- Email: [email protected]
Serpina1 (α1-AT) is synthesized in the osteoblastic stem cell niche
Experimental Hematology 2009;37:641–647
Serpina1 (a1-AT) is synthesized in the osteoblastic stem cell niche H. Bea Kuiperija,*, Melissa van Pela, Karien E. de Rooijb, Rob C. Hoebenc, and Willem E. Fibbea a Department of Immunohematology and Bloodtransfusion, Leiden University Medical Center, Leiden, The Netherlands; bDepartment of Endocrinology and Metabolic Diseases, Leiden University Medical Center, Leiden, The Netherlands; cDepartment of Molecular Cell Biology, Leiden University Medical Center, Leiden, The Netherlands
(Received 28 August 2008; revised 3 February 2009; accepted 4 February 2009)
Objective. Previously, we identified Serpina1 as a potent inhibitor of hematopoietic stem and progenitor cell (HSC/HPC) mobilization. Serpina1 protein is found in the bone marrow (BM) extracellular fluid and concentrations are decreased during granulocyte colony-stimulating factor–induced HSC/HPC mobilization in mice. In addition, administration of exogenous Serpina1 protein inhibits HSC/HPC mobilization. BM cells responsible for production and secretion of Serpina1 remain unknown. Here, we examined the expression of Serpina1 in order to identify cell populations of the BM that synthesize Serpina1. Materials and Methods. Osteoblast (OB) and hematopoietic BM cell fractions were isolated from femurs, tibias, and humeri obtained from untreated mice. Subsequently, each BM fraction was examined for the production of Serpina1 messenger RNA and protein by quantitative real-time polymerase chain reaction, Western blotting, and immunohistochemistry. Results. Quantitative real-time polymerase chain reaction analysis showed that Serpina1 messenger RNA is produced at high levels by OB compared to hematopoietic BM cells. Furthermore, Western blot analysis indicated that Serpina1 protein was secreted by OB. In contrast, no Serpina1 protein could be detected in the supernatant obtained from overnight cultured hematopoietic BM cells. Finally, in BM sections obtained from the femurs of untreated mice, Serpina1 protein was detected in OB cells lining the bone. Conclusion. Serpina1 protein in the BM extracellular fluid is predominantly produced by OB. This indicates that Serpina1 may play a regulatory role in the maintenance of HSC in the OB stem cell niche. Ó 2009 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc.
Serpina1 (also known as a1-antitrypsin or a1-proteinase inhibitor) belongs to a superfamily of serine protease inhibitors with highly conserved secondary structural elements. They inhibit their target proteases by covalent binding, and this inhibition is irreversible and required for effective control of the proteolytic cascade that may be initiated by the release of just a few protease molecules. In mammals, serpins play a role in a range of proteolytic processes, including blood clotting, inflammation, and turnover of extracellular matrix [1,2].
Offprint requests to: Melissa van Pel, Ph.D., Section of Stem Cell Biology, Department of Immunohematology and Bloodtransfusion, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, The Netherlands; E-mail: [email protected] *Current address: Department of Neurology and Donders Centre for Neuroscience, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.
Serpina1 is, together with a2-macroglobulin, the major plasma protease inhibitor. Besides being present in plasma, Serpina1 is also found in the bone marrow (BM) extracellular fluid [3,4]. Although the main substrate of Serpina1 is neutrophil elastase, also various other proteases, including trypsin and chymotrypsin, can be inhibited [5]. The specificity of the protease inhibitor is determined by its so-called reactive center loop, a domain of approximately 20 residues that extends from the body of the Serpina1 polypeptide. Binding of the protease to Serpina1 is directed via this reactive center loop, which acts as a substrate for the protease. The protease cleaves the reactive center loop, resulting in a conformational change, followed by crushing and subsequent destruction of the protease [1,6]. Proteases have shown to be regulatory mediators of cytokine-induced hematopoietic stem and progenitor cell (HSC/HPC) mobilization. Studies from our laboratory and
0301-472X/09 $–see front matter. Copyright Ó 2009 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc. doi: 10.1016/j.exphem.2009.02.004
642
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647
others indicated a critical role for the matrix metalloproteinase-9 (gelatinase B), neutrophil elastase, and cathepsin G in cytokine- and cyclophosphamide-induced HSC/HPC mobilization [7–9]. Recently, we reported that Serpina1 is a potent inhibitor of HSC/HPC mobilization [3,4]. We showed that administration of Serpina1 prior to interleukin-8 (IL-8) injection inhibited HSC/HPC mobilization. Furthermore, low-dose total body irradiation increased Serpina1 levels in the BM extracellular fluid, and inhibited both IL-8 and granulocyte colony-stimulating factor– induced HSC mobilization [4]. HSC are localized in the BM in a specialized microenvironment called the hematopoietic stem cell niche, which is located at the endosteal surface of the bone. Herein, primitive HSC are in close contact with stromal cells that form the hematopoietic microenvironment of the BM. The stromal tissue of the marrow creates a three-dimensional network comprised of cells (including endothelial cells, fibroblasts, and osteoblasts [OB]), extracellular matrix components, and hematopoietic growth factors that collectively constitute a unique microenvironment. It is believed that the stem cell niche directs regulation of self-renewal and differentiation of primitive HSC. Cytokines play a major role in this process and act both via cell-to-cell contact, or as soluble factors [10,11]. OBs are important regulators of the HSC niche [12–14]. By secreting osteopontin the number of HSC is controlled [15,16]. Angiopoietin-1 (Ang-1), which is secreted by OB, binds Tie2 and regulates HSC activity on the level of quiescence and repopulating ability as well as on the level of interaction of HSC with the niche [17]. Furthermore, OB produce Jagged 1, which regulates HSC maintenance and differentiation via Notch signaling [12,18]. Here we show that the serine protease inhibitor Serpina1 is predominantly expressed by OB. We hypothesize that Serpina1 is involved in HSC maintenance in the OB niche by inhibiting proteases, including neutrophil elastase.
Materials and methods Mice Eight- to 12-week-old male Balb/c mice were purchased from Charles River Laboratories (Maastricht, The Netherlands). The animals were fed commercial rodent food and acidified water ad libitum. The Institutional Ethical Committee on Animal Experiments approved all experimental protocols. Isolation of BM and OB-enriched cell fractions Femurs, tibias, and humeri were isolated from Balb/c mice and the bones were flushed with ice-cold phosphate-buffered saline (PBS) to harvest hematopoietic BM cells. Thereafter, the flushed bones were crushed using scissors and incubated at 37 C alternating with 1 mg/mL collagenase I (Sigma, Zwijndrecht, The Netherlands) plus 0.25% trypsin in PBS and 4 mM ethylene diamine tetraacetic acid in PBS. After each incubation step, cells were harvested and numbered consecutively OB fraction 1 through 5. After the last
incubation step, bone fragments were treated with Trizol reagent (Invitrogen, Breda, The Netherlands) to obtain OB fraction 6. The cells in BM and OB fraction 1 to 5 cells were either lysed in Trizol reagent for RNA isolation or cultured. Previous studies have shown that these cells have osteoblastic characteristics and have been extensively used to map OB maturation [19,20]. Analysis of Serpina1 secretion by Western blotting To analyze Serpina1 protein secretion, 3 106 hematopoietic BM cells and approximately 0.8 106 OB fraction 5 cells were seeded in a well of a six-well plate containing a–minimum essential medium supplemented with 10% fetal bovine serum (Gibco, Grand Island, NY, USA), penicillin, streptomycin, and L-glutamine (all from Gibco). After 24 hours of incubation, cells were washed twice with PBS and grown in 1 mL a–minimum essential medium without fetal bovine serum for 24 hours. The serum-free culture supernatant was harvested and cleared for remaining cells by centrifugation. Subsequently, proteins from 15 mL of these culture supernatants were separated on sodium dodecyl sulfate polyacrylamide gels and blotted onto polyvinylidene difluoride membranes (Roche, Almere, The Netherlands). Membranes were blocked using 1% blocking reagent (Roche) in Tris-buffered saline, followed by incubations with a primary antibody directed against mouse Serpina1 (previously obtained by immunization of rabbits [4]) and a horseradish peroxidase–conjugated goat anti-rabbit Ig secondary antibody (Promega, Leiden, The Netherlands). Proteins were visualized using standard enhanced chemiluminescence (Roche) and autoradiography. Alkaline phosphatase activity assay To assess alkaline phosphatase (ALP) activity, OB fraction 5 cells were cultured up to passage 3 in a–minimum essential medium supplemented with 10% fetal bovine serum, pen-strep and L-glutamine. Cells were fixed in methanol, washed with PBS and incubated with ALP substrate (0.02% Naphthol AS-MX salt; Sigma, N5000), 0.06% Fast Blue RR salt (Sigma, F0500), 0.1 M Tris-HCl [pH 8.8], 0.01% MgSO4) for 5 minutes and washed twice with PBS. ALP staining was analyzed by inverted light microscopy. Quantitative reverse transcription polymerase chain reaction (qPCR) RNA was isolated from hematopoietic BM and OB fraction 1 to 6 Trizol lysates and subsequently complementary DNA was generated. Primer sets specific for Serpina1, Osteocalcin, Runx2 splice variant (MASN), Sclerostin, PECAM, N-cadherin (Ncad), glyceraldehyde phosphate dehydrogenase, b2M (Table 1), and Cathepsin K (QuantiTect primer assay set, Qiagen, Venlo, The Netherlands) were used for qPCR experiments. The qPCR was performed using SYBR Green qPCR Mastermix (Eurogentec, Maastricht, The Netherlands) and the iQ qPCR analyzer (Bio-Rad, Veenendaal, The Netherlands), according to the manufacturers instructions. Relative gene expression was calculated using the comparative threshold cycle (CT) method, with Gapdh and b2M as the endogenous reference genes. Immunohistochemistry on bone sections To identify Serpina1 in situ, femurs of mice were fixed in 10% formalin in PBS, decalcified in 10% ethylene diamine tetraacetic acid, embedded in paraffin and sectioned (5 mm). Immunohistochemistry was performed essentially as described previously [21], with some modifications. To block nonspecific activity,
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647
643
Table 1. Quantitative reverse transcription polymerase chain reaction primer sets Gene Serpina1 OC MASN SOST PECAM Ncad GAPDH b2M
Forward primer
Reverse primer
GATGGGAAGATGCAGCATC ACAGACAAGTCCCACACAGCAGC AAGTGCGGTGCAAACTTTCT TCCTCCTGAGAACAACCAGAC ACGAGGTACAGTCTGGAAGTG GAAGGCAATCCCACTTATGG ATGGCCTTCCGTGTTCCTAC CCTGTATGCTATCCAGAAAACC
TCCAGAGATGGACAGTCTG TGAAGGCTTTGTCAGACTCAGGGC TCTCGGTGGCTGGTAGTGA TGTCAGGAAGCGGGTGTAGTG TTGTTTGTTGTGCCAAAACG CTTCTCCGTAGAAAGTCATG CCTGCTTCACCACCTTCTTG GTTCAGTATGTTCGGCTTCC
sections were pretreated with hydrogen peroxide. Antigen retrieval was performed using 5 mg/mL hyaluronidase in 0.5% Boehringer milk protein (Boehringer, Mannheim, Germany) in PBS (pH 5). Sections were blocked using 5% normal goat serum and 0.5% Boehringer milk protein in Tris-buffered saline-Tween, followed by incubation with the primary antibody directed against mouse Serpina1 and a biotinylated goat anti-rabbit secondary antibody (Dako, Heverlee, Belgium). For detection of Serpina1, sections were incubated with streptavidin–horseradish peroxidase (Dako). The signal was enhanced using biotinyl tyramide, followed by a second incubation step with streptavidin–horseradish peroxidase. Serpina1 was visualized using 3,30 -diaminobenzidine as a substrate for horseradish peroxidase, and hematoxylin was used as counterstaining. To determine the specificity of the staining, the antibody directed against mouse Serpina1 protein was pre-incubated with 25 mg purified mouse Serpina1 protein [4] for 30 minutes. Subsequently, bone sections were incubated with the blocked antibody, followed by a staining procedure as described here.
and hematopoietic BM (Fig. 1A). Similarly, MASN levels were significantly increased in fraction 5 and 6, compared to fraction 1 to 4 and hematopoietic BM (Fig. 1A). Sclerostin mRNA levels were strongly induced in OB fraction 6, compared to other OB fractions, but could not be detected in hematopoietic BM cells (data not shown). As a control,
Analysis of Serpina1 isoforms For the analysis of the different isoforms of Serpina1 in murine tissues, Serpina1 mRNA fragments were amplified by PCR using either the Serpina1 forward and reverse primers for qPCR (Table 1), or the qPCR Serpina1 forward primer and the 30 -Serpina1 reverse primer (CAATTCAGAAGGAAGGATG). The product of respectively 117 and 454 bp was subcloned into the pCRIITOPO vector (Invitrogen), and sequences of several clones were analyzed using either the M13-reverse sequence primer or the 30 -Serpina1 primer in the sequence reactions.
Results Isolation of OB-enriched cell fractions from femurs, tibia and humeri To identify the cell populations in BM that produce/express Serpina1, hematopoietic BM cells and OB cell fractions were isolated from murine femurs, tibias, and humeri. To confirm the presence of OB in each fraction, mRNA levels of the OB markers osteocalcin, the Runx2 splice variant MASN, and Sclerostin were measured [22–24]. The results of OB fraction 1 to 4 were taken together, since no differences between these fractions were observed. Osteocalcin messenger RNA (mRNA) levels were significantly increased in OB fraction 6, compared to fraction 1 to 5
Figure 1. Serpina1 messenger RNA (mRNA) levels in primary osteoblast (OB) cell fractions. (A) mRNA levels of several OB markers were quantified by quantitative reverse transcription polymerase chain reaction. Relative mRNA expression in OB cell fraction 1-4, 5 and 6, compared to bone marrow (BM) is plotted. (B) The relative mRNA expression of Serpina1 is shown in OB cell fraction 1 to 6, compared to mRNA levels in BM. Data represent the mean fold induction 6 standard deviation. *p ! 0.05, **p ! 0.01 vs BM (A) or vs previous fraction (B).
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647
644
the mRNA level of the endothelial marker PECAM was measured. PECAM mRNA levels remained unaltered in all the OB cell fractions, and were comparable to the levels in hematopoietic BM cells (Fig. 1A). Interestingly, Ncad mRNA levels were also significantly increased in OB fraction 5 and 6, compared to hematopoietic BM cells and OB fraction 1–4 (Fig. 1A). To investigate a possible contamination of osteoclasts in the OB fractions, Cathepsin K mRNA levels were analyzed. Compared to total BM cells, 15-fold lower levels of Cathepsin K were detected in OB fractions 1-5 and a 22.9-fold reduction in OB fraction 6 (data not shown). Together, these data indicate that during isolation of OB fractions, the differentiation level increased compared to the preceding fraction. Serpina1 mRNA is highly expressed in OB-enriched cell fractions To identify the cell populations that synthesize Serpina1, Serpina1 mRNA levels were determined in the OBenriched cell fractions, and compared to Serpina1 mRNA levels in hematopoietic BM cells. Serpina1 mRNA levels were up to 6 104 times increased in OB-enriched cell fractions (range, 1 102 to 6 104 in fractions 1 to fraction 6, respectively) compared to hematopoietic BM cells (Fig. 1B). Furthermore, in fraction 5, the relative Serpina1 mRNA level was significantly increased compared to fraction 1–4, and this increase was even more pronounced in fraction 6. These data indicate that Serpina1 mRNA levels are significantly increased in OB fraction, compared to hematopoietic BM cell fraction. In addition, upon maturation of OBs, the levels of Serpina1 increase accordingly.
previously published results of other groups, the expression of three isoforms in the liver (Serpina1a, Serpina1b, and DOM6) was confirmed (Table 2) [26]. Interestingly, in hematopoietic BM cells, only the Serpina1b isoform was identified, whereas in OB fraction 5/6 cells, a mixture of isoforms was detected (Table 2). Besides the isoforms present in the liver, this mixture included a fourth isoform, DOM7, which has not been identified in Balb/c mice previously. Of note, due to significant sequence overlap, part of the sequences could only be determined as being either Serpina1a or DOM6. Serpina1 protein is produced by cultured OB To further assess whether osteoblasts secrete Serpina1 protein, we cultured OB fraction 5 cells and hematopoietic BM cells in serum-free medium for 24 hours. The cultured fraction 5 cells phenotypically resembled OB. Staining for ALP activity indicated that about 30% of cultured cells showed a clear purple stain after incubation with the chromogenic conversion substrate for ALP, indicating the presence of OB (Fig. 2A). Next, Serpina1 protein levels were determined in the culture supernatants of the OB fraction 5 and hematopoietic BM cells by Western blot analysis. Serpina1 protein was
Differential expression of Serpina1 isoform mRNA in murine tissues To date, seven isoforms of Serpina1 have been identified in the mouse [6,25]. Three of these isoforms are expressed in the liver of Balb/c mice: DOM1 (Serpina1a), DOM2 (Serpina1b), and DOM6 [26]. Because Serpina1 isoforms exhibit different substrate specificities [6,27], we determined the isoform identity of at least 10 clones in liver, BM, and in OB fraction 5/6 cells (Table 2). In line with Table 2. Expression of serpina1 isoforms in murine tissues Tissue Isoforma
Liver
Bone marrow
Osteoblast fraction 5/6
Serpina1a Serpina1b DOM6 DOM7 Serpina1a or DOM6b
2/19 6/19 5/19 0/19 6/19
0/10 10/10 0/10 0/10 0/10
ND 13/24 ND 2/24 9/24
ND 5 not determined. Expressed as number of cases out of total. b Cases for which the sequence could not be discriminated for Serpina1a or DOM6. a
Figure 2. Cultured primary osteoblasts (OB) produce Serpina1 protein. (A) Cultured OB fraction 5 cells were stained for alkaline phosphatase (ALP) activity. ALP activity is shown in purple. (B) Culture supernatant obtained from overnight-cultured hematopoietic bone marrow (BM) cells and OB fraction 5 cells were processed for Western blot analysis. Serpina1 protein was visualized by immunoblotting and standard enhanced chemiluminescence.
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647
found in the culture supernatant obtained from the culture of OB-enriched fraction 5 cells, whereas no Serpina1 protein could be detected in the culture supernatant obtained from the culture of hematopoietic BM cells (Fig. 2B). This indicates that the mRNA levels in OB correlate with Serpina1 protein production and secretion. Serpina1 protein is produced in bone-lining OB To investigate whether the Serpina1 protein is also expressed in vivo by OBs, we stained bone sections obtained from femurs of untreated mice with anti-murine Serpina1 antibodies. Serpina1 was observed in OBs lining the bone, but also in the more differentiated osteocytes embedded in the bone matrix (Fig. 3A). Because there was also Serpina1 staining in and between hematopoietic BM cells, we tested the specificity of the Serpina1 antibody. Hereto, the primary antibody was incubated with mouse Serpina1 protein prior to the staining of the femur sections. This resulted in specific blocking of the Serpina1 signal in
Figure 3. Serpina1 protein is produced by bone-lining osteoblasts (OB). (A) Immunohistochemistry of Serpina1 in bone sections. Serpina1 protein is shown in brown, whereas nuclei are stained purple by hematoxylin counterstaining. Scale bar: 80 mm; the insert has 4 extra magnification. (B) Immunohistochemistry on bone sections as in (A), except that here the primary Serpina1 antibody was incubated with mouse Serpina1 protein before use.
645
both OB and hematopoietic BM (Fig. 3B) indicating that a specific Serpina1 signal was obtained. Taken together, these results indicate that Serpina1 protein is produced by bone-lining osteoblasts and osteocytes.
Discussion In the present article, we show that Serpina1 is synthesized and secreted by OB lining the inner surface of the bone cavity (bone-lining OB), which are the cells that are important for maintenance of HSC [12,13]. This implies that Serpina1 is produced in the OB stem cell niche. The OB stem cell niche serves as a specific microenvironment for maintenance of primitive HSC via both adhesive and soluble interactions between OB and HSC, including Ncad [28]. Ncad is the only member of the cadherin family that is both present on HSC and OB, and Ncad expression has been reported to support the interaction of OB with HSC, and keep HSC quiescent [13]. Recently, Lymperi et al. reported that an increase in OB number, without increased Ncad expression levels, did not result in increased HSC numbers [29]. They speculated that for increased HSC numbers, an increase in special Ncadpositive OB is needed. Our data clearly show that an increase in Ncad expression levels correlates with increased differentiation status of OB and with increased Serpina1 expression. This suggests correlation between Ncad expression levels, Serpina1 expression levels, and stem cell maintenance. Serpina1 is predominantly synthesized in the liver, where the protein is delivered to the bloodstream. Its main function is protecting the lung from proteolytic damage by inhibiting neutrophil elastase. Besides liver cells, macrophages, monocytes, platelets, and recently chondrocytes have been reported to produce Serpina1 protein [30–32]. Production of Serpina1 by chrondrocytes strengthens our results, because both chondrocytes and OB differentiate from mesenchymal stem cells. However, while Serpina1 produced by OB likely protects against the proteolytic action of enzymes in the BM, the function of Serpina1 expression in chondrocytes remains elusive. In the mouse, two classical and five nonclassical Serpina1 isoforms are expressed. The classical isoforms, Serpina1a and Serpina1b, can inhibit neutrophil elastase. For this function, it is important that the so-called p1 residue is a methionine. In the nonclassical isoforms of Serpina1 (Serpina1c-e, DOM6, and DOM7) the p1 residue is a different amino acid, and these isoforms have specificity for other proteases than neutrophil elastase [6]. Interestingly, we detected DOM7 mRNA in Balb/c OB cell fractions. The presence of this isoform in Balb/c mice has not been reported before, and this may be due to the fact that others have focused on Serpina1 isoform expression in liver tissue. We observed that the expression of Serpina1 isoforms in OB, besides DOM7 expression, resembled those of liver
646
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647
cells, containing a mix of classical and nonclassical isoforms. Interestingly, hematopoietic BM cells did synthesize Serpina1 mRNA molecules up to several log scales less than did OB, but they purely produced the classical isoform Serpina1b. More research has to be done to determine the consequence of these results. Proteases, including neutrophil elastase, Cathepsin G, and matrix metalloproteinase–9, play an important role in HSC mobilization [7–9]. We and others have hypothesized that cytokine-induced HSC/HPC mobilization is determined by a critical balance between serine proteases and serine protease inhibitors in the BM extracellular fluid [3,4]. Indeed, Serpina1 protein levels in the BM space are inversely correlated with mobilization of HSC ([3]; unpublished results [H.B.K., M.vP.]). This suggests that under steady-state conditions, Serpina1 may be involved in retention of HSC in the niche, and thus may play an important role in the protection of HSC from mobilizing proteases. Moreover, we hypothesize a possible role for Serpina1 in regulating the exit of HSC from the stem cell niche. Therefore, for further research it will be interesting to see whether Serpina1 mRNA and/or protein levels are regulated locally in the stem cell niche during HSC/HPC mobilization.
Acknowledgments We would like to thank Lianne van der Wee-Pals for help in preparing bone sections for immunohistochemistry. This study was financially supported by a grant from Landsteiner Foundation for Blood transfusion Research (Amsterdam, The Netherlands) (L.S.B.R., H.B.K.).
References 1. Silverman GA, Bird PI, Carrell RW, et al. The serpins are an expanding superfamily of structurally similar but functionally diverse proteins. Evolution, mechanism of inhibition, novel functions, and a revised nomenclature. J Biol Chem. 2001;276:33293–33296. 2. Carrell RW, Lomas DA. Alpha1-antitrypsin deficiencyda model for conformational diseases. N Engl J Med. 2002;346:45–53. 3. Winkler IG, Hendy J, Coughlin P, Horvath A, Levesque JP. Serine protease inhibitors serpina1 and serpina3 are down-regulated in bone marrow during hematopoietic progenitor mobilization. J Exp Med. 2005;201:1077–1088. 4. van Pel M, van Os R, Velders GA, et al. Serpina1 is a potent inhibitor of IL-8-induced hematopoietic stem cell mobilization. Proc Natl Acad Sci U S A. 2006;103:1469–1474. 5. Beatty K, Bieth J, Travis J. Kinetics of association of serine proteinases with native and oxidized alpha-1-proteinase inhibitor and alpha-1-antichymotrypsin. J Biol Chem. 1980;255:3931–3934. 6. Barbour KW, Goodwin RL, Guillonneau F, Wang Y, Baumann H, Berger FG. Functional diversification during evolution of the murine alpha(1)-proteinase inhibitor family: role of the hypervariable reactive center loop. Mol Biol Evol. 2002;19:718–727. 7. Pruijt JF, Fibbe WE, Laterveer L, et al. Prevention of interleukin8-induced mobilization of hematopoietic progenitor cells in rhesus monkeys by inhibitory antibodies against the metalloproteinase gelatinase B (MMP-9). Proc Natl Acad Sci U S A. 1999;96:10863–10868.
8. Levesque JP, Takamatsu Y, Nilsson SK, Haylock DN, Simmons PJ. Vascular cell adhesion molecule-1 (CD106) is cleaved by neutrophil proteases in the bone marrow following hematopoietic progenitor cell mobilization by granulocyte colony-stimulating factor. Blood. 2001;98:1289–1297. 9. Levesque JP, Hendy J, Takamatsu Y, Williams B, Winkler IG, Simmons PJ. Mobilization by either cyclophosphamide or granulocyte colony-stimulating factor transforms the bone marrow into a highly proteolytic environment. Exp Hematol. 2002;30:440–449. 10. Wilson A, Trumpp A. Bone-marrow haematopoietic-stem-cell niches. Nat Rev Immunol. 2006;6:93–106. 11. Li Z, Li L. Understanding hematopoietic stem-cell microenvironments. Trends Biochem Sci. 2006;31:589–595. 12. Calvi LM, Adams GB, Weibrecht KW, et al. Osteoblastic cells regulate the haematopoietic stem cell niche. Nature. 2003;425: 841–846. 13. Zhang J, Niu C, Ye L, et al. Identification of the haematopoietic stem cell niche and control of the niche size. Nature. 2003;425:836–841. 14. Mayack SR, Wagers AJ. Osteolineage niche cells initiate hematopoietic stem cell mobilization. Blood. 2008;112:519–531. 15. Nilsson SK, Johnston HM, Whitty GA, et al. Osteopontin, a key component of the hematopoietic stem cell niche and regulator of primitive hematopoietic progenitor cells. Blood. 2005;106:1232–1239. 16. Stier S, Ko Y, Forkert R, et al. Osteopontin is a hematopoietic stem cell niche component that negatively regulates stem cell pool size. J Exp Med. 2005;201:1781–1791. 17. Arai F, Hirao A, Ohmura M, et al. Tie2/angiopoietin-1 signaling regulates hematopoietic stem cell quiescence in the bone marrow niche. Cell. 2004;118:149–161. 18. Stier S, Cheng T, Dombkowski D, Carlesso N, Scadden DT. Notch1 activation increases hematopoietic stem cell self-renewal in vivo and favors lymphoid over myeloid lineage outcome. Blood. 2002;99: 2369–2378. 19. Taichman RS, Hauschka PV. Effects of interleukin-1 beta and tumor necrosis factor-alpha on osteoblastic expression of osteocalcin and mineralized extracellular matrix in vitro. Inflammation. 1992;16: 587–601. 20. Aronow MA, Gerstenfeld LC, Owen TA, Tassinari MS, Stein GS, Lian JB. Factors that promote progressive development of the osteoblast phenotype in cultured fetal rat calvaria cells. J Cell Physiol. 1990; 143:213–221. 21. van der Pluijm G, Sijmons B, Vloedgraven H, Deckers M, Papapoulos S, Lowik C. Monitoring metastatic behavior of human tumor cells in mice with species-specific polymerase chain reaction: elevated expression of angiogenesis and bone resorption stimulators by breast cancer in bone metastases. J Bone Miner Res. 2001;16:1077–1091. 22. Stein GS, Lian JB. Molecular mechanisms mediating proliferation/differentiation interrelationships during progressive development of the osteoblast phenotype. Endocr Rev. 1993;14:424–442. 23. van Bezooijen RL, Roelen BA, Visser A, et al. Sclerostin is an osteocyte-expressed negative regulator of bone formation, but not a classical BMP antagonist. J Exp Med. 2004;199:805–814. 24. Banerjee C, Javed A, Choi JY, et al. Differential regulation of the two principal Runx2/Cbfa1 n-terminal isoforms in response to bone morphogenetic protein-2 during development of the osteoblast phenotype. Endocrinology. 2001;142:4026–4039. 25. Goodwin RL, Barbour KW, Berger FG. Expression of the alpha 1-proteinase inhibitor gene family during evolution of the genus Mus. Mol Biol Evol. 1997;14:420–427. 26. Barbour KW, Wei F, Brannan C, Flotte TR, Baumann H, Berger FG. The murine alpha(1)-proteinase inhibitor gene family: polymorphism, chromosomal location, and structure. Genomics. 2002;80:515–522. 27. Paterson T, Moore S. The expression and characterization of five recombinant murine alpha 1-protease inhibitor proteins. Biochem Biophys Res Commun. 1996;219:64–69.
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647 28. Arai F, Suda T. Maintenance of quiescent hematopoietic stem cells in the osteoblastic niche. Ann N Y Acad Sci. 2007;1106:41–53. 29. Lymperi S, Horwood N, Marley S, Gordon MY, Cope AP, Dazzi F. Strontium can increase some osteoblasts without increasing hematopoietic stem cells. Blood. 2008;111:1173–1181. 30. Boeuf S, Steck E, Pelttari K, et al. Subtractive gene expression profiling of articular cartilage and mesenchymal stem cells: serpins
647
as cartilage-relevant differentiation markers. Osteoarthritis Cartilage. 2008;16:48–60. 31. Nachman RL, Harpel PC. Platelet alpha2-macroglobulin and alpha1antitrypsin. J Biol Chem. 1976;251:4512–4521. 32. Perlmutter DH, Cole FS, Kilbridge P, Rossing TH, Colten HR. Expression of the alpha 1-proteinase inhibitor gene in human monocytes and macrophages. Proc Natl Acad Sci U S A. 1985;82:795–799.
Serpina1 (a1-AT) is synthesized in the osteoblastic stem cell niche H. Bea Kuiperija,*, Melissa van Pela, Karien E. de Rooijb, Rob C. Hoebenc, and Willem E. Fibbea a Department of Immunohematology and Bloodtransfusion, Leiden University Medical Center, Leiden, The Netherlands; bDepartment of Endocrinology and Metabolic Diseases, Leiden University Medical Center, Leiden, The Netherlands; cDepartment of Molecular Cell Biology, Leiden University Medical Center, Leiden, The Netherlands
(Received 28 August 2008; revised 3 February 2009; accepted 4 February 2009)
Objective. Previously, we identified Serpina1 as a potent inhibitor of hematopoietic stem and progenitor cell (HSC/HPC) mobilization. Serpina1 protein is found in the bone marrow (BM) extracellular fluid and concentrations are decreased during granulocyte colony-stimulating factor–induced HSC/HPC mobilization in mice. In addition, administration of exogenous Serpina1 protein inhibits HSC/HPC mobilization. BM cells responsible for production and secretion of Serpina1 remain unknown. Here, we examined the expression of Serpina1 in order to identify cell populations of the BM that synthesize Serpina1. Materials and Methods. Osteoblast (OB) and hematopoietic BM cell fractions were isolated from femurs, tibias, and humeri obtained from untreated mice. Subsequently, each BM fraction was examined for the production of Serpina1 messenger RNA and protein by quantitative real-time polymerase chain reaction, Western blotting, and immunohistochemistry. Results. Quantitative real-time polymerase chain reaction analysis showed that Serpina1 messenger RNA is produced at high levels by OB compared to hematopoietic BM cells. Furthermore, Western blot analysis indicated that Serpina1 protein was secreted by OB. In contrast, no Serpina1 protein could be detected in the supernatant obtained from overnight cultured hematopoietic BM cells. Finally, in BM sections obtained from the femurs of untreated mice, Serpina1 protein was detected in OB cells lining the bone. Conclusion. Serpina1 protein in the BM extracellular fluid is predominantly produced by OB. This indicates that Serpina1 may play a regulatory role in the maintenance of HSC in the OB stem cell niche. Ó 2009 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc.
Serpina1 (also known as a1-antitrypsin or a1-proteinase inhibitor) belongs to a superfamily of serine protease inhibitors with highly conserved secondary structural elements. They inhibit their target proteases by covalent binding, and this inhibition is irreversible and required for effective control of the proteolytic cascade that may be initiated by the release of just a few protease molecules. In mammals, serpins play a role in a range of proteolytic processes, including blood clotting, inflammation, and turnover of extracellular matrix [1,2].
Offprint requests to: Melissa van Pel, Ph.D., Section of Stem Cell Biology, Department of Immunohematology and Bloodtransfusion, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, The Netherlands; E-mail: [email protected] *Current address: Department of Neurology and Donders Centre for Neuroscience, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.
Serpina1 is, together with a2-macroglobulin, the major plasma protease inhibitor. Besides being present in plasma, Serpina1 is also found in the bone marrow (BM) extracellular fluid [3,4]. Although the main substrate of Serpina1 is neutrophil elastase, also various other proteases, including trypsin and chymotrypsin, can be inhibited [5]. The specificity of the protease inhibitor is determined by its so-called reactive center loop, a domain of approximately 20 residues that extends from the body of the Serpina1 polypeptide. Binding of the protease to Serpina1 is directed via this reactive center loop, which acts as a substrate for the protease. The protease cleaves the reactive center loop, resulting in a conformational change, followed by crushing and subsequent destruction of the protease [1,6]. Proteases have shown to be regulatory mediators of cytokine-induced hematopoietic stem and progenitor cell (HSC/HPC) mobilization. Studies from our laboratory and
0301-472X/09 $–see front matter. Copyright Ó 2009 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc. doi: 10.1016/j.exphem.2009.02.004
642
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647
others indicated a critical role for the matrix metalloproteinase-9 (gelatinase B), neutrophil elastase, and cathepsin G in cytokine- and cyclophosphamide-induced HSC/HPC mobilization [7–9]. Recently, we reported that Serpina1 is a potent inhibitor of HSC/HPC mobilization [3,4]. We showed that administration of Serpina1 prior to interleukin-8 (IL-8) injection inhibited HSC/HPC mobilization. Furthermore, low-dose total body irradiation increased Serpina1 levels in the BM extracellular fluid, and inhibited both IL-8 and granulocyte colony-stimulating factor– induced HSC mobilization [4]. HSC are localized in the BM in a specialized microenvironment called the hematopoietic stem cell niche, which is located at the endosteal surface of the bone. Herein, primitive HSC are in close contact with stromal cells that form the hematopoietic microenvironment of the BM. The stromal tissue of the marrow creates a three-dimensional network comprised of cells (including endothelial cells, fibroblasts, and osteoblasts [OB]), extracellular matrix components, and hematopoietic growth factors that collectively constitute a unique microenvironment. It is believed that the stem cell niche directs regulation of self-renewal and differentiation of primitive HSC. Cytokines play a major role in this process and act both via cell-to-cell contact, or as soluble factors [10,11]. OBs are important regulators of the HSC niche [12–14]. By secreting osteopontin the number of HSC is controlled [15,16]. Angiopoietin-1 (Ang-1), which is secreted by OB, binds Tie2 and regulates HSC activity on the level of quiescence and repopulating ability as well as on the level of interaction of HSC with the niche [17]. Furthermore, OB produce Jagged 1, which regulates HSC maintenance and differentiation via Notch signaling [12,18]. Here we show that the serine protease inhibitor Serpina1 is predominantly expressed by OB. We hypothesize that Serpina1 is involved in HSC maintenance in the OB niche by inhibiting proteases, including neutrophil elastase.
Materials and methods Mice Eight- to 12-week-old male Balb/c mice were purchased from Charles River Laboratories (Maastricht, The Netherlands). The animals were fed commercial rodent food and acidified water ad libitum. The Institutional Ethical Committee on Animal Experiments approved all experimental protocols. Isolation of BM and OB-enriched cell fractions Femurs, tibias, and humeri were isolated from Balb/c mice and the bones were flushed with ice-cold phosphate-buffered saline (PBS) to harvest hematopoietic BM cells. Thereafter, the flushed bones were crushed using scissors and incubated at 37 C alternating with 1 mg/mL collagenase I (Sigma, Zwijndrecht, The Netherlands) plus 0.25% trypsin in PBS and 4 mM ethylene diamine tetraacetic acid in PBS. After each incubation step, cells were harvested and numbered consecutively OB fraction 1 through 5. After the last
incubation step, bone fragments were treated with Trizol reagent (Invitrogen, Breda, The Netherlands) to obtain OB fraction 6. The cells in BM and OB fraction 1 to 5 cells were either lysed in Trizol reagent for RNA isolation or cultured. Previous studies have shown that these cells have osteoblastic characteristics and have been extensively used to map OB maturation [19,20]. Analysis of Serpina1 secretion by Western blotting To analyze Serpina1 protein secretion, 3 106 hematopoietic BM cells and approximately 0.8 106 OB fraction 5 cells were seeded in a well of a six-well plate containing a–minimum essential medium supplemented with 10% fetal bovine serum (Gibco, Grand Island, NY, USA), penicillin, streptomycin, and L-glutamine (all from Gibco). After 24 hours of incubation, cells were washed twice with PBS and grown in 1 mL a–minimum essential medium without fetal bovine serum for 24 hours. The serum-free culture supernatant was harvested and cleared for remaining cells by centrifugation. Subsequently, proteins from 15 mL of these culture supernatants were separated on sodium dodecyl sulfate polyacrylamide gels and blotted onto polyvinylidene difluoride membranes (Roche, Almere, The Netherlands). Membranes were blocked using 1% blocking reagent (Roche) in Tris-buffered saline, followed by incubations with a primary antibody directed against mouse Serpina1 (previously obtained by immunization of rabbits [4]) and a horseradish peroxidase–conjugated goat anti-rabbit Ig secondary antibody (Promega, Leiden, The Netherlands). Proteins were visualized using standard enhanced chemiluminescence (Roche) and autoradiography. Alkaline phosphatase activity assay To assess alkaline phosphatase (ALP) activity, OB fraction 5 cells were cultured up to passage 3 in a–minimum essential medium supplemented with 10% fetal bovine serum, pen-strep and L-glutamine. Cells were fixed in methanol, washed with PBS and incubated with ALP substrate (0.02% Naphthol AS-MX salt; Sigma, N5000), 0.06% Fast Blue RR salt (Sigma, F0500), 0.1 M Tris-HCl [pH 8.8], 0.01% MgSO4) for 5 minutes and washed twice with PBS. ALP staining was analyzed by inverted light microscopy. Quantitative reverse transcription polymerase chain reaction (qPCR) RNA was isolated from hematopoietic BM and OB fraction 1 to 6 Trizol lysates and subsequently complementary DNA was generated. Primer sets specific for Serpina1, Osteocalcin, Runx2 splice variant (MASN), Sclerostin, PECAM, N-cadherin (Ncad), glyceraldehyde phosphate dehydrogenase, b2M (Table 1), and Cathepsin K (QuantiTect primer assay set, Qiagen, Venlo, The Netherlands) were used for qPCR experiments. The qPCR was performed using SYBR Green qPCR Mastermix (Eurogentec, Maastricht, The Netherlands) and the iQ qPCR analyzer (Bio-Rad, Veenendaal, The Netherlands), according to the manufacturers instructions. Relative gene expression was calculated using the comparative threshold cycle (CT) method, with Gapdh and b2M as the endogenous reference genes. Immunohistochemistry on bone sections To identify Serpina1 in situ, femurs of mice were fixed in 10% formalin in PBS, decalcified in 10% ethylene diamine tetraacetic acid, embedded in paraffin and sectioned (5 mm). Immunohistochemistry was performed essentially as described previously [21], with some modifications. To block nonspecific activity,
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647
643
Table 1. Quantitative reverse transcription polymerase chain reaction primer sets Gene Serpina1 OC MASN SOST PECAM Ncad GAPDH b2M
Forward primer
Reverse primer
GATGGGAAGATGCAGCATC ACAGACAAGTCCCACACAGCAGC AAGTGCGGTGCAAACTTTCT TCCTCCTGAGAACAACCAGAC ACGAGGTACAGTCTGGAAGTG GAAGGCAATCCCACTTATGG ATGGCCTTCCGTGTTCCTAC CCTGTATGCTATCCAGAAAACC
TCCAGAGATGGACAGTCTG TGAAGGCTTTGTCAGACTCAGGGC TCTCGGTGGCTGGTAGTGA TGTCAGGAAGCGGGTGTAGTG TTGTTTGTTGTGCCAAAACG CTTCTCCGTAGAAAGTCATG CCTGCTTCACCACCTTCTTG GTTCAGTATGTTCGGCTTCC
sections were pretreated with hydrogen peroxide. Antigen retrieval was performed using 5 mg/mL hyaluronidase in 0.5% Boehringer milk protein (Boehringer, Mannheim, Germany) in PBS (pH 5). Sections were blocked using 5% normal goat serum and 0.5% Boehringer milk protein in Tris-buffered saline-Tween, followed by incubation with the primary antibody directed against mouse Serpina1 and a biotinylated goat anti-rabbit secondary antibody (Dako, Heverlee, Belgium). For detection of Serpina1, sections were incubated with streptavidin–horseradish peroxidase (Dako). The signal was enhanced using biotinyl tyramide, followed by a second incubation step with streptavidin–horseradish peroxidase. Serpina1 was visualized using 3,30 -diaminobenzidine as a substrate for horseradish peroxidase, and hematoxylin was used as counterstaining. To determine the specificity of the staining, the antibody directed against mouse Serpina1 protein was pre-incubated with 25 mg purified mouse Serpina1 protein [4] for 30 minutes. Subsequently, bone sections were incubated with the blocked antibody, followed by a staining procedure as described here.
and hematopoietic BM (Fig. 1A). Similarly, MASN levels were significantly increased in fraction 5 and 6, compared to fraction 1 to 4 and hematopoietic BM (Fig. 1A). Sclerostin mRNA levels were strongly induced in OB fraction 6, compared to other OB fractions, but could not be detected in hematopoietic BM cells (data not shown). As a control,
Analysis of Serpina1 isoforms For the analysis of the different isoforms of Serpina1 in murine tissues, Serpina1 mRNA fragments were amplified by PCR using either the Serpina1 forward and reverse primers for qPCR (Table 1), or the qPCR Serpina1 forward primer and the 30 -Serpina1 reverse primer (CAATTCAGAAGGAAGGATG). The product of respectively 117 and 454 bp was subcloned into the pCRIITOPO vector (Invitrogen), and sequences of several clones were analyzed using either the M13-reverse sequence primer or the 30 -Serpina1 primer in the sequence reactions.
Results Isolation of OB-enriched cell fractions from femurs, tibia and humeri To identify the cell populations in BM that produce/express Serpina1, hematopoietic BM cells and OB cell fractions were isolated from murine femurs, tibias, and humeri. To confirm the presence of OB in each fraction, mRNA levels of the OB markers osteocalcin, the Runx2 splice variant MASN, and Sclerostin were measured [22–24]. The results of OB fraction 1 to 4 were taken together, since no differences between these fractions were observed. Osteocalcin messenger RNA (mRNA) levels were significantly increased in OB fraction 6, compared to fraction 1 to 5
Figure 1. Serpina1 messenger RNA (mRNA) levels in primary osteoblast (OB) cell fractions. (A) mRNA levels of several OB markers were quantified by quantitative reverse transcription polymerase chain reaction. Relative mRNA expression in OB cell fraction 1-4, 5 and 6, compared to bone marrow (BM) is plotted. (B) The relative mRNA expression of Serpina1 is shown in OB cell fraction 1 to 6, compared to mRNA levels in BM. Data represent the mean fold induction 6 standard deviation. *p ! 0.05, **p ! 0.01 vs BM (A) or vs previous fraction (B).
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647
644
the mRNA level of the endothelial marker PECAM was measured. PECAM mRNA levels remained unaltered in all the OB cell fractions, and were comparable to the levels in hematopoietic BM cells (Fig. 1A). Interestingly, Ncad mRNA levels were also significantly increased in OB fraction 5 and 6, compared to hematopoietic BM cells and OB fraction 1–4 (Fig. 1A). To investigate a possible contamination of osteoclasts in the OB fractions, Cathepsin K mRNA levels were analyzed. Compared to total BM cells, 15-fold lower levels of Cathepsin K were detected in OB fractions 1-5 and a 22.9-fold reduction in OB fraction 6 (data not shown). Together, these data indicate that during isolation of OB fractions, the differentiation level increased compared to the preceding fraction. Serpina1 mRNA is highly expressed in OB-enriched cell fractions To identify the cell populations that synthesize Serpina1, Serpina1 mRNA levels were determined in the OBenriched cell fractions, and compared to Serpina1 mRNA levels in hematopoietic BM cells. Serpina1 mRNA levels were up to 6 104 times increased in OB-enriched cell fractions (range, 1 102 to 6 104 in fractions 1 to fraction 6, respectively) compared to hematopoietic BM cells (Fig. 1B). Furthermore, in fraction 5, the relative Serpina1 mRNA level was significantly increased compared to fraction 1–4, and this increase was even more pronounced in fraction 6. These data indicate that Serpina1 mRNA levels are significantly increased in OB fraction, compared to hematopoietic BM cell fraction. In addition, upon maturation of OBs, the levels of Serpina1 increase accordingly.
previously published results of other groups, the expression of three isoforms in the liver (Serpina1a, Serpina1b, and DOM6) was confirmed (Table 2) [26]. Interestingly, in hematopoietic BM cells, only the Serpina1b isoform was identified, whereas in OB fraction 5/6 cells, a mixture of isoforms was detected (Table 2). Besides the isoforms present in the liver, this mixture included a fourth isoform, DOM7, which has not been identified in Balb/c mice previously. Of note, due to significant sequence overlap, part of the sequences could only be determined as being either Serpina1a or DOM6. Serpina1 protein is produced by cultured OB To further assess whether osteoblasts secrete Serpina1 protein, we cultured OB fraction 5 cells and hematopoietic BM cells in serum-free medium for 24 hours. The cultured fraction 5 cells phenotypically resembled OB. Staining for ALP activity indicated that about 30% of cultured cells showed a clear purple stain after incubation with the chromogenic conversion substrate for ALP, indicating the presence of OB (Fig. 2A). Next, Serpina1 protein levels were determined in the culture supernatants of the OB fraction 5 and hematopoietic BM cells by Western blot analysis. Serpina1 protein was
Differential expression of Serpina1 isoform mRNA in murine tissues To date, seven isoforms of Serpina1 have been identified in the mouse [6,25]. Three of these isoforms are expressed in the liver of Balb/c mice: DOM1 (Serpina1a), DOM2 (Serpina1b), and DOM6 [26]. Because Serpina1 isoforms exhibit different substrate specificities [6,27], we determined the isoform identity of at least 10 clones in liver, BM, and in OB fraction 5/6 cells (Table 2). In line with Table 2. Expression of serpina1 isoforms in murine tissues Tissue Isoforma
Liver
Bone marrow
Osteoblast fraction 5/6
Serpina1a Serpina1b DOM6 DOM7 Serpina1a or DOM6b
2/19 6/19 5/19 0/19 6/19
0/10 10/10 0/10 0/10 0/10
ND 13/24 ND 2/24 9/24
ND 5 not determined. Expressed as number of cases out of total. b Cases for which the sequence could not be discriminated for Serpina1a or DOM6. a
Figure 2. Cultured primary osteoblasts (OB) produce Serpina1 protein. (A) Cultured OB fraction 5 cells were stained for alkaline phosphatase (ALP) activity. ALP activity is shown in purple. (B) Culture supernatant obtained from overnight-cultured hematopoietic bone marrow (BM) cells and OB fraction 5 cells were processed for Western blot analysis. Serpina1 protein was visualized by immunoblotting and standard enhanced chemiluminescence.
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647
found in the culture supernatant obtained from the culture of OB-enriched fraction 5 cells, whereas no Serpina1 protein could be detected in the culture supernatant obtained from the culture of hematopoietic BM cells (Fig. 2B). This indicates that the mRNA levels in OB correlate with Serpina1 protein production and secretion. Serpina1 protein is produced in bone-lining OB To investigate whether the Serpina1 protein is also expressed in vivo by OBs, we stained bone sections obtained from femurs of untreated mice with anti-murine Serpina1 antibodies. Serpina1 was observed in OBs lining the bone, but also in the more differentiated osteocytes embedded in the bone matrix (Fig. 3A). Because there was also Serpina1 staining in and between hematopoietic BM cells, we tested the specificity of the Serpina1 antibody. Hereto, the primary antibody was incubated with mouse Serpina1 protein prior to the staining of the femur sections. This resulted in specific blocking of the Serpina1 signal in
Figure 3. Serpina1 protein is produced by bone-lining osteoblasts (OB). (A) Immunohistochemistry of Serpina1 in bone sections. Serpina1 protein is shown in brown, whereas nuclei are stained purple by hematoxylin counterstaining. Scale bar: 80 mm; the insert has 4 extra magnification. (B) Immunohistochemistry on bone sections as in (A), except that here the primary Serpina1 antibody was incubated with mouse Serpina1 protein before use.
645
both OB and hematopoietic BM (Fig. 3B) indicating that a specific Serpina1 signal was obtained. Taken together, these results indicate that Serpina1 protein is produced by bone-lining osteoblasts and osteocytes.
Discussion In the present article, we show that Serpina1 is synthesized and secreted by OB lining the inner surface of the bone cavity (bone-lining OB), which are the cells that are important for maintenance of HSC [12,13]. This implies that Serpina1 is produced in the OB stem cell niche. The OB stem cell niche serves as a specific microenvironment for maintenance of primitive HSC via both adhesive and soluble interactions between OB and HSC, including Ncad [28]. Ncad is the only member of the cadherin family that is both present on HSC and OB, and Ncad expression has been reported to support the interaction of OB with HSC, and keep HSC quiescent [13]. Recently, Lymperi et al. reported that an increase in OB number, without increased Ncad expression levels, did not result in increased HSC numbers [29]. They speculated that for increased HSC numbers, an increase in special Ncadpositive OB is needed. Our data clearly show that an increase in Ncad expression levels correlates with increased differentiation status of OB and with increased Serpina1 expression. This suggests correlation between Ncad expression levels, Serpina1 expression levels, and stem cell maintenance. Serpina1 is predominantly synthesized in the liver, where the protein is delivered to the bloodstream. Its main function is protecting the lung from proteolytic damage by inhibiting neutrophil elastase. Besides liver cells, macrophages, monocytes, platelets, and recently chondrocytes have been reported to produce Serpina1 protein [30–32]. Production of Serpina1 by chrondrocytes strengthens our results, because both chondrocytes and OB differentiate from mesenchymal stem cells. However, while Serpina1 produced by OB likely protects against the proteolytic action of enzymes in the BM, the function of Serpina1 expression in chondrocytes remains elusive. In the mouse, two classical and five nonclassical Serpina1 isoforms are expressed. The classical isoforms, Serpina1a and Serpina1b, can inhibit neutrophil elastase. For this function, it is important that the so-called p1 residue is a methionine. In the nonclassical isoforms of Serpina1 (Serpina1c-e, DOM6, and DOM7) the p1 residue is a different amino acid, and these isoforms have specificity for other proteases than neutrophil elastase [6]. Interestingly, we detected DOM7 mRNA in Balb/c OB cell fractions. The presence of this isoform in Balb/c mice has not been reported before, and this may be due to the fact that others have focused on Serpina1 isoform expression in liver tissue. We observed that the expression of Serpina1 isoforms in OB, besides DOM7 expression, resembled those of liver
646
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647
cells, containing a mix of classical and nonclassical isoforms. Interestingly, hematopoietic BM cells did synthesize Serpina1 mRNA molecules up to several log scales less than did OB, but they purely produced the classical isoform Serpina1b. More research has to be done to determine the consequence of these results. Proteases, including neutrophil elastase, Cathepsin G, and matrix metalloproteinase–9, play an important role in HSC mobilization [7–9]. We and others have hypothesized that cytokine-induced HSC/HPC mobilization is determined by a critical balance between serine proteases and serine protease inhibitors in the BM extracellular fluid [3,4]. Indeed, Serpina1 protein levels in the BM space are inversely correlated with mobilization of HSC ([3]; unpublished results [H.B.K., M.vP.]). This suggests that under steady-state conditions, Serpina1 may be involved in retention of HSC in the niche, and thus may play an important role in the protection of HSC from mobilizing proteases. Moreover, we hypothesize a possible role for Serpina1 in regulating the exit of HSC from the stem cell niche. Therefore, for further research it will be interesting to see whether Serpina1 mRNA and/or protein levels are regulated locally in the stem cell niche during HSC/HPC mobilization.
Acknowledgments We would like to thank Lianne van der Wee-Pals for help in preparing bone sections for immunohistochemistry. This study was financially supported by a grant from Landsteiner Foundation for Blood transfusion Research (Amsterdam, The Netherlands) (L.S.B.R., H.B.K.).
References 1. Silverman GA, Bird PI, Carrell RW, et al. The serpins are an expanding superfamily of structurally similar but functionally diverse proteins. Evolution, mechanism of inhibition, novel functions, and a revised nomenclature. J Biol Chem. 2001;276:33293–33296. 2. Carrell RW, Lomas DA. Alpha1-antitrypsin deficiencyda model for conformational diseases. N Engl J Med. 2002;346:45–53. 3. Winkler IG, Hendy J, Coughlin P, Horvath A, Levesque JP. Serine protease inhibitors serpina1 and serpina3 are down-regulated in bone marrow during hematopoietic progenitor mobilization. J Exp Med. 2005;201:1077–1088. 4. van Pel M, van Os R, Velders GA, et al. Serpina1 is a potent inhibitor of IL-8-induced hematopoietic stem cell mobilization. Proc Natl Acad Sci U S A. 2006;103:1469–1474. 5. Beatty K, Bieth J, Travis J. Kinetics of association of serine proteinases with native and oxidized alpha-1-proteinase inhibitor and alpha-1-antichymotrypsin. J Biol Chem. 1980;255:3931–3934. 6. Barbour KW, Goodwin RL, Guillonneau F, Wang Y, Baumann H, Berger FG. Functional diversification during evolution of the murine alpha(1)-proteinase inhibitor family: role of the hypervariable reactive center loop. Mol Biol Evol. 2002;19:718–727. 7. Pruijt JF, Fibbe WE, Laterveer L, et al. Prevention of interleukin8-induced mobilization of hematopoietic progenitor cells in rhesus monkeys by inhibitory antibodies against the metalloproteinase gelatinase B (MMP-9). Proc Natl Acad Sci U S A. 1999;96:10863–10868.
8. Levesque JP, Takamatsu Y, Nilsson SK, Haylock DN, Simmons PJ. Vascular cell adhesion molecule-1 (CD106) is cleaved by neutrophil proteases in the bone marrow following hematopoietic progenitor cell mobilization by granulocyte colony-stimulating factor. Blood. 2001;98:1289–1297. 9. Levesque JP, Hendy J, Takamatsu Y, Williams B, Winkler IG, Simmons PJ. Mobilization by either cyclophosphamide or granulocyte colony-stimulating factor transforms the bone marrow into a highly proteolytic environment. Exp Hematol. 2002;30:440–449. 10. Wilson A, Trumpp A. Bone-marrow haematopoietic-stem-cell niches. Nat Rev Immunol. 2006;6:93–106. 11. Li Z, Li L. Understanding hematopoietic stem-cell microenvironments. Trends Biochem Sci. 2006;31:589–595. 12. Calvi LM, Adams GB, Weibrecht KW, et al. Osteoblastic cells regulate the haematopoietic stem cell niche. Nature. 2003;425: 841–846. 13. Zhang J, Niu C, Ye L, et al. Identification of the haematopoietic stem cell niche and control of the niche size. Nature. 2003;425:836–841. 14. Mayack SR, Wagers AJ. Osteolineage niche cells initiate hematopoietic stem cell mobilization. Blood. 2008;112:519–531. 15. Nilsson SK, Johnston HM, Whitty GA, et al. Osteopontin, a key component of the hematopoietic stem cell niche and regulator of primitive hematopoietic progenitor cells. Blood. 2005;106:1232–1239. 16. Stier S, Ko Y, Forkert R, et al. Osteopontin is a hematopoietic stem cell niche component that negatively regulates stem cell pool size. J Exp Med. 2005;201:1781–1791. 17. Arai F, Hirao A, Ohmura M, et al. Tie2/angiopoietin-1 signaling regulates hematopoietic stem cell quiescence in the bone marrow niche. Cell. 2004;118:149–161. 18. Stier S, Cheng T, Dombkowski D, Carlesso N, Scadden DT. Notch1 activation increases hematopoietic stem cell self-renewal in vivo and favors lymphoid over myeloid lineage outcome. Blood. 2002;99: 2369–2378. 19. Taichman RS, Hauschka PV. Effects of interleukin-1 beta and tumor necrosis factor-alpha on osteoblastic expression of osteocalcin and mineralized extracellular matrix in vitro. Inflammation. 1992;16: 587–601. 20. Aronow MA, Gerstenfeld LC, Owen TA, Tassinari MS, Stein GS, Lian JB. Factors that promote progressive development of the osteoblast phenotype in cultured fetal rat calvaria cells. J Cell Physiol. 1990; 143:213–221. 21. van der Pluijm G, Sijmons B, Vloedgraven H, Deckers M, Papapoulos S, Lowik C. Monitoring metastatic behavior of human tumor cells in mice with species-specific polymerase chain reaction: elevated expression of angiogenesis and bone resorption stimulators by breast cancer in bone metastases. J Bone Miner Res. 2001;16:1077–1091. 22. Stein GS, Lian JB. Molecular mechanisms mediating proliferation/differentiation interrelationships during progressive development of the osteoblast phenotype. Endocr Rev. 1993;14:424–442. 23. van Bezooijen RL, Roelen BA, Visser A, et al. Sclerostin is an osteocyte-expressed negative regulator of bone formation, but not a classical BMP antagonist. J Exp Med. 2004;199:805–814. 24. Banerjee C, Javed A, Choi JY, et al. Differential regulation of the two principal Runx2/Cbfa1 n-terminal isoforms in response to bone morphogenetic protein-2 during development of the osteoblast phenotype. Endocrinology. 2001;142:4026–4039. 25. Goodwin RL, Barbour KW, Berger FG. Expression of the alpha 1-proteinase inhibitor gene family during evolution of the genus Mus. Mol Biol Evol. 1997;14:420–427. 26. Barbour KW, Wei F, Brannan C, Flotte TR, Baumann H, Berger FG. The murine alpha(1)-proteinase inhibitor gene family: polymorphism, chromosomal location, and structure. Genomics. 2002;80:515–522. 27. Paterson T, Moore S. The expression and characterization of five recombinant murine alpha 1-protease inhibitor proteins. Biochem Biophys Res Commun. 1996;219:64–69.
H.B. Kuiperij et al./ Experimental Hematology 2009;37:641–647 28. Arai F, Suda T. Maintenance of quiescent hematopoietic stem cells in the osteoblastic niche. Ann N Y Acad Sci. 2007;1106:41–53. 29. Lymperi S, Horwood N, Marley S, Gordon MY, Cope AP, Dazzi F. Strontium can increase some osteoblasts without increasing hematopoietic stem cells. Blood. 2008;111:1173–1181. 30. Boeuf S, Steck E, Pelttari K, et al. Subtractive gene expression profiling of articular cartilage and mesenchymal stem cells: serpins
647
as cartilage-relevant differentiation markers. Osteoarthritis Cartilage. 2008;16:48–60. 31. Nachman RL, Harpel PC. Platelet alpha2-macroglobulin and alpha1antitrypsin. J Biol Chem. 1976;251:4512–4521. 32. Perlmutter DH, Cole FS, Kilbridge P, Rossing TH, Colten HR. Expression of the alpha 1-proteinase inhibitor gene in human monocytes and macrophages. Proc Natl Acad Sci U S A. 1985;82:795–799.