For human tissues, fresh tissues were obtained with appropriate IRB approval from the UCLA Department of Pathology Translational Pathology Core Laboratory.
All animal studies were approved by the animal research committee at UCLA, IRB #2005-169. The UCLA facility is an AALAC-accredited facility. This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All efforts were made to minimize suffering.
Myofibroblasts were isolated from 5-day-old C57BL/6 wild type mice from our own breeding colony. Six-week-old immunocompromised NOD-SCID-IL2Rγnull (NSG) mice (Jackson Laboratory, Bar Harbor, Maine) were used for human epithelial implantation studies. Both strains of mouse were housed in the UCLA animal facility. The mouse pups were sacrificed per UCLA Division of Laboratory Animal Medicine (DLAM) protocol using an isoflurane overdose followed by decapitation. NSG mice were placed into a CO2 chamber and gas added per DLAM protocol. The UCLA facility is an AALAC-accredited facility. This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All efforts were made to minimize suffering.
Human Intestinal Tissue
Non-diseased small intestinal samples were obtained fresh from intestinal specimens excised for Roux-en-Y gastric bypass procedures and uncomplicated ileostomy takedown procedures. Samples were obtained from the Surgical Pathology Department within 45 minutes of resection and placed into ice-cold Dulbecco's Phosphate Buffered Solution (PBS). Fresh tissues obtained with appropriate IRB approval from the UCLA Department of Pathology Translational Pathology Core Laboratory.
Isolation of intestinal crypts
Tissue was removed from PBS solution and washed multiple times with ice-cold PBS washes until the solution remained clear. The specimen was then placed in a Petri dish containing PBS on ice with the mucosal surface facing upward. Using a razor blade, excess mucoid material was scrapped from the epithelial surface. The specimen was then divided into approximately 0.5 cm2
pieces. These pieces were placed into a 2.5 mmol/L EDTA solution in PBS for 30 minutes of incubation with gentle shaking at 4°C. After this incubation period, the fragments were allowed to settle and the supernatant was discarded. 10 ml of cold PBS was added to the sample, and subsequently vortexed for 10 seconds with 1-second bursts. The fragments were allowed to settle, and the supernatant was removed and saved on ice. Again 10 ml of PBS were added and the process was repeated eight times. Samples were spun down at 100 g for 2 minutes. The supernatant was discarded. The contents of the pellets were examined under light microscopy using a Nikon TMS microscope to assess purity of crypt fractions. Typically, all fractions were pooled together to increase yield of epithelial crypts. The pooled fractions were then purified using a 100-µm pore filter (BD Biosciences, Bedford, MA). Fetal Bovine Serum at 10% per volume was then used to suspend the contents of the filtrate. These clusters were examined under light microscopy and counted. 500 crypt clusters were suspended in 50 µL Matrigel (BD Biosciences) as previously described in Sato's 3-D Matrigel culture system developed for murine intestines
. The crypt cell/Matrigel suspension was placed directly upon previously plated mouse/human myofibroblasts. Matrigel was allowed to polymerize on the myofibroblasts. Crypt culture medium was then added to the wells. The media consisted of Advanced DMEM/F12 (Invitrogen, Carlsbad, CA) with penicillin-streptomycin (Invitrogen), GlutaMax supplement (Invitrogen, 2 mmol/L), HEPES buffer (Invitrogen, 10 mmol/L), N-2 supplement (Invitrogen), B-27 supplement (Invitrogen), EGF (PeproTech, Rocky Hill, NJ, 50 µg/mL), Murine noggin (PeproTech, 100 µg/mL) and R-spondin (R&D Systems, Minneapolis, MN, 1 µg/mL)
. Subsets of studies utilized various doses of Wnt3a (R&D Systems, 100 ng/mL), and FGF10 (R&D Systems, 100 ng/mL). The medium was replaced every two days with the same factors.
Myofibroblast isolation and culture
Small intestine was excised from 7 day-old mice. The tissue was placed into a Petri dish containing calcium and magnesium free Hank's Buffered Salt Solution (Invitrogen) with D-Glucose (Sigma, 20 mg/mL), penicillin-streptomycin (Invitrogen), and L-glutamine (Invitrogen, 4 mmol/L) (HBSS* solution). The intestines were washed out and rinsed. The intestinal tissue was diced into 0.3–0.5 mm2 pieces. The diced material was transferred into a T25 flask. 30 mL of cold HBSS* solution was added to the flask after which the flask was shaken for 2 minutes at room temperature. The flask was then allowed to settle and the supernatant discarded. This process was repeated until the solution was clear.
Once the last supernatant was discarded, a 20 mL dispase (Invitrogen, 0.31 mg/mL)/collagenase Type XI (Sigma, St. Louis, MO, 0.25 mg/mL) solution was added to the tissue. The flask was gently rocked moderately for 30 minutes at room temperature.
The flask contents were then transferred to a 50 mL conical tube and vigorously shaken for 30 seconds. 10 mL cold HBSS* was added to the solution and entire contents allowed to settle. The supernatant was transferred to a new 50 mL conical tube. This was repeated 6 times in 50 ml conical tubes.
The samples were then suspended in 25 mL of high glucose Dulbecco's Modified Eagle Medium with fetal bovine serum (Invitrogen, 5% v/v), L-glutamine (Invitrogen, 4 mmol/L), D-Sorbitol (Sigma, 20 mg/mL), and penicillin-streptomycin (Invitrogen) (DMEM-S solution). The solution was inverted until well mixed and then centrifuged at 100 g at 4°C for two minutes. The tube was then placed back on ice and the supernatant discarded. The pellet was transferred to a 5 mL centrifuge tube. The contents were allowed to settle and any supernatant was discarded. The pellet was resuspended in HBSS with magnesium and calcium supplemented with penicillin-streptomycin (Invitrogen) and L-glutamine (Invitrogen, 4 mmol/L). The entire contents were spun at high speed for 10 seconds. The supernatant was discarded and the pellet was suspended in Basic Growth Media for myofibroblasts. Basic Growth Media consisted of DMEM (Invitrogen), with Antibiotic-Antimycotic (Invitrogen), fetal bovine serum (Invitrogen, 10% v/v), EGF (PeproTech, Rocky Hill, NJ, 50 µg/mL), transferrin (Sigma, St. Louis, MO, 10 µg/mL), and insulin (Sigma, St. Louis, MO, 0.25 U/mL) added.
Immunohistochemical studies were undertaken using paraffin-embedded culture samples that were prepared as follows: culture samples were washed once with PBS. The samples were then fixed for ~12 hours with 10% buffered formalin solution. The formalin solution was then removed and 80% ethanol solution added for 10 minutes then removed. A 95% ethanol solution was added for 15 minutes twice. Finally, 100% ethanol was then added for 10 minutes. The culture contents were then carefully removed mechanically from the culture dish. The samples were then paraffin embedded. Serial 8 µm cuts of the tissue were obtained for microscopic evaluation and staining. Immunohistochemical staining was performed using the DAKO (Carpinteria, CA) automated Flex system. Primary antibodies CDX-2, E-cadherin, SMA, Synaptophysin were obtained from DAKO and were at manufacturer concentrations. Antibody to lysozyme (DAKO) was diluted 1
1500 in Antibody Diluent (DAKO). Antibody to Chromogranin A (Immunostar, Hudson, WI) was diluted 1
200 in Antibody Diluent (DAKO).
7–11 days old culture samples were allowed to elevate and partially detach from the culture plate as part of their natural growth process. These samples were placed on non-woven 5 mm polyglycolic acid (PGA) felt disks (Synthecon, Houston, TX). Immunocompromised NOD-SCID-IL2Rγ null (NSG) mice were anesthetized in a manner consistent with protocols establish by the UCLA DLAM group (http://www.ncbi.nlm.nih.gov/pubmed/19052619
). A subcutaneous pocket was created in the anterior abdominal wall. The PGA felt with the cultured cells was placed into the pocket and 6–0 Prolene suture (Ethicon, Somerville, NJ) was used to suture the scaffold to underlying muscle. The overlying incision was closed without tension using 3–0 silk (Ethicon, Somerville, NJ) suture. The mouse was sacrificed after 28 days and the implantation was excised and fixed in buffered 10% formalin solution. The sample was embedded, sectioned, and stained for microscopic evaluation.
Quantitative real-time PCR
mRNA was isolated from the samples with the RNeasy RNA Isolation Kit (Qiagen, Valencia, CA) following the manufacturer's protocol. The mRNA samples were then prepared for the RT-PCR reaction with the Quantitect Probe RT-PCR Kit (Qiagen) and the TaqMan Gene Expression Assay (Applied Biosystems, Carlsbad, CA) for smooth muscle actin (Assay ID Mm01546133_m1), desmin (Mm00802455_m1), vimentin (Mm00449208_m1), and GAPDH (Mm99999915_g1). GAPDH was used as the house keeping gene to normalize RNA quantities. The samples were analyzed with the LightCycler 480 Real-Time PCR System (Roche, Indianapolis, IN) with settings described in the Quantitect Probe Kit. The comparative CT method was used to calculate the relative gene expression.