Effect of fetal bovine serum on mineralization in silkfibroin scaffoldsCitation for published version (APA):Vetsch, J. R., Paulsen, S. J., Müller, R., & Hofmann, S. (2015). Effect of fetal bovine serum on mineralization insilk fibroin scaffolds. Acta Biomaterialia, 13, 277-285. DOI: 10.1016/j.actbio.2014.11.025

DOI:10.1016/j.actbio.2014.11.025

Document status and date:Published: 01/01/2015

Document Version:Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can beimportant differences between the submitted version and the official published version of record. Peopleinterested in the research are advised to contact the author for the final version of the publication, or visit theDOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and pagenumbers.Link to publication

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, pleasefollow below link for the End User Agreement:

www.tue.nl/taverne

Take down policyIf you believe that this document breaches copyright please contact us at:

openaccess@tue.nl

providing details and we will investigate your claim.

Download date: 07. Aug. 2019

Acta Biomaterialia 13 (2015) 277–285

Contents lists available at ScienceDirect

Acta Biomaterialia

journal homepage: www.elsevier .com/locate /actabiomat

Effect of fetal bovine serum on mineralization in silk fibroin scaffolds

http://dx.doi.org/10.1016/j.actbio.2014.11.0251742-7061/� 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Technische Universiteit Eindhoven, P.O. Box 513,5600MB Eindhoven, The Netherlands. Tel.: +31 40 247 3494; fax: +31 40 247 3744.

E-mail address: S.Hofmann.Boss@tue.nl (S. Hofmann).

Jolanda R. Vetsch a, Samantha J. Paulsen a, Ralph Müller a, Sandra Hofmann a,b,c,⇑a Institute for Biomechanics, Swiss Federal Institute of Technology Zürich (ETHZ), Vladimir-Prelog-Weg 3, Zurich 8093, Switzerlandb Department of Biomedical Engineering, Eindhoven University of Technology, PO Box 513, Eindhoven 5600MB, The Netherlandsc Institute for Complex Molecular Systems, Eindhoven University of Technology, PO Box 513, Eindhoven 5600MB, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history:Received 21 August 2014Received in revised form 5 November 2014Accepted 13 November 2014Available online 20 November 2014

Keywords:Fetal bovine serumSilk fibroinScaffoldsBone tissue engineeringSpontaneous mineralization

Fetal bovine serum (FBS) is a common media supplement used in tissue engineering (TE) cultures. Thechemical composition of FBS is known to be highly variable between different brands, types or batchesand can have a significant impact on cell function. This study investigated the influence of four differentFBS types in osteogenic or control medium on mineralization of acellular and cell-seeded silk fibroin (SF)scaffolds. In bone TE, mineralized tissue is considered as the final product of a successful cell culture. Cal-cium assays and micro-computed tomography scans revealed spontaneous mineralization on SF scaffoldswith certain FBS types, even without cells present. In contrast, cell-mediated mineralization was foundunder osteogenic conditions only. Fourier transform infrared spectroscopy analysis demonstrated a sim-ilar ion composition of the mineralization present in scaffolds, whether cell-mediated or spontaneous.These results were confirmed by scanning electron microscopy. This study shows clear evidence forthe influence of FBS type on mineralization on SF scaffolds. The suitability of FBS medium supplementa-tion in TE studies is highly questionable with regard to reproducibility of studies and comparability ofobtained results. For future TE studies, alternatives to conventional FBS such as defined FBS or serum-freemedia should be considered, as suggested decades ago.

� 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Fetal bovine serum (FBS) is a nutritional serum supplementused for most cell cultures. It contains important basic proteinssuch as growth factors and hormones for maintaining cell survival,growth and division. The chemical composition of FBS typicallyvaries from batch to batch, between different types and among dif-ferent brands due to biological variance [1,2]. FBS contains manyunknown substances with unclear functions on cultured cells thatmay alter the outcome of cell experiments [1]. Today, various typesof defined FBS are commercially available. Defined FBS is chro-matographically purified, then individual constituents are sepa-rated and recombined into a defined composition.

Biomineralization is a process describing mechanisms of miner-alized tissue formation by organisms in nature [3,4]. Biomineralsare hybrid structures composed of both minerals and organic com-ponents [4]. Mineral deposition starts by the formation of prenu-cleation clusters at a templating surface initiated by localsupersaturation of ions [5,6]. The formation of these ion clustersis highly dependent on the fluid surrounding the underlying struc-

ture of the inorganic matrix [4,6]. The proteins of the inorganicmatrix interact with ions of the surrounding fluid and facilitatethe formation of mineralized macromolecules [5].

The most abundant mineralized tissue in the human body isbone [5]. Bone is a composite material consisting of hydroxyapatite(HA) and collagen type I (Col I) fibrils [7]. In vivo bone mineraliza-tion is a cell-mediated process. Osteoblasts are responsible for thedeposition of bone matrix into the extracellular space. Col I is themajor component of bone matrix [8] and is a fibrous protein con-taining repetitive amino acids (Fig. 1A). Col I builds the three-dimensional (3-D) framework of bone on which bone mineral isdeposited [4]. The mineralization process in bone occurs by nucle-ation of HA out of calcium and phosphate ions in solution. Theorganic Col I acts as a substrate for the mineralization of HA crys-tals that mineralize in thin layers between sequences of Col I mol-ecules (Fig. 1A) [9].

Like Col I, silk fibroin (SF) is a fibrous protein. It is synthesizedby Lepidoptera larvae [10]. SF protein side chains interdigitateand form antiparallel plated b-sheets (Fig. 1B). Highly and lessordered b-sheets are connected by amorphous network chains[11]. SF is a widely used scaffold material for bone tissue engineer-ing (TE) applications [12–16], due to its excellent biocompatibility[17], controllable degradation [18] and favorable mechanical prop-erties [19]. SF is an interesting scaffold material for bone TE consid-

278 J.R. Vetsch et al. / Acta Biomaterialia 13 (2015) 277–285

ering its ability to regulate the formation of HA nanocrystals whenexposed to simulated body fluid (SBF). SBF mimics ion concentra-tions of human blood serum [20,21]. It was shown that SF hasthe potential to mineralize spontaneously in SBF by inducing apa-tite deposition on its surface and provoking continuous growth andenrichment of HA crystals [21,22]. Marelli et al. [11] have shownthat the amorphous connections between the b-sheets of SF actas nucleation sites for HA crystals similar to Col I in bone. Sponta-neous deposition of mineralization on SF was also shown in cal-cium chloride (CaCl2) solution. Choi et al. [23] managed topromote calcium deficient HA formation on SF particles in CaCl2

solution, due to electrostatic interactions between the calcium ionsand the functional groups of SF [23]. Ion compositions in FBS andSBF are very similar and the pH, an important environmental factorinfluencing spontaneous mineralization, of both solutions is buf-fered to 7.4 [20,24].

In bone TE applications, the formation of mineralized extracel-lular matrix by cells seeded on a 3-D scaffold is considered as thefinal product of a successful culture. Still, obtained results varyhighly within and between research groups and reproducibilityof studies is often not given. If FBS type (or even batch) has aninfluence on mineralization due to variations in chemical composi-tion between different suppliers, this effect needs to be quantifiedor preferably avoided in order to draw conclusions on how the cel-lular environment influences TE outcomes. There are some studieswhich investigated the effect of FBS on mineralization focusing onthe cellular response, but without having a closer look at the role ofthe materials used [25,26]. The objective of this paper was to

Fig. 1. Comparison of collagen and silk fibroin structures. Both fibrous proteinsshow a hierarchical structure with repeating amino acids and have been reported totake part in mineralization processes. (A) Structure of collagen with repeatingamino acids of glycine (Gly), proline (Pro) and hydroxyproline (Hyp). (B) Structureof silk fibroin with repeating amino acids of glycine (Gly), serine (Ser) and alanine(Ala).

investigate the influence of four different FBS types on mineralizedtissue formation in cell-seeded and acellular SF scaffolds. Two con-ventional and two defined FBS types were compared against eachother. Additionally, the influence of osteogenic factor supplemen-tation on mineralized tissue formation was evaluated.

2. Materials and methods

Details of all FBS types used (full label, order number, batchnumber and detailed supplier information), all experimental con-ditions (FBS type, medium type and cell use) and all assays per-formed (including number of samples evaluated for eachexperimental condition) can be found in Table 1.

2.1. Materials

The four different FBS types used were chosen according to thefollowing criteria. Gibco Standard is the standard FBS currentlyused in our lab. Like Gibco Standard, PAA Standard is a conven-tional FBS and was chosen as a direct control for Gibco Standard.PAA Gold and Biochrom Superior are both defined FBS types andwere chosen because the suppliers assert no need for batch-to-batch testing. Dulbecco’s modified Eagle medium (DMEM), antibi-otic/antimycotic (Anti-Anti) and trypsin were from Gibco (Zug,Switzerland). 1,1,1,3,3,3-hexafluoroisopropanol (HFIP) wasobtained from abcr chemicals (Karlsruhe, Germany). Methanol(MeOH) was from Merck (Zug, Switzerland) and lithium bromide(LiBr) from Thermo Fisher Scientific (Reinach, Switzerland). Phos-phate buffered saline (PBS) was supplied from Medicago (Uppsala,Sweden). All other substances were of analytical or pharmaceuticalgrade and obtained from Sigma (Buchs, Switzerland). Silkwormcocoons were kindly provided by Trudel Inc. (Zurich, Switzerland).

2.2. Scaffolds

SF scaffolds were produced as previously described [14,27].Briefly, silk from Bombyx mori L. silkworm cocoons was degummedby boiling in 0.2 M Na2CO3 twice for 1 h. Dried silk was dissolved in9 M LiBr and dialyzed against ultra pure water (UPW) for 36 husing Slide-A-Lyzer cassettes (molecular weight cutoff: 3.5 K;Thermo Fisher Scientific, Reinach, Switzerland). Dialyzed silk solu-tion was lyophilized (Alpha 1-2, Martin Christ GmbH, Osterode amHarz, Germany) for 4 days and dissolved in HFIP, resulting in a 17%(w/v) solution. 1 ml of dissolved silk was added to 2.5 g NaCl with agranule size of 300–400 lm and was allowed to air dry for 3 days.Silk-salt blocks were immersed in 90% MeOH for 30 min to induceb-sheet formation [28]. NaCl was extracted from dried blocks indeionized water for 2 days. Scaffolds were cut into disks of 5 mmin diameter and 3 mm in height and autoclaved in PBS at 121 �Cfor 20 min.

2.3. Cell culture

Human mesenchymal stem cell (hMSC) isolation from humanbone marrow (Lonza, Walkersville, MD, USA) was performed aspreviously described [29]. Passage 3 hMSCs were expanded for7 days. At day 7 hMSCs were trypsinized and half of all scaffoldswere seeded with 1 million cells per scaffold, while the remainingscaffolds were left acellular. All scaffolds were incubated in wells ofa 24-well plate at 37 �C for 90 min to allow cell attachment. Subse-quently, half of all cell-seeded scaffolds and half of all acellularscaffolds were provided with 1 ml control medium (DMEM supple-mented with 10% FBS and 1% Anti-Anti) and were incubated at37 �C and 5% CO2. The remaining scaffolds were incubated in1 ml osteogenic medium (control medium supplemented with

Table 1Supplier information of each FBS type, experimental conditions with group numbers indicated and assays performed including numbers of samples evaluated for eachexperimental condition per time point.

FBS supplier information PAA Gold PAA Standard Biochrom Superior Gibco Standard

Full label PAA FBS Gold, EU approved PAA FBS standard quality, EUapproved

Biochrom FBS Superior Gibco�, FBS, Qualified, EU approved, SouthAmerica Origin

Abbreviation PAA Gold PAA Standard Biochrom Superior Gibco StandardOrder number A15-151 A15-101 S 0615 10270-106Batch number A15110-2039 A10110-2069 0306 A 41Q6208 KSupplier Chemie Brunschwig, Basel, Switzerland Chemie Brunschwig, Basel,

SwitzerlandBiochrom AG, Berlin,Germany

Gibco, Zug, Switzerland

Experimental conditions PAA Gold PAA Standard Biochrom Superior Gibco Standard

# group number 1. Control acellular2. Osteo acellular3. Control cell-seeded4. Osteo cell-seeded

1. Control acellular2. Osteo acellular3. Control cell-seeded4. Osteo cell-seeded

1. Control acellular2. Osteo acellular3. Control cell-seeded4. Osteo cell-seeded

1. Control acellular2. Osteo acellular3. Control cell-seeded4. Osteo cell-seeded

N = 44 per group

Assays performed Week 3: Week 5: Week 7:

(sample number)Calcium assay (N = 10) Calcium assay (N = 10) Calcium assay (N = 10)

lCT (N = 10)FTIR (N = 2)SEM (N = 2)

J.R. Vetsch et al. / Acta Biomaterialia 13 (2015) 277–285 279

50 lg ml�1 L-ascorbic acid 2-phosphate, 100 nM dexamethasoneand 10 mM b-glycerophosphate (b-GP)). Every culture mediumwas prepared with every type of FBS, resulting in 16 experimentalconditions (Table 1).

2.4. Calcium assay

A spectrophotometric calcium assay was performed accordingto the manufacturer’s instruction (Calcium CPC FS, Rolf GreinerBioChemica, Flacht, Germany). After 3, 5 and 7 weeks of culture,10 scaffolds per group were washed with PBS and disintegratedin 5% aqueous trichloroacetic acid (TCA) using steel balls and aMinibeadBeater™ (Biospec, Bartlesville, OK, USA). Scaffolds werefurther incubated in 5% TCA at room temperature for 48 h and fro-zen until evaluation.

For the correlation of calcium assay and micro-computedtomography (lCT) data, 20 scaffolds were analyzed after lCT scan-ning for calcium amount. After the lCT scans, scaffolds were disin-tegrated and incubated as described before. For analysis, allscaffolds were centrifuged at 3000g for 10 min. Supernatant wasfurther diluted in 5% TCA at a ratio of 1:20 (v/v) and the colorintensity of the resulting solution was measured with a microplatereader (Infinite 200 PRO, Tecan Group Ltd, Männedorf,Switzerland).

2.5. lCT

lCT scans were performed on scaffolds after 7 weeks of culture(N = 10 per group). Scaffolds were washed with PBS, lyophilizedovernight (Alpha 1-2, Martin Christ GmbH, Osterode am Harz, Ger-many) and scanned in air on a lCT 40 (Scanco Medical, Brütisellen,Switzerland). The energy level was set to 45 kVp, intensity to177 lA, integration time to 200 ms and a frame averaging of 4was applied. The scans were executed at medium resolution modewith a nominal resolution of 12 lm. After reconstruction, scanswere filtered applying a 3-D constrained Gaussian filter with finitefilter support (1 voxel) and filter width (sigma = 1.2). Filtered gray-scale images were segmented at a global threshold of 17% of themaximal grayscale value (corresponding to a density value of125.17 mg HA cm�3). Unconnected objects smaller than 50 voxelswere removed and neglected for further analysis. For the final

evaluation an overall mask (OM) with radius = Ro and height = Hwas generated (Fig. 2A). Ro and H were chosen to fit to the outerboundaries of each scaffold individually due to small differencesin scaffold dimensions especially present between mineralizedand non-mineralized scaffolds. The resulting 3-D volume was eval-uated morphometrically for mineralized tissue volume (BV) andmineralized tissue volume fraction (BV/TV = mineralized tissuevolume/total volume), as described previously for bone [30,31].All scaffolds were grouped into either an ‘‘acellular’’ or a‘‘cell-seeded’’ group to assess differences in spatial distribution ofspontaneously formed mineralization on acellular scaffolds andof cell-mediated mineralization observed on cell-seeded scaffolds.An inner mask (IM) was generated based on the geometry of theOM (Fig. 2B). The percentage of total BV within the IM was calcu-lated based on the BV of the OM. The radius (Ri) of the IM equaled50% � Ro. H was kept constant (Fig. 2B). Bone mineral content(BMC) per scaffold as determined by lCT was correlated with cal-cium amount per scaffold as measured biochemically.

2.6. Fourier transform infrared spectroscopy (FTIR)

After 7 weeks of culture, scaffolds (N = 2 per group) werewashed with PBS, frozen at �80 �C overnight and lyophilized for1 day. For FTIR analysis dried scaffolds were ground with KBr at4% (w/w). 100% KBr was used as background medium. The groundsamples were analyzed using a Bruker Tensor 27 FTIR spectroscope(Bruker Optics GmbH, Fällanden, Switzerland) at a resolution of4 cm�1. Scans were performed in full range from 4000 cm�1 to400 cm�1 at a scan time of 100 ms. Measurements were convertedto absorbance spectra in OPUS Application Executable 5.0.5(Bruker Optics GmbH, Fällanden, Switzerland). SigmaPlot 12.2(Systat Software, Inc., San Jose, CA, USA) was used todetermine exact wavenumbers for peaks and Adobe IllustratorCS5.1 (Adobe Systems, Inc., San Jose, CA, USA) was used to markpeaks.

2.7. Scanning electron microscopy (SEM)

After 7 weeks of culture, scaffolds (N = 2 per group) werewashed in PBS and fixed in 2.5% glutaraldehyde solution in 0.1 Mcacodylate buffer (pH 7.4) at 4 �C for 4 h. After rinsing in 0.1 M

Fig. 2. Masks applied for the evaluation of lCT data. (A) OM. The OM was fitted to the whole volume of each scaffold individually. An outer scaffold radius (Ro) and thescaffold height (H) were chosen to fit to the dimensions of each scaffold. (B) IM. The IM mask was created based on the OM with a radius of Ri = 50% * Ro. The height (H) of bothmasks was kept constant.

Fig. 3. Calcium assay and lCT data. Calcium amount per scaffold in lg after: (A) 3 weeks, (B) 5 weeks and (C) 7 weeks of culture. (D) Mineralized tissue volume fraction after7 weeks of culture. Lines represent significant differences between FBS types under the same culture conditions. abc represent significant differences from control acellular,osteogenic acellular and control cell-seeded conditions, respectively, within the same type of FBS. Error bars represent mean ± standard deviation. Results are consideredstatistically significant at P 6 0.05. N = 10 per group for calcium assay and lCT data.

280 J.R. Vetsch et al. / Acta Biomaterialia 13 (2015) 277–285

cacodylate buffer a second fixation step was performed using 0.04%aqueous osmium tetroxide in 0.1 M cacodylate buffer at roomtemperature in the dark. Scaffolds were again rinsed in 0.1 Mcacodylate buffer, dehydrated for 10 min in serial EtOH baths of37%, 67% and three times 96%, frozen overnight in 100% EtOH at�80 �C and lyophilized. Scaffolds were coated with gold using asputter coater (Balzers SCD 030, former Balzers Union Ltd,Liechtenstein) for 90–120 s at a current of 40 mA and imaged at850� and 5000�magnification in an SEM (Leo 1530 Gemini, Zeiss,Oberkochen, Germany) operated at a voltage of 5 kV to investigatescaffold surface morphology.

2.8. Statistical analysis

Quantitative data are represented as mean ± standard deviation.IBM SPSS Statistics 20 (SPSS Inc., Chicago, IL, USA) was used for theevaluation of statistically significant differences for calcium assayand quantitative lCT data. Under the assumption of normally dis-tributed data, analysis of variance was performed followed by posthoc assessment using the Bonferroni method. Differences betweengroups were considered statistically significant at a level ofP 6 0.05. Images in Fig. 4 represent the upper median BV/TV sam-ple of each group.

Fig. 4. 3-D lCT images of scaffold mineralization for each FBS type after 7 weeks ofculture. The upper median sample of each group is shown in top view (upper row)and side view (bottom row): (A) control acellular, (B) osteogenic acellular, (C)control cell-seeded and (D) osteogenic cell-seeded conditions. Scale bar = 1 cm.

Fig. 5. BV in percentage of total BV per scaffold for the IM volume after 7 weeks ofculture. Samples with a BV < 0.001 mm3 were excluded from evaluation leading toN = 77 for the acellular and N = 52 for the cell-seeded condition. Higher BV wasobserved in the IM for acellular scaffolds. Error bars represent mean ± standarddeviation. Results are considered statistically significant at P 6 0.05.

J.R. Vetsch et al. / Acta Biomaterialia 13 (2015) 277–285 281

3. Results

3.1. Calcium assay

Calcium assays were performed to assess total calcium contentof the scaffolds, as a method to represent the total mineralized tis-sue volume per scaffold (Fig. 3A–C). As expected, cells responded tothe different culture media used. Cells seeded on scaffolds culturedin control medium remained undifferentiated, whereas cellsseeded on scaffolds in osteogenic medium underwent osteogenicdifferentiation and started to form mineralized tissue. Scaffoldscultured in osteogenic medium exhibited the highest calcium con-tents after 7 weeks (Fig. 3C). Interestingly, considerable amounts ofcalcium were detected for acellular scaffolds with certain FBS typesstarting as early as week 3 up to week 7. Elevated calcium levelswere observed particularly for acellular scaffolds cultured withBiochrom Superior and Gibco Standard (Fig. 3A–C).

Cell-seeded scaffolds cultured in osteogenic medium showedsignificantly increased calcium contents compared to cell-seededscaffolds cultured in control medium and all acellular scaffolds ateach time point (P < 0.05). No differences were observed betweencell-seeded scaffolds cultured in control medium at any time point.

Acellular scaffolds cultured in osteogenic medium with GibcoStandard showed significantly increased calcium deposition com-pared to acellular and cell-seeded scaffolds cultured in controlmedium with the same FBS type at week 3 (P < 0.05; Fig. 3A). In

week 5 a similar result was observed for acellular scaffolds cul-tured in control and osteogenic medium with Biochrom Superiorwith significantly higher calcium levels compared to their corre-sponding cell-seeded scaffolds cultured in control medium(P < 0.05; Fig. 3B). The same result was observed for acellular scaf-folds cultured in control and osteogenic medium with Gibco Stan-dard at week 7 (P < 0.05; Fig. 3C).

3.2. lCT

lCT measurements of BV/TV (Fig. 3D) confirmed calcium assaydata of week 7 (Fig. 3C). Cell-seeded scaffolds cultured in osteo-genic medium exhibited the highest BV/TV values compared tocell-seeded scaffolds cultured in control medium and all acellularscaffolds (P < 0.001). Acellular scaffolds cultured with BiochromSuperior and Gibco Standard showed significantly increased cal-cium levels compared to their corresponding cell-seeded scaffoldscultured in control medium (P < 0.05).

Mineralized tissue was located primarily on scaffold edges andthe top of the scaffolds in cell-seeded scaffolds. Acellular scaffoldscultured with Biochrom Superior and Gibco Standard showed sub-stantial mineralization (Fig. 4A and B). Very little mineralizationwas observed in acellular scaffolds cultured with PAA Gold orPAA Standard or in cell-seeded scaffolds cultured in control med-ium (Fig. 4A–C). Cell-seeded scaffolds cultured in osteogenic med-ium displayed highly mineralized structures with every type of FBS(Fig. 4D). For the evaluation of differences in spatial distribution ofacellular scaffolds and cell-seeded scaffolds only mineralized scaf-folds (BV of OM > 0.001 mm3) were considered. Acellular scaffolds(N = 77) showed significantly increased BV percentage in the IM(18.87 ± 18.32%) compared to cell-seeded scaffolds (N = 52) thatdisplayed 10.66 ± 11.16% of the total BV in the IM (P < 0.05;Fig. 5). BMC per scaffold correlated highly with calcium amountper scaffold (R2 = 0.96; Supplementary Fig. S.1A). Both measure-ment methods were in agreement and no systematic differencescould be observed between calcium assay and lCT data (Bland Alt-man Plot; Supplementary Fig. S.1B).

3.3. FTIR

FTIR was performed to analyze differences in phosphate and car-bonate precipitates on the scaffolds (Fig. 6; SupplementaryTable S.1). Typical absorption spectra of the underlying SF scaffoldwere observed between 1700 and 1000 cm�1 for every culturecondition [32]. All spectra showed features of different apatite

Fig. 6. FTIR data: (A) control acellular, (B) control cell-seeded, (C) osteogenic acellular and (D) osteogenic cell-seeded conditions. Spectra are shown between wavenumbers of1700–800 cm�1.

282 J.R. Vetsch et al. / Acta Biomaterialia 13 (2015) 277–285

precipitates. Carbonate was present in all samples [11,33] and phos-phate bands were visible between 1200 and 900 cm�1 [34–37].Cell-seeded scaffolds cultured in osteogenic medium displayed acharacteristic peak for phosphate ions (PO4

3�) at 1038 cm�1, indicat-ing a higher amount of mineralized tissue formation (Fig. 6D)[11,34]. Overlapping phosphate bands between 1200 and900 cm�1 present in spectra of acellular scaffolds might havemasked less prominent phosphate peaks.

3.4. SEM

SEM micrographs confirmed the results of the calcium assayand lCT data (Fig. 7; Supplementary Fig. S.2). Cell-seeded scaffoldsshowed extracellular matrix covering the surface of the scaffolds(Supplementary Fig. S.2I–P). No mineralization clusters were visi-ble on acellular scaffolds cultured in osteogenic medium scaffoldswith PAA Gold (Fig. 7A and C). In contrast, clearly visible mineral-ization clusters were observed with Gibco Standard under thesame culture conditions (Fig. 7B and D).

4. Discussion

FBS has been used for more than a century as a cell culture addi-tive for most cell types cultured in vitro [38,39]. It has been previ-ously shown that differences in serum composition lead tostatistical differences in experimental outcomes [40]. The disad-vantages of FBS, such as undefined composition or batch-to-batch

variations, have been known for decades. Alternatives to FBS havebeen proposed and include options such as sera from other species,reduced serum media, chemically defined FBS or chemicallydefined serum-free media [38,39]. Using defined FBS would elimi-nate the necessity to test every new batch of FBS and would min-imize experimental differences due to different FBS batches ortypes. Nevertheless, if correct batch testing is performed, experi-mental differences could already be reduced. It is important thatevery FBS batch is tested, preferably under real culture conditions.

Very few publications can be found discussing differences inexperimental outcomes provoked by FBS. The exact ordering andbatch number of the FBS used in experiments is not required forpublications and therefore a direct comparison of experimentsfrom different laboratories seems impossible. It is very difficultto assess whether experimental outcomes occur due to the exper-imental conditions or if they are attributed to the FBS type orbatch. Differences in FBS composition may be a major reasonwhy different labs are not able to reproduce data published inthe literature.

In this study, the effect of four different FBS types on mineral-ized tissue formation on SF scaffolds was investigated. The influ-ence of FBS on mineralization is not new and has been addressedearlier. For example, mineral precipitation on decalcified newbornrat tibia in the absence of cells has been shown to occur when cul-ture media were supplemented with FBS and osteogenic factors (b-GP) [26], but not with FBS-containing control media only as shownwith this study. Evident effects of FBS on mineralization, alkalinephosphatase activity and osteogenic marker expression based on

Fig. 7. SEM images of acellular scaffolds cultured in osteogenic medium. (A, B) Overview pictures of one characteristic scaffold per group at a magnification of 850� and (C, D)close-up pictures of the area marked in the respective pictures in (A) or (B) at a magnification of 5000�. Mineralized clusters are present on the surface of scaffolds culturedwith Gibco Standard. No mineralized clusters and a smooth scaffold surface can be observed for scaffolds cultured with PAA Gold. Scale bar: 60 lm (A, B) and 10 lm (C, D).

J.R. Vetsch et al. / Acta Biomaterialia 13 (2015) 277–285 283

serum conditions have also been shown for human adipose stemcells, but the results were not compared to acellular culture condi-tions and therefore the influence of the material could not bedetermined [25]. The results observed in the present studyrevealed that the FBS type influences mineralization formationon acellular and cell-seeded SF scaffolds significantly. The additionof osteogenic factors did not have an influence on mineralized tis-sue formation in acellular scaffolds, but led to osteogenic differen-tiation with cell-seeded scaffolds.

Acellular scaffolds showed an unexpected and completely dif-ferent mineralization pattern compared to cell-seeded scaffolds.Scaffolds cultured with Biochrom Superior and Gibco Standardshow strong evidence of spontaneous mineralization on acellularSF scaffolds presenting calcium values up to one third of the high-est calcium level measured overall. The amount of spontaneousmineralization of acellular scaffolds was therefore clearly depen-dent on the FBS type used. Interestingly, spontaneous mineraliza-tion occurred even under control conditions. This result iscontradictory to previous results where mineral precipitation incell-free conditions only occurred with FBS containing media sup-plemented with b-GP [26].

The amount of spontaneous mineralization is dependent on theion concentration of the solution surrounding the substrate [21].Differences in spontaneous mineralization between the four FBStypes used can be explained by differences in elemental composi-tion. It is known that sera exhibit significant differences in compo-sition between different types and batches [2,40]. The elementalcomposition of FBS, or sera in general, is not defined. Only somecommon components are defined for all FBS types (SupplementaryTable S.2). Basic properties like pH or osmolality, primary compo-nents like Hemoglobin or proteins and the absence of bacteria,fungi and mycoplasma were tested for all the four FBS types used.

The ion concentration was stated for Biochrom Superior only (Sup-plementary Table S.2). Hamlin and Price [26] hypothesized thatthere exist one or more nucleators of bone mineralization in serumthat escaped from the bone matrix to the blood. For future studiesit would be helpful to identify key components that contribute tomineralization.

Cell-seeded scaffolds showed mineralization when culturedwith osteogenic medium only (Figs. 3 and 4). It is known thathMSCs differentiate towards the osteogenic lineage with the addi-tion of osteogenic factors to the medium [41]. The presence of min-eralization in these scaffolds shows a successful differentiation ofthe hMSCs towards osteoblasts. The amount of mineralizationformed was dependent on the FBS type, especially at early timepoints (week 3 and 5; Fig. 3A and B), which is in agreement withprevious results [25].

In cell-seeded scaffolds, we assume that cells prevented sponta-neous mineralization by either changing the ion concentration ofthe surrounding culture medium or by inhibiting the interactionbetween ions in the culture medium with either the SF surfaceitself or the proteins adsorbed to the surface of the scaffold. Underosteogenic conditions cells seem to metabolize the ions in the cul-ture medium to build up mineralized tissue.

The formation of spontaneous mineralization under controlconditions on acellular scaffolds compared to cell-seeded scaffolds,however, is not fully understood. One possibility is that cells cul-tured with control medium take up ions from the surroundingmedium like cells cultured under osteogenic conditions. In con-trast, cells subsequently are not able to form mineralized tissue,because they are not differentiated towards the osteogenic lineage.Another possible mechanism inhibiting spontaneous mineraliza-tion in cell-seeded scaffolds under control conditions could be anactive inhibition of mineralization by the hMSCs itself. It is known

284 J.R. Vetsch et al. / Acta Biomaterialia 13 (2015) 277–285

that some cell types are able to inhibit mineralization [42]. So far,active inhibition of mineralization by hMSCs could not be shownand further investigation would be needed to test this hypothesis.Spontaneous mineralization of acellular SF has been describedbefore in SBF or CaCl2 [21–23] but, to our knowledge, it was nevershown in control or osteogenic medium or on other scaffoldmaterials.

The underlying silk structure is thought to provide an idealenvironment for spontaneous mineralization, because its chemicalstructure is very similar to Col I (Fig. 1). Comparisons of minerali-zation on different scaffold materials are therefore critical, becausesome materials could induce spontaneous mineralization moreeffectively than others because of their different chemical struc-ture. Therefore, it is necessary to include acellular control groupsto test spontaneous mineralization potential of each scaffold mate-rial with the respective culture media used.

It is known that in static cell cultures seeded cells tend to con-centrate on the outer scaffold surface, leading to a poor nutrientand waste exchange to the center of the scaffold [43]. Because ofthis effect the cell-mediated mineralization is expected to belocated predominantly at the outer scaffold surface in the pre-sented study setup. This effect was confirmed by lCT scans show-ing mineralized tissue formation located mostly at the edges andthe top of the scaffolds (Fig. 4D). Ions are small enough to diffusethroughout the whole scaffold volume easily. Because spontaneousmineralization is dependent on local ion concentration [21], thespatial distribution of spontaneous mineralization on acellularscaffolds was expected to be more homogeneous compared tocell-seeded scaffolds. Quantitative evaluation of spatial distribu-tion of mineralization volume confirmed this hypothesis andshowed a significantly increased percentage of total BV in the cen-ter of acellular scaffolds (Fig. 5).

SEM provided additional visual confirmation of the resultsdetected with the calcium assay and lCT (SupplementaryFig. S.2). The absence of cells in acellular scaffolds was confirmed,and indicated that the mineralization in acellular scaffolds ispurely attributable to the spontaneous mineralization phenomena.

FTIR revealed that the mineralization formed in all of the 16experimental groups, whether cell-mediated or not, consisted ofsimilar phosphate and carbonate precipitates. Unfortunately, FTIRlacks the possibility to clearly distinguish between all differentprecipitates present, especially when they are available in smallamounts. FTIR gives information about the chemical compositionof the samples only. In addition to FTIR, X-ray diffraction couldbe used to determine crystallinity (size and orientation) to com-pare the mineralization formed to in vivo bone mineralization. Inaddition to that, transmission electron microscopy can be per-formed to investigate the orientation of mineral crystals relativeto the orientation of collagen fibrils. To characterize mineralmechanical properties of the formed mineralization atomic forcemicroscopy can be used [44]. The solubility behavior of a mineralphase could reveal further information about the composition ofthe solid phase [45].

With respect to clinical applications it is of great importancethat the effects of FBS on cell cultures are known and animal com-ponents are avoided. Human serum is an interesting alternative toanimal-derived serum. It seems attractive for the culture of tissue-engineered grafts for future implantation and clinical cell therapyapplications. Because of the clinical relevance of human serum, itwould be interesting to investigate its effect on the mineralizationon acellular and cell-seeded SF scaffolds. Nevertheless, results thathave been observed with human serum are still conflicting. The useof human serum, however, makes sense only if enough autologousserum is available. Due to the limited availability and high variabil-ity within different donors, human serum is not a reasonableoption for large-scale cell cultures in clinical applications [46]

and therefore it is highly recommended to look for serum-freealternatives for in vitro cultures.

The use of conventional FBS for cell culture is questionable andshould be considered with reservation. Not only cell growth, butalso cell viability, cell attachment, cell density, geometrical envi-ronment (2-D/3-D) and scaffold material should be consideredfor every FBS type used [38]. Chemically defined FBS, like PAA Gold,could prevent excessive and time-consuming batch-to-batch test-ing. Ideally, the use of defined FBS is to be preferred over conven-tional FBS.

It should be noted that the results of this paper are valid only forthe described example of bone TE with hMSCs and 3-D SF scaffoldsprepared by the methods described herein. However, it can beassumed that similar results could be expected for any other out-put parameter in any other type of static cell culture. The effectof different FBS types in dynamic cell cultures needs to be investi-gated more precisely. Results are expected to be different fordynamic cultures because the scaffold-surrounding medium ismoving, which could highly influence total ion concentration inthe scaffold and therefore the process of spontaneous mineraliza-tion. This study is also limited to the four different FBS types usedand batch variations for each individual FBS type were not investi-gated. It has to be expected that other FBS types or batches showdifferent mineralization amounts and patterns on SF scaffolds.

It is important that these differences will be revealed in the nearfuture. Hopefully, cell culture will soon be performed using serum-free medium if possible. If, for any reason, serum-free medium can-not be used, suitable FBS pre-testing needs to be performed and itshould be a matter to state at least FBS type and batch used in pub-lications. Alternatively, standards for batch testing procedures forspecific applications should be determined, similar to cell charac-terization protocols. If these recommendations are followed inthe future, it is possible that the complexity of TE cultures can bereduced, leading to more robust and more comparable results.

5. Conclusions

The objective of this study was to investigate the effect of differ-ent FBS types on TE cultures; in this example, on bone TE culturesof hMSCs on 3-D SF scaffolds. The results provide clear evidence ofthe strong influence of FBS type on mineralization in SF scaffoldsusing different evaluation methods. Unexpectedly, spontaneousmineralization was observed in acellular scaffolds dependent onFBS type. Only two of the four FBS types used induced spontaneousmineralization under acellular conditions. The addition of osteo-genic factors to the medium did not influence spontaneous miner-alization. No mineralization was present in cell-seeded scaffoldscultured with control medium, which indicates that cells areinvolved in the prevention of spontaneous mineralization. Theobserved results emphasize a clear dependence of experimentaloutcomes on FBS type. The use of FBS in TE cultures is thereforehighly questionable and alternatives to conventional FBS shouldbe considered, as already suggested decades ago. It is importantto solve these issues now in order to reduce variability andimprove experimental comparability to promote the entire fieldof TE.

Conflicts of Interest

The authors declare that there is no conflict of interest.

Acknowledgements

The authors would like to acknowledge Trudel Inc. (Zurich,Switzerland) for the kind supply of silkworm cocoons. We thank

J.R. Vetsch et al. / Acta Biomaterialia 13 (2015) 277–285 285

Dr. Dirk Mohn for his help in FTIR data acquisition and evaluation.This project was supported by the European Union (EU Project No.FP7-NMP-2010-LARGE-4: BIODESIGN – Rational Bioactive Materi-als Design for Tissue Regeneration).

Appendix A. Figures with essential colour discrimination

Certain figure in this article, particularly Fig. 6 is difficult tointerpret in black and white. The full colour images can be foundin the on-line version, at http://dx.doi.org/10.1016/j.actbio.2014.11.025.

Appendix B. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.actbio.2014.11.025.

References

[1] Jochems CEA, van der Valk JBF, Stafleu FR, Baumans V. The use of fetal bovineserum: ethical or scientific problem? Altern Lab Anim 2002;30:219–27.

[2] Price PJ, Gregory EA. Relationship between invitro growth promotion andbiophysical and biochemical-properties of the serum supplement. In Vitro-JTissue Cult Assoc 1982;18:576–84.

[3] Sarikaya M. Biomimetics: materials fabrication through biology. Proc NatlAcad Sci USA 1999;96:14183–5.

[4] Weiner S, Dove PM. An overview of biomineralization processes and theproblem of the vital effect. Biomineralization 2003;54:1–29.

[5] Addadi L, Weiner S. Control and design principles in biological mineralization.Angew Chem Int Ed Engl 1992;31:153–69.

[6] Dey A, Bomans PH, Muller FA, Will J, Frederik PM, de With G, et al. The role ofprenucleation clusters in surface-induced calcium phosphate crystallization.Nat Mater 2010;9:1010–4.

[7] Rho JY, Kuhn-Spearing L, Zioupos P. Mechanical properties and the hierarchicalstructure of bone. Med Eng Phys 1998;20:92–102.

[8] Mackie EJ. Osteoblasts: novel roles in orchestration of skeletal architecture. IntJ Biochem Cell Biol 2003;35:1301–5.

[9] Fratzl P, Fratzl-Zelman N, Klaushofer K, Vogl G, Koller K. Nucleation andgrowth of mineral crystals in bone studied by small-angle X-ray scattering.Calcif Tissue Int 1991;48:407–13.

[10] Kaplan DL, Mello CM, Arcidiacono S, Fossey S, Senecal K, Muller W. Silk. In:McGrath K, Kaplan D, editors. Protein-based materials. Boston: Birkhäuser;1997. p. 103–31.

[11] Marelli B, Ghezzi CE, Alessandrino A, Barralet JE, Freddi G, Nazhat SN. Silkfibroin derived polypeptide-induced biomineralization of collagen.Biomaterials 2012;33:102–8.

[12] Hagenmüller H, Hitz M, Merkle HP, Meinel L, Müller R. Design and validationof a novel bioreactor principle to combine online micro-computedtomography monitoring and mechanical loading in bone tissue engineering.Rev Sci Instrum 2010;81:014303.

[13] Meinel L, Fajardo R, Hofmann S, Langer R, Chen J, Snyder B, et al. Silk implantsfor the healing of critical size bone defects. Bone 2005;37:688–98.

[14] Meinel L, Karageorgiou V, Hofmann S, Fajardo R, Snyder B, Li C, et al.Engineering bone-like tissue in vitro using human bone marrow stem cells andsilk scaffolds. J Biomed Mater Res Part A 2004;71:25–34.

[15] Sofia S, McCarthy MB, Gronowicz G, Kaplan DL. Functionalized silk-basedbiomaterials for bone formation. J Biomedical Mater Res 2001;54:139–48.

[16] Uebersax L, Hagenmuller H, Hofmann S, Gruenblatt E, Muller R, Vunjak-Novakovic G, et al. Effect of scaffold design on bone morphology in vitro.Tissue Eng 2006;12:3417–29.

[17] Altman GH, Horan RL, Lu HH, Moreau J, Martin I, Richmond JC, et al. Silk matrixfor tissue engineered anterior cruciate ligaments. Biomaterials2002;23:4131–41.

[18] Arai T, Freddi G, Innocenti R, Tsukada M. Biodegradation of Bombyx mori silkfibroin fibers and films. J Appl Polym Sci 2004;91:2383–90.

[19] Perez-Rigueiro J, Viney C, Llorca J, Elices M. Mechanical properties of single-brin silkworm silk. J Appl Polym Sci 2000;75:1270–7.

[20] Kokubo T, Takadama H. How useful is SBF in predicting in vivo bonebioactivity? Biomaterials 2006;27:2907–15.

[21] Takeuchi A, Ohtsuki C, Miyazaki T, Tanaka H, Yamazaki M, Tanihara M.Deposition of bone-like apatite on silk fiber in a solution that mimicsextracellular fluid. J Biomed Mater Res Part A 2003;65:283–9.

[22] Lin F, Li YC, Jin J, Cai YR, Wei KM, Yao JM. Deposition behavior and properties ofsilk fibroin scaffolds soaked in simulated body fluid. Mater Chem Phys2008;111:92–7.

[23] Choi Y, Cho SY, Park DJ, Park HH, Heo S, Jin HJ. Silk fibroin particles astemplates for mineralization of calcium-deficient hydroxyapatite. J BiomedMater Res Part B 2012;100:2029–34.

[24] Lindl T. Zell- und gewebekultur. 4th ed. Heidelberg: Spektrum AkademischerVerlag; 2002.

[25] Kyllonen L, Haimi S, Mannerstrom B, Huhtala H, Rajala KM, Skottman H, et al.Effects of different serum conditions on osteogenic differentiation of humanadipose stem cells in vitro. Stem Cell Res Ther 2013;4:17.

[26] Hamlin NJ, Price PA. Mineralization of decalcified bone occurs under cellculture conditions and requires bovine serum but not cells. Calcif Tissue Int2004;75:231–42.

[27] Nazarov R, Jin HJ, Kaplan DL. Porous 3-D scaffolds from regenerated silkfibroin. Biomacromolecules 2004;5:718–26.

[28] Tsukada M, Gotoh Y, Nagura M, Minoura N, Kasai N, Freddi G. Structural-changes of silk fibroin membranes induced by immersion in methanolaqueous-solutions. J Polym Sci Part B: Polym Phys 1994;32:961–8.

[29] Hofmann S, Hagenmüller H, Koch AM, Müller R, Vunjak-Novakovic G, KaplanDL, et al. Control of in vitro tissue-engineered bone-like structures usinghuman mesenchymal stem cells and porous silk scaffolds. Biomaterials2007;28:1152–62.

[30] Hildebrand T, Laib A, Müller R, Dequeker J, Ruegsegger P. Direct three-dimensional morphometric analysis of human cancellous bone:microstructural data from spine, femur, iliac crest, and calcaneus. J BoneMiner Res 1999;14:1167–74.

[31] Van Lenthe GH, Hagenmüller H, Bohner M, Hollister SJ, Meinel L, Müller R.Nondestructive micro-computed tomography for biological imaging andquantification of scaffold-bone interaction in vivo. Biomaterials2007;28:2479–90.

[32] Wilson D, Valluzzi R, Kaplan D. Conformational transitions in model silkpeptides. Biophys J 2000;78:2690–701.

[33] Hossain S, Yamamoto H, Chowdhury EH, Wu X, Hirose H, Haque A, et al.Fabrication and intracellular delivery of doxorubicin/carbonate apatitenanocomposites: effect on growth retardation of established colon tumor.PLoS One 2013;8:e60428.

[34] Chang MC, Tanaka J. FT-IR study for hydroxyapatite/collagen nanocompositecross-linked by glutaraldehyde. Biomaterials 2002;23:4811–8.

[35] Stoch A, Jastrzebski W, Brozek A, Stoch J, Szaraniec J, Trybalska B, et al. FTIRabsorption-reflection study of biomimetic growth of phosphates on titaniumimplants. J Mol Struct 2000;555:375–82.

[36] Drouet C. Apatite formation: why it may not work as planned, and how toconclusively identify apatite compounds. Biomed Res Int 2013;2013:490946.

[37] Li C, Vepari C, Jin HJ, Kim HJ, Kaplan DL. Electrospun silk-BMP-2 scaffolds forbone tissue engineering. Biomaterials 2006;27:3115–24.

[38] Jayme DW, Epstein DA, Conrad DR. Fetal bovine serum alternatives. Nature1988;334:547–8.

[39] Brunner D, Frank J, Appl H, Schoffl H, Pfaller W, Gstraunthaler G. Serum-freecell culture: the serum-free media interactive online database. ALTEX2010;27:53–62.

[40] Bryan N, Andrews KD, Loughran MJ, Rhodes NP, Hunt JA. Elucidating thecontribution of the elemental composition of fetal calf serum to antigenicexpression of primary human umbilical-vein endothelial cells in vitro. BiosciRep 2011;31:199–210.

[41] Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, et al.Multilineage potential of adult human mesenchymal stem cells. Science1999;284:143–7.

[42] Oshima T, Abe M, Asano J, Hara T, Kitazoe K, Sekimoto E, et al. Myeloma cellssuppress bone formation by secreting a soluble Wnt inhibitor, sFRP-2. Blood2005;106:3160–5.

[43] Yu X, Botchwey EA, Levine EM, Pollack SR, Laurencin CT. Bioreactor-basedbone tissue engineering: the influence of dynamic flow on osteoblastphenotypic expression and matrix mineralization. Proc Natl Acad Sci USA2004;101:11203–8.

[44] Boskey AL, Roy R. Cell culture systems for studies of bone and toothmineralization. Chem Rev 2008;108:4716–33.

[45] Driessens FCM, Vandijk JWE, Borggreven JMPM. Biological calcium phosphatesand their role in the physiology of bone and dental tissues I. Composition andsolubility of calcium phosphates. Calcif Tissue Res 1978;26:127–37.

[46] Bieback K, Hecker A, Kocaomer A, Lannert H, Schallmoser K, Strunk D, et al.Human alternatives to fetal bovine serum for the expansion of mesenchymalstromal cells from bone marrow. Stem Cells 2009;27:2331–41.