Bone marrow mononuclear cell fraction (BM MNC) is commonly used source for cell therapies. Only last year (2012) there were 24 clinical trials, assessing BM MNC in variety of diseases. Density gradient centrifugation (DGC) by a Ficoll (Ficoll-Paque) used in all clinical trials and most pre-clinical studies to isolate MNC. It is well known that DGC procedure associated with a big cell loss. This issue could be critical if cell therapy efficacy depends on administered cell number or if MNC undergo further processing. Also, differences in DGC procedures by Ficoll could affect cell function, and therefore, cell therapy outcome.

The recent study, published in PLoS ONE, addresses the problem of cell yield after DGC and compares 3 methods for BM MNC cell isolation. The authors compared 3 following methods for MNC isolation:

  • Ficoll-Paque DGC
  • Percoll DGC
  • immunomagnetic polymorphonuclear cells (PMNs) depletion

The study was done on rat lysed (red blood cells depleted) bone marrow. The authors assessed cell composition after each step of precessing by flow cytometry.  Lysis depleted 90% of erythroid (CD45-) cells.

Freshly isolated rat bone marrow consists of 60% CD45+ and 40% CD45- cells. Lysis caused a significant decrease of CD45- (−85%) and CD45+ (−36%) cells.

Percoll was used as a first alternative to Ficoll. Distribution of cells in Percoll layers:

For magnetic depletion, the authors used CD15 antibody for labeling of PMNs.

As expected, all isolation protocols caused a nearly complete depletion of PMNs (Ficoll: 2.5±1.3%, Percoll 11.5±6.1% and MACS 9.5±8.5% recovery of total PMNs in lysed bone marrow) and, to a lesser extent, of the CD45- population, particularly of the erythroid cells.
Depletion of the CD45- cell population was significantly more severe after Ficoll DGC compared to Percoll DGC and MACS (14.6±8.3% versus 41.2±5.1% and 51.0±6.0% recovery of total CD45- cells in lysed bone marrow).

The loss of MNC fraction from these 3 methods was the following:

  • after Ficoll DGC = 14.6%
  • after Percoll DGC = 41.2%
  • after MACS = 51%

Further, researchers measured progenitor/ stem cell activity by CFU-GM and CFU-F (for hematopoietic and mesenchymal progenitors respectively) colony assays. Unlike, Percoll and MACS, Ficoll DGC led to ~2 times enrichment in hematopoietic colonies number. Frequency of CFU-GM colonies after Percoll and MACS didn’t differ from whole unseparated lysed bone marrow.

However, considering the significant different BMNC yields (Figure 4A), only one-third of all HSCs were harvested after Ficoll DGC. By contrast, after adjusted Percoll DGC and immunomagnetic separation, the HSC yield was two-fold higher as compared to Ficoll DGC and increased to two-third of the total amount in lysed bone marrow (Figure 5B).

As for mesenchymal progenitors activity:

As described previously [29], CFU-F frequency was extremely low in lysed bone marrow (0.00007±0.00002%) representing only 7 of non-hematopoietic progenitors per 1×10E7 bone marrow cells. We did not detect any CFU-F within the BMNCs after Ficoll, Percoll or immunomagnetic separation nor within the discharge of the latter two.

I wonder why Percoll and MACS didn’t enrich hematopoietic progenitors and why was no any CFU-F activity detected after any of these cell separation methods. The authors discuss one possible explanation:

…it was recently described that most of the bone marrow MSCs were discarded during Ficoll DGC due to their aggregate nature [23], and that whole bone marrow, but not BMNCs obtained by Ficoll DGC improved functional recovery after experimental myocardial infarction [41].

Finally, the authors compared two magnetic separation methods (kits from Miltenyi Biotech and PluriSelect) for isolation of human BM MNC. They used combined immunodepletion of PMN (CD15+) and erythrocytes (CD235+). Magnetic sorting led to better erythrocytes depletion and leukocytes enrichment than conventional lysis method. Taking in account significance of red blood cells depletion from therapeutic cell products, these findings could be very useful for developers and cell trialists.

Both methods were effective to significantly decrease PMN numbers to less than 1% of the content in whole bone marrow and recovered almost 60% of the lysed bone marrow MNCs (Figure 6B). The combined depletion of erythrocytes and PMNs by either MACS (two-step approach) or PluriB (one-step approach) showed a comparable BMNC yield and purity…

They assess progenitors content by flow cytometry:

The subsequent isolation of BMNCs was responsible for a significant loss of both HSCs and MSCs (on average of 30% for HSCs and of 55% for MSCs). The decrease of HSCs was thereby mostly pronounced after one-step depletion of erythrocytes and PMNs using PluriB (Figure 6D). However, the proportion of progenitors to the amount of total nucleated cells (i.e. frequency) increased after BMNC isolation, simply as a consequence of selective PMN depletion.

Overall, this very interesting study can teach us how to improve bone marrow cell processing for therapeutic purposes.