Adobe Acrobat Required  Click here to download

Methods for the isolation of SP cells from adult murine skeletal muscle and skin

The isolation of SP cells from adult mouse skeletal muscle and skin is labor intensive. It is performed in one single day unless you decide to dissect the tissues the evening before and store them in the fridge for use the next morning. Dissection is really the only step after which the procedure can be paused. Once tissue dissociation is started, you have to go till the end. We tried freezing down the dissociated cells. It works, but cell loss is considerably increased and we feel the number of SP cells recovered is less than in fresh preparations. If you decide to use frozen cells, you have to rinse them well after thawing and culture them overnight to eliminate all the DMSO. Since DMSO makes cells permeable, it interferes with Hoechst dye exclusion by SP cells.

When trying this protocol for the first time, start early in the morning. Do not be surprised if you start at 8:00 am and reach the cell sorter only around 5:00 pm. As your manual dexterity improves, you will be able to start at 9:00 am and expect to be at the cell sorter by 3:00 pm. So don't give up!

Solutions

Bovine serum albumin (0.5% w/v; Sigma, cat# A9647) was dissolved in phosphate buffered saline without calcium, nor magnesium (PBS/BSA). The solution was filter sterilized using a low protein-binding filter and stored at 4 degrees C. The absence of calcium and magnesium and the addition of BSA prevent cells from adhering to each other forming clumps during incubations.

Red blood cell lysis buffer (RBCLB): 0.2% w/v Tris pH 7.5, 0.747% w/v NH 4 Cl. Store at 4 0 C. This buffer is formulated for the lysis of mouse red blood cells and does not work well on human erythrocytes.

Enzymes: Dispase (Worthington; cat# LS02104), collagenase IV (Worthington; cat# LS004188) and collagenase I (Worthington; cat# LS004196) are dissolved in PBS containing 5 mM CaCl 2 . Calcium is necessary to activate these enzymes. Dispase stock solutions are prepared at 2.4 U/ml for muscle and 6 U/ml for skin. Collagenase IV and I stock solutions are prepared at 10 mg/ml. Trypsin (Worthington; cat# LS003667) is dissolved in 0.001N HCl at a concentration of 100 mg/ml.   All stock solutions are aliquoted and stored at -20 degrees C. The final digestion mixes are prepared fresh before use as follows:

Muscle enzyme mix (MuE): after cleaning and mincing the tissue, weigh it. The muscle weight (in grams) is multiplied by 3.5 and the number obtained is used as the volume in ml of each enzyme to be added. The final enzyme mix has the following concentrations: 1.2 U/ml dispase, 5 mg/ml collagenase IV and 5 mM CaCl 2 .

Skin enzyme mix with trypsin (SkET): 4 ml collagenase I stock solution, 4 ml dispase stock solution (6 U/ml), 200 m l trypsin stock solution, 12 ml PBS. Final concentrations: 2 mg/ml collagenase I, 1.2 U/ml dispase, 1 mg/ml trypsin, and 2 mM CaCl 2 .

Skin enzyme mix without trypsin (SkE): 4 ml collagenase I stock solution, 4 ml dispase stock solution (6 U/ml), 12 ml PBS. Final concentrations: 2 mg/ml collagenase I, 1.2 U/ml dispase and 2 mM CaCl 2 .

Hoechst, verapamil and PI: Hoechst 33342 (Sigma, cat#B2261), verapamil ( Sigma, cat# V-4629) and PI (Sigma, cat# P4170) are dissolved in distilled deionised water at 1 mg/ml, 5 mM and 1 mg/ml respectively. All solutions are filter sterilized. Aliquots are kept frozen at -20 0 C. Hoechst and verapamil are allowed to thaw at room temperature protected from light before use . Under these conditions the same aliquot can be re-frozen and re-used multiple times for Hoechst and up to 5 times for verapamil. PI is also thawed at room temperature and subsequently stored at 4 degrees C.

Skeletal muscle medium (SkM): Ham's F10 with 20% FBS, 50 U/ml penicillin, 50 m g/ml streptomycin

Isolation of muscle cells

Dissect the hind limb and the proximal fore limb muscles and either use them fresh or store them overnight at 4 degrees C in SkM. Transfer the muscle into a Petri dish, clean it from connective tissue and fat, mince* it until no clumps are visible and weigh it to calculate the volume of enzymes needed. Digest it for 45 min at 37 degrees C in the appropriate volume of MuE solution. During the digestion, pass the tissue 10 times through a 5 ml pipette every 15 min to facilitate its dissociation. Terminate the enzymatic digestion by adding 20-25 ml SkM and filter sequentially through a 70 m m and a 40 m m cell strainer. Centrifuge the cells for 10 min at 514xg at 4 degrees C and resuspend the pellet in 3 ml SkM. Lyse red blood cells by adding 17 ml RBCLB, invert a few times to mix, then centrifuge for 10 min at 514xg. Resuspend the cells for cell count in 2-5 ml PBS/BSA pre-warmed to 37 degrees C.

Isolation of skin cells

Shave the mice and remove the dorsal and ventral skin. Carefully remove sub-cutaneous fat and muscle tissues by scraping with a scalpel blade. Thoroughly mince* the skin and digest it for 45 min at 37 degrees C in 20 ml SkET or SkE. During the digestion, pass the tissue 10 times through a 5 ml pipette every 15 min to facilitate its dissociation. Terminate the enzymatic digestion by adding 20-25 ml DMEM supplemented with 10% FBS, and filter the sample sequentially through a 70 m m and a 40 m m cell strainer. Centrifuge the cells for 10 min at 514xg at 4 degrees C. Skin samples usually only contain traces of blood and do not need to be treated with RBCLB. If there is a significant amount of blood in the pellet, add 3 ml DMEM supplemented with 10% FBS and 17 ml RBCLB. Invert a few times then centrifuge again for 10 min at 514xg at 4 degrees C. Resuspend the pellet for cell counting in 2-5 ml PBS/BSA pre-warmed to 37 degrees C.

* Note: Do not let the tissue dry out. Add PBS if necessary.

Hoechst 33342 staining

Count the total cell number (live and dead nucleated cells) using a hemacytometer. Be precise, count all four chambers and be consistent from sample to sample . There will be a lot of small particulate matter but do not get discouraged. It is normal. Count anything that you feel might contain DNA, anything remotely similar to a cell. It is important to count both live and dead cells since Hoechst binds indiscriminately anything that contains DNA and the proportion of dead cells varies from sample to sample.   Counting both live and dead cells will increase the reproducibility of the SP profile from different samples. Add the appropriate volume of warm PBS/BSA to the samples to obtain a final concentration of 10 6 cells/ml. Set aside one aliquot of at least 1 ml for staining in the presence of 100 m M verapamil. Add verapamil first, let stand at 37 degrees C for 5min, then add the appropriate volume of Hoechst. Add Hoechst to the remaining samples. When trying this protocol for the first few times, test different concentrations of Hoechst such as 3, 5, 7.5, 10, 12.5, 15, and 20 m g/ml. From this Hoechst concentration curve select the concentration where the number of SP cells consistently reaches a plateau and the corresponding verapamil sample shows the lowest amount of SP cells. This concentration is affected by several parameters, such as the way you count the cells and the background of your mouse strain. Wrap your tubes in aluminum foil and incubate them in a 37 degrees C water bath ( make sure the temperature is stable during the incubation period ) for 60 min for muscle and skin. Longer incubation times, such as 90 min, can be used but we found no difference with an incubation of 60 min and it has the advantage of leaving the cells less time in the presence of Hoechst, which is toxic. Immediately transfer the samples on ice and keep them on ice and protected from light for all subsequent steps. Add at least five volumes of ice cold PBS/BSA, then centrifuge the cells for 10 min at 514xg at 4 degrees C and resuspend the pellet in 1-3 ml ice cold PBS/BSA. Before FACS analysis, add 2 m g/ml propidium iodide (PI) to exclude dead cells .

Antibody staining

Antibody staining for Sca-1 and CD45 should be performed when the technique is being set up in the lab to ensure that the SP gate is set properly and that the correct concentration of Hoechst has been used. It is also important to know how consistent your samples are from day to day, and to have a clear notion of the composition of the SP population you are isolating. Other tissue-specific markers can be added later.

Right after Hoechst staining, prepare five (six if you are doing double labelling) aliquots with at least 10 to the sixth power cells per aliquot in 15ml conical tubes then fill the tubes with ice cold PBS/BSA and centrifuge for 10 min at 514xg at 4 degrees C. From now on, keep the tubes on ice at all time and use only ice-cold solutions . Resuspend the pellets in 0.1ml ice-cold PBS/BSA, use one aliquot as a negative control where nothing is added, then add 2 m g primary antibody or isotype control per 10 6 cells to the remaining aliquots, and incubate on ice for 15 min. The following primary and isotype control monoclonal antibodies are commonly used: rat anti-mouse Sca-1 conjugated to FITC (Pharmingen, cat#553335, clone E13-161.7), rat anti-mouse CD45 conjugated to PE (Pharmingen, cat#553081, clone 30-F11), rat IgG2a,k isotype control conjugated to PE (Pharmingen, cat#553930, clone R35-95),   and rat IgG2a,k isotype control conjugated to FITC (Pharmingen, cat#553929, clone R35-95). Add 5 ml of ice cold PBS/BSA, mix by inverting the tubes several times then centrifuge for 10 min at 514xg at 4 degrees C . Resuspend the pellet in 0.5 ml ice-cold PBS/BSA and just before FACS analysis, add 2 m g/ml PI to each sample . For each tissue, use the control samples labelled with both isotype control antibodies or with a single primary antibody to determine the background noise due to non-specific antibody binding, and to set the proper compensation for optimum separation between the FITC and PE signals.

We acquire the data using the CellQuest software (Becton Dickinson) and subsequently analyze it with Flowjo (Treestar). In our hands, analysis of all live cells in bone marrow, muscle or skin preparations following antibody staining reveals reproducible percentages of Sca1-positive and CD45-positive cells for each tissue (see table). These markers could be used as indicators of the efficiency of digestion and reproducibility of sample preparation. Treatment of skin tissue with trypsin (for the digestion time used in this protocol) does not significantly affect the epitopes on Sca1 and CD45 recognized by the antibodies used here (see table).

FACS analysis

FACS analysis was performed on a dual laser FACSVantage SE (Becton-Dickinson) using the following settings. Lasers: 488nm Argon ion laser (Coherent, Innova C) and multiline high UV output laser (Coherent, Innova 90-6). Laser power: 200 mW for the 488 nm laser, 150 mW for the 365nm laser. Filters: PI has 2 excitation peaks and can be excited with both lasers. It should be noted that the PI excitation from the 365 nm laser was used to visualize SP cells. Emission signals from Hoechst and PI were first separated using a 500 SP dichroic mirror, then detected using a 400 LP and a 600 LP filter respectively. FITC and PE were excited with the 488 nm laser and emission signals were detected using a 530/30 BP filter for FITC and a 575/25 BP filter for PE (see figure). In order to ensure that the FITC and PE signals were properly separated and that the PI signal elicited by excitation in the visible range did not interfere with the detected antibody signals, FACS compensation was performed using three control samples: 1- cells labeled with a mix of isotype control antibodies directly conjugated to FITC and PE, 2- cells labeled with the anti-CD45-FITC antibody only, and 3- cells labeled with the anti-Sca1-PE antibody only.

Figure: FACS visualization of Hoechst and PI

Excitation/emission spectra of DNA-bound Hoechst 33342 and PI. The excitation and emission ranges for each dye are indicated by gray and black double-ended arrows, respectively. Excitation and emission peaks are denoted by a vertical bar. Solid black bars indicate the wavelength range that is detected using the FACS filters for Hoechst and PI described above.  

Figures of the FACS profiles and descriptions of the effects of changing important protocol variables on SP cell isolation, are described in Montanaro et al. (2004) Exp. Cell Biol. The article is available online ahead of press at the Experimental Cell Research web site.

Should you need further details or assistance please contact one of the following people:

•Muscle preparations:

Kalliopi Liadaki: liadaki@enders.tch.harvard.edu

Jaclyn Schienda: jschienda@enders.tch.harvard.edu

Federica Montanaro: fmontana@enders.tch.harvard.edu

Emanuela Gussoni: gussoni@enders.tch.harvard.edu

•Skin preparations:

Federica Montanaro: fmontana@enders.tch.harvard.edu

•FACS Setup

Alan Flint: flint@enders.tch.harvard.edu