Recovery of Highly Functional Human Hematopoietic Stem and Progenitor Cells from 27-Year Old Cryopreserved Umbilical Cord Blood

Umbilical cord blood (CB) is a promising source of hematopoietic stem cells (HSCs) and progenitor cells (HPCs) for hematopoietic cell transplantation (HCT). CB is cryopreserved following collection, allowing for the amassment of nearly 800,000 CB units in public banks and an estimated 4 million in private banks. Cryopreservation of CB is essential for successful CB transplantation, and it is critical to understand how long CB can be stored with efficient recovery of engraftable HSCs. Prior work demonstrated efficient recovery of engraftable HSCs from 10-, 15-, and 21-year old cryopreserved CB. Here, we demonstrate efficient recovery of highly engraftable HSCs from 27-year old cryopreserved CB units.

3 CB units cryopreserved for >27 years were thawed with clinical protocols, and total nucleated cells were recovered with 76-87% viability. 3-6 fresh CB units were used as a comparison for cryopreserved units. Immunophenotyping assays to determine numbers of HSCs/HPCs showed that cryopreserved CB units had on average 4.25x10³ CD34+ cells per 1x10⁶ nucleated low density CB cells (LDCB), similar to fresh CB with 4.55x10³CD34+ per 1x10⁶ LDCB. Within the CD34+ population, cryopreserved CB when compared to fresh CB had on average equal or greater numbers of stringently immunophenotypically defined HSCs (9.1x10³ vs 6.2x10³), multipotent progenitors (MPPs) (2.1x10⁵ vs 8.6x10⁴), and multilymphoid progenitors (MLPs) (1.9x10⁴ vs 1.0x10⁴) per 1x10⁶ CD34+ cells. Conversely, there were fewer numbers of common myeloid progenitors (CMPs) (2.9x10⁴ vs 1.1x10⁵), megakaryocyte-erythroid progenitors (MEPs) (4.5x10⁴ vs 1.0x10⁵), and granulocyte macrophage progenitors (GMPs) (4.8x10⁴ vs 1.4x10⁵). Colony forming unit (CFU) assays demonstrated that cryopreserved CB units exhibited lower functional HPC numbers per 1x10⁶ CD34+ enriched cells including CFU-GM (1.9x10⁴ vs 8.3x10⁴), BFU-E (1.2x10⁴ vs 3.4x10⁴), and CFU-GEMM (1.1x10⁴ vs 5.8x10⁴). In ex vivo analysis of HSC/HPC proliferative capacity, cryopreserved CB units showed similar or greater capacity for expansion in growth stimulating media compared to fresh CB, including HSCs (7.7-fold vs 6.2-fold increase), MPPs (3.5-fold vs 4-fold increase), MLPs (44-fold vs 1.7-fold increase), CFU-GM (29-fold vs 13.3-fold increase), and CFU-GEMM (28.6-fold vs 4-fold increase).

For in vivo functional HSC analysis, 10,000 CD34+ enriched cells were transplanted to sublethally irradiated NSG mice. Cryopreserved CB units exhibited early and late human engraftment in peripheral blood (PB) at months 1, 2, 4, and 5. Bone marrow (BM) human chimerism at month 5 ranged from low (0.5%) to robust levels of engraftment (52.7%). Limiting dilution analysis revealed that 2 cryopreserved CB units had similar SCID repopulating cell (SRC) frequency, ranging from 314 to 1392 SRCs per 1x10⁶ CD34+ cells while the third had a much lower 65 to 393 SRCs per 1x10⁶ CD34+ cells. The third CB unit also exhibited lower average human chimerism in the BM (0.55%) than the other 2 units (15% and 16.2%). However, the SRC frequencies in all 3 CB units are similar to those in fresh CB units published by our group and others (range 51 to 929 SRCs per 1x10⁶ CD34+ cells). Secondary transplants showed that all 3 CB units have secondary repopulation capacity demonstrated by human chimerism in PB.

These data suggest functional HSCs with short- and long-term engrafting and sustained self-renewal capacity can be cryopreserved for at least 27 years. We show that different cryopreserved CB units contain variable functional HSC frequency. Similar results in ex vivo characterization suggest cryopreservation is not the cause of these functional differences, but rather there were functional differences in the CB units prior to cryopreservation. This underscores the need to have reliable measures of CB functionality prior to cryopreservation for more efficient selection of donor units and to test potential effects of cryopreservation on individual units. Transcriptomic studies are ongoing to determine molecular programs that may contribute to differences in SRC frequency. These data will provide insight into mechanistic explanations for differences in HSC function in different CB units.

Authors’ note: This project was conceived by the authors prior to the death of Hal E. Broxmeyer. Dr. Broxmeyer initiated the study as a continuation of his life's work demonstrating the utility of CB for HCT. This work is dedicated to Hal.

No relevant conflicts of interest to declare.