Marrow Stromal Cells Expression in Bone Tac1 System Through

Stromal Derived Growth Factor-1α: Another Mediator in Neural-Emerging Immune System through Tac1 Expression in Bone Marrow Stromal Cells This information is current as of September 28, 2021. Kelly E. Corcoran, Nitixa Patel and Pranela Rameshwar J Immunol 2007; 178:2075-2082; ; doi: 10.4049/jimmunol.178.4.2075 http://www.jimmunol.org/content/178/4/2075 Downloaded from

References This article cites 31 articles, 7 of which you can access for free at: http://www.jimmunol.org/content/178/4/2075.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 28, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2007 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Stromal Derived Growth Factor-1␣: Another Mediator in Neural-Emerging Immune System through Tac1 Expression in Bone Marrow Stromal Cells1

Kelly E. Corcoran,* Nitixa Patel,* and Pranela Rameshwar2†

Stromal cell-derived growth factor-1␣ (SDF-1␣) is a member of the CXC chemokines and interacts with the G protein, seven- transmembrane CXCR4 receptor. SDF-1␣ acts as a chemoattractant for immune and hemopoietic cells. The Tac1 gene encodes peptides belonging to the tachykinin family with substance P being the predominant member. Both SDF-1␣ and Tac1 peptides are relevant hemopoietic regulators. This study investigated the effects of SDF-1␣ on Tac1 expression in the major hemopoietic supporting cells, the bone marrow stroma, and addresses the consequence to hemopoiesis. Reporter gene assays with the 5؅ flanking region of Tac1 showed a bell-shaped effect of SDF-1␣ on luciferase activity with 20 ng/ml SDF-1␣ acting as stimulator, Downloaded from whereas 50 and 100 ng/ml SDF-1␣ acted as inhibitors. Gel shift assays and transfection with wild-type and mutant I␬B indicate NF-␬B as a mediator in the repressive effects at 50 and 100 ng/ml SDF-1␣. Northern analyses and ELISA showed correlations among reporter gene activities, mRNA (␤-preprotachykinin I), and protein levels for substance P. Of relevance is the novel finding by long-term culture-initiating cell assays that showed an indirect effect of SDF-1␣ on hemopoiesis through substance P production. The results also showed neurokinin 1 and not neurokinin 2 as the relevant receptor. Another crucial finding is that substance ␣ ␣

P does not regulate the production of SDF-1 in stroma. The studies indicate that SDF-1 levels above baseline production in bone http://www.jimmunol.org/ marrow stroma induce the production of substance P to stimulate hemopoiesis. Substance P, however, does not act as autocrine stimulator to induce the production of SDF-1␣. This study adds SDF-1␣ as a mediator within the neural-immune-hemopoietic axis. The Journal of Immunology, 2007, 178: 2075–2082.

ac1 (also referred to as preprotachykinin I or PPT-I)3 is a NK1 and NK2 mediate stimulatory and inhibitory effects on single copy gene that is conserved by evolution (1). Tac1 hemopoiesis, respectively (5). NK2 is constitutively expressed in T is ubiquitously expressed, although its functions are organ unstimulated BM stroma, whereas NK1 expression requires induc- specific (2). Tac1 peptides exhibit neurotransmission function in tion by cytokines with stimulatory effects on hemopoiesis and by the nervous system, hemopoietic regulation in bone marrow (BM), broad-acting cytokines such as IL-1 (7). The expression of NK1 by guest on September 28, 2021 and immunomodulatory properties in the immune system (3). Tac1 correlates with down-regulation of NK2 on normal hemopoietic produces several peptides that belong to the tachykinin family with cells and BM stroma (5). Although the mechanism involved in this substance P (SP) and neurokinin-A as its major gene products (4). yin-yang type of expression between NK1 and NK2 is unclear, the SP and neurokinin-A interact with different binding affinities to experimental evidence suggests mechanisms involving intracellu- three related seven-transmembrane G protein-coupled receptors: lar crosstalk (5). The regulated expressions of NK1 and NK2 in the neurokinin (NK) 1, NK2, and NK3 (5, 6). BM stroma affect hemopoiesis, mainly because one receptor sub- type appears to modulate the functions of the other (1, 4). Similar to the tachykinins, the chemokine stromal-derived growth factor (SDF)-1␣ is also involved in hemopoiesis (8). The ␣ *Graduate School of Biomedical Sciences, University of Medicine and Dentistry of production of SDF-1 in BM stromal cells is membrane-bound New Jersey, Newark, NJ 07107; and †Department of Medicine, University of Med- and available for interaction with the CXCR4-expressing hemo- icine and Dentistry of New Jersey, New Jersey Medical School, Newark, NJ 07103 poietic stem cells (HSCs) (9, 10). SDF-1␣ does not appear to act Received for publication July 31, 2006. Accepted for publication November 14, 2006. as a hormone because it is localized at the source of production. The costs of publication of this article were defrayed in part by the payment of page Specifically, SDF-1␣ is retained within the BM niche as membrane charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. bound or is attached to the surrounding proteoglycans of extracel- 1 This work was supported by grants from the Department of Defense and the Uni- lular matrix proteins (11). versity Hospital Cancer Center, University of Medicine and Dentistry of New Jersey, The levels of SDF-1␣ follow a gradient pattern across the BM New Jersey Medical School. This work was performed in partial fulfillment for a cavity (12). SDF-1␣ is important for the retention of HSCs within Ph.D. thesis by K.E.C. and was done at the Department of Medicine, Division of Hematology/Oncology, University of Medicine and Dentistry of New Jersey, New the stromal compartment under homeostatic conditions. However, Jersey Medical School, Newark, NJ 07103. changes in this gradient facilitate HSC mobilization of the BM; 2 Address correspondence and reprint requests to Dr. Pranela Rameshwar, University into the peripheral circulation (9). Stromal-derived SDF-1␣ is im- of Medicine and Dentistry of New Jersey, New Jersey Medical School, Medical portant for the quiescent state of HSC close to the BM niche, Science Building, Room E-579, 185 South Orange Avenue, Newark, NJ 07103. ␣ E-mail address: [email protected] whereas changes in the levels of SDF-1 have been shown to 3 Abbreviations used in this paper: PPT-I, preprotachykinin I; Alk Phos, alkaline stimulate hemopoiesis (8, 13). phosphatase; BM, bone marrow; ␤-gal, ␤-galactosidase; HSC, hemopoietic stem cell; In mice, SDF-1␣ has been shown to mobilize HSCs into the LTC-IC, long-term culture-initiating cell; NK, neurokinin; SDF, stromal-derived peripheral circulation (14). Similarly, antagonists to the SDF-1␣ growth factor; SP, substance P. receptor CXCR4 have also been shown to mobilize HSCs, sug- Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00 gesting that the concentration of SDF-1␣ determines the fate of www.jimmunol.org 2076 SDF-1␣-Tac1 IN HEMOPOIESIS

HSC (15). Together, these findings suggest that at high levels of gene-specific insert of PPT-I/1.2 is equivalent to 1.2 kb and includes intron SDF-1␣ the interaction between stromal cells and HSCs could be 1, exon 1, and upstream sequences. Exon 1 and intron 1 are omitted in broken, in part due to a localized imbalance in SDF-1␣ levels. PPT-I/N0. The pSEAP vector system, which consists of three vectors ␣ (pSEAP-Basic, pSEAP-Enhancer, and pSEAP-Promoter) was purchased Although a large increase in SDF-1 levels could mobilize the from BD Clontech. The exon 1 construct of Tac1 was inserted into pSEAP- HSCs from the BM cavity to the periphery, it is unclear how a Basic, pSEAP-Enhancer, and pSEAP-Promoter in the sense orientation us- small increase around the stromal cell niche could affect hemopoi- ing primers linked to sequences for HindIII and EcoR1. Exon 1 was placed esis. This study addresses the functional relationship between in the antisense orientation into pSEAP-Promoter by amplifying the insert ␣ with primers containing XhoI and KpnI linkers. Insert direction was con- SDF-1 and the Tac1 gene in BM stroma and the consequence to firmed by a restriction digest and sequencing. pCMV-I␬B␣ (wild type) hemopoiesis. We show that a modest increase in SDF-1␣ levels allows for activation of NF-␬B, whereas pCMV- I␬BM (mutant) sequesters stimulates stromal cells to induce the hemopoietic regulator gene NF–␬B in the cytosol by preventing its phosphorylation. Both vectors are Tac1, leading to the production of SP. The latter stimulates hemo- part of the Mercury I␬B dominant-negative vector set (BD Clontech). poiesis, specifically through NK1 receptor. The relevance of these Transient transfection and reporter gene assay studies to neural-hemopoietic regulation is discussed. BM stromal cells were cotransfected with plasmid ␤-galactosidase (␤-gal) Materials and Methods and PPT-I/1.2 or PPT-I/N0 using the Effectene reagent (Qiagen). At 16 h, the transfectants were stimulated with SDF-1␣ at 20, 50, or 100 ng/ml in Reagents serum-free medium. After 16 h, the cells were scraped in 30 ␮l of lysis ␣-MEM, IMDM, Ficoll-Hypaque, the bicyclam AMD3100 CXCR4 antag- buffer (Promega) and then lysed by repeated freeze-thaw cycles in a dry onist, and SP were purchased from Sigma-Aldrich. Antagonists to NK-1 ice/ethanol bath. Cell-free lysates were obtained by centrifugation at ϫ (CP-99,994) and NK-2 (SR-48968) were provided by Pfizer and Sanofi 15,000 g for 5 min at 4°C. Luciferase activities were quantitated with 10 Downloaded from ␮ ␤ Research, respectively. The methods for dissolving and storing SP and the l of lysates using the Luciferase Assay System and -galactosidase using ␤ antagonists were previously described (16). kits from Promega or the Luminescent -galactosidase detection kit II from Clontech, respectively. Luciferase activity was presented per microgram of Cytokines and antibodies total protein in levels normalized with cells transfected with vector alone. Total protein was determined with a kit from Bio-Rad. Recombinant human GM-CSF and SDF-1␣ were purchased from R&D Systems, prolyl-4-hydroxylase mAb from DakoCytomation, PE-anti-CD14 Alk Phos reporter gene assays from BD Pharmingen, p50 from Promega, rabbit anti-SP from Biogenesis, http://www.jimmunol.org/ alkaline phosphatase (Alk Phos)-goat anti-rabbit IgG from Kirkegaard & pSEAP-Basic, pSEAP-Enhancer, or pSEAP-Promoter containing the exon Perry, and rabbit anti-p65 and HRP-goat anti-rabbit IgG from Santa Cruz 1 (sense and antisense) inserts of Tac1 was cotransfected with p␤-gal- Biotechnology. Control (0.5 ␮g each) in 80% confluent BM stroma using SuperFect (Qia- gen). After 48 h, the culture medium was analyzed for Alk Phos using the BM stromal cells Great EscAPe SEAP detection kit (BD Clontech). Cells were scraped in 1 ml PBS and then lysed by repeated cycles of freezing and thawing in a dry BM aspirates were obtained from healthy donors between 18 and 25 years. ice/ethanol bath. Cell-free lysates (24 ␮l) were obtained by centrifugation The use of BM aspirates followed the guidelines of a protocol approved by at 15,000 ϫ g for 5 min at 4°C and then diluted with 5ϫ cell culture lysis the Institutional Review Board of University of Medicine and Dentistry of buffer (Promega). ␤-gal activities were quantitated as described above. New Jersey (Newark, NJ). Nucleated cells (107) were added to 25-cm2 ␣ tissue culture flasks (Falcon 3109) in stromal culture medium ( -MEM Modified long-term culture-initiating cell (LTC-IC) assays by guest on September 28, 2021 with 20% FCS). Flasks were incubated at 37°C. At day 3, the RBC and neutrophils were removed by Ficoll-Hypaque density gradient and the Modified LTC-IC assays were performed in 25-cm2 tissue flasks as de- mononuclear cells were readded in fresh stromal medium. The flasks were scribed (13). The assay used supporting layers of stroma. At confluence, incubated until confluence with a weekly replacement of 50% fresh stromal the supporting cells were ␥-irradiated with 150 gray. After 16 h, the me- medium. At confluence, the trypsin-sensitive adherent cells were passaged. dium was replaced with 5 ml of fresh medium containing BM mononuclear Cells were passaged at least five times before being used in experiments. cells at 107/ml. The cultures were performed in the presence or absence of Flow cytometry studies indicated Ͼ99% of the cells were negative for different concentrations of SDF-1␣ and/or NK1-specific antagonists. There CD14 and positive for prolyl 4-hydroxylase. were weekly replacements of 50% culture medium. At week 12, the adherent and nonadherent cells from each flask were combined and then Northern analysis studied in clonogenic cultures for granulocytic-monocytic progenitors 5 BM stromal cells were stimulated for 6 h with 20 ng/ml SDF-1␣ or were (CFU-GM) as described (13). Briefly, 10 cells were resuspended in 1.2% ␤ methylcellulose in medium supplemented with 3 U/ml GM-CSF and each unstimulated. Northern blots were performed for Tac1 mRNA ( -PPT-A) Ͼ as described (16). Total RNA (15 ␮g) was analyzed with a specific cDNA culture was assayed in duplicate. Colonies with 15 cells were counted on probe for ␤-PPT-I, as previously described (17). The probe was randomly day 10. labeled with [␣-32P]dATP, 3,000 Ci/mM, (DuPont Pharmaceuticals), as described (18). Membranes were stripped and reprobed with cDNA for 18S Western blot rRNA. Hybrids were detected by exposures in PhosphorImager cassettes Nuclear proteins were extracted with the Nxtract kit (Sigma-Aldrich) and (Amersham Biosciences). Cassettes were scanned on a Typhoon 9410 Mo- total protein concentrations were determined using the Bio-Rad DC protein lecular Imager PhosphorImager (Amersham Biosciences). assay. Extracts (15 ␮g) were analyzed by Western blotting using electro- ELISA phoresis with 12% SDS-polyacrylamide gels and the proteins were trans- ferred onto polyvinylidene difluoride membranes (PerkinElmer). The Competitive ELISA quantitated SP as described (19). Streptavidin-coated membranes were incubated overnight with primary Abs followed by 2-h 96-well plates (Sigma-Aldrich) were incubated with biotinylated SP. Equal incubation with HRP-conjugated IgG. The primary and secondary Abs volumes (100 ␮l each) of cell-free medium and rabbit anti-SP at 1/3000 were used at final dilutions of 1/1000 and 1/2000, respectively. HRP was dilution were added to the wells. Each sample was tested in triplicate. After developed with a chemiluminescence detection reagent (PerkinElmer). this, wells were incubated with Alk Phos-conjugated goat anti-rabbit IgG and Sigma 104 phosphatase substrate (Sigma-Aldrich). SP levels were EMSA calculated from a standard curve produced from 12 serial dilutions of EMSA for NF-␬B binding was performed as described (18). Double- known SP concentrations and the absorbance was read at an OD of 405 ␣ stranded oligonucleotides were synthesized at the Molecular Core Facility nm. SDF-1 levels were quantitated with a Quantikine ELISA colori- of the New Jersey Medical School (Newark, NJ). Sequences were synthe- metric quantitation kit (R&D Systems) according to the manufacturer’s sized based on the 5Ј flanking region of Tac1 (GenBank accession no. instruction. AF252261). Wild-type (ϩ791/ϩ824) sequences was 5Ј-CCCGCGGGA Vectors CTGTCCGTCGCAGTAAGTGCCCGCG-3Ј (sense). Both the sense and antisense sequences have TG overhangs in the 5Ј ends that served as end pGL3-Basic vectors containing inserts of the 5Ј flanking regions of Tac1 fillings with reverse transcriptase (SuperScript; Invitrogen Life Technolo- (PPT-I/1.2 and PPT-I/N0) were previously described (18). Briefly, the gies) and [32P]CTP and dATP (50 ␮Ci of 3000 Ci/mM; PerkinElmer). The The Journal of Immunology 2077

FIGURE 1. Effects of SDF-1␣ on the activity of the 5Ј ﬂanking regions of Tac1. A, Schematic showing the relative sizes within the 5Ј ﬂanking region of Tac1. B, Stromal cells were cotransfected with pGL3-PPT-I-1.2 and pGL3-␤-gal. After 16 h, transfectants were stimulated with various concentrations of SDF-1␣, and then 16 h later cell extracts were analyzed for luciferase activities and ␤-gal. Luciferase activities Downloaded from were normalized based on ␤-gal activities and presented as the mean Ϯ SD, n ϭ 4. C and D, Stromal cells were transfected as in B, except that pGL3-PPT-I/N0 were included and then subjected to similar analyses; the results are presented as the mean Ϯ SD, n ϭ 7. http://www.jimmunol.org/ by guest on September 28, 2021

consensus sequence for NF-␬B is underlined (18). Mutant sequences concentration of AMD3100 was based on dose-response studies changed the third G to A (G3A and the eighth T to G (T3G). Double- ranging between 0.1–100 ng/ml. The effects of SDF-1␣ were re- ␮ stranded probes were prepared with 2.5 g of the forward and reverse versed by AMD3100, indicating specific effects via the CXCR4 oligonucleotides. The reaction mix consisted of 2.5 ␮g of dsDNA, 3 ␮gof poly(deoxyinosinic-deoxycytidylic acid) (Sigma-Aldrich), and 25 ␮gof receptor (Fig. 1C). proteins in the presence or absence of anti-p65 at 1/2000 final dilution. The Exon 1 has been shown to have a consensus sequence for control reaction contained 1 ␮g of p50. NF-␬B site and might be responsible for the suppression observed beyond 20 ng/ml SDF-1␣ (Fig. 1A) (18). To determine the signif- Data analyses icance of exon 1, we omitted this region in PPT-I-1.2, designated Statistical evaluations of the data were done with ANOVA and Tukey- pGL3-PPT-I-N0 (Fig. 1A), and then transfected stroma. The results Ͻ Kramer multiple comparisons test. A p value of 0.05 was considered showed a dose-response effect (Fig. 1D), indicating that exon 1 is significant. important in Tac1 regulation in BM stroma at concentrations of Results SDF-1␣ above the endogenous level. Effects of SDF-1␣ on the activation of the 5Ј flanking region of Tac1 Characterization of Tac1-exon 1 in BM stromal cells The effects of SDF-1␣ at levels above baseline in stromal cells on The differences in luciferase activities between PPT-I-1.2 and the expression of Tac1 were studied with reporter gene constructs. PPT-I-N0 led us to examine exon 1 closer, because this region We first studied reporter gene activities of the 5Ј flanking region of accounts for the variations in activities between the two fragments Tac1, previously designated as pGL3-PPT-I/1.2 (18). The frag- (Fig. 1, C and D). Luciferase levels (Fig. 1) suggest that exon 1 ment includes 722 bp upstream of the transcription start site, exon could have repressor activity at 50 and 100 ng/ml SDF-1␣. If this 1, and part of intron 1 (Fig. 1A). We first performed dose-response region is indeed a repressor, then it should suppress the activity of studies by stimulating pGL-1-PPT-1.2-transfected BM stroma a heterologous promoter. We therefore inserted exon 1 into with SDF-1␣ ranging between 5 and 100 ng/ml. Luciferase activ- pSEAP-Basic, which secretes Alk Phos, making its activity quan- ities indicated the optimum response at 20 ng/ml and decreased tifiable at various times after transfection. At the end of the ex- responses at 50 and 100 ng/ml (Fig. 1B). We next focused on 20, periment, cellular ␤-gal activities were quantitated for transfection 50, or 100 ng/ml SDF-1␣ by repeating the studies in the presence efficiency. The results are shown for Alk Phos levels at 48 h after or absence of the CXCR4 antagonist AMD3100 at 10 ng/ml. The transfection. The fold change from transfectants with vector alone 2078 SDF-1␣-Tac1 IN HEMOPOIESIS

the response is concentration dependent. We therefore transfected stromal cells with pSEAP-Basic-exon 1 and then stimulated the cells with SDF-1␣ at 20, 50, and 100 ng/ml. After 48 h, Alk Phos activities showed increased levels in the presence of 20 ng/ml SDF-1␣ and signiﬁcant ( p Ͻ 0.05) reduction at 50 and 100 ng/ml (Fig. 3A). Exon 1 has a consensus region for NF-␬B (18). We now ask whether this region could be involved in the repression at high SDF-1␣ levels. To address this question, we needed to determine whether the consensus region is a functional NF-␬B site. To this end, a gel shift assay was used to study the binding of p50 to wild-type and mutant 32P-labeled nucleotide probes spanning the consensus region. Controls with puriﬁed p50 showed a band with the wild-type probe but not with mutant probe (Fig. 3B, second and third lanes from the left). To test binding with NF-␬B subunits present in nuclear extracts, we selected extracts from IL-1␣-stimulated stroma, shown to be positive for p65 by Western blotting (not shown). The results showed a sharp band that supershifted

with anti-p65 (Fig. 3B, second and third lanes from the right). Downloaded from There was a light to almost undetectable band when the nuclear extract was incubated with a mutant probe (Fig. 3B, right lane). The role of NF-␬B in the SDF-1␣-mediated activity of exon 1 was studied by cotransfecting stroma with pSEAP-Basic/exon 1 and wild-type or mutant I␬B. The levels of Alk Phos after 48 h

showed increases at 50 and 100 ng/ml SDF-1␣ but no effect for http://www.jimmunol.org/ unstimulated transfectants or for 20 ng/ml SDF-1␣ (Fig. 3D). Because exon 1 acted as a repressor for PPT-I-1.2 (Fig. 1, C and D), we asked whether NF-␬B can also mediates repressive functions for PPT-I-1.2 following stimulation with high SDF-1␣ levels. FIGURE 2. Function of Tac1 exon 1 in BM stromal cells. A, BM stro- We repeated the studies described for Fig. 3D, except that the mal cells were transfected with pSEAP-Basic, pSEAP-Enhancer, or whole cell extracts were studied for luciferase activity. The results pSEAP-Promoter containing exon 1. At different times up to 48 h, Alk showed increased luciferase in the presence of mutant I␬B, with 50 Phos activities were quantitated with aliquots of culture medium. The re- and 100 ng/ml SDF-1␣ (Fig. 3E, two right groups). Similar effects sults are shown for values at 48 h and presented as activity normalized with ␣ by guest on September 28, 2021 ␤ Ϯ ϭ were not observed in the cases of 20 ng/ml SDF-1 and unstimu- -gal, mean SD, n 5. B, BM stromal cells were transfected with ␬ pSEAP-Promoter with exon 1 inserted in the sense (S) or antisense (AS) lated transfectants (Fig. 3E, two left groups). In summary, NF- B orientation. After 48 h, Alk Phos activities were determined and presented is shown to mediate the repressive activity of PPT-I in stromal .p Ͻ 0.05 vs unstimulated. cells at high SDF-1␣ concentrations ,ء .as normalized values, mean Ϯ SD, n ϭ 5 Induction of endogenous Tac1 by SDF-1␣ was 20-fold for pSEAP-Basic (Fig. 2A, open bar). Similar trans- We next determined whether SDF-1␣ could induce the expression fection with the upstream sequences resulted in Ͼ92-fold increase of endogenous Tac1. BM stromal cells were stimulated with 20 (not shown). Thus, in the absence of the upstream sequences, exon ng/ml SDF-1␣ and after 6 h total RNA was extracted and then 1 appears to be a weak promoter. analyzed by Northern analyses for Tac1 mRNA (␤-PPT-I). Com- The next set of transfections confirms promoter activity for exon pared with unstimulated stroma, there were significantly denser 1 by the insertion into pSEAP of a heterologous enhancer (pSEAP- bands in stimulated stroma (representative blots are shown in Fig. Enhancer) or a promoter (pSEAP-Promoter). In both transfections, 4A, n ϭ 4). We next determined whether the increase in ␤-PPT-I Alk Phos levels were significantly ( p Ͻ 0.05) increased over p- correlated with protein increase. SP levels served as indicators of SEAP-Basic (Fig. 2A, middle and right bars). The 3-fold increase ␤-PPT-I expression because it could be produced by each of the for pSEAP-Promoter suggests two reasons: exon 1 could have an four possible Tac1 transcripts (1). enhancer function, or perhaps the increase is due to the outcome of Confluent stromal cells were stimulated with 20, 50, and 100 two promoters, the pSEAP-Promoter and exon 1 promoter activity, ng/ml SDF-1␣. At 24, 36, and 48 h, culture medium was collected as suggested in the left open bar of Fig. 2A. and then quantitated for SP levels by ELISA. At 20 ng/ml SDF-1␣, To ascertain that exon 1 is indeed a weak promoter and not an there was significant ( p Ͻ 0.05) increase in SP levels at 24 and enhancer, we next inserted the sequence in the sense and anti-sense 36 h, but a reduction at 48 h (Fig. 4B, left bars). At 50 ng/ml orientations in the pSEAP-Promoter vector and then transfected SDF-1␣, there was similar trend but significantly ( p Ͻ 0.05) re- BM stroma. After 48 h, Alk Phos activities showed no significant duced SP (Fig. 4B, middle bars). At 100 ng/ml SDF-1␣ the levels ( p Ͼ 0.05) change in reporter gene activity (Fig. 2B), indicating were minimal (Fig. 4B, right bars). Together, this section shows that exon 1 does not have enhancer functions in stroma and there- Tac1 induction by SDF-1␣ at 20 ng/ml over baseline production in fore supporting weak promoter function. BM stroma but reduced production at high levels. NF-␬B as a mediator on the repressor effects at high SDF-1␣ Effects of exogenous SDF-1␣ on LTC-IC cultures levels SP has been reported to induce the production of cytokines with Because exon 1 shows weak promoter activity (Fig. 2), we next hemopoietic stimulatory effects (20) and has also been shown to determined whether it is responsive to SDF-1␣ and, if so, whether enhance hemopoiesis, both at the level of granulocytic-monocytic The Journal of Immunology 2079 Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 3. Role of NF-␬B in SDF-1␣-mediated activation of reporter gene activities. A, Cells were transfected as in Fig. 2A except that the transfectants were stimulated with SDF-1␣ at 20, 50, and 100 ng/ml. The results are mean Ϯ SD, n ϭ 4. B, Representative of three gel shift assays done with nuclear extracts from IL-1␣-stimulated stroma. Positive control was done with wild-type probe and 1 ␮g of purified p50. Specificity of binding was studied by supershift with anti-p65 at 1/2000 final dilution. Parallel studies were done with mutant probes. C, Representative of three Western blots for p65 done with nuclear extracts from stromal cells stimulated for 2 h with various concentrations of SDF-1␣. The membrane was stripped and reprobed with anti-␤-actin. D, The studies described in A were repeated, except that the cells were cotransfected with wild-type or mutant I␬B. The results are the mean Ϯ SD, n ϭ 4. E, Stromal cells were cotransfected with pGL3-PPT-I-1.2 and wild-type or mutant I␬B. After ;␣p Ͻ 0.05 vs 20 ng/ml SDF-1 ,ء .this, the experimental steps followed those in Fig. 1B. The results are presented as the mean Ϯ SD, n ϭ 4 .p Ͻ 0.05 vs mutant I␬B ,ءء

progenitors and at the level of immature progenitors/LTC-IC cells more, at 100 nM neither antagonist affected the viability of BM (21). We therefore determined whether SDF-1␣, added exog- mononuclear cells and stromal cells in 12-wk cultures, as indicated enously at different levels (20, 50 and 100 ng/ml), could enhance by trypan blue exclusion. Positive controls were done in parallel LTC-IC proliferation. The studies were done for 6- and 12-wk with SP, which is known to stimulate LTC-IC cells (1). Baseline cultures. These two time points were selected because the 6-wk cultures (medium) showed no change in 6- and 12- wk cultures in assay evaluates hemopoietic progenitors that are relatively more the presence of NK1 antagonists (Fig. 6, A and B). However, the mature than those in 12-wk cultures. Thus, the analyses would NK2 antagonist resulted in a significant ( p Ͻ 0.05) increase in provide information on both the mature and immature hemopoietic CFU-GM (Figs. 6, A and B). progenitors. The results showed significant ( p Ͻ 0.05) increases in LTC-IC cultures (6- and 12-wk) stimulated with SDF-1␣ in the CFU-GM (readout) for 6-wk cultures at all SDF-1␣ levels but only presence of the NK1 antagonist showed significant reduction in at 20 ng/ml for 12-wk cultures (Fig. 5). CFU-GM ( p Ͻ 0.05) as compared with cultures without the antagonist (Fig. 6, A and B). Similarly for LTC-IC with medium, in ␣ SP is a mediator in the activation of LTC-IC cells by SDF-1 the presence of SDF-1␣ and the NK-2 antagonist there were sig- SDF-1␣ has been shown to induce the production of SP in BM nificant ( p Ͻ 0.05) increases in CFU-GM colonies (Fig. 6, A and stroma (Fig. 4). We asked whether SP might mediate the enhanced B). The results showed that the increases in LTC-IC activity at 6- hemopoietic effects by SDF-1␣ at 20 ng/ml (Fig. 5). The LTC-IC and 12-wk cultures by 20 ng/ml SDF-1␣ were mediated by the assays were repeated in the presence or absence of 100 nM antag- NK1 receptor. In the absence of NK2 activation, there was signif- onists specific for NK1 (CP-99,994) or NK2 (SR-48968). These icant ( p Ͻ 0.05) increase in CFU-GM over cultures with only concentrations were determined in dose-response studies. Further- SDF-1␣. 2080 SDF-1␣-Tac1 IN HEMOPOIESIS Downloaded from

FIGURE 4. Effects of SDF-1␣ on the expression of Tac1. A, Represen- http://www.jimmunol.org/ tative of four Northern blots for Tac1 mRNA (␤-PPT-I) with total RNA from stroma, unstimulated or stimulated with 20 ng/ml SDF-1␣. The lanes were normalized with 18S rRNA. B, ELISA quantitated SP in serum-free medium obtained from stroma, unstimulated or stimulated for various times with different levels of SDF-1␣. The results are presented as the mean SP levels Ϯ SD (pg/ml; n ϭ 5).

Effect of SP on the induction of SDF-1␣ in BM stroma by guest on September 28, 2021 It is possible that SP induced by SDF-1␣ in stroma could lead to FIGURE 6. SP is a secondary mediator in the activation of LTC-IC cells ␣ 7 autocrine regulation of hemopoiesis. In this case, SP production by SDF-1 . BM mononuclear cells (10 /ml) were studied in LTC-IC assays in the presence of 20 ng/ml SDF-1␣ and/or 100 nM an antagonist to would be expected to increase SDF-1␣ production in stroma. To NK1, CP-99,994 (A), or to NK2, SR-48968 (B). Positive controls were address this question, we stimulated stroma with 10 nM SP in the stimulated with 10 nM SP. CFU-GM were analyzed as for Fig. 5 and then p Ͻ 0.05 vs unstimulated ,ء .presence or absence of 100 nM NK1 antagonist. At different times presented as the mean CFU-GM Ϯ SD, n ϭ 5 p Ͻ 0.05 vs SDF-1␣ stimulated without NK1 ,ءء ;or SDF-1␣ stimulated antagonist.

the culture medium was collected and then quantitated for SDF-1␣ levels by ELISA. The lower end of the standard curve, 0.5 pg/ml, was included in the linear section of the graph. The results showed

FIGURE 5. Effects of exogenous SDF-1␣ on hemopoiesis. BM mononuclear cells (107/ml) were added to confluent gamma-irradiated stroma in the presence or absence of 20, 50, and 100 ng/ml SDF-1␣.Atwks6and FIGURE 7. Effects of SP on SDF-1␣ production in BM stroma. ELISA 12, aliquots of mononuclear cells were studied in clonogenic assays for quantitated the levels of SDF-1␣ in BM stroma, unstimulated or stimu- CFU-GM. The number of CFU-GM colonies are presented as the mean Ϯ lated, for 24 h with 10 nM SP in the presence or absence of 100 nM NK1 SD, n ϭ 5. Each experiment was performed with cells from a different antagonist (CP-99,994) in sera-free medium. The results are presented as donor. the mean SDF-1␣ levels Ϯ SD (pg/ml, n ϭ 5). The Journal of Immunology 2081 no significant ( p Ͼ 0.05) change in SDF-1␣ levels over baseline Although there are several possible explanations for this observa- (Fig. 7), indicating that, at least in a pure culture of BM stroma, SP tion, perhaps high SDF-1␣ might cause the movement of hemo- does not induce the production of SDF-1␣. poietic stem cells from the marrow region. If this occurs, then the 12-wk LTC-IC cells will undergo cell death due to the lack of Discussion cellular support of stroma. These are intriguing observations that This study reports on the link between SDF-1␣ and the Tac1 gene should be explored further in future studies. Also, the lack of ev- in BM stroma. We have found that the concentrations of SDF-1␣ idence for autocrine stimulation of stroma by SP to produce affect the expression of Tac1, specifically via CXCR4 (Fig. 1). SDF-1␣ further supports receptor crosstalk between NK1 and NK2 Although relatively low concentrations (20 ng/ml) of SDF-1␣ ac- (5). Despite the evidence of intracellular crosstalk between NK1 and tivated the 5Ј flanking region of Tac1, higher concentrations (50– NK2, the molecular mechanism of this occurrence is unknown. 100 ng/ml) were inhibitory (Fig. 1, B and C). Because the CXCR4 These studies have implications for the neural-immune-hemo- is a G protein-coupled receptor, one can argue that receptor de- poietic axis. The BM is innervated by peptidergic fibers including sensitization could be operative at high SDF-1␣ levels (22). How- those that are positive for SP. Because SP is a hemopoietic stim- ever, our studies have ruled out this mechanism because we ulator, it is logical to assume that this report showing a link be- showed a dose-dependent effect in reporter gene activities when tween SP-SDF-1␣-hemopoietic axis could be linked to the nervous the 3Ј region of reporter gene fragment was omitted (Figs. 1D). We system. Evidence has shown that often the signal to increase hemo- concluded that mechanisms other than receptor desensitization poiesis comes from the central nervous system and is regulated could be possible. Analyses of exon 1 revealed weak promoter secondarily by cytokines (23, 24). For example, the SDF-1␣ function in BM stroma (Figs. 2, A and B). Exon 1 could also act as shown in this report could be a mediator in the neural-hemopoietic Downloaded from a negative regulator of PPT-I-1.2 (Fig. 1). response through SP and perhaps other neurotransmitters. It would The dual role of exon 1, weak promoter vs repressor activity, be interesting in future studies to determine whether SDF-1␣ and makes this fragment a region of interest with regard to Tac1 reg- other mediators, through retrograde uptake, may serve as a nega- ulation. Analyses of the NF-␬B region in exon 1 indicate that this tive feedback on the nervous system to turn off the signals coming transcription factor is relevant to the repressive function of high from the central nervous system (25–28).

SDF-1␣ levels (Fig. 2). This does not, however, fulfill the broad This report also has direct relevance to functions of the BM http://www.jimmunol.org/ role of NF-␬B in the enhancing effects of low SDF-1␣ levels. microenvironment. In BM, HSCs are sequestered is a quiescent Previous studies have reported enhancing functions of cAMP re- state close to the endosteum bound to stromal cells through SDF- sponse element regions on Tac1 (18). NF-␬B might be involved in 1␣/CXCR4 interaction (10). To enter hemopoiesis, the HSCs need an antagonistic effect on CREBs at high levels of SDF-1␣. These to move from the stem cell niche into the proliferative niche (29). mechanistic pathways are important and represent ongoing studies This occurs through up-regulation of various proteases and cyto- in the laboratory. kines in the stem cell niche. Both SP and SDF-1␣ have been shown The verification of enhanced Tac1 promoter activity following to function as hemopoietic enhancers (5, 8). Importantly, SP has SDF-1 stimulation confirmed enhanced expression of Tac1 at both been shown to up-regulate GM-CSF, which has been reported to the mRNA and protein levels (Fig. 4). These effects of exogenous mobilize HSCs into the periphery through down-regulation of by guest on September 28, 2021 SDF-1␣ on Tac1 were relevant to hemopoiesis as shown by in- SDF-1␣ (30, 31). creased hemopoietic activity in LTC-IC cultures (Fig. 5). Because Therefore, the relationship that is demonstrated between the studies were done by the LTC-IC assays, we concluded that the SDF-1␣ and SP in hemopoiesis could involve secondary produc- level of hemopoietic regulation occurred at the most primitive tion of GM-CSF. This could result in hemopoietic stimulation or level. perhaps antagonize the increase in SDF-1␣ to move the HSC into Although the NK1 antagonist blunted the hemopoietic effects, the proliferative niche, consequently increasing hemopoiesis. In we showed a significant increase in hemopoietic activity in the summary, this report shows a novel mechanism by which a neu- presence of NK2 antagonists (Fig. 6). This suggested that NK2 rotransmitter, SP, interacts with SDF-1␣ within the BM microen- activation is important to prevent exacerbated stimulation by NK1 vironment to affect hemopoiesis. These studies have implications on hemopoiesis. This observation is consistent with other studies for the neural-hemopoietic axis. showing functional crosstalk between NK1 and NK2 (5). Another possibility for this enhanced hemopoietic effect of the NK2 antag- Disclosures onist is that SP, in the absence of antagonist, could bind weakly to The authors have no financial conflict of interest. NK2 and might elicit negative effects through the production of cytokines with inhibitory effects on hemopoiesis (1). References These studies do not show evidence that SP regulates the pro- 1. Greco, S. J., K. E. Corcoran, K. J. Cho, and P. Rameshwar. 2004. Tachykinins in duction of SDF-1␣, either negatively or positively. We have ob- the emerging immune system: relevance to bone marrow homeostasis and main- ␣ tenance of hematopoietic stem cells. Front. Biosci. 9: 1782–1793. served similar SDF-1 levels in stromal cells stimulated with ex- 2. Rameshwar, P., A. Poddar, and P. Gascon. 1997. Hematopoietic regulation ogenous SP and unstimulated stroma (Fig. 7). This ruled out the mediated by interactions among the neurokinins and cytokines. Leuk. Lymphoma possibility that SP, in the presence of the NK1 antagonist, could 28: 1–10. 3. Singh, D., D. D. Joshi, M. Hameed, J. Qian, P. Gascoń, P. B. Maloof, A. act as negative feedback on SDF-1 production to reduce hemopoi- Mosenthal, and P. Rameshwar. 2001. Increased expression of preprotachykinin-I etic activity (Fig. 6). The observed lack of autocrine stimulation by and neurokinin receptors in human breast cancer cells: implications for BM me- SP not being able to induce the production of SDF-1␣ further in tastasis. Proc. Natl. Acad. Sci. USA 97: 388–393. 4. Kang, H. S., K. A. Trzaska, K. Corcoran, V. T. Chang, and P. Rameshwar. 2004. stroma might be explained by the normal biology of the BM. Spe- Neurokinin receptors: relevance to the emerging immune system. Arch. Immunol. cifically, hemopoietic stem cells and stroma interact close to the Ther. Exp. 52: 338–347. ␣ 5. Bandari, P., J. Qian, H. S. Oh, J. A. Potian, G. Yehia, J. S. Harrison, and P. endosteum to retain the stem cells (9). SDF-1 increase could lead Rameshwar. 2003. Crosstalk between neurokinin receptors is relevant to hema- to mobilization of the stem cells into the periphery. Thus, physi- topoietic regulation: cloning and characterization of neurokinin-2 promoter. ologically it would be a disadvantage if SDF-1␣-mediated produc- J. Neuroimmunol. 138: 65–75. ␣ 6. Bellucci, F., F. Carini, C. Catalani, P. Cucchi, A. Lecci, S. Meini, R. Patacchini, tion of SP causes autocrine production of SDF-1 . In fact, high L. Quartara, R. Ricci, M. Tramontana, et al. 2002. Pharmacological profile of the levels of SDF-1␣ showed suppression of 12-wk LTC-IC (Fig. 5). novel mammalian tachykinin hemokinin 1. Br. J. Pharmacol. 135: 266–274. 2082 SDF-1␣-Tac1 IN HEMOPOIESIS

7. Rameshwar, P., and P. Gascon. 1995. Substance P (SP) mediates production of 19. Rameshwar, P., and P. Gascon. 1996. Induction of negative hematopoietic reg- stem cell factor and interleukin-1 in bone marrow stroma: potential autoregula- ulators by neurokinin-A in bone marrow stroma. Blood 88: 98–106. tory role for these cytokines in SP receptor expression and induction. Blood 86: 20. Rameshwar, P., G. Zhu, R. J. Donnelly, J. Qian, H. Ge, K. R. Goldstein, 482–490. T. N. Denny, and P. Gascoń. 2001. The dynamics of bone marrow stromal cells 8. Arai, A., A. Jin, W. Yan, D. Mizuchi, K. Yamamoto, T. Nanki, and O. Miura. in the proliferation of multipotent hematopoietic progenitors by substance P: an 2005. SDF-1 synergistically enhances IL-3-induced activation of the Raf-1/MEK/ understanding of the effects of a neurotransmitter on the differentiating hemato- Erk signaling pathway through activation of Rac and its effector Pak kinases to poietic stem cell. J. Neuroimmunol. 121: 22–31. promote hematopoiesis and chemotaxis. Cell. Signal. 17: 497–506. 21. Gascon, P., J. Qian, D. D. Joshi, T. Teli, A. Haider, and P. Rameshwar. 2000. 9. Lapidot, T., A. Dar, and O. Kollet. 2005. How do stem cells find their way home? Effects of preprotachykinin-I peptides on hematopoietic homeostasis. A role for Blood 106: 1901–1910. bone marrow endopeptidases. Ann. N Y Acad. Sci. 917: 416–423. 10. Kucia, M., R. Reca, K. Miekus, J. Wanzeck, W. Wojakowski, A. Janowska- 22. Dar, A., P. Goichberg, V. Shinder, A. Kalinkovich, O. Kollet, N. Netzer, R. Wieczorek, J. Ratajczak, and M. Z. Ratajczak. 2005. Trafficking of normal stem Margalit, M. Zsak, A. Nagler, I. Hardan, et al. 2005. Chemokine receptor cells and metastasis of cancer stem cells involve similar mechanisms: pivotal role CXCR4-dependent internalization and resecretion of functional chemokine of the SDF-1-CXCR4 axis. Stem Cells 7: 879–894. SDF-1 by bone marrow endothelial and stromal cells. Nat. Immunol. 6: 11. Neiva, K., Y. X. Sun, and R. S. Taichman. 2005. The role of osteoblasts in 1038–1046. regulating hematopoietic stem cell activity and tumor metastasis. Braz. J. Med. 23. Miyan, J. A., C. S. Broome, and A. M. Afan. 1998. Coordinated host defense Biol. Res. 38: 1449–1454. through an integration of the neural, immune and haemopoietic. Domest. Anim. 12. Laurence, A. D. 2006. Location, movement and survival: the role of chemokines Endocrinol. 15: 297–304. in haematopoiesis and malignancy. Br. J. Haematol. 132: 255–267. 24. Wrona, D. 2006. Neural-immune interactions: an integrative view of the bidirec- 13. Moharita, A. L., M. Taborga, K. E. Corcoran, M. Bryan, P. S. Patel, and P. ␣ tional relationship between the brain and immune systems. J. Neuroimmunol. Rameshwar. 2006. SDF-1 regulation in breast cancer cells contacting bone mar- 172: 38–58. row stroma is critical for normal hematopoiesis. Blood 108: 3245–3252. 25. Felten, D. L., S. Y. Felten, D. L. Bellinger, and D. Lorton. 1992. Noradrenergic 14. Perez, L. E., O. Alpdogan, J. H. Shieh, D. Wong, A. Merzouk, H. Salari, R. J. and peptidergic innervation of secondary lymphoid organs: role in experimental O’Reilly, M. R. van den Brink, and M. A. Moore. 2004. Increased plasma levels rheumatoid arthritis. Eur. J. Clin. Invest. 1: 37–41. of stromal-derived factor-1 (SDF-1/CXCL12) enhance human thrombopoiesis and mobilize human colony-forming cells (CFC) in NOD/SCID mice. Exp. Hematol. 3: 26. Weihe, E., D. Nohr, S. Michel, S. Muller, H. J. Zentel, T. Fink, and J. Krekel. Downloaded from 300–307. 1991. Molecular anatomy of the neuro-immune connection. Int. J. Neurosci. 15. Flomenberg, N., J. DiPersio, and G. Calandra. 2005. Role of CXCR4 chemokine 59: 1–23. receptor blockade using AMD3100 for mobilization of autologous hematopoietic 27. Campenot, R. B. 1994. NGF and the local control of nerve terminal growth. progenitor cells. Acta Haematol. 114: 198–205. J. Neurobiol. 25: 599–611. 16. Rameshwar, P., A. Poddar, G. Zhu, and P. Gascon. 1997. Receptor induction 28. Ko¨bbert, C., R. Apps, I. Bechmann, J. L. Lanciego, J. Mey, and S. Thanos. 2000. regulates the synergistic effects of substance P with IL-1 and PDGF on the pro- Current concepts in neuroanatomical tracing. Prog. Neurobiol. 62: 327–351. liferation of bone marrow fibroblasts. J. Immunol. 158: 3417–3424. 29. Heissig, B., K. Hattori, S. Dias, M. Friedrich, B. Ferris, N. R. Hackett, R. G. Crystal, P. Besmer, D. Lyden, M. A. Moore, et al. 2002. Recruitment of stem and 17. Oh, H. S., A. Moharita, J. G. Potian, I. P. Whitehead, J. C. Livingston, T. A. http://www.jimmunol.org/ Castro, P. S. Patel, and P. Rameshwar. 2004. Bone marrow stroma influences progenitor cells from the bone marrow niche requires MMP-9 mediated release transforming growth factor-␤ production in breast cancer cells to regulate c-myc of kit-ligand. Cell 109: 625–637. activation of the preprotachykinin-I gene in breast cancer cells. Cancer Res. 64: 30. Hattori, K., B. Heissig, and S. Rafii. 2003. The regulation of hematopoietic stem 6327–6336. cell and progenitor mobilization by chemokine SDF-1. Leuk. Lymphoma 44: 18. Qian, J., G. Yehia, C. Molina, A. Fernandes, R. J. Donnelly, D. J. Anjaria, 575–582. P. Gascoń, and P. Rameshwar. 2001. Cloning of human preprotachykinin-I pro- 31. Gazitt, Y. 2002. Comparison between granulocyte colony-stimulating factor and moter and the role of cAMP response elements in its expression by IL-1 and stem granulocyte-macrophage colony-stimulating factor in the mobilization of periph- cell factor. J. Immunol. 166: 2553–2561. eral blood stem cells. Curr. Opin. Hematol. 9: 190–198. by guest on September 28, 2021