- Top of page
- Materials and methods
AC133+ cells may represent an alternative source of transplantable haemopoietic progenitor cells to CD34+ cells. Here, we have addressed the characterization of umbilical cord blood (UCB) AC133+ cells and compared their immunophenotypic and functional features with those of UCB CD34+ cells. UCB AC133+ and CD34+ cell fractions were purified by magnetic cell sorting, analysed by flow cytometry, tested for their content in blast cell colony-forming units (CFU-Bl), erythroid and granulocyte–macrophage colony-forming units before and after expansion in the presence of various haemopoietic growth factor combinations. Median AC133+ cell yield was 62·3%, and median AC133+ population purity was 97·9%. AC133+ cells were found to contain significantly more CFU-Bl than CD34+ cells; furthermore, the replating efficiency, i.e. the number of CFU-Bl capable of generating secondary colonies, was higher in the former than in the latter cells. Both AC133+ and CD34+ cells displayed an increased ability to give rise to committed progenitors after 7-day expansion in liquid cultures. These data suggest that the AC133+ cell subset is a heterogeneous pool of immature and more differentiated cells that can be maintained and expanded in well-defined culture conditions. In comparison with CD34+ cells, UCB AC133+ cells appear to contain a higher number of early haemopoietic progenitors.
Expression of the CD34 cell surface antigen has allowed the identification of early haemopoietic progenitor cells in bone marrow, peripheral blood and umbilical cord blood (UCB). CD34 expression has been widely used for the selection of primitive haemopoietic progenitor cells by means of sophisticated strategies. The availability of these techniques has also fostered research on the ex vivo expansion of CD34+ haemopoietic progenitor cells, a procedure that could find clinical application in the setting of transplantation.
The function of AC133 is so far unknown. The expression of AC133 has been evaluated within the CD34bright haemopoietic stem and progenitor cells derived from human fetal liver, bone marrow and blood ( Yin et al, 1997 ), in normal donors undergoing granulocyte colony-stimulating factor (G-CSF) mobilization ( Durett et al, 1998 ), in chronic myeloid leukaemia ( Hömer et al, 1998 ), in acute myeloid ( Horn et al, 1998 ) and lymphoblastic leukaemia ( Snell et al, 1998 ) and in myelodysplastic syndromes ( Snell et al, 1998 ).
Because of the important role of new immunophenotypic markers in the characterization of stem/progenitor cells and the preliminary evidence that AC133 antibody could constitute an alternative approach to CD34 for the selection of cells capable of both short- and long-term engraftment, we have addressed in this study the characterization of AC133+ isolated from UCB cells and compared the functional features of these latter cells with those of CD34+ cells purified from the same source.
- Top of page
- Materials and methods
The immunophenotypic and functional characterization of a subset of CD34+ cells, which includes early haemopoietic stem/progenitor cells, is relevant to the potential applications of such cells in both laboratory and clinical investigation. FACS analysis showed that, in our UCB samples, AC133+ cells not only were exclusively contained in the population of CD34+ cells, but were also strictly related to the CD34bright cells. The latter cells are known to include most of the CD34+CD38− cells and contain the majority of CFU-GM, a proportion of CFU-GEMM and a minor population of BFU-E ( Mayani et al, 1993 ). Co-expression of AC133 and CD34 antigens might suggest that their expression is similarly regulated in haemopoietic development, even though the relationships between the two antigens have to be clarified. Further information on AC133/CD34 antigen expression in haemopoietic differentiation might be inferred from the characterization of immature cell populations in malignancies. For example, it has been reported that AC133 may be found on myeloid leukaemia blasts in the absence of CD34 ( Kratz-Albers et al, 1998 ) .
In our samples, AC133+ cells ranged from 0·3% to 2·4% of total MNCs. This range appears to be higher than that reported by others ( de Wynter et al, 1998 ). The higher percentage of AC133+ cells we found could be the result of physiologic variability, the modalities of sample collection and time of harvesting, as well as differences in laboratory procedures (percentage defined on unseparated samples vs. Ficoll gradient fractionated MNCs). Reports published so far on the AC133+ cell population in bone marrow, UCB and mobilized peripheral blood are limited ( de Wynter et al, 1998 ), and comparison between reported data is made difficult by the expression of the frequency of AC133+ cells as a percentage within the CD34+ cell population ( Yin et al, 1997 ; Durett et al, 1998 ). However, the high percentage of AC133+ cells that we also detected in the CD34+ cell population (median value 78·7%) and the finding that CD34+/AC133+ cells are capable of repopulating NOD/SCID mice ( de Wynter et al, 1998 ), coupled with the demonstration that AC133+ cells can be isolated at a high purity (> 91%), prompted us to investigate further the in vitro response to growth factors of this subset of stem/progenitor haemopoietic cells.
In this study, AC133+ cells isolated from UCB were found to contain 0·2–4·5% CD38− cells, and similar proportions (0·5–5%) of the latter cells were detected in CD34+ cells from the same source in a previous study by our group ( Pasino et al, 1998 ). Whether or not CD38− cells found in AC133+ vs. CD34+ UCB cells are completely, or only partially, overlapping remains to be ascertained in further studies. Likewise, how other early markers of primitive haemopoietic progenitors, such as c-kit or Thy-1, are comparatively expressed in AC133+ vs. CD34+ cells deserves further investigation.
Conditio sine qua non for the use of UCB as an alternative source of transplantable cells is that the AC133+ cell subset contains both short- and long-term reconstituting cells. A major advantage could also derive from the demonstration that AC133+ cells are more enriched in early progenitor cells than CD34+ cells. Assays suitable for the detection of human putative stem cells have been developed, including CFU-Bl ( Leary & Ogawa, 1987). Both AC133+ and CD34+ cells isolated by Mini-Macs from UCB contained an aliquot of cells capable of giving rise to second-generation colonies. Comparing equal concentrations of cells seeded, we found that CFU-Bls derived from AC133+ cells were significantly more numerous than CFU-Bls derived, under identical conditions, from CD34+ cells (Table IV). Moreover, the replating efficiency, measured as the number of CFU-Bls capable of generating secondary colonies, was significantly higher for AC133+-derived CFU-Bls than for CD34+-derived CFU-Bls. During the short-term HGF-supplemented culture, the absolute number of cells capable of generating secondary colonies, normalized for the absolute number of counted cells and compared with the input value, was significantly higher for AC133+-derived CFU-Bls than for CD34+-originated CFU-Bls. Taken together, these data suggest that the early stem/progenitor cells are present and enriched in the pool of AC133+ isolated cells compared with CD34+ isolated cells.
Ex vivo expansion has been regarded as a potential tool for increasing the number of more immature haemopoietic progenitor cells. However, the experimental conditions employed have rarely been encouraging in obtaining an effective increase in stem and early progenitor cells ( Piacibello et al, 1997 ). This failure has been ascribed to the lack of accessory cells, to the predominant effect of induction of differentiation by the cytokines added ( Haylock et al, 1992 ) or, alternatively, to the presence of very immature, deeply quiescent cells, which proliferate after a more prolonged time and/or do not respond to known cytokines ( Berardi et al, 1995; Hao et al, 1996 ). In our stroma-free culture system, the simple combination of two early acting cytokines, i.e. SCF and IL-6, resulted in an output of CFU-Bls that was augmented when FL was associated. This finding represents an additional demonstration of the capability of FL to sustain the growth of primitive progenitors ( Piacibello et al, 1997 ). The presence of lineage-restricted growth factors in the 7-day expansion culture before the CFU-Bl assay did not further enhance the CFU-Bl output (P = 0·201, Student's t-test for paired data).
It would appear that AC133+ cells behave differently from the whole pool of CD34+ cells, in the expansion culture conditions used, in terms of the production of clonogenic precursors from primitive cells. Undoubtedly, the AC133+ cell fraction is a heterogeneous pool of immature and more mature cells, as we have demonstrated in clonogenic assays in which CFU-GM, BFU-E and CFU-Bl colonies were generated. However, the possibility of maintaining and expanding, although with small numbers, more immature elements adds new support to the hypothesis that the AC133+ cell subset could become increasingly important in all those clinical procedures that benefit from the availability of a pool of expanded and/or gene-manipulated immature haemopoietic cells.