INTRODUCTION

Neurological disorders affect people in all countriesirrespective of age, sex, education or income, and strokeremains the second-most common cause of death worldwide.Thrombolytic therapy is able to restore cerebral blood flow insome patients with acute ischemic stroke and can lead to theimprovement or resolution of neurologic deficits. However,only a limited number of patients can be treated in this way,because the treatment must occur within 3 hours of the onset ofstroke (1). Neurodegenerative disorders such as Alzheimer’sdisease, Parkinson’s disease, amyotrophic lateral sclerosis andothers affect millions of people worldwide. These diseasesresult in the gradual and progressive loss of neural cells,leading to nervous system dysfunction. Spinal cord injury andoptic nerve injury results in loss of neurons, degeneration ofaxons, formation of glial scar, and severe functionalimpairment. Due to the lack of effective therapies forneurodegenerative diseases, the development of stem cell-based therapeutic strategies aiming to exert trophicneuroprotective activity or replace the lost neurons have begunand show promise for treating these diseases.

In recent decades, a number of studies have been performed toassess the feasibility and benefits of stem cell-based experimentalapproaches to regeneration and trophic support for neural tissue.Stem/progenitor cells (SPCs) have the potential to self-renew andto differentiate into mature cells. These cells have also beendemonstrated to exert trophic activity by producing importantgrowth factors (2). The cell membrane protein CD34 is animportant marker of early SPCs, and the CD34+ cell population isthought to comprise early hematopoietic SPCs and endothelialprogenitor cells. Since it has been reported that the infusion of largenumbers of CD34+ cells has had a positive impact on the outcomein patients undergoing bone marrow (BM) transplantation in termsof faster engraftment and fewer infectious episodes (3), theestimation of the number of CD34+ cells has been widely used tomonitor hematopoietic stem cell transplantation. The efficacy andsafety of intravenous administration of immunoselected CD34+

cells in hematopoietic reconstitution following high-dosechemotherapy in patients with cancer has also been demonstrated(4). CD34+ cells isolated by magnetic sorting have also been usedin clinical trials studying SPC transplantation for the purpose oftissue regeneration (5). In addition to their hematopoietic potential,CD34+ cells are of potential use in stem cell-based therapies for

JOURNALOF PHYSIOLOGYAND PHARMACOLOGY 2016, 67, 1, 151-159

www.jpp.krakow.pl

E. PACZKOWSKA1, K. PIECYK1, K. LUCZKOWSKA1, M. KOTOWSKI2, D. ROGINSKA1, E. PIUS-SADOWSKA1, K. ORONOWICZ1, M. OSTROWSKI2, B. MACHALINSKI 1

EXPRESSION OFNEUROTROPHINS AND THEIR RECEPTORS IN HUMAN CD34+ BONE MARROW CELLS

1Department of General Pathology, Pomeranian Medical University, Szczecin, Poland; 2Department of General Surgery and Transplantation, Pomeranian Medical University, Szczecin, Poland

Bone marrow (BM) CD34+ cells have the ability to secrete growth factors, cytokines, and chemotactic factors. Wesought to better characterize this population and to investigate whether human BM CD34+ cells express neurotrophins(NTs) and their relevant receptors. We also compared their expression levels with BM nucleated cells (NCs). BM CD34+

cells were evaluated with respect to the expression levels of neurotrophins using qRT-PCR, immunofluorescent staining,and Western blotting. Next, the expression of specific (TrkA, TrkB, TrkC) and non-specific (p75NTR) neurotrophinreceptors was detected by qRT-PCR and immunofluorescent staining in BM CD34+ cells. Using qRT-PCR, we show thateven in the absence of inducing factors, CD34+ cells spontaneously express neurotrophins such as NGF, BDNF, NT-3,and NT-4. In addition, the NTexpression levels in BM CD34+ cells are considerably higher than in NCs. Furthermore,we confirmed intracellular NTexpression in BM CD34+ cells at the protein level using immunofluorescent staining andWestern blotting. Using qRT-PCR, we found that immunomagnetically separated BM CD34+ cells spontaneouslyexpress high-affinity neurotrophin receptors (TrkA, TrkB, and TrkC) and the low-affinity receptor p75NTR at higherlevels than NCs. Immunomagnetic CD34+ cell separation enables for the rapid and gentle sorting of stem/progenitorcells (SPCs) to prepare specific cell types for use in research and clinical applications. Our study suggests that BMCD34+ cells have the potential to support trophic factors for neural tissue and could contribute towards the protectionand regeneration of neural cells.

K e y w o r d s :CD34+ cells, bone marrow, neurotrophins, neurotrophin receptors, neurological disorders, stem cell-based therapy

neural diseases, including spinal cord injury (6) and stroke (7).Human CD34+ cells have been administered intravenously toimmunocompromised mice 48 hours prior to experimentallyinduced ischemia and have been shown to induce long-termneovascularization in the ischemic zone (8). Similarly, a recentstudy demonstrated that human CD34+ cells injected intravenouslyto immunocompromised mice 48 hours after neonatal strokemodestly reduced ischemic brain damage with a transientaugmentation of cerebral blood flow in the peri-infarct area (9).

We and others have previously demonstrated that endogenousBM CD34+ cells are mobilized into the peripheral blood inresponse to central nervous system (CNS) injury followingischemic or hemorrhagic stroke and acute spinal cord injury (10-12). This finding suggests an intrinsic mechanism forameliorating tissue damage that involves circulating SPCs.Moreover, it has been shown that, in patients with intracerebralhemorrhage, the concentration of circulating CD34+ progenitorcells at day 7 is independently associated with good functionaloutcome at 3 months (13). This finding suggests that CD34+

progenitor cells might participate in the functional recovery ofthese patients. These observations have provided the basis fortwo experimental approaches. The first was the intra-arterialtransplantation of BM mononuclear cells (MNCs) in anexperimental model of CNS injury and in clinical trials. Thisprocedure has been shown to be feasible and safe in patients;however, there were no significant differences in neurologicalfunction in stroke patients at 180 days after transplantation (14).The second study used granulocyte-colony stimulating factor (G-CSF) as a SPC-mobilizing cytokine and a neuroprotective factorinitially in an experimental stroke model (15). Recent clinicaltrials have suggested that G-CSF is safe when administeredsubacutely in stroke patients; however, it is too early to knowwhether G-CSF improves functional outcomes (16, 17).

The characterization of the SPC population and optimalchoice of SPCs in terms of their ability to exert paracrine activityfollowing their transplantation into the neural microenvironmentare important prerequisites for the successful application of stemcells in future experimental studies. BM CD34+ cells areheterogeneous and easily available from whole BM aspirates,which can be enriched by immunomagnetic separation.Therefore, in the present study, we sought to better characterizethis population specifically by investigating whether human BMCD34+ cells express NTs and their relevant receptors along witha comparison of their expression levels with those ofunseparated BM nucleated cells (NCs).

MATERIALS AND METHODS

Cells

Bone marrow cells were collected from ten adult heparinizeddeceased organ donors (HDODs). The mean age of the HDODswas 46 ± 15 years (5 men, 5 woman). The donors’families gavetheir consent in each case. All procedures were approved by theLocal Ethics Committee and in accordance with the HelsinkiDeclaration.

BM cells were aspirated from pelvic bones and subsequentlyresuspended in collecting medium phosphate-buffered saline(PBS) and heparin (20 U/mL; Life Technologies, Paisley, UK).Before organ harvest, every HDOD was infused with heparin(25,000 U/donor, Teva, Kutno, Poland) and subsequently BMwas aspirated from iliac crest before disconnection from therespirator. Whole human BM samples were lysed in BDPharmLyse Lysing Solution (BD Biosciences, San Jose, CA,USA) for 15 min at room temperature in the dark and washedtwice in PBS. A part of the obtained suspension of BM NCs was

subjected to immunomagnetic separation procedure. CD34+ cellpopulation enriched in SPCs were isolated from NCs usingcommercially available CD34 MicroBead Kit (Miltenyi Biotec,Auburn, CA, USA). Isolation procedure was performedaccording to the manufacturer’s instructions, as described (18).

qRT-PCR

Total mRNA was isolated from NCs or BM CD34+ cellsusing the RNeasy Mini Kit (Qiagen Inc., CA, USA).Subsequently, the mRNAwas reverse-transcribed using the FirstStrand cDNA Synthesis Kit (Thermo Fisher Scientific, MA,USA). A quantitative assessment of gene expression wasperformed using real time qRT-PCR carried out on a Bio-RadCFX96 Real-Time PCR Detection System (Bio-Rad Inc., CA,USA). The 15 µLreaction mixture contained 7.5 µLof SYBRGreen PCR Master Mix, 10 ng of cDNAtemplate, and one pairof forward and reverse primers. The primers used for thesereactions are listed in Table 1. The threshold cycle (Ct), i.e. thecycle number at which the amount of the amplified gene ofinterest reached a fixed threshold, was subsequently determined.The relative target gene mRNAexpression was quantified usingthe comparative Ct method. The relative quantification value ofthe target was normalized to the endogenous control BMG geneand expressed as 2∆Ct, where ∆Ct = [Ct of endogenous controlgene (BMG)] – [Ct of target genes].

Immunofluorescence

Freshly isolated BM CD34+ cells and NCs were subjected toimmunofluorescence (IF) staining of neurotrophic factors. First,the cells were fixed with 3.7% paraformaldehyde and then smeared

152

BDNF f. 5’ GATGCTCAGTAGTCAAGTGCC 3’

r. 5’ GCCGTTACCCACTCACTAATAC 3’

NGF f. 5’ GCAAGCGGTCATCATCCCAT 3’

r. 5’ TGTTGTTAATGTTCACCTCTCCC 3’

NT-3 f. 5’ GGTACGCGGAGCATAAGAGTC 3’

r. 5’ GAGAGTTGCCCGTTTTGATCT 3’

NT-4 f. 5’ GCGAGGTGGAGGTGTTGG 3’

r. 5’ CCTTCCTCAGCGTTATCAGC 3’

p75NTR f. 5’ CCTACGGCTACTACCAGGATG 3’

r. 5’ CACACGGTGTTCTGCTTGTC 3’

TrkA f. 5’ GTCAGCCACGGTGATGAAATC 3’

r. 5’ CAGCACGTCACGTTCTTCCT 3’

TrkB f. 5’ CTGGTGCATTCCATTCACTG 3’

r. 5’ CGTGGTACTCCGTGTGATTG 3’

TrkC f. 5’ TGGCTGGACTATGTGGGCT 3’

r. 5’ CCCATTGCTGTTCCCTGAATC 3’

BMG f. 5’ AATGCGGCATCTTCAAACCT 3’

r. 5’ TGACTTTGTCACAGCCCAAGATA 3’

Table 1. The primers used for real time qRT-PCR reactions.

on polylysine-coated slides. After permeabilization in 0.5% Tween20 (Bio-Rad Inc.) and blocking with 10% normal goat serum, thesmears were incubated at 4°C overnight with one of the followingprimary antibodies: anti-BDNF, anti-NGF, anti-NT3, anti-NT4,anti-TrkA, anti-TrkB, anti-TrkC, or anti- p75NTR. Subsequently,the cells were incubated in the dark with the relevant secondaryantibodies. The antibodies used for immunofluorescence studiesare listed in Table 2. Upon termination, all of the sections werecounterstained with DAPI solution (Thermo Fisher Scientific,Waltham, MA, USA), mounted, and examined using an LSM700confocal system (Carl Zeiss, Jena, Germany). For quantification ofthe percentage of cells expressing a specific marker in any givenexperiments the number of positive cells was determined inrelation to the total number of DAPI labeled nuclei. Counts ofimmunoreactive cells were made in 10 random fields in two slidesfor each with a 20 objective.

Western blot

Western blot analysis was performed to evaluate the expressionof the BDNF, NGF, NT-3, and NT-4 as well as TrkA, TrkB, TrkC,p75NTR. The BM CD34+ cells and NCs (2 × 106) were lysed for10 min on ice in M-Per lysing buffer (Pierce, Rockford, IL)containing protease and phosphatase inhibitors (Sigma-Aldrich,MO, USA) (10 µg/ml leupeptin, 10 µg/ml aprotinin, 1 µg/mlpepstatin A, 1 mM sodium fluoride, and 2 mM Na3VO4). The celllysates were clarified via centrifugation at 14,000 rpm for 10 minat 4°C, and the protein concentrations were determined using the

Bradford protein assay (Sigma-Aldrich). Three cell lysates werepooled and equal amounts of protein (20 µg/well) were loaded andseparated on a 4 – 0% sodium dodecyl sulfate polyacrylamide gelvia electrophoresis (SDS-PAGE, mini-PROTEAN IIelectrophoresis system, Bio-Rad Inc.) and then transferred to a 0.2µm polyvinylidene fluoride (PVDF) membrane (Bio-Rad Inc.).Kaleidoscope polypeptide standard wide range (10 – 250 kD)protein markers (Bio-Rad Inc.) were used to determine themolecular weights of the analyzed proteins. After blocking non-specific binding for 2 hours at room temperature with a 3% BSA,Tris-HCl and NaCl solution with 0.05% Tween 20, the membranewas probed with a specific monoclonal/polyclonal IgG antibodydirected against amino acid sequences of the selected proteins(BDNF, NGF, NT-3, NT-4, TrkA, TrkB, TrkC, p75NTR) andincubated overnight at 4°C. The antibodies used for Westernblotting studies are listed in Table 2. Immunoreactive bands weredetected using horseradish peroxidase-conjugated secondaryantibody specific to the primary antibody used in the previous step.Chemiluminescence detection was performed using the ECLSelect Detection Kit (Amersham Life Sciences, Buckinghamshire,UK), and the bands were subsequently visualized with a UVPcamera (Gel DOC-It Imaging system, Bio-Rad).

Statistical analysis

Because the distribution of most variables significantlydeviated from normal distribution, non-parametric tests wereused. The significance of differences between the CD34+ cell

153

Antibody name Company Dilution

Antibodies used for immunofluorescence studies

Primary antibodies

rabbit anti-BDNF GenTex 1:50

rabbit anti-NGF Santa Cruz Biotechnology 1:50

rabbit anti-NT3 Acris Antibodies GmbH 1:50

mouse anti-NT4 Santa Cruz Biotechnology 1:50

mouse anti-TrkA Biorbyt 1:500

rabbit anti-TrkB Abcam 1:50

rabbit anti-TrkC Abcam 1:500

rabbit anti-p75 Abcam 1:50

Secondary antibodies

chicken anti- rabbit Life Technologies 1:100

goat anti-mouse Alexa Fluor 488 Life Technologies 1:100

goat anti-rabbit-TR Vector Laboratories 1:100

Antibodies used for Western blot

Primary antibodies

rabbit anti-BDNF Santa Cruz Biotechnology 1:375

rabbit anti-NGF Santa Cruz Biotechnology 1:400

goat anti-NT-3 Santa Cruz Biotechnology 1:375

mouse anti-NT-4 Santa Cruz Biotechnology 1:375

rabbit anti-TrkA Abcam 1:400

rabbit anti-TrkB Abcam 1:300

rabbit anti-TrkC Santa Cruz Biotechnology 1:300

rabbit anti-NGFR p75 Proteintech 1:600

Secondary antibodies

horseradish peroxidase-conjugated

secondary antibody

Santa Cruz Biotechnology 1:2500 – 1:5000

Table 2. The antibodies used in this study.

and NC populations was assessed using the Kruskal-Wallis testfollowed by the Mann-Whitney test. Pvalues < 0.05 wereconsidered statistically significant.

RESULTS

Human BM is a rich source of heterogeneous SPCs. In thisstudy, we isolated BM CD34+ cells from NCs obtained fromHDODs and subsequently used these cells to better characterizethis SPC-enriched population. Non-separated NCs obtained afterthe lysis of erythrocytes were used as a control. The average totalnumber of NCs recovered from a single collection was 544 ± 262× 106 cells. We successfully isolated the CD34+ cells from NCsusing immunomagnetic separation. The number of CD34+ cellsrecovered from 100 × 106 NCs was 1.13 ± 0.63 × 106. Thus, in ourstudies, the NC fraction contained 1.13 ± 0.63% BM CD34+ cells.

Human bone marrow CD34+ cells spontaneously expressneurotrophins

Accumulating data indicate that CD34+ have theneuroprotective properties. By employing qRT PCR, weevaluated the expression of various NTs (BDNF, NGF, NT-3,and NT-4) at the mRNAlevel in BM CD34+ cells and comparedto those observed in unsorted NCs. We found that NTmRNAsare robustly expressed in human BM CD34+ cells; the expressiondata are shown in Fig. 1A. Univariate statistical analysisrevealed that BM CD34+ cells expressed significantly higherlevels of each analyzed neurotrophic factor compared to NCs.Furthermore, we observed a tendency towards an increase in theexpression of NTs in BM CD34+ cells at the protein level, asdetermined by Western blot analysis (Fig. 1B).

Finally, we confirmed the expression of NTs in CD34+ cellsat the protein level by visualizing their intracellular expressionthrough immunofluorescence analyses. We determined that theisolated CD34+ population expressed select NTs (BDNF, NGF,NT-3, and NT-4), as shown by immunofluorescent staining(Fig. 2A). Freshly isolated BM CD34+ cells wereimmunostained for specific neurotrophins including BDNF,NGF, NT-3, and NT-4, which showed that 3.41 ± 7.53% ofCD34+ cells strongly expressed BDNF; 2.13 ± 4.0% expressedNGF; approximately 3.32 ± 6.34% of CD34+ cells were positivefor NT-3; and a few cells (2.62 ± 5.15%) expressed NT-4. Thesedata indicate that human BM CD34+ cells express neurotrophicand neuroprotective factors. Similarly, NCs wereimmunostained for specific neurotrophins and we could observevery few cells expressing NTs, i.e. 0.49 ± 1.27% of NCsexpressed BDNF; 0.26 ± 0.72% expressed NGF; approximately0.22 ± 0.68% of NCs were positive for NT-3; and 0.37 ± 0.92%expressed NT-4 (Fig. 2B). We observed a tendency towardshigher number of cells expressing NTs in BM CD34+ cellscompared to NCs.

Collectively, we found that human BM CD34+ cells expressall of the examined NTs at higher levels than NCs. These resultsstrongly support the hypothesis that BM CD34+ cells might exerta neurotrophic and neuroprotective effect on neural cells.

The expression of neurotrophin receptors is more pronouncedin human bone marrow CD34+ cells compared to nucleatedcells

In recent years, numerous studies have shown that BM cell-induced neuroprotection involves neurotrophic factors actingthrough paracrine and/or autocrine interactions between theneural microenvironment and transplanted cells (19, 20). Using

154

Fig. 1. BM CD34+ populations spontaneously express neurotrophins (NTs).(A) NT mRNA expression levels in BM CD34+ cells and NCs. Relative of mRNAexpression values are normalized against BMGexpression values. Data are presented as median, quartiles, interquartile range, minimum, and maximum (N= 6 for CD34+ cells,N = 10 for NCs). *P< 0.05 and #P < 0.001 versus NCs.(B) Representative Western blot analysis of NTprotein levels in BM CD34+ and NC cell lysates.

qRT-PCR, we evaluated the expression levels of theneurotrophin receptors TrkA, TrkB, TrkC, and p75NTR in BMCD34+ cells compared to NCs.

All neurotrophin receptors were expressed at detectablelevels in each cell type, as shown in Fig. 3A. Univariatestatistical analysis revealed that BM CD34+ cells expressed

155

Fig. 3. Human BM CD34+ cells spontaneously express neurotrophin receptors.(A) NT receptor mRNAexpression levels in BM CD34+ cells and NCs. Relative mRNAexpression values are normalized againstBMG levels. Data are shown as median, quartiles, interquartile range, minimum, and maximum (N= 6 for CD34+ cells, N= 10 forNCs). *P< 0.05 and #P < 0.001 versus NCs.(B) Representative Western blot analysis of NTreceptor protein levels in BM CD34+ and NC cell lysates.

Fig 2. (A) Immunofluorescent analysis of intracellular NTexpression in CD34+ cells. Nuclei were visualized by DAPI staining.Pseudo-coloring was assigned to each stain as follows: anti-BDNF, anti-NGF are red, anti-NT-3, and anti-NT-4 are green, and nucleiare blue.(B) Immunofluorescent analysis of intracellular NTexpression in NCs. Nuclei were visualized by DAPI staining. Pseudo-coloring wasassigned to stain as follows: anti-BDNF, anti-NGF are red, anti-NT-3, and anti-NT-4 are green, and nuclei are blue. All images werecaptured using an LSM700 confocal system (Carl Zeiss, Jena, Germany). Representative data are shown from two independentexperiments.

significantly higher levels of each neurotrophin receptorcompared to unsorted NCs. Moreover, we observed a tendencytowards an increase in the expression of NTreceptors in BMCD34+ cells at the protein level, as determined by Western blotanalysis (Fig. 3B). Furthermore, we assessed the percentage ofBM CD34+ cells that expressed NT-specific receptors byvisualizing the surface protein on freshly isolated BM CD34+

cells using immunofluorescent analyses. We determined that theisolated BM CD34+ cell population expressed the specific NTreceptors (TrkA, TrkB, TrkC), as shown in Fig. 4A. Freshlyisolated BM CD34+ cells were also immunostained for thenonspecific NTreceptor p75NTR. Immunostaining showed that3.81 ± 5.48% of CD34+ cells expressed TrkA (Fig. 4A); 1.70 ±2.97% expressed TrkB; approximately 1.5 ± 0.83% of CD34+

cells were positive for TrkC; and a few cells (1.0 ± 0.53%)expressed p75NTR. These data indicate that human BM CD34+

cells express neurotrophin receptors on their cell membranes.Similarly, we performed analysis of NCs usingimmunofluorescence staining. We observed few cells expressingNT receptors among NCs (0.45 ± 1.46% TrkA positive cells,0.27 ± 0.66% TrkB positive cells, 0.25 ± 0.62% TrkC positivecells, and 0.26 ± 0.88% positive for p75NTR) (Fig. 4B).

Taken together, these results demonstrate that human BMCD34+ cells express specific NTreceptors at the mRNAandprotein levels, suggesting that these cells could respond toneurotrophic factors in neural tissue.

DISCUSSION

The involvement of trophic support in addition to thereplacement of damaged neurons has been proposed as anexplanation for the beneficial effects observed following the

infusion/transplantation of SPCs in experimental animal modelsof neural tissue injury (20, 21). We and others have shown thathuman SPCs could be a source for growth factors (2, 18, 22). Inthis study, we report the expression of neurotrophins in humanBM CD34+ cells at the mRNAand protein levels. Moreover,these cells expressed mRNAs encoding BDNF, NGF, NT-3, andNT-4 on a higher level than nucleated cells. Taken together, thesefindings support the hypothesis that human BM SPCs exertparacrine effects following their transplantation into injuredneural tissue. These cells could be a source for neuroprotectivefactors that inhibit apoptosis and promote neurogenesis andneovascularization in neural tissue. Because BM cells have beendemonstrated to exert positive effects in an experimental modelof nervous tissue injury, it could be speculated thatneurotrophins produced by these cells interact with othereffector cells in the injured neural tissue. A thoroughcharacterization of the SPC population is an important factor indetermining the optimal choice of stem cells for experimentalstudies of specific tissue injury models. In our study, we usedBM harvested from HCOD, because our previous studies havedemonstrated that HCOD BM cells have similar characteristicsto BM cells from healthy subjects (23). We have not observedstatistically significant differences in the transplantationpotential of HSPCs harvested from the BM of HDODs comparedto healthy donors.

CD34+ cells appear to have a number of significantadvantages when considering their use in stem cell-basedtherapies. The CD34+ cell population has been demonstrated tosecrete various growth factors (kit ligand, fibroblast growthfactor-2, vascular endothelial growth factor, hepatocyte growthfactor, insulin-like growth factor-1, and thrombopoietin),cytokines (tumor necrosis factor-α, Fas ligand, interferon α,interleukin 1, and interleukin 16), and chemokines (macrophage

156

Fig. 4. (A) Immunofluorescent analysis of intracellular NTreceptor expression in CD34+ cells. Nuclei were visualized by DAPIstaining. Pseudo-coloring was assigned to each stain, consequently anti-TrkA, anti-TrkB, anti-TrkC, and anti-p75NGR are green, andnuclei are blue.(B) Immunofluorescent analysis of intracellular NTreceptor expression in NCs. Nuclei were visualized by DAPI staining. Pseudo-coloring was assigned to each stain, consequently anti-TrkA, anti-TrkB, anti-TrkC, and anti-p75NGR are green, and nuclei are blue.All images were captured using an LSM700 confocal system (Carl Zeiss, Jena, Germany). Representative data are shown from twoindependent experiments.

inflammatory protein-1α, and 1β, monocyte chemotacticprotein-3, interleukin 8, and others) (2). Consequently, in thisstudy, we show that human BM CD34+ cells express crucialneurotrophins that are required for the repair of neural tissues.Neurotrophins are a family of proteins that are wellcharacterized by their influence on the survival, differentiationand apoptosis of cells in the nervous system (24, 25). Theclassical group of neurotrophins includes NGF, BDNF, NT-3,and NT-4. The effects of neurotrophins are mediated via high-affinity tropomyosin receptor kinase (Trk) receptors TrkA, TrkB,and TrkC and the low-affinity receptor p75NTR, which is amember of the tumor necrosis factor receptor family (26). It hasbeen hypothesized that under some pathological conditions,neural cells fail to be exposed to a necessary level ofneurotrophic factors and consequently undergo apoptosis (27,28). Here we demonstrated that freshly isolated BM CD34+ cellsproduce BDNF, NGF, NT-3, NT-4 on protein level. We did notculture BM CD34+ cells and conduct secretion analysis byELISA or Western blot on the cell culture supernatants.However, the secretion of NTs by BM cells and neuroprotectionof conditioned media has been demonstrated in numerousstudies (29, 30). One study demonstrated that BM mesenchymalstem cells produce a number of neuroprotective proteinsincluding BDNF and suggested that secretion of these factorsmay play an important role in observed mesenchymal stem cell-mediated retinal ganglion cell neuroprotection. Authorsconfirmed the secretion of BDNF by BM cells using ELISA(29). The quantification of secreted NTs by ELISAanalysis ofcell culture supernatants, the defining cell culture conditions aswell as strategies involving pre-treatment of BM CD34+ beforeusing in experimental studies could be a subject of furtherresearch.

Here, we report that mRNAs for neurotrophin receptors arehighly expressed in human BM CD34+ cells. A comparison ofreceptor mRNAexpression showed a higher level of expressionof these surface proteins in BM CD34+ cells compared to NCs.It has been previously shown that TrkA is expressed by andmediates functional responses to NGF in immaturehematopoietic cell lines and mouse precursor BM cells (31).Another study by Laurenzi et al. has demonstrated theexpression of neurotrophin mRNAin human granulocytes andBM cells (32). The presence of NGF receptors, which creates thephysiological microenvironment for hematopoietic stem cells,on BM stromal cells has been reported previously (33). Bracci-Laudiero and co-workers have provided the first evidence thathuman umbilical cord blood (UCB) CD34+ stem cells andprogenitors are able to respond to NGF through the expressionof TrkA receptors and that they are able to synthesize NGFthemselves (34). These findings are in agreement with ourprevious studies showing that UCB CD34+ cells express NGFand its relevant receptor at the mRNAand protein levels (18).NGF expression together with the expression of its relevantreceptor, TrkA, has been observed in whole BM cells (35) and inBM stromal cells (36); however, the expression of NGF inCD34+ cells from BM had not been investigated previously.Here, we clearly demonstrate that human BM CD34+ cellsexpress NGF and its relevant receptor, TrkA, at a significantlyhigher level than NCs, suggesting that the cells are moreresponsive to NGF stimulation than mononuclear cells.

In a previous study, it was demonstrated that B lymphocytesand their precursors express the BDNF receptors p75NTR andTrkB, while no BDNF expression was found in these cells (37).In the BM, BDNF has been found to be expressed inmegakaryocytes, endothelial cells, and osteoblasts (35). Thepresence of BDNF receptors on lymphocyte precursors togetherwith strong BDNF expression in BM stromal cells suggests arole for BDNF in normal B lymphocyte development (36). TrkB

expression has also been shown in BM stromal cells in culture(36). Here, we observed the differences in the spontaneousexpression levels of NTRs between selected BM CD34+ cellswith immature characteristics and BM nucleated cells,suggesting that NTs may play different functional rolesdepending on the level of differentiation and maturity of thecells. In mice with experimental stroke injury, it has beenobserved that following a systemic injection of CD34+-enrichedhuman cord blood cells, there was a significant increase insubgranular zone proliferation compared to control mice (38).Neurotrophin receptors could be involved in the recruitment ofSPCs to specific sites. In an interesting finding by Kermani et al,the induction of neoangiogenesis in an ischemic limb model wasobserved following treatment with BDNF (39). The mechanismby which BDNF enhances capillary formation was mediated inpart through the local activation of the TrkB receptor and by therecruitment of specific subsets of BM-derived hematopoieticcells that co-express TrkB, which provide peri-endothelialsupport for the newly formed vessels.

It is interesting that in a clinical study involving intra-arterial BM MNC transplantation in stroke patients, there was anegative correlation between the total number of CD34+ cellsinjected and MMP-2 levels 4 days post-transplantation, andlower plasma levels of MMP-2 at this time point was associatedwith better functional outcomes (40). In addition, Moniche etal. observed changes in the serum levels of GM-CSF andPDGF-BB even 3 months post-transplantation, which could beassociated with better functional outcomes. These resultssuggest that the effects of BM MNC transplantation onfunctional outcomes in stroke patients are achieved bymultifocal mechanisms that depend on changes in growthfactors, enzymes and possibly the expression of other genes.

The availability of SPCs and their ease of isolation areimportant factors in considering their use in clinical applications.Autologous BM CD34+ cells can be obtained by aspiration ofBM under local anesthesia and separated in a closed system,which is in contrast to endothelial progenitor cells andmesenchymal stem cells that require ex vivo culture. CD34+ cellsisolated by magnetic sorting have been used in SPCtransplantation clinical trials for the purposes of tissueregeneration (5, 41). In a previous clinical study, the possibilityof delivering autologous BM-derived CD34+ cells into the spinalcord of patients with spinal cord injury via the lumbar puncturetechnique has been demonstrated. The results of studies usingmagnetic resonance imaging to monitor the fate of cells labeledwith magnetic nanoparticles suggest that BM CD34+ cellsmigrate into the injured site in patients with chronic spinal cordinjury (42). The neuroprotective effects of CD34+ cells havebeen observed by Boltze et al., who reported that in vivo,intravenously transplanted human UCB MNCs and CD34+ andCD34– cells reduced neurofunctional deficits and the lesionvolume in rats following middle cerebral artery occlusion. Thisstudy also showed that human CD34+ cells within MNCs werepreferably attracted to damaged hippocampal tissue in vitro andthat the depletion of CD34-expressing cells abolished areduction in neural damage, indicating that these cells wereparticularly involved in the protective action of MNCs (43).Separation to remove mature cells, e.g., granulocytes, prior totransplantation could decrease the cellular and cytokineinteractions that can lead to inflammation and increasedcapillary permeability at the site of transplantation.

Conclusions

We found evidence that BM CD34+ cells have the potential torespond to NTs by expressing their relevant receptors TrkA, TrkB,TrkC, p75NTR and that they are also able to synthesize NTs

157

(BDNF, NGF, NT-3, and NT-4) themselves. These cells showhigher levels of NTexpression compared to NCs. The spontaneousexpression of neurotrophins and neurotrophin receptors at themRNA and protein levels demonstrates the potential for theneurotrophic activity of BM CD34+ cells. However, a betterunderstanding of the trophic influence of CD34+ cells on neuraltissue following transplantation requires additional functionalsecretome studies using proteomic techniques and in vivoexperiments. The simplicity and safety of immunomagneticisolation, which has been confirmed in clinical settings, and thepotential to respond to neural stimuli and to produce neurotrophicfactors lays the foundation for stem cell-based regenerative therapyfor neurological diseases using CD34+ isolated from autologousBM as an alternative source of cells for transplantation purposes.

Acknowledgments: This work was supported by TheNational Centre for Research and Development grantSTRATEGMED1/234261/2NCBR/2014 (to BM).

E. Paczkowska and K. Piecyk equally contributed to this work.

Conflict of interests: None declared.

REFERENCES

1. Emberson J, Lees KR, Lyden P, et al. Effect of treatmentdelay, age, and stroke severity on the effects of intravenousthrombolysis with alteplase for acute ischaemic stroke: ameta-analysis of individual patient data from randomisedtrials. Lancet 2014; 384: 1929-1935.

2. Majka M, Janowska-Wieczorek A, Ratajczak J, et al.Numerous growth factors, cytokines, and chemokines aresecreted by human CD34(+) cells, myeloblasts,erythroblasts, and megakaryoblasts and regulate normalhematopoiesis in an autocrine/paracrine manner. Blood2001; 97: 3075-3085.

3. Bittencourt H, Rocha V, Chevret S, et al. Association ofCD34 cell dose with hematopoietic recovery, infections, andother outcomes after HLA-identical sibling bone marrowtransplantation. Blood 2002; 99: 2726-2733.

4. Chou T, Sano M, Ogura M, Morishima Y, Itagaki H, TokudaY. Isolation and transplantation of highly purified autologousperipheral CD34+ progenitor cells: purging efficacy,hematopoietic reconstitution following high dosechemotherapy in patients with breast cancer: results of afeasibility study in Japan. Breast Cancer 2005; 12: 178-188.

5. Fujita Y, Kinoshita M, Furukawa Y, et al. Phase II clinicaltrial of CD34+ cell therapy to explore endpoint selection andtiming in patients with critical limb ischemia. Circ J 2014;78: 490-501.

6. Walczak P, Chen N, Hudson JE, et al. Do hematopoietic cellsexposed to a neurogenic environment mimic properties ofendogenous neural precursors? J Neurosci Res 2004; 76:244-254.

7. Willing AE, Vendrame M, Mallery J, et al. Mobilized peripheralblood cells administered intravenously produce functionalrecovery in stroke. Cell Transplant 2003; 12: 449-454.

8. Taguchi A, Soma T, Tanaka H, et al. Administration ofCD34+ cells after stroke enhances neurogenesis viaangiogenesis in a mouse model. J Clin Invest 2004; 114:330-338.

9. Tsuji M, Taguchi A, Ohshima M, et al. Effects of intravenousadministration of umbilical cord blood CD34(+) cells in amouse model of neonatal stroke. Neuroscience 2014; 263:148-158.

10. Paczkowska E, Golab-Janowska M, Bajer-Czajkowska A, etal. Increased circulating endothelial progenitor cells in

patients with haemorrhagic and ischaemic stroke: the role ofendothelin-1. J Neurol Sci 2013; 325: 90-99.

11. Paczkowska E, Roginska D, Pius-Sadowska E, et al.Evidence for proangiogenic cellular and humoral systemicresponse in patients with acute onset of spinal cord injury.J Spinal Cord Med 2015; 38: 729-744.

12. Taguchi A, Nakagomi N, Matsuyama T, et al. CirculatingCD34-positive cells have prognostic value for neurologicfunction in patients with past cerebral infarction. J CerebBlood Flow Metab 2009; 29: 34-38.

13. Sobrino T, Arias S, Perez-Mato M, et al. CD34+ progenitorcells likely are involved in the good functional recovery afterintracerebral hemorrhage in humans. J Neurosci Res 2011;89: 979-985.

14. Moniche F, Gonzalez A, Gonzalez-Marcos JR, et al. Intra-arterial bone marrow mononuclear cells in ischemic stroke:a pilot clinical trial. Stroke 2012; 43: 2242-2244.

15. Piao CS, Gonzalez-Toledo ME, Gu X, Zhao LR. Thecombination of stem cell factor and granulocyte-colonystimulating factor for chronic stroke treatment in agedanimals. Exp Transl Stroke Med 2012; 4: 25. doi:10.1186/2040-7378-4-25

16. England TJ, Abaei M, Auer DP, et al. Granulocyte-colonystimulating factor for mobilizing bone marrow stem cells insubacute stroke: the stem cell trial of recovery enhancementafter stroke 2 randomized controlled trial. Stroke 2012; 43:405-411.

17. Bath PM, Sprigg N, England T. Colony stimulating factors(including erythropoietin, granulocyte colony stimulatingfactor and analogues) for stroke. Cochrane Database SystRev 2013; 6: CD005207.

18. Paczkowska E, Kaczynska K, Pius-Sadowska E, et al.Humoral activity of cord blood-derived stem/progenitorcells: implications for stem cell-based adjuvant therapy ofneurodegenerative disorders. PLoS One 2013; 8: e83833.

19. Machalinska A, Roginska D, Pius-Sadowska E, et al.Neuroprotective and antiapoptotic activity of lineage-negative bone marrow cells after intravitreal injection in amouse model of acute retinal injury. Stem Cells Int 2015;2015: 620364. doi: 10.1155/2015/620364.

20. Uemura M, Kasahara Y, Nagatsuka K, Taguchi A. Cell-basedtherapy to promote angiogenesis in the brain followingischemic damage. Curr Vasc Pharmacol 2012; 10: 285-288.

21. Chen SH, Wang JJ, Chen CH, et al. Umbilical cord blood-derived CD34+ cells improve outcomes of traumatic braininjury in rats by stimulating angiogenesis and neurogenesis.Cell Transplant 2014; 23: 959-979.

22. Racz GZ, Kadar K, Foldes A, et al. Immunomodulatory andpotential therapeutic role of mesenchymal stem cells inperiodontitis. J Physiol Pharmacol 2014; 65: 327-339.

23. Baumert B, Kawa MP, Kotowski M, et al. Bone marrow ofmultiorgan donors underutilized: implications forimprovement of accessibility of hematopoietic cells fortransplantations. Transplantation 2012; 93: 165-171.

24. Sakr HF, Abbas AM, El Samanoudy AZ. Effect of vitamin Eon cerebral cortical oxidative stress and brain-derivedneurotrophic factor gene expression induced by hypoxia andexercise in rats. J Physiol Pharmacol 2015; 66: 191-202.

25. Murawska-Cialowicz E, Wojna J, Zuwala-Jagiello J.Crossfit training changes brain-derived neurotrophic factorand irisin levels at rest, after wingate and progressive tests,and improves aerobic capacity and body composition ofyoung physically active men and women. J PhysiolPharmacol 2015; 66: 811-821.

26. Machalinski B, Lazewski-Banaszak P, Dabkowska E,Paczkowska E, Golab-Janowska M, Nowacki P. The roleof neurotrophic factors in regeneration of the nervous

158

system. [article in Polish] Neurol Neurochir Pol 2012; 46:579-590.

27. Rashedi I, Panigrahi S, Ezzati P, Ghavami S, Los M.Autoimmunity and apoptosis - therapeutic implications.Curr Med Chem 2007; 14: 3139-3151.

28. Sakr HF, Khalil KI, Hussein AM, Zaki MS, Eid RA,Alkhateeb M. Effect of dehydroepiandrosterone (DHEA) onmemory and brain derived neurotrophic factor (BDNF) in arat model of vascular dementia. J Physiol Pharmacol 2014;65: 41-53.

29. Johnson TV, DeKorver NW, Levasseur VA, et al.Identification of retinal ganglion cell neuroprotectionconferred by platelet-derived growth factor through analysisof the mesenchymal stem cell secretome. Brain 2014; 137:503-519.

30. Mead B, Logan A, Berry M, Leadbeater W, Scheven BA.Paracrine-mediated neuroprotection and neuritogenesis ofaxotomised retinal ganglion cells by human dental pulp stemcells: comparison with human bone marrow and adipose-derived mesenchymal stem cells. PLoS One 2014; 9:e109305. doi: 10.1371/journal.pone.0109305.

31. Chevalier S, Praloran V, Smith C, et al. Expression andfunctionality of the trkA proto-oncogene product/NGFreceptor in undifferentiated hematopoietic cells. Blood 1994;83: 1479-1485.

32. Laurenzi MA, Beccari T, Stenke L, Sjolinder M, Stinchi S,Lindgren JA. Expression of mRNAencoding neurotrophinsand neurotrophin receptors in human granulocytes and bonemarrow cells-enhanced neurotrophin-4 expression inducedby LTB4. J Leukoc Biol 1998; 64: 228-234.

33. Cattoretti G, Schiro R, Orazi A, Soligo D, Colombo MP.Bone marrow stroma in humans: anti-nerve growth factorreceptor antibodies selectively stain reticular cells in vivoand in vitro. Blood 1993; 81: 1726-1738.

34. Bracci-Laudiero L, Celestino D, Starace G, et al. CD34-positive cells in human umbilical cord blood express nervegrowth factor and its specific receptor TrkA. J Neuroimmunol2003; 136: 130-139.

35. Labouyrie E, Dubus P, Groppi A, et al. Expression ofneurotrophins and their receptors in human bone marrow.Am J Pathol 1999; 154: 405-415.

36. Li N, Yang H, Lu L, Duan C, Zhao C, Zhao H. Spontaneousexpression of neural phenotype and NGF, TrkA, TrkB genesin marrow stromal cells. Biochem Biophys Res Commun2007; 356: 561-568.

37. Schuhmann B, Dietrich A, Sel S, et al. A role for brain-derivedneurotrophic factor in B cell development. J Neuroimmunol2005; 163: 15-23.

38. Kadam SD, Chen H, Markowitz GJ, et al. Systemic injectionof CD34(+)-enriched human cord blood cells modulatespoststroke neural and glial response in a sex-dependentmanner in CD1 mice. Stem Cells Dev 2015; 24: 51-66.

39. Kermani P, Rafii D, Jin DK, et al. Neurotrophins promoterevascularization by local recruitment of TrkB+ endothelialcells and systemic mobilization of hematopoieticprogenitors. J Clin Invest 2005; 115: 653-663.

40. Moniche F, Montaner J, Gonzalez-Marcos JR, et al. Intra-arterial bone marrow mononuclear cell transplantationcorrelates with GM-CSF, PDGF-BB, and MMP-2 serumlevels in stroke patients: results from a clinical trial. CellTransplant 2014; 23 (Suppl. 1): 57-64.

41. Cordes AL, Jahn K, Hass R, et al. Intramedullary spinal cordimplantation of human CD34+ umbilical cord-derived cellsin ALS. Amyotroph Lateral Scler 2011; 12: 325-330.

42. Callera F, de Melo CM. Magnetic resonance tracking ofmagnetically labeled autologous bone marrow CD34+ cellstransplanted into the spinal cord via lumbar puncturetechnique in patients with chronic spinal cord injury: CD34+

cells’ migration into the injured site. Stem Cells Dev 2007;16: 461-466.

43. Boltze J, Reich DM, Hau S, et al. Assessment ofneuroprotective effects of human umbilical cord bloodmononuclear cell subpopulations in vitro and in vivo. CellTransplant 2012; 21: 723-737.

R e c e i v e d :September 3, 2015A c c e p t e d :January 20, 2016

Author’s address: Prof. Boguslaw Machalinski, Department ofGeneral Pathology, Pomeranian Medical University, 72Powstancow Wlkp. Street, 70-111 Szczecin, Poland.E-mail: machalin@pum.edu.pl

159