed the bioavailability of Midazolam in both Gut/Liver MPS models, a drug known to predominately undergo intestinal clearance by the CYP3A4 enzyme4. Midazolam was rapidly cleared by the liver following IV dosing, reflecting its high intrinsic hepatic clearance rate (19 mL/min/Kg) (Figure 6A). We observed greater clearance of Midazolam by the primary RepliGut/Liver MPS, compared to the Caco-2 Gut/Liver MPS (Figure 6B). The primary RepliGut/Liver MPS also delivered an oral bioavailability estimation that more closely represented human clinical observations (Figure 6C).
Figure 1. Set-up of a standard Gut/Liver MPS in the PhysioMimix Dualorgan
plate.
A) Set-up of the PhysioMimix Multi-organ System and its Multi-chip Dualorgan
plate, showcasing the location of the gut and liver compartments
within the plate. Each plate can culture up to six Gut/Liver models. B)
Schematic diagram of a RepliGut® model cultured on a biomimetic
scaffold within a Transwell insert, forming an intestinal epithelial barrier. C)
Schematic representation of fluidic flow within the liver compartment of
the Dual-organ plate.
B
C
Epithelial
monolayer
Luminal
reservoir
Biomimetic
scaffold
Cassette
Basal
reservoir
Porous
membrane
Liver
microtissue
3D scaffold
Micropump
B
C
A
C
Figure 2. RepliGut primary model of the intestine is more human
relevant compared to the standard Caco-2 cell line model.
A) Crypt epithelium stem/ progenitor cells are isolated from jejunum
samples and expanded on a biomimetic scaffold in static conditions.
B) Barrier integrity measured by TEER and relative gene expressions,
in the expansion and differentiation stages of the RepliGut model. C)
Immunofluorescence images of the RepliGut model. The barrier is stained
for DAPI (dark blue), the tight junction protein, ZO-1 (light blue), and
other intestinal markers including MUC2 (yellow), villin (purple) and CDX2
(green). The image scale bars equate to 100 μm for CDX2 and 200 μm for
the other three markers. D) Histology sections of the RepliGut and Caco-2
models. Monolayer cross-sections are stained with hematoxylin and eosin
to visualize nuclei and with Alcian blue to visualize the mucus layer. E)
qPCR to demonstrate relative gene expressions of key transporters and
metabolic enzymes in the RepliGut and Caco-2 models.
A
B
-3 -2 -1 0 1 2 3 4 5 6 7 8 9 10
0
500
1000
1500
2000
Days of Differentiation Corrected TEER (Ω*cm2)
Cell Expansion
4-6 days
Mostly Stem & Progenitor Cells
Cell Differentiation
6-8 days
Mostly Differentiated Cells
MKI67 LGR5 SOX9 SI ALPI ANPEPFABP6
0.001
0.01
0.1
1
10
100
1000
Relative Gene Expression
Proliferative Cells Differentiated Cells
Proliferative Cell Genes Differentiated Cell Genes
D
E
RepliGut®
Caco-2 only
C
Figure 3. Liver and gut functionality markers are maintained
throughout the primary cell Gut/Liver co-culture.
A) Experiment timeline to establish the primary RepliGut/Liver MPS.
RepliGut was first cultured independently, in static, for 13 days before
being transferred to the Transwell compartment of the PhysioMimix Dualorgan
plate, 4 days after PHH seeding. The co-culture was maintained for
48 hours in a chemically defined media. Markers of liver cell health and
functionality were measured at 0, 24 and 48 hours of co-culture, with a
liver only MPS cultured in parallel as a control. We investigated B) LDH
release C) CYP3A4 activity and D) Albumin production. Barrier integrity of
RepliGut in the Gut/Liver MPS was assessed by E) TEER and F) permeability
to Lucifer yellow and compared to independent RepliGut cultured in static,
for its full duration in standard RepliGut® Maturation media.
A
B C
Day 0 13 14 15
9
PHH seeded onto
dual organ plate
Jejunum
cells seeded
Jejunum transferred to dual organ
plate, co-culture begins
QC
Drug dosed
Experiment
endpoints
Media sampling
Media sampling at
0, 1, 2, 4, 6, 24, 48 hours
Media samples sent for LC-MS for
drug quantification
LDH CYP3A4 Metabolism
Liver only Gut/Liver, RepliGut®
D
F
Albumin E Barrier Integrity
Permeability to Lucifer Yellow
Liver only
Gut/Liver, RepliGut®
Static, RepliGut®
Gut/Liver, RepliGut®
Figure 4. First pass metabolism can be modeled by the primary cell
Gut/Liver MPS.
A) Schematic representation of 7-HC dosing in the Gut/Liver MPS model.
1mM 7-HC was dosed in the apical side of the Transwell compartment,
containing the RepliGut barrier, within the Dual-organ plate. 7-HC was
then transported across the barrier into the media circulating between
the gut and liver tissues. The gut only control was prepared by inserting
the RepliGut model into the Dual-organ plate Transwell compartment
that was connected to a blank liver compartment (without PHH).
Likewise, the liver only control was prepared by dosing 7-HC into a blank
Transwell insert (without gut cells), in the Transwell compartment of
the Dual-organ plate. B) Pathway of 7-HC metabolism. 7-HC undergoes
phase II metabolism by UGT enzymes to form the non-fluorescent
metabolite 7-hydroxycoumarin glucuronide. C) Changes in 7-HC
concentration in the circulating media over time.
Gut only, RepliGut®
Gut/Liver, RepliGut®
Liver only
A
B C Metabolism of 7-hydroxycoumarin
7-hydroxycoumarin
(7-HC)
7-hydroxycoumarin
glucuronide
Figure 5. Case study 1, Temocapril. Resistance of Temocapril to intestinal
clearance observed in the primary cell Gut/Liver MPS, correlated with
isoenzyme expression in the human intestine.
Temocapril was added as either an oral dose (100 μM) in the apical side of
the Transwell or as an IV dose (10 μM) and directly mixed with circulating
media in the liver compartment. Two configurations of the Gut/Liver MPS
were studied: Caco-2/Liver and RepliGut/Liver. Temocapril is A) primarily
metabolized by carboxylesterase (CES) 1 isoenzyme into the active
metabolite Temocaprilat. Pattern of CES1 isoenzyme B) RNA and C) Protein
expressions in human liver, human small intestine, and the Caco-2 cell
line3. D) Pattern of CES gene expression in the Jejunum model over time
in expansion media (EM) and differentiation media (DM); data courtesy
of Scott Magness, UNC Chapel Hill E) Concentration of Temocapril in the
circulating media over time in the Gut/Liver models, measured by LC-MS. F)
Concentration of Temocapril over time in the liver only model. G) Total molar
fraction of Temocapril remaining across all compartments in each model
at 48 hours. H) Relative amount of the metabolite Temocaprilat in the gut
apical compartment in both Gut/Liver models at 0 and 48 hours after drug
dosing. Peak area is a qualitative measure of compound concentration.
A
B C D
Temocapril
Pattern of
isoenzyme
expression3 (RNA)
Pattern of
isoenzyme
expression3
(Protein)
CES expression
pattern in
jejunum model
Temocaprilat
E
G
F
H
Oral, Gut/Liver,
100 μM dose
Temocapril at 48 hrs in all
compartments
IV, liver only, 10 μM dose
Temocaprilat at 0 and 48 hrs
in Gut Apical compartment
Figure 6. Case study 2, Midazolam. Improved correlation with human
bioavailability of Midazolam by the primary cell Gut/Liver MPS.
Midazolam was added as either an oral dose (50 μM) in the apical side of
the Transwell, within the gut compartment, or as an IV dose (5 μM) directly
mixed with circulating media in the liver compartment. Two configurations
of the Gut/Liver MPS were studied: Caco-2/Liver and RepliGut/Liver. Media
samples were taken over 48 hours and quantified by LC-MS to determine
the A) concentration of Midazolam in the circulating media following
different dosing regimens and the B) total molar fraction of Midazolam
remaining across all compartments in each model at 48 hours. C) We
estimated oral bioavailability with both Gut/Liver MPS models by taking
the ratio of the area under the curves and dose and compared with human
and animal data in published literature1.
A
C
Oral, Gut/Liver, 100 μM dose B
Estimation of bioavailability
compared to human and animal data
Midazolam at 48 hrs in all
compartments
Conclusion
This study demonstrated that the primary RepliGut/Liver MPS more accurately recapitulates the physiological conditions of oral drug dosing in the human. By combining intestinal absorption and hepatic metabolism, the primary RepliGut/Liver MPS generates in vivo-like drug concentrations that cannot be replicated using standard preclinical in vitro models. The primary RepliGut/Liver MPS offers a more accurate method to study the pharmacokinetics of pro-drugs that undergo CES metabolism compared to an equivalent Caco-2 Gut/Liver MPS, which failed to profile their human bioavailability. The results demonstrate that the primary MPS model offers a viable alternative to circumvent the human-relevance limitations of the Caco-2 cell line for this drug type. By generating more human-relevant data earlier in drug discovery, means that observed issues can be flagged and addressed before costly preclinical in vivo studies.
As the primary RepliGut/Liver MPS is entirely made up of primary human cells, there are no interspecies differences to account for, therefore, the model can be utilized to help overcome the poor correlation between animal model ADME predictions and human outcomes. Bridging the gap between in vitro assays and in vivo studies, the model enables researchers to confidently progress only the most promising drug candidates to support reductions in cost and the number of animals required.
References
1.
Musther, H., Olivares-Morales, A., Hatley, O. J. D., Liu, B. & Hodjegan, A. R. Animal versus human oral drug bioavailability: Do they correlate? European Journal of Pharmaceutical Sciences 57, 280 (2014).
2.
Edington, C. D. et al. Interconnected Microphysiological Systems for Quantitative Biology and Pharmacology Studies. Sci Rep 8, (2018).
3.
Imai, T., Imoto, M., Sakamoto, H. & Hashimoto, M. Identification of esterases expressed in Caco-2 cells and effects of their hydrolyzing activity in predicting human intestinal absorption. Drug Metab Dispos 33, 1185–1190 (2005).
4.
Paine, M. F. et al. First-pass metabolism of midazolam by the human intestine. Clin Pharmacol Ther 60, 14–24 (1996).
Notes
Notes
+44 (0) 1223 737941 | enquiries@cn-bio.com
visit cn-bio.com
332 Cambridge Science Park
Milton Road, Cambridge, CB4 0WN
visit cn-bio.com
Connecting the gut and liver: a human relevant multi-organ microphysiological system for preclinical profiling of oral bioavailability
Yassen Abbas1, Hailey Sze1, Ashley A. Spreen2, Elizabeth M. Boazak2, William R. Thelin2 and Tomasz Kostrzewski1
1. CN Bio Innovations. 2. Altis Biosystems
cn-bio.com/208
APPLICATION NOTE
CN Bio’s organ-on-chip systems, which include the PhysioMimix® Single- and Multi-organ lab-benchtop instruments, enable researchers to model human biology in the lab through rapid and predictive human tissue-based studies.
The technology bridges the gap between traditional cell culture and human studies, advancing towards the simulation of human biological conditions to support the accelerated development of new therapeutics in application areas including oncology, infectious diseases, metabolism and inflammation.
Learn more at cn-bio.com
Introduction
Absorption, distribution, metabolism, and excretion (ADME) are four key processes that indicate the behavior of a drug following administration, and therefore play a key role in defining a compound’s pharmacokinetic (PK) properties and bioavailability. Oral bioavailability is defined as the fraction of a drug that reaches systemic circulation following absorption across the intestinal wall and first pass metabolism in the liver. ADME and bioavailability are central in determining the safety and toxicology profiles of compounds and are therefore crucial measurements at the preclinical stage of drug development.
Currently, a combination of simple in vitro assays that either model the gut (Caco-2 cell line) or the liver (liver microsomes and suspension hepatocytes), and in vivo animal models is used to profile oral bioavailability; however, significant limitations exist with both approaches. Caco-2 cells, which have been the workhorse for assessing in vitro intestinal permeability, cannot account for liver metabolism, plus the cell line has absent or low levels of enzyme and transporter expression. Liver microsomes and suspension hepatocytes are used for in vitro drug metabolism screening studies, but do not consider intestinal absorption. Collectively, these limit the accuracy of their estimations. Furthermore, in a seminal study investigating 184 compounds, animal models were found to have a weak correlation with bioavailability in humans (R2=0.34)1. A new human relevant approach, which combines oral absorption and hepatic metabolism, is therefore required to more accurately estimate drug bioavailability.
In the last decade, microphysiological systems (MPS), also known as organ-on-a-chip (OOC), have shown their potential to improve the human translatability of ADME studies. They are designed to recapitulate the structural and functional biomarkers of cells and tissues in a more physiologically relevant manner through the culture of primary human cells on perfused 3D scaffolds. Efforts to improve the in vitro to in vivo translation of drug efficacy and safety data has led to the emergence of more complex MPS where multiple organs, such as gut and liver, are fluidically linked together to simulate processes such as drug absorption and first pass metabolism2.
Aim
Here, we introduce a dual-organ MPS that links our established primary human liver MPS with a primary model of the human intestine. For the intestinal barrier, primary cells are isolated from the human jejunum and expanded on a biomimetic scaffold (RepliGut®). To link the primary gut and liver tissues together, we developed a chemically defined media that supports both organ models in the dual-organ MPS. This media enables both the maintenance of hepatic metabolic functionality and intestinal barrier integrity. Using well-studied drug compounds, we aimed to demonstrate the improved predictive capacity of this primary Gut/Liver MPS for profiling the ADME behavior of oral drugs compared to an equivalent Caco-2 Gut/Liver MPS.
Methods
Here, we describe a dual-organ MPS that links a primary human intestine (RepliGut® Planar-Jejunum, Altis Biosystems) with a primary human liver-on-a-chip (CN Bio). The dual-organ MPS is cultured using the PhysioMimix® Multi-organ System and its bespoke “Multi-chip” Dual-organ consumable plate (Figure 1A). The dual-organ plate consists of six wells, each with two compartments (i) a Transwell® compartment and (ii) a liver compartment. Fluidic flow can be independently controlled in each compartment (intestine and liver), and in the interconnecting channel between the organs (Figure 4A). The intestinal barrier was established through the expansion of human jejunum stem/progenitor cells on a Transwell® coated with a biomimetic scaffold, followed by differentiation into a polarized barrier comprised of all post-mitotic lineages found in the human intestine. Media was changed every 48 hours with either RepliGut® Growth or Maturation media. Intestinal barrier integrity was assessed with transepithelial electrical resistance (TEER) measurements every 48 hours or by determining the permeability of Lucifer yellow across the Transwell at the end of the dual-organ experiment. Expression of markers confirming intestinal origin (Villin and CDX2) and the presence of the tight-junction marker ZO-1 was confirmed by fluorescent microscopy. Mucus production was confirmed by histological staining of the RepliGut® Planar-Jejunum
cross-section with Alcian blue and by in-well immunofluorescent staining for the Muc-2 protein. Expression of metabolic and transporter genes were evaluated using RNA isolated at day 15 of Caco-2 culture or day 7 post differentiation of RepliGut® Planar-jejunum. Gene expressions of metabolic enzymes and transporters were measured by qPCR using TaqManTM Gene Expression Assays, with relative expression determined using ΔΔCT analysis.
For the liver, primary human hepatocytes (PHH) were seeded in the liver compartment of the PhysioMimix Dual-organ plate, on a porous 3D collagen-coated scaffold (Figure 1C). On day 4, following PHH seeding and the formation of microtissues, differentiated RepliGut cultures were added to the PhysioMimix Dual-organ plate to establish the Gut/Liver co-culture. A chemically defined media was used to maintain functionality of the gut and liver tissues for at least 48 hours, during which compounds were added to study their ADME profile. In this study, we compared the performance of the primary RepliGut/liver MPS versus a Caco-2 Gut/liver MPS. The Caco-2 Gut/liver MPS was established by adding Transwells with differentiated Caco-2 monolayers (at 15-17 days post seeding) to the PhysioMimix Dual-organ plate on day 4, following PHH seeding and the formation of microtissues.
We used 7-hydroxycoumarin (7-HC), a fluorescent compound that undergoes Phase II metabolism by glucuronidation, in a proof-of-concept study to demonstrate absorption through the intestinal barrier and subsequent first pass metabolism in the co-culture model. The PhysioMimix Dual-organ plate allows for flexibility in compound dosing with either Gut/Liver in co-culture or gut only (no PHH) and liver only (no gut barrier, with compounds dosed into a blank Transwell with no cells).
In this study, two compounds (Temocapril and Midazolam) who’s human ADME properties were not predicted by existing models, were investigated. Compounds were added at day 4 post seeding of PHH and addition of the gut tissues to Dual-organ plates by either oral (drug added to the apical surface of the Transwell) or intravenous (IV) dosing (liver only, drug mixed into co-culture media). Samples of media were taken at 0, 1, 4, 6, 24, and 48 hours, and analyzed by liquid chromatography–mass spectrometry (LC-MS) to determine the concentration of parent compounds in the liver compartment. An estimation of area under the curve (AUC) of both oral and IV concentration profiles was made using GraphPad Prism.
Results and Discussion
Using the PhysioMimix Multi-organ System and its Multi-chip Dual-organ plate (Figure 1), we established a primary Gut/Liver MPS to overcome the human-relevance limitations of current models in profiling oral drug bioavailability. In the RepliGut model (Figure 2A), jejunum stem and progenitor cells were expanded to confluence on a biomimetic scaffold before undergoing differentiation, which resulted in increasing barrier strength (Figure 2B). Gene expression confirmed downregulation of proliferative cell genes and upregulation of differentiated enterocyte genes relative to cells in the proliferative phase (Figure 2B). In the differentiation phase, the RepliGut model expressed tissue-specific markers confirming intestinal origin as well as markers of tight junctions and mucins, confirming multi-cellular lineages (Figure 2C). We confirmed this via histology with observations of a distinct and continuous layer of mucus at day 7 post-differentiation. By comparison, mucus production was absent in the epithelial cell line, Caco-2 model (Figure 2D). To assess the RepliGut model’s potential to improve predictivity over the Caco-2 model, we investigated the expression of major metabolic enzyme and transporter genes that are important in drug metabolism and active drug transport. Gene expression was found to be improved compared to the Caco-2 cell line (Figure 2E).
One of the challenges with the co-culture of two, or more, tissues together is establishing the conditions that maintain functionality of both tissue types. We designed a chemically defined media, which is added to the Transwell basolateral and liver compartments of the Dual-organ plate (Figure 3A). This media maintained hepatic cell health and functionality in co-culture for at least 48 hours, measured by LDH release, CYP3A4 activity and albumin production (Figure 3 B-D). Intestinal barrier integrity was also maintained for the 48 hours of co-culture, as measured by TEER (Figure 3E) and a Lucifer yellow permeability assay (Figure 3F), after the addition of the RepliGut model to the dual-organ plate. Functionality was further demonstrated by 7-HC dosing (Figure 4A) which undergoes metabolism through glucuronidation (Figure 4B), with both intestinal and hepatic tissues contributing to its clearance (Figure 4C).
To demonstrate the improved predictive capability of the primary RepliGut/Liver MPS versus an equivalent Caco-2 Gut/Liver MPS, we investigated the ADME properties of two drugs where both animal models and the Caco-2 cell line failed to accurately profile their ADME behavior and bioavailability in human. The first drug, Temocapril, is a pro-drug designed to be resistant to intestinal hydrolysis that is metabolized to Temocaprilat by carboxylesterase 1 (CES1) (Figure 5A). The isoenzyme pattern in human is well-characterized with the liver and intestine expressing CES1 and CES2 respectively3. In Caco-2 cells, there is a miss-match as CES1 is predominantly expressed, resulting in an overestimation of drug clearance (Figure 5B-C). In contrast, the more human representative ratios of both CES1 & CES2 expression by the RepliGut model make it more relevant for pro-drug studies (Figure 5G). Temocapril clearance was profiled with oral and IV dosing (Figure 5E). Rapid clearance was observed by the liver within 24 hours following IV dosing (Figure 5F). In line with intestinal CES expression, the primary RepliGut/Liver MPS correctly reported resistance to intestinal hydrolysis of Temocapril and less Temopcarilat was produced at 48 hours in the gut apical compartment, following oral dosing.
Finally, we assessed the bioavailability of Midazolam in both Gut/Liver MPS models, a drug known to predominately undergo intestinal clearance by the CYP3A4 enzyme4. Midazolam was rapidly cleared by the liver following IV dosing, reflecting its high intrinsic hepatic clearance rate (19 mL/min/Kg) (Figure 6A). We observed greater clearance of Midazolam by the primary RepliGut/Liver MPS, compared to the Caco-2 Gut/Liver MPS (Figure 6B). The primary RepliGut/Liver MPS also delivered an oral bioavailability estimation that more closely represented human clinical observations (Figure 6C).
Figure 1. Set-up of a standard Gut/Liver MPS in the PhysioMimix Dualorgan
plate.
A) Set-up of the PhysioMimix Multi-organ System and its Multi-chip Dualorgan
plate, showcasing the location of the gut and liver compartments
within the plate. Each plate can culture up to six Gut/Liver models. B)
Schematic diagram of a RepliGut® model cultured on a biomimetic
scaffold within a Transwell insert, forming an intestinal epithelial barrier. C)
Schematic representation of fluidic flow within the liver compartment of
the Dual-organ plate.
B
C
Epithelial
monolayer
Luminal
reservoir
Biomimetic
scaffold
Cassette
Basal
reservoir
Porous
membrane
Liver
microtissue
3D scaffold
Micropump
B
C
A
C
Figure 2. RepliGut primary model of the intestine is more human
relevant compared to the standard Caco-2 cell line model.
A) Crypt epithelium stem/ progenitor cells are isolated from jejunum
samples and expanded on a biomimetic scaffold in static conditions.
B) Barrier integrity measured by TEER and relative gene expressions,
in the expansion and differentiation stages of the RepliGut model. C)
Immunofluorescence images of the RepliGut model. The barrier is stained
for DAPI (dark blue), the tight junction protein, ZO-1 (light blue), and
other intestinal markers including MUC2 (yellow), villin (purple) and CDX2
(green). The image scale bars equate to 100 μm for CDX2 and 200 μm for
the other three markers. D) Histology sections of the RepliGut and Caco-2
models. Monolayer cross-sections are stained with hematoxylin and eosin
to visualize nuclei and with Alcian blue to visualize the mucus layer. E)
qPCR to demonstrate relative gene expressions of key transporters and
metabolic enzymes in the RepliGut and Caco-2 models.
A
B
-3 -2 -1 0 1 2 3 4 5 6 7 8 9 10
0
500
1000
1500
2000
Days of Differentiation Corrected TEER (Ω*cm2)
Cell Expansion
4-6 days
Mostly Stem & Progenitor Cells
Cell Differentiation
6-8 days
Mostly Differentiated Cells
MKI67 LGR5 SOX9 SI ALPI ANPEPFABP6
0.001
0.01
0.1
1
10
100
1000
Relative Gene Expression
Proliferative Cells Differentiated Cells
Proliferative Cell Genes Differentiated Cell Genes
D
E
RepliGut®
Caco-2 only
C
Figure 3. Liver and gut functionality markers are maintained
throughout the primary cell Gut/Liver co-culture.
A) Experiment timeline to establish the primary RepliGut/Liver MPS.
RepliGut was first cultured independently, in static, for 13 days before
being transferred to the Transwell compartment of the PhysioMimix Dualorgan
plate, 4 days after PHH seeding. The co-culture was maintained for
48 hours in a chemically defined media. Markers of liver cell health and
functionality were measured at 0, 24 and 48 hours of co-culture, with a
liver only MPS cultured in parallel as a control. We investigated B) LDH
release C) CYP3A4 activity and D) Albumin production. Barrier integrity of
RepliGut in the Gut/Liver MPS was assessed by E) TEER and F) permeability
to Lucifer yellow and compared to independent RepliGut cultured in static,
for its full duration in standard RepliGut® Maturation media.
A
B C
Day 0 13 14 15
9
PHH seeded onto
dual organ plate
Jejunum
cells seeded
Jejunum transferred to dual organ
plate, co-culture begins
QC
Drug dosed
Experiment
endpoints
Media sampling
Media sampling at
0, 1, 2, 4, 6, 24, 48 hours
Media samples sent for LC-MS for
drug quantification
LDH CYP3A4 Metabolism
Liver only Gut/Liver, RepliGut®
D
F
Albumin E Barrier Integrity
Permeability to Lucifer Yellow
Liver only
Gut/Liver, RepliGut®
Static, RepliGut®
Gut/Liver, RepliGut®
Figure 4. First pass metabolism can be modeled by the primary cell
Gut/Liver MPS.
A) Schematic representation of 7-HC dosing in the Gut/Liver MPS model.
1mM 7-HC was dosed in the apical side of the Transwell compartment,
containing the RepliGut barrier, within the Dual-organ plate. 7-HC was
then transported across the barrier into the media circulating between
the gut and liver tissues. The gut only control was prepared by inserting
the RepliGut model into the Dual-organ plate Transwell compartment
that was connected to a blank liver compartment (without PHH).
Likewise, the liver only control was prepared by dosing 7-HC into a blank
Transwell insert (without gut cells), in the Transwell compartment of
the Dual-organ plate. B) Pathway of 7-HC metabolism. 7-HC undergoes
phase II metabolism by UGT enzymes to form the non-fluorescent
metabolite 7-hydroxycoumarin glucuronide. C) Changes in 7-HC
concentration in the circulating media over time.
Gut only, RepliGut®
Gut/Liver, RepliGut®
Liver only
A
B C Metabolism of 7-hydroxycoumarin
7-hydroxycoumarin
(7-HC)
7-hydroxycoumarin
glucuronide
Figure 5. Case study 1, Temocapril. Resistance of Temocapril to intestinal
clearance observed in the primary cell Gut/Liver MPS, correlated with
isoenzyme expression in the human intestine.
Temocapril was added as either an oral dose (100 μM) in the apical side of
the Transwell or as an IV dose (10 μM) and directly mixed with circulating
media in the liver compartment. Two configurations of the Gut/Liver MPS
were studied: Caco-2/Liver and RepliGut/Liver. Temocapril is A) primarily
metabolized by carboxylesterase (CES) 1 isoenzyme into the active
metabolite Temocaprilat. Pattern of CES1 isoenzyme B) RNA and C) Protein
expressions in human liver, human small intestine, and the Caco-2 cell
line3. D) Pattern of CES gene expression in the Jejunum model over time
in expansion media (EM) and differentiation media (DM); data courtesy
of Scott Magness, UNC Chapel Hill E) Concentration of Temocapril in the
circulating media over time in the Gut/Liver models, measured by LC-MS. F)
Concentration of Temocapril over time in the liver only model. G) Total molar
fraction of Temocapril remaining across all compartments in each model
at 48 hours. H) Relative amount of the metabolite Temocaprilat in the gut
apical compartment in both Gut/Liver models at 0 and 48 hours after drug
dosing. Peak area is a qualitative measure of compound concentration.
A
B C D
Temocapril
Pattern of
isoenzyme
expression3 (RNA)
Pattern of
isoenzyme
expression3
(Protein)
CES expression
pattern in
jejunum model
Temocaprilat
E
G
F
H
Oral, Gut/Liver,
100 μM dose
Temocapril at 48 hrs in all
compartments
IV, liver only, 10 μM dose
Temocaprilat at 0 and 48 hrs
in Gut Apical compartment
Figure 6. Case study 2, Midazolam. Improved correlation with human
bioavailability of Midazolam by the primary cell Gut/Liver MPS.
Midazolam was added as either an oral dose (50 μM) in the apical side of
the Transwell, within the gut compartment, or as an IV dose (5 μM) directly
mixed with circulating media in the liver compartment. Two configurations
of the Gut/Liver MPS were studied: Caco-2/Liver and RepliGut/Liver. Media
samples were taken over 48 hours and quantified by LC-MS to determine
the A) concentration of Midazolam in the circulating media following
different dosing regimens and the B) total molar fraction of Midazolam
remaining across all compartments in each model at 48 hours. C) We
estimated oral bioavailability with both Gut/Liver MPS models by taking
the ratio of the area under the curves and dose and compared with human
and animal data in published literature1.
A
C
Oral, Gut/Liver, 100 μM dose B
Estimation of bioavailability
compared to human and animal data
Midazolam at 48 hrs in all
compartments
Conclusion
This study demonstrated that the primary RepliGut/Liver MPS more accurately recapitulates the physiological conditions of oral drug dosing in the human. By combining intestinal absorption and hepatic metabolism, the primary RepliGut/Liver MPS generates in vivo-like drug concentrations that cannot be replicated using standard preclinical in vitro models. The primary RepliGut/Liver MPS offers a more accurate method to study the pharmacokinetics of pro-drugs that undergo CES metabolism compared to an equivalent Caco-2 Gut/Liver MPS, which failed to profile their human bioavailability. The results demonstrate that the primary MPS model offers a viable alternative to circumvent the human-relevance limitations of the Caco-2 cell line for this drug type. By generating more human-relevant data earlier in drug discovery, means that observed issues can be flagged and addressed before costly preclinical in vivo studies.
As the primary RepliGut/Liver MPS is entirely made up of primary human cells, there are no interspecies differences to account for, therefore, the model can be utilized to help overcome the poor correlation between animal model ADME predictions and human outcomes. Bridging the gap between in vitro assays and in vivo studies, the model enables researchers to confidently progress only the most promising drug candidates to support reductions in cost and the number of animals required.
References
1.
Musther, H., Olivares-Morales, A., Hatley, O. J. D., Liu, B. & Hodjegan, A. R. Animal versus human oral drug bioavailability: Do they correlate? European Journal of Pharmaceutical Sciences 57, 280 (2014).
2.
Edington, C. D. et al. Interconnected Microphysiological Systems for Quantitative Biology and Pharmacology Studies. Sci Rep 8, (2018).
3.
Imai, T., Imoto, M., Sakamoto, H. & Hashimoto, M. Identification of esterases expressed in Caco-2 cells and effects of their hydrolyzing activity in predicting human intestinal absorption. Drug Metab Dispos 33, 1185–1190 (2005).
4.
Paine, M. F. et al. First-pass metabolism of midazolam by the human intestine. Clin Pharmacol Ther 60, 14–24 (1996).
Notes
Notes
+44 (0) 1223 737941 | enquiries@cn-bio.com
visit cn-bio.com
332 Cambridge Science Park
Milton Road, Cambridge, CB4 0WN
