| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Pathology and Microbiology, University of Nebraska Medical Center, Omaha, Nebraska, USA
Key Words. Matrix metalloproteinase • G-CSF • Flt-3 ligand • Cytokine-induced mobilization • Engraftment
Simon N. Robinson, Ph.D., Department of Genetics, Cell Biology and Anatomy, 986395 Nebraska Medical Center, Omaha, Nebraska 68198-6395, USA. Telephone: 402-559-4390; Fax: 402-559-7328; e-mail: snrobins{at}unmc.edu
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Increased numbers of HSCs in the blood after G-CSF administration correlate with elevated serum levels of interleukin (IL)-8 [8, 9] and matrix metalloproteinase (MMP)-9 [9, 10]. Elevated serum levels of MMP-9 are also observed in mice following IL-8 administration [11]. This, together with the ability of MMP-9-neutralizing antibodies to prevent IL-8-induced hematopoietic progenitor cell (HPC) mobilization in nonhuman primates, suggests that MMP-9 has a role in mobilization [12, 13]. Further, MMP-9 clips six N-terminal amino acids from IL-8, significantly increasing its potency [14] and providing additional evidence that MMP-9 has a role in G-CSF and IL-8-induced mobilization [15].
Granulocytes express receptors for G-CSF and IL-8 [16] and their activation may have a role in G-CSF and IL-8-induced mobilization [11, 17–19]. When activated by G-CSF or IL-8, granulocytes synthesize and release IL-8 [20, 21] and enzymes including MMPs [15, 22–24], neutrophil elastase [23, 25], and cathepsin G [23]. These may disrupt cell-cell and/or cell-extracellular matrix interactions leading to HSC/HPC mobilization [26–29]. Other potential mechanisms may include changes in cell adhesion molecule expression by HSCs [30] and/or stroma [23] during mobilization and/or the expression of MMPs by HSCs [31]. Indeed, the expression of MMPs by HSCs may play a role in their transmigration through basement membranes during mobilization or engraftment [31].
Given the evidence that MMP-9 has an important role in growth factor (GF)-induced mobilization, we investigated whether the absence of MMP-9 would impact G-CSF or Flt-3 ligand (Flt-3L)-induced mobilization. We report that G-CSF and Flt-3L-induced mobilization is intact in MMP-9 knockout (KO) mice, suggesting that MMP-9 is not required for G-CSF- or Flt-3L-induced mobilization. While these results appear contrary to previous evidence suggesting that MMP-9 has a key role in mobilization, we suggest that mobilization in the KO model may be a consequence of the upregulation of compensatory enzymes. These data suggest that, while MMP-9 is not an absolute requirement for G-CSF and Flt-3L-induced mobilization, it may be required for mobilization in wild-type (WT) animals. Further, we demonstrate that the absence of MMP-9 expression by HSCs does not impair their ability to reconstitute hematopoiesis in sublethally irradiated recipients.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recombinant GF Administration
G-CSF (Neupogen Filgrastim; Amgen Inc.; Thousand Oaks, CA; http://www.amgen.com) was administered s.c. twice daily (approximately 8 a.m. and 5 p.m.) at 2.5 µg G-CSF per injection (5 µg G-CSF per day) for 5 consecutive days and a final 2.5-µg injection of G-CSF was given approximately 2 hours prior to assay on day 6. Sterile Dulbecco’s phosphate-buffered saline (DPBS) was administered as an excipient control. Recombinant human Flt-3L (rhuFlt-3L) (Immunex; Seattle, WA; http://www.immunex.com) was administered s.c. at 10 µg per mouse per day for 10 days [33, 34]. Mice were assayed on day 11.
Blood, Spleen, and Bone Marrow
Blood was obtained from the retro-orbital plexus of heparinized, anesthetized mice. Red blood cell counts (RBCs), WBCs, and platelet numbers were determined (System 9000 Hematology Series Cell Counter; Serono-Baker Diagnostics Inc.; Allentown, PA; http://www.abx.com), blood smears were prepared and stained, and a differential WBC was performed. Prior to assay, RBCs in all blood samples were lysed by hypotonic shock and cellularities were determined using a hemacytometer. Single-cell suspensions of spleen and femur BM cells (BMCs) were prepared and cellularities were determined using a H-2000 Hematology Analyzer (Careside; Culver City, CA; http://www.careside.com).
In Vitro Colony-Forming Cell Assays
For each mouse in each group, triplicate granulocyte-macrophage colony-forming unit (CFU-GM) and high proliferative potential colony-forming cell (HPP-CFC) assays were performed for blood, spleen, and BM as previously described [35]. Briefly, CFU-GM assays were performed in semisolid culture media containing 20 ng/ml recombinant murine interleukin (rmuIL)-3 (BioSource International; Camarillo, CA; http://www.biosource.com) and cultured for 7 days. HPP-CFC assays were performed in semisolid culture media containing 20 ng/ml rmuIL-3 and 35% (v/v) L-cell-conditioned medium (a source of M-CSF) and cultured for 14 days. Twenty-four hours prior to counting, cultures were stained with 1 mg/ml p-Iodonitrotetrazolium violet solution (Sigma; St. Louis, MO; http://www.sigmaaldrich.com) [36]. Colonies derived from CFU-GM and HPP-CFC assays were quantified using a dissecting microscope. Means were calculated for the triplicate dishes and frequencies and total numbers of CFU-GM and HPP-CFC per tissue were calculated.
Cobblestone Area-Forming Cell (CAFC) Assay
The frequencies of CAFC in samples of spleen and BM were determined by limiting dilution analysis (LDA), as originally described [37]. Cultures were re-fed weekly and scored positive or negative (assayed on day 35 for the G-CSF study and on day 28 for the Flt-3L study. Assay of cobblestone areas in culture ≥4 weeks was considered to provide a measure of relatively primitive HPCs within the hematopoietic system) for the presence or absence, respectively, of cobblestone areas, using an inverted microscope. The proportion of negative wells at each cell concentration was used in a Poisson-based LDA calculation to determine CAFC frequency [37].
Zymographic Analysis of Gelatinase B (MMP-9) Levels in Isolated Serum
Gelatinase activity in serum samples was determined by electrophoresis through a 4% SDS-polyacrylamide stacking gel into a 7.5% SDS-polyacrylamide separating gel containing 0.15% (w/v) gelatin (gelatin A from porcine skin; Sigma) as an enzyme substrate. One microliter of test serum was diluted in running and loading buffers and electrophoresed at 200 volts using a Mini-Protean II electrophoresis apparatus (Bio-Rad Laboratories; Richmond, CA; http://www.bio-rad.com). The volumes of serum and protein levels loaded per lane were identical, allowing comparisons to be made between the gelatinase activities of different samples. After electrophoresis, SDS was removed from the gels by repeated washing (30 minutes x 2) in 2.5% Triton X-100. Gels were subsequently incubated overnight at 37°C in a solution of 150 mM NaCl, 1 mM MnCl2, 10 mM CaCl2, 0.05% sodium azide, and 50 mM Tris-HCl buffer (pH 7.4). Gels were stained with Coomassie Brilliant Blue R-250 (1% w/v) in 10% (v/v) acetic acid and 30% (v/v) methanol and destained with 10% (v/v) acetic acid and 30% (v/v) methanol. Areas of gelatinase activity appeared transparent against the Coomassie blue-stained background. As a reference, the supernatant from a cell line known to produce MMP-9 and MMP-2 was used. Electronically captured images were analyzed to allow semiquantitative determination of gelatinase activity by densitometry (ImageQuantTM for Windows NT, v 5.2; Molecular Dynamics; Amersham BioSciences; Sunnyvale, CA; http://www.apbiotech.com).
Hematologic Reconstitution Following the Transplantation of WT or MMP-9 KO Marrow
Eight-week-old WT, 129S6/SvEv mice (Taconic) were maintained on irradiated food and acidified water (pH 2.2) for 2 weeks prior to 7 Gy sublethal, whole-body 
-irradiation (Picker V90 cobalt-60 source, 0.61 Gy/minute). Approximately 24 hours postirradiation, mice (five mice per group) were anesthetized and received either 106 WT (129S6/SvEv) BMCs or 106 MMP-9 KO BMCs intravenously. Marrow cells were delivered in 100 µl Hank’s balanced salt solution (HBSS). One group of irradiated mice (n = 5) received excipient alone. Blood was analyzed on days 10, 14, 17, 20, 23, and 36 posttransplantation.
Statistical Analysis
Statistical analyses of data were performed using SPSS 10.0 for Windows (SPSS Inc.; Chicago, IL; http://www.spss.com). Where appropriate, means were compared using the Student’s two-sample t-test; otherwise, data were compared using nonparametric Mann-Whitney analysis. Significance was assumed at p ≤ 0.05.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Femur marrow and spleen cellularities were both significantly lower (by 23% and 46%, respectively) in MMP-9 KO mice compared with age- and sex-matched WT animals. In contrast, the frequency of splenic CFU-GM in MMP-9 KO mice was significantly greater (2.7-fold) than that in WT mice (Fig. 1). However, due to the significantly reduced steady-state spleen cellularity, there was no significant difference in the absolute number of CFU-GM per spleen.
|
|
Changes in the Blood
Granulocyte colony-stimulating factor resulted in significantly greater WBC cellularity of both the WT (6.2-fold) and MMP-9 KO (7.5-fold) mice when compared with their respective DPBS-treated cohorts (Table 1
). The greater WBC cellularities in both the WT and MMP-9 KO mice were not statistically different between the two groups. This was associated with a significant increase in granulocyte frequency (4.7- and 3.8-fold, respectively) and number (28.9- and 27.3-fold, respectively) in the blood. In WT mice, but not in MMP-9 KO mice, a significantly lower (20%) platelet count was also observed compared with the respective DPBS-treated cohort (Table 1).
Following G-CSF, there were no significant differences in the numbers and frequencies of CFU-GM and HPP-CFC between the blood of WT and MMP-9 KO animals and that of DPBS-treated animals (Fig. 2). The number and frequency of CFU-GM in the blood were significantly increased, 26- and 14-fold, respectively, in WT mice, and 23- and 13-fold, respectively, in MMP-9 KO mice. Similarly, the number and frequency of HPP-CFC in the blood were significantly increased, 40- and 52-fold, respectively, in WT mice and 42- and 15-fold, respectively, in MMP-9 KO mice.
|
Changes in the Spleen
In WT mice, G-CSF resulted in significantly greater spleen cellularity (twofold) (Table 1), frequencies of HPP-CFC and day-35 CAFC (fivefold and twofold, respectively), and numbers of CFU-GM, HPP-CFC, and day-35 CAFC (threefold, ninefold, and fourfold, respectively) (Fig. 1) in the spleen compared with the DPBS-treated cohort. Similarly, in MMP-9 KO mice, G-CSF also resulted in significantly greater spleen cellularity (threefold); numbers of CFU-GM, HPP-CFC, and day-35 CAFC (ninefold, fivefold, and sixfold, respectively); and frequencies of CFU-GM, HPP-CFC, and day-35 CAFC (all threefold) (Fig. 1). Following G-CSF, the spleen cellularity of MMP-9 KO mice was increased, but to a significantly lesser extent (25%) than similarly treated WT mice. The increase in the numbers and frequencies of splenic HPP-CFC and day-35 CAFC was not significantly different between the MMP-9 KO and WT mice. However, the number and frequency of CFU-GM were significantly greater in the MMP-9 KO mice compared with the WT mice (threefold and twofold, respectively).
Changes in the Femur
In WT and MMP-9 KO mice, G-CSF did not result in significant differences in femur marrow cellularity or the number or frequency of HPP-CFC and day-35 CAFC in the marrow compared with the DPBS-treated cohorts (Table 1). However, the femur marrow cellularity of the MMP-9 KO mice was significantly lower than that of WT mice during steady-state hematopoiesis and remained significantly less than that in WT mice following G-CSF administration. In contrast, there were no significant differences between the numbers and frequencies of HPP-CFC and day-35 CAFC in the marrow of WT and MMP-9 KO mice following G-CSF administration.
Zymographic Analysis of Serum Gelatinase Activity
Zymography confirmed the absence of MMP-9 activity in serum from steady-state and G-CSF-treated MMP-9 KO mice (Fig. 3A). In contrast, MMP-2 activity was present in serum samples from MMP-9 KO mice and did not vary significantly regardless of G-CSF injection. A similar lack of MMP-2 upregulation was observed in WT mice receiving G-CSF. However, MMP-9 activity, evident in the serum of WT mice during steady-state hematopoiesis (PBS-treated), was significantly greater (>sixfold) in the serum of mice receiving G-CSF (quantified by image analysis).
|
). Flt-3L significantly increased the number of CFU-GM and day-28 CAFC in the blood of WT (17- and 11-fold, respectively) and MMP-9 KO mice (12- and 30-fold, respectively) (Fig. 4). However, the number of CFU-GM in the blood of the WT mice receiving Flt-3L was significantly greater (2.7-fold) than that in MMP-9 KO mice receiving Flt-3L.
|
|
) and the numbers of CFU-GM, HPP-CFC, and day-28 CAFC (Fig. 5) were significantly increased in both the Flt-3L-treated WT (3-, 67-, 118-, and 5-fold, respectively) and MMP-9 KO mice (2-, 20-, 12-, and 27-fold, respectively). A statistical comparison of the two groups revealed that there were significantly more HPP-CFC (fourfold) and significantly fewer day-28 CAFC (twofold) in the spleens of Flt-3L-treated WT mice than in the spleens of Flt-3L-treated MMP-9 KO mice (Fig. 5).
|
). In contrast, the numbers of CFU-GM and HPP-CFC per femur were significantly greater in the Flt-3L-treated MMP-9 KO mice (4.4- and 3.8-fold, respectively) than in DPBS-treated mice. However, the number of CAFC per femur was significantly greater (3.1-fold) in the Flt-3L-treated WT mice.
Hematologic Reconstitution Following Transplantation of WT or MMP-9 KO Marrow
In studies examining the role of MMP-9 in HPC engraftment, we examined the ability of WT and MMP-9 KO BMCs to reconstitute sublethally irradiated WT mice. Mice received 7 Gy whole-body -irradiation and all survived irrespective of the source of infused BMCs. When examined on day 36, WBC, RBC, and platelet numbers and WBC differentials were at, or approaching, preirradiation levels (data not shown). In these studies, no significant differences in hematologic recovery were demonstrable between irradiated mice receiving 106 WT or 106 MMP-9 KO marrow cells (Fig. 6). However, significant differences were observed in hematologic recovery when nontransplanted WT mice were compared with WT mice receiving either WT or MMP-9 KO BMCs. The significant nadirs in RBC, platelet, and granulocyte numbers after irradiation in nontransplanted mice were absent or significantly lower in mice receiving either WT or MMP-9 KO BMCs.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The mobilization of HSCs from the marrow to the blood is associated with significantly elevated levels of serum IL-8 [8, 9], which is thought to activate granulocytes and stimulate the release of a number of enzymes, including MMP-9 [16]. Indeed, elevated levels of serum MMP-9 are detected following G-CSF administration in humans [9, 10]. Consistent with and extending these findings are our observations of a sixfold elevation in serum MMP-9 levels in WT mice following G-CSF administration. The importance of MMP-9 to mobilization is demonstrated by the inhibition of GF-induced mobilization with antibodies to MMP-9 in primates and the correlation of elevated serum MMP-9 levels with the appearance of mobilized HPCs in the blood [12].
To examine the role of MMP-9 in HPC mobilization, G-CSF or Flt-3L were given to MMP-9 KO mice. They were both found to result in significantly greater spleen cellularities and numbers and frequencies of CFU-GM and HPP-CFC per spleen and per milliliter of blood compared with DPBS-treated cohorts. These hematopoietic changes were similar to those observed in WT mice receiving G-CSF or Flt-3L, suggesting that the genetic absence of MMP-9 activity does not significantly impact G-CSF- or Flt-3L-induced HPC mobilization. These studies extend prior studies demonstrating that IL-8-induced HPC mobilization is also intact in MMP-9 KO mice [19].
These results are paradoxical when compared with the absence of IL-8-induced HPC mobilization following the administration of MMP-9-neutralizing antibodies in primates. This discrepancy may be a consequence of the acute blockade of MMP-9 activity in WT animals by MMP-9-neutralizing antibodies, while in genetically MMP-9-deficient animals there may be a compensatory upregulation of proteases that masks the MMP-9 deficiency. One candidate protease is MMP-2, which shares a similar spectrum of enzymatic activity with MMP-9. However, consistent with previous reports [10, 38], zymographic analysis of serum samples suggests that levels of MMP-2 are unchanged during steady-state and G-CSF-mobilized hematopoiesis (although elevated MMP-2 levels are observed in extracellular fluids of mice during cyclophosphamide- or cyclophosphamide/G-CSF-induced mobilization [38]). Other enzymes, such as neutrophil elastase [10, 25] and/or cathepsin G [23], also have been shown to have a role in GF-induced HPC mobilization and may undergo compensatory upregulation in MMP-9 KO mice. Thus, we suggest that MMP-9 is not an absolute requirement for G-CSF-, Flt-3L-, or, as reported by others [19], IL-8-induced mobilization, although it may have an important role in mobilization in WT animals. Recently, the role of MMP-9 was also investigated in sulfated polysaccharide-induced HPC mobilization using the MMP-9 KO mouse model [39–41]. Results from the studies demonstrated that sulfated polysaccharide-induced mobilization was intact in MMP-9 KO mice, suggesting that, as with G-CSF, IL-8, and Flt-3L, the presence of MMP-9 activity is not an absolute requirement for mobilization.
Hematopoietic stem cells also express MMPs [31, 42], and their expression plays an important role in their in vitro migration through reconstituted basement membrane [31]. To further investigate the role of MMP-9 in HSC transmigration, we transplanted WT or MMP-9 KO BMCs into sublethally irradiated WT mice and followed hematologic reconstitution. However, the transplantation of MMP-9-deficient (KO) HSCs did not impair the hematologic recovery of the sublethally irradiated recipients compared with those receiving MMP-9-competent (WT) HSCs. Indeed, hematologic recovery was similarly and significantly improved in the sublethally irradiated mice regardless of the HSC source. It should be emphasized that with the transplantation of both WT and MMP-9 KO BMCs, significantly improved hematologic recovery was observed relative to nontransplanted mice.
Based on our observation that progenitor cell mobilization by G-CSF appears intact in the MMP-9-deficient mouse, our conclusion that MMP-9 is not an absolute requirement for progenitor cell mobilization by G-CSF initially appears contrary to the report of Heissig et al. [43]. They reported that transplantation of cells from G-CSF-mobilized, MMP-9-deficient mice failed to rescue lethally irradiated syngeneic mice and concluded that MMP-9 plays an important role in G-CSF-induced stem cell mobilization. Their conclusion is further strengthened by the demonstration that cells from G-CSF-mobilized WT mice previously treated with a potent MMP inhibitor similarly failed to rescue lethally irradiated recipients. This later observation is, in part, consistent with the demonstration that antibodies against MMP-9 impair IL-8-induced progenitor cell mobilization [11, 12]. However, Heissig et al. [43] reported the use of G-CSF at 50 µg/kg body weight daily for 5 days prior to harvest of peripheral blood mononuclear cells. In contrast, we and other groups [23, 38] report the use of recombinant human G-CSF at 250 µg/kg body weight for 5 days for HPC mobilization in mice. At this dose of recombinant human G-CSF, Levesque et al. [38] reported the highly proteolytic nature of the murine BM microenvironment. We suggest that, in MMP-9-deficient mice, enzymes that may compensate for the deficiency in MMP-9 may be part of this highly proteolytic milieu. In contrast, the administration of G-CSF at 50 µg/kg body weight for 5 days may induce a less proteolytic environment such that any putative compensatory enzymes are produced at lower levels.
Indeed, a similar rationale may explain the observed variation in the extent of HPC mobilization to the spleen in WT and MMP-9 KO mice following Flt-3L administration (10 µg/day x 10 days). This treatment may induce a proteolytic milieu such that the potential compensatory enzymes are induced at low levels, resulting in lower and more variable HPC mobilization. Further, data from Liu et al. [44] have suggested that the mechanism of action for HPC mobilization by Flt-3L differs from that of G-CSF. Thus, despite the MMP-9 deficiency, HPCs are mobilized by both G-CSF and Flt-3L, and we suggest that this is associated with the upregulation of compensatory enzymes. The data presented herein suggest that HPC mobilization is observed, with some variation, in both WT and MMP-9 KO mice by optimal G-CSF and Flt-3L mobilization protocols. Thus, we conclude that MMP-9 is not an absolute requirement for progenitor cell mobilization by Flt-3L.
In conclusion, studies using MMP-9 KO mice demonstrate that MMP-9 activity is not an absolute requirement for G-CSF and Flt-3L-induced HPC mobilization, or for the functional engraftment of HSCs in sublethally irradiated mice. However, this conclusion does not preclude a role for MMP-9 in the mobilization of HSCs in competent animals, since the compensatory upregulation of other proteases may mask the effects of the MMP-9 deficiency.
|
ACKNOWLEDGMENT |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Tjwa, S. Janssens, and P. Carmeliet Plasmin therapy enhances mobilization of HPCs after G-CSF Blood, November 15, 2008; 112(10): 4048 - 4050. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Iwata, M. Pillai, A. Ramakrishnan, R. C. Hackman, H. J. Deeg, G. Opdenakker, and B. Torok-Storb Reduced expression of inducible gelatinase B/matrix metalloproteinase-9 in monocytes from patients with myelodysplastic syndrome: correlation of inducible levels with the percentage of cytogenetically marked cells and with marrow cellularity Blood, January 1, 2007; 109(1): 85 - 92. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Cheretakis, R. Leung, C. X. Sun, Y. Dror, and M. Glogauer Timing of neutrophil tissue repopulation predicts restoration of innate immune protection in a murine bone marrow transplantation model Blood, October 15, 2006; 108(8): 2821 - 2826. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Aicher, A. M. Zeiher, and S. Dimmeler Mobilizing Endothelial Progenitor Cells Hypertension, March 1, 2005; 45(3): 321 - 325. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Deem and J. M. Cook-Mills Vascular cell adhesion molecule 1 (VCAM-1) activation of endothelial cell matrix metalloproteinases: role of reactive oxygen species Blood, October 15, 2004; 104(8): 2385 - 2393. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Pelus, H. Bian, A. G. King, and S. Fukuda Neutrophil-derived MMP-9 mediates synergistic mobilization of hematopoietic stem and progenitor cells by the combination of G-CSF and the chemokines GRO{beta}/CXCL2 and GRO{beta}T /CXCL2{Delta}4 Blood, January 1, 2004; 103(1): 110 - 119. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
) and MMP-9-deficient (
) mice. Data are presented as mean ± standard error (SE). *significantly different (p ≤ 0.05) from respective DPBS-treated cohort. #significantly different (p ≤ 0.05) from similarly treated WT mice.
) or 106 MMP-9 KO (
) BMCs into sublethally irradiated (7 Gy whole body 
). The hatched area (///////) provides data from control, nonirradiated mice. Irrespective of whether BMCs were of WT or MMP-9 KO origin, RBC, platelet, and granulocyte reconstitution was similarly and significantly improved compared with mice receiving HBSS alone. Data are presented as mean ± standard error (5 mice per group). #significantly different (p ≤0.05) from control (nonirradiated mice). @significantly different (p ≤0.05) from mice receiving bone marrow transplant (either 106 WT or 106 MMP-9 KO cells).