Mitochondrial and glycolytic extracellular flux analysis optimization for isolated pig intestinal epithelial cells

Animals

Intestines were either harvested from pigs at a slaughterhouse or from control pigs from dedicated animal experiments sacrificed at our Animal Facility, that were all approved by the Animal Care and Use Committee of Wageningen University. However no animals were sacrificed specifically for the purpose of this study. The majority of material was derived from slaughterhouses. We were unable to get all details for pig breeds, age and sex, because we choose for a rapid, relatively easy organ collection procedure instead of a detailed dissection. This offers ease in performing multiple experiments on scheduled cell isolation days, and shows the flexibility of the isolation procedure. Overall, the pigs were from both sexes, multiple breeds, weighed between 20 and 100 kg and were aged between 12 and 32 weeks.

IEC isolation

Following excision from the abdominal cavity, an approximately 20 cm long segment of the colon was taken for IEC isolation and placed in aerated Krebs Henseleit Buffer containing 5 mM glucose (#K3753, Sigma Aldrich; hereafter referred to as modified-KHB), containing 2.5 g/L Bovine serum albumin (BSA, #A7906, Sigma-Aldrich). Samples were transferred to the lab and isolation commenced within 2 h after killing the animal.

Multiple steps were taken to optimize the isolation procedure. Initially, two methods described in literature were used as a basis for the design of the procedure. First, a method described by Roedinger and Truelove²⁷ was assessed. We refer to this method as the vigorous method. The colon segments were first flushed with room temperature (RT) modified-KHB, inverted and then a sac was created using dialysis clamps (#Z371092, Sigma Aldrich) and the sac was filled with modified-KHB. The sacs were then placed in Ca²⁺-free KHB with 5 mM Ethylenediaminetetra-aceticacetic acid (EDTA) and 2.5 g/L BSA. After a 30-min incubation in a shaking 37 °C water bath, the buffer was removed and replaced by fresh Ca²⁺-free KHB containing 2.5 g/L BSA. The intestines were stirred vigorously by hand for two minutes to dissociate the IECs. IECs were then passed over a 70 µM cellulose filter top to remove large tissue pieces and debris. After washing cells twice using modified-KHB contained 2.5 g/L BSA, cells were taken up in pH 7.4 buffered XF DMEM medium (#103575-100, Agilent Technologies) supplemented with 10 mM glucose (#103577-100, Agilent Technologies), 2 mM glutamine (#103579-100, Agilent Technologies) and 1 mM pyruvate (#103578-100, Agilent Technologies) and counted using a Cellometer K4 (Nexcelom Bioscience) and viability was simultaneously assessed by staining with ViaStain (#CS2-0106, Nexcelom Bioscience). The second method tested, to which we refer as the gentle method, was a modification of the one described by Darcy-Vrillon et al.¹⁹. In this method, the intestine was first flushed with modified-KHB, and then immediately a sac was created using dialysis clamps. The sac was filled with Ca²⁺-free KHB containing 10 mM EDTA, 5 mM Dithiothreitol (DTT), and 2.5 /L BSA. After a 20-min incubation in a shaking 37 °C water bath, the sac was emptied and refilled with the same buffer, followed by another fifteen-minute incubation. Afterwards, the intestines were gently massaged, and cells were collected, washed and counted, as described for the first procedure.

The third method we tested (which was also the optimized method we used for downstream analysis of metabolic function) was a combination of steps from the above two methods combined with a hyaluronidase enzymatic dissociation step and optional washing steps. We refer to this method as the ‘enzyme’ method or ‘enzyme + extra wash’ method (see supplementary materials for a stepwise lab protocol of the final optimized method). First, intestines were flushed with modified-KHB. Then, they were inverted, and a sac was created using dialysis clamps by filling them with modified-KHB. Inversion of the intestines at this stage facilitated exchange between buffer and mucosa, since the amount of buffer can be much higher than if the intestines are not inverted. The sacs were first incubated for 20 min in Ca²⁺-free KHB buffer containing 20 mM EDTA and 10 mM DTT in a shaking 37 °C water bath. Following this washing step, intestines were re-verted and filled with an isolation buffer containing Ca²⁺-free KHB buffer, 2.5 g/L BSA and 400 U/mL hyaluronidase type IV (#3884, Sigma-Aldrich), an enzyme that catalyzes the breakdown of hyaluronic acid which is present in the extracellular matrix of IECs. The re-version of the sacs at this stage is convenient, since cells will be collected in a smaller volume. In addition, the amount of buffer and enzyme needed can thus be reduced, which is cost-effective. After a fifteen-minute incubation, the intestines were gently massaged and cells were collected, washed and counted as previously described. This protocol was finally adjusted by adding a 20 min washing step before the enzymatic digestion to facilitate increased removal of mucus from the intestines.

For the experiments where we analyzed the effects of antibiotics on metabolic function of IECs, all washing and isolation buffers used in the isolation procedure were supplemented with 1% v/v penicillin–streptomycin (#15140122, Fisher Scientific).

Metabolic flux analysis with seahorse XFe96 analyzer

Isolated IECs were plated in a XF96 cell plates that were coated with Cell-Tak (#354240, Corning, New York, USA) according to manufacturer’s protocol, no longer than one week prior to the assay. Cells were plated at concentrations ranging from 25,000 to 150,000 cells/well in 50 µL pH 7.4 balanced XF DMEM assay medium supplemented with 10 mM XF glucose, 2 mM XF glutamine and 1 mM XF pyruvate. For the normalization optimization and Oligomycin response experiments, cells were plated at 100,000 cells/well, left to settle for 5 min prior to spin-down (200 × g for 2’ with zero break). After spin-down, cell plates were imaged as described below, while kept at 37 °C. Following imaging, an additional volume of 130 µL assay medium was added and cell plates were incubated for another 20 min in a non-CO₂ 37 °C incubator. For the optimization of carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP; #C2920, Sigma-Aldrich), cells were isolated using the ‘gentle’ method and final concentrations of 0.5–1.18 µM were injected into the wells. For optimization of Oligomycin (a mix of A, B and C Oligomycin, #O4875, Sigma-Aldrich) concentration, a final concentration of either 1.5 or 15 µM was used. Extracellular flux analyses (XF assays) was performed using the Seahorse XFe96 (Seahorse Bioscience, Agilent Technologies, Santa Clara, USA). Most often, XF assays were performed using serial injections of 1.5 µM Oligomycin, 1 µM FCCP, a combination of 1.25 µM Rotenone and 2.5 µM Antimycin A and finally 50 mM 2-deoxyglucose (2-DG). The XF assay protocol typically consisted of 12 measurement cycles of 3 min, with 2 min of mixing in between measurements. For the measurements with antibiotics, assay medium also contained 1% v/v penicillin–streptomycin. Cell plates were kept in a non-CO₂ 37 °C incubator for 1.5 h prior to the start of the assay for the delayed measurement experiments.

Imaging procedure

Brightfield images of the inner probe area of each well in the XF96 cell plates were obtained prior to the XF assay run using a 37 °C equilibrated Cytation 1 Cell Imaging Multi-Mode Reader (BioTek, Winooski, Vermont, USA) using a 4 × objective. A LED intensity of 5 and integration time of 80 ms was kept constant for all cell plates, image focus height was adjusted as needed to get the optimal image quality, as was determined by visual inspection. For optimization of the normalization procedure cell nuclei were either stained using 8 µM Hoechst 33342 (Hoechst, #B2261, Sigma-Aldrich), or fixed using 4% paraformaldehyde (#252549, Sigma-Aldrich) and then stained with 4′,6-diamidino-2-phenylindo (DAPI, #D9564, Sigma-Aldrich), followed by image acquisition using a 4 × objective with a 365 nm LED in combination with an EX337/EM447 filter cube.

Brightfield image analysis in R

Brightfield images obtained prior to the XF assay run were processed and quantified using an in-house generated R-script that uses the EBImage package available for Bioconductor⁵⁰. Image processing was performed in a similar manner as previously published²⁹, with an adjustment of the image quantification. Briefly, a Gaussian blur low-pass filter was applied to generate a background image, followed by subtraction of the background image from the original. The background corrected image was then inverted to generate a “white-objects-on-black-background image”. This image was subsequently cropped by 5% to remove potential noise from the XF assay plate molded stops, that are present on the plates to prevent the sensors from disrupting the cell monolayer. Images were then analyzed to calculate pixel intensity values for all the pixels in the image. All the pixels with an intensity > 1 was counted as representing the presence of a cell, and we refer to these as “cell-pixels”. For conversion of cell-pixels back to cells, an external calibration curve was generated. To do this, a second order polynomial fit analysis was performed on the combined data of three individual pig standard curves. The best-fit curve that matched the data was then used as an external calibration curve to convert cell-pixel values of every plate back to cell numbers. These cell numbers were subsequently used for normalization of the Seahorse XF assays. The R-script is available from GitHub (https://github.com/vcjdeboer/seahorse-data-analysis-PIXI).

Statistical analysis and data visualization

Data are presented as mean ± s.d., unless stated otherwise. The standard score, or z-score, was calculated using Eq. (1):

Statistical analyses and data visualizations were performed using GraphPad Prism v.9 (GraphPad Software, CA, USA). Statistical testing was performed using student’s t-test or one-way ANOVA when appropriate and as stated in the figure legends. A p-value of < 0.05 was considered statistically significant.