Purification of cells from primary human breast tissue
Sibgat Choudhury 03/16/1

  1. Before grinding the tissues, find an area of tissue without fat covered and cut 2-4 pieces (thickeness no more than 5 mm) to fix in 10% formalin. Fixed tissue should be processed or save in PBS at 4 0C after 24 hours. Tissue stored in PBS has to be processed within a week. Mince breast tissue into small (1-2mm) pieces using a food processor.
  2. Digest tissue in ~5x tissue volume of DMEM/F12 medium (experiments requiring estrogen-free conditions, phenol red free medium and charcoal stripped serum is necessary) containing 2 mg/ml collagenase IV (Worthington, LS004189), 2 mg/ml hyaluronidase (Sigma H3506) and 2mg/ml BSA (Sigma A7906) with constant agitation at 370C. We set up the digest in a 500-1000 ml bottle with stir bar and immerse in a water bath heated to 370C on a heated magnetic stirrer. Small-scale purifications can be performed in 50 ml tubes with a stir bar and similarly immersed in a 370C water bath. After 45 min of digestion, remove the top fat layer. Digest for 2-8 hours. Shake bottle occasionally during the digestion to break apart any large clumps. Check the digested medium under the microscope. Stop the digestion before the organoids are digested into single cells.
  3. Spin down cells by pouring the digested tissue through a 500 mm mesh into 50/250 ml tubes and centrifuge at 3,000 rpm for 10 minutes at room temperature. Then wash with HBSS/PBS (we use PBS and HBSS interchangeably) 1x and cf again (combine the pellet from different tubes into one).
  4. Resuspend pellet in 10-20 mls of HBSS and filter through a 500 μm, a 250 μm, a 100 μm, and then a 40 μm cell strainer (500 μm, 250 μm and 20μm filter meshes are from Tetko, cat #03-500-47, 03-250-50 and 03-20-14 respectively, now Fisher, NC9168854 while the 100 μm and 40 μm filters are from Fisher cat #: 08-771-19 and 08-771-1, respectively). Collect organoids that remain in 250 μm, 100 μm, and 40 μm strainer, combine all organoids from all three strainers. For the 100 and 40 μm strainers use multiple (2-4) at the same time since they get clumped pretty quickly. It is very important to wash filters with HBSS to flush out any single cells attaching to the organoids. This will ensure a more pure organoid fraction. Whatever flows through the 40 μm strainer, filter through a 20 μm mesh to ensure nearly complete removal of epithelial cells/breast ducts. Collect organoids as well as flow through (the flow through is the stroma fraction) by cf. At this point both organoids and stromal fraction can be stored in freezing media in liquid nitrogen for future use. If you freeze the organoids, it is better to do the trypsinisation and Percoll steps after thawing, since Percoll also removes dead cells that may stick to the beads and lead to non-specific binding. Thawing of the frozen organoids/stromal fraction should be done as usual (i.e., take out the vial from nitrogen, quickly warm it at 370C, dilute with warm media or HBSS (prior to trypsinization) and gently spin out (2000 rpm) cells from DMSO. Let the cells recover in medium at 37 0C for 15 mins at least before the percoll centrifugation. For stromal cells go directly to STEP 7.
  5. Resuspend organoid pellet in 5 ml (or 10x vol) trypsine, incubate at 37 0C for 5 min., pipette up and down rigorously to disrupt clumps, then dilute with ice cold medium+10% BSA to inactivate trypsin, repeat the pipetting step. Double check the cells under the microscope to make sure the majority are single cells. Then filter through a 100μ, 40μ, and 20μ mesh, save flow through (these should to be single cells.) and collect cells by cf. 3,000 rpm for 5 min.
  6. Wash cells once with ice cold PBS to remove residual trypsine. Cf. 3,000 rpm for 5 min. If organoid pellet has significant RBC or leukocytes contamination, then go to step 7, otherwise go to step 8 directly. The number of cells for bead purification (step 8) should be around 15-20 million. The rest of the cells can be stored.
  7. Percoll gradient cf. (Percoll-GE Healthcare cat# 1708901, mix: 9.5 ml percoll+1.5ml 10xPBS+4 ml dh2O, layer Percoll underneath the cells, cf for 3,000 rpm 15-20 min.) to remove dead cells and erythrocytes, you can see a cloudy cell layer in the ~middle, gently collect this with a 10 ml pipette and wash cells in PBS 1x to remove Percoll, collect them by cf. at 3,000 rpm for 5 min.
  8. Resuspend cell pellet in 200-1000 μl (volume depends on the number of cells, you don’t want to have more than ~1 million cells/100 μl) PBE (PBE is PBS, 0.5 % BSA and 2mM EDTA)-first resuspend cells in 200μl, pipette up and down vigorously then add more PBE. For 10 million cells have to use 1 ml volume, minimum.
  9. Removal of leukocytes. This is a critical step, since both the CD44 and CD24 antigens are present on leukocytes. For 10 million cells, cell suspension in 1 ml PBE is added to 100μl (depends on the total # of cells and the estimated # of leukocytes) of 1X pre-washed “leukocyte removal cocktail” (100ul mix of pre-washed CD45 cat# 111.53, CD15 cat# 111-37D, CD14Cat#111-49D, CD19 cat#  111-43D, CD3 Cat# 111-51D Invitrogen/Dynal beads) after removing the solvent from the mix, incubate for 8-12 min. on ice with occasional “flicking” of the tube. Capture bound cells on magnet for 2 min. The captured cells are the LEUKOCYTES, freeze this immediately on dry ice. Wash bound cells 3x200 μl PBE to remove contaminating cells. Transfer unbound cells and first wash into new Epp. Tubes and repeat magnetic capturing to ensure that all bead bound cells are removed. Combine first wash with the unbound fraction and if volume is too large, you can collect them by centrifugation (3000 rpm, 1 min. in epp. Cf.) and resuspend them in 200-1000 μl PBE to the optimal cell density. For more than 10 million cells, this leukocytes depletion step should be performed separately by adjusting the volume to the number of cells described earlier to avoid non-specific binding. After leukocute removal step, check for the presence of leukocytes under a microscope with 10ul of the last wash on a slide.
  10. Removal of endothelial cells. This is another critical step, since PROCR antigen is present on endothelial cells. To the unbound cells from step 9 add 5 mg CD146 (P1H12, MCAM) (Serotec – cat# MCA2141), incubate on ice for 30 min. Then add 100 ml prewashed Invitrogen/Dynal Pan mouse IgG (cat# 110-41) and incubate it on ice for another 10 min, then wash 3x with 200ml PBE to remove unbound antibody and resuspend it in 100ml PBE. Incubate, capture, and wash the cells as described above. The captured cells are the ENDOTHELIAL CELLS. Wash bound cells 3x200 ul PBE then freeze captured cells immediately on dry ice. Transfer unbound cells and first wash into new Epp. Tubes and repeat magnetic capturing to ensure that all bead bound cells are removed. Take unbound cells and go to the next step (Fibroblast isolation).

    NB: For purifying stromal cells, after step 10, an additional purification step with 100ul of pre-washed Breast epithelial enrich Dynabeads (contains BerEP4 antibody specific for epithelial cells) for 10 million cells is required to remove any contaminating epithelial cells. This will ensure a more purified stromal cell population. Furthermore, myofibroblasts (from tumors) can be isolated by using CD10 beads as described above. This stromal fibroblast-enriched fraction is then subjected to testing the purity by semi-quantitative RT-PCR described below.
    ​
  11. Fibroblasts: From step 10 take unbound cells, pellet cells by cf. and resuspend in 200-1000 μl PBE the same way as before and add 100 μl of pre-washed/bound IB10 beads (bind 15μl IB10 antibody- sigma cat#F4771 - to 100 μl beads-Invitrogen/Dynal Pan mouse IgG cat# 110-41- in 100 μl PBE at 4C for at least 1 hour) to the cells and capture/wash as described above. The captured cells are the FIBROBLASTS. Wash bound cells 3x200 ul PBE then freeze captured cells immediately on dry ice. Transfer unbound cells and first wash into new Epp. Tubes and repeat magnetic capturing to ensure that all bead bound cells are removed. Take unbound cells and go to the next step.
  12. MUC1+ cells: to the cell suspension add 100 μl prewashed beads prepared as follows: mix 100 μl prewashed Dynal Pan mouse IgG (cat# 110-41) with 3μl MUC1 antibody (clone DF3, gift from Don Kufe, DFCI) in 100μl PBE and incubate it at 4C for 0.5-1 hour, then wash 3x with 200μl PBE to remove unbound antibody and resuspend it in 100μl PBE. Incubate, capture, and wash the cells as described above. Freeze captured fraction as MUC1+ cells. Transfer unbound cells and first wash into new Epp. Tubes and repeat magnetic capturing to ensure that all bead bound cells are removed. Take unbound cells and go to the next step.
  13. CD24+ cells: to the cell suspension add 100 μl prewashed beads prepared as follows: mix 100 μl prewashed Invitrogen/Dynal Pan mouse IgG (cat# 110-41) with 5-10 μl CD24 antibody (BD Pharmingen cat# 555426) in 100μl PBE and incubate it at 4C for 0.5-1 hour, then wash 3x with 200μl PBE to remove unbound antibody and resuspend it in 100μl PBE. Incubate, capture, and wash the cells as described above. Freeze captured fraction as CD24+ cells. Repeat the step if you anticipate high CD24+ cells that overwhelmed the beads and remained in the unbound fraction as carryover from this step can affect purity of CD44+ population. Transfer unbound cells and first wash into new Epp. Tubes and repeat magnetic capturing to ensure that all bead bound cells are removed. Take unbound cells and go to the next step.
  14. CD10+ cells: From step 13 take unbound cells, pellet cells by cf. and resuspend in 200-1000 μl PBE the same way as before and add 100 μl of pre-washed/bound CD10 beads (bind 5-10 μl CD10 antibody-Dako cat#M0727- to 100 μl beads- Invitrogen/Dynal Pan mouse IgG cat# 110-41 - in 100 μl PBE at 4C for at least 1 hour) to the cells and capture/wash as described above. The captured cells are the MYOEPITHELIAL CELLS. Repeat the step if you anticipate high CD10+ cells that overwhelmed the beads. Wash bound cells 3x200 ul PBE then freeze captured cells immediately on dry ice. Transfer unbound cells and first wash into new Epp. Tubes and repeat magnetic capturing to ensure that all bead bound cells are removed. Take unbound cells and go to the next step (CD44+ cell isolation).
  15. CD44+ cells: to the cell suspension add 100 ul prewashed beads prepared as follows: mix 100 μl prewashed Invitrogen/Dynal Pan mouse IgG (cat# 110-41) with 5-10 μl CD44 antibody (BD Pharmingen cat# 550988) in 100μl PBE and incubate it at 4C for 0.5 -1 hour, then wash 3x with 200μl PBE to remove unbound antibody and resuspend it in 100μl PBE. Incubate, capture, and wash the cells as described above. Freeze captured fraction as CD44+ cells. Transfer unbound cells and first wash into new Epp. Tubes and repeat magnetic capturing to ensure that all bead bound cells are removed. Take unbound cells save it as “unbound organoid” fraction.
 

Testing the purity of the cells:
For testing purity of different epithelial cell fractions and stroma, isolate mRNA and DNA from the cells using uMACS kit (Miltenyi Biotec), flow through is used for DNA prep. For epithelial fractions, confirm purity of the fractions by performing RT-PCR on cDNA made from 10% of the polyARNA using primers for the following genes: CD45 (leukocyte marker), MUC1, CD24, CDH1 (luminal markers), CD10 (myoepithelial marker), and IGFBP7, LRP1, CD44 (progenitor markers), and CDH5 (endothelial marker). Design small (~<200bp) amplicons, so that you get efficient amplification even from very small amount of template. Also the amplicon size should be about the same for each marker tested to avoid differences due to efficiency of PCR.  RT-PCRs are performed for 25/30/35 cycles for each gene; take out a 5ul aliquot after 25/30/35 cycles and run them side-by-side on gel. Both CD24 and CD44 fractions should be negative for CD45, use the leukocytic fraction as a positive control for these genes. The CD24 fraction has to be negative for IGFBP7 and MUC1, and can be somewhat positive for CD44, while the CD44+ fraction has to be negative for MUC1, CD24, CD10, and positive for CD44, and IGFBP7 and LRP1.
For stromal cells, RT-PCR should be performed for genes CD45, CDH5, and CDH1. A pure stromal fibroblast-enriched fraction should be negative for these markers.
 
 
RT-PCR:
  1. For making cDNA
 
for small amounts of polyA RNA use the OMNISCRIPT or SENSISCRIPt kits from Qiagen
after cDNA synthesis is done (in 20ul volume) Add 180ul of H2O and use 1-2ul for 25ul PCR reaction
 
PCR reaction:

Template cDNA 1-2
10X PCR buffer                            2.5ul
25mM dNTP                                 1.5ul
DMSO                                          1.5ul
Forward primer (350ng/ul)            0.5ul
Reverse Primer (350 ng/ul)         0.5ul
Plat. Taq                                      0.25ul
H2O                                            to 25ul
 
PCR condition:
 
950C 3min
950C, 30sec, 640C,30sec, 700C 30sec (3 cycles)
950C, 30sec, 610C,30sec, 700C 30sec (3 cycles)
950C, 30sec, 580C,30sec, 700C 30sec (35 cycles)
700C, 5min
 
Extension time should be adjusted according to product sizE (1kb/min).