Differential Effects of Granulocyte Colony-Stimulating Factor on Marrow

Download PDF
7 downloads 0 Views 219KB Size Report
Comment
ABSTRACT. A recent phase III trial comparing granulocyte colony-stimulating factor (G-CSF)–stimulated bone marrow. (G-BM) and G-CSF–mobilized peripheral ...
Biology of Blood and Marrow Transplantation 10:624-634 (2004) 䊚 2004 American Society for Blood and Marrow Transplantation 1083-8791/04/1009-0005$30.00/0 doi:10.1016/j.bbmt.2004.05.009

Differential Effects of Granulocyte Colony-Stimulating Factor on Marrow- and Blood-Derived Hematopoietic and Immune Cell Populations in Healthy Human Donors Luke R. Shier,1 Kirk R. Schultz,2 Suzan Imren,3 Jill Regan,1 Andrew Issekutz,4 Irene Sadek,5 Andrew Gilman,6 Zhijuan Luo,2 Tony Panzarella,7 Connie J. Eaves,3 Stephen Couban1 1 Department of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada; 2Department of Pediatrics, University of British Columbia, Vancouver, British Columbia, Canada; 3Terry Fox Laboratory, British Columbia Cancer Agency, Vancouver, British Columbia, Canada; 4Department of Pediatrics, Dalhousie University, Halifax, Nova Scotia, Canada; 5Department of Pathology, Dalhousie University, Halifax, Nova Scotia, Canada; 6 Department of Pediatrics, The Children’s Mercy Hospital, Kansas City, Missouri; 7Department of Biostatistics, Princess Margaret Hospital, Toronto, Ontario, Canada

Correspondence and reprint requests: Stephen Couban, MD, Queen Elizabeth II Health Sciences Centre, Bethune Bldg., Room 417, 1278 Tower Rd., Halifax NS, Canada B3H 2Y9 (e-mail: [email protected]). Received March 1, 2004; accepted May 24, 2004

ABSTRACT A recent phase III trial comparing granulocyte colony-stimulating factor (G-CSF)–stimulated bone marrow (G-BM) and G-CSF–mobilized peripheral blood (G-PB) in matched sibling allograft recipients showed that G-BM produced a similar hematologic recovery but a reduced incidence of extensive chronic graft-versus-host disease, indicating differences in the cell populations infused. As a first step toward identifying these differences, we treated a group of healthy adult humans with 4 daily doses of G-CSF 10 ␮g/kg and monitored the effects on various hematopoietic and immune cell types in the PB and BM over 12 days. G-CSF treatment caused rapid and large but transient increases in the number of circulating CD34ⴙ cells, colony-forming cells, and long-term culture-initiating cells and in the short-term repopulating activity detectable in nonobese diabetic/severe combined immunodeficiency/␤2-microglobulin–null mice. Similar but generally less marked changes occurred in the same cell populations in the BM. G-CSF also caused transient perturbations in some immune cell types in both PB and BM: these included a greater increase in the frequency of naive B cells and CD123ⴙ dendritic cells in the BM. The rapidity of the effects of G-CSF on the early progenitor activity of the BM provides a rationale for the apparent equivalence in rates of hematologic recovery obtained with G-BM and G-PB allotransplants. Accompanying effects on immune cell populations are consistent with a greater ability of G-BM to promote tolerance in allogeneic recipients, and this could contribute to a lower rate of chronic graft-versus-host disease. © 2004 American Society for Blood and Marrow Transplantation

KEY WORDS G-CSF ● Bone marrow Dendritic cells

●

Peripheral blood

INTRODUCTION The safety and ease of procuring granulocyte colony-stimulating factor (G-CSF)–mobilized peripheral blood (G-PB) cells from healthy donors and the faster engraftment obtained in allogeneic recipients of these cells as compared with normal steady-state bone marrow (BM) cells are now widely recognized [1– 4]. In addition, 2 randomized trials comparing transplants of allogeneic G-PB and 624

●

Stem cells

●

Immune function

●

B cells

●

steady-state BM have demonstrated a benefit to the recipients of G-PB cells in both overall [3] and disease-free [1] survival. In contrast, the related question of whether allogeneic G-PB transplants are associated with a significantly higher incidence or severity of acute or chronic graft-versus-host disease (GVHD) has not yet been resolved [5– 8]. A meta-analysis demonstrated a small but significant increase in both acute and chronic GVHD in recip-

G-CSF Effects on Marrow and Peripheral Blood Cells

ients of G-PB [9]. Whether this compromises quality of life or survival is not known. Recently, a number of groups have evaluated the use of allogeneic BM transplants obtained from G-CSF–stimulated healthy donors [10 –13]. These studies have established the safety of using this source of cells and the ability of G-CSF–stimulated bone marrow (G-BM) to produce rapid and sustained engraftment. A small, randomized phase III trial comparing matched sibling G-BM with G-PB allografts indicated comparable rates of hematologic recovery and a significant reduction in the incidence of both overall (47% versus 90%; P ⬍ .02) and extensive chronic (22% versus 80%; P ⬍ .002) GVHD in G-BM recipients, although the incidence of chronic GVHD in the G-PB population was high [14]. Taken together, these studies have suggested that G-CSF may differentially modulate the BM and PB content of the immune cells responsible for GVHD while causing similar effects on the cells responsible for hematologic recovery. G-PB contains more T cells than does steady-state BM, although the relationship between the number of CD4⫹ T cells infused and the incidence of GVHD in patients receiving G-PB is uncertain. Evaluation of the dendritic cell (DC) populations in G-PB has indicated that an increased number of plasmacytoid DCs or interferon (IFN)–producing DCs is associated with induction of a T-helper type 2–type response and induction of tolerance [15,16]. The number of CD34⫹ cells infused has also been associated with the rate of onset of GVHD in patients receiving G-PB [17]. This has suggested that immature CD34⫹ myeloid-lineage DCs may play a significant role in the development of GVHD in recipients of G-PB transplants. More recently, GCSF treatment was shown to produce an increase in expression of GATA-3 [18], an important transcription factor in the induction of T-helper type 2 T cells and their secretion of interleukin (IL)– 4. However, relatively little is known about the effects that G-CSF administration may have on immune cell populations that are released into the PB as compared with those that remain in the BM. Similarly, whereas the ability of G-CSF treatment to increase the concentration of primitive hematopoietic progenitor cells in the blood is well established [19 –22], only a few studies have investigated the effects of G-CSF on the progenitor activity of human BM [23–26]. This study was designed to obtain quantitative comparative data on the effects of a 4-day course of G-CSF on various phenotypically and functionally defined populations of cells with immune and hematopoietic activities in the PB and BM of healthy individuals followed up serially over 12 days.

BB&MT

METHODS Volunteer Selection, G-CSF Treatment and Sample Collection, Distribution, and Cell Preparation

Ten healthy volunteers between 18 and 30 years of age were recruited at the Halifax site after informed consent was obtained under the aegis of an institutionally approved protocol. Exclusion criteria were any of the following: medical comorbidities (specifically, cardiopulmonary disease, autoimmune disease, chronic skin conditions, or obesity), pregnancy or lactation, splenomegaly, known allergy to lidocaine, known hypersensitivity to Escherichia coli– derived products, or abnormalities on screening complete blood count or liver enzymes. Volunteers received G-CSF 10 ␮g/kg subcutaneously once daily for 4 consecutive days (days 0, 1, 2, and 3). All volunteers were followed up to 90 days after G-CSF administration, and no adverse reactions were noted. Heparinized PB and BM aspirate samples were obtained before, during, and up to 12 days after the first injection of G-CSF. All BM aspirates were performed by 2 individuals (S.C. and L.R.S.) by using a standard technique: 2 mL of BM was aspirated in a single draw and placed immediately in a tube. An aliquot of each PB and BM sample was analyzed on the day of collection in Halifax to determine the concentration of total nucleated cells (TNCs), with an automated cell counter, and the concentration of CD34⫹ cells and colony-forming cells (CFCs) after lysis of the red blood cells, as described below. Additional aliquots were sent by overnight courier to Vancouver for immune cell analyses, long-term culture– initiating cell (LTC-IC) assays, measurement of short-term repopulating cell (STRC) activity in immunodeficient mice, and repeat CFC assays on CD34⫹ cell-enriched populations. The latter were isolated immunomagnetically from low-density PB and untreated BM cells using the EasySep procedure (StemCell Technologies, Vancouver, BC, Canada). In a few cases, PB or BM cells were first cryopreserved in 10% dimethyl sulfoxide and 90% fetal calf serum. In these cases, progenitor values measured on the thawed samples were corrected by the estimated losses incurred from the freezing and thawing. These were estimated by comparing the differences in CFC values measured in Halifax (before freezing) and subsequently after thawing in Vancouver for the same number of prefreezing cells. Phenotyping Studies

Flow cytometric determinations of BM CD34⫹ cell counts were undertaken in Halifax as recommended by the International Society for Cellular Therapy [27] by using CD45–fluorescein isothiocyanate (FITC) and CD34-phycoerythrin (PE) monoclo625

L. R. Shier et al.

nal antibodies from StemKit (Beckman Coulter, Miami, FL). The same method was used for PB CD34⫹ cell enumeration after correction of the white blood cell count for the presence of nucleated erythroid cells. Phenotype analyses of immune cells were performed in Vancouver on whole blood by using the following monoclonal antibodies: CD4-CyC, CD8PE, T-cell receptor (TCR)␣␤–FITC, CD45RO–antigen-presenting cell (APC), CD45RA-CyC, CD123PE, CD11c-APC, lin-FITC, HLA-DR-PerCP, CD33-FITC, CD19-CyC, immunoglobulin D–PE, CD25-CyC, perforin-FITC, and G-CSF receptor (GCSF-R)– biotin-SA-APC (Becton Dickinson, San Jose, CA). T cells were evaluated by flow cytometry after staining with anti–CD4-CyC, anti–CD8-PE, and anti–TCR␣␤-FITC monoclonal antibodies. Memory and naive CD4⫹ or CD8⫹ T cells were distinguished by positive co-staining with anti– CD45RO-APC or anti–CD45RA-CyC antibodies, respectively. DCs were identified as lin⫺/HLA-DR⫹ cells and were classified as myeloid (CD11c⫹/ CD123⫺) or plasmacytoid (CD11c⫺/CD123⫹). B cells were identified by their expression of CD19 and absence of CD33. Natural killer (NK) and NK T cells were identified by their coexpression of CD3, CD8, and CD56. Lymphocyte values were expressed as an absolute number of cells per volume of whole blood (cells ⫻ 104/mL). The cell concentration was calculated by the mononuclear cell count upon arrival to the laboratory (cells/mL) multiplied by the percentage of marker-positive cells. Immune Cell Cytokine Assays

TCR␣␤⫹CD8⫺ and TCR␣␤⫹CD8⫹ cells were classified as Th/Tc1 and Th/Tc2 according to their cytoplasmic staining with antibodies for various cytokines, as previously described [28]. All cytokine assays were performed on low-density (⬍1.077 g/mL) cells isolated by centrifugation on Ficoll-Hypaque (Amersham, Uppsala, Sweden) and stimulated as described below. Cytokine and perforin monoclonal antibodies used were anti–IL-1-PE, anti–IL-6-PE, anti–IL-10APC, anti–IL-2-APC, anti–IFN-␥-APC, anti–tumor necrosis factor-␣-APC, anti–IL-4-APC, and anti–IL12-APC perforin-FITC (Becton Dickinson). Because phorbol myristate acetate (PMA)–ionomycin decreases CD4 expression, CD4⫹ T cells were identified as TCR␣␤⫹CD8⫺ cells. Mononuclear cells were treated with PMA/ionomycin for stimulation of Tand NK-cell populations, with lipopolysaccharide (LPS) for activation of B cells and DCs, and with IFN-␥ for stimulation of IL-10 and IL-12 secretion by CD19⫹CD33⫺ cells. All samples were incubated with monensin to block cell glycoprotein export and enhance intracellular cytokine accumulation. Results 626

were expressed as the stimulation index, which is the ratio of cytokine-positive cells in gated populations (eg, CD4⫹ T cells) of stimulated cells compared with unstimulated cells. The value of unstimulated cells was set at a minimum of 1%. CFC Assays

CFC assays were performed by plating red blood cell– depleted (Halifax) or CD34⫹ cell– enriched (Vancouver) suspensions in methylcellulose-based medium (Methocult; StemCell Technologies) containing stem cell factor 50 ng/mL, granulocyte-macrophage colony-stimulating factor 20 ng/mL, G-CSF 20 ng/mL, IL-3 20 ng/mL, IL-6 20 ng/mL, and erythropoietin 3 U/mL to detect colony forming unit– erythrocytes plus burst-forming unit– erythroids plus colony forming unit– granulocyte-macrophages plus colony forming unit– granulocytes, erythrocytes, megakaryocytes, and macrophages (total ⫽ CFCs) by using standard procedures for colony enumeration after a 2-week incubation period [29]. LTC-IC Assays

CD34⫹ cell– enriched PB or BM cells were assayed for LTC-ICs by plating aliquots of 5 ⫻ 104 cells into replicate 2.5-mL cultures containing preestablished irradiated feeder layers of murine fibroblasts engineered to secrete human G-CSF, IL-3, and stem cell factor as previously described [29]. After 6 weeks, all cells in the cultures were harvested by trypsinization, and appropriate aliquots were plated in CFC assays. The concentration of CFCs thus detected was then used to calculate the total number present in each LTC and, hence, the number of input LTC-ICs originally seeded into the culture, assuming that, on average, 1 G-PB LTC-IC produces 25 CFCs and 1 BM LTC-IC produces 18 CFCs [29]. The number of LTC-ICs per 107 cells in the original low-density G-PB or G-BM population (and, hence, per liter of sample) was then calculated, by knowing its content of CD34⫹ cells and assuming that all LTC-ICs were low-density CD34⫹ cells [29,30] and that there were no losses in the various cell-isolation procedures. Xenotransplant Assays of STRC Activity

A total of 105 CD34⫹ cell– enriched human PB and BM cells were injected intravenously together with 106 irradiated (carrier) human BM cells into sublethally irradiated (350 cGy) nonobese diabetic/ severe combined immunodeficiency (NOD/SCID)/ ␤2-microglobulin–null mice as previously described [30]. Three weeks after transplantation, femoral BM aspirates were obtained [31]. After 8 weeks, mice were killed, and all BM cells were isolated from their femurs. Each BM sample obtained from the mice was processed individually as follows: red blood cells were

G-CSF Effects on Marrow and Peripheral Blood Cells

lysed with ammonium chloride, and the cells were stained with anti-human CD45, CD71, and propidium iodide (Sigma Chemicals, St. Louis, MO) before flow cytometric analysis by using gates set after staining parallel aliquots of the same cells with fluorochrome-labeled isotype control antibodies as described previously [31]. short-term repopulating cell– myeloid (STRC-M) and short-term repopulating cell–myeloid, lymphoid (STRC-ML) activity was expressed, respectively, as the percentages of cells in the BM of the mice 3 and 8 weeks after transplantation that were human CD45⫹/71⫹ per 107 cells in the original low-density human G-PB or G-BM populations (and, hence, per liter of starting sample) by using the same assumptions as for LTC-ICs. STRC assays were not performed on any of the pre–G-CSF (day 0) PB samples because of the small number of CD34⫹ cells they contained and the expectation that these would not contain detectable numbers of STRCs [32]. Therefore, to calculate -fold increases in these samples, the day 0 value was assumed to be the minimum detectable value. Statistics

All data are shown as the mean ⫾ SEM per liter of starting PB or BM sample or as a ratio relative to the pre–G-CSF (steady-state) value. In each case, values were calculated for each individual donor, and the mean changes for the group were then plotted over time. Statistical significance was assigned for P values of ⬍.05, calculated by using the unpaired Student t

test on mean values and the Fisher exact 2-tailed test for discrete variables.

RESULTS Effects on Total Cells and CD34ⴙ Cell Concentrations in PB and BM

Healthy individuals given 4 daily injections of G-CSF showed the expected rapid transient increase in the PB white blood cell count [33], which, in this study, reached a maximum value after 3 to 4 days (approximately 6-fold above baseline values; P ⬍ .05) and then returned to normal by days 9 to 12. The early increase in circulating white blood cells was accompanied by a significant (P ⬍ .05), but less marked, increase in the concentration of TNCs (per liter) in BM aspirate samples, peaking at approximately 3-fold above baseline on days 3 to 4, as first reported by Dicke et al. [23] and Isola et al. [24]. However, in contrast to the PB, the cellularity of the BM aspirates was still high after the additional 5 to 9 days of follow-up (Figure 1A). Changes in the concentration of CD34⫹ cells in both PB and BM mirrored the changes seen in the TNC values. Peak concentrations were attained in both PB and BM by days 3 to 4, and these were then sustained in the BM but returned to normal values in the PB by days 9 to 12 (Figure 1B). However, the magnitude of the G-CSF–stimulated increases in CD34⫹ cell numbers in PB and BM samples was quite different, with maximum increases of 26-fold (P ⬍ .05) and 1.5- to 1.7-fold (P ⬎ .05), respectively. These findings confirm those reported by others [23–25,34]. Effects on CFCs, LTC-ICs, and STRC Activity

Figure 1. Kinetics of G-CSF–induced changes in PB and BM cellularity and CD34⫹ cell numbers. TNCs (A) or CD34⫹ cells (B) per liter of PB (E) or BM aspirate (F) are shown relative to the pre–G-CSF value at different times after the initiation of a 4-day schedule of G-CSF administration (G-CSF given on days 0, 1, 2, and 3). Values shown are derived from 10 healthy adults. The number of individuals was as follows: before G-CSF administration, n ⫽ 10; day 1, n ⫽ 4; day 3, n ⫽ 9; day 4, n ⫽ 7; day 5, n ⫽ 6; and days 9 to 12, n ⫽ 10. The values are expressed as a ratio of the post–G-CSF measurement at each time point compared with the pre–G-CSF reference point. The SEM is present with each data point, but some of the error bars were smaller than the size of the symbol.

BB&MT

As shown in Figure 2A, G-CSF treatment also induced the expected increases in the concentration of CFCs in the PB and BM aspirate samples with very similar kinetics to those seen for the total CD34⫹ cells(peak increases of approximately 20-fold and 3-fold seen on days 3 to 4 in PB and BM, respectively, returning to baseline by days 9 to 12). These results are again similar to those documented in a number of studies [24,25,34 –36]. Figure 2B shows that the same pattern of change in CFC numbers (per liter of PB or BM) was obtained from assays performed the same day in Halifax and the next day in Vancouver, despite the 24-hour delay and differences in how the cells were processed before plating. Figure 3 shows the changes seen in the concentrations of more primitive progenitors detected in vitro as LTC-ICs (Figure 3A and B) and in vivo as STRCs. Both myeloid-restricted STRC activity (Figure 3C and 3D; STRC-Ms, 3-week repopulation end point in NOD/SCID-␤2-microglobulin–null mice) 627

L. R. Shier et al.

Figure 2. Kinetics of G-CSF–induced changes in CFC numbers in the PB and BM. A, Changes in the absolute number of CFCs per liter of PB (E; data from Vancouver assays) or BM aspirate samples (F; data from Vancouver assays) from the same individuals as in Figure 1. The values are expressed as a ratio of the post–G-CSF measurement at each time point compared with the pre–G-CSF reference point. B, The same data expressed relative to the pre–GCSF values (䊐; CFC Halifax; PB data from Halifax assays).

No significant changes were seen in the concentration of T cells activated to produce IL-2, IL-4, or tumor necrosis factor-␣ (data not shown). Significantly increased proportions of IFN-␥⫹CD8⫹ T cells and CD8⫺/TCR␣␤⫹ (primarily CD4⫹) T cells were seen in the day 3 BM samples (Figure 4E and 4G; P ⫽ .03 and .02, respectively). However, evaluation of the ratio of the IFN-␥–producing CD8⫹ (Figure 4F) and CD4⫹ (Figure 4H) T cells revealed that, even though there was a significant increase in IFN-␥–producing CD4⫹ and CD8⫹ T cells in the BM, the difference quickly disappeared between days 3 and 9. The concentration of IFN-␥–producing NK T cells (CD3⫹/ CD56⫹) in the BM was also increased on day 5,

and STRC activity with lymphoid and myeloid differentiation potential (Figure 3E and 3F; STRC-MLs, 6to 8-week repopulation end point in NOD/SCID-␤2microglobulin–null mice) [23] were assessed. The activity of all 3 of these primitive cell types increased rapidly, albeit variably, in PB and BM, again with greater increases in the PB (peak increases of 700- to 8000-fold) than in the BM (50- to 90-fold) on days 3 to 4. The increases in LTC-IC activity were also somewhat larger than those reported previously [25,26,35,36], possibly because, at least in part, of the use of different LTC-IC assay conditions, which detect different progenitor subsets with different efficiencies [29]. We also noted that, in contrast to the more sustained increases in CFCs and total CD34⫹ cells (Figures 1 and 2), the increase in LTC-IC activity seen after 3 to 4 days was short-lived and, in both PB and BM samples, had consistently declined by day 5. Effects on T and NK Cells

The increase in CD3/TCR␣␤⫹ T cells on day 5 compared with pre–G-CSF values was also higher in the PB than in the BM samples (2.2 ⫾ 0.6 versus 0.9 ⫾ 0.3; P ⫽ .10), although the absolute concentrations of CD4⫹ and CD8⫹ T cells on day 5 were actually higher in the BM aspirates than in the PB samples (Figure 4A and 4B). The relative change in both CD4⫹ and CD8⫹ T cells (Figure 4B and 4D) was not significant. The greater numbers of CD3/TCR␣␤⫹ T cells in the day 5 PB samples could be attributed primarily to increased numbers of CD8⫹/TCR␣␤⫹ T cells (Figure 4A; P ⫽ .07) that were CD45RO⫹ (data not shown; P ⫽ .05). No significant changes were detected in the concentrations of either NK or NK T cells (data not shown). 628

Figure 3. Kinetics of G-CSF–induced changes in primitive progenitors in the PB and BM. A, Absolute number of LTC-ICs per liter of PB (E) or BM aspirate samples (F) from a subset of the same individuals as in Figure 1 (n ⫽ 5; P ⱕ .05 only for PB on day 5). B, Same data expressed relative to the pre–G-CSF values (P ⱕ .05 on days 3 and 4). C and E, Absolute STRC activity per liter of the same PB (E) or BM aspirate samples (F) (STRC-Ms: panel C, n ⫽ 5, P ⬎ .05; STRC-MLs: panel E, n ⫽ 5, P ⬎ .05). D and F, The same data as in C and E, respectively, expressed relative to the pre–G-CSF activity.

G-CSF Effects on Marrow and Peripheral Blood Cells

Figure 4. Effect of G-CSF treatment on T-cell populations. Shown are the changes in the absolute (A and C) and relative (B and D) numbers of TCR␣␤⫹CD8⫹ T cells (A and B) and CD4⫹ T cells (C and D) measured on the same PB (E) and BM samples (F) as in Figure 1 (10 healthy donors). E to H, Corresponding results for the IFN-␥–producing CD8⫹ and CD4⫹ T cells detected after stimulation with PMA/ionomycin. Differences between G-BM and G-PB that were significant (P ⱕ .05) by a Student t test are identified with an asterisk (*). The number of individuals per time period was as follows: before G-CSF administration, n ⫽ 10; day 1, n ⫽ 4; day 3, n ⫽ 9; day 5, n ⫽ 6; and day 9, n ⫽ 4. The values in panels A, C, and E are expressed as the number of cells ⫻ 104/mL of whole blood; values in panels B, D, F are expressed as the ratio of post– G-CSF to pre–G-CSF values.

BB&MT

Figure 5. G-CSF induces greater increases in the concentration of B cells and plasmacytoid DCs (pDCs) in the BM as compared with the PB. Results for absolute (A) and relative (B) numbers of total CD19⫹CD33⫺ B cells in PB (E) and BM aspirate samples (solid symbols) are shown for the same samples analyzed in Figures 1 to 4. Significant differences between PB and BM were seen on days 0 and 5 (*). DCs were defined as lin⫺/HLA-DR⫹ cells and were subclassified as myeloid (mDC) CD123⫺/CD11c⫹ or pDC CD123⫹/ CD11c⫺ cells. The difference between BM and PB CD123⫹ DCs was significant at both day 3 (P ⫽ .01) and day 5 (P ⫽ .02) and is represented by an asterisk. The number of individuals per time period was as follows: before G-CSF administration, n ⫽ 10; day 1, n ⫽ 4; day 3, n ⫽ 9; day 5, n ⫽ 6; and day 9, n ⫽ 4. The values in panels A, C, E, and G are expressed as the number of cells ⫻ 104/mL of whole blood; values in panels B, D, F, and H are expressed as the ratio of post–G-CSF to pre–G-CSF values. 629

L. R. Shier et al.

Table 1. Evaluation of Cell Populations in G-CSF–Primed Bone Marrow and PB (n ⫽ 4) Before GSF (ⴛ 104)

3 days (ⴛ 104)

5 days (ⴛ 104)

Cell Markers

PB

BM

P Value

PB

BM

P Value

PB

BM

P Value

CD4ⴙCD25ⴙTCR␣/␤ TCR␣/␤ⴙ T cells perforinⴙ (CD3ⴙ/CD56ⴚ) NK cells perforinⴙ (CD3ⴚ/CD56ⴙ)

12.6 ⴞ 6.5 5.7 ⴞ 2.8

13.9 ⴞ 8.7 24.9 ⴞ 12.8

.90 .20

7.8 ⴞ 3.8 35.3 ⴞ 15

12.1 ⴞ 9.3 57.0 ⴞ 14

.50 .30

8.0 ⴞ 4.8 58.8 ⴞ 53.8

8.1 ⴞ 4.9 30.3 ⴞ 9.4

1.0 .60

18.3 ⴞ 7.2

42.0 ⴞ 16.3

.20

12.6 ⴞ 2.0

25.1 ⴞ 12

.40

17.6 ⴞ 17.6

18.3 ⴞ 9.6

1.0

12.4 ⴞ 4.0

50.7 ⴞ 21.5

.10

20.3 ⴞ 4.9

30.0 ⴞ 3.9

.20

19.6 ⴞ 8.5

17.9 ⴞ 7.9

although this increase was not significant (P ⫽ .06; data not shown). Effects on Accessory Cells

An increase in the concentration of CD19⫹CD33⫹ B cells in both the PB and BM was seen on day 3 (P ⫽ .05), which then rapidly declined, although absolute levels of these cells remained higher in the BM (Figure 5A). However, the B cells in the BM did not seem to be activated (ie, there was no increase in IL-1 or IL-6 production). The relative increase in B cells (Figure 5B) was greater in the PB after G-CSF because of the low concentration of B cells in pre–G-CSF PB. By day 5, G-CSF treatment produced a significantly larger population of plasmacytoid (lin⫺/HLADR⫹/CD123⫹) DCs (Figure 5C and 5D) in the BM than in the PB (8.8 ⫾ 2.7 versus 1.3 ⫾ 0.5 ⫻ 104/mL; P ⫽ .02), whereas the levels of myeloid (lin⫺/HLADR⫹/CD11c⫹) DCs (Figure 5E and 5F) became similar (4.5 ⫾ 1.3 versus 3.1 ⫾ 2.3; P ⫽ .63), despite significantly higher concentrations of both CD11c⫹ and CD123⫹ DCs in steady-state BM samples (Figure 5C and 5E). As seen for B cells, the relative increase in both plasmacytoid DCs (Figure 5F) and myeloid DCs (Figure 5D) was less striking in BM because of the higher concentration of these populations in steadystate BM. The number of IL-6 –producing CD33⫹ monocytes was also increased in the PB on days 3 and 5 (data not shown; P ⫽ .05).

.90

numbers of CD4⫹/CD25⫹ T cells present in the BM as compared with the PB after G-CSF. There was, however, a trend toward expansion of G-CSF-R⫹ T cells in the day 5 BM aspirates. There was also a large decrease in perforin-expressing T and NK cells in the day 5 BM samples that approached the levels seen in the PB. In the same samples, naive immunoglobulin D⫹/G-CSF-R⫺ B cells were present at higher levels in the BM (Figure 5). However, none of these differences was significant, likely because of the small numbers of samples compared (Figure 6). DISCUSSION In this study, we describe the parallel kinetics of G-CSF–induced effects on the BM and PB content of several progenitor cell populations thought to contribute to the early regeneration of hematopoiesis after transplantation. These include the first measurements of 2 types of human cells with short-term repopulating activity in engrafted immunodeficient mice [22,39,40], in addition to the total CD34⫹ cell population, CFCs, and LTC-ICs, defined with a

Effects on G-CSF-Rⴙ T and B Cells

G-CSF-Rs are expressed on both T cells and B cells, and G-CSF-R⫹ T cells can be induced to increase expression of GATA-3 [18]. We hypothesized that the greater effect of G-CSF treatment on B cells in BM versus PB (Figure 4A) could be due to (1) a selective expansion and retention of G-CSF-R⫹ B cells in the BM, (2) a decrease in perforin-containing T or NK cells needed to inhibit B-cell expansion [37], (3) an increase in immature or naive B cells [38], or (4) a skewed effect of G-CSF on regulatory CD4⫹CD25⫹ T cells. We therefore evaluated the samples from the last 4 volunteers for potential changes in these cell populations (Table 1). No difference was found in the 630

Figure 6. G-CSF-R expression on B cells in G-CSF–stimulated BM versus PB. Immunophenotyping was performed by 4-color flow cytometry after staining for CD19, CD33, immunoglobulin D, and G-CSF-R 5 days after the initiation of G-CSF treatment. B cells were identified as CD19⫹/CD33⫺ and were further evaluated for expression of immunoglobulin D, G-CSF-R, or both. The concentration of cells (⫻ 104/mL) is expressed in the figure. The number of individuals was 4.

G-CSF Effects on Marrow and Peripheral Blood Cells

6-week CFC readout [29]. Population size comparisons were made for fixed volumes of BM aspirate or PB samples by using standardized culture reagents and detection criteria. To minimize variability in the extent of PB contamination of BM aspirates, all BM aspirates were obtained by 2 individuals, who used a standardized technique. Average changes over time were then assessed after first normalizing the data for each sample to the corresponding pre–G-CSF value for the same individual. All hematopoietic cell types evaluated had similar rapid kinetics of expansion, usually reaching a peak on day 3 or 4 after the 4-day G-CSF treatment protocol was started. However, the magnitude of the increases obtained and their durability varied for the various types of progenitors. In general, the magnitude of the expansions was approximately 10-fold higher in PB than in BM, but it varied over a wide range, depending on the cell type being assessed. Considerable variability between individual donor responses was also seen. As anticipated from previous measurements of changes in CD34⫹ cells, CFCs and LTC-ICs [25,35,36]—the more primitive cell populations— showed the greatest G-CSF–induced effects. Thus, although CFC numbers in the BM showed very little change, there was an approximately 50-fold increase in LTC-IC activity and an approximately 90-fold increase in STRC activity. Similarly, in the PB, the increases in LTC-IC and STRC activity were 40- to 200-fold higher than the increases in CFC numbers. However, these effects of G-CSF on circulating STRC activity need to be interpreted with caution because they are based on the assumption that before G-CSF administration, STRC numbers were at or below the detection limit. It is also important to note that, despite the large expansion seen in the numbers of circulating primitive progenitors, these still remained lower than in the BM, except in the case of the LTC-ICs, where PB values up to 35 times those measured in the BM were seen. In addition to differences in the extent to which various progenitor populations were expanded, there were also differences in the period over which these expansions were sustained. The more primitive progenitors showed the shortest-lived responses to the 4 days of G-CSF administration, as evidenced by a decline in progenitor activity in both PB and BM immediately after the peak value reached on the last day of G-CSF treatment or a day later. These findings suggest that the regenerative activity of an allograft harvest could be significantly affected by the timing of the BM or PB cell collection after the cessation of G-CSF administration, with an optimal collection time being the day after completion of a 3- or 4-day course. A small randomized trial comparing G-PB and G-BM allografts has demonstrated that transplanta-

BB&MT

tion of G-PB is associated with a much higher frequency of chronic GVHD compared with G-BM [14]. Such an observation suggests differences in the infused immune populations in G-BM compared with G-PB and requires confirmation by a larger clinical trial. In this study, we identified 2 cell populations (B cells and plasmacytoid DCs) that were increased to much higher levels in G-BM compared with G-PB. B cells seem to play a role in the development of chronic GVHD as producers of both autoreactive immunoglobulin and APCs. It is interesting to note that B cells have been shown to play a role in the induction of GVHD in murine models [41], and recent human studies have demonstrated that treatment with rituximab, a monoclonal antibody against CD20⫹ cells, can influence the onset of GVHD [42] and can be used as therapy for GVHD [43]. Similarly, B-cell production of autoreactive immunoglobulins and T cell/B cell interactions play an important role in GVHD [44,45]. We hypothesized that the expanded B-cell population found in G-BM may be produced either by a loss of cells with negative regulatory activities— eg, perforincontaining cytolytic (NK and/or CD8⫹) T-cell populations— or by direct stimulatory effects of G-CSF on B cells with expansion of G-CSF-R⫹ B cells [37,38]. We therefore evaluated the type of B-cell populations that were expanded in the G-BM samples obtained from the last 4 donors. These analyses failed to detect any differences in the number of perforinpositive T or NK cells, although the small number of individuals evaluated likely precluded a definitive conclusion. Our findings also failed to support the hypothesis that there was direct effect of G-CSF on B cells that resulted in a selective expansion of G-CSF-R⫹ B cells in G-BM. In contrast, an increased concentration of immunoglobulin D⫹ B cells was noted in the G-BM samples, suggesting a shift in favor of naive B-cell expansion in G-BM. In a transplant setting, this might, in turn, inhibit T-cell activation and promote the induction of tolerance [46,47]. All 3 mechanisms of B-cell function—including presentation of antigen to T cells, antibody production, and cytokine production— have been shown to play an important role in the development of chronic GVHD [41,44,48]. Plasmacytoid DCs were present in greater numbers in the G-BM samples than in the G-PB samples. We observed an overall expansion of both myeloid (CD11c⫹) and plasmacytoid (CD123⫹) subpopulations of DCs but noted a shift toward a predominance of plasmacytoid DCs in G-BM. Plasmacytoid DCs have been associated with the presence of chronic GVHD in humans and are of donor origin [49]. There is good evidence that decreased chronic GVHD is associated with a predominance of plasmacytoid DCs; this is supported by the fact that extracorporeal photopheresis results in a shift from myeloid DCs to 631

L. R. Shier et al.

plasmacytoid DCs [50]. Others have shown that a lower incidence of chronic GVHD and a higher leukemia relapse rate were associated with higher numbers of infused plasmacytoid DCs. This may be due to the ability of plasmacytoid DCs to induce a shift to a T-helper type 2/Tc2 response after BMT [51]. Some of the conflicting results regarding the role of DCs in chronic GVHD may reflect whether the DCs are of donor or host origin. Unlike in acute GVHD, in which host DCs seem to be a major antigen presenting cell, both donor plasmacytoid DCs [49] and host DCs [52] have been associated with increased chronic GVHD in humans. These cells respond differently to activation factors such as LPS and CpG oligodeoxynucleotides [53], with subsequent IFN production [54], consistent with their playing a different role in activating other populations that contribute to GVHD. The response to TLR9 stimulation by CpG oligodeoxynucleotides has been associated with the development of GVHD [28]. The decreased rate of response by cord blood DCs to CpG stimulation [55] and the profound inhibition of B-cell responses in GVHD to CpG stimulation by 4-aminoquinolines, such as hydroxychloroquine [28,41], highlight this potential mechanism in GVHD. Moreover, significant interaction exists between B-cell and plasmacytoid DC populations in altering the immune response [56]. We were not able to demonstrate any significant differences in cytokine production after mitogen stimulation (PMA/ionomycin for T cells and LPS for B cells). We did not evaluate T- or B-cell responses to either alloantigens or pathogens. It is possible that differences between G-PB and G-BM T cells may have been observed in such responses. In conclusion, the results presented here describe differences in subsets of cells within the PB and BM compartments in response to G-CSF. Although these differences may explain site-specific effects of G-CSF in vivo, our study was limited by being solely descriptive in nature. Therefore, it does not afford a definitive explanation for the differences in hematologic recovery and GVHD seen with different allograft sources. Our results may also have implications for the timing of leukapheresis harvests from G-CSF–treated healthy donors. However, because of how the apheresis product is collected, both the relative and absolute numbers of all the cell types we have measured would likely be different from those documented here for whole PB. Similarly, the G-BM results in this study were derived from a first aspiration that would have been minimally diluted with PB. A BM harvest sufficient for a transplant represents a more diluted product containing more PB cells. The results from these studies, although representing biological differences between G-CSF–stimulated marrow and PB, cannot be directly correlated to the infused products collected from G-CSF–stimulated donors. Further analyses 632

comparing the composition of harvested G-BM and apheresed G-PB with clinical hematologic recovery and other clinical outcome parameters—including GVHD—will be needed to determine the ultimate predictive utility of the type of BM and PB assessments reported here. Such studies are planned as part of a large Canadian Bone Marrow Transplant Group study comparing G-BM with G-PB.

ACKNOWLEDGMENTS This study was supported in part by research grants from Amgen Canada and the Capital District Health Authority Foundation (S.C.). The hematopoietic progenitor studies in Vancouver were supported by a grant from the National Cancer Institute of Canada with funds from the Terry Fox Run (C.J.E.). The immune cell function studies were funded by a grant from the Children’s Mercy Hospital Foundation. K.R.S. is supported by a Wyeth/Canadian Insitutes of Health Resarch Clinical Research Chair in Transplantation.

REFERENCES 1. Bensinger WI, Martin PJ, Storer B, et al. Transplantation of bone marrow as compared with peripheral-blood cells from HLA-identical relatives in patients with hematological cancers. N Engl J Med. 2001;344:175-181. 2. Blaise D, Kuentz M, Fortanier C, et al. Randomized trial of bone marrow versus lenograstim-primed blood cell allogeneic transplantation in patients with early-stage leukemia: a report from the Socie´te´ Franc¸aise de Greffe de Moelle. J Clin Oncol. 2000;18:537-546. 3. Couban S, Simpson DR, Barnett MJ, et al. A randomized multicenter comparison of bone marrow and peripheral blood in recipients of matched sibling allogeneic transplants for myeloid malignancies. Blood. 2002;100:1525-1531. 4. Schmitz N, Beksac M, Hasenclever D, et al. Transplantation of mobilized peripheral blood cells to HLA-identical siblings with standard-risk leukemia. Blood. 2002;100:761-767. 5. Watanabe T, Takaue Y, Kawano Y, et al. HLA-identical sibling peripheral blood stem cell transplantation in children and adolescents. Biol Blood Marrow Transplant. 2002;8:26-31. 6. Ringden O, Labopin M, Bacigalupo A, et al. Transplantation of peripheral blood stem cells as compared with bone marrow from HLA-identical siblings in adult patients with acute myeloid leukemia and acute lymphoblastic leukemia. J Clin Oncol. 2002;20:4655-4664. 7. Eapen M, Klein JP, Champlin RE, et al. Increased chronic graft versus host disease and mortality after peripheral blood stem cell transplantation in older children and adolescents with acute leukemia. Blood. 2002;100:545a (abstr.). 8. Couban S, Barnett M. The source of cells for allografting. Biol Blood Marrow Transplant. 2003;9:669-673. 9. Cutler C, Giri S, Jeyapalan S, et al. Acute and chronic graftversus-host disease after allogeneic peripheral-blood stem-cell and bone marrow transplantation: a meta-analysis. J Clin Oncol. 2001;19:3685-3691.

G-CSF Effects on Marrow and Peripheral Blood Cells

10. Couban S, Messner HA, Andreou P, et al. Bone marrow mobilized with granulocyte colony-stimulating factor in related allogeneic transplant recipients: a study of 29 patients. Biol Blood Marrow Transplant. 2000;6:422-427. 11. Ji SQ, Chen HR, Wang HX, et al. Comparison of outcome of allogeneic bone marrow transplantation with and without granulocyte colony-stimulating factor (lenograstim) donor-marrow priming in patients with chronic myelogenous leukemia. Biol Blood Marrow Transplant. 2002;8:261-267. 12. Isola L, Scigliano E, Fruchtman S. Long-term follow-up after allogeneic granulocyte colony-stimulating factor-primed bone marrow transplantation. Biol Blood Marrow Transplant. 2000;6: 428-433. 13. Serody JS, Sparks SD, Lin Y, et al. Comparison of granulocyte colony-stimulating factor (G-CSF)-mobilized peripheral blood progenitor cells and G-CSF-stimulated bone marrow as a source of stem cells in HLA-matched sibling transplantation. Biol Blood Marrow Transplant. 2000;6:434-440. 14. Morton J, Hutchins C, Durrant S. Granulocyte-colony-stimulating factor (G-CSF)-primed allogeneic bone marrow: significantly less graft-versus-host disease and comparable engraftment to G-CSF-mobilized peripheral blood stem cells. Blood. 2001;98:3186-3191. 15. Arpinati M, Green CL, Heimfeld S, et al. Granulocyte-colony stimulating factor mobilizes T helper 2-inducing dendritic cells. Blood. 2000;95:2484-2490. 16. Klangsinsirikul P, Russell NH. Peripheral blood stem cell harvests from G-CSF-stimulated donors contain a skewed Th2 CD4 phenotype and a predominance of type 2 dendritic cells. Exp Hematol. 2002;30:495-501. 17. Zaucha JM, Gooley T, Bensinger WI, et al. CD34 cell dose in granulocyte colony-stimulating factor-mobilized peripheral blood mononuclear cell grafts affects engraftment kinetics and development of extensive chronic graft-versus-host disease after human leukocyte antigen-identical sibling transplantation. Blood. 2001;98:3221-3227. 18. Franzke A, Piao W, Lauber J, et al. G-CSF as immune regulator in T cells expressing the G-CSF receptor: implications for transplantation and autoimmune diseases. Blood. 2003;102:734739. 19. To LB, Haylock DN, Simmons PJ, et al. The biology and clinical uses of blood stem cells. Blood. 1997;89:2233-2258. 20. van der Loo JC, Hanenberg H, Cooper RJ, et al. Nonobese diabetic/severe combined immunodeficiency (NOD/SCID) mouse as a model system to study the engraftment and mobilization of human peripheral blood stem cells. Blood. 1998;92: 2556-2570. 21. Korbling M, Anderlini P. Peripheral blood stem cell versus bone marrow allotransplantation: does the source of hematopoietic stem cells matter? Blood. 2001;98:2900-2908. 22. Glimm H, Eisterer W, Lee K, et al. Previously undetected human hematopoietic cell populations with short-term repopulating activity selectively engraft NOD/SCID-beta2 microglobulin-null mice. J Clin Invest. 2001;107:199-206. 23. Dicke KA, Hood DL, Arneson M, et al. Effects of short-term in vivo administration of G-CSF on bone marrow prior to harvesting. Exp Hematol. 1997;25:34-38. 24. Isola LM, Scigliano E, Skerrett D, et al. A pilot study of allogeneic bone marrow transplantation using related donors stimulated with G-CSF. Bone Marrow Transplant. 1997;20: 1033-1037. 25. Chiang KY, Lamb L, Clark J, et al. Assessment of G-CSF

BB&MT

26.

27.

28.

29.

30.

31.

32.

33. 34.

35.

36.

37.

38.

39.

40.

41.

stimulated BM hematopoietic stem cells in normal donors. Cytotherapy. 2002;4:55-63. Dahl E, Burroughs J, DeFor T, et al. Progenitor content of autologous grafts: mobilized bone marrow vs mobilized blood. Bone Marrow Transplant. 2003;32:575-580. Sutherland DR, Anderson L, Keeney M, et al. The ISHAGE guidelines for CD34⫹ cell determination by flow cytometry. International Society of Hematotherapy and Graft Engineering. J Hematother. 1996;5:213-216. Schultz KR, Su WN, Hsiao CC, et al. Chloroquine prevention of murine MHC-disparate acute graft-versus-host disease correlates with inhibition of splenic response to CpG oligodeoxynucleotides and alterations in T-cell cytokine production. Biol Blood Marrow Transplant. 2002;8:648-655. Hogge DE, Lansdorp PM, Reid D, et al. Enhanced detection, maintenance, and differentiation of primitive human hematopoietic cells in cultures containing murine fibroblasts engineered to produce human steel factor, interleukin-3, and granulocyte colony-stimulating factor. Blood. 1996;88:3765-3773. Fujisaki T, Berger MG, Rose-John S, et al. Rapid differentiation of a rare subset of adult human lin⫺CD34⫺CD38⫺ cells stimulated by multiple growth factors in vitro. Blood. 1999;94: 1926-1932. Verlinden SF, van Es HH, van Bekkum DW. Serial bone marrow sampling for long-term follow up of human hematopoiesis in NOD/SCID mice. Exp Hematol. 1998;26:627-630. Lapidot T, Petit I. Current understanding of stem cell mobilization: the roles of chemokines, proteolytic enzymes, adhesion molecules, cytokines, and stromal cells. Exp Hematol. 2002;30: 973-981. Korbling M. Effects of granulocyte colony-stimulating factor in healthy subjects. Curr Opin Hematol. 1998;5:209-214. Korbling M, Huh YO, Durett A, et alAllogeneic blood stem cell transplantationperipheralization and yield of donor-derived primitive hematopoietic progenitor cells (CD34⫹ Thy-1dim) and lymphoid subsets, and possible predictors of engraftment and graft-versus-host disease. Blood. 1995;86:2842-2848. Fujisaki T, Otsuka T, Harada M, et al. Granulocyte colonystimulating factor mobilizes primitive hematopoietic stem cells in normal individuals. Bone Marrow Transplant. 1995;16:57-62. Prosper F, Stroncek D, Verfaillie CM. Phenotypic and functional characterization of long-term culture-initiating cells present in peripheral blood progenitor collections of normal donors treated with granulocyte colony-stimulating factor. Blood. 1996;88:2033-2042. Shustov A, Luzina I, Nguyen P, et al. Role of perforin in controlling B-cell hyperactivity and humoral autoimmunity. J Clin Invest. 2000;106:R39-R47. Storek J, Wells D, Dawson MA, et al. Factors influencing B lymphopoiesis after allogeneic hematopoietic cell transplantation. Blood. 2001;98:489-491. Kerre TC, De Smet G, De Smedt M, et al. Both CD34⫹38⫹ and CD34⫹38⫺ cells home specifically to the bone marrow of NOD/LtSZ scid/scid mice but show different kinetics in expansion. J Immunol. 2001;167:3692-3698. Hogan CJ, Shpall EJ, Keller G. Differential long-term and multilineage engraftment potential from subfractions of human CD34⫹ cord blood cells transplanted into NOD/SCID mice. Proc Natl Acad Sci U S A. 2002;99:413-418. Schultz KR, Paquet J, Bader S, et al. Requirement for B cells in T cell priming to minor histocompatibility antigens and devel633

L. R. Shier et al.

42.

43.

44.

45.

46.

47.

48.

634

opment of graft-versus-host disease. Bone Marrow Transplant. 1995;16:289-295. Ratanatharathorn V, Ayash L, Reynolds C, et al. Treatment of chronic graft-versus-host disease with anti-CD20 chimeric monoclonal antibody. Biol Blood Marrow Transplant. 2003;9: 505-511. Ratanatharathorn V, Carson E, Reynolds C, et al. Anti-CD20 chimeric monoclonal antibody treatment of refractory immune-mediated thrombocytopenia in a patient with chronic graft-versus-host disease. Ann Intern Med. 2000;133:275-279. Miklos DB, Kim HT, Zorn E, et al. Antibody response to DBY minor histocompatibility antigen is induced after allogeneic stem cell transplantation and in healthy female donors. Blood. 2004;103:353-359. Zorn E, Miklos DB, Floyd BH, et al. Minor histocompatibility antigen DBY elicits a coordinated B and T cell response after allogeneic stem cell transplantation. J Exp Med. 2004;199:11331142. Soulas P, Koenig-Marrony S, Julien S, et al. A role for membrane IgD in the tolerance of pathological human rheumatoid factor B cells. Eur J Immunol. 2002;32:2623-2634. Arpinati M, Chirumbolo G, Urbini B, et al. Role of plasmacytoid dendritic cells in immunity and tolerance after allogeneic hematopoietic stem cell transplantation. Transpl Immunol. 2003; 11:345-356. Schots R, Kaufman L, Van Riet I, et al. Proinflammatory cytokines and their role in the development of major trans-

49.

50.

51.

52.

53.

54. 55.

56.

plant-related complications in the early phase after allogeneic bone marrow transplantation. Leukemia. 2003;17:1150-1156. Clark FJ, Freeman L, Dzionek A, et al. Origin and subset distribution of peripheral blood dendritic cells in patients with chronic graft-versus-host disease. Transplantation. 2003;75:221-225. Gorgun G, Miller KB, Foss FM. Immunologic mechanisms of extracorporeal photochemotherapy in chronic graft-versus-host disease. Blood. 2002;100:941-947. Waller EK, Rosenthal H, Sagar L. DC2 effect on survival following allogeneic bone marrow transplantation. Oncology (Huntingt). 2002;16(1 suppl 1):19-26. Chan GW, Gorgun G, Miller KB, et al. Persistence of host dendritic cells after transplantation is associated with graft versus host disease. Biol Blood Marrow Transplant. 2003;9:170-176. Kadowaki N, Ho S, Antonenko S, et al. Subsets of human dendritic cell precursors express different toll-like receptors and respond to different microbial antigens. J Exp Med. 2001; 194:863-869. Iho S. Type I IFN synthesis in plasmacytoid dendritic cells. J Immunol. 2003;171:2767. De Wit D, Olislagers V, Goriely S, et al. Blood plasmacytoid dendritic cell responses to CpG oligodeoxynucleotides are impaired in human newborns. Blood. 2004;103:1030-1032. Poeck H, Wagner M, Battiany J, et al. Plasmacytoid dendritic cells, antigen and CpG-C license human B cells for plasma cell differentiation and immunoglobulin production in the absence of T cell help. Blood. 2004;103:3058-3064.