This article was downloaded by: [Nipissing University]On: 10 October 2014, At: 18:00Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Biomaterials Science,Polymer EditionPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/tbsp20

Structure and assembly of PEO-PPO-PEO co-polymers in mammalian cell-culture mediaPraveen K. Sharma , Julie E. Matthew & Surita R. BhatiaPublished online: 02 Apr 2012.

To cite this article: Praveen K. Sharma , Julie E. Matthew & Surita R. Bhatia (2005) Structure andassembly of PEO-PPO-PEO co-polymers in mammalian cell-culture media, Journal of BiomaterialsScience, Polymer Edition, 16:9, 1139-1151, DOI: 10.1163/1568562054798545

To link to this article: http://dx.doi.org/10.1163/1568562054798545

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms& Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

J. Biomater. Sci. Polymer Edn, Vol. 16, No. 9, pp. 1139–1151 (2005) VSP 2005.Also available online - www.vsppub.com

Structure and assembly of PEO-PPO-PEO co-polymersin mammalian cell-culture media

PRAVEEN K. SHARMA, JULIE E. MATTHEW and SURITA R. BHATIA ∗

Department of Chemical Engineering, 159 Goessmann Laboratory, 686 North Pleasant Street,University of Massachusetts, Amherst, MA 01003, USA

Received 28 June 2004; accepted 19 January 2005

Abstract—We investigate the structure of a poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) co-polymer, Pluronic® F127, in mammalian cell-culture media. It is well knownthat aqueous solutions of F127 gel at physiological temperatures with this transition, which corre-sponds to the formation of a close-packed cubic assembly of spherical micelles. Previous work hasshown that in both mammalian cell minimum essential medium (MEM) and MEM with added fetalbovine serum, the gel phase boundary shifts to lower temperatures and concentrations as comparedto pure water. Using DLS, we have found that at 25◦C the critical micelle concentration (CMC) de-creases in the presence of MEM. Our SANS studies at 25◦C indicate that F127 in MEM-D2O alsoforms a close-packed cubic micellar gel, suggesting that the mechanism of gelation is the same inboth pure water and MEM. Fits to the neutron spectra on 2 wt% F127 in D2O and MEM-D2O show alarge difference in the micelle aggregation number (Nagg) and a small difference in the micelle size,with both Nagg and the micelle size larger in the presence of MEM. In addition, our SANS spectrain MEM-D2O indicate repulsive interactions between micelles at 2 wt% polymer, whereas no cor-relation peak is observed for this concentration in water. Finally, moderately concentrated samples(5–18 wt% polymer) in MEM exhibit slightly stronger ordering and sharper peaks, perhaps indicatingstronger intermicellar interactions in the presence of MEM. This stronger repulsive interaction maybe the cause of the shift in the liquid–gel phase boundary that is observed.

Key words: Pluronic F127; micelles; cell culture; triblock co-polymers; cell encapsulation; poloxam-ers.

INTRODUCTION

Triblock co-polymers consisting of poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) have been widely studied for various pharmaceutical andbiomedical applications [1–3]. Due to the difference in hydrophobicity between

∗To whom correspondence should be addressed. Tel.: (1-413) 545-0096. Fax: (1-413) 545-1647.E-mail: sbhatia@ecs.umass.edu

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

1140 P. K. Sharma et al.

poly(ethylene oxide) (PEO) and poly(propylene oxide) (PPO) blocks, these co-polymers are able to self-assemble into micellar structures in water. Pluronic®

F127 (EO100PO70EO100) is one member of this family of co-polymers that exhibitslow toxicity, and has been approved by the FDA for use in the human body.Interest for drug-delivery purposes stems from the fact that moderately concentratedsolutions have been found to form gels when warmed to physiological temperatures[4–6]. This type of gelation behavior also provides a potentially gentle method forencapsulating cells for tissue-engineering applications. For example, F127 has beenused to encapsulate both sheep tracheal epithelial cells [7] and HEpG2 human livercells [8, 9], and to promote the attachment of human gingival fibroblasts [10]. F127has also been used to provide a scaffold for porcine chondrocytes [11, 12], wheregrowth in the co-polymer scaffold was shown to be more organized [11], with a lowinflammatory response [12].

The structure of F127 gels in water is widely accepted to be a close-packedarrangement of spherical micelles on a cubic lattice [6]. The gels are highlyordered, with SANS spectra often exhibiting 2–3 higher-order peaks. Enthalpiesof gel formation are orders of magnitude lower than the enthalpy of micellization,suggesting that gelation is a nearly athermal process [13]. This is consistent witha gelation mechanism that is primarily entropic in nature, arising from repulsiveinteractions between micelles (i.e., formation of a close-packed gel).

Although there have been several studies regarding the various uses of F127 gelsas biomaterials, relatively few studies exist on the phase behavior and structure ofthese gels in the presence of components that are commonly added for biologicalapplications, such as cell media, proteins and nutrients. An understanding of theeffects that such components may cause is crucial for formulating block co-polymersolutions for use in the body or in tissue-engineering applications. For example,changes in F127 self-assembly might impact gel formation, possibly resulting inlower cell viability and an increased inflammatory response. Changes in the criticalmicelle concentration or micelle size could also impact transport of nutrients orproteins.

Recently, we examined the lower liquid-gel phase boundary of F127 in thepresence of minimum essential medium (MEM) and MEM with added fetal bovineserum (MEM-FBS) [14]. In both MEM and MEM-FBS, we found that the gel-phase boundary shifted to lower temperatures and concentrations as compared topure water. The enthalpy of gel formation was found to be similar in water, MEMand MEM-FBS, suggesting that the mechanism of gelation was similar in all threesolvents [14]. However, dynamic light scattering on samples of 10 wt% F127 inMEM revealed large aggregates with hydrodynamic radii of 652 nm. Samplesin pure water showed no evidence of aggregation, yielding hydrodynamic radiicharacteristic of individual micelles (10–11 nm). The formation of aggregatessuggested that components in the cell media could influence the intermicellarinteractions, which could lead to a shift in the liquid–gel phase boundary and acorresponding change in the gel structure. Thus, in order to examine the effect of

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

PEO-PPO-PEO co-polymers in cell-culture media 1141

MEM on the micellar structure and elucidate the mechanism of gel formation, weperformed SANS on F127 gels and dilute solutions in MEM at 25◦C. Additionally,changes in the gelation behavior have been linked to changes in the critical micelleconcentration [15]. We, therefore, utilized dynamic light scattering (DLS) on dilutesolutions at 25◦C to determine if MEM influenced the self-assembly of F127, i.e.,the critical micelle concentration (CMC).

MATERIALS AND METHODS

Experimental

Minimum essential medium (MEM) was obtained as a solid powder from Gibco-BRL (Life Technologies) and was dissolved in D2O (99.9% D, Cambridge IsotopeLaboratories) and supplemented with penicillin, dihydrostreptomycin, HEPESbuffer and sodium bicarbonate (Sigma-Aldrich). Cell culture tested grade Pluronic®

F127 was obtained from Sigma-Aldrich and used to make solutions on a weight/weight basis of between 2 wt% and 25 wt% co-polymer in either D2O or MEM-D2O. Addition of the co-polymer was done using the cold method described bySchmolka [16], whereby prescribed amounts of the co-polymer were added to theappropriate solvent at ambient temperature, stirred and refrigerated until dissolution(24–120 h).

Small-angle neutron scattering (SANS) experiments were performed on the SmallAngle Diffractometer (SAD) at the Intense Pulsed Neutron Source at ArgonneNational Laboratory at 25◦C in quartz sample cells with a path length of 1.0 mm.Spectra were collected for 1–3 h, depending on the contrast. D2O was used toquantify the solvent scattering for all samples, which was subsequently subtractedoff. The incoherent scattering from each sample was estimated from the signal athigh q and was also subtracted from the data. Data were obtained for 0.006 Å−1 >

q > 1 Å−1.Dynamic light scattering (DLS) was used to determine the critical micelle

concentration of the co-polymer in the absence and presence of MEM. Two stocksolutions were made, each being comprised of 5 wt% co-polymer dissolved in eithernanopure water (obtained from a Barnstead NANOpure® Infinity UF filtration unit)or MEM-nanopure water (made in the same manner as the MEM-D2O solvent).From these stock solutions, nine samples in each solvent (nanopure water or MEM-nanopure water) at concentrations of between 0.01 wt% and 1 wt% co-polymer weremade by diluting a convenient amount of the stock solution with the appropriatesolvent. These samples were then filtered slowly through 0.45 µm Millex®-HVsyringe-driven filters (Millipore) into clean light-scattering tubes that had beenthoroughly washed with nanopure water, dried and cleaned with a Kensington DustBlaster™ compressed gas duster prior to use. Light scattering experiments werecarried out at a fixed angle of 90◦ using a 514.5 nm Lexel 95 8W Argon IonLaser working at a constant power output of 200 mW. Samples were allowed to

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

1142 P. K. Sharma et al.

equilibrate to 25◦C for at least 30 min in a jacketed decahydronaphthalene vat ofthe BI 200SM Goniometer set up (Brookhaven Instruments), prior to data collection.Time-dependent intensity fluctuation data were collected over a delay range of 25 nsto 100 ms (with 342 channels plus four extended channels) using a photomultiplierdetector tube (200 µm pinhole aperture size) and correlated with a BI 9000ATdigital correlator (both from Brookhaven Instruments).

Data analysis

The scattered intensity, I (q), of a monodisperse system can be expressed as theproduct of the form factor and structure factor:

I (q) = N(�ρb)2F(q)S(q), (1)

where N is the number density of the micelles and �ρb is the difference in thescattering length density. The form factor, F(q), describes the individual scatterersand the structure factor, S(q), characterizes interactions between scatterers. For theform factor, we utilized the core-corona model, which was first used to describe themorphology of ion containing co-polymers [17] and is commonly used for PEO-PPO-PEO co-polymer micelles [18]:

F(q)(�ρb)2 =

{(4πR3

1

3

)(ρ1 − ρ2)

[3J1(x1)

x1

]+

(4πR3

2

3

)(ρ2 − ρs)

[3J1(x2)

x2

]},

(2)

where R1 and R2 are the radii of the micelle core and the entire micelle, respectively;ρ1, ρ2 and ρs are the scattering length densities of the core, corona and solvent,respectively; J1(xi) is the first-order spherical Bessel function and xi = qRi. Thescattering length densities of the core and corona account for the presence of solventin these regions of the micelle. The two radii can be related to the amounts of solventin the core and corona through the micelle aggregation number, Nagg, reducing thenumber of fitting parameters for the core-corona model to three: R1, R2 and Nagg.

We chose to use a hard sphere potential to model the structure factor [19, 20]. Useof such a model for the structure factor is fairly standard in the analysis of data frompolymeric micelles [20, 21] and details of the mathematical form of the structurefactor are given elsewhere [19]. The fitted parameter in this model is Rint, the radiusof interaction; with the volume fraction, φ, given by:

φ = 4

3πR3

intN. (3)

In addition, we accounted for polydispersity in the micellar size, since non-uniformity of micelles tends to smear the features of observed SANS spectra.However, there was not enough resolution to distinguish between polydispersityin the core size or the micelle size. Thus, in order to keep the number of fittedparameters to a minimum, we chose to only account for polydispersity in the overallmicelle size. This was done by assuming that the micelle size followed a Gaussian

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

PEO-PPO-PEO co-polymers in cell-culture media 1143

distribution with a mean value of R2 and a half-width of σ . The scattered intensitywas then obtained by integrating this distribution over all positive values of themicelle radius.

I (q) = N

(∫ ∞

0

1

σ√

2πe

[−(r − R2)2

2σ 2

](�ρb)

2F(q) dr

)S(q). (4)

For the DLS experiments, analysis of the autocorrelation function obtained fromthe correlator in terms of particle size distribution was done numerically using aNon-Negatively Constrained Least Squares (Regularized CONTIN) method over aparticle size range of 1.00 nm to 100 nm. Most samples showed a small diameterpopulation corresponding to unimers, with samples at higher concentrations show-ing a population of greater diameter (i.e., micelles). The lowest concentration atwhich this micellar population was observed was taken to be the critical micelleconcentration (CMC). The error on the observation was estimated as half the differ-ence between this concentration and the highest concentration that did not show amicelle-sized population.

RESULTS AND DISCUSSION

Critical micelle concentration (CMC)

From dynamic light scattering experiments, the CMC of F127 in water was foundto be 0.26 ± 0.03 wt%. In the presence of MEM, the CMC was found to decreaseto 0.105 ± 0.015 wt%. These results, together with our previously measured gel–liquid boundaries, are consistent with literature data on other additives to F127,which have shown that co-solvents that affect gelation also alter the CMC. Ingeneral, these effects occur in parallel, such that a co-solvent that decreases thegelation temperature also shows a decrease in the CMC. This is expected, since alower CMC leads to a greater number density of micelles at a given co-polymerconcentration. This results in a larger potential for micellar entanglements and,hence, a lower temperature at which gelation may occur [15]. For example,Malmsten and Lindman [4] showed that the lower gelation boundary of PluronicF127 in water was decreased by about 20◦C in the presence of 1.0 M NaCl. Desaiet al. [22] reported that the CMC of Pluronic F127 in water was 2.0 g/dl, butdecreased to 0.6 g/dl in the presence of 1.0 M NaCl. The gelation temperatureof 20 wt% F127 in water was found to decrease by up to 10◦C in the presenceof either 20 wt% propylene glycol, glycerol or PEG 400 [15]. These co-solventsalso showed a marked decrease in the CMC from 0.12 wt% in water to 0.041 wt%,0.032 wt% and 0.028 wt%, in the presence of 20 wt% propylene glycol, glyceroland PEG 400, respectively [15]. It is also worth noting that there is a large variationin the value of the CMC reported in the literature. This is most likely to be dueto the differences in experimental methods used and therefore comparisons shouldonly be made between measurements carried out using the same technique. Our

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

1144 P. K. Sharma et al.

two measurements of the CMC of F127 were made using the same procedure and,hence, are comparable. It is clear that the presence of MEM promotes the formationof micelles from F127 unimers, resulting in a lower required concentration of co-polymer chains for the onset of micellization. From the findings of Desai et al.,it is likely that the presence of specific components of MEM, such as NaCl itself,may be the cause of the observed decrease in the CMC. The measured values of theCMC were used in fitting the SANS data.

The CMC of F127 in water has been found to decrease by almost two orders ofmagnitude from 25◦C to 40◦C [23]. At 37◦C, we expect the CMC of F127 in MEMto remain lower than the value in water, but to significantly decrease according tothis type of trend.

Dilute solutions

Spectra taken on samples containing 2 wt% F127 are shown in Fig. 1, along with fitsto the data using the core-corona form factor with polydisperse micelles. At this lowconcentration, interactions between micelles were initially assumed to be negligible(i.e., S(q) = 1). The data from 2 wt% F127 in D2O could be fitted satisfactorilyby the core-corona form factor with polydisperse micelles, and the parameters fromthis fit are listed in Table 1. However, the spectra from 2 wt% F127 in MEM-D2Ocan be seen in Fig. 1 to show a slight increase in the scattered intensity with thedevelopment of a broad peak at around 0.023 Å−1. This value of q corresponds

Figure 1. SANS spectra on 2 wt% F127 in D2O and MEM-D2O, fitted to the core-corona modelwith polydisperse micelles. In the case of MEM-D2O, the hard sphere repulsion interaction has beenincluded in the model. Data are shown between 0.007 Å−1 and 0.2 Å−1, due to the large uncertaintyin scattered intensity at low and high values of q.

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

PEO-PPO-PEO co-polymers in cell-culture media 1145

to a length scale of around 280 Å, which is much larger than previously reportedsizes for F127 micelles [5, 24]. A correlation peak at this length scale suggests theexistence of interactions between micelles. A fit with S(q) = 1 is unable to accountfor this feature; instead, the model over-predicts the scattered intensity at very lowvalues of q. It was, therefore, necessary to include a model for S(q) in the data fit inorder to account for such interactions and fit the data from this system satisfactorily.The fit for the MEM-D2O data shown in Fig. 1 includes a hard sphere repulsioninteraction between micelles, and the parameters from this fit are shown in Table 1.It is clear from Fig. 1 that the inclusion of a structure factor allows the overall modelto better estimate the features of the data from this system at very low values of q.

From Table 1, the micelle core radius, R1, in D2O is approx. 46 Å. In the MEM-D2O solvent, the core radius increases slightly to approx. 52 Å. In addition, theoverall micelle size can be seen to increase in the presence of MEM, from approx.71 Å to 82 Å, with a corresponding increase in polydispersity from 10.4% to11.5%. Since the gelation of F127 solutions is thought to be due to an increasein the effective micellar volume fraction [24, 25], this increase in micelle size maypartially account for the reason why solutions of F127 in MEM are able to formgels at lower temperatures and concentrations [14]. A more dramatic change is thatshown by the micelle aggregation number upon addition of MEM. In D2O, F127micelles show an aggregation number of 89. In MEM-D2O, Nagg increases to avalue of 143, indicating that many more chains of F127 are involved in the formationof a single micelle. This change also indicates that the presence of MEM causes agreater attraction between PPO segments of F127 chains, once again promotingthe formation of micelles. Although this would mean that the absolute number ofmicelles might actually decrease, the associated increase in the size of these micellesmay possibly outweigh the effect this may have on the micellar volume fraction.The increase in attraction between PPO segments may be due to the presence ofsome media components less polar than water, in the MEM system. By far, themost important change that is observed in the MEM-D2O solvent as compared tothe D2O solvent is the interaction between micelles. In D2O, there are no stronginteractions between micelles at 2 wt%. In MEM-D2O, micelles show repulsiveinteractions between them, with the interaction distance having a radius of 140 Å.The effect of such an increase in Rint would be to effectively hinder the movement of

Table 1.Parameters from the data fits using the core-corona model with polydisperse micelles, and a hardsphere repulsion interaction for all cases other than 2 wt% F127 in D2O

Concentration Solvent Nagg R1 (Å) R2 (Å) Rint (Å) σ (%)(wt%)2 D2O 88.9±1.4 46.0 ± 0.2 70.9 ± 0.4 — 10.4 ± 0.1

MEM-D2O 142.5±1.4 52.2 ± 0.2 81.6 ± 0.3 140 ± 3 11.4 ± 0.15 D2O 73.4±0.3 42.8 ± 0.1 66.5 ± 0.1 101.3 ± 0.4 10.4 ± 0.03

MEM-D2O 77.4±0.5 43.8 ± 0.1 67.7 ± 0.2 98.9 ± 0.5 10.0 ± 0.1

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

1146 P. K. Sharma et al.

the micelles, allowing the system to form a gel at a lower concentration. In reality,it is more likely to be a combination of several effects that results in the formationof a gel at lower temperatures and concentrations.

The aggregation number and micelle size of F127 in water have been shownto increase with increasing temperature [5]. We expect a similar trend for F127in MEM, yielding a larger aggregation number and micelle size at 37◦C. Due tothe greater micellar volume fraction that would result from larger micelles, wealso expect the intermicellar interactions to be stronger at this temperature. Theaggregation number and micelle size are expected to remain higher in MEM incomparison to water.

Gelation has been attributed to an increase in the micellar volume fraction abovea critical value [25], a decrease in the critical micelle concentration and increasein the aggregation number [26] and dehydration of the PPO groups in the micellecore [5]. Our results indicate an increase in the micellar volume fraction withMEM as compared with just D2O (due to an increase in micellar size), a decreasein the critical micelle concentration and an increase in the aggregation number.Therefore, although we do not have any information on the dehydration of the PPOcores, it would appear that according to these arguments, our evidence suggests thatthe presence of MEM would indeed promote the process of gelation. This couldresult in the liquid-to-gel transition being able to occur at lower temperatures andconcentrations as has been observed previously [14].

Due to the components present in MEM, there may be weak electrostatic interac-tions between adjacent F127 chains, causing the micelles to swell and interact withone another, even at 2 wt% co-polymer. In solvents with a dielectric constant lowerthan water (such as methanol and acetonitrile), the presence of a salt containing amonovalent cation (at low ionic strength) has been shown to result in complex for-mation, causing PEO chains to behave as weak polyelectrolytes [27, 28]. This hasbeen reported to be due to binding between the monovalent cations and the etheroxygens of PEO in solvents less polar than water, leading to electrostatic repulsionsbetween PEO chains and thus an increase in the radius of gyration [27, 28]; theseeffects are not exhibited in pure water [27, 28]. Since MEM contains several aminoacids with non-polar functional groups, it is not unreasonable to expect the dielec-tric constant to be lower than that of water. Furthermore, since MEM also containssalts of the monovalent cations K+ and Na+ (at relatively low concentrations), itfollows that the PEO segments of F127 could perhaps undergo complexation withthese cations, leading to electrostatic repulsions between the PEO chains of a mi-celle and thus some degree of micellar swelling. This could explain the increase inmicelle size and strong intermicellar interactions observed in the presence of MEMfrom our SANS data at 2 wt% co-polymer.

Concentrated liquids and gels

The data from 5 wt% F127 samples in D2O and MEM-D2O were fitted using thesame model as the dilute 2 wt% samples. At this concentration, a correlation peak

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

PEO-PPO-PEO co-polymers in cell-culture media 1147

Figure 2. SANS spectra on 5 wt% F127 in D2O and MEM-D2O, fitted to the core-corona modelwith polydisperse micelles. In both cases, the hard sphere repulsion interaction has been included inthe model. Data are shown between 0.007 Å−1 and 0.5 Å−1, due to the large uncertainty in scatteredintensity at low and high values of q.

was visible for both spectra, so a hard sphere repulsion interaction was includedin both the model fits to the data. The SANS spectra and fits are shown in Fig. 2,and the parameters from the fits are given in Table 1. It is interesting to note thatthe large differences observed for the 2 wt% samples in D2O and MEM-D2O havedecreased considerably. However, the aggregation number, overall micelle size andmicelle core size remain slightly higher in the presence of MEM. Since both D2Oand MEM-D2O systems exhibit intermicellar interactions at this higher co-polymerconcentration, and since they are not significantly different, it is likely that anyswelling of the micelles in the presence of MEM observed in the dilute system isoverwhelmed by the greater number and proximity of adjacent micelles exertingexcluded volume repulsions at this higher concentration.

The SANS spectra of 10 wt% F127 in D2O and MEM-D2O are shown in Fig. 3. Atthis concentration, the samples are in the liquid phase, but strong correlation peakscorresponding to intermicellar interactions are present in both D2O and MEM-D2O,along with a broad shoulder, which is evidence of the micelles starting to becomearranged into a cubic order, but not yet forming a liquid-crystal gel. Hence, sincethe gel state is being approached [14], micelles have much less space to move in,resulting in stronger interactions between them. The correlation peaks for D2Oand MEM-D2O in Fig. 3 appear to occur at about the same value of q, with thelatter being sharper and more intense. This suggests that there is a slightly greaterdegree of ordering in the MEM-D2O solvent, and could be evidence that the stronger

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

1148 P. K. Sharma et al.

Figure 3. SANS spectra on 10 wt% F127 in D2O and MEM-D2O. Both systems are in the liquidphase. Data are shown between 0.008 Å−1 and 0.3 Å−1, due to the large uncertainty in scatteredintensity at low and high values of q.

Figure 4. SANS spectra on 18 wt% F127 in D2O and MEM-D2O. Both systems are in the gel phase.Data are shown between 0.01 Å−1 and 0.5 Å−1, due to the large uncertainty in scattered intensity atlow and high values of q.

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

PEO-PPO-PEO co-polymers in cell-culture media 1149

Figure 5. SANS spectra on 25 wt% F127 in D2O and MEM-D2O. Both systems are in the gel phase.Data are shown between 0.01 Å−1 and 0.6 Å−1, due to the large uncertainty in scattered intensity atlow and high values of q.

interactions observed at 2 wt% are indeed still present in the system. Furthermore,this indicates that in MEM-D2O the micelles are closer to the gel transition pointthan in D2O, which agrees with the previously reported gelation boundary data [14].

Data taken on samples in the gel region of the phase diagram (18–25 wt% F127)are shown in Figs 4 and 5. Spectra in both MEM and D2O are characteristic of acubic arrangement of micelles and show several higher-order peaks. At 18 wt% co-polymer, the peaks in MEM-D2O are slightly sharper and stronger than in D2O;however, at 25 wt% co-polymer, the spectra are virtually indistinguishable. At18 wt% co-polymer, both samples are just inside the gel state envelope [14], sothere is still some difference visible between the intermicellar interactions of eachsystem. At 25 wt% co-polymer, samples are well within the gel state boundary [14],so micelles in both D2O and MEM-D2O are strongly ordered, leaving the spectrapractically indistinguishable. In general, the observations at 18 wt% and 25 wt% co-polymer confirm that the mechanism of gelation and gel structure remain unchangedin the presence of MEM, with perhaps slightly stronger interactions and strongerordering in MEM-D2O than in D2O.

CONCLUSIONS

Our SANS data showed that for dilute solutions of F127, there was a significantincrease in the micellar aggregation number in MEM-D2O (accompanied by slightincreases in the micelle core and overall sizes). More significantly, the data in

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

1150 P. K. Sharma et al.

MEM-D2O required the inclusion of a structure factor in the model, in orderto satisfactorily describe the data. The size of the repulsive interaction distanceassociated with this structure factor was found to be much larger than the micellesize. At higher F127 concentrations, although the differences in the scatteringpatterns became less obvious, the data indicated a greater degree of ordering inMEM-D2O. Our DLS experiments showed a decrease in the CMC of Pluronic F127micelles in the presence of MEM. These observations are likely to be a direct resultof some of the components of MEM, such as salts and amino acids. Combinedwith the observations from SANS, our findings may account for the lower gelationtemperatures previously observed for F127 solutions in MEM. Changes in the self-assembly and structure of F127 micelles may also impact the transport of nutrientsand proteins through Pluronic gels in tissue-engineering applications.

Acknowledgements

We gratefully acknowledge Dr. Jyotsana Lal and Ed Lang (IPNS, ANL) for assis-tance with the SANS experiments and Prof. Susan Roberts (UMass) for discussionson mammalian cell culture protocols. This work was partially supported throughthe Hamilton Sundstrand Summer Internship Program, a Commonwealth CollegeSophomore Honors Fellowship and a UMass Engineering Alumni Association Fac-ulty/Student Project Grant for J. E. M., and a 3M Nontenured Faculty Grant andDupont Young Professor Award for S. R. B. This work benefited from the use ofIPNS beam lines at Argonne National Laboratory, which are funded by the Officeof the Basic Energy Sciences, US DOE and was carried out under the auspices ofthe Office of the Basic Energy Sciences of the US DOE under contract # W-31-109-ENG-38. This work partially utilized central facilities of the NSF-sponsored UMassMRSEC on Polymeric Materials (DMR-0213695).

REFERENCES

1. M. Malmsten, in: Amphiphilic Block Co-polymers: Self-assembly and Applications, P. Alexan-dridis and B. Lindman (Eds), p. 319. Elsevier, New York, NY (2000).

2. S. M. Moghimi and A. C. Hunter, Trends Biotechnol. 18, 412 (2000).3. A. V. Kabanov and V. Y. Alakhov, in: Amphiphilic Block Co-polymers: Self-assembly and

Applications, P. Alexandridis and B. Lindman (Eds), p. 347. Elsevier, New York, NY (2000).4. M. Malmsten and B. Lindman, Macromolecules 25, 5440 (1992).5. G. Wanka, H. Hoffmann and W. Ulbricht, Macromolecules 27, 4145 (1994).6. R. K. Prud’homme, G. W. Wu and D. K. Schneider, Langmuir 12, 4651 (1996).7. K. M. Spangenberg, M. M. Farr, A. K. Roy, C. Vacanti, L. Bonassar and J. Cortiella, Tissue Eng.

4, 476 (1998).8. S. Khattak, S. R. Bhatia and S. C. Roberts, Tissue Eng. 11 (2005) (in press).9. J. E. Matthew, S. R. Bhatia and S. C. Roberts, Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.)

43, 769 (2002).10. S. D. Hokett, M. F. Cuenin, R. B. O’Neal, W. A. Brennan, S. L. Strong, R. R. Runner,

J. C. McPherson and T. E. Van Dyke, J. Periodontol. 71, 803 (2000).

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014

PEO-PPO-PEO co-polymers in cell-culture media 1151

11. Y. L. Cao, A. Rodriguez, M. Vacanti, C. Ibarra, C. Arevalo and C. A. Vacanti, J. Biomater. Sci.Polymer Edn 9, 475 (1998).

12. C. A. Arevalo-Silva, R. D. Eavey, Y. Cao, M. Vacanti, Y. Weng and C. A. Vacanti, Arch.Otolaryngol. Head Neck Surg. 126, 1448 (2000).

13. G. E. Yu, Y. L. Deng, S. Dalton, Q. G. Wang, D. Attwood, C. Price and C. Booth, J. Chem. Soc.,Faraday Trans. 88, 2537 (1992).

14. J. E. Matthew, Y. L. Nazario, S. C. Roberts and S. R. Bhatia, Biomaterials 23, 4615 (2002).15. N. K. Pandit and H. J. McIntyre, Pharm. Dev. Technol. 2, 181 (1997).16. I. R. Schmolka, J. Biomed. Mater. Res. 6, 571 (1972).17. D. J. Yarusso and S. L. Cooper, Macromolecules 16, 1871 (1983).18. I. Goldmints, G. E. Yu, C. Booth, K. A. Smith and T. A. Hatton, Langmuir 15, 1651 (1999).19. L. Yang, P. Alexandridis, D. C. Steytler, M. J. Kositza and J. F. Holzwarth, Langmuir 16, 8555

(2000).20. J. S. Pedersen, Adv. Colloid Interf. Sci. 70, 171 (1997).21. V. Castelletto and I. W. Hamley, Curr. Opin. Colloid Interf. Sci. 7, 167 (2002).22. P. R. Desai, N. J. Jain, R. K. Sharma and P. Bahadur, Colloids Surf. A Phys. Eng. Asp. 178, 57

(2001).23. P. Alexandridis, J. F. Holzwarth and T. A. Hatton, Macromolecules 27, 2414 (1994).24. P. K. Sharma and S. R. Bhatia, Int. J. Pharm. 278, 361 (2004).25. M. J. Song, D. S. Lee, J. H. Ahn, D. J. Kim and S. C. Kim, Polym. Bull. 43, 497 (2000).26. M. Bohorquez, C. Koch, T. Trygstad and N. Pandit, J. Colloid Interf. Sci. 216, 34 (1999).27. I. F. Hakem and J. Lal, Appl. Phys. A Mater. Sci. Process. 74, S531 (2002).28. I. F. Hakem and J. Lal, Europhys. Lett. 64, 204 (2003).

Dow

nloa

ded

by [

Nip

issi

ng U

nive

rsity

] at

18:

00 1

0 O

ctob

er 2

014