Conferring materials with antibacterial properties 3

For example, the addition of CS nanoparticles up to a CS/bone cement powder ratio of 15% does not result in a significant loss of Young’s and bending modulus both values ≥ 97% of the val

CHAPTER 6 NOVEL STRATEGIES FOR CONFERRING

ANTIBACTERIAL PROPERTIES TO BONE CEMENT

6.1 Antibacterial and Mechanical Properties of Bone Cement Impregnated with Chitosan Nanoparticles

6.1.1 Introduction

The use of poly(methyl methacrylate) (PMMA)-based bone cement for the anchoring

of artificial joints started in 1958 when Sir John Charnley first succeeded in anchoring femoral head prostheses in the femur using auto-curing PMMA (Kühn, 2000) The PMMA cements in use today bear many similarities with the original cement used by Charnley (Daniels et al., 1998) Commercial bone cements are supplied as two component systems consisting of a powder and a liquid that are mixed in the operating room during the surgical procedure and delivered to the implant site The powdered portion of the cement contains PMMA particles (~ 10 to 150 μm in diameter), an initiator for polymerization (benzoyl peroxide) and a radiopaque medium (such as barium sulfate or zirconium dioxide) The main ingredient of the liquid portion is

MMA monomer The other ingredient, N,N-dimethyl-p-toluidine (DMpT), acts as an

activator in the polymerization reaction and the liquid is stabilized with a small amount

of hydroquinone to prevent polymerization during storage of the product and provide a sufficient shelf life When the solid and liquid components of the bone cement system

are mixed, the BPO from the powder and the DMpT from the liquid participate in a

redox reaction that produces free radicals which initiate addition polymerization of the

MMA monomer As the polymerization reaction proceeds, the powder/liquid mixture

undergoes a transition from the liquid state to a dough and finally to the solid state, forming a rigid, amorphous polymer

The infection rates in primary joint replacement range from 1% to 3% despite strict antiseptic operative procedures, including systemic antibiotic prophylaxis and special

to treat and eradicate biomaterial-associated infection with systemic antibiotic regimens is usually due to the fact that implant infection is associated with biofilm formation Implant-associated infections in orthopedic surgery are thought to be

mainly due to S aureus, S epidermidis and Pseudomonas or other Gram-negative rods

(Van de Belt, 2001) Such infections can rarely be eradicated without removal or revision of the infected implant It is expected that as the population ages, more patients will require joint replacement surgery, and when such orthopaedic prostheses are complicated by infection, management is usually highly problematic, and significant morbidity and increased costs are expected (Segreti, 2000)

Antibiotic-loaded bone cements (ALBC) have been in use for over 30 years and have been the standard of care in arthroplasty The clinical benefits of ALBC combined with systemic antibiotic prophylaxis have been reported in a study with more than 22,000 patients (Engesæter et al., 2003) However, a number of problems related to their use still persist It is known that the antibotics are released from the ALBC in a biphasic fashion, with an initial peak release followed by slow release which may continue for days to months (Hendriks et al., 2004) Hence, while the ALBC may be effective in preventing bacterial infection in the initial period, this protective effect is generally lost

if bacterial contact with the cement occurs after a delay of several weeks Furthermore, there is a worry that the long-term low concentrations of antibiotic around the implant site may well lead to the occurrence of antibiotic-resistant strains, which may then require the addition of a second antibiotic to the bone cement (Neut et al., 2005) An increase in the antibiotic content may extend the period of protection against biofilm formation but aminoglycosides including gentamicin have considerable nephrotoxic side-effects (Parlakpinar et al., 2006) and the mechanical strength of the cement may

also be compromised since high quantities of antibiotics may lead to incomplete polymerization of the cement (Neut et al., 2005)

As described in Section 2.2, a number of surface functionalization techniques have been developed to confer substrates with antibacterial properties However, the use of such techniques is confined to preformed substrates and is not suitable for bone cement since the cement is mixed into a doughy mass just prior to insertion into a prepared cavity Previous works on conferring antibacterial properties to bone cement have relied on loading with antibiotics (Van de Belt et al., 2001) or silver (Dueland et al., 1982; Vik et al., 1985) However, the increasing use of ALBC may lead to the development of resistant strains of bacteria, and the antibiotic action is also relatively short-lived Although silver has a broad antibacterial effect, high concentrations of silver ions may have cytotoxic effects (Dueland et al., 1982; Vik et al., 1985), and the use of nanosized silver in bone cement has been proposed to be a better alternative (Alt

et al., 2004)

In our present work, we explored a strategy based on the incorporation of chitosan (CS) into bone cement CS (poly(1,4),-β-D-glucopyranosamine), an abundant natural biopolymer (structure shown below), is derived by the deacetylation of chitin obtained from the shells of crustaceans and can be fabricated into film, fiber, bead, and powder forms

O O H

O

O H

NH O

O O H

O

O H

NH2

It has great potential as a biomaterial due to its biological activities, low toxicity toward mammalian cells (Kumar et al., 2004; Sashiwa and Aiba, 2004) and antibacterial activity in controlling growth of bacteria and inhibiting viral multiplication (Huh et al., 2001; No et al., 2002; Rabea et al., 2003) To increase the antibacterial activity and solubility of CS, quaternary ammonium CS derivatives (QCS) have been prepared and their antibacterial activities were shown to increase with increasing chain length of the alkyl substituent (Kim et al., 1997)

However, one concern is that the incorporation of CS into bone cement may be expected to result in degradation of the mechanical properties of the bone cement To address this concern, CS and QCS in nanoparticulate form (NP) were used in our modification of two types of commercial bone cement Mechanical testing was performed, and the antibacterial activities of the modified bone cements were assessed

against Gram-positive S aureus and S epidermidis These two bacteria were chosen

since they are commonly associated with infections of orthopaedic implants, wounds and indwelling medical devices (Gristina, 1987; Schierholz and Beuth, 2001)

6.1.2 Experimental

Materials and reagents

Two types of bone cement with and without gentamicin were used: CMW Smartset and CMW Smartset-G (DePuy International Ltd., UK), and Palacos R and Palacos R-G (Biomet, Merck, Germany) The compositions of the bone cements, as given in the manufacturers’ information leaflets, are shown in Table 6.1 CS was purchased from CarboMer Inc and refined twice by dissolving it in dilute acetic acid (HOAc) solution The solution was filtered and the CS was precipitated with aqueous sodium hydroxide and then dried in a vacuum oven for 24 h at 40 oC The viscosity-average molecular weight was about 2.2×105 as determined by the viscometric method (Zhang et al., 2004) The degree of deacetylation was 84% as determined by elemental analysis using the Perkin-Elmer Model 2400 elemental analyzer (Zhang et al., 2004)

Smartset

(w/w%)

Palacos R (w/w%)

Smartset G (w/w%)

Palacos R-G (w/w%)

Powder

Methylmethacrylate

methacrylate copolymer 84.00 83.88 80.46 82.15 Zirconium dioxide 15.00 15.32 14.37 15.01 Benzoyl peroxide 1.00 0.80 0.96 0.78

S aureus (ATCC 25923) and S epidermidis (ATCC 12228) were obtained from

American Type Culture Collection The minimum inhibitory concentration (MIC) of gentamicin as determined by the broth dilution method recommended by the National

Committee for Clinical Laboratory Standards (2003) is 0.12 μg/ml for S aureus and 0.025 μg/ml for S epidermidis

Synthesis of QCS

QCS was synthesized according to the method previously reported (Huh et al., 2001)

In brief, 1 g of CS was added to 50 ml N-methyl-2-pyrrolidinone and suspended by stirring at room temperature for 12 h The temperature of the suspension solution was then lowered to 4 oC using an ice water bath A 1.5 N NaOH aqueous solution (15 ml), potassium iodide (1.2 g) and hexylbromide (13 g) were added to this solution and its temperature was raised to 45 oC and maintained at this value for 48 h while stirring The reaction solution was then filtered using a mesh (120 mesh) to remove the insoluble portion The filtrate was precipitated into a large excess of acetone and filtered using a filter paper The precipitate was re-dispersed and washed with acetone

3 times and the resulting product was dried under vacuum

Preparation of CS NP and QCS NP

CS NP was prepared using the ionic gelation method (Calvo et al., 1997) CS was dissolved in 1 v/v% HOAc solution at a concentration of 0.5 w/v% and the pH was raised to 4.6-5 with 10 N NaOH CS NP was formed upon adding 5 ml of 0.25% TPP

in water to 15 ml CS solution under stirring at a speed of 1000 rpm The nanoparticles were separated by centrifugation at 20000 rpm for 30 min The supernatant was discarded and the CS NP was extensively rinsed with water to remove any NaOH and then freeze-dried before further use QCS NP was obtained from QCS using the same method

Characterization

The chemical composition of the surfaces was analyzed by XPS as described in Section 3.1.2 The nanoparticles were examined using a FE-SEM (JEOL JSM 6700F) and the size and size distribution were determined by laser light scattering with a Brookhaven LLS 90 Plus Particle Size Analyzer The dried nanoparticles were first suspended in water and sonicated to obtained a homogeneous suspension before the measurement The zeta potential of the nanoparticles was measured by a zeta potential analyzer (Zeta Plus from Brookhaven Instruments) with palladium electrodes, and the mean of six readings was calculated

Preparation of cements

Chitosan in the form of powder or nanoparticles was mixed with bone cement powder

at weight ratios of 15:100 and 30:100 All the cements were prepared by manually mixing the powder with the liquid monomer in a ratio of 2 g/ml in a bowl in a laminar flow hood, in accordance with the manufacturer's instructions The polymer powder was placed in a bowl and the monomer was added and stirred using a spatula until the powder was fully wetted The mixture was subsequently inserted into the mould at approximately dough time, usually about 1 min The filled mould was pressed between two glass plates for 1 h After the cement had hardened, it was pulled out of the mould and stored under dark, sterile conditions at room temperature Different moulds were used to make samples for the different tests Rectangular beams (25 × 10 × 2 mm3) were used for the bending tests, while the specimens for the tensile tests were 75 mm

in length, 5 mm in width, 3 mm in thickness, with a gauge length of 25 mm (Harper and Bonfield, 2000) Cylindrical specimens (6 mm in diameter and 2 mm in height) and rectangular specimens were prepared for the cytotoxicity and antibacterial assays

respectively The preparation of bone cement with the QCS nanoparticles was carried out in a similar manner

Testing of mechanical properties

The tensile test and three-point bending test were performed on the Instron universal materials testing machine (Model 5544) The tensile test (according to ASTM D638-03) was conducted at a cross-head speed of 1 mm/min For the three-point bending test (according to ASTM D790-3) the span length was 20 mm and the loading rate was 1 mm/min Five specimens were used in each mechanical test The bending modulus (EB) was calculated according to Eq (1):

where L is support span (mm), b is width of beam tested (mm), d is depth of beam tested (mm), and m is slope of the tangent to the initial straight-line portion of the load-deflection curve (N/mm)

Determination of antibacterial activity

Two Gram-positive bacterial strains S aureus and S epidermidis were cultivated as

described in Section 3.1.2 An aliquot (2 ml) of culture was then added to the yeast-dextrose broth and incubated for 6-8 h at 37 oC until the exponential growth phase was reached This culture was then adjusted by spectrophotometric measurement

at 600 nm to provide a final density of 108 colony forming units (CFU)/ml in broth (From calibrations, an optical density of 1 at 600 nm is equivalent to 109 bacterial cells/ml)

Bone cement substrates (before and after 3 weeks immersion in phosphate-buffered saline (PBS) at 37 oC) were immersed in 15 ml of the bacteria suspension and shaken

at 100 rpm at 37 oC After 3h, the substrates were removed with sterile forceps and gently washed with sterile PBS The substrates were then placed in broth and the bacteria retained on substrates were dislodged by mild ultrasonication (for 5-7 min) in

a 100 W ultrasonic bath operating at a nominal frequency of 50 Hz followed by rapid vortex mixing (10 s) Serial ten-fold dilutions were performed and viable counts estimated following the spread plate method The number of CFU on each sample surface were counted and expressed relative to the surface area of the sample (CFU/cm2) All experiments were performed in triplicate with three substrates and the mean values were calculated

Viability of bacteria on bone cement surfaces

The viability of bacteria on the surface of bone cements was investigated by staining with a combination dye as described in Section 5.2

In vitro cytotoxicity assay

3T3 mouse fibroblasts cells (3T3-Swiss albino, ATCC) culture and cell viability testing were carried out as described in Section 4.1

Statistical analysis

The data obtained was evaluated for statistical significance using the two sample t-test

The results are reported as mean ± SD and the differences observed between substrates

were considered significant at P < 0.05

6.1.3 Results and discussion

Characterization of CS and QCS powders and NP

The FTIR spectra of CS and QCS powders are shown in Figure 6.1

Figure 6.1 FTIR spectrum of (a) CS and (b) QCS powders

can be fitted with three peaks due to neutral amine at 399.4 eV, amide at 400.0 eV and protonated amines at 401.8 eV The sum of NH2 and N+ components is 87%, close to 84% which is expected for the CS (84% deacetylated) For QCS, the high BE peak (above 400 eV) is assigned to the positively charged nitrogen (N+) The N+/N ratio is 0.65, indicating that about 65% of the nitrogen in CS have been quaternized in the hexylbromide alkylation process

Figure 6.2 XPS N 1 s core-level spectra of (a) CS and (b) QCS powders

-NH-C=O

In the ionic gelation method for the preparation of the NP, an anionic cross-linking agent, TPP, was added to an aqueous solution of CS in acetic acid CS or QCS NP was formed through interactions between the positively charged CS or QCS and negatively charged phosphate groups of TPP Table 6.2 shows the size and size distribution of CS and QCS NP as determined by the laser light scattering technique The average diameter of the CS NP and QCS NP were determined to be 220 nm and 284 nm respectively In contrast, the FE-SEM images of these NP (Figure 6.3) show that the average size of the CS NP is around 70 nm, while that of the QCS NP is around 100

nm The larger size of the nanoparticles determined from laser scattering measurements is probably due to some degree of aggregation upon suspension in water (as can be seen from Figure 6.3(b) and (c)) The surface charge (zeta potential) of the

CS and QCS are also shown in Table 6.2 The zeta potential of NP can greatly influence the stability of nanoparticles in suspension through electrostatic repulsion between the particles, and it also determines the extent of interaction with the cell membrane of bacteria, which is usually negatively charged As shown in Table 6.2, the zeta potential increases from 48 mV to 67 mV after quaternization

Table 6.2 Characteristics of CS NP and QCS NP

Nanoparticles Mean diameter (nm) Polydispersity Zeta potential (mV)

48.87 ± 0.46 67.32 ± 0.14

(a)

Figure 6.3 FE-SEM images of CS NP ((a) and (b)) and QCS NP (c)

Mechanical properties of CS and QCS loaded bone cements

The mechanical properties of the acrylic bone cement used in orthopedic surgery play

an influential role in determining the successful long-term stability of a prosthesis (Harper and Bonfield, 2000; Ries et al., 2006) The CS powder can be easily mixed with the PMMA bone cement powder and cast However, mechanical testing of such samples shows that the Young’s modulus and bending modulus are significantly compromised when the CS/bone cement powder weight ratio is 30% (Figure 6.4(a) and Figure 6.5(a)) Even when the CS loading was decreased to 15%, the Young’s modulus and bending modulus are about 90% of the corresponding properties of the original bone cement On the other hand, the mechanical tests showed when the CS was added in the form of nanoparticles instead of powder, the bone cement can better retain its mechanical properties For example, the addition of CS nanoparticles up to a CS/bone cement powder ratio of 15% does not result in a significant loss of Young’s and bending modulus (both values ≥ 97% of the values of the original bone cement, see Figure 6.4(a) and 6.5(a)) The improved mechanical properties of the composite cement when chitosan of nano size was used is attributed to its more uniform dispersion in the PMMA matrix such that no “macroscopic” weak links are present in the cement The results in Figure 6.4(a) and Figure 6.5(a) also show that after 3 weeks immersion in PBS (pH=7.4), the Young’s modulus and bending modulus of the original and modified bone cements decrease by ~10% or less The biggest decrease was observed in the Young’s modulus of the cement loaded with CS powder at weight ratios of 15% and 30% The mechanical properties of the QCS NP-loaded cement are not significantly different from those of the CS NP-loaded cement, either in the freshly prepared form or after extended immersion in PBS The mechanical tests on the various gentamicin-loaded cements before and after 3 weeks in PBS also yield a similar

Figure 6.4 Young’s modulus of original and CS (30%), CS, CS NP and QCS

NP-loaded Smartset plain (a) and gentamicin-loaded (b) bone cements With the exception of CS (30%), the other substrates were mixed at a weight ratio of

CS or QCS to PMMA bone cement powder of 15% (#), (*) denote significant

differences (P < 0.05) compared with original bone cement which is freshly

prepared or after 3 weeks in PBS, respectively

(b)

*

Original CS (30%) CS CS NP QCS NP 0.0

(b)

*

Figure 6.5 Bending modulus of original and CS (30%), CS, CS NP

and QCS NP-loaded Smartset plain (a) and gentamicin-loaded (b)

bone cements With the exception of CS (30%), the other substrates

were mixed at a weight ratio of CS or QCS to PMMA bone cement

powder of 15%

trend (Figure 6.4(b) and Figure 6.5(b)) In view of these results, the antibacterial assays were carried out with composites having a CS or QCS to bone cement powder weight ratio of 15% in order to achieve a balance between mechanical strength and antibacterial effectiveness

Antibacterial assay

The antibacterial effects of the various bone cements were investigated by a

comparison of the number of viable S aureus or S epidermidis cells after contact with

the different substrates From Figure 6.6(a), it can be seen that with the addition of CS

to the Smartset cement (at a CS to bone cement powder ratio of 15%), the number of

viable S aureus cells on this cement surface decreased by 50% compared to pristine

cement When CS NP and QCS NP were used instead of CS at the same ratio, the viable cell number decreased by about two orders and three orders of magnitude, respectively It can also be seen from Figure 6.6(a) that after the composite cement samples have been immersed in PBS for 3 weeks at 37 °C, the antibacterial effect is still preserved especially for the substrates with CS NP and QCS NP From Figure

6.6(b), it can be observed that the CS NP and QCS NP are just as effective against S epidermidis

The exact mechanisms of the antibacterial action of CS and its derivatives are still unknown, although different mechanisms have been proposed Interactions between positively charged CS and negatively charged bacterial cell membranes lead to altered cell permeability, which prevents the transport of essential solutes into the cell (Choi et al., 2001; Hu et al., 2003) and results in leakage of proteinaceous and other intracellular components, thus killing the bacteria cells (Jung et al., 1999) CS NP

exhibits higher antibacterial activity than CS powder since the polycationic CS NP has higher surface area and charge density than the CS powder and can interact to a greater degree with the negatively charged surface of the bacterial cell

0 1 2 3 4 5 6 7

2 )

Freshly prepared After 3 weeks in PBS at 37 o C

Figure 6.6 Number of viable adherent S aureus (a) and S epidermidis (b)

cells on the different substrates based on Smartset bone cement without gentamicin

The QCS NP exhibits even higher antibacterial activity than the CS NP The average zeta potential of QCS NP (67 mV) is higher than that of CS NP (48 mV) due to the presence of the quaternary ammonium groups The effectiveness of such groups bearing alkyl substituents in disrupting bacterial cell membranes and causing cell lysis has been documented earlier (Hu et al., 2005) Hence, the high surface charge density

of the QCS NP which increases the affinity for the negatively charged bacterial cell membrane coupled with the effectiveness of the C6 alkyl substituent in penetrating this membrane is responsible for their observed higher antibacterial activity in Figure 6.6

The release mechanisms of gentamicin loaded in ALBC are poorly understood and difficult to control, and damage to the kidneys by gentamicin is also a potential problem (Hendriks et al., 2004) The results of the antibacterial assay of the gentamicin-loaded Smartset bone cement are given in Figure 6.7 It can be seen that

some S aureus bacteria can remain viable after contact with the gentamicin-loaded

cement for 3 h (Figure 6.7(a)), which is probably due to the high bacterial cell concentration in the nutrient broth After 3 weeks immersion in PBS at 37 °C, the antibacterial effectiveness of the gentamicin-loaded bone cement decreases significantly and the number of viable cells on the cement surface increases by almost

a factor of 4 as a result of the loss of gentamicin into the PBS From a comparison of Figure 6.7(a) and 6.7(b), it can be seen that the gentamicin-loaded cement is more

effective against S epidermidis than S aureus, which is expected due to the lower MIC for S epidermidis than S aureus.However, after 3 weeks immersion in the PBS, its effectiveness against this bacterium is also greatly reduced It has been reported that

Figure 6.7 Number of viable adherent S aureus (a) and S epidermidis (b)

cells on the different substrates based on Smartset bone cement with gentamicin

0 200

(b)

*

0 200

(a)

*

ALBC generally prevents attempts by bacteria to infect the bone cement haematogenously, if it occurs within an hour after implantation However, if bacterial contact with the implant occurs after a delay of several weeks (which may be the case

of infection in some other parts of the body circulating via the bloodstream to the implant site), this protective effect no longer exists (Blomgren and Lindgren, 1981) Our results in Figure 6.7 areconsistent with this postulate

The results in Figure 6.7 show that the addition of CS or CS NP and QCS NP further enhances the antibacterial property of the gentamicin-loaded bone cements Similar to the results in Figure 6.6, the QCS NP possessed the highest antibacterial effectiveness For both bacteria, the number of cells that remained viable upon contact with the QCS

NP incorporated gentamicin-loaded bone cements is 10 times less than that in contact with the commercial gentamicin-loaded cement The loss of antibacterial efficacy after

3 weeks immersion in PBS is also much less pronounced in the bone cement loaded with CS NP or QCS NP Another important advantage in using CS NP or QCS NP as bactericidal agents is the low probability of the bacteria developing resistance after repeated exposure since the polycations are not selective in penetrating the bacterial membrane (Milović et al., 2005) In contrast, there is a risk that the use of gentamicin-loaded bone cement may lead to the emergence of resistance in subjected strains (Hendriks et al., 2004) since release rates from such cements are generally low and sub-inhibitory antibiotic concentrations over extended periods of time may stimulate antibiotic resistance among infectious microorganisms

The distribution of viable and dead bacteria which adhered on the surface of the substrates after immersion in the bacterial suspension of 108 cells/ml for 3 h at 37 oC

was observed via staining with the combination dye Figure 6.8 shows the fluorescence microscopy images (observed under green and red filters) of the various Smartset bone

cements after exposure to S aureus The presence of many viable cells (stained green)

can be seen on the plain bone cement surface (Figure 6.8(a)), while at the same time, there were very few dead cells (Figure 6.8(b)) With CS in the bone cement, the number of viable cells decreases, as shown in Figure 6.8(c) and the number of dead cells increases (Figure 6.8(d)) The number of viable cells is further decreased after CS was added in the form of nanoparticles (Figure 6.8(e) and (f)) In the case of the QCS

NP loaded bone cement, there are very few viable cells (Figure 6.8(g) and the vast majority of the cells are stained red (Figure 6.8(h)) The fluorescence microscopy results confirmed the effectiveness of CS, and particularly the QCS NP, in killing the bacteria upon contact The surviving cells are likely to be those which had adhered to spots without nanoparticles or onto other cells which were in contact with the surface

Figure 6.8 Fluorescence microscopy images of Smartset bone cement, and CS,

CS NP and QCS NP-loaded bone cement substrates under green filter (a, c, e, g) and red filter (b, d, f, h), respectively, after immersion in bacterial suspension (108 S aureus cells/ml, for 3 h at 37 oC)

Effect of different bone cements

The antibacterial assay results discussed so far in Section 6.1.3 have been obtained with the Smartset and Smartset-G bone cements Similar tests were carried out with the Palacos R and Palacos R-G bone cements and the results are summarized in Table 6.3

the different substrates based on Palacos bone cement

1.3×10 5 (2.5×10 5 ) 7.9×10 4 (1.0×10 5 ) 2.5×10 3 (5.0×10 3 ) 2.0×10 2 (7.9×10 1 )

** CS and QCS to bone cement powder weight ratio of 15%

The results essentially show trends that are similar to those obtained with Smartset bone cement, ie:

(i) The effectiveness of the CS NP and QCS NP-loaded cements without

gentamicin against S epidermidis is similar to that against S aureus

(ii) QCS NP is more effective than CS or CS NP in killing the bacteria The number of viable bacterial cells on the QCS NP-loaded Palacos bone cement (without gentamicin) is 3 orders of magnitude less than that on the original Palacos cement

(iii) The number of viable bacterial cells on QCS NP-loaded cement which has been immersed in PBS for 3 weeks is still 3 orders of magnitude lower than

on the original commercial cement

(iv) The addition of CS NP and QCS NP to gentamicin-loaded Palacos bone cement confers an added antibacterial effect This effect is retained to a

significantly larger extent than that conferred by gentamicin if the cement is immersed for a prolonged period in an aqueous medium

Cytotoxicity assay

Since the ultimate objective of our work is to confer antibacterial properties in implanted materials, the issue of cytotoxicity of the modified bone cement and its degradation products has to be addressed The results of the MTT cytotoxicity assay using 3T3 mouse fibroblasts are shown in Table 6.4 These cells were selected for this assay as they are substrates-dependent, nonspecific cell lines The results in Table 6.4 show that there is no statistical difference in cytotoxicity among CS, CS NP and QCS

NP loaded bone cements (P > 0.05) as compared to the plain cement It should also be

noted that the results obtained with these bone cements are not significantly different from that of the non-toxic control (growth culture medium)

Table 6.4 Cytotoxicity assay of 3T3 cells on the different bone cements

Substrates Cell viability % of control

Growth medium

PMMA bone cement

* CS-loaded bone cement

* CS NP-loaded bone cement

* QCS NP-loaded bone cement

* CS and QCS to bone cement powder weight ratio of 15%

6.1.4 Conclusions

This in-vitro study has demonstrated that the incorporation of nanoparticles of chitosan and quaternary ammonium chitosan derivative in bone cements can provide effective

antibacterial action against S aureus and S epidermidis These nanoparticles also

enhance the antibacterial efficacy of gentamicin-loaded bone cements and this property

is retained even after an extended period of immersion of the modified bone cement in

an aqueous medium The nanoparticles provide a high surface charge density for interacting with and disrupting bacterial cell membranes Since the nanoparticles can

be uniformly mixed with the bone cement, the mechanical properties of the bone cement are not significantly compromised even at a chitosan/bone cement powder weight ratio of 15% The MTT assay showed that there is no significant difference in cytotoxicity between the CS NP, QCS NP and the non-toxic control

6.2 Antibacterial and Mechanical Properties of Bone Cement Containing a Monomer or Polymer with Norfloxacin Moieties 6.2.1 Introduction

In this section, another strategy for confering long-lasting antibacterial properties to bone cement is described This strategy is based on the incorporation of the antibiotic norfloxacin (NOR), either in the form of an acryl monomer moiety or a polymer form

in the bone cement It can be expected that with the antibiotic incorporated in this manner, the mechanical properties of the bone cement can be preserved since norfloxacin in the modified form may be more compatible with the PMMA component Norfloxacin is the first in a series of new fluoroquinolone antibiotics that have been introduced into medical practice for treatment of bacterial infections It is a broad spectrum bactericidal agent that is much more potent than the earlier analogs such as nalidixic acid, and is less likely to result in resistant mutants compared to nalidixic acid (Gadebusch and Shungu, 1991; Hooper and Rubinstein, 2003) Its water solubility

is 0.37 mg/ml at 25 °C, significantly lower than that of gentamicin (>50 mg/ml) (Yu et al., 1994) Recently, norfloxacin was conjugated to mannosylated dextran in order to increase the drug’s uptake by cells, enabling it to gain faster access to microorganisms (Roseeuw et al., 1999; Roseeuw et al., 2003) Yang et al incorporated norfloxacin as a monomer into a polyurethane backbone structure and their antimicrobial activities were tested (Yang and Santerre, 2001) Poly(acrylated quinolone) with norfloxacin as the pendant group on the polymer chain has been shown to have very good antibacterial activities against Gram-positive and Gram-negative bacteria (Kim et al., 2005) In this work, glycidyl methacrylate (GMA) was reacted with NOR to result in

an antibiotic-containing monomer (M-NOR) The homopolymer (P-NOR) was then synthesized using conventional free-radical solution polymerization M-NOR

When the liquid and solid components of the bone cement are mixed together, the norfloxacin-containing monomer is expected to copolymerize with the methyl methacrylate This allows for a higher concentration of antibiotic to be incorporated in the cement without compromising the mechanical strength Furthermore, since norfloxacin has a lower solubility than gentamicin, its release profile is moderated and its concentration on the cement surface can be maintained at a sufficiently high level to inhibit biofilm formation for a longer period of time Mechanical properties of the modified bone cements were tested and the antibacterial activities of the modified bone

cements were assessed against Gram-positive S aureus and S epidermidis

6.2.2 Experimental

Materials and reagents

Norfloxacin was purchased from Sigma-Aldrich The minimum inhibitory concentration (MIC) of norfloxacin as determined by the broth dilution method recommended by the National Committee for Clinical Laboratory Standards (2003) is

0.25 μg/ml for S aureus and 0.05 μg/ml for S epidermidis The materials and other

reagents involved in this work were all the same as those specified in Section 3.1.2

Synthesis of M-NOR and P-NOR

M-NOR was synthesized according to the method previously reported (Kim et al., 2005) (Figure 6.9) In brief, NOR (5.0 g, 0.016 mole) and GMA (2.2 g, 0.016 mole)

were mixed in 30 ml N,N-dimethylformamide (DMF) and stirred at 40 oC for 24 h The reaction mixture was then cooled, precipitated into water and extracted with dichloromethane (CH2Cl2) Excess CH2Cl2 was evaporated by a rotary evaporator The desired product was washed with diethyl ether and then recrystallized from CH2Cl2

and dried under vacuum Characterization of the product was carried out by FTIR, XPS and elemental analyses

P-NOR was synthesized using free-radical polymerization A mixture of M-NOR (2 g, 4.3 mmol) and AIBN (0.05 g, 0.30 mmol) in DMF was stirred at 40 oC for 24 h The reaction mixture was cooled and precipitated into a 50/50 (v/v) water/acetone mixture The solid yellow product was filtered and washed with acetone and dried under vacuum

Characterization

The as-synthesized M-NOR was mixed with KBr to make test pellet specimens for FTIR analysis using a Shimadzu FTIR 8400 spectrophotometer The bulk chemical composition was analyzed using a Perkin-Elmer model 2400 CHN elemental analyzer Molecular weight determination was carried out using gel permeation chromatography

Figure 6.9 Schematic representation of synthesis of M-NOR and P-NOR

O O

O

N N F

F

N

OH OH

N

OH OH

N N O

P-NOR

(GPC) on an HP 1100 HPLC, equipped with a HP 1047A refractive index detector Tetrahydrofuran (THF) was used as the eluent at a flow rate of 1 ml/min at 35 °C Polystyrene standards were used as the reference The XPS measurements were carried out as described in Section 3.1.2 SEM images and EDX spectra of the surfaces of bone cements were obtained using the JEOL JSM 5600L scanning electron microscope with a EDX accessory

Preparation of cements

NOR, M-NOR or P-NOR in the form of powder was mixed with bone cement powder

at weight ratios of 4:100 or 15:100 All the cements were prepared by manually mixing the powder with the liquid monomer in a ratio of 2 g/ml in a bowl in a laminar flow hood as described in Section 6.1.2

Testing of mechanical properties

The tensile test and three-point bending test were performed as described in Section 6.1.2

Determination of antibacterial activity

Bacterial cell culture and determination of antibacterial activity were carried out as described in Section 6.1.2

In vitro cytotoxicity assay of M-NOR and P-NOR

3T3 mouse fibroblasts cell culture and cell viability testing were carried out as described in Section 6.1.2

6.2.3 Results and Discussion

Characterization of M-NOR and P-NOR

Figure 6.10 shows the FTIR spectra of NOR and M-NOR The absorption bands at around 1750 cm-1 and 1250 cm-1 are attributed to the C=O and C-O groups respectively of carboxylic acid in NOR The peak at around 1720-1750 cm-1 in the M-NOR spectrum is broader and more intense compared to that of NOR due to the stretching of the additional C=O groups of M-NOR (see Figure 6.9) The results of the elemental analysis of the as-synthesized M-NOR compare well with the theoretical values expected for C23H28FN3O6 (calculated values (wt%): C: 59.86%; H: 6.12%; N: 9.11%; experimental values: C: 59.79%; H: 6.17%; N: 9.08%) The C/N mass ratio obtained from XPS analysis is about 6.41/1 The results from the FTIR and elemental analysis of the as-synthesized monomer are thus consistent with those expected of M-NOR The polymerization of M-NOR monomer using AIBN as initiator at 40 °C for 24 h in DMF yielded a polymer (P-NOR) of weight-average molecular weight of

16204 and polydispersity index of 1.86

Figure 6.10 FTIR spectrum of (a) NOR and (b) M-NOR powders

Characterization of bone cement surfaces

The SEM images of the surfaces of the various bone cements are shown in Figure 6.11 From this figure it can be observed that the PMMA beads in the original cement (Figure 6.11(a)) are homogeneously distributed on the surface and held together by the PMMA from the polymerization of the liquid MMA monomer initiated by the BPO The SEM image of the M-NOR-loaded cement (Figure 6.11(b)) shows a different morphology, which may be indicative of the copolymerization of the MMA monomer and M-NOR The copolymerization would be facilitated by the presence of the C=C group in MMA and the GMA moiety linked to NOR The surface morphology of the P-NOR-loaded cement (Figure 6.11(c)) also differs from that of original cement The more irregularly shaped component observed in Figure 6.11(c) is likely to be P-NOR, and it appears to be more separated from the PMMA component than that observed in Figure 6.11(b) This is understandable since C=C groups in P-NOR would not be available for copolymerization with the MMA The surface morphology of the gentamicin-loaded cement (Figure 6.11(d)) is similar to that of the original cement The EDX signal at 0.68 keV in the insets of Figure 6.11(b) and (c) is attributed to fluorine, which confirms the presence of norfloxacin on the surface of the respective cements Similarly, the signal at 2.3 keV in Figure 6.11(d) is attributed to the sulfur moiety of gentamicin

(a)

(b)

0 5 1 0 1 5 2 0 2 5

P t O

K e V C

P t C

O

K e V F

Figure 6.11 Scanning electron micrographs and EDX of (a) original, (b)

M-NOR, (c) P-NOR and gentamicin-loaded CMW Smartset bone cements For (b) and (c), the weight ratio of M-NOR or P-NOR to PMMA bone cement powder is 15%

Mechanical properties of NOR-modified cements

The desire for increased antibiotic content in bone cement must be balanced with the necessity for mechanical integrity of the bone cement One potential disadvantage of ALBC is the decrease in mechanical properties of bone cement after adding a high dose of antibiotics (Lautenschlager et al., 1976; Hanssen, 2004) The compressive and tensile strength of bone cement has been shown to decrease significantly with addition

of more than 2 g of gentamicin to 40 g of cement powder (Lautenschlager et al., 1976) Figure 6.12 compares the Young’s modulus and bending modulus of the Smartset and Smartset G bone cements with the bone cements modified with norfloxacin It can be seen that there is no significant change in the mechanical properties when NOR was added to bone cement powder at a weight ratio of 4% However, when the NOR to bone cement powder weight ratio is increased to 15%, the mechanical properties decrease significantly On the other hand, it can be seen that the Young’s modulus and bending modulus increase significantly when M-NOR was added to bone cement powder, even up to a weight ratio of 15% (Figure 6.12) The improved mechanical properties of the M-NOR loaded cement may be due to the copolymerization of M-NOR with MMA monomer, which results in more crosslinking between PMMA beads, as can be seen from the SEM image in Figure 6.11(b) In the case of P-NOR modified cement (15% loading), the Young’s modulus and bending modulus are not significantly different from those of the original and gentamicin-loaded bone cements The results in Figure 6.12(a) and Figure 6.12(b) also show that after the original and modified bone cements have been immersed for 3 weeks in PBS (pH=7.4), the Young’s modulus and bending modulus decrease by ~10% or less

*

Freshly prepared After 3 weeks in PBS at 37oC

(a)

#

*

Figure 6.12 Young’s modulus (a) and bending modulus (b) of original, NOR,

M-NOR, P-NOR and gentamicin-loaded bone cements NOR, M-NOR or P-NOR were mixed with PMMA bone cement powder at a weight ratio of either 4% or 15%

as indicated on the X-axis (#), (*) denote significant differences (P < 0.05)

compared with original bone cement which is freshly prepared or after 3 weeks in PBS, respectively