at SciVerse ScienceDirect

Biomaterials 34 (2013) 1657e1668

Contents lists available

Biomaterials

journal homepage: www.elsevier .com/locate/biomater ia ls

Stem cell recruitment and angiogenesis of neuropeptide substance P coupledwith self-assembling peptide nanofiber in a mouse hind limb ischemia model

Ji Hyun Kim a, Youngmee Jung a, Byung-Soo Kimb, Soo Hyun Kim a,c,*

aCenter for Biomaterials, Biomedical Research Institute, Korea Institute of Science and Technology, Seoul 136-791, Republic of Koreab School of Chemical and Biological Engineering, Seoul National University, Seoul 151-742, Republic of KoreacGraduate School, Korea University, Seoul, 136-705, Republic of Korea

a r t i c l e i n f o

Article history:Received 26 October 2012Accepted 8 November 2012Available online 1 December 2012

Keywords:Bioactive peptidesMesenchymal stem cellsAngiogenesisSubstance PIschemia

* Corresponding author. Biomedical Research Instituand Technology, L1216, 39-1, Hawolgok-dong, WoRepublic of Korea. Tel.: þ82 2 958 5343; fax: þ82 2 9

E-mail address: soohkim@kist.re.kr (S.H. Kim).

0142-9612/$ e see front matter � 2012 Elsevier Ltd.http://dx.doi.org/10.1016/j.biomaterials.2012.11.008

a b s t r a c t

For the successful treatment of ischemia, it is important to resupply sufficient blood into ischemicregions by inducing angiogenesis. Many stem cell transplantation studies have been reported to enhanceangiogenesis, especially those relating to mesenchymal stem cells (MSCs); however cell transplantationhas a number of limitations, such as the low rate of cell survival and donor cell shortage. In this study, wedeveloped bioactive peptides by immobilizing substance P into self-assembling peptides, and their MSCsrecruiting ability and therapeutic effects were evaluated by using ischemic hind limb models. Limbischemia was produced in athymic mice, and 1% (wt/vol) peptides were injected into ischemic sites(n ¼ 6 in each group: ischemia, substance P, RADA16-II, RADA16-II þ substance P, and RADA16-II þ RADA-SP (bioactive peptides)). The tissues were harvested for histological analysis and tissueperfusion measurement at 1, 3, 7, and 28 days after injection. We observed that bioactive peptidesassembled themselves (<10 nm nanofibers) and formed 3-dimensional (3D) microenvironments withinischemic regions. In the animal study, it was observed that by applying bioactive peptides, substance Pcontinued to be released at 28 days, and consequently, MSCs were successfully recruited into ischemicregions. Bioactive peptides could prevent fibrosis, promote neovascularization, enhance tissue perfusion,and prevent limb salvages. Our results demonstrated that bioactive peptides are one of the mostpowerful tools for the treatment of ischemia, through their recruitment of autologous MSCs andpromotion of angiogenesis without cells transplantation.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Ischemic tissue disease remains one of the foremost causes ofmorbidity and mortality worldwide [1]. Critical limb ischemia isa particularly common injury, which is caused by poor circulationto lower limbs, and it is estimated to have an associated 20e40%amputation [2]. Large numbers of patients are not eligible forrevascularization by surgical or percutanous procedures. Therefore,it is important to supply sufficient blood to ischemic regions byregenerating the blood vessels. The use of a number of angiogenicfactors, such as growth factors, cytokines, various types of stromal/stem/progenitor cells, have been shown to have therapeutic effectsin ischemic diseases [3].

te, Korea Institute of Sciencelsong-gil 5, Seoul 136-791,58 5308.

All rights reserved.

Stem cell transplantation is currently generating a great deal ofinterest in the treatment of ischemic diseases. Among the varioustypes of stem/progenitor cells, mesenchymal stem cells (MSCs) arepotential therapeutic cells for a wide variety of tissue damages anddiseases. Numerous transplantation studies of MSCs, includingbone marrow-derived MSCs, showed tissue repair for myocardialinfarction, bleomycin-induced lung injuries, liver injury and skininjuries, either by differentiating into many different cell types orby their trophic effects [4]. MSCs have previously been demon-strated to cause angiogenesis. Wu YJ et al. demonstrated that bonemarrow-derived MSCs enhanced wound healing through differ-entiation and release of pro-angiogenic factors [5]. However, celltransplantation often encounters problems, such as low rates of cellsurvival, tumorigenesis, and donor cell shortage [6]. Although thefundamental principles of cell-based therapies have been success-fully applied clinically, it is essential to obtain appropriate cellsources for conventional tissue engineering applications. In thecase of MSCs, they are usually obtained by aspirating bone marrowfrom the ileac crest bone, which does not yield sufficient amounts

J.H. Kim et al. / Biomaterials 34 (2013) 1657e16681658

of MSCs for the wide variety of potential applications to which theyare suited. Therefore, more efficient strategies to retrieve thera-peutic MSCs from bone marrow are anticipated [7].

To overcome such obstacles of stem cell therapy, there isanother promising approach, where the recruitment of circulatinghost cells to the site to be regenerated is enhanced in vivo [6]. In situtissue regeneration, which induces mobilization of host stem cellsand controls cell differentiation into tissue-specific cell types, isa new concept in the field of tissue engineering [8,9]. Unlike cell-based tissue engineering, this strategy does not include donor cellprocurement and in vitro cell manipulation [10]. For efficient andsuccessful in situ tissue regeneration, appropriate numbers of hoststem cells must be recruited. Therefore, exogenous granulocytecolonyestimulating factor (G-CSF) or granulocyte-macrophagecolony-stimulating factor (GM-CSF) has been applied to mobilizehost stem cells to injured sites for tissue repair. Orlic D et al. re-ported that bone marrow stem cells were mobilized along withhematopoietic stem cells (HSCs) and endothelial precursor cells(EPCs) in reaction to exogenous G-CSF, and differentiated into car-diomyocytes in the damaged tissue of myocardial infarction [11].However, this application is limited for EPC mobilization inischemic tissue damages because of a strong inflammatoryresponse [12].

Recently, a new role of substance P has been reported, asa systemically acting injury-inducible wound messenger thataccelerates wound healing by mobilizing stromal-like cells [4].Substance P is an 11eamino acid neuropeptide secreted from theperipheral terminals of sensory nerve fibers, where it acts asa neurotransmitter or hormone. Substance P mediates painperceptions, neuroimmune modulation, bone marrow fibrosis,tumor cell proliferation and enhanced proliferation and differen-tiation of endothelial cells (ECs), all of which are expected from itslocal action, direct nerve innervation and direct cellular contacts[4,6,13,14]. In addition to its local action, Hong HS et al. demon-strated its systemic action in thewound healingmodel of the alkali-burned corneal injury that appears to recruit CD29þ stromal-likecells from the periphery to the site of injury, resulting in acceler-ated wound healing [4]. Using this role of substance P, Ko IK et al.developed a delivery method which was used to enhance recruit-ment of endogenous stem cells such as MSCs and HSCs intoimplanted scaffolds, and to subsequently increase angiogenesis[15]. Kohara H et al. confirmed that the local release of substance Pachieved the recruitment of circulating cells from blood flowaround the site released, resulting in enhanced angiogenesis [6]. Incontrast to G-CSF, substance P does not evoke any general phar-macology action or genetic toxicity [7]. Taken together, substance Pis one of the factor candidates which can induce angiogenesis andtissue repair through an enhanced recruitment of autologous stemcells.

Self-assembling peptides are promising biomaterials in tissueengineering and regenerative medicine [16,17]. This injectablebiodegradable material forms fibers (<10 nm) and assembles intoa 3-dimensional (3D) scaffold in the physiological environment.The scaffolds closely mimic the porosity and gross structure ofextracellular matrices, allowing cells to reside and migrate in themicroenvironment, supporting cell attachment and differentiationof a variety of cells [17e21]. Self-assembling peptides are composedof synthetic amino acids, and unlike other natural hydrogelsderived from animals, they can be easily designed and modified ina variety of ways, making it is possible to control degradation rateswith non-immunogenicity. Self-assembling peptides can beapplied for protein delivery [22]. In previous study, we reportedthat self-assembling peptides contributed to stability of growthfactors and enabled the controlled delivery of growth factors intoischemic regions, leading to improved angiogenesis for the

treatment of myocardial infarction [23]. Since self-assemblingpeptides can be designed with great flexibility, various modifica-tions of self-assembling peptides have been proposed to deliverindividual proteins or biologically active molecules such as incor-porating growth factors, RGDs or YIGSR [16,17,24].

In this study, we developed a bioactive peptide which has theability to recruit MSCs bymodifying self-assembling peptides usingsubstance P sequence. The therapeutic effects of the bioactivepeptides were evaluated in ischemic hind limb models by injectingthem into ischemic regions. The structural characteristics ofbioactive peptides were analyzed, and their degradation rates aswell as their sustained delivering ability of substance P and MSCsrecruiting ability were investigated. In long term animal studies,the effects of ischemic tissue repair, neovascularizaion and tissueperfusionwere estimated. By using bioactive peptides, we intendedto improve MSC recruiting activity with increasing stability andlong term release of substance P into ischemic sites. We postulatedthat the injection of suitable self-assembling peptides into ischemicregions would create 3-dimensional microenvironments, subse-quently improve infiltration and attachment of recruited MSCswithin them, and promote neovascularization.

2. Materials and methods

2.1. Bioactive peptides synthesis and preparation

The peptides RADA16-II (AcN-RARADADARARADADA-CONH2), RADA-SP (AcN-RARADADARARADADAGGRPKPQQFFGLM-CONH2) and biotinylated RADA 16-II(biotin-RADA16-II) were synthesized and purified to 95% (Peptron, Daejeon,Korea). For bioactive peptides hydrogel, RADA16-II and RADA-SP were mixed(200:1). These peptides were dissolved in sterile sucrose at 1% (wt/vol) and weresonicated for 30 min [23,24].

2.2. Structural characterization

For transmission electron microscopy (TEM), 3 samples were diluted in distilledwater to a working concentration of 0.01%. After staining with 2% uranyl acetate,TEM images were acquired (CM30, Philips, Amsterdam, Netherlands). For circulardichroism (CD) measurements, all samples (RADA16-II, RADA-SP, and RADA16-II þ RADA-SP) were prepared by diluting stock peptide solutions in distilled waterto working concentrations of 0.02% (wt/vol) that were incubated at room temper-ature overnight. CD data were collected on a J-715 spectrophotometer (JASCO, Inc.,Easton, MD) using a quartz cuvettewith a 1mm path length (Hellma Standard Cuvet110QS (Quartz Suprasil), Hellma Analytics, Müllheim, Germany) and each spectrumwas repeated three times [16,25].

2.3. Hind limb ischemia and injection of peptides

All animals were treated in accordance with the “Recommendations forHandling of Laboratory Animals for Biomedical Research” compiled by theCommittee on the Safety and Ethical Handling Regulation for Laboratory Experi-ments at Korea Institute of Science and Technology and Seoul National University.Hind limb Ischemia was induced in 5 week old male BALB/c-nu Slc mice (22e24 g).Mice were anesthetized with xylazine and zoletil and the femoral artery and itsbranches were ligated and excised [26,27]. To ensure the successful establishedmodel for hind-limb ischemia, the ratio (ischemic/non ischemic limb) was imme-diately verified by a laser doppler perfusion imaging (LDPI, Moor Instruments,Devon, United Kingdom). One day after surgery, the mice were randomly dividedinto five experimental groups (total 120 mice): sucrose (IS), Substance P (SP),RADA16-II (R), RADA16-II þ SP (R þ SP) and RADA16-II þ RADA-SP (R þ RSP,bioactive peptides). The materials (200 ml) were injected into ischemic zones at 4locations. Total amount of substance P (Substance P, Merck, Darmstadt, Germany)was 5 mg in SP, R þ SP, and R þ RSP groups. To detect the injected peptide and toconfirm the peptide microenvironment, a biotinylated peptide was injected (0.1% ofthe total peptide was biotinylated). After surgery, the animals were allowed torecover under a heating pad. Ischemic limb muscles were harvested 1, 3, 7, and 28days after injection (n ¼ 6 in each treatment and each duration, total n ¼ 120).

2.4. MSCs recruitment

To investigate the MSCs recruiting activity of bioactive peptides, samples werestained with MSC markers (CD29, CD90, and CD105) [4]. For histological analysis,tissues were fixed with 10% buffered formalin, embedded in paraffin, and sectionedinto slices with 5 mm thickness. At day 1, 3 and 7, tissues were double stained with

J.H. Kim et al. / Biomaterials 34 (2013) 1657e1668 1659

CD29 (mouse anti-human integrin beta1 monoclonal antibody, Millipore, Temecula,CA, 1:1000) and CD90 (Anti-CD90/Thy1 antibody, Abcam, Cambridge, UnitedKingdom, 1:250) and also CD105 (Anti-Endoglin monoclonal antibody, Millipore,1:500) and CD90 (�200 magnification, n ¼ 6 in each group). Alexa Fluor 488 goatanti-mouse IgG (Invitrogen, Eugene, Oregon, 1:500) and Alexa Fluor 594 goat anti-rabbit IgG (Invitrogen, 1:1000) were used for secondary antibodies. Raters did thiswithout knowledge of treatment groups.

2.5. Sustained delivery test of substance P

To assess the feasibility of using bioactive self-assembling peptides for sustaineddelivery of substance P, retained substance P was stained with substance P antibody(substance P antibody, Santa Cruz Biotechnology, INC., CA, 1:500) 1, 3, 7 and 28 daysafter treatment (n ¼ 6 in each group). After staining with anti-Substance P antibody,remained substance P (mm2/mm2) was quantitatively analyzed by image analysissystem (ImageJ 1.43, National Institute of Mental Health, Bethesda, MD, Image Proversion 4.5, Media Cybernetics, Silver Spring, MD) in 5 different fields (�200magnification). Raters did this without knowledge of treatment groups.

2.6. Peptides degradation measurement

Degradation tendency of peptides was estimated by staining of biotin-RADA16-II [28]. In R, R þ SP, and R þ RSP groups, biotin-RADA16-II was stained withDyLightTM 594 conjugated Streptavidin (Thermo scientific, Rockford, IL, 1:500).Peptides which were retained in ischemic tissues (mm2/mm2) were quantitativelyanalyzed by image analysis system in 5 different fields (�200magnification, n¼ 6 ineach group). Raters did this without knowledge of treatment groups.

2.7. Fibrosis

To characterize the fibrosis of ischemic limb muscles, tissues were prepared asdescribed previously, and were stained with Masson’s trichrome [27]. Fibrotictissues of all groups at 7 and 28 days after treatment were observed (�100magnification, n ¼ 6 in each group). The area of fibrosis in 10 different fields wasanalyzed as the mean per unit area (1 mm2). Raters were again blinded to anyknowledge of the treatment groups.

2.8. TUNEL assay

Apoptosis of ischemic limb muscles was detected by DeadEndTM RluoromericTUNEL System (Promega, Madison, WI) 7 and 28 days after injection. The number ofTUNEL-positive nuclei and the total number of nuclei in the 5 different fields (�400magnification) were counted by a blinded rater in the border zone (n ¼ 6 in eachgroup). TUNELþ cell density (%) is expressed as the ratio of TUNEL-positive nuclei tothe total number of nuclei [23].

2.9. Angiogenesis

To evaluate vascular cells density 7 and 28 days after injection, ECs and vascularsmooth muscle cells (SMCs) were stained with polyclonal rabbit anti-human vonWillebrand factor antibody (vWF, DAKO, Glostrup, Denmark, 1:100) and monoclonalmouse anti-human a-smooth muscle actin antibody (a-SMA, DAKO, 1:100),respectively. The vWF and a-SMA positive vessels in 5 random fields (�200magnification) were evaluated, and vWFþ cell density and a-SMAþ vessel densitywere quantified (n ¼ 6 in each group) [23]. The area of blood vessels was presentedas the mean per unit area (1 mm2). The maturation index was quantified as the ratioof a-SMA-positive vessels to the total number of vessels [23]. Raters were blinded toany knowledge of the treatment groups.

2.10. Laser Doppler perfusion image and limb salvage score

Tissue perfusion was measured via LDPI on postoperative day 3, 7, 14, 21, and 28after treatment to evaluate blood flow in both the ischemic and non-ischemic limbs.After surgery, LDPI was performed to confirm the successful established model forhind-limb ischemia, as described above. The perfusion unit (PU) was calculated bythe blood flow ratio of ischemic to non-ischemic limbs [26]. All LDPI measurementswere taken on a 37 �C heating pad. Additionally, tissue damage in the ischemic limb(limb salvage score) was graded as full recovery (grade 6), minor necrosis or nail loss(grade 5), partial toe amputation (grade 4), total toe amputation (grade 3), partial/total foot amputation (grade 2), or partial/total limb amputation (grade 1) bya blinded rater [26].

2.11. Statistical analysis

Results were analyzedwith OriginPro 7.5 (OriginLab Co, Northampton, MA, USA)and SPSS software (SPSS, version 17; SPSS Inc, Chicago, IL). Continuous variables areexpressed as mean � standard deviation, and are compared between any twogroups using student’s t-test, and two-tailed probability values of p < 0.05 wereconsidered statistically significant. Multiple comparisons were performed using

one-way analysis of variance (ANOVA) followed by post hoc test of Dunnett’s T3 orbonferroni.

3. Results

3.1. Characterization of bioactive peptides

The structural characterization of bioactive peptides wasanalyzed using TEM and CD. TEM imaging showed that RADA16-IIformed nanofibers of 5e10 nm (Fig. 1A); hence, RADA-SP did notshow any structural characteristics (Fig. 1B). In the case of bioac-tive peptides (RADA16-II þ RADA-SP, 200:1), it formed a fibrousnanostructure, the structure of which was similar to RADA16-II(Fig. 1C). CD spectra were obtained for each peptide to investi-gate their secondary structure (Fig. 1D). A molecular ellipticity(deg cm2/decimole) of RADA16-II observed a typical spectrum forb-sheet structures in the 216 nm and 195 nm regions [25,29]. Thebioactive peptides exhibited similar structural properties to theRADA16-II sequences. However, RADA-SP observed a spectrum forrandom structures. Therefore, bioactive peptides as well asRADA16-II were self-assembled by forming stable b-sheets intonanofibers.

3.2. MSCs recruitment of bioactive peptides

The MSCs recruiting ability of bioactive peptides was observedin ischemic regions at 1, 3, and 7 days after treatments. In thisstudy, we hypothesized that CD29-CD90 or CD105-CD90 doublestained cells were MSCs. As the results of qualitative analysis(Fig. 2 and Supplemental Fig. 1), all groups showed an increasingtendency of double stained cells as over time. The IS group showeda few double stained cells in ischemic regions at each time pointdetermined. In SP and R groups, the number of double stainedcells was higher than that in the IS group. In the R þ SP andR þ RSP groups, many double stained cells were also observed,and the R þ RSP group showed the highest quantities of doublestained cells of all the groups at each time point. Therefore,bioactive peptides were the most effective to recruit MSCs intoischemic regions.

3.3. Sustained delivery test of substance P

After staining with anti-substance P antibody, retainedsubstance P in injected sites was quantitatively analyzed at 1, 3, 7and 28 days after treatment (mm2/mm2, n ¼ 6 in each group, Fig. 3and Supplemental Fig. 2). The quantity of substance P in the ISgroup increased at 3 days and then decreased at 7 days(7218.38 � 811.98 mm2/mm2 at 3 days in IS group, SupplementalFig. 2A); in the other groups, the amount of substance Pdecreased gradually. Substance P in the SP group decreased to 50%of the amount of that in R þ SP and R þ RSP groups at 1 days, andonly 4.75% of substance P was remained within injected sites at 28days, which was not statistically different from that of the ISgroup (1064.08 � 486.58 mm2/mm2 in SP group at 28 days,p > 0.05); however, relatively large amounts of substance P weredetected in the R þ RSP and R þ RSP groups at 28 days. Especially,at 28 days, 25.19% of substance P remained at injected sites in theR þ RSP group compared to 1 day, and it was 3.03 times higherthan that in R þ SP group (18791.45 � 1687.87 mm2/mm2 inR þ RSP group and 6200.37 � 576.41 mm2/mm2 in R þ SP group at28days, p < 0.05). Unexpectedly, substance P in R group showedno statistically different from SP group at 1 day, and was higher at3, 7, and 28 days (35405.55 � 3133.23 mm2/mm2 in R group and33751.54 � 2387.63 mm2/mm2 in SP group at 1 day, p > 0.05).

Fig. 2. Mesenchymal stem cells (MSCs) recruitment of bioactive peptides. Bioactive peptides are effective for the recruitment of MSCs into ischemic regions. Representative imagesof ischemic regions from each group after CD29, CD90, and CD105 staining at 7 days after injection. (A) CD29-CD90 and (B) CD105-CD90. Scale bars: 100 mm (�200).

Fig. 1. Structural characteristics of bioactive peptides. Bioactive peptides are self-assembled by forming stable b-sheets into nanofibers. (A) Transmission electron microscopy (TEM)image of RADA16-II. (B) TEM image of RADA-SP. (C) TEM image of bioactive peptides (RADA16-II þ RADA-SP). (D) Circular dichroism (CD) of each peptide.

J.H. Kim et al. / Biomaterials 34 (2013) 1657e16681660

Fig. 3. Sustained delivery of substance P and degradation test of bioactive peptides. Injection of bioactive peptides can provide sustained delivery of substance P for 28 days. (AeF)Representative pictures of injected regions from each group by staining of substance P and biotin-RADA16-II at 7 and 28 days after injection. (A) Normal, (B) IS, (C) SP, (D) R, (E)R þ SP, and (F) R þ RSP group. Scale bars: 100 mm (�200). (G) Quantification of retained substance P of each group (mm2/mm2). *p > 0.05 with 1 day, þp > 0.05 with 3 days, zp > 0.05with SP group. (H) Quantification of peptides degradation rates of each group (mm2/mm2). *p < 0.05 with 1 day, þp < 0.05 with 3 days, zp < 0.05 with 7 days.

J.H. Kim et al. / Biomaterials 34 (2013) 1657e1668 1661

These results indicate that bioactive peptides are the most effec-tive for the sustained delivery of substance P into ischemic region.

3.4. Peptide degradation measurement

To investigate peptide degradation at ischemic regions, biotin-RADA16-II in R, R þ SP and R þ RSP group was quantitativelyanalyzed (mm2/mm2, n ¼ 6 in each group). As shown in Fig. 3 andSupplemental fig. 2, peptides were gradually degraded in all groupsand remained until 28 days. At that time, 18% of peptides remainedat ischemic regions in all groups compared to 1 day and there wasno statistical difference among groups (Fig. 3H).

3.5. Fibrosis

Histological examination of the ischemic hind limbs were con-ducted 7 and 28 days after treatment to observe fibrotic tissueformations (Fig. 4). In the IS group, fibrotic tissues were observed inbroad regions at 7 days (0.65� 0.05mm2/mm2). Fibrotic tissueswereobserved in the R and SP groups, even though they were reducedcompared to those of the IS group. In the Rþ SP and Rþ RSP groups,fibrotic tissue formation was markedly attenuated, and especially inR þ RSP group, little was observed which were similar to normaltissues at 28 days (0.07 � 0.03 mm2/mm2 at 7 days and0.02�0.01mm2/mm2at28days in theRþRSPgroup,p<0.05, Fig.4B

Fig. 4. Fibrotic tissue formation of ischemic limbs. Injection of bioactive peptides prevents fibrotic tissue formations 7 and 28 days after treatment. (A) and (B) Representativepictures of ischemic limbs from each group after Masson’s trichrome staining at 7 and 28 days, respectively. Scale bars: 100 mm (�100). (C) Quantification of fibrosis (mm2/mm2).*p > 0.05 with SP group, þp < 0.05 with 7 days.

J.H. Kim et al. / Biomaterials 34 (2013) 1657e16681662

and C). These results indicate that the injection of bioactive peptidesprotects limb muscles against necrotic damage caused by ischemia.

3.6. TUNEL assay

Apoptosis of ischemic limbmuscles (TUNELþ cell density (%))wasinvestigated in each groups by using TUNEL assay (Fig. 5, n ¼ 6 ineach group). The percentage of apoptotic cells in the IS group was21.47 times increased compared to normal limbs at 7 days(44.23 � 3.09% in IS group and 2.06 � 1.45% in normal, p < 0.05). Incontrast, cell apoptosis was decreased to 59.26% in the SP group and62.88% in the R group of the IS group, and there was no significantdifferencebetween the SPandRgroups at 7days (26.21�2.76% in SPgroup and 27.81 � 2.30% in R group, p > 0.05). Moreover, cellapoptosiswas decreased to 31.61% in the Rþ SP group and 13.93% inthe R þ RSP group of IS group, and it was 2.27 times higher in theR þ RSP group than in the R þ SP group at 7 days (13.98 � 1.33% inRþ SP group and 6.16�1.55% in Rþ RSP group, p< 0.05). Therewasno difference between day 7 and day 28 data for cell apoptosis ofeach group. Injection of bioactive peptides was themost effective inpreventing cell apoptosis, showing a significant difference fromMI.

3.7. Capillary density, arterial density and maturation index

To evaluate the angiogenic activity of bioactive peptides, vWFþ

cell density (mm2/mm2), a-SMAþ vessel density (mm2/mm2), and

maturation index (%) of each group were analyzed in ischemicregions 7 and 28 days after treatments (Fig. 6, n ¼ 6 in each group).The vWFþ cell density of SP and R groups increased to 6.67 and 7.09times that of IS group at day 7 (80997.79 � 8411.52 mm2/mm2 in Rgroup, 76272.93 � 15338.48 mm2/mm2 in SP group, and11431.01 �8888.81 mm2/mm2 in IS group, p > 0.05 for R versus SP).The value for Rþ SP and Rþ RSP groups increased to 11.12 and 15.62times that of IS group, respectively (127128.14� 6870.72 mm2/mm2

in R þ SP group and 178590.32 � 13979.69 mm2/mm2 in R þ RSPgroup). Especially the capillary density of RþRSP group increased to2.34 times that of SP group, 2.20 times that of R group, and1.40 timesthat of Rþ SP group. The values at 28daysweredecreased comparedto the values at 7 days, except for the IS and R þ RSP group. TheR þ RSP group was 3.94 and 5.09 times higher than the R and SPgroup at 28 days, respectively (155466.26 � 27577.54 mm2/mm2 inR þ RSP group, 39468.16 � 3164.49 mm2/mm2 in R group, and30541.18 � 2318.44 mm2/mm2 in SP group). Especially, this groupshowed a 1.85 fold increase compared to R þ SP(83813.86 � 6497.43 mm2/mm2 in R þ SP group).

The a-SMAþ vessel density for R þ RSP group showed thehighest among all groups which increased to 1.54 and 23.15 timesthat of R þ SP and IS groups, respectively (55747.07� 3658.37 mm2/mm2 in R þ SP group and 3698.89 � 694.71 mm2/mm2 in IS group).The value for R and SP groups increased to 11.09 and 9.49 times thatof IS group, respectively (41009.78 � 2757.26 mm2/mm2 in R groupand 35109.11 � 3061.26 mm2/mm2 in SP group). At 28 days, the a-

Fig. 5. Cells apoptosis of ischemic limbs. Injection of bioactive peptides decreases TUNELþ cells apoptosis 7 and 28 days after treatments. (A) and (B) Immunostaining of T EL staining in the ischemic regions at 7 and 28 days,respectively. Apoptotic cells: green. Scale bars: 50 mm (�400). (C) The ratio of TUNEL-positive to total nuclei (expressed as a %) was quantified at 7 and 28 days. *p > 0.05 with S group. (For interpretation of the references to colour inthis figure legend, the reader is referred to the web version of this article.)

J.H.K

imet

al./Biom

aterials34

(2013)1657

e1668

1663

UNP

Fig. 6. Capillary density, arterial density and maturation index. Injection of bioactive peptides improves capillary density, arterial density and the maturation index. (A) and (B)Representative immunostaining of endothelial cells (ECs) and smooth muscle cells (SMCs) of each group 7 and 28 days after treatment, respectively. In immunofluorescence, ECs arestained with von Willebrand factor (vWF, green) and SMCs are stained with a-smooth muscle actin (a-SMA, red). Scale bars: 100 mm (x200). (C) Quantification of vWFþ cell density(mm2/mm2). (D) Quantification of a-SMAþ vessel density (mm2/mm2). (E) Maturation index (%). Quantification of the maturation index was based on comparison of the percentage ofa-SMAþ vessels to the total number of vessels. *p > 0.05 with SP group, þp > 0.05 with 7 days. (For interpretation of the references to colour in this figure legend, the reader isreferred to the web version of this article.)

J.H. Kim et al. / Biomaterials 34 (2013) 1657e16681664

SMAþ vessel density for R þ RSP group showed no differences fromday 7 data, but the values for Rþ RSP, R, and SP groups were slightlydecreased (88226.90 � 9050.09 mm2/mm2 in R þ RSP group at 28days). Moreover, it increased to 1.77 and 3.32 times that of R þ SPand SP group, respectively (49815.27� 3631.88 mm2/mm2 in Rþ SPgroup and 26606.92 � 3989.93 mm2/mm2 in SP group at 28 days).

To determine the percentage of matured vessels, the maturationindex was calculated based on double staining results at 28 days. Asshown in Fig. 6E 95.63� 2.95% for Rþ RSP group and 84.44� 2.14%for Rþ SP group were observed, and Rþ RSP group was the highestamong all groups. For SP and R groups, the values were70.86 � 1.68% and 71.71 � 1.58%, respectively, but there was nodifference (p < 0.05). Based on these data, injection of bioactivepeptides is the most effective way to increase capillary density,arterial density, and maturation index.

3.8. Laser Doppler perfusion image and limb salvage score

In order to estimate the recovery of blood flow, the perfusionratio was calculated by using LDPI immediately after surgery, and 3,7, 14, 21, and 28 days after treatments (Fig. 7 and SupplementalFig. 3, n ¼ 6 in each group). Immediately after inducing ischemia,

the mean perfusion ratio was 0.04 � 0.04. The perfusion ratio for Rand SP groupwas gradually increased compared to the IS group andthe value at 28 days was 0.33 � 0.03 and 0.33 � 0.02, respectively(p > 0.05). The value for R þ SP group was 0.35 � 0.04 at 7 days(0.16� 0.03 in IS group at 7 days, p< 0.05 for Rþ SP group versus ISgroup) and was 1.88 times higher than that of the IS group at 28days (0.24 � 0.03 in IS group and 0.45 � 0.01 in R þ SP group,p < 0.05). The perfusion ratio for R þ RSP group showed a signifi-cant difference to other groups at 3 days, and increased to 3.13times that of IS group, 2.26 times that of R and SP group, and 1.67times that of R þ SP group at 28 days, respectively (0.74 � 0.08 inR þ RSP group at 28 days).

To estimate the therapeutic potential of bioactive peptides, thelimb salvage score was determined by physical observation ofischemic limbs (Fig. 7A and C, n¼ 6 in each group). In IS group, totalfoots or partial limbs of all mice were amputated at 28 days(1.83� 0.41 in IS group at 28 days). The limb salvage score for R andSP groups was 2.67 � 0.52 and 2.33 � 0.52, respectively, and therewas no statistical difference. The value for R þ SP group was5.00 � 1.26 and showed minor necrosis or toe amputation in 3mice. Remarkably, all mice of the R þ RSP group were fully recov-ered even though there was no significant difference in the limb

Fig. 7. Laser doppler perfusion image (LDPI) and limb salvage score. Injection of bioactive peptides enhances tissue perfusion that lead to full recovery of ischemic hind limb. (A) Representative pictures of ischemic limbs and LDPI fromeach group at 28 days, respectively. (B) Perfusion ratio of each group. (C) Limb salvage score of each group. *p < 0.05 with IS group, þp < 0.05 with SP group, zp < 0.05 with R group, xp < 0.05 with R þ SP grosup.

J.H.K

imet

al./Biom

aterials34

(2013)1657

e1668

1665

J.H. Kim et al. / Biomaterials 34 (2013) 1657e16681666

salvage scores, compared to the R þ SP group (Supplemental Fig. 3,6.00 � 0.00 in R þ RSP group in 28 days).

4. Discussion

It is important to supply sufficient blood flow into ischemicregion by forming stable vessel networks for treatment of ischemicdiseases. There are various studies the treatment of ischemicdisease by using stem cells, but some obstacles remain, such as lowtherapeutic effects and donor cell shortage [6]. One promisingsolution in the field of cell therapy for tissue repair is the utilizationof an endogenous wound healing mechanism, which is the body’sown ability for mobilization of stromal-like or stem cells, which aresubsequently recruited to the injured tissue. In this study, injectablebioactive peptides which can recruit MSCs were developed andwere applied to the treatment of ischemic hind limbs without theneed for transplantation of cells. By covalently immobilizingsubstance P sequences to self-assembling peptides, substance Pwas released over a longer-term, and subsequently, MSCs wereefficiently recruited into ischemic regions. Moreover, bioactivepeptides provided 3D microenvironments by assembling nano-fibrous structures. Therefore, bioactive peptides showed increasedrecruitment of MSCs, preventing fibrosis, and promoting neo-vascularization with high tissue perfusion, which led to muchimproved therapeutic effects of ischemic hind limb.

First of all, structural analyzes of bioactive peptides was per-formed using TEM and CD spectra. In TEM images, we confirmedthat RADA16-II formed nanofibrous structures. Since self-assembling peptides are composed of short amino acid chains ofalternating ionic hydrophilic and hydrophobic amino acids, self-assembling peptides form stable b-sheet structures and undergomolecular self-assembly into nanofibers (5e10 nm) throughhydrophobic van der Waals and ionic interactions [17e21,25,29].Unlike RADA16-II, RADA-SP showed no specific structure. It isspeculated that substance P sequences, which are bonded toRADA16-II, interrupt RADA-SP to form b-sheet structures. On theother hand, bioactive peptides formed nanofibers with stable b-sheet similar to RADA16-II. It means that 0.5% RADA-SP does notinterrupt bioactive peptides molecular self-assembly into nano-fibers. Additionally, it does not affect degradation rates of bioactivepeptides. Peptide degradation rates for R þ RSP group showed nodifference to R þ SP and R groups, and peptides remained until 28days in the injected sites.

This study confirmed that bioactive peptides recruited MSCsefficiently in vivo. SP, Rþ SP, and Rþ RSP group weremore effectiveto MSCs recruiting than IS groups at all time points determined.These results confirm that substance P promotes MSCs recruitmentinto injected sites as reported by Hong HS et al., even thoughsubstance P was physically adsorbed (R þ SP) or covalently bonded(R þ RSP) [4,15]. Moreover, in R þ SP and R þ RSP group, a largenumber of MSCs was observed compared to that of SP group.Especially, R þ RSP group was the most effective in MSCs recruit-ment due to the sustained delivery of substance P. Since water-soluble factors are rapidly excreted from the injected site and areoften enzymatically digested or deactivated [30,31]. Therefore, toincrease protein retention in tissue, local sustained delivery oftherapeutic proteins is important [22]. Since substance P is gener-ally unstable in the body, it is a prime necessity to developa delivery system to allow the factor to function in vivo for angio-genesis. In particular, Kohara H et al. reported that the controlledrelease of substance P was achieved by hydrogels prepared from ananionic gelatin derivative, resulting in enhanced recruitment ofgranulocytes and induced angiogenesis [6]. In our previous study,self-assembling peptides could successfully deliver dual growthfactors into injected sites after 28 days in vivo [23]. In this study,

retention and sustained delivery of substance P was improved until28 days, by physically adsorbing and covalently binding to self-assembling peptides. Especially, bioactive peptides covalentlyimmobilized substance P, increasing stability and diminishing thediffusion of substance P. Consequently, bioactive peptides releasedsubstance P over a longer term, which led to increased MSCsrecruitment. Unexpectedly, substance P and MSCs were detected inR group. Some factors of RADA16-II might stimulate substance P,and so it has a MSCs recruiting ability, even though the mechanismhas not been clarified.

Injection of bioactive peptides was the most effective atprevention of fibrosis and cells apoptosis in the ischemic limbmodel. It is anticipated that this was due to the structural proper-ties of bioactive peptides.We confirmed that bioactive peptides hadnanofibrous structures and formed 3D microenvironments inischemic regions. It is known that 3D microenvironments createdby self-assembling peptides are closely mimic the porosity andgross structure of extracellular matrices and are highly hydrated,>99% in water [19,20,28]. They can support cell attachment,migration, and differentiation of a variety of cells and inhibit scartissue formation [21,29,32]. Among the various types of self-assembling peptides, the RADA16-II used in bioactive peptides issuperior for infiltrating vascular cells, inhibiting ECs apoptosis, andpromoting long-term survival relative to other hydrogels [18,28].Ellis-Behnke RG et al. and Guo J et al. reported that self-assemblingpeptides enabled the regeneration of axons by preventing scartissue formation in damaged brain tissue [21,32]. Yoshimi R et al.reported that MSCs maintained their own potential by increasingattachment and survival rates of transplanted cells in the 3Dmicroenvironment using self-assembling peptides [33]. Addition-ally, Koon HW et al. reported that substance P possessed anti-apoptotic effects by activating Akt, which prevents apoptosis andmediates tissue recovery [34]. With these anti-apoptotic effects ofsubstance P, it is anticipated that paracrine factors secreted fromrecruited MSCs reduced cells apoptosis. Therefore, 3D microenvi-ronment formation and substance P retention by bioactive peptidescontributed to reduced fibrosis formation and apoptosis inischemic limbs.

Bioactive peptides developed in this study are one of thecandidates which can induce angiogenesis through an enhancedrecruitment of MSCs, with therapeutic potential for ischemia. InR þ RSP group, vWFþ cell density and a-SMAþ vessel density was44.63 and 27.83 times greater than that of the IS group, respec-tively. Remarkably, this was the only group for which increasedvascular density was maintained until 28 days with a high matu-ration index up to 95%. Moreover, tissue perfusion was tremen-dously improved and ischemic limbs were fully recovered in allmice, as demonstrated by the limb salvage score. The improvementin tissue perfusion and limb salvage score is due to fibrosisprevention as well as the recovery of blood flow with mature andfunctional vessels of bioactive peptides. We speculated that thesebeneficial effects of bioactive peptides result from (i) their MSCsrecruiting ability with angiogenic activity, (ii) 3D microenviron-ments provided by self-assembling peptides, and (iii) long-termrelease of substance P.

Substance P-immobilized bioactive peptides have a potential torecruit cells with angiogenic activities, including MSCs. It has beenreported that substance P recruits not only MSCs, but also variousangiogenic cells in vitro and in vivo, and these cells have beenshown to participate in angiogenesis [5,35,36]. Ko IK et al. and HongHS et al. demonstrated that substance P treatment recruited MSCs,EPCs, pericytes and HSCs, and resulted in the formation of bloodvessels [7,15]. Kohara H et al. reported that the controlled release ofSP efficiently induced the recruitment of granulocytes from theblood circulation into the site implanted, resulting in enhanced

J.H. Kim et al. / Biomaterials 34 (2013) 1657e1668 1667

angiogenesis [6]. Our results also confirmed that bioactive peptideswere effective in the recruitment of MSCs which could induceangiogenesis. Due to the recruitment of cells, it is considered thatthe biological regulation of angiogenesis by substance P is mediatedby the direct effect on ECs. Ziche M et al. demonstrated thatsubstance P induced proliferation and differentiation of ECs, andstimulated neovascularization in vivo [37,38]. Overall, it is possiblethat recruited stem or progenitor cells and stimulated vascular cellsby substance P-immobilized bioactive peptides were proliferated orsometimes differentiated, and consequently, enhanced angiogen-esis with recovery of blood flow. Although this study focused onMSCs recruitment of bioactive peptides, the recruiting ability forvarious other cell types with angiogenic activities remains to bedetermined.

Bioactive peptides provided 3D microenvironments whichcould promote neovascularization. Among the various self-assembling peptides, RADA16-II, composed of bioactive peptides,is suitable for enhancing angiogenesis. As described above,RADA16-II provided angiogenic microenvironments that promotedcapillary-like formation. The RADA16-II is superior for infiltratingECs and SMCs, inhibiting ECs apoptosis, and promoting long-termsurvival relative to other hydrogels [18,28]. Based on these prop-erties, RADA16-II has been used to treat MI by promoting angio-genesis. Davis ME et al. reported that injectable microenvironmentsin the infarcted myocardium recruit EC and SMC and promote theirvascularization [23,28]. Additionally, self-assembling peptides canprovide a suitable microenvironment for cells to adhere to, andperform normal cellular functions, so they support cell attachment,migration, and the differentiation of a variety of cells includingMSCs and ECs [29]. Yoshimi R et al. reported that MSCs maintainedtheir own potential by increasing attachment and survival rates oftransplanted cells in the 3D microenvironment by self-assemblingpeptides and promoting tissue regeneration [33]. Therefore,bioactive peptides provided 3D microenvironments withinischemic limbs, which promoted vascularization by infiltrating andattaching vascular cells as well as MSCs.

In this study, bioactive peptides could maintain increasedvascular density by sustained delivery of substance P. As describedpreviously, bioactive peptides which were covalently bonded withsubstance P, enabled substance P to release until 28 days, and thatled to the enhanced recruitment of MSCs extending their angio-genic activity. Enhanced vascular density was maintained until 28days only in the bioactive peptides-injected group, being slightlydecreased in the other groups. Moreover, the bioactive peptides-injected group showed a high maturation index up to 95%. Itmeant that almost all vessels were stable and functional, whichconsisted with ECs and SMCs. It is essential to supply sufficientblood flow by forming functional vessels to treat ischemia. There-fore, to apply bioactive peptides by long-term delivery of substanceP is one of themost powerful strategies to enhance angiogenesis forthe treatment of ischemia.

The key to in situ tissue regeneration in the initial stage is theproficient recruitment of host stem or progenitor cells into animplanted scaffold. However, adult stem cell populations in thebody are generally too low in number to have a significant effect ontissue regeneration. Therefore, we attempted to improve theprotein delivery system to effectively mobilize host stem cells,especially MSCs. Even though MSCs mobilization mechanism ofsubstance P from bone marrow into injured sites has not been fullyelucidated, it is certain that substance P can recruit and proliferateMSCs [4,7,15]. Moreover, self-assembling peptides have theadvantages of non-immunogenicity and non-inflammatoryresponse [19e21]. Therefore, bioactive peptides can substitute forG-CSF usage for MSCs mobilization in a variety of ischemic tissueinjuries [7]. In addition to the potential therapeutic uses of

bioactive peptides as a wound-healing agent and inducer of ther-apeutic MSCs mobilization, its side effects, such as neuro-inflammation and metastatic stimulation, must also be consideredand finely regulated [39].

5. Conclusion

We developed injectable bioactive peptides which could recruitMSCs, and their therapeutic potential was evaluated on ischemichind limb model. By applying bioactive peptides, substance P wasreleased long-term and 3D microenvironments were provided intoischemic regions. Consequently, bioactive peptides could recruitMSCs successfully, prevent fibrosis, and promote neo-vascularization, with enhanced tissue perfusion that led to fullrecovery of ischemic hind limb. Our results confirmed that bioac-tive peptides are one of the most powerful tools to treat ischemicdisease without cells transplantation by recruiting autologous stemcells.

Acknowledgments

This study was supported by a grant of the National ResearchFoundation of Korea Grant funded by the Korean Government(MEST) (NRF-2010-C1AAA001-2010-0028939).

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.biomaterials.2012.11.008.

References

[1] Roger VL, Go AS, Lloyd-Jones DM, Adams RJ, Berry JD, Brown TM, et al. Heartdisease and stroke statisticse2011 update: a report from the American HeartAssociation. Circulation 2011;123:e18e209.

[2] Varu VN, Hogg ME, Kibbe MR. Critical limb ischemia. J Vasc Surg 2010;51:230e41.

[3] Sieveking DP, Ng MK. Cell therapies for therapeutic angiogenesis: back to thebench. Vasc Med 2009;14:153e66.

[4] Hong HS, Lee J, Lee E, Kwon YS, Lee E, Ahn W, et al. A new role of substance Pas an injury-inducible messenger for mobilization of CD29(þ) stromal-likecells. Nat Med 2009;15:425e35.

[5] Wu YJ, Chen L, Scott PG, Tredget EE. Mesenchymal stem cells enhance woundhealing throughdifferentiation andangiogenesis. StemCells 2007;25:2648e59.

[6] KoharaH, Tajima S, YamamotoM, Tabata Y. Angiogenesis induced by controlledrelease of neuropeptide substance P. Biomaterials 2010;31:8617e25.

[7] Hong HS, Kim DY, Yoon KJ, Son Y. A new paradigm for stem cell therapy:substance-P as a stem cell-stimulating agent. Arch PharmRes 2011;34:2003e6.

[8] Lee CH, Cook JL, Mendelson A, Moioli EK, Yao H, Mao JJ. Regeneration of thearticular surface of the rabbit synovial joint by cell homing: a proof of conceptstudy. Lancet 2010;376:440e8.

[9] Huebsch N, Mooney DJ. Inspiration and application in the evolution ofbiomaterials. Nature 2009;462:426e32.

[10] Lee SJ, Van Dyke M, Atala A, Yoo JJ. Host cell mobilization for in situ tissueregeneration. Rejuvenation Res 2008;11:747e56.

[11] Orlic D, Kajstura J, Chimenti S, Jakoniuk I, Anderson SM, Li B, et al. Bonemarrow cells regenerate infarcted myocardium. Nature 2001;410:701e5.

[12] Lapidot T, Petit I. Current understanding of stem cell mobilization: the roles ofchemokines, proteolytic enzymes, adhesion molecules, cytokines, and stromalcells. Exp Hematol 2002;30:973e81.

[13] Rameshwar P, Joshi DD, Yadav P, Qian J, Gascon P, Chang VT, et al. Mimicrybetween neurokinin-1 and fibronectin may explain the transport and stabilityof increased substance P immunoreactivity in patients with bone marrowfibrosis. Blood 2001;97:3025e31.

[14] Castagliuolo I, Valenick L, Liu J, Pothoulakis C. Epidermal growth factorreceptor transactivation mediates substance P-induced mitogenic responsesin U-373 MG cells. J Biol Chem 2000;275:26545e50.

[15] Ko IK, Ju YM, Chen T, Atala A, Yoo JJ, Lee SJ. Combined systemic and localdelivery of stem cell inducing/recruiting factors for in situ tissue regeneration.FASEB J 2012;26:158e68.

[16] Genove E, Shen C, Zhang S, Semino CE. The effect of functionalized self-assembling peptide scaffolds on human aortic endothelial cell function.Biomaterials 2005;26:3341e51.

J.H. Kim et al. / Biomaterials 34 (2013) 1657e16681668

[17] Horii A, Wang X, Gelain F, Zhang S. Biological designer self-assemblingpeptide nanofiber scaffolds significantly enhance osteoblast proliferation,differentiation and 3-D migration. PLoS One 2007;2:e190.

[18] Narmoneva DA, Oni O, Sieminski AL, Zhang S, Gertler JP, Kamm RD, et al. Self-assembling short oligopeptides and the promotion of angiogenesis. Bioma-terials 2005;26:4837e46.

[19] Zhao XJ, Zhang SG. Fabrication of molecular materials using peptideconstruction motifs. Trends Biotechnol 2004;22:470e6.

[20] Zhang SG. Fabrication of novel biomaterials through molecular self-assembly.Nat Biotechnol 2003;21:1171e8.

[21] Ellis-Behnke RG, Liang YX, You SW, Tay DK, Zhang S, So KF, et al. Nano neuroknitting: peptide nanofiber scaffold for brain repair and axon regenerationwith functional return of vision. Proc Natl Acad Sci U S A 2006;103:5054e9.

[22] Segers VFM, Lee RT. Local delivery of proteins and the use of self-assemblingpeptides. Drug Discov Today 2007;12:561e8.

[23] Kim JH, Jung Y, Kim SH, Sun K, Choi J, KimHC, et al. The enhancement of maturevessel formation and cardiac function in infarcted hearts using dual growthfactor delivery with self-assembling peptides. Biomaterials 2011;32:6080e8.

[24] Hsieh PCH, Davis ME, Gannon J, MacGillivray C, Lee RT. Controlled delivery ofPDGF-BB for myocardial protection using injectable self-assembling peptidenanofibers. J Clin Invest 2006;116:237e48.

[25] Yokoi H, Kinoshita T, Zhang S. Dynamic reassembly of peptide RADA16nanofiber scaffold. Proc Natl Acad Sci U S A 2005;102:8414e9.

[26] Webber MJ, Tongers J, Newcomb CJ, Marquardt KT, Bauersachs J, Losordo DW,et al. Supramolecular nanostructures thatmimic VEGF as a strategy for ischemictissue repair. Proc Natl Acad Sci U S A 2011;108:13438. 2012;109:9220.

[27] Bhang SH, Cho SW, La WG, Lee TJ, Yang HS, Sun AY, et al. Angiogenesis inischemic tissue produced by spheroid grafting of human adipose-derivedstromal cells. Biomaterials 2011;32:2734e47.

[28] Davis ME, Motion JP, Narmoneva DA, Takahashi T, Hakuno D, Kamm RD, et al.Injectable self-assembling peptide nanofibers create intramyocardial micro-environments for endothelial cells. Circulation 2005;111:442e50.

[29] Zhang S, Holmes TC, DiPersio CM, Hynes RO, Su X, Rich A. Self-complementaryoligopeptide matrices support mammalian cell attachment. Biomaterials1995;16:1385e93.

[30] Post MJ, Laham R, Sellke FW, Simons M. Therapeutic angiogenesis in cardi-ology using protein formulations. Cardiovasc Res 2001;49:522e31.

[31] Gupta R, Tongers J, Losordo DW. Human studies of angiogenic gene therapy.Circ Res 2009;105:724e36.

[32] Guo J, Leung KK, Su H, Yuan Q, Wang L, Chu TH, et al. Self-assembling peptidenanofiber scaffold promotes the reconstruction of acutely injured brain.Nanomedicine 2009;5:345e51.

[33] Yoshimi R, Yamada Y, Ito K, Nakamura S, Abe A, Nagasaka T, et al. Self-assembling peptide nanofiber scaffolds, platelet-rich plasma, and mesen-chymal stem cells for injectable bone regeneration with tissue engineering.J Craniofac Surg 2009;20:1523e30.

[34] Koon HW, Zhao D, Zhan Y, Moyer MP, Pothoulakis C. Substance P mediatesantiapoptotic responses in human colonocytes by Akt activation. Proc NatlAcad Sci U S A 2007;104:2013e8.

[35] Crisan M, Yap S, Casteilla L, Chen CW, Corselli M, Park TS, et al. A perivascularorigin for mesenchymal stem cells in multiple human organs. Cell Stem Cell2008;3:301e13.

[36] Meirelles Lda S, Fontes AM, Covas DT, Caplan AI. Mechanisms involved in thetherapeutic properties of mesenchymal stem cells. Cytokine Growth FactorRev 2009;20:419e27.

[37] Ziche M, Morbidelli L, Pacini M, Geppetti P, Alessandri G, Maggi CA. SubstanceP stimulates neovascularization in vivo and proliferation of cultured endo-thelial cells. Microvasc Res 1990;40:264e78.

[38] Wiedermann CJ, Auer B, Sitte B, Reinisch N, Schratzberger P, Kahler CM.Induction of endothelial cell differentiation into capillary-like structures bysubstance P. Eur J Pharmacol 1996;298:335e8.

[39] Kidd BL, Inglis JJ, Vetsika K, Hood VC, De Felipe C, Bester H, et al. Inhibition ofinflammation and hyperalgesia in NK-1 receptor knock-out mice. Neuroreport2003;14:2189e92.