Nanoenhanced hydrogel system with sustained release capabilities

Sonali Karnik,1 Kanesha Hines,1* David K. Mills1,2

1Center for Biomedical Engineering and Rehabilitation Science, Louisiana Tech University, Ruston, Louisiana 712722The School of Biological Sciences, Louisiana Tech University, Ruston, Louisiana 71272

Received 24 August 2014; revised 4 November 2014; accepted 19 November 2014

Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.a.35376

Abstract: An alginate/halloysite nanotube (HNT) nanocompo-

site was developed with sustained release of bone morphoge-

netic proteins (BMPs) at picogram low levels. BMP-2, 4, and 6

and osteoblasts were chosen as our model “growth factor”

and “cell type” as the interaction of BMPs with osteoblasts is

well known and thoroughly investigated. Alginate hydrogels

with HNTs doped with BMP-2, 4, or 6 only or BMP-4 and 6 in

combination. Osteoblasts were seeded within the hydrogels

and studied for changes in cell proliferation, phenotypic

expression, and mineralization over a 28-day experimental

period. Osteoblast behavior was enhanced in BMP doped

hydrogel/HNTs nanocomposites as compared with control

groups. Release profiles showed that BMP-2 was released in a

sustained fashion over a 7-day period and at picogram levels.

Mineralization, as showed by Von Kossa staining, and protein

synthesis peaked at 28 days, for all three growth factor combi-

nations. BMP-4 provided a marked stimulus for osteoblast

functionality base and was comparable to BMP-6 in terms of

osteoblast differentiation and mineralization. BMP-4 and 6, in

combination, showed a marked enhancement in osteoblast

differentiation and functionality; however, the response

seemed to be delayed when compared with BMP-4 and 6

release. Hydrogel surfaces had a complex surface topography

and greater structural integrity with increased halloysite

addition. The data suggest that these nanocomposites may

provide a mechanism to enhance repair and regeneration in

damaged or diseased tissues, reducing the need for more

invasive treatment modalities. VC 2014 Wiley Periodicals, Inc. J

Biomed Mater Res Part A: 00A:000–000, 2014.

Key Words: halloysite nanotubes, osteoblasts, growth factors,

sustained release, tissue repair

How to cite this article: Karnik S, Hines K, Mills DK. 2014. Nanoenhanced hydrogel system with sustained release capabilities.J Biomed Mater Res Part A 2014:00A:000–000.

INTRODUCTION

In the field of regenerative medicine, an array of scaffoldshave been studied for use in the repair or replacement ofdamaged or injured tissues and organs1–4 No single scaffoldtype produced has been the “perfect package” possessing allthe critical features such as cyto- and tissue compatibility,the appropriate tissue architecture, bioactivity and has therequisite mechanical properties5–7. One area seeing anintense research effort that directly addresses this issue isthe design of novel constructs that are biocompatible, cellsupportive, and able to deliver bioactive molecules that helpthe body’s native regenerative response8–11 and Refs. 10and 11 for an extensive review on the topic. Hydrogels, inparticular, have been used to deliver drugs and bioactiveagents in various forms12,13 and using a variety of releasingsystems, including alginate,14 chitosan,15 and calcium phos-phate16 (see also Refs. 16 and 17 for a complete review ofthe use of these materials as drug delivery systems).

Our hypothesis was that halloysite nanotubes (HNTs)could provide sustained release of bone morphogenetic pro-teins (BMPs) from alginate hydrogels at extremely low levelsand over a sustained time period. We used a novel approach

in our design by combining calcium alginate hydrogels withan organic clay nanotube, halloysite, as a nanocontainer andcarrier.18–20 Halloysite is a naturally occurring two-layeredaluminosilicate, chemically similar to kaolinite and has a pre-dominantly hollow nanotubular structure with an inner diam-eter ranging from 15 to 50 nm, an outer diameter from 30 to50 nm, and a length between 100 and 2000 nm dependingon its extraction site.21,22 The surface and inner lumenalcharacteristics, cytocompatibility, and biocompatibility ofHNTs make it a potential drug delivery system. Both hydro-phobic and hydrophilic agents can be entrapped after anappropriate pretreatment of the halloysite surface.23–25

Smaller molecular-sized drugs are trapped within the innerlumen of the HNTs, and drugs of a larger molecular size typi-cally attach to the outer surface of HNTs.26 A wide range ofactive agents, including antibiotics, cancer drugs, tetracycline,marine biocides, can be entrapped within the inner lumen,and within void spaces of the multilayered aluminosilicateshells (extensively reviewed in Abdllayev and Lvov26). Thisentrapment of diverse biological and industrial agents, fol-lowed by their retention and sustained release, supports theconcept of halloysite as a nanocontainer well suited for

*Kanesha Hines contributed to this manuscript as part of her participation in Louisiana Tech University’s NSF REU program.Correspondence to: D.K. Mills; e-mail:dkmills@latech.eduContract grant sponsor: Golden Key International Honor Society (to S.K.)

Present address: LDR Spine, 13785 Research Blvd. Suite 200, Austin, TX 78750

VC 2014 WILEY PERIODICALS, INC. 1

molecular delivery applications. Growth factor doped HNT/alginate hydrogels were studied for their ability to deliver acombination of growth factors and their subsequent effect onosteoblast functionality.

For assessment of our nanocomposites, we selected a“model protein,” the BMPs, including BMPs-2, 4, and 6.BMPs are proteins that stimulate ectopic bone growth andare members of the transforming growth factor beta (TGF-b) superfamily. BMP-2 has been shown to stimulate boneproduction27–29 and has been used in diverse bioengineer-ing applications [reviewed in Refs. 30 and 31]. Recombinanthuman protein (rhBMP-2) is also approved by the Food andDrug Administration for orthopedic and dental applicationsin the United States.32 BMP-4 is a protein that stimulatesectodermal tissue differentiation and BMP-6 induces osteo-blast differentiation in mesenchymal stem cells and subse-quent osteogenesis.30 The cells used in this experiment arebone-forming cells, osteoblasts responsible for bone tissueformation during fracture repair. We chose the osteoblast asour “model cell type” as the interactions of BMPs-2, 4, and 6on osteoblast behavior are well known and characterized.30–32

Cell-seeded alginate HNT/hydrogel nanocomposites(beads) were doped with one of four BMP combinations:BMP-2, BMP-4, and BMP-6 only, and BMP-4 and 6 in combi-nation. Nanocomposites were cultured over a 28-day experi-mental period and samples were removed for analysis ondays 7, 14, 21, and 28. As compared with controls (HNTsonly), cell proliferation, mineralization, and protein contentincreased over the experimental period. Mineralization, asshowed by Von Kossa staining and Alizarin Red assay,peaked at 28 days for all three growth factor combinations.Osteoblast behavior was significantly enhanced in BMP-2and 4 doped hydrogel/HNT nanocomposites as comparedwith control groups and also showed enhanced stimulationbased on increased mineralization content. BMP-6 alsoshowed marked increases in mineralization over time andwas comparable to BMP-4 in terms of osteoblast differentia-tion and mineralization. BMP-4 and 6, in combination, alsoshowed a marked enhancement in osteoblast differentiationand functionality, but the response seemed to be delayedwhen compared with the effects of a single dose of BMP-2,4, and 6.

We propose that HNT doped, cell-seeded alginate hydro-gels may have potential as a repair material for damagedand compromised cartilage and bone, an irreversibly dam-aged disc, structurally altered ligaments, or tendons andencourage the restoration of normal tissue function, thusreducing the need for more invasive treatment modalities.Furthermore, our results suggest that HNTs could be dopedwith a suite of bioactive factor and released in a combinato-rial effect for either synergistic or antagonistic effects.

MATERIALS AND METHODS

MaterialsSyringes, 12-well plates, centrifuge and microcentrifuge tubes,ethanol, pipettes, and cell culture plastics were obtained fromMid Scientific, St. Louis, MO. All cell culture media, buffers,and serum were purchased from Life Technologies (Grand

Island, NY). Sodium citrate, sodium alginate, sodium chloride,calcium chloride, and HNTs were purchased from SigmaAldrich (St. Louis, MO). BMPs (BMP-4, BMP-6) were purchasedfrom Prospec (Rehovat, Israel). The Quantikine ELISA kit forBMP-2 was obtained from R&D Systems (MN). The MC 3T3subclone E1 preosteoblast cell line was purchased from ATCC(ATCC CRL-2593, Manassas, VA). This cell line is used as amodel for studying in vitro osteoblast differentiation andexhibits a high level of osteoblast differentiation after additionof ascorbic acid.33 Once differentiated, cellular behavior is sim-ilar to that of primary calvarial osteoblasts. Preosteoblastbehavior was tested in the medium with and without BMP-2and ascorbate. Cell characterization studies were designed toassess protein production, differentiation, and mineralizationand to check the extent of mineralization and extracellularmatrix (ECM) production and used as a model for studying invitro osteoblast differentiation and for comparison with cellu-lar behavior in response to BMP-doped HNTs.

Osteoblast cell culture, proliferation, and differentiationOsteoblasts were cultured according to the protocols pro-vided by the supplier under aseptic conditions. The mediumused was a-MEM 90% (GIBCO Invitrogen, Grand Island,NY), 10% fetal bovine serum (Phenix, Candler, NC), and 1%penicillin–streptomycin (Phenix, Candler). Cells were grownto passage 3 and then 0.2M L-ascorbic acid was added tothe medium to initiate osteoblasts differentiation. Afterascorbate addition, cells changed their morphology andexpressed bone marker proteins 3 days after addition of L-ascorbic acid.

For seeding into hydrogel/HNT composites, cells wereharvested after osteoblast differentiation as describedabove. Cells were detached from the flasks by adding Try-plE, an animal-free trypsin substitute (GIBCO Invitrogen,Grand Island, NY) and mixed with Hank’s balanced salt solu-tion (HBSS). Cells were then centrifuged, the cell pelletsresuspended in complete a-MEM, and cell number for sub-sequent seeding in alginate hydrogels determined using ahemocytometer.

Preparation of alginate beadsFor all study groups, a solution of 2% w/v sodium alginatein autoclaved reverse osmosis (RO) water was used to pre-pare alginate beads. Two control and two experimentalgroups were used. Control group #1 consisted of cell-seeded 2% w/v sodium alginate beads. For control group#2, cells were seeded in a 2% sodium alginate solution, towhich 1% w/v HNTs were added. For experimental group#1, cells were seeded in a 2% sodium alginate and 1%HNTs loaded with BMP solution, mixed and added to auto-claved RO water. The alginate solutions, except the experi-mental, were placed under a UV lamp in a laminar flowhood for 30 min for sterilization. The HNTs used in theexperimental groups were sterilized before loading by dip-ping in 70% ethanol and drying inside a laminar air hoodand then vacuum loaded with BMPs under sterile condi-tions. BMP-2 was prepared by adding 1 mL sterile 10 mMacetic acid to prepare a stock solution of 10 lg/mL. From

2 KARNIK, HINES, AND MILLS NANOENHANCED HYDROGEL WITH SUSTAINED RELEASE CAPABILITIES

this stock solution, a 2-mL solution of concentration 5 lg/mL was prepared. This was the concentration used to loadinto HNTs under sterile conditions. The BMP-4 solution wasprepared by adding 1 mL of 100 mM acetic acid to lyophi-lized powder. A total of 100 mL from this stock solution wasremoved and put into the BMP-4-labeled tubes and filledwith 900 mL of water to make 0.01% v/v concentration ofBMP-4. For BMP-6, a 20-mM acetic acid was added to thelyophilized powder. A total of 20 mL was removed from thestock solution and 980 mL of water was added to make upa total volume of 1000 mL and a working concentration of2% v/v. For the BMP-4 and 6 tubes, 0.05 g of BMP-4 andBMP-6 doped HNTs were used. Group #1 beads were cul-tured in complete a-MEM, to which no ascorbate wasadded. Experimental group #2 consisted of cells seeded into2% sodium alginate/1% HNT/BMP-2 solution and culturedin complete a-MEM, to which ascorbate was added.

Solutions of 1% w/v calcium chloride were prepared inautoclaved RO water and then filter sterilized. Cells, sus-pended in HBSS, were then gently mixed within the three-alginate solutions at a cell density of 1 3 106cells/bead.The alginate solutions were loaded in a 27-G syringe anddrops of this solution were added to a 1% w/v sterile cal-cium chloride solution by gentle agitation. Beads formedimmediately upon contact with the calcium chloride solu-tion. The solution with formed beads was left standing for15 min for proper gelation. The beads were then collectedand washed in sterile HBSS twice to remove traces of resid-ual calcium chloride. Control cultures were fabricated in thesame manner but contained undoped HNTs.

Five, 12-well plates were prepared with complete a-MEM medium (6ascorbate) in the wells. Beads were thenplaced in each well and kept within a cell culture incubatorset at 37oC, 95% air, and 5% CO2. The plates with undopedalginate/HNT beads and doped alginate/doped HNT beadswere then placed in the incubator at 37oC and cultured for28 days, fed every other day with fresh complete a-MEMand samples were removed and fixed on 7, 14, 21, and 28days in culture. All procedures were performed under alaminar flow hood maintaining sterile conditions.

Sample preparation for scanning electron microscopyTo study bead morphology after the addition of HNTs, sur-face morphology and overall shape and size was studiedusing scanning electron microscopy. In this study, samplesof alginate, with and without HNTs (2% w/v sodium algi-nate in autoclaved RO water), were prepared. HNTs in algi-nate were prepared in the following concentrations: 0, 0.25,0.50, 0.75, and 1% HNTs (% w/v). The beads, after forma-tion, were isolated and washed in sterile RO water. Thesamples were dried at 56oC overnight. The beads were thenfirmly attached on a two-sided adhesive conducting tape.Samples were imaged using a S4800 field emission scanningelectron microscope (SEM), HITACHI SEM.

Dissolving alginate hydrogels for cell isolationA 1% sodium citrate and 150 mM sodium chloride solutionwas prepared in autoclaved RO water and then filter-sterilized.

The alginate hydrogel beads were then removed from the wellplates and washed in HBSS. Beads were put in the solution ofsodium citrate and sodium chloride and were swirled gently tospeed up the process of dissolution. Beads were dissolved in30 min. The solution was collected and then centrifuged at12,000 rpm for 5 min. The supernatant was discarded andthen the pellet of cells was resuspended in HBSS. This suspen-sion was used for various assays performed in this study.

Loading of the HNTs with BMPsThe HNTs were first washed in ethanol by dipping in 70%ethanol and then air dried in a laminar air hood. The vac-uum chamber was then placed inside the hood after thesurface sterilization. A 5-mg/mL solution was prepared froma stock of 10 mg/mL BMP-2 solution by diluting the stocksolution in autoclaved RO water. Microfuge tubes were filledwith HNTs-BMP-2 solution and the vacuum was kept run-ning for 24 h. The HNTs were then removed and washedwith sterile RO water to remove traces of BMP-2 on the sur-face. These were then added to the alginate solution.

Cell viability and proliferation assayThe cells were counted on days 0, 1, 3, 7, and 14 by trypanblue (Sigma Aldrich) method. Osteoblasts were isolatedfrom the beads by dissolving the beads in a 1% sodiumcitrate/HBSS solution. Cells were then centrifuged at10,000 rpm for 5 min. The pellet was resuspended in HBSS.Cell viability was quantified using the cell titre blue assay(Promega, Sunnyvale, CA) and an absorbance microplatereader (Fluostar Optima BMG Labtech, Cary, NC). The resaz-urin in the assay kit reacts with live cells to form fluores-cent resorufin. Only live cells have the ability to reduceresazurin to resorufin, which fluoresces at 590 nm. Theabsorbance is directly proportional to the number of viablecells in the well. The resazurin is deep blue in color and isreduced to pink-colored resorufin by viable cells. Theabsorbance was measured using wavelengths 490 and630 nm.

Embedding and sectioningAlginate beads were collected at days 0, 1, 3, 7, 14, and 21days and washed with DPBS. Fixation of the tissue con-structs was performed by two methods: (1) embedding thebeads in 1:1 2% w/v gelatin and 2% w/v agar noble mix-ture in autoclaved RO water and then fixing in 2% bufferedparaformaldehyde, overnight and (2) fixation of beadsdirectly in 95% ethanol for 10 min. The molds and beadswere then divided using a sharp blade.

HistologyTo visualize the extent of ECM formation and mineralization,sections were stained with Alcian blue, Picrosirius red, andVon Kossa. Alcian blue staining was performed to visualizeglycosaminoglycan content and distribution in the ECM. Pic-rosirius red stains the collagen as bright red mass. VonKossa staining was performed to visualize mineral deposi-tion and these appear as dark brown or black spots. Eitherhematoxylin or nuclear fast red was used as a nuclear stain.

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00 3

Total protein content and ECM proteinsTotal protein content was calculated by performing a MicroBCA assay (Thermo Scientific). The medium was collectedfor days 0, 1, 3, 7, and 14. Readings were taken on BiotekMicroplate reader at 562 nm.

For statistical analyses of the data, the Excel 2010 dataanalysis toolpak and SPSS software were used. For experi-mental analyses, a minimum of three samples per time ortesting point from each group (controls 1 and 2, and experi-mental 1 and 2) was used. All experiments were repeatedtwice and results from both sets of experiments were inaccord. A two-way analysis of variance (ANOVA) was per-formed using SPSS software to understand the interactionsand trends of the control and experimental groups.

BMP-2 release study from HNTsThe HNTs were loaded for 24 h in a vacuum-loading cham-ber. The HNTs were taken out and washed in autoclaved ROwater. A magnet was inserted in the tube carrying theHNTs. The tube was placed on a magnetic stirrer set at 80-rpm speed. The BMP-2 release was estimated using theQuantikine ELISA kits for BMP-2, and due to the nature ofthe kit, sampling was limited to a 7-day period. Sampleswere collected at 5 min, 10 min, 15 min, 20 min, 30 min,1 h, 2 h, 4 h, 6 h, 8 h, 12 h, 24 h, 36 h, 48 h, 60 h, 72 h, 84h, and every 24 h afterward up to 7 days.

RESULTS

Scanning electron microscopyAlginate hydrogels, with and without HNTs, were examinedusing a SEM to characterize hydrogel morphology, uniformity,and surface features. SEM images suggest that increased con-centration of HNTs leads to increased surface roughness and

mechanical strength. The beads without HNTs had a smoothmorphology and were transparent [Fig. 1(A)]. They also col-lapsed immediately after dehydration. The beads with 0.25and 0.5% HNTs had morphology almost similar to the beadswithout HNTs only few HNTs can be seen clumped in groups,but these beads were not transparent [Fig. 1(B)]. The beadswith 0.75% HNTs showed that the HNTs are scattered acrossthe surface and are distinct in their roughness [Fig. 1(C)],whereas beads with 1% HNTs showed most uniform distri-bution of HNTs and the roughest surface [Fig. 1(E)].

Alginate/HNT hydrogels did not collapse even after com-plete dehydration. During processing for histological analy-sis, hydrogels with increasing HNT concentration resistedcomplete desiccation and became more difficult to macerate,suggesting that they became stiffer with HNT addition.Accordingly, beads with 1% w/v HNTs were selected forfurther experimental analysis.

In cross-section, beads had a hollow inner core [Fig. 2(A,C)]and an outer porous core [Fig. 2(B,D)] with HNTs concentratedwithin the outer core with a few HNT clusters protruding fromthe surface [Fig. 2(B)]. The best analogy for morphologicalcomparison would be a coconut shell cut in half.

Cell viability assayThis assay uses resazurin (blue colored) to resorufin (pinkcolored) reduction reaction. The viable cells have the abilityto reduce resazurin to resorufin. The absorbance was takenat 490 and 630 nm on BioTek microplate reader. The resultsare as shown in Figure 3.

The cell viability on day 0 is almost similar in all thefour sets. The cell viability is highest for the constructs withBMP-2 loaded HNTs with and without 0.2M ascorbic acid on

FIGURE 1. (A) Alginate beads without HNTs, (B) Alginate beads 0.25% HNTs and 0.5% HNTs (inset) (C) Alginate beads 0.75% HNTs (D) Alginate

beads 1% HNTs.

4 KARNIK, HINES, AND MILLS NANOENHANCED HYDROGEL WITH SUSTAINED RELEASE CAPABILITIES

day 7. The cell viability decreased on day 14 for all the fourtypes of constructs compared to day 7.

Total protein contentThe total protein content was measured by Micro BCAmethod. Total protein content assay was performed to esti-mate the cell activity. The results are as shown in Figure 4.

Statistical analysisThe samples for each group (controls 1 and 2 and experi-mental 1 and 2) had three constructs each and the experi-

ments were repeated twice. A two-way ANOVA wasperformed using SPSS software to understand the interac-tions and trends of the control and experimental groups.The total protein content assay for hydrogel constructsshowed significant differences with increased time in cul-ture, however, but no significant difference was observedbetween conditions, that is, no significant trend wasobserved between the behavior of control and experimentalgroups. The graphical trend shown by the software confirmsthe increase in total protein content in experimental groups1 and 2 but the total protein content dips after day 7. This

FIGURE 2. SEM micrographs of nanocomposites. (A, C) Inner surface of hydrogel. (B,D) Outer surface of hydrogel. Arrows show HNTs. A and B

= lower power SEM, C and D = higher power SEM.

FIGURE 3. Cell titre blue assay day 0–14. All the values for cell viability across the days from 0 to 7 were significant (p< 0.05).

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00 5

trend is also observed in the graph for total protein content(Fig. 4) in the results section. No significant difference couldbe observed between the groups for cell viability assay aswell but a qualitative trend can be observed as seen in thegraph for cell viability assay (Fig. 3) in the results section.Since the sample size was not large enough for the accuratecomputation of variance between the controls and experi-mental groups, the results show a qualitative trend only andthe insignificant results across conditions should be viewedcautiously due to sample constraints.

BMP-2 releaseThe graph for the release profile for BMP-2 from HNTsshowed a controlled and sustained release of the BMP 2molecule over a period of 7 days (Fig. 5). The concentrationof the BMP-2 molecule released was in picogram per millili-ter and the highest being approximately 3,600 pg/mL,which is 36 ng/mL.

Alcian blue stainingControl HNT/hydrogel (no BMP-2) cultures showed amarked decrease in Alcian blue staining in comparison with

both experimental cultures (Fig. 6). In addition, across theexperimental time period, Alcian blue staining showed onlya marginal increase in stain intensity. In contrast, bothexperimental groups showed major increases in Alcian blueperiod over the experimental period. Patches of intensestaining were observed on day 7 [Fig. 6(B)] and increasedin size and stain intensity by day 21 [Fig. 6(C)].

Picrosirius red stainingPicrosirius red staining is routinely used in the histologicalidentification of collagen types I and III fibers using brightfield microscopy. Control groups showed less intense stain-ing compared with experimental groups, a condition similarto Alcian blue staining (Fig. 7). Staining in all experimentalgroups showed marked collagen deposition by day 7 andincreased in stain intensity by days 14 and 21 and leveledoff by day 28 based on stain intensity [Fig. 7(B,C)].

Von Kossa stainingThe trends observed with Alcian blue and Picrosirius redwere also observed with Von Kossa staining. Stain intensitywas greater in the experimental group than in controls andmineralization was greatest in both experimental groups atday 21 (Fig. 8).

DISCUSSION

Most of the approaches used in designing bioartificial tis-sues have emphasized the method of in vitro cell seedingonto a supportive and resorbable scaffold followed by mech-anisms designed to enhance cellular response within theseconstructs.34–36 In most studies, the addition of exogenouslydelivered growth factors was used to modify or redirect cel-lular growth and functionality. Several systems have beendeveloped to combine growth factors into scaffolds, includ-ing incorporation within electrospun scaffolds,37,38 depos-ited within surface coatings, mixing growth factors with

FIGURE 4. Total protein content assay (Micro BCA protein assay). All the values for total protein content across the days from 0 to 14 were sig-

nificant (p< 0.05).

FIGURE 5. Graph showing BMP-2 cumulative release for 7 days.

[Color figure can be viewed in the online issue, which is available at

wileyonlinelibrary.com.]

6 KARNIK, HINES, AND MILLS NANOENHANCED HYDROGEL WITH SUSTAINED RELEASE CAPABILITIES

FIGURE 6. Alcian blue and hematoxylin staining. Control groups showed only a slow increase in the ECM deposition over a period of 21 days. (A)

control group #1, day 7 with insert control group #2, day 21. Patchy regions of ECM appeared by day 7 experimental groups, BMP-2 (B) experimen-

tal group #1 and insert experimental group #2, BMP-2. Patchy areas increased in size (C) experimental group #1 day 14 and insert experimental

group #2, day 21, BMP-2. Similar results were obtained with BMP-4 and 6 and in combination, (D) BMP-4, day 7. (E) BMP-6, day 14. (F) BMP-4 and

6, day 14. Patchy regions = arrows.

FIGURE 7. Picrosirius red stained hydrogel constructs. Picrosirius red stain specifically stains the collagen fibers that mark the onset of differen-

tiation. All showed less collagen deposition when compared with experimental groups, A = control #1 day 21. The collagen deposition in experi-

mental 1 and 2 was highest on day 21, BMP-2. (B) = Experimental group #1 on day 7, BMP-2. (C) Experimental group #1, day 21, BMP-2. (BMPs

4, 6 and in combination showed a parallel pattern in Picrosirius red staining. D) BMP-4. (E) BMP-6. (F) BMP-4 and 6.

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00 7

polymer particles before processing the composites intomicroparticles, scaffolds, or 3D constructs,39–41 and encap-sulating the growth factors into microspheres42,43 orhydrogels.44–46

Growth factors are widely used clinically to treat growthabnormalities, neuropathies, pediatric and adult oncologypatients, bone injuries, and wound healing.47–49 These pro-teins trigger a series of metabolic pathways leading to cellu-lar proliferation, differentiation, and functionality. Thechallenge is to supply these proteins in the proper amounts,which lie in the range of nano or even picograms, and in asustained fashion over a defined period of time.50 Cell-seeded hydrogels that serve as a drug delivery systemdesigned to stimulate the body’s native regenerativeresponse is an area of intense research effort.51,52 Hydrogelshave long been used to deliver drugs and bioactive agentsin various forms53,54 and using a variety of releasing sys-tems, including alginate,55 calcium phosphate,56 chitosan,15

collagen,57 and hyaluronic acid58 among others.The focus of our design was to develop a construct

designed to deliver bioactive molecules BMPs, singly or incombination, at picogram levels and in a sustained fashion.The concept of using HNTs in biomedical applications wasfirst proposed in 200559,60 and the potential of an alginate/growth factor-doped HNTs hydrogel system was firstdescribed in 201161 and its use in applications for TMJrepair in 2012.62,63 This is the first detailed report on theuse of doped HNTs in alginate hydrogels for tissue engineer-ing applications. Our approach was to develop a reparativematerial that is natural, cytocompatible, histogenic, and bio-

degradable as means for in vivo tissue repair or regenera-tion, thus reducing the need for more invasive treatmentmodalities. Alginate hydrogels, loaded with HNTs dopedwith BMP-2, 4, and 6, were seeded with preosteoblasts andstudied in vitro over a 28-day period. Hydrogels wereassessed for BMP-2 release, osteoblast differentiation, func-tionality, and mineralization.

The hypothesis that HNTs could be used for the releaseof growth factors has been confirmed. We were successfulin loading three different growth factors and effecting fourdifferent release combinations. Hydrogels stiffened and theirsurfaces became rougher with increased HNT addition. HNTaddition may have reduced pore size in the hydrogels con-tributing to the observed stiffness. Further testing isrequired to confirm this observation. Release profilesshowed that BMP-2 was released in a sustained fashionover a 7-day period and at picogram levels. Results suggestthat all growth factor combinations enhanced osteoblast dif-ferentiation, functionality, and mineralization. The pattern ofosteoblast cell behavior was nearly identical except theonset of mineralization was delayed in the BMP-4 and 6-combination treatment. In terms of producing bone mineral,BMP-4 seemed to have a more enhanced effect leading tothe largest amount of bone mineralization as evidence byVon Kossa staining. The data support the potential use of ahydrogel-growth-factor doped HNT system as part of a noveltherapeutic system that can deliver growth factors in therequisite amounts and over a sustained period. Our nano-composite delivery system unites the chemoattractive, pro-liferative, and inductive properties of growth factors with

FIGURE 8. Von Kossa staining of hydrogel constructs. Control groups #1 and #2 (insert) showed less mineralization through 28 days (A and B,

BMP-2) when compared against the experimental #1 and #2 on all days (B = day 14, C = day 21). The mineral deposition was highest on day 21

in the experimental constructs. BMP-6 showed marked increases in mineralization over time and was comparable to BMP-4 in terms of osteo-

blast differentiation and mineralization (D). Osteoblast response to the combination of BMP-4 and 6 appeared delayed when compared with the

effects of BMP-4 or BMP-6 (E and F).

8 KARNIK, HINES, AND MILLS NANOENHANCED HYDROGEL WITH SUSTAINED RELEASE CAPABILITIES

the carrier capabilities of HNTs. Our design is flexible, scal-able, and suitable for delivery of single or multiple types ofinstructional packets. We envision a “local” rather than sys-temic strategy wherein the nanocomposites are inserteddirectly into the affected tissue or wound site to accelerateand enhance tissue regeneration and repair.

We have attempted to graphically depict our cumulativeobservations and data in a proposed model that describespresumed osteoblast behavior in BMP-doped HNT/alginatenanocomposites (Fig. 9). We propose that our nanocompo-sites served as a histogenic microenvironment with HNTsserving as a chemoattractant. Osteoblasts migrated towardthe HNTs and the released BMP-2, differentiated in situ andproduced a mineralized matrix that increased in aggregatesize over the experimental period [Fig. 9(A,B)]. Numerousclusters of osteoblasts were observed surrounded by a typeI collagen and a mineral-rich matrix [Fig. 9(C,D)].

However, although we are encouraged by our results,they must be viewed with a certain degree of caution.Although the BMP doped hydrogels triggered osteoblast dif-

ferentiation and functionality, leading to a mineralizedmatrix, additional investigation is needed to conclude quali-tatively and quantitatively the exact character of the matrixproduced and how this compares with in vivo bone tissue.Furthermore, a more detailed characterization of the patternof BMP release and its transductive effect on seeded osteo-blasts must be studied at the gene expression level. Alonger-term study is currently underway with the aim ofaddressing the concerns raised above and as a means totest the predictions derived from our model preparatory totesting the nanocomposites in a critical bone defect in vivomodel.

CONCLUSIONS

A hydrogel growth factor–doped HNT system was developedas a novel treatment modality for tissue repair and regenera-tion. HNTs provided sustained released of BMP-2 at the pico-gram levels. BMP-doped halloysite/alginate hydrogels inducedosteoblast proliferation, differentiation, ECM synthesis, and its

FIGURE 9. (A) Proposed model that summarizes cumulative observations regarding hydrogel structure, osteoblast cell behavior, matrix deposi-

tion and its mineralization. (B) Drawing illustrating BMP-2 release, osteoblast differentiation, collagen deposition, and mineralization. (C) and (D)

Clusters of osteoblasts surrounding HNTs and mineralized collagenous matrix. C= Experimental group #1, day 7. D = Experimental group #2,

day 14. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00 9

mineralization and sustained release of the growth factorBMP-2, 4, and 6. Our hydrogel-growth-factor–doped HNT sys-tem is flexible, scalable, and suitable for delivery of single ormultiple types of instructional packets.

ACKNOWLEDGMENTS

The authors acknowledge the Louisiana’s Governor’s Biotech-nology Initiative, LASPACE (awarded to Dr. David K. Mills) andthank Dr. Alfred Gnasekaran for his help in Electron micros-copy, Mr. Udaybhanu Murthy Jammalamadaka for his help inthe release studies, and Ms. Bharati Belwalkar for her help instatistical analyses by using SPSS software.

REFERENCES1. Parenteau-Bareil R, Gauvin R, Berthod F. Collagen-based biomate-

rials for tissue engineering applications. Materials 2010;3:1863–

1887.

2. Ungaro F, Biondi M, Indolfi L, De Rosa G, La Rotonda M, Quaglia

F, Netti P. Bioactivated polymer scaffolds for tissue engineering.

In: Ashammakhi N, Reis RL, editors. Topics in Tissue Engineering,

Vol.2; 2005.

3. Sohier J, van Blitterswilk ML, De Groot K, Bezemer JM. Critical

factors in the design of growth factor releasing scaffolds for carti-

lage tissue engineering. Exp Opin Drug Deliv 2008;5:543–566.

4. Sauter T, Weigle T, Kratz K, Lendlein A. Polymer scaffolds for

regenerative therapies- Design of hierarchically organized

structures and their morphological characterization. Nano LIFE

2012;2:1230005.

5. Tamayol A, Akbari M, Annabi N, Paul A, Khademhosseini A,

Juncker D. Fiber-based tissue engineering: Progress, challenges,

and opportunities. Biotech Adv 2013;21:669–687.

6. Nichol JW, Khademhosseini A. Modular tissue engineering: Engi-

neering biological tissues from the bottom up. Soft Mater 2009;5:

1312–1319.

7. Lutolf MP, Hubbell JA. Synthetic biomaterials as instructive

extracellular microenvironments for morphogenesis in tissue

engineering. Nat Biotechnol 2005;23:47–55.

8. Sun G, Mao JJ. Engineering dextran-based scaffolds for drug

delivery and tissue repair. Nanomedicine 2012;7:1771–1784.

9. Pagliari S, Romanazzo S, Mosquiera D, Pinto-do-O P, Aoyag T,

Forte G. Adult stem cells and biocompatible scaffolds as smart

drug delivery tools for cardiac tissue repair. Curr Med Chem 2013;

20:3429–3447.

10. Sezer AD. Recent Advances in Novel Drug Delivery Systems.

ISBN 978-953-51-0810-8: InTech; 2012. p 512.

11. 11). Patravale V, Dandekar P, Jain R. Nanoparticulate Drug Deliv-

ery, 1st ed, ISBN 9781907568985: Woodhead Publishing; 2012. p

244.

12. Bindu S, Ashok V, Chatterjee A. Review on hydrogels as drug

delivery in the pharmaceutical field. Int J Pharm Chem Sci 2012;1:

642–661.

13. Ganji F, Vasheghani-Farahani E. Hydrogels in controlled drug

delivery systems. Iranian Poly J 2009;18:63–68.

14. Kamoun E. Alginate Hydrogel as Drug Delivery System: Prepara-

tion of Alginate Hydrogel and its Application as Drug Delivery

System. VDM Verlag Dr. Muller; 2010.

15. Prabakaran M, Mano JF. Chitosan-based particles as controlled

drug delivery systems. Drug Deliv 2005;12:41–57.

16. Slaughter BV, Khurshid SS, Fisher OZ, Khademhosseini A,

Peppas NA. Hydrogels in regenerative medicine. Adv Mater 2009;

21:3307–3329.

17. Sohier J, van Blitterswilk ML, De Groot K, Bezemer JM. Critical

factors in the design of growth factor releasing scaffolds for carti-

lage tissue engineering. Exp Opin Drug Deliv 2008;5:543–566.

18. Lvov YM, Shchukin DG, M€ohwald H, Price RR. Halloysite clay

nanotubes for controlled release of protective agents. ACS Nano

2008;2:814–820.

19. Abdullayev E, Price RR, Shchukin DG, Lvov TM, M€ohwald H, Price

RR. Halloysite tubes as nanocontainers for anticorrosion coating

with benzotriazole. Appl Mater Interfaces 2009;1:437–443.

20. Wei W, Abdllayev E, Goeders A, Hollister A, Mills DK, Lvov YM.

Clay nanotube/poly(methyl methacrylate) bone cement composite

with sustained antibiotic release. Macromol Mater Eng RCS Adv

2012;4:488–494.

21. Veerabardran NG, Price RR, Lvov YM, Clay nanotubes for encap-

sulation and sustained release of drugs. Nano 2007;2:115–120.

22. Lvov YM, Shchukin DG, Mohwald H, Price RR. Halloysite clay

nanotubes for controlled release of protective agents. ACS Nano

2008;2:814–820.

23. Price RR, Gaber BP, Lvov Y. In-vitro release characteristics of tet-

racycline HCl, khellin and nicotinamide adenine dineculeotide

from halloysite; a cylindrical mineral. J Microencapsul 2001;18:

713–722.

24. Wenbo W, Minullina R, Abdullayev E, Fakhrullin R, Mills DK, Lvov

Y. Enhanced efficiency of antiseptics with sustained release from

clay nanotubes. RSC Advances 2013;4:488–494.

25. Wei W, Abdllayev E, Goeders A, Hollister A, Lvov YM, Mills K.

Clay nanotube/poly(methyl methacrylate) bone cement composite

with sustained antibiotic release. Macromol Mater Eng 2012;297:

645–653.

26. Abdllayev E, Lvov Y. Functional polymer clay nanotube compo-

sites with sustained release of chemical agents. Prog Polym Sci

2013;38:1690–1719.

27. Urist MR. Bone: Formation by autoinduction. Science 1965;12:

893–899.

28. Marie PJ, Debiais F, Hay E. Regulation of human cranial osteo-

blast phenotype by FGF-2, FGFR-2 and BMP-2 signaling. Histol

Histopathol 2002;17:877–885.

29. Wozney JM, Rosen V, Bryne M, Celeste AJ, Moustsatos I, Wang

EA. Growth factors influencing bone development. J Cell Sci

Suppl 1990;13:149–156.

30. Nakashima M, Reddi AH. The application of bone morphogenetic

proteins to dental tissue engineering. Nat Biotechnol 2003;21:

1025–1032.

31. Bessa PC, Casal M, Reis RL. Bone morphogenetic proteins in tis-

sue engineering: The road from the laboratory to the clinic, part I

(basic concepts). J Tissue Eng Regen Med 2008;2:1–13.

32. Khan SN, Lane JM. The use of recombinant bone morphogenetic

protein-2 (rhBMP-2) in orthopaedic applications. Exp Opin Biol

Ther 2004;4:741–748.

33. Wang D, Christensen K, Chawla K, Xiao G, Krebsbach P,

Franceschi R. Isolation and characterization of MC3T3-E1 preos-

teoblast subclones with distinct in vitro and in vivo differentia-

tion/mineralization potential. J Bone Miner Res 1999;14:893–903.

34. Weinand C, Xu JW, Peretti GM, Bonassar LI, Gill TJ. Conditions

affecting cell seeding onto three-dimensional scaffolds for

cellular-based biodegradable implants. J Biomed Mar Res B Appl

Biomater 2009;91:80–87.

35. Francioli SE, Candrian C, Martin K, Heberer M, Martin I, Barbero A.

Effect of three-dimensional expansion and cell seeding density on

the cartilage-forming capacity of human articular chondrocytes in

type II collagen sponges. J Biomed Mater Res A 2010;95:924–931.

36. Dai W, Kawazoe N, Lin X, Dong J, Chen G. The influence of struc-

tural design of PLGA/collagen hybrid scaffolds in cartilage tissue

engineering. Biomaterials 2010;31:2141–2152.

37. Ji W, Sun Y, van den Beuken JJ, Fan M, Chen Z, Jansen JA. Bioac-

tive electrospun scaffolds delivering growth factors and genes for

tissue engineering applications. Pharm Res 2011;28:1259–1272.

38. Chung HK, Park TG. Surface engineered and drug releasing pre-

fabricated scaffolds for tissue engineering. Adv Drug Deliv Res

2007;59:249–262.

39. Majeti NV, Kumar R. Nano and microparticles as controlled drug

delivery devices. J Pharm Pharm Sci 2000;3:234–258.

40. Mourino V, Boccaccini AR. Bone tissue engineering therapeutics:

Controlled drug delivery in three-dimensional scaffolds. J R Soc

Interface 2010;7:209–227.

41. Liu X, Zhao K, Gong T, Bao C, Weng J, Zhou S. Delivery of

growth factors using a smart porous nanocomposite scaffold to

repair a mandibular bone defect. Biomacromolecules 2014;15:

1019–1030.

42. Ramteke KH, Jadhav VB, Dhole SN. Microspheres: As carriers

used for novel drug delivery system. IOSR J Pharm 2012;2:44–48.

10 KARNIK, HINES, AND MILLS NANOENHANCED HYDROGEL WITH SUSTAINED RELEASE CAPABILITIES

43. Saralidze K, Koole LH, Knetsch MLW. Polymeric microspheres for

medical applications. Materials 2010;3:3537–3564.

44. Van Vlierberghe S, Dubruel P, Schacht E. Biopolymer-based

hydrogels as scaffolds for tissue engineering applications: A

review. Biomacromolecules 2011;12:1387–1408.

45. Bindu S, Ashok V, Chatterjee A. Review on hydrogels as drug delivery

in the pharmaceutical field. Int J Pharm Chem Sci 2012;1:642–661.

46. Huynh CT, Nguyen MK, Lee DS. Injectable block copolymer

hydrogels: Achievements and future challenges for biomedical

applications. Macromolecules 2011;44:6629–6636.

47. Wingard JR, Demetri GD. Clinical applications of cytokines and

growth factors. Dev Oncol 1999;80:407.[WorldCat]

48. Berrientos S, Stojaninovic O, Golinko MS, Brem H, Tomic-Canic

M. Growth factors and cytokines in wound healing. Wound Rep

Regen 2008;16:585–601.

49. Rose FR, Hou Q, Oreffo RO. Delivery systems for bone growth

factors - The new players in skeletal regeneration. J Pharm Phar-

macol 2004;56:415–427.

50. Vo TN, Kasper FK, Mikos AG. Strategies for controlled delivery of

growth factors and cells for bone regeneration. Adv Drug Deliv

Rev 2012;64:1292–1309.

51. Paolino D, Sinha P, Fresta M, Ferrari M. Drug delivery systems.

In: Webster JG, editor. Encyclopedia of Medical Devices and

Instrumentation. Wiley Online Library; 2006.

52. Ganji F, Vasheghani-Farahani E. Hydrogels in controlled drug

delivery systems. Iranian Poly J 2009;18:63–68.

53. Bindu S, Ashok V, Chatterjee A. Review on hydrogels as drug delivery

in the pharmaceutical field. Int J Pharm Chem Sci 2012;1:642–661.

54. Suri SS, Hicham F, Singh B. Nanotechnology-based drug delivery

systems. J Occup Med Toxicol 2007;2:16–21.

55. Kamoun E. Alginate hydrogel as drug delivery system: Prepara-

tion of alginate hydrogel and its application as drug delivery sys-

tem. VDM Verlag Dr. Muller; 2010.

56. Juhasz JA, Best SM, Bonfield W. Preparation of novel bioactive

nano-calcium phosphate hydrogel composites. Sci Technol Adv

Mater 2010;11:014103.

57. Antoine EE, Vlachos PP, Rylander MN. Review of collagen I

hydrogels for bioengineered tissue microenvironments: Charac-

terization of mechanics, structure, and transport. Tissue Eng Part

B Rev 2014;20:683–696.

58. Burdick JA, Prestwich GD. Hyaluronic acid hydrogels for biomedi-

cal applications. Adv Mater 2011;32:H41–H56.

59. Kommireddy DA, Lvov Y, Mills DK. Nanoparticle multilayers:

Surface modification for cell attachment and growth. J Biomed

Nanotechnol 2005;1:286–290.

60. Kommireddy DA, Patel S, Shutava T, Mills DK, Lvov Y. Layer-

by-layer assembly of TiO2 nanoparticles for stable hydrophilic

biocompatible coatings. J Nanosci Nanotechnol 2005;5:1081–1087.

61. Karnik S. Bioactive hydrogels for TMJ repair. MS thesis: Louisiana

Tech University; 2011.

62. Karnik SJ, Mills DK. Bioactive hydrogels for TMJ repair. FASEB J

2012;26:917.

63. Karnik S, Mills DK. Clay nanotubes as growth factor delivery vehi-

cle for bone tissue engineering. J Nanomed Nanotechnol 2013;4:

104.

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2014 VOL 00A, ISSUE 00 11

ORIGINAL ARTICLE