Patent Application
20130302888
This invention focuses on the development of a coculture model, based on Caco-2 absorption cells and HT29 goblet cells, for testing the transport of active substances and nutrients in the intestine. The purpose of the proposed coculture model is to create cell culture models that are as realistic as possible and which can be adapted to the specific properties of different parts of the human intestinal epithelium.
The Caco-2 cell line was isolated from a human colorectal cancer. Caco-2 cells differentiate spontaneously into well-developed polarised monolayers of columnar absorption cells expressing a brush border with typical enzymes (e.g. alkaline phosphatase, sucrase-isomaltase, aminopeptidase) on their apical surface. Active transporters for amino acids, nucleosides, bile acid, vitamins, oligopeptides and monocarboxylic acids are expressed on the apical side. The cells in these monolayers are joined by tight intercellular junctions that restrict paracellular passage of drug molecules and ions.1,2 However, tight bonding of the enterocytes in these monolayers is more like that of the colon than of the small intestine, which results in paracellular permeability in these monolayers which is too low for hydrophilic molecules.13 In addition, P-glycoprotein or P-gp (encoded by the multidrug resistance gene-1) is strongly expressed in Caco-2 cells.
The HT29 cell line is derived from a human colon adenocarcinoma. Post-confluent cultures of HT29 cells form a heterogeneous multilayer in which the majority of cells are undifferentiated. A subpopulation of mucus-producing goblet cells (HT29-MTX) has been obtained from the parental HT29 cell line by gradual adaptation to methotrexate (MTX). Several passages of HT29 cells in exponential growth were incubated with increasing amounts of MTX (10−7, 10−6, 10−5 mol). This results initially in high mortality, but leads to subpopulations with stable growth rates. These cells do not need to be maintained in a medium containing MTX to differentiate into mucus-producing goblet cells after confluence.3,4,5 The thickness of the mucus layer which spreads over the whole apical surface is on average 50 to 150 μm. HT29-MTX cells at passage 13 are available for example from Thecla Lesuffleur of INSERM U 938.
Cultures of Caco-2 cells have been extensively studied to predict the absorption of drugs from the gastrointestinal tract, despite less than optimal conditions. Indeed, the P-gp activity in Caco-2 monolayers in standard culture conditions is much higher than in the human colon in vivo.6,7 In addition, TEER (Trans Epithelial Electrical Resistance) values are very high, greater than 250 ohm.cm2, in comparison with the human intestine (12 to 69 ohm.cm2),7 which results in paracellular permeability of hydrophilic active substances that is too low. Furthermore, the role of the mucus layer naturally occurring on the luminal side of the intestinal barrier is not taken into account in evaluating the permeability of active substances.8,9,10
Proposals have been made to modify Caco-2 cells in an attempt to get closer to the in vivo situation. The document EP 1 158 045 describes a strain derived from Caco-2 with increased expression of cytochrome CYP3A4.
Furthermore, it has been proposed that cocultures could be made combining absorption cells or enterocytes and goblet cells. The first studies were by Wikman-Larhed and Artursson11 on cocultures of Caco-2/HT29-H, using a mixture of different proportions of the cells at the time of seeding. Walter et al.12 (Caco-2/HT29-MTX) and Meaney et al.9 (Caco-2/HT29GlucH) used only one ratio. Hilgendorf et al.13 selected three different Caco-2/HT29-MTX cell ratios, in single factor experiments. A coculture without serum in a single ratio of Caco-2/HT29-5M21 cells was developed by Nollevaux et al.14 Poquet et al.15 developed a coculture model with a single Caco-2/HT29-MTX cell ratio to analyse the transport of ferulic acid. The latest studies have used different Caco-2/HT29-MTX cell ratios to predict iron bioavailability.16 Chen et al., in recent work in 2010,18 studied various factors affecting the permeability of active substances in Caco-2/HT29-MTX coculture models. The authors showed that the length of culture and the culture medium have a considerable effect on TEER values and permeability coefficients. However, in all these studies, seeding with HT29-MTX cells was carried out simultaneously with seeding with Caco-2 cells.
There is therefore an obvious need to develop new in vitro intestinal barrier systems which are both more flexible and more realistic.
In the context of the invention, the Applicant has demonstrated that it is possible to modify the characteristics of enterocytes, particularly Caco-2 cells, by varying the time of seeding with goblet cells. It is thus possible, in an ingenious way, to control the important parameters of the cell culture obtained, particularly the transport and metabolism of the compounds, to obtain an in vitro system as close as possible to the real conditions of the intestinal barrier in vivo.
More specifically, the cell culture process according to the invention consists of:
Characteristically, and unlike the prior art, the seeding of the two cell types according to the invention is not, therefore, simultaneous.
Enterocytes, also called absorption cells or brush border cells, are characterised by their ability to exhibit properties of the small intestine in vitro, particularly in terms of transport for biopharmaceutical studies and analyses. The enterocytes are preferably Caco-2 cells. They may, for example, be the CBBel clone at passage 47, with the American Type Culture Collection (ATCC) reference number CRL-2102, which is itself derived from a clone of Caco-2, ATCC reference number HTB-37. In practice, the latter can be used at a passage between passage 55 and passage 65.
Goblet cells, also called mucus-producing cells, are characterised by their ability to produce mucus. As far as the goblet cells are concerned, to advantage they are HT29-MTX cells. The process for obtaining them has been described in publications by Lesuffleur et a/.3,4,5 They can, for example, be the cell line HT29-MTX10-6 M at passage 13, available from Dr Theela Lesuffleur of INSERM UMR S 938 in Paris, France. This line can be used preferably at a passage between passage 14 and passage 22. Alternatively, they can be E12 and D1 clones described in the publication by Behrens et al. (Behrens I, Stenberg P., Artursson P., Kissel T. (2001) Transport of lipophilic drug molecules in a new mucus-secreting cell culture model based on HT29-MTX cells. Pharmaceutical Research 18, 1138-1145) or HT29GlucH described in the publication by Meaney et al.9
In the context of this application, it has been shown that it is possible to modify the characteristics of the cell culture obtained by varying the timing of seeding the two cell types involved.
Unlike the prior art which advocated simultaneous seeding of the two cell types, the method according to the invention proposes seeding the two types of cell separately, in other words, at different times.
More specifically, the culture medium concerned is first seeded with enterocytes, preferably Caco-2 cells.
The culture medium is therefore advantageously suited for optimal growth of the enterocytes. In the case of Caco-2 cells, and as already described in the prior art, the medium is to advantage a complete nutrient medium such as DMEM (Dulbecco's Modified Eagle Medium). To even greater advantage, a growth factor is added to it, such as heat-inactivated foetal bovine serum (FBS), to advantage at a concentration of 15% (v/v).
The culture medium may also contain:
Cells are cultured to advantage at 37° C. in a moist atmosphere at 5% CO2. The culture medium is to advantage changed twice a week, or each day at the end of culture.
The total cell culture time, for the invention and classically, is to advantage between 21 and 35 days after seeding with enterocytes, to greater advantage between 21 and 30 days.
The second type of cells, i.e. the goblet cells, is not therefore seeded simultaneously, but on the contrary at a later time relative to the first cell type, i.e. the enterocytes. In practice, the goblet cells are seeded after the enterocytes.
More precisely and advantageously, this should not be until after the enterocytes have begun their cycle of cell division. In other words, the enterocytes must have started to proliferate. To even greater advantage, in order to see a difference from the coseeding, the second cell type is seeded at least one day (24 hours) after the first cell type.
As has already been said, the goblet cells are seeded in the medium containing the enterocytes which have started to proliferate. Seeding is preferably performed when the enterocytes form a monolayer. Nevertheless, seeding the second cell type should not be too late, either, and should lead, in particular, to mucus production.
In practice, the goblet cells are to advantage seeded on the surface of the enterocytes. To even greater advantage, the goblet cells are seeded directly onto the monolayer of enterocytes.
The time needed for the monolayer of enterocytes to form depends in practice on the conditions of enterocyte culture, particularly on the initial enterocyte seeding density. To modify the properties of the coculture obtained further, it is possible to adjust both the enterocyte seeding density and the time of seeding the goblet cells. Other factors may influence the enterocyte growth rate, including the growth medium and the culture temperature.
Under the conditions of this application, it was observed that it was particularly suitable to seed goblet cells between 1 and 8 days after seeding the enterocytes, preferably between 2 and 6 days, or even between 2 and 3 days. The coculture obtained, which produces mucus, has characteristics that are thus intermediate between the two cell types, particularly in terms of the transport and metabolism of compounds, and therefore tends to reflect the real in vivo conditions.
Moreover, the two cell types are seeded in a controlled manner. For this application, the following culture conditions proved to be particularly suitable:
Moreover, the cells, particularly the enterocytes, are classically seeded onto a permeable culture insert, for example onto a polycarbonate filter. The Transwell™ system is particularly suitable for implementing the method according to the invention.
In another embodiment, this invention also concerns the cell culture which is likely to be obtained using the process described above, particularly after 21 to 35 days, or even 21 to 30 days following seeding the enterocytes. Cocultures thus produced are new, since it has been shown in this application that the seeding sequence has an effect on various characteristics (such as efflux, paracellular transport, etc.) which thus distinguish them from cultures with only one cell type or even from a coculture obtained after simultaneous seeding.
To advantage, the cell culture thus obtained has at least one of the following characteristics:
To advantage, the cell culture is differentiated by its TEER value and/or P-gp activity and/or cytochrome CYP3A4 activity, to advantage by its P-gp activity. As shown in this application, this is particularly marked when the goblet cells are seeded after an enterocyte monolayer has formed, typically between 2 and 8 days after seeding.
To advantage, the TEER value and/or P-gp activity and/or cytochrome CYP3A4 activity obtained using the coculture according to the invention differs from the TEER value obtained for a coculture produced under the same conditions but with simultaneous seeding of the two cell types.
In conclusion, the in vitro model which has been developed in the context of this application has two major advantages. Firstly, it allows the functional capacity of monolayers to be modified, particularly expression of P-gp (P-glycoprotein—multidrug resistance gene-1) and paracellular transport, to make them as close as possible to in vivo conditions. Furthermore, the monolayer is covered by a layer of mucus as in the natural intestinal barrier.
Such a cell culture can thus be used as an in vitro model of the intestinal barrier.
This cell culture may in particular allow bioavailability to be assessed, including the transport and/or metabolism of a compound of interest, especially an active substance or nutrient.
As is apparent from this description, using the method according to the invention the following aspects of a coculture consisting of enterocytes and goblet cells can be modified:
The way in which the invention can be implemented and the advantages ensuing from it are best illustrated by the non-exhaustive examples below, provided for information purposes, supported by the attached figures, where:
An asterisk (*) indicates the values obtained for the cocultures, which are statistically different from values obtained with each strain.
Two types of cells were used for this work. The CBBel clone of Caco-2 cells was obtained from the American Type Culture Collection (ATCC) at passage 47 and was used in experiments at passages 55 to 65. The cell line HT29-MTX10-6 M was provided by Dr Thecla Lesuffleur of INSERM UMR S 938, Paris, France, at passage 13 and was used in experiments at passages 14 to 22.
The two cell types were grown on a routine basis in 25 or 75 cm3 culture flasks maintained at 37° C. in a humidified atmosphere of 5% CO2 in a complete culture medium of DMEM (Dulbecco's Modified Eagle Medium) containing GlutaMAX™, D-glucose (4500 mg/L), sodium pyruvate (110 mg/L) and phenol red (15 mg/L). 15% foetal bovine serum (heat-inactivated FBS), 1% non-essential amino acids and 1% antibiotics (100 μg/mL streptomycin and 100 IU/mL penicillin) were added to the DMEM. The medium was changed twice a week. The cells were subcultured when confluence of about 70-80% was reached. A trypsin-EDTA mixture was used to detach the cells at a concentration of 0.125 and 0.25% for HT29-MTX and Caco-2, respectively.
The Transwell™ (TW) system, which was used for the coculture, consisted of a 24-well plate with polycarbonate membrane filtering inserts (diameter: 6.5 mm; area of membrane: 0.33 cm2; pore size: 0.4 μm; 108 pores/cm2; membrane thickness: 10 μm). Before seeding, the Transwell™ inserts were equilibrated at 37° C., 5% CO2 with culture medium (upper compartment: 100 μL, lower compartment: 600 μL) for at least 1 hour. The medium was then aspirated and the cells seeded.
The day of Caco-2 seeding was considered as day 0 (D0). Caco-2 cells were seeded at a density of 3.105 cells/mL in the upper compartment (3.104 cells/0.33 cm2 or 9.104 cells/cm2). The lower compartment was then filled with culture medium. For the coculture, HT29-MTX cells were seeded at various times between day 0 and day 18 (D0 to D18) after the Caco-2 seeding. For day 0 (D0), the Caco-2 and HT29-MTX were seeded simultaneously by taking 100 μL of a cell suspension with concentrations of 3.105 cells/mL of Caco-2 cells and 1.105 cells/mL of HT29-MTX cells. For other times (D4, D6, D8, D10, D12, D14 and D18), the HT29-MTX cells were seeded directly onto the Caco-2 monolayers at a concentration of 1.105 cells/mL (100 μL/insert), after removing the culture medium from the upper compartment. The Caco-2/HT29-MTX ratio was therefore 75/25 for each coculture. Seeding was performed in quadruplicate. Controls were made with monolayers of Caco-2 and HT29-MTX only. The culture medium was changed every two days for the first two weeks of culture. From the third week of culture until the transport experiments, the growth medium was changed every day. Monolayers were used for transport experiments between days 21 and 30 after Caco-2 seeding.
The transepithelial electrical resistance (TEER) was determined using the Millicell-ERS (Electrical Resistance System), which is a special voltohmmeter for measuring the resistance of monolayers of cultured cells. This device measures the health of cell monolayers qualitatively and cell confluence quantitatively. To take the measurement, ‘chopstick’ electrodes (Ag/AgCl) of different lengths were immersed in a prewarmed culture medium (incubator at 37° C., 5% CO2) of the two compartments of the Transwell™ insert. The TEER was monitored every two days after the sixth day of culture and before and after each transport experiment. The final resistance of the monolayer of cells was calculated by entering the net TEER value in the following equation:
TEER [Ω*cm2]=(RTranswell−Rblank)*A
Rblank was determined before each measurement in a Transwell™ insert without cells, maintained in the same conditions as the inserts containing cells. The RTranswell value is the total resistance of the cell monolayer and the polycarbonate membrane. A is the area of the filter (0.33 cm2).
The apparent permeability Papp (10−6cm/s) describes the absorption permeability.17 Papp is calculated using the following equation:
P
app[10−6cm/s]=dQ/dt*1/(A*C0)*106
dQ/dt [μmol/sec] is the flow of active substance across the cell monolayer over time.
A is the area of the membrane in cm2.
C0 is the initial concentration of active substance in the donor compartment (μmol).
The transport buffer (TB) used for all permeability experiments was composed of 25 mM D-(+)-glucose and 10 mM HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulphonic acid) in a Hank's balanced salt solution, adjusted to pH=7.4 with sodium hydroxide (1 M). The solution was then filtered with a 0.20 μm filter. The TEER of the monolayers was measured before and after each transport study to check the integrity of the monolayer.
For the transport studies, permeability markers including Lucifer yellow (LY), propranolol and rhodamine123 (Rho123) were used as follows:
The results obtained in this series of experiments are presented in
The data concerning these experiments are shown in
In line with the literature, an LY Papp of less than 0.4 (10−6 cm/s) is considered as being acceptable for a monolayer of Caco-2. The paracellular permeability is explained by the presence of tight junctions between the cells. In contrast, the Papp for HT29-MTX is much higher with values around 2-3 (10−6 cm/s). Adding HT29-MTX to Caco-2 cells drastically increases the Papp. A direct relationship can be seen between the Papp values and the time of HT29-MTX seeding. HT29-MTX cells may therefore facilitate the paracellular transport of molecules, which can be adjusted depending on the day of HT29-MTX seeding. These results agree perfectly with the TEER values.
The results for these experiments are shown in
In
The effect of verapamil, a P-gp inhibitor, is shown in
In this figure, when verapamil is added to the monolayers, the Papp values for all the monolayers are similar, demonstrating the absence of efflux. Additionally, no significant difference was observed with HT29-MTX with or without verapamil, suggesting that expression of P-gp is absent in this strain. This explains why the presence of HT29-MTX in the coculture significantly reduces efflux due to a lower level of P-gp expression.
Additional experiments were performed to show the possibility of modifying the characteristics of the coculture obtained even more, depending on the initial Caco-2 density:
| Number | Date | Country | Kind |
|---|---|---|---|
| 1151296 | Feb 2011 | FR | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP12/52780 | 2/17/2012 | WO | 00 | 7/9/2013 |
| Number | Date | Country | |
|---|---|---|---|
| 61443778 | Feb 2011 | US |
