SEARCH

SEARCH BY CITATION

Keywords:

  • Stem cells;
  • Hematopoietic stem cells;
  • T cells;
  • NK cells

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Side-population (SP) cells are a recently described rare cell population detected in selected tissues of various mammalian species, but not yet described in the peripheral circulation. In the present study, we have identified for the first time SP cells in lineage-negative adult human blood and have provided an extensive functional and immunophenotypic characterization of these cells. Adult peripheral blood was depleted of mature leukocyte cell types by density gradient and immunomagnetic separation. The SP cell population was identified by its characteristic Hoechst 33342 profile. Immunophenotypic analysis revealed that blood SP cells expressed high levels of CD45, CD59, CD43, CD49d, CD31, and integrin markers and lacked CD34. Highly purified SP cells were put into cobblestone area-forming cell (CAFC), long-term culture-initiating cell (LTC-IC), and liquid cell culture assays; repopulating assays were performed utilizing nonobese diabetic/severe combined immunodeficient mice. Circulating SP cells were shown to exhibit verapamil sensitivity and a low growth rate. LTC-IC, CAFC, and engraftment assays indicated that circulating SP cells had lost the multipotentiality described in murine bone marrow SP cells. However, outgrowth of mature cell types from liquid cell culture suggests the presence of common lymphoid (T and natural killer) and dendritic cell precursor(s) within circulating SP cell populations. The absence of SP cell growth in the LTC-IC, CAFC, and repopulating assays might be intrinsic to the tissue source (marrow versus blood) or species (mouse versus human) tested. Thus, human blood SP cells, although rare, may serve as a source of selected leukocyte progenitor cells. The immunophenotype of circulating SP cells may provide clues to their seeding and homing capacity.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Hematopoietic stem cells (HSCs) are characteristically quiescent, multipotent cells with the capacity for both self-renewal and differentiation. After development in the fetus, HSCs reside in adult bone marrow (BM) and serve to replenish lymphoid [1], megakaryocytic, erythroid, and myeloid [2] hematopoietic lineages throughout the lifetime of the host. Additional studies have demonstrated the possible plasticity [3] of BM-derived HSCs, as evidenced by their capacity to develop into hepatocytes [4–, 6]; skeletal muscle [7]; glial cells [8]; and epithelial cells of the lung, gastrointestinal tract, and skin [9]. Murine HSCs lack the immunophenotypic antigens that define mature peripheral blood cell lineages (i.e., are lineage negative [Lin]), but do express Sca-1 (Ly-6/E), CD117 [10], CD27, and CD90 [11] and have the capacity to rescue lethally irradiated animals [12,, 13].

Murine side-population (SP) cells were initially defined by their relatively active efflux of Hoechst 33342 dye, via a p-glycoprotein multidrug/ATP-binding cassette transporter protein [14–, 16]. BM-derived SP cells were shown to be enriched 1,000-fold for in vivo lymphoid and myeloid hematopoietic reconstituting activity. These earliest studies showed SP cells to be about 10% of the Sca-1+ CD34 murine BM, expressing CD117, CD43, and CD45 [17]. Cell cycle analysis of SP cells revealed their relative quiescence; the few SP cells in growth phase (<2%) had a repopulating capacity indistinguishable from those in G0/G1. In other studies, SP cells obtained from murine muscle were shown to express Sca-1, but lacked the CD117+, CD43+, and CD45+ immunophenotype seen in BM. Muscle-derived SP cells were capable of reconstituting the hematopoietic compartment, but not as efficiently as BM-derived SP cells, suggesting the importance of microenvironmental factors on plasticity-related SP cell function [18]. More recent studies, showing that murine BM-derived SP cells could regenerate injured cardiac muscle, indicate that these cells uniformly express CD31 in addition to CD117 [19]. SP cells also have been described in miniature swine [17] and are present among murine neural progenitor cells [20].

Limited immunophenotypic information is available about antigens expressed on primate SP cells. Human Lin BM SP cells have been characterized as CD45+, variably CD38+, CD34low/–, CD90low/–, and CD117low/–. Rhesus Lin BM SP cells are CD45+, CD59+, major histocompatibility complex class-I+, and variably CD7+, CD38+, and HLA-DR+. Rhesus SP cells can acquire CD34 in vitro, include cells with long-term culture initiating cell (LTC-IC) capacity, and are capable of generating T cells after culture on primate thymic stroma [17]. SP cells derived from human cord blood are predominantly CD34 and are CD7+, CD11b+, and CD45RA+ [21].

Here, for the first time, we identify SP cells obtained from normal human peripheral blood. These cells were characterized immunophenotypically and functionally in comparison to those obtained from other tissues and species.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Cell Preparation—Lineage Depletion

Peripheral blood white cells (buffy coats) were obtained from normal healthy volunteer donors at the Massachusetts General Hospital blood bank. These preparations (25-35 ml) were diluted 1:10 with phosphate-buffered saline (PBS) and layered over density gradient separation medium (Isolymph; Galard Schlesinger; Garden City, NY; http://www.gallard-schlesinger.com) to separate polymorphonuclear cells and mature erythrocytes. The remaining lymphocytes and monocytes were incubated for 15 minutes with titered volumes of biotinylated antibodies to CD3, CD5 (Becton Dickinson; San Jose, CA; http://www.bd.com), CD19, CD7, CD14, CD34 (Caltag; Burlingame, CA; http://www.caltag.com), and CD16 (PharMingen; San Diego, CA). This cocktail of monoclonal antibodies (mAbs) was shown by flow cytometry to select out, with high purity (>98%), mature T, B, and natural killer (NK) cells, monocytes, and CD34+ cells from the residual Lin cells on a MACS B6 column (Miltenyi Biotec; Bergisch Gladback, Germany; GmbH; http://www.miltenyibiotec.com).

Cell Staining and Flow Cytometry

To identify SP cells for cell sorting, the Lin CD34 cells passing through the depletion column were stained with 5 μg/ml of the A-T intercalating Hoechst 33342 dye [22–, 24] for 2 hours at 37°C, washed, and resuspended in PBS containing 2% fetal calf serum and 1 mM HEPES. Prior to cell sorting, 2 μg/ml propidium iodide ([PI] Sigma Chemicals; St. Louis, MO; http://www.sigmaaldrich.com) were added; cells demonstrating uptake of this dye were omitted from analysis and subsequent purification. SP cells were identified and electronically gated on their characteristic light-scatter properties and singular Hoechst 33342 red versus blue fluorescence emission pattern [14], after excitation with 100 mW of 350-360 nm ultraviolet light on a HiPerFACS Vantage Cell Sorter (Becton Dickinson). SP fluorescence emissions were directed toward a 610-nm dichroic filter, then captured simultaneously through both a 450-nm band-pass (blue) and a 675-nm long-pass (red) filter on a linearly amplified fluorescence scale. Cell viability was assessed either by PI or 7 aminoactinomycin-D dye exclusion, also captured through the 675-nm filter. Circulating human SP cells were present at approximately 9-35 × 10−6 peripheral blood mononuclear buffy coat cells (PBMCs).

Immunophenotypic characterization of SP cells was done with titered fluorescein isothiocyanate (FITC), phycoerythrin (PE), or allophycocyanin mAb conjugates of CD2, CD9, CD11a, CD11b, CD11c, CD31, CD33, CD38, CD44, CD45, CD49d, HLA-DR, CXCR4, CCR5 (Becton Dickinson), CD35, CD59, CD90, CD117 (Beckman-Coulter-Immunotech; Miami, FL; http://www.beckman.com), AC133 (Miltenyi), kinase insert domain receptor ([KDR] Sigma), CD43 (Dako; Carpinteria, CA; http://www.dako.dk), and HLA-class 1 (One Lambda, Inc.; Canoga Park, CA; http://www.onelambda.com). Up to 106 cells were collected for each analysis. Both cell sorting and immunophenotypic characterization of Lin SP cells involved computer gating on light scatter to omit debris and contaminating cells on 0.01%-0.2% of the characteristic SP tail presented in a Hoechst 33342 red versus blue bivariate plot. When testing for multidrug resistance 1 (MDR1)-like sensitivity, verapamil (verapamil hydrochloride, 50 μg/ml; Sigma) was additionally added. To enhance visualization of selected SP populations, cells with the highest expression of Hoechst dye were electronically omitted from the analysis.

Cell Cycle Analysis

Lin SP cells were stained with Hoechst 33342 dye, identified and sorted as described, and fixed with 70% ethanol for 24 hours. Cells were washed in PBS and then stained with 20 μg/ml PI and 1 mg/ml RNAse (Type IIA; Sigma) and collected on an LSR configured with CellQuest Pro (Becton Dickinson) software.

Liquid Cell Cultures

Cultures of SP cells sorted to a >99% purity [25] were initiated with a tenfold excess of autologous x-irradiated (3,000 rads) feeders and optimal mitogenic concentrations (1-2 mg/ml) of phytohemagglutinin ([PHA] Leukoagglutinin; Pharmacia; http://www.pnu.com) within the wells of 48-well culture plates, as previously described. All cultures were grown at 37°C in 5% CO2 in RPMI 1640 with L-glutamine (Whittaker MA Bioproducts; Walkersville, MD; http://www.biowittaker.com) supplemented with 10% fetal bovine serum (FBS), HEPES (2.4 g/l), nonessential amino acids (1 mM/l), sodium pyruvate (1 mM/l), gentamicin (0.1 mg/ml) (GIBCO; Grand Island, NY; http://www.lifetech.com), and 1,000 U/ml recombinant interleukin-2 ([IL-2] Cetus; Emeryville, CA). Control wells containing only irradiated feeder cells were established at the initiation of all cell cultures to exclude outgrowth of feeder cells as a source of cell contamination. SP and control cell cultures were subsequently restimulated with PHA and auto- or allogeneic irradiated feeder cells approximately every 3 weeks to maintain cell growth.

LTC-IC and Cobblestone Area-Forming Cell (CAFC) Assays

LTC-IC cultures were assessed according to described methods [26] and methods developed within our laboratory [27]. Briefly, low-density BM cells were cultured at 37°C for 3-4 days prior to transfer to 33°C in long-term culture (LTC) medium (α-minimal essential medium with 12.5% horse serum, 12.5% FBS, 0.2 mM I-inositol, 20 mM folic acid, 10−4 M 2-mercaptoethanol, 2 mM L-glutamine [StemCell Technologies Inc.; Vancouver, BC, Canada; http://www.stemcell.com], and 10−6 M hydrocortisone [Sigma]). The confluent stroma layer was trypsinized, irradiated (15 Gy), and subcultured in 96-well flat-bottomed plates at a density of 1.25 × 104/well. Within 1 week, sorted Lin CD34 SP and control (Lin CD34+) cells from the same human blood donor were cultured at 33°C for 5 weeks with half-volume medium changes weekly. CAFCs were assayed after 3 and 5 weeks of culture. Culture plates were then centrifuged. Iscove's modified Dulbecco's medium semisolid medium (StemCell Technologies) containing 0.9% methylcellulose, 30% FBS, 1% bovine serum albumin, 10−4 M 2-mercaptoethanol, and 2 mM L-glutamine, supplemented with 20 ng/ml IL-3, 20 ng/ml GM-CSF, 50 ng/ml stem cell factor ([SCF] R & D Systems; Minneapolis, MN; http://www.rndsystems.com), 20 ng/ml G-CSF, and 3 U/ml erythropoietin (Amgen Inc.; Thousand Oaks, CA; http://www.amgen.com) was overlaid. Following 10 days at 37°C, 5% CO2, colonies were quantitated by phase-contrast microscopy, and the absolute number of LTC-ICs was calculated by Poisson statistics.

Nonobese Diabetic/Severe Combined Immunodeficient (NOD/SCID) Engraftment Assay

Seven-week-old female NOD/LtSz-Prkc-scid/scid (NOD/SCID) mice (Jackson Laboratory; Bar Harbor, ME; http://www.jax.org) were sublethally irradiated with 3 Gy of radiation (137Cs source) and intravenously injected with purified SP cells, with a purity of >75% (3 × 103 cells/mouse), and 20 Gy-irradiated ficolled PBMCs (carrier cells, 1 × 105/mouse). Control mice were injected with PBMCs (1 × 105/mouse) handled similarly as the carrier cells. White blood cells and BM cells were prepared at week 9 after transplant and were analyzed for the presence of human cells by flow cytometry. Cells were stained with PE-conjugated anti-mouse Ly-5 (rat IgG2b anti-mouse CD45; PharMingen) in combination with FITC-conjugated anti-human CD45 (Becton Dickinson). Human CD45+ cells unstained by anti-mouse CD45 mAb were considered to be of human (donor) origin. Nonspecific binding of labeled mAbs was blocked with 2.4G2 (rat anti-mouse FcγR mAb). FITC-HOPC1 (PharMingen), and PE-rat IgG2a (PharMingen), either lacking reactivity to mouse or human cells, served as negative control antibodies.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Immunophenotype of Circulating SP Cells—Lin SP Cells do not Express CD34

If CD34 antibody was excluded from the Lin separation cocktail, analysis of the SP cells in the remaining blood revealed these to be predominantly CD45bright+, without expression of CD34 (Fig. 1, R1). Characteristically stained CD45dim+ CD34+ cells could be identified in the remaining Lin blood, arising from a portion of the Hoechst red versus blue plot that did not include any SP cells (Fig. 1, R2). Analysis of an alternate area of the Hoechst red versus blue plot indicated the absence of CD34+ cells within this location as well (Fig. 1, R3).

thumbnail image

Figure Figure 1.. Immunophenotype of Lin SP and Lin non-SP cells.Circulating Lin SP cells are CD34. Electronic gating on the SP tail (R1 = 0.1% cells; top right panel, Hoechst blue versus Hoechst red dyes) indicates these are predominantly CD45bright+ cells, completely lacking expression of CD34 (top left panel; CD45 versus CD34. The percentages of cells in each immunofluorescence quadrant are shown, throughout). Electronic gates on R2 and R3, two alternate regions in the Hoechst blue versus Hoechst red plot show the CD45dimt+ CD34+ cells (3.44%, bottom left panel) may be found within the boundaries of gate R2, but not gate R3. The percentage of clearly distinguishable CD34+ cells (3.44%) is separable from a fraction of events (0.49%) contiguous with CD34 cells that fall just past the PE+ threshold. The relatively greater percentage of blood CD34+ cells seen in the lower left plot is the result of lineage depletion of mature cell types.

Download figure to PowerPoint

Further immunophenotypic analysis of Lin human peripheral blood SP cells stained with Hoechst 33342 dye revealed a wide spectrum of cell-surface-antigen expression. Circulating SP cells were, on average, greater than 80% CD59+, CD45+, CD43+, CD49d+, CD31+, and HLA-class 1+ (n = 6 analyses). From 30%-79% of SP cells were CD44+, CD35+, CD33+, CD25+, CD18+, CD11a/b/c+, CD2+, and HLA-DR+ (n = 5 analyses), while 15%-29% of SP cells were CD117+, CD90+, CD62L+, CD54+, and CD38+ (n = 4 analyses; Fig. 2). SP cells were also examined for CD122, CD9, KDR, CXCR4, and CCR5, but these antigens were infrequently detected (2%-11% positive; n = 2-3 analyses).

thumbnail image

Figure Figure 2.. Immunophenotypic analysis of circulating Lin CD34 human SP cells.Shown are the average percentages of positive cells, as indicated by scale on a twelve o'clock ray. Expression of selected antigens decreases in a clockwise direction. The number of analyses per antigen tested is shown inResults.

Download figure to PowerPoint

Inhibition of MDR1-Like Gene Activity by Verapamil—Cell Cycle Analysis

The MDR1 gene encodes a p-glycoprotein, which is expressed in a variety of normal cell types. This, and other ATP-binding cassette proteins are responsible for biliary excretion, lipid translocation, and drug elimination and for the exclusion of Hoechst 33342 dye from SP cells, permitting their identification [14,, 15,, 28]. Figure 3 demonstrates that the capacity to identify SP cells was abrogated when Lin peripheral blood cells were stained with Hoechst 33342 dye in the presence of verapamil.

thumbnail image

Figure Figure 3.. Circulating SP cells are verapamil sensitive.Circulating human SP cells, like those identified from other species and different tissues, possess an active MDR1-like pump. A) The control cells (SP cells are 0.2% of dot plot) were unexposed to verapamil. B) Abrogation of SP cell visualization after incubation with 50 μg/ml verapamil, as described inMaterials and Methods.

Download figure to PowerPoint

Figure 4 indicates that few cells were in growth phase in the SP cells, or in any of the remaining peripheral blood Lin subpopulations that stained with greater intensities of Hoechst 33342 dye.

thumbnail image

Figure Figure 4.. Circulating SP cells were relatively quiescent.Lin CD34 SP cells and cells with increased Hoechst dye uptake were in G0/G1 growth phase. Examination of the SP cell cluster (R1) and three alternate non-SP clusters (R2-R4) indicated all to be in a similar (S phase <1%), resting cell cycle phase.

Download figure to PowerPoint

Liquid Cell Culture

On four separate occasions, an average of 2,800 Lin CD34 purified SP cells were distributed into 24-well plates containing tenfold excess autologous x-irradiated feeders stimulated with PHA and supplemented with 1,000 U/ml IL-2 (Table 1, Figs. 5 and 6, Figure 6.). The resulting cultures were sufficiently cellular to permit earliest analysis at day 27 postpurification and were periodically analyzed over a subsequent 2-month period, based on available cellularity. Outgrowth of feeder cells was discounted based upon the absence of growth in cultures containing only x-irradiated feeder cells.

thumbnail image

Figure Figure 5.. One month postpurification of an SP cell population (experiment 4), two distinct light-scatter populations were identified in liquid cell cultures.Lymphocytes in gate 1 (51% of cells) consisted of CD3+ T cells (A, black dots, 62% of gate) expressing CD7 (B) bright CD45 (C, E), and CD2 (D). CD56+CD3 cells (A, gray dots, 33% of gate) were CD7dim+ or CD7 (B), lacked CD16 and CD2 (D), and had a lower expression of CD45 (E). No expression of CD34 was detected on these lymphocytes (C). Larger cells in gate 2 (36% of cells) coexpress CD33, CD83 (F), CD34, and CD45 (G). Very high levels of HLA-DR were also expressed on these cells (not shown). Abbreviation: FSC = forward scatter.

Download figure to PowerPoint

thumbnail image

Figure Figure 6.. Day 34 postpurification, only lymphocytes were identified in the liquid cell culture obtained from SP cells purified from experiment 1.CD3+ T cells (black dots throughout, approximately 4% of cells) uniformly expressed CD2 with a predominance of CD8+ T cells detected; the cells also expressed CD5 (not shown). The majority of cells in the plot were CD16+CD3 (gray dots, approximately 96% of cells) cells, shown expressing CD2 with a subpopulation of CD8dim+. Additionally, these cells uniformly expressed CD56 and CD7 (not shown). The percentages of cells in each quadrant are indicated throughout the figure. These results are similar to those from day 27 in Table1.

Download figure to PowerPoint

Table Table 1.. Immunophenotype of liquid-cultured SP cells
   Percentage positive cells in lymphocyte light-scatter gate
ExperimentDay post-purificationInput number of SP cellsCD45+CD2+CD3+CD56+CD3CD34+TdT+HLA-DR+
1272,30099961476<1<1
2563,300947618<112
3692,800999711<138
4342,82099696233<144

At initial analysis in all four experiments, all lymphocytes expressed CD45 and when assayed in experiments 1 and 4, CD2. These cell cultures consisted of mixtures of CD3+ T cells and CD3CD56+ (NK) cells. In experiment 1, NK cells predominated (Fig. 6). In experiments 2, 3, and 4, T cells were the majority (>63%) cell type. The T cells present expressed CD3, CD7, T-cell receptor αβ, and CD5, with more CD8+ T cells detected than CD4+ T cells. There was no CD34 or terminal deoxynucleotidyl transferase (TdT) detected on the T cells.

The NK cells from experiment 1 demonstrated lytic capacity with efficient killing of K562 targets (data not shown) approximately 3 months postpurification. These cells were noted to express CD2, CD16, CD45RA, and CD8dim+, characteristic of normal circulating NK cells (Fig. 6). Thirty-four days after initial culture, cells from experiment 4 were shown to lack the commonly detected CD2, CD16, CD7bright+ of mature NK cells, although variable levels of CD7 were evident on a subpopulation (Fig. 5). These cells also expressed lower levels of CD45 but could not be characterized further due to cessation of cell growth.

In experiments 3 and 4, cells with the light-scatter characteristics of dendritic cells were detected after approximately 1-2 months in culture. These cells expressed CD45, CD83, CD33, CD34, and very high levels of HLA-DR (Fig. 5).

In Vitro and In Vivo Studies

In the presence of BM feeder layers containing myeloid-inducing cytokines, SP cells contained approximately a twelvefold reduced frequency of LTC-ICs compared with autologous control CD34+ cells (31.01 versus 370.52 per 100,000 cells; n = 2). In two experiments, CAFCs were detected originating from CD34+ control cells at both 3 and 5 weeks; however, no CAFCs were detected from cultures established with SP cells at these time points.

Human cells were virtually undetectable in NOD/SCID mice transplanted with either sorted SP cells (n = 3) or control PBMCs (n = 1). Although scant human cell engraftment was detected in the BM 9 weeks after transplant, the levels were extremely low in both groups (human CD45+ cells ≤0.05%).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

SP cells are a rare cell population that has been identified in a variety of tissue types variously obtained from primate, rodent, and porcine sources. BM-derived murine SP cells have been characterized by their exceptionally high repopulating activity when injected into lethally irradiated animals. However, relatively little information is available concerning the immunophenotype of SP cells. Little has been reported regarding the cell-surface-antigen expression of human Lin BM-derived SP cells other than their expression of CD45 and variable level of CD38; immunophenotypic data from nonhuman primates is similarly scant [17]. The data presented here serve to first identify and more fully immunophenotype SP cells in human blood, and to distinguish the differentiative and functional capacities of these cells.

SP cells are characteristically identified by their predilection for active efflux of Hoechst 33342 dye, and the abrogation of this capacity by exposure to verapamil. Data presented here conclusively show that Lin SP cells obtained from adult human blood possess a verapamil-sensitive Hoechst efflux pump similar to SP cells obtained from other tissues and species [17]. Other described functional attributes of SP cells are their low proliferative rate, repopulating capacity, and variable plasticity, based upon tissue origin [7].

While circulating human SP cells also demonstrate low proliferative capacity, what appears distinctive is their inability, relative to autologous CD34+ cells, to grow in an in vitro LTC-IC assay. This is in contrast to rhesus BM SP cells, which, when placed in an LTC-IC assay, developed into myeloid, erythroid, and megakaryocytic cells at markedly greater frequencies than autologous control CD34+ cells [17]. This difference between human and rhesus may be related to either an intrinsic difference between the species, requirement for some undefined cytokine, or specific tissue source of SP cells utilized.

Injection of human blood SP cells into NOD/SCID mice in an engraftment assay, well established in our own [29–, 31] and other [32,, 33] laboratories, similarly showed no growth or engraftment. These results correlate with the lack of growth in the LTC-IC assay, although these failed repopulating data are tempered by the absence of an available positive control cell group. A number of factors could explain these results, not the least of which is the imperfect nature of the NOD/SCID model. This model is most useful for cord-blood-derived cells [34], with other tissue sources performing poorly. Furthermore, relatively large numbers of cells are required, and the homology of the model marrow space compared with that of humans, with regard to homing, adhesion or retention molecules, or cytokine or chemokine profiles, may differ. Conservatively, our results show that human circulating SP cells injected at the given dose were unable to achieve detectable hematopoiesis.

Alternatively, the absence of repopulating activity might be a reflection of the blood derivation of these SP cells and consequent level of lymphoid/dendritic commitment that precludes growth in repopulating, CAFC, or LTC-IC assays. SP cells derived from muscle have a lower capacity to repopulate BM than SP cells derived from BM [7], suggesting that the site from which SP cells are obtained has a bearing on their inherent functional attributes. In both in vivo and in vitro assays described, SP cell toxicity related to cell sorting or Hoechst dye exposure is worthy of consideration; however, the growth of SP cells in liquid culture systems makes these less likely possibilities.

Despite the lack of growth in the CAFC, LTC-IC, and repopulation cell assays, when blood SP cells were cocultured with PHA + IL-2-stimulated feeder cells in liquid cell culture, T, NK, and dendritic-like cells were obtained. Although largely undefined, the supernatant of PHA-stimulated lymphocytes is a rich source of numerous cytokines; presumably these, and additional IL-2, were responsible for the resulting output cells. However, conditions favoring monomyeloid or B-cell differentiation in liquid culture were not examined, nor was the capacity of circulating SP cells to generate nonhematopoietic cells tested. Thus, the growth observed in liquid cell culture, known to favor the growth of lymphocytes [25], should be viewed within this context. These results do support the possibility that a less-differentiated common T, NK, and dendritic cell precursor(s) may exist within this SP cell population, similar to those detected in human BM [35] or thymus [36,, 37] or reported in the murine system [38]. Further clonal studies are needed to determine if the multiple mature cell types obtained from these studies arose from single cell progenitors or if multiple progenitors are present within the SP cell subpopulation.

Stromal cultured CD7+ CD34 Lin cord blood SP cells are reported to differentiate specifically into NK cells [21]. It is not straightforward to compare these reported results with those here, since CD7+ and CD16+ cells were initially depleted in the Lin cocktail prior to purification of the SP population. Although stringent cell-sorting criteria were adhered to [25], the possibility of contaminating non-SP cells could not be formally ruled out, although the demonstration of an NK population unusually lacking CD2 and CD16 and with low levels of CD7 argues against contamination of a mature cell type. Additionally, it is unlikely that non-SP CD34+ cells were the origin of the output cells differentiating in these liquid cultures, since in addition to the presort depletion column removing CD34+ cells, SP cells were shown to be intrinsically CD34. Dendritic cells can be generated in vitro from rare proliferating CD34+ or more frequent nonproliferating CD14+ cells [39]; both these cell types also should have been depleted in the initial Lin separation column prior to further purification by cell sorting.

The NK cells grown from one SP-cell purification expressed CD7, CD16, and CD2 at the first available monitoring point, 1 month postpurification and culture. Cells grown for 60 additional days from this culture were noted to robustly kill K562 targets (data not shown). Prior to the first immunophenotypic analysis, too few cells were available for inspection, and a possible time point lacking CD7, CD16, and CD2 could not be assessed in this culture.

The NK cells obtained from a second experiment were unlike (CD7, CD16, CD2, CD45dim+) normal NK cells. Due to loss of this cell culture, it could not be determined if any of these markers might have been eventually upregulated.

The expression of the leukocyte differentiation antigens detected on circulating SP cells may provide insight into their ability to home and seed appropriate tissue sites, although their actual role is unknown. Their uniform expression of CD45 and differentiation in liquid cell culture into mature cell types support their role as circulating multipotential CD34 hematopoietic precursors. CD44, also highly expressed, has been associated with adhesion and the homing and seeding of hematopoietic stem cells [40]; HLA-class I is expressed on virtually all stages of human hematopoiesis [41]. The low expression of CD38 and absence of CD34 clearly distinguish them from non-SP blood progenitor cells, while their absence of CD90 and CD117 expression is similar to that reported for human BM SP cells [17].

CD59, expressed on rhesus BM SP cells, is a member of the Ly-6 family of molecules, which includes the Sca-1 marker found on murine HSCs [17]; both CD59 and CD35 prevent complement activation [42]. It can be speculated that the high levels of expression of these proteins impart a protective effect that saves circulating SP cells from autologous complement-mediated degradation [43].

Platelet/endothelial cell adhesion molecule 1 (CD31) is crucial to the process of leukocyte transmigration through intercellular junctions of vascular endothelial cells [44] and is highly expressed on murine BM-derived SP cells capable of forming myocytes [19]. The relatively high expression levels of the integrin markers CD18, CD11a/b/c, and CD31, and CD43 suggest that SP cells might have the capacity to move into and through extralymphoid tissue. However, the absence of L-selectin (CD62L) on circulating SP cells suggests their inability to enter lymphoid tissue, which correlates with their lack of mature leukocyte markers and presumed absence of mature function. CD49d (VLA4) binds fibronectin and vascular cell adhesion molecule; its presence on SP cells suggests these cells might have the capacity to enter a wound area or a site of chronic inflammation. CXCR4 is the receptor for stromal-cell-derived factor-1. This receptor was found to be crucial for the ability of human CD34+ cells' to home to SCID BM [45]. Although human circulating SP cells lack CD34, the additional absence of this chemokine receptor correlates with the inability of SP cells to implant in the described repopulating experiments.

In contrast to the abundant literature on the characterization and functional attributes of progenitor and stem cells obtained from BM, cord blood, or fetal sources, relatively little is known about the characteristics of progenitor cells derived from adult tissues. These data support the capacity to detect and separate early progenitor cells from adult human blood, despite their relative scarceness. Although not intuitively thought of as a source of undifferentiated progenitor cells, the peripheral blood is markedly accessible and available in higher volumes than most other tissue. Although SP cells are rare in peripheral blood, it might be possible to mobilize them to this compartment by methods similar to those used to mobilize CD34+ cells with cytokines. In conclusion, data presented here show that SP cells obtained from peripheral blood expressed a broad immunophenotype and shared the hallmark-identifying characteristic of dye efflux and quiescence with SP cells obtained from other sources. However, they appeared to have a more restricted progenitor capacity and different growth characteristics in the LTC-IC assay than those derived from primate and murine BM.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors gratefully acknowledge the contributions of Dr. Nobuyuki Miura for help with in vitro progenitor cell assays, Dr. R. Colvin for helpful comments and L. Duckett, J. Cardott, T. Christian, and S. Ziganti for help with HiPerFACs. This work was supported by a National Center for Research Resources grant S10-RR14904-01 [F.I.P.]. This manuscript is dedicated to the memory of H.L. Dombkowski.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References