IntroductionThe majority of the preclinical intestinal screening modelsdo not properly reflect the complex physiology of thehuman intestinal tract, resulting in low translational valueto the clinical situation. A major drawback of the currentintestinal models is the use of single cell lines and thestatic environment, which is in contrast with the dynamicprocesses in vivo.

GoalWe aim to develop a physiological in vitro humanintestinal model that can be used to study (drug)absorption and impact of drugs, nutrition and microbialenvironment on gut health. The ultimate goal is todevelop a population-on-a-chip model to aid in thedevelopment of precision medicines by targeting patientvariability using intestinal tissue and microbiota fromvarious individuals reflecting populational variation.

Approach

TNO has developed two ex vivo predictive models, theInTESTine-on-a-chip model in which chronic drugexposure effects can be studied using a (dual flow)microfluidic device, and the I-screen platform which is ascreening multi-well platform simulating the humancolonic microbiota conditions to study microbiota inducedmetabolic transformation of drugs using pooled humancolonic microbiota under fully anaerobic conditions.

Results

First steps in development of the gut on-a-chip model:• Sustained functionality of human intestinal tissue

mounted within microfluidic model (Fig. 1)• Representative functions of ex vivo human intestinal

tissue when mounted in gut on-a-chip model (Fig. 2)

Demonstrating the microbiota-based drug metabolismusing the I-Screen model• Among the 12 drugs examined, five were

demonstrated to generate metabolites. Allmetabolic transformations were known to occur inhuman in vivo (Fig. 3)

• Able to culture fecal samples and showingdifferences in microbiota composition of differentage groups (Fig. 3)

Human ex vivo model to study

intestinal processes and

microbiome induced metabolism

Lianne Stevens1

Irene Nooijen1

Steven Erpelinck1

Scott Obach2

Gregory Walker2

Frank Schuren1

Angelique Speulman1

Mariska Gröllers1

Wouter Vaes1

Evita van de Steeg1

1TNO, Microbiology & Systems Biology, The Netherlands2 Pfizer, Inc., Groton, CT, USA

E: Lianne.stevens@tno.nl

Conclusions / next steps:

• We have successfully developed an in vitro human intestinal model, based on human intestinal, which can be applied as a reliable tool for longer-term incubations

• We here present an in vitro tool (I-screen) containing human colonic microbiota to study microbiota driven metabolism of drugs. We identified different types of drug transformations that may occur in the human gut and are known to occur in vivo.

• Next, the two platforms can be combined in order to create a physiological representation of human intestinal processes including intestinal permeability as well as (anaerobic) host-microbe-immune responses.

I

IV

Sustained functionality of human intestinal tissue in microfluidic model

Representative functions of human intestinal tissue

A

J e ju n u m C o lo n

0

2 5

5 0

7 5

1 0 0

1 2 5

1 5 0

T o ta l in tra c e llu la r L D H

a fte r 2 8 h in c u b a t io n

% I

ntr

ac

ell

ula

r L

DH

(%

co

ntr

ol)

S ta tic

t = 0

M ic ro f lu id ic s

B

4 5 1 0 5 1 6 5 2 2 5 2 4 h o u r s0

1

2

3

4

5

S ta tic m o d e l

M ic ro f lu id ic s

0 .5

T im e (m in )

% F

D4

le

ak

ag

e

% F D 4 le a k a g e

2 4 h o f in c u b a t io n

C

F G H

Metabolism of drugs by intestinal microbiotasulfasalazine

Active metabolite (sulfapyridine)(>90% of total)

Sulfasalazine (6% of total)

CompoundMolecular

TransformationFormed Metabolite

Risperidone N-O reduction

Imine intermediate

Reduced metabolite,

complete conversion

SulindacS-oxide

reduction

Deoxysulindac, complete

conversion

SulfinpyrazoneS-oxide

reductionDeoxysulfinpyrazone

Nizatidine N-O reduction Deoxynizatidine

Sulfasalazine ReductionReductively cleaved

metabolite of sulfasalazine

MetronidazoleDecline of

parentNo metabolites detectable

Omeprazole

No decline of parent, no metabolites

detectable

Simvastatin

Levodopa

Acetaminophen

Zonisamide

Dapsone

Fig 1. Improved functionality of human intestinal tissue. A) Intracellular LDH content showing the addition of an apical and basolateral flow significantly extends the viability of the tissue. B) Integrity of intestinal barrier determined by the leakage of FD4 in each incubation. Data represent the mean ± SD (n=4). C) Viability of human intestinal tissue (ileum) before (A) and after (B) 24h incubation in microfluidic gut on-a-a chip model shown by H&E staining.

Fig 2. Demonstrating representative functions of human intestinal tissue in vitro gut on-a-chip model. A,B) Metabolic activity of human intestinal tissue (ileum) as demonstrated by linear rate of testosterone metabolism. C) Cytokine response of human intestinal tissue under static control and stimulated with flagellin (5h) and microfluidic conditions of control and after stimulation with flagellin for 24h) D) Ileum tissue, stimulated with LPS showing increase in IL-6 response and the suppressing effect of different probiotics (n=4) All data represent the mean ± SD (n=4

Fig 3. Demonstrating the microbiota-based drug metabolism A) List of 12 investigated drugs potentially susceptible to microbial metabolism in I-screen B) Sulfasalazine as a quality reference drug in all experiments to screen for anaerobic metabolism. C)Sulfasalazine is metabolized into 5-ASA and sulfapyridine by azoreductases of gut microbiota D) showing populational variation in microbiota composition between newborns, infants, toddlers, teenagers and adults

C o n tro l + F la g e l l in C o n tro l + F la g e l l in

0

4

8

2 0

3 0

4 0

IL-8

co

nc

en

tra

tio

n

(pg

/mg

tis

su

e)

5 h s ta t ic in c u b a t io n 2 4 h f lu id ic in c u b a t io n

H u m a n Ile u m tis s u eB C

A B C

D

D

Visit us at booth #306

P114

Bla

nk

LP

S

Pro

bio

t ic

1

Pro

bio

t ic

2

Pro

bio

t ic

3

Pro

bio

t ic

4

0

10

20

30

40

50

60

70

IL-6

co

nce

ntr

at i

on

(pg

/mg

t is

su

e)

Ileum tissue

6h static incubation