Available online at www.sciencedirect.com

www.elsevier.com/locate/jmbbm

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 3 1 3 – 3 1 9

http://dx.doi.org/101751-6161/& 2014 El

nCorresponding autE-mail addresses

Research Paper

Nanomechanical properties of multi-block copolymermicrospheres for drug delivery applications

P.R. Moshtagha,b, J. Raukera, M.J. Sandkerc, M.R. Zuiddamd, F.W.A. Dirned,E. Klijnstrae, L. Duquee, R. Steendame, H. Weinansa,b, A.A. Zadpoora,n

aFaculty of Mechanical, Maritime, and Materials Engineering, Delft University of Technology (TU Delft), Mekelweg 2,Delft 2628 CD, The NetherlandsbDepartment of Orthopaedics and Department of Rheumatology, University Medical Center Utrecht, Utrecht,The NetherlandscDepartment of Orthopaedics, Erasmus Medical Centre, Rotterdam, The NetherlandsdKavli Institute of Nanoscience, Delft University of Technology, Lorentzweg 1, 2628 CJ Delft, The NetherlandseInnoCorePharmaceuticals, L.J. Zielstraweg 1, Groningen 9713GX, The Netherlands

a r t i c l e i n f o

Article history:

Received 9 December 2013

Received in revised form

28 February 2014

Accepted 9 March 2014

Available online 19 March 2014

Keywords:

Nanoindentation

Mechanical properties

Hydrophilic multi-block copolymer

Monodisperse microspheres

Controlled release

.1016/j.jmbbm.2014.03.002sevier Ltd. All rights rese

hor. Tel.: þ31 15 2781021;: a.a.zadpoor@tudelft.nl,

a b s t r a c t

Biodegradable polymeric microspheres are interesting drug delivery vehicles for site-specific

sustained release of drugs used in treatment of osteoarthritis. We study the nano-

mechanical properties of microspheres composed of hydrophilic multi-block copolymers,

because the release profile of the microspheres may be dependent on the mechanical

interactions between the host tissues and the microspheres that aim to incorporate between

the cartilage surfaces. Three different sizes of monodisperse microspheres, namely 5, 15, and

30 μm, were tested in both dry and hydrated (swollen) states. Atomic force microscopy was

used for measuring nanoindentation-based force–displacement curves that were later used

for calculating the Young's moduli using the Hertz's contact theory. For every microsphere

size and condition, the measurements were repeated 400–500 times at different surface

locations and the histograms of the Young's modulus were plotted. The mean Young's

modulus of 5, 15, and 30 μm microspheres were respectively 56.1771.1 (mean7SD),

94.67103.4, and 57.6758.6 MPa under dry conditions and 226.4754.2, 334.57128.7, and

342.57136.8 kPa in the swollen state. The histograms were not represented well by the

average Young's modulus and showed three distinct peaks in the dry state and one distinct

peak in the swollen state. The peaks under dry conditions are associated with the different

parts of the co-polymeric material at the nano-scale. The measuredmechanical properties of

swollen microspheres are within the range of the nano-scale properties of cartilage, which

could favor integration of the microspheres with the host tissue.

& 2014 Elsevier Ltd. All rights reserved.

rved.

fax: þ31 15 2784717.z_amirabbas@yahoo.com (A.A. Zadpoor).

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 3 1 3 – 3 1 9314

1. Introduction

Osteoarthritis is the most common joint disorder (Arden andNevitt, 2006) and the most frequently reported cause of long-term disability (Badley, 1995). Since disease-modifying drugs arenot currently available for osteoarthritis, medicinal treatmentoften includes non-steroidal anti-inflammatory (NSAID) drugs(Lin et al., 2004) that are often used for pain management. Thedelivery of such drugs is a challenging task due to the chance ofadverse reactions via conventional drug injections on the onehand (Lin et al., 2004; Mamdani et al., 2002) and fast drugclearance that causes the drug concentration to rapidly dropbelow the therapeutically effective levels on the other hand(Ayral, 2001). Recently proposed approaches for the delivery ofNSAID include the use of liposomes, nanoparticles, and micro-spheres for sustained and adjustable drug release (Gerwin et al.,2006; Sandker et al., 2013). Among those, microspheres have theadvantages of being easily adjustable in their release kinetic,allowing well-controlled sustained drug release varying fromdays to months (Sandker et al., 2013).

It is known that the mechanical properties of the micro-spheres used for drug delivery applications influence theirperformance (Chan et al., 2008; Mercadé-Prieto and Zhang,2012). The effects of mechanical properties are even moreimportant for skeletal diseases where strong mechanical forcesare transferred through the tissues and the microspheresinteract with the surrounding tissues both chemically andmechanically. The release kinetics of the microspheres maychange due to those mechanical interactions and the modula-tions that mechanics may have with diffusion kinetics andbiodegradation behavior of the microspheres.

In this paper, we study the mechanical properties of mono-disperse microspheres composed of hydrophilic phase-separatedmulti-block copolymers that are developed for delivery of drugsused in the management of osteoarthritis. We use atomic forcemicroscopy (AFM) for nanoindentation tests and the Hertz'scontact theory to calculate the nano-mechanical properties ofthe microspheres. The mechanical properties of microsphereswith different sizes are measured both in the dry and swollenstates. In the vast majority of studies that report the mechanicalproperties of microspheres, a limited number of measurementsare carried out and the obtained mechanical properties areaveraged to calculate the mean values of the mechanical proper-ties. In this study, we repeated the measurements severalhundred times for every case to be able to measure the proper-ties more accurately and to reveal nano-scale propertiesof the microspheres that cannot be identified using a fewmeasurements.

Fig. 1 – Molecular structure of SynBiosys 20LP10L20-LL40multi-block copolymer.

2. Materials and methods

Monodisperse microspheres of different sizes, namely 5,15 and 30 μm, were made of a biodegradable hydrophilicmulti-block copolymer (SynBiosyss20LP10L20-LLA40, InnoCorePharmaceuticals, Groningen, The Netherlands) using the Micro-sieve™ membrane emulsification equipment (Nanomi, Olden-zaal, The Netherlands) and were tested in both dry and swollenstates. The hydrophilic 20LP10L20-LLA40 multi-block copolymer

is composed of hydrophilic poly (DL-Lactide)-PEG-poly (DL-Lac-tide) and hydrophobic poly (L-Lactide) segments that are ran-domly chain-extended with 1,4-butanediisocyanate. The poly(DL-Lactide)-PEG-poly (DL-Lactide) segments form amorphousdomains, whereas the poly (L-Lactide) segments form crystal-line domains, thereby forming a phase-separated multi-blockcopolymer in which the crystalline domains act as physicalcrosslinks (Fig. 1). Differently sized 20LP10L20-LLA40 micro-spheres were tested in both dry and swollen states to determinetheir mechanical properties. Before starting the test with AFM,the microspheres were attached to a firm substrate. A thin layerof a water-resistant epoxy-based glue (Bison, Netherlands) wascarefully applied to fix the microspheres on the microscopeslide. For measurement under hydrated conditions, prior totesting, the microspheres were submerged in deionized waterfor three days to allow them reach their equilibrium swellingdegree. The tests were carried out at room temperature.

An AFM with a Nanoscope controller (Bruker, Dimension V,Japan) and a standard fluid cell (Bruker) was used for perform-ing nanoindentation tests on the microspheres (Fig. 2).A symmetric triangular AFM probe (Bruker, Camarillo, USA)with a nominal diameter of 2 nm and a nominal cantileverspring constant of 0.35 N/m was used for the measurements tofind the more symmetric characteristics of the microspheres(Zhu et al., 2011). The actual cantilever spring constant wasdetermined using the thermal fluctuations technique (Notbohmet al., 2011). The standard calibration process for findingthe sensitivity factor for converting voltage to deflection wasfollowed before the start of the measurements. Using around 30individual microspheres, 400–500 indentation curves wereobtained at a frequency of 1 Hz and an indentation depth ofaround 500 nm. The Nanoscope analysis software (Bruker,version 1.4) was used for analyzing the obtained force–displace-ment curves and calculating the Young's modulus accordingto the Sneddon theory (Zhu et al., 2011). According to theSneddon theory, force and displacement are related to eachother through the following relationship:

F¼ π tan φ

2γ2E

ð1�υ2Þ h2 ð1Þ

where F is force, h is displacement, φ is the half angle of thecone, γ¼π/2, υ is the Poisson's ratio and E is the Young'smodulus of the microsphere (Franke et al., 2007; Oyen andCook, 2009; Zhu et al., 2011). A Poisson's ratio of 0.5 wasassumed which is close to the value used in some other studieson cartilage nanomechanics (Loparic et al., 2010; Stolz et al.,2004). With such a value of the Poisson's ratio, the materialbehaves incompressibly.

The Young's moduli calculated for the swollen micro-spheres with different sizes were statistically assessed using

Fig. 2 – SEM pictures of 5 (a), 15 (b), and 30 μm (c)microspheres.

(2)

400 600 800Displacement (nm)

Displacement (nm)

Displacement (nm)

Displacement (nm)

Displacement (nm)

Displacement (nm)

Displacement (nm)

Displacement (nm)

Displacement (nm)

(1)

0 200 400

(2)

0 200 400-200

-150

-100

-50

0 (3)

-100

0-200

(3)

-180-160-140-120

00-80

Load

(nN

)

Load

(nN

)

Load

(nN

)

Load

(nN

)

Load

(nN

)

Load

(nN

)

Load

(nN

)

Load

(nN

)

Load

(nN

)

-200

-150

-100

-50

0 50 100 150 200 250 300 350 400

20

40

60

80

100

120

Modulus (MPa)

Freq

uenc

y(1)

(2)(3)

200 400 600

-50

0 (1)

200 400

-100

0 (2)

0 200 400-200-150-100-50

050 (3 )

0 50 100 150 200 250 300 350 4000

20

40

60

80

100

120

(1)

(2)

(3)

0 200 400-50

0

50(1)

0 200 400

-50

0

50(2)

400 600 800-100-50

050

100150 (3)

Freq

uenc

y

20

40

60

80

100

120

Freq

uenc

y

Modulus (MPa)

00 50 100 150 200 250 300 350 400

Modulus (MPa)

0

(1)

Fig. 3 – Histograms of the modulus of dry microspheres withtypical force–displacement curve obtained from individualmicrospheres: (a) 5, (b) 15, and (c) 30 μm microspheres.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 3 1 3 – 3 1 9 315

the one-way ANOVA test followed by the Tukey–Kramer post-hoc analysis. Statistical significance threshold was set atpo0.05. For the dry micro-spheres, the data was first rank-transformed using the non-parametric Kruskal–Wallis test.The post-hoc analysis could then be performed using theTukey–Kramer test and the rank-transformed data.

3. Results

In the dry state, the histogram of the Young's modulus showsseveral distinct peaks (Fig. 3). The peaks are more or lesssimilar between the different microsphere sizes (Table 1). The

first and by far the largest peak is between 1 and 40 MPa for5 μm microspheres, between 10 and 20 MPa for 15 μm micro-spheres and between 1 and 20 MPa for 30 μm microspheres(Fig. 3). A number of smaller peaks can be observed for allthree microsphere sizes (Table 1), although the second andthird peaks are more visible for 30 μm microspheres ascompared to both other sizes. For 30 μm microspheres, thesecond peak occurs between 30 and 60 MPa while the thirdpeak can be found around 120–130 MPa (Fig. 3c). Some lessclear peaks can be observed for both the other microspheresizes around the same values of the Young's modulus(Fig. 3a–b). When averaged over all measurements, Young's

Table 1 – The main characteristics of the Young's modulus histograms found for microspheres with different sizes in dryand swollen states.

Size (μm) Young's modulus Primary and secondary peaks

Dry (MPa) Swollen (kPa) Dry (MPa) Swollen (kPa)

Mean SD Median Mean SD Median

5 56 71 25 226b 54 216 1–40a 197a

(p¼1.29� 10�68) 40–50120–130

15 95b 103 39 334 129 304 10–20a 295a

(p¼7.90�10�8) 20–50 625190–200

30 58 59 38 342 137 324 1–20a 10030–60a 315a

120–130a

a Peak with the highest frequency.b Microsphere size with significant difference in the Young's modulus from both the other sizes.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 3 1 3 – 3 1 9316

modulus of dry microspheres of 5, 15, and 30 μm wererespectively 56.17 71.1 (mean7SD), 94.67103.4, and 57.6758.6 MPa. In dry state, the mean Young's moduli of themicrospheres with different sizes were found to be signifi-cantly different (Table 1). Post-hoc analysis showed that theYoung's modulus of 15 μm dry microspheres is significantlyhigher than those of both the other sizes (also in dry state).

The range of the Young's moduli measured by AFM ofmicrospheres in the swollen state was several orders of magni-tude lower than the Young's moduli of the microspheresmeasured in the dry state (Fig. 4). As opposed to the dry statewhere several peaks were observed in the histogram of theYoung's modulus, there was only one clear peak in thehistogram of the Young's modulus of the swollen microspheres(Fig. 4). This enabled us to fit a Gaussian distribution to themeasured data points (Fig. 4a–c). When averaged over allmeasurements, the Young's moduli of swollen microspheresof 5, 15, and 30 μm were respectively 226.4754.2 (mean7SD),334.57128.7, and 342.57136.8 kPa. Statistical analysis of theYoung's moduli of the swollen microspheres with differentsizes showed that there is a significant difference between theirYoung's moduli (Table 1). Post-hoc analysis showed that, in theswollen state, the Young's moduli of 15 and 30 μmmicrospheresare higher than those of the 5 μm microspheres. In the swollenstate, there was no significant difference between the Young'smoduli of 15 and 30 μm microspheres. Moreover, the Young'smoduli of swollen microspheres generally tended to increase asthe size of the microsphere increased.

4. Discussion

The hydrophilic 20LP10L20-LLA40 multi-block copolymer stu-died here contains two major components: an amorphouspoly (DL-Lactide)-PEG-poly(DL-Lactide) (PDLA–PEG–PDLA) and acrystalline poly (L-Lactide) (PLLA) segment. The PDLLA–PEG–PDLLA part has water absorption functionality due to thepresence of polyethylene glycol (PEG). PEG is known for its

good biocompatibility and hydrophilicity (Hutson et al., 2011;Ostroha, 2006; Zustiak and Leach, 2010), whereas PDLA iswidely used in various sustained release drug delivery pro-ducts due to its good biocompatibility and excellent biode-gradability. The crystalline PLLA segments act as physicalcrosslinks and also give the microsphere appropriate stiffnessthat is needed for drug delivery applications in a cartilaginousenvironment. PLLA is widely used in biomedical implantsespecially where load-bearing properties are required. Thespecific molecular architecture and phase-separated mor-phology of the used multi-block copolymer gives the micro-spheres the desired hydrophilicity in combination withmechanical characteristics that are fairly similar to those ofspecially soft biological tissues such as cartilage, making itsuitable for local delivery and controlled release applications(Brown et al., 2005; Garlotta, 2001; Ostroha, 2006; Wang et al.,2012).

4.1. Dry state

The first peak seen in the histogram of the Young's moduli ofmicrospheres in the dry state can be attributed to the Young'smodulus of PEG as the values are within the range ofpreviously reported Young's moduli for PEG (Drira andYadavalli, 2013; Gäbler, 2009). As compared to crystallinePLLA, the PEG part is the softer part of the multi-blockcopolymer that constitutes the microspheres. The meanvalues of the second apparent peak in the histogram of theYoung's modulus of 5, 15, and 30 μm microspheres arearound 50 MPa. This second peak may be ascribed to thenon-crystalline PDLA component of the PDLA–PEG–PDLAsegment (Ostroha, 2006; Yang, 2012). It should be noted thatthe PDLA chains present in the PDLA–PEG–PDLA block studiedhere are relatively short. It is therefore not clear to whatextent their properties are close to those of high molecularweight PDLA often studied elsewhere. The average values ofthe other recognizable peak, especially in 15 and 30 μmmicrospheres, exceed 100 MPa. This last peak might beindicative of the crystalline portion of the copolymer.

0 100 200 300 400 500 600 700 8000

50

100

150

Modulus (kPa)

Modulus (kPa)

Modulus (kPa)

Freq

uenc

yFr

eque

ncy

Freq

uenc

y

mean = 226.4SD = 54.1

0 100 200 300 400 500 600 700 8000

50

100

150

mean = 334.5SD = 128.9

0 100 200 300 400 500 600 700 8000

50

100

150

mean = 342.5SD = 137.1

Fig. 4 – Histograms and corresponding normal distributioncurves of swollen microspheres: (a) 5, (b) 15, and (c) 30 μmmicrospheres.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 3 1 3 – 3 1 9 317

It should be noted that the above-mentioned values of theYoung's modulus and their relationship with the differentparts of the copolymer are merely indicative. The actualvalues of the Young's modulus depend on many parametersincluding the molecular weight of the macromolecules, thelength of the chains within polymeric network, the cross-linking density of the matrix, the ratio of the length of thePDLA portion to that of the PEG length in the PDLA–PEG–PDLAsegment, and the optimized proportion between the PDLA–PEG–PDLA segment used for hydrophilicity and the PLLAblock used for mechanical stiffness (Brown et al., 2005;Ostroha, 2006; Wang et al., 2012; Zustiak and Leach, 2010).

4.2. Swollen state

The final functionality of the microsphere is in the swollenstate, as they will be hydrated and perform their drug deliveryrole in the target cartilage tissue based on a combination ofswelling, diffusivity (Drira and Yadavalli, 2013; Ebenstein andPruitt, 2004), and biodegradation. It is clear that there is a sharpdecrease in the Young's modulus of the swollen microspheresas compared to the values found under dry conditions. Thewater molecules, with their polar groups, tend to interact withthe hydrophilic domains of the multi-block copolymer and actas a plasticizer reducing the stiffness of the co-polymeric matrix(Yang, 2012). The water uptake ratio can therefore be consid-ered as a major factor for optimizing the mechanical propertiesof microspheres based on the principles of the swelling process.Moreover, the molecular weight of the crystalline PLLA segmenthave been narrowly tailored with respect to the molecularweight of the hydrophilic PDLA–PEG–PDLA segment (Ostroha,2006; Wang et al., 2012).

In the swollen state, the Young's moduli of 15 and 30 μmmicrospheres are significantly higher than those of 5 μm micro-spheres. This is due to the fact that the water adsorption abilityof microspheres is size-dependent and is inversely related tothe surface area, meaning that smaller microspheres showhigher water adsorption tendency and lower stiffness values(Eichhorn and Sampson, 2010; Suttiponparnit et al., 2010).

4.3. The required number of measurements

This study is one of the rare studies in which more than ahandful data points are used for characterizing the nano-mechanical properties of soft co-polymeric biomaterials parti-cularly microspheres. In most other studies, the average of afew measurements is reported as the representative Young'smodulus of the material at the nano-scale. The results of thisstudy (Figs. 3 and 4) show that the mean of a few measure-ments may not necessarily be a good representative value.As for microspheres in the swollen state, the mean Young'smodulus is a relatively good representative of the mechanicalproperties of the microspheres (Fig. 4) as most values seen inthe histogram are distributed around the mean Young's mod-ulus. In comparison, in the dry state, the mean Young's moduliof the microspheres vary between 55 and 95 MPa (Fig. 3). Thesemean values are several times larger than the first and by farthe largest peak of the histograms of the young's modulus(Fig. 3a–c). Using the mean value of the Young's modulus as therepresentative Young's modulus of the microspheres will there-fore grossly misrepresent the actual mechanical properties ofmicrospheres at the nano-scale. Moreover, the histograms ofthe Young's modulus are very information-rich and couldreveal additional information regarding the range and distribu-tion of the mechanical properties of the microspheres at thenano-scale. This kind of information is totally lost when theaverage of a few data points is reported as the mechanicalproperties of the polymeric material.

4.4. Relevance for cartilage and drug delivery applications

Stolz et al. (2009) measured the age-related dynamic elasticmodulus of the human osteoarthritis articular cartilage tissue

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 3 1 3 – 3 1 9318

and fount it to be between 15.3 and 142 kPa. Han et al. (2011)analyzed the modulus of the surface layer of the bovinearticular tissue and found it to be between 120 and 200 kPadepending on the loading rate (0.1, 1, 10 mm/s). The Young'smodulus of the microspheres in the swollen state underhydrated conditions are therefore well in the range of themechanical properties of cartilage at the nano-scale, meaningthat the microspheres should be able to well integrate withinthe host tissue and perform their functions.

The mechanical properties of drug delivery devices suchas micro-spheres play important roles in regulating theircontrolled release performance both directly and indirectly.For example, it is known that mechanical interactionsbetween the drug delivery devices and the surroundingtissues could influence the bio-degradation, surface area,and hydration and/or hydrolysis behavior of the involvedbiomaterials (Kranz et al., 2000; Makadia and Siegel, 2011).Moreover, the mechanical properties of drug delivery devicescould be used as an indirect measure of their drug deliveryperformance. For example, it is known (Galeska et al., 2005)that the degree of cross-linking in polyvinyl alcohol (PVA)hydrogels is intimately linked to both mechanical propertiesand drug delivery characteristics of the polymer (Hassan andPeppas, 2000). One could therefore use mechanical propertiesof the polymer as an indirect measure of drug deliverycharacteristics. Probing the mechanical properties of themicrosphere material is best done with AFM (Chen, 2013)and at the nano-scale, because accurate estimation of themechanical properties using micro-scale indenters is extre-mely challenging particularly when micro-spheres are assmall as 5 μm. That is partly because the surface of themicro-sphere cannot be considered to be flat anymore andthe entire micro-sphere may start to deform considerably asa result of micro-scale indentations. These kind of micro-scale experiments are much more difficult to interpret andmay necessitate the use of advanced computational techni-ques. Moreover, the mechanical properties at the micro-scaleare related to the mechanical properties at the nano-scaleand can theoretically be obtained from the nano-scalemechanical properties using homogenization techniques.

4.5. Limitations

This study has certain limitations that need to be consideredwhen interpreting the presented data. First, the nano-indentation technique used for characterizing the mechan-ical properties of the microspheres is associated with certaincomplexities and sources of uncertainty. For example, the tipand cantilever geometry, the type of material from whichthey are made, the depth and rate of indentation, adhesionforces, intermolecular forces, and electrostatic interactionscould all play important roles in the accuracy of AFMmeasurements (Calabri et al., 2008; Chen et al., 2011;Darling et al., 2010; Salerno et al., 2010; Thalhammer et al.,2001). Second, the analytical solution used for interpretationof the force–displacement data has been derived based onspecific assumptions that are never perfectly satisfied inpractice. Deviations from those assumptions could causeinaccuracy in the measured mechanical properties. Finally,the mechanical properties of microspheres might be different

in vivo, for example, due to the presence of different typesof ions and enzymatic factors. Testing the microspheres inwater is therefore only an approximation of the actual in vivoconditions.

5. Conclusions

In summary, the nano-mechanical properties of co-polymericmicrospheres with different sizes were measured in both thedry and swollen states using atomic force microscopy. It wasshown that the mechanical properties of the microspheresare in general much lower in the swollen state as comparedto the dry state. In addition, the mechanical properties werefound to be different for different microsphere sizes. In theswollen state, the Young's modulus tended to increase as thesize of the microsphere increased. The mean Young's moduliof microspheres of 5, 15, and 30 μm were respectively 56.1771.1 (mean7SD), 94.67103.4, and 57.6758.6 MPa in the drystate and 226.4754.2, 334.57128.7, and 342.57136.8 kPa inthe swollen state. However, the mean values of the Young'smodulus were not good representatives of the mechanicalproperties of the microspheres at the nano-scale, specificallyin the dry state. A large number of measurements are there-fore needed to establish the histogram of the Young's moduliin order to study the range and distribution of the Young'smodulus of the microspheres at the nano-scale. Consideringthe range of cartilage tissue mechanical characteristics, theswollen microspheres demonstrate desired nano-scale elasticmodulus data, due to which the microspheres are expectedto exhibit the desired biocompatibility upon intra-articularadministration besides well integration with their cartilagi-nous host tissue.

r e f e r e n c e s

Arden, N., Nevitt, M.C., 2006. Osteoarthritis: epidemiology. BestPract. Res. Clin. Rheumatol. 20, 3–25.

Ayral, X., 2001. Injections in the treatment of osteoarthritis. BestPract. Res. Clin. Rheumatol. 15, 609–626.

Badley, E., 1995. The effect of osteoarthritis on disability andhealth care use in Canada. J. Rheumatol.Suppl. 43, 19–22.

Brown, C.D., Stayton, P.S., Hoffman, A.S., 2005. Semi-interpenetrating network of poly(ethylene glycol) and poly(D,L-lactide) for the controlled delivery of protein drugs. J.Biomater. Sci. Polym. Ed. 16, 189–201.

Calabri, L., Pugno, N., Menozzi, C., Valeri, S., 2008. AFMnanoindentation: tip shape and tip radius of curvatureeffect on the hardness measurement. J. Phys. Condens.Mater. 20, 1–7.

Chan, B., Li, C., Au-Yeung, K., Sze, K., Ngan, A., 2008. A microplatecompression method for elastic modulus measurement ofsoft and viscoelastic collagen microspheres. Ann. Biomed.Eng. 36, 1254–1267.

Chen, J., 2013. Understanding the nanoindentation mechanismsof a microsphere for biomedical applications. J. Phys. D: Appl.Phys. 46, 495303.

Chen, Y., Feng, B., Zhu, Y., Weng, J., Wang, J., Lu, X., 2011.Preparation and characterization of a novel porous titaniumscaffold with 3D hierarchical porous structures. J. Mater. Sci.:Mater. Med. 22, 839–844.

j o u r n a l o f t h e m e c h a n i c a l b e h a v i o r o f b i o m e d i c a l m a t e r i a l s 3 4 ( 2 0 1 4 ) 3 1 3 – 3 1 9 319

Darling, E.M., Wilusz, R.E., Bolognesi, M.P., Zauscher, S., Guilak, F.,2010. Spatial mapping of the biomechanical properties of thepericellular matrix of articular cartilage measured in situ viaatomic force microscopy. Biophys. J. 98, 2848–2856.

Drira, Z., Yadavalli, V.K., 2013. Nanomechanical measurements ofpolyethylene glycol hydrogels using atomic force microscopy.J. Mech. Behav. Biomed. Mater. 18, 20–28.

Ebenstein, D.M., Pruitt, L.A., 2004. Nanoindentation of softhydrated materials for application to vascular tissues. J.Biomed. Mater. Res. Part A 69, 222–232.

Eichhorn, S.J., Sampson, W.W., 2010. Relationships betweenspecific surface area and pore size in electrospun polymerfibre networks. J. R. Soc. Interface 7, 641–649.

Franke, O., Durst, K., Maier, V., Goken, M., Birkholz, T., Schneider,H., Hennig, F., Gelse, K., 2007. Mechanical properties of hyalineand repair cartilage studied by nanoindentation. ActaBiomater. 3, 873–881.

Gabler, S., 2009. Determination of the viscoelastic properties ofhydrogels based on polyethylene glycol diacrylate (PEG-DA)and human articular cartilage. Int. J. Mater. Eng. Innov. 1, 3–20.

Galeska, I., Kim, T.-K., Patil, S.D., Bhardwaj, U., Chatttopadhyay,D., Papadimitrakopoulos, F., Burgess, D.J., 2005. Controlledrelease of dexamethasone from PLGA microspheresembedded within polyacid-containing PVA hydrogels. AAPS J.7, E231–E240.

Garlotta, D., 2001. A literature review of poly(lactic acid). J. Polym.Environ. 9, 63–84.

Gerwin, N., Hops, C., Lucke, A., 2006. Intraarticular drug deliveryin osteoarthritis. Adv. Drug Deliv. Rev. 58, 226–242.

Han, L., Frank, E.H., Greene, J.J., Lee, H.Y., Hung, H.H., Grodzinsky,A.J., Ortiz, C., 2011. Time-dependent nanomechanics ofcartilage. Biophys. J. 100, 1846–1854.

Hassan, C.M., Peppas, N.A., 2000. Structure and applications ofpoly (vinyl alcohol) hydrogels produced by conventionalcrosslinking or by freezing/thawing methods. Biopolym. PVAHydrogels Anion. Polym. Nanocompos. Adv. Polym. Sci. 153,37–65.

Hutson, C.B., Nichol, J.W., Aubin, H., Bae, H., Yamanlar, S., Al-Haque, S., Koshy, S.T., Khademhosseini, A., 2011. Synthesisand characterization of tunable poly(ethylene glycol): gelatinmethacrylate composite hydrogels. Tissue Eng. Part A 17,1713–1723.

Kranz, H., Ubrich, N., Maincent, P., Bodmeier, R., 2000.Physicomechanical properties of biodegradable poly (D, L-lactide) and poly (D, L-lactide-co-glycolide) films in the dry andwet states. J.Pharm. Sci. 89, 1558–1566.

Lin, J., Zhang, W., Jones, A., Doherty, M., 2004. Efficacy of topicalnon-steroidal anti-inflammatory drugs in the treatment ofosteoarthritis: meta-analysis of randomised controlled trials.Br. Med. J. 329, 324–326.

Loparic, M., Wirz, D., Daniels, A.U., Raiteri, R., Vanlandingham, M.R., Guex, G., Martin, I., Aebi, U., Stolz, M., 2010. Micro- andnanomechanical analysis of articular cartilage by indentation-type atomic force microscopy: validation with a gel-microfibercomposite. Biophys. J. 98, 2731–2740.

Makadia, H.K., Siegel, S.J., 2011. Poly lactic-co-glycolic acid (PLGA)as biodegradable controlled drug delivery carrier. Polymers 3,1377–1397.

Mamdani, M., Rochon, P.A., Juurlink, D.N., Kopp, A., Anderson, G.M., Naglie, G., Austin, P.C., Laupacis, A., 2002. Observationalstudy of upper gastrointestinal haemorrhage in elderlypatients given selective cyclo-oxygenase-2 inhibitors orconventional non-steroidal anti-inflammatory drugs. Br. Med.J. 325, 624–629.

Mercade-Prieto, R., Zhang, Z., 2012. Mechanical characterizationof microspheres-capsules, cells and beads: a review. J.Microencapsul. 29, 277–285.

Notbohm, J., Poon, B., Ravichandran, G., 2011. Analysis ofnanoindentation of soft materials with an atomic forcemicroscope. J. Mater. Res. 27, 229–237.

Ostroha, J.L., 2006. PEG-based degradable networks for drugdelivery applications. Drexel. 165, (Ph.D. Thesis). DrexelUniversity, Philadelphia.

Oyen, M.L., Cook, R.F., 2009. A practical guide for analysis ofnanoindentation data. J. Mech. Behav. Biomed. Mater. 2,396–407.

Salerno, M., Dante, S., Patra, N., Diaspro, A., 2010. AFMmeasurement of the stiffness of layers of agarose gelpatterned with polylysine. Microsc. Res. Tech. 73, 982–990.

Sandker, M.J., Petit, A., Redout, E.M., Siebelt, M., Muller, B., Bruin,P., Meyboom, R., Vermonden, T., Hennink, W.E., Weinans, H.,2013. In situ forming acyl-capped PCLA–PEG–PCLA triblockcopolymer based hydrogels. Biomaterials 34, 8002–8011.

Stolz, M., Gottardi, R., Raiteri, R., Miot, S., Martin, I., Imer, R.,Staufer, U., Raducanu, A., Duggelin, M., Baschong, W., Daniels,A.U., Friederich, N.F., Aszodi, A., Aebi, U., 2009. Early detectionof aging cartilage and osteoarthritis in mice and patientsamples using atomic force microscopy. Nat. Nanotechnol. 4,186–192.

Stolz, M., Raiteri, R., Daniels, A.U., VanLandingham, M.R.,Baschong, W., Aebi, U., 2004. Dynamic elastic modulus ofporcine articular cartilage determined at two different levelsof tissue organization by indentation-type atomic forcemicroscopy. Biophys. J. 86, 3269–3283.

Suttiponparnit, K., Jiang, J., Sahu, M., Suvachittanont, S.,Charinpanitkul, T., Biswas, P., 2010. Role of surface area,primary particle size, and crystal phase on titanium dioxidenanoparticle dispersion properties. Nanoscale Res. Lett., 6–27.

Thalhammer, S., Heckl, W.M., Zink, A., Nerlich, A.G., 2001. Atomicforce microscopy for high resolution imaging of collagenfibrils—a new technique to investigate collagen structure inhistoric bone tissues. J. Archaeol. Sci. 28, 1061–1068.

Wang, D.K., Varanasi, S., Hill, D.J.T., Rasoul, F., Symons, A.L.,Whittaker, A.K., 2012. The influence of composition on thephysical properties of PLA–PEG–PLA-co-Boltorn basedpolyester hydrogels and their biological performance. J. Mater.Chem. 22, 6994–7004.

Yang, T., 2012. Mechanical and swelling properties of hydrogels.KTH67.

Zhu, Y., Dong, Z., Wejinya, U.C., Jin, S., Ye, K., 2011. Determinationof mechanical properties of soft tissue scaffolds by atomicforce microscopy nanoindentation. J. Biomech. 44, 2356–2361.

Zustiak, S.P., Leach, J.B., 2010. Hydrolytically degradable poly(ethylene glycol) hydrogel scaffolds with tunable degradationand mechanical properties. Biomacromolecules 11, 1348–1357.