Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx – xxx nanomedjournal.com
Challenges facing sterilization and depyrogenation of nanoparticles: Effects on structural stability and biomedical applications, Melissa A. Vetten, Msc a, b,⁎, Clarence S. Yah, PhD c, d , Tanusha Singh, PhD e, f , Mary Gulumian, PhD a, b a
Toxicology and Biochemistry Section, National Institute for Occupational Health, National Health Laboratory Services, Johannesburg, South Africa b Department of Molecular Medicine and Haematology, School of Pathology, University of the Witwatersrand, Johannesburg, South Africa c Human Sciences Research Council, Newton Park, South Africa d Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA e Immunology and Microbiology Section, National Institute for Occupational Health, National Health Laboratory Services, Johannesburg, South Africa f Department of Immunology, School of Pathology, University of the Witwatersrand, Johannesburg, South Africa Received 11 December 2013; accepted 29 March 2014
Abstract This review outlines and compares techniques that are currently available for the sterilization of nanoparticles and addresses the topic of endotoxin contamination. Several techniques are available for the removal of microbial contamination from nanoparticles developed for use in nanomedicine applications. These techniques include filtration, autoclaving and irradiation, as well as formaldehyde, ethylene oxide and gas plasma treatments. Of these sterilization methodologies, filtration may potentially remove microbial contamination without altering the physicochemical properties of the carrier nanoparticles, nor affecting their toxicity and functionality. However, no single process may be applied to all nanoparticle preparations and, therefore, it is recommended that each nanoparticle-drug system be validated on a case-by-case basis. © 2014 Elsevier Inc. All rights reserved. Key words: Nanoparticles; Sterilization; Endotoxin
Abbreviations: 60Co, Cobalt-60; 137Cs, Cesium-137; AuNPs, gold nanoparticles; β-CDC6, Amphiphilic β-cyclodextrin; Cu/LDPE, Copper/ low-density polyethylene; DLS, dynamic light scattering; e-beam, electron beam; EO, Ethylene oxide; EDS, Energy-dispersive spectroscopy; ESD, Emulsion Solvent Diffusion; FDA, Food and Drug Administration; FTIR, Fourier transform infrared; Hap, hydroxyapatite; HS, hydrothermal synthesis; IUDs, intrauterine devices; LAL, limulus amoebocyte lysate; LPS, LPS lipopolysaccharide; NP, nanoparticle; PBCA, poly(butyl cyanoacrylate); PBLG, poly(γ-benzyl-L-glutamate); PEC, poly(ɛ -caprolactone); PEG, poly (ethylene glycol); PEG-AuNPs, poly(ethylene glycol) coated gold nanoparticles; PLGA, poly(DL-lactide-co-glycolide); SAL, sterility assurance level; SLNs, solid lipid nanoparticles; TEM, transmission electron microscopy; TiO2, titanium dioxide; tiopronin-AuNPs, tiopronin coated gold nanoparticles; UV-Vis, ultraviolet-visible; WCS, wet chemical synthesis. The authors acknowledge the financial support from the Department of Science and Technology, South Africa, and the National Institute for Occupational Health. The authors declare that there are no competing interests. ⁎Corresponding author at: Toxicology and Biochemistry Section, National Institute for Occupational Health, National Health Laboratory Services, Johannesburg, South Africa. E-mail address:
[email protected] (M.A. Vetten).
The unique physical and chemical properties of nanoparticles (NPs) have resulted in substantial research being performed on their potential applications in various fields including biology and medicine. Nanomedicine applications include the development of “lab-on-chip” technology in order to assess biomarkers, drug and gene delivery, tissue engineering, and cancer therapy. 1,2 Any nanoparticle-drug based formulation requires the solvent to be sterile and apyrogenic, in addition to being safe, non-toxic and non-irritating for both in vitro and in vivo applications. The term sterility refers to the absence of viable microorganisms that could pose a risk when administered. The current accepted sterility assurance level (SAL) is limited to 10 −6, that is to say not more than one viable microorganism in one million parts of final product is allowed. 3 Manufacturers of medical devices are required to ensure that their products meet established quality requirements and specifications, including regulations regarding microbial contamination. Potential sources of microbial contamination during production of pharmaceutics include the raw materials, equipment and processes used during production, in addition to the facility and personnel. 4
http://dx.doi.org/10.1016/j.nano.2014.03.017 1549-9634/© 2014 Elsevier Inc. All rights reserved. Please cite this article as: Vetten M.A., et al., Challenges facing sterilization and depyrogenation of nanoparticles: Effects on structural stability and biomedical applications. Nanomedicine: NBM 2014;xx:1-9, http://dx.doi.org/10.1016/j.nano.2014.03.017
2
M.A. Vetten et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
Table 1 Summary of the effects of filtration on nanoparticles. Pore size of filter
Type of nanoparticle
Mean approximate size
Effect of sterilization on nanoparticle
Outcome of sterilization method
Authors
0.2 μm
Poly(ε-caprolactone) nanospheres PEGylated poly(γ-benzyl-L-glutamate) nanoparticles Polyester nanoparticles
130 nm; 160-180 nm 50 nm
No aggregation and no effect on the nanoparticle morphology No change in size or polydispersity index. Slight change in zeta potential.
Not tested
7 8
sub 200 nm
Poly(DL-lactide-co-glycolide) nanospheres
200-300 nm 103-163 nm
No clogging of the membrane filters. No significant change in particle size and size distribution Less than 10% of nanospheres passed through filter. 100-98% of nanospheres passed through filter.
No detectable growth of bacteria, yeast or fungi No bacterial contamination Passed bacterial sterility test
0.22 μm 0.22 μm 0.2 μm
9 12
Potential contaminating microorganisms include bacteria, fungi and mould; however the removal of endotoxins, the lipopolysaccharides in the cell membrane of Gram-negative bacteria, must also be addressed. Magalhães and colleagues 5 provided a review of endotoxins and their removal from biological preparations. They discussed the pathophysiological effects of endotoxins, as caused by activation of the immune system and release of proinflammatory mediators, which lead to endotoxin shock, tissue injury and sometimes death. Since the effects of endotoxin are related to the amount of endotoxin present, in the case of drug products, this would imply that the endotoxin limit would be dependent on the amount of drug product administered to the patient. The formula for the endotoxin limit provided by the Food and Drug Administration (FDA) is K/M; where K is 5.0 EU/kilogram, and M is the maximum recommended dose of product per kilogram of body weight administered in 1 h. 6 Although numerous well-established sterilization techniques exist, concerns have been raised regarding the adverse effects that these techniques may have on the physicochemical characteristics of the nanoparticles. A change in these characteristics could potentially affect both the toxicity and the efficacy of the sterilized nanoparticles. This paper reviews various sterilization methodologies that have been assessed for the removal of microbial and endotoxin contamination from nanoparticle preparations with the aim of providing a possible recommendation on a suitable methodology for nanomedicine production and sterilization.
not appear to have any adverse effects on the nanoparticles (Table 1). Poly(ε-caprolactone) (PEC) nanospheres of mean diameter below 200 nm were successfully sterilized by filtration with 0.2 μm cellulose acetate membrane filters without alteration of their size, morphology or concentration. 7 Filter sterilization through 0.22 μm filters has also successfully been implemented for the sterilization of PEGylated poly(γ-benzyl-L-glutamate) (PBLG) NPs 8 and polyester NPs. 9 The use of 0.2 μm or 0.22 μm filters may not always be possible if the NPs are larger than, or close to, the pore size of the filters since clogging can occur resulting in a decreased yield. 10,11 For example, filtration of 200-300 nm poly(DL-lactide-co-glycolide) (PLGA) nanospheres prepared using the standard Emulsion Solvent Diffusion (ESD) method resulted in less than 10% of the nanospheres passing through the membrane filter. 12 This problem was circumvented by optimizing the synthesis methodology to produce nanospheres with a particle diameter of 103-163 nm, of which 100-98% could pass through the membrane filter and successfully pass bacterial sterility tests. However adjusting the size of the nanoparticles in order to enable them to pass through a membrane filter may not always be feasible. Sterile filtration may therefore present itself as a reasonable method for the removal of bacterial contamination, provided that a sufficiently high percentage of nanoparticles can be recovered following filtration.
Methodologies implemented for the sterilization of nanoparticles
Autoclaving
A literature survey of publications obtained from reputable journals shows that several conventionally used methodologies such as filtration, autoclaving, irradiation, as well as treatment with formaldehyde, ethylene oxide and gas plasma, have been implemented for the sterilization of nanoparticles. Filtration Sterile filtration is a commonly used method for the physical removal of microorganisms from chemically and thermally sensitive liquids, through the use of 0.22 μm membrane filters. This technique has been shown to be widely applicable as it does
Autoclaving kills microbes with high pressurized steam, at a minimum temperature of 121 °C, within 15-20 min depending on the size of the load and the contents to be sterilized. This methodology has been shown to have a number of effects on the nanoparticles sterilized (Table 2). Fesharaki and colleagues synthesized selenium nanoparticles using Klebsiella pneumonia bacteria followed by recovery of the nanoparticles from the bacteria by autoclaving at 121 °C, 17 psi for 20 min. Energydispersive spectroscopy (EDS) confirmed chemical stability of the selenium nanoparticles before and after sterilization. 13 On the other hand, autoclaving PBLG NPs at 121 °C for 20 min resulted in the aggregation and a drastic increase in size of these nanoparticles. 8 Further investigation of the effect of autoclaving on aggregation was conducted on PEC nanospheres where it was
Table 2 Summary of the effects of autoclaving on nanoparticles. Type of nanoparticle
Mean approximate size
Effect of sterilization on nanoparticle
Outcome of sterilization method
Authors
121 °C for 20 min
Poly(ε-caprolactone) nanospheres
Aggregation of nanoparticles dependent on polymer or surfactant. Decrease in pH observed.
Not tested
7
121 °C for 20 min
PEGylated poly(γ-benzyl-L-glutamate) nanoparticles
130 nm; 160-180 nm; 220 nm 50 nm
Massive aggregations of the nanoparticles, changes in polydispersity index and zeta potential.
8
121 °C for 20 min
Amphiphilic β-cyclodextrin nanoparticles:
No detectable growth of bacteria, yeast or fungi Not tested
10
Chemically stability according to energy-dispersion spectroscopy. WCS: agglomeration, morphology change, crystallite growth. HS: no significant morphology change, surface charge less negative.
Not tested Not tested
13 14
Minor aggregation of some PEG-AuNPs. No effect on poly(ethylene glycol) layers. Tiopronin-AuNPs were slightly bigger. No effect on tiopronin molecules. Significant increase in particle size.
No bacterial growth
15
Not tested
16
Minor changes in size, polydispersity index and zeta potential dependent on lipid used and aqueous solution. Presence of diazepam did not influence stability.
Not tested
17
No significant change in size or morphology. Increased tendency to cause platelet aggregation in vitro. Different autoclaving methods result in changes in cell adherence to the nanotubes, which was dependent on nanotube size At 110 °C, no changes in particle size or polydispersity, and no toxicity observed using Neutral Red assay. At 121 °C, increase in particle size and toxicity was observed.
Not tested
18
Not tested
19
Not tested
20
121 °C for 20 min 120 °C for 20 min
• Blank nanosphere • Blank nanocapsule • Tamoxifen citrate loaded nanosphere • Tamoxifen citrate loaded nanocapsule Selenium nanoparticles Hydroxyapatite nanoparticles synthesized by either WCS⁎ or HS †
All nanoparticles showed aggregation and increased polydispersity index. Significant reduction of Tamoxifen citrate entrapment. 205 nm 290 nm 235 nm 310 nm 245 nm WCS: needle-like shape, crystallite size 17 nm HS: rod-like shape, crystallite size 38 nm 60 nm 2 nm
134 °C for 40 min
PEG-AuNPs ‡ Tiopronin-AuNPs §
121 °C for 20 min
Unloaded poly(butyl cyanoacrylate) nanoparticles Solid lipid nanoparticles, with or without Diazepam
Diameter in water; trehalose; Pluronic F38:
• Acidan N12 nanoparticles • Behenic acid nanoparticles • Stearic acid nanoparticles Citrate-stabilized silver nanoparticles
60 nm; 85 nm; 78 nm 70 nm; 90 nm; 85 nm 55 nm; 56.5 nm; 76 nm 20, 40, 60, and 80 nm
Titanium dioxide nanotubes
30, 50, 70, 100 nm
Cyclosporine A loaded solid lipid nanoparticles
225.9 nm
121 °C for 15 min
121 °C for 30 min Dry and wet autoclaving (Details not provided) 110 °C for 30 min, or 121 °C for 15 min
Ranged from 116-524 nm
M.A. Vetten et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
Temperature and duration of autoclaving
⁎ WCS: wet chemical synthesis. † HS: hydrothermal synthesis. ‡ PEG-AuNPs: poly(ethylene glycol) coated gold nanoparticles. § AuNPs: gold nanoparticles. 3
4
Type of irradiation
Type of nanoparticle
Mean approximate size
Effect of sterilization on nanoparticle
Outcome of sterilization method
Authors
γ-Irradiation (2.5 MRad)
Poly(ε-caprolactone) nanospheres
130 nm; 160-180 nm; 220 nm
Not tested
7
γ-Irradiation at room temperature to deliver a dose of 1.88 kGy/h
Amphiphilic β-cyclodextrin nanoparticles:
Decrease in pH. Increase in poly(ε-caprolactone) molecular weight, however no change in nanoparticle diameter. Thimerosal-containing formulations showed degradation. Hydroxyethyl-cellulose containing formulations showed depolymerisation. All nanoparticles showed: No effect on size or polydispersity. Slight changes in zeta potential. No significant change in drug entrapment or drug release profile.
Not tested
10
No effect on PEG-AuNP Aggregation of tiopronin-AuNPs. Dramatic changes in the size, morphology and biocompatibility of the nanoparticles.
No bacterial growth
15
Not tested
18
Degradation from γ irradiation could be restored by UV annealing.
Not tested
22
No changes in size of either particle. No changes in molecular mass with doses up to 25 kGy. Irradiation at 35 kGy lead to increase in molecular mass and polydispersity of drug-loaded nanoparticles.
Sterilization of bacteria achieved with a dose ≥ 15 kGy irrespective of type of irradiation.
23
UV irradiation at room temperature for approximately 12 h γ-Irradiation at room temperature (20-23 °C) to deliver 15, 25, or 50 kGy. γ-Irradiation of a total dose 25 kGy, followed by UV annealing at room temperature γ-Irradiation and e-beam irradiation at ambient temperature to deliver doses of 10, 15, 25, 35 kGy.
• Blank nanosphere • Blank nanocapsule • Tamoxifen citrate loaded nanosphere • Tamoxifen citrate loaded nanocapsule PEG-AuNPs⁎ Tiopronin-AuNPs † Citrate-stabilized silver nanoparticles
Polysilicon wire pH sensors coated with 3-aminopropyltriethoxysilane/silica nanoparticle nanocomposite Empty poly(butyl cyanoacrylate) nanoparticles Doxorubicin-loaded poly(butyl cyanoacrylate) nanoparticles
⁎ PEG-AuNPs: poly(ethylene glycol) coated gold nanoparticles. † AuNPs: gold nanoparticles.
205 nm 290 nm 235 nm 310 nm 60 nm 2 nm 20, 40, 60 and 80 nm Silica particle size: 14 nm 301 nm 245 nm
M.A. Vetten et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
Table 3 Summary of the effects of irradiation on nanoparticles.
M.A. Vetten et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
shown to be dependent on the polymer or surfactant used during synthesis. Nanospheres prepared with Cremophor RH40 showed aggregation following autoclaving, while those prepared with Pluronic remained the same size. 7 The manner of synthesis was also shown to influence effects of autoclaving on hydroxyapatite (Hap) nanoparticles, where nanoparticles synthesized using wet chemical synthesis (WCS) showed significant aggregation and morphology change following autoclaving, in comparison to the Hap nanoparticles produced by hydrothermal synthesis (HS) which showed no change. 14 In addition, the surface charge of the HS Hap nanoparticles became less negative due to autoclaving, while the charge of the WCS Hap nanoparticles remained unchanged. The effects of various sterilization techniques on the physicochemical properties of poly(ethylene glycol) (PEG) coated gold nanoparticles (PEG-AuNPs) and tiopronin coated gold nanoparticles (tiopronin-AuNPs) were determined by transmission electron microscopy (TEM), ultraviolet-visible (UV-Vis) spectroscopy, and Fourier transform infrared (FTIR). 15 Autoclaving of the PEG-AuNPs produced minor aggregation of particles and no changes in the PEG coating. In contrast, autoclaving of tiopronin-AuNPs resulted in the growth of the particles from approximately 2 nm to 5 nm, a process likely caused by a growth and recrystallization process known as Ostwald ripening. A significant increase in particle size was also observed after autoclaving unloaded polybutylcyanoacrylate NPs. 16 Amphiphilic β-cyclodextrin (β-CDC6) nanospheres and nanocapsules, in addition to their Tamoxifen citrate loaded counterparts, aggregated and increased in size, possibly due to heat-induced chemical modifications of β-CDC6. 10 This disruption of NP integrity resulted in a significant decrease in the quantity of Tamoxifen citrate entrapped by both the nanospheres and the nanocapsules. On the other hand, it has been reported that sterilization at 121 °C of either drug-free or drug-loaded solid lipid nanoparticles (SLNs) had no significant effect on NPs. 17 Apart from physicochemical changes that may or may not arise during autoclaving, this method may also produce changes in the biological effects of the sterilized nanoparticles that should be investigated. Autoclaving of citrate-stabilized silver nanoparticles at 121 °C for 30 min has been shown to have no significant effect on the size or morphology of the NPs, as determined by dynamic light scattering (DLS) and TEM. Despite this, the autoclaved NPs showed an increased tendency to cause platelet aggregation, a sensitive in vitro indicator of thrombogenicity. 18 Dry and wet autoclaving of titanium dioxide (TiO2) nanotubes was also shown to differently affect the adhesion, proliferation and alkaline phosphatase activity of osteoblasts. 19 Finally, in vitro cytotoxicity studies on Cyclosporine A-loaded SLNs, which were autoclaved at 110 °C for 30 min, showed no toxicity, whereas nanoparticles autoclaved at 121 °C for 15 min demonstrated significant toxicity. This increase in toxicity was explained to be due to release of the surfactant Tween 80 during sterilization at higher temperatures, which lead to cell damage. 20 The majority of literature reviewed on sterilization via autoclaving showed that the aggregation of nanoparticles, as well as changes in their biological activities, is of concern. Moreover,
5
it would appear that these effects are not only dependent on the type of nanoparticles and their coating but also on the methodology and reagents used during their synthesis.
Irradiation Sterilization using irradiation can be achieved through gamma (γ) irradiation, electron beam (e-beam), X-rays, UV light irradiation or through the use of subatomic particles. The most common form used due to its high penetration is γ rays, which are generated by the self disintegration of Cobalt-60 ( 60 Co) or Cesium-137 ( 137 Cs). The advantages of this technique include its application for the sterilization of many materials since it is independent from chemicals or heat, and it leaves no residue after sterilization. E-beam sterilization is also commonly used where very high electron energy is used for a shorter duration than γ-irradiation. UV light irradiation is generally only used for sterilization of surfaces using a germicidal lamp. 21 γ-Rays and e-beams provide very efficient disinfection since they are able to penetrate through materials but both can cause damage to the material being sterilized. 22 An irradiation dose of ≥ 15 kGy of either γ-irradiation or e-beam irradiation has been shown to be effective in the sterilization of poly(butyl cyanoacrylate) (PBCA) NPs that were deliberately contaminated with spore-forming bacteria. 23 Despite the efficacy of this mode of sterilization to kill bacteria, various adverse effects have been observed on nanoparticles when irradiation was used for their sterilization (Table 3). Where both γand e-beam irradiation at 10, 15, 25, and 35 kGy were used to sterilize doxorubicin-loaded or empty shell PBCA NPs, no effect on size or molecular mass could be observed. Only the drug-loaded NPs sterilized by γ-irradiation at 35 kGy could show an increase in the molecular mass of PBCA and the polydispersity. 23 In another study, a decrease in pH and an increase in molecular weight of PEC nanospheres could be observed, with no alteration in particle size when sterilized by γ-irradiation. Moreover, this sterilization method also caused the degradation of both the antimicrobial agent thimerosal and the anti-sedimentation agent hydroxyethyl-cellulose which had been added to the nanosphere formulation. 7 A study of citrate-stabilized silver NPs demonstrated that γ-irradiation at 15, 25 and 50 kGy had a significant effect on the size and morphology of the NPs. Following γ-irradiation, NPs were shown to lose their typical polycrystalline structure and form both smaller particulates and larger irregular aggregates. 18 In addition, these γ-irradiated NPs exhibited a four- to five-fold increase in their ability to cause platelet aggregation and, therefore, γ-irradiation would not be suitable in the sterilization of these nanoparticles. On the other hand, no significant changes in size were reported following γ-irradiation of either blank or Tamoxifen citrate loaded β-CDC6 nanospheres and nanocapsules. Although slight changes in zeta potential were observed, this did not result in any significant changes in nanoparticle properties. In addition, it had no effect on the amount of drug entrapped by the NPs or on the in vitro drug release profile, thereby demonstrating that γ-irradiation could be applied for the sterilization of these NPs. 10 Other studies have also been conducted on the sterilization of microparticles developed as drug delivery systems. For example, a study investigating the effects of βand γ-irradiation on gentamycin-loaded collagen/PLGA
6
M.A. Vetten et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
Table 4 Summary of the effects of sterilization on nanoparticles treated with formaldehyde, ethylene oxide, and gas plasma. Sterilization method
Type of nanoparticle
Mean approximate size
Effect of sterilization on nanoparticle
Outcome of sterilization method
Authors
Formaldehyde treatment at 60° for 60 min Formaldehyde treatment at 60 °C Ethylene oxide treatment at 54 °C for 60 min 500 mg/L ethylene oxide treatment at 50 ± 5 °C, 60% relative humidity, for 2 h
PEG-AuNPs⁎ Tiopronin-AuNPs †
60 nm 2 nm
Minor aggregation of some PEG-AuNPs. Aggregation of tiopronin-AuNPs.
No bacterial growth
15
Unloaded poly(butyl cyanoacrylate) nanoparticles (lyophilized powder) PEG-AuNPs Tiopronin-AuNPs
Ranged from 116 to 524 nm
Powders showed impaired resuspension characteristics and increase in size.
Not tested
16
60 nm 2 nm
No effect on either Tiopronin-AuNPs or PEG-AuNPs morphologies or coatings.
No bacterial growth
15
Not specified. Nanocomposite intrauterine devices consisted of 15.0 wt% copper nanoparticles. 60 nm 2 nm
No influence on the internal structure, surface functional groups, mechanical property and cupric ion release rate of the devices.
Not tested
27
PEG-AuNP aggregation. Alterations of poly(ethylene glycol) layer. No effect on Tiopronin-AuNPs
No bacterial growth
15
Gas plasma treatment at 45 °C for 50 min
Copper/low-density polyethylene nanocomposite
PEG-AuNPs Tiopronin-AuNPs
⁎ PEG-AuNPs: poly(ethylene glycol) coated gold nanoparticles. † AuNPs: gold nanoparticles.
microparticle composites showed a decrease in molecular weight and glass transition temperature following sterilization. However, these changes did not influence the morphology or the gentamycin release profile from the composites. 24 Similarly, a decrease in molecular weight of PLGA microspheres was observed after γirradiation without any changes in tetracycline-HCl content. 25 Despite the changes observed following sterilization in these examples, irradiation could still potentially be implemented for sterilization since the changes observed did not appear to adversely affect the attached drug, thereby maintaining the end-use function of the drug delivery system. γ-Irradiation could, however, be used for the sterilization of biosensors where, although γ-irradiation was shown to degrade polysilicon wire pH sensors covered with 3aminopropyltriethoxysilane/silica NP nanocomposites, the authors found that the sensors could be restored following γ-irradiation using UV annealing. 22 One study has used UV irradiation at room temperature for 12 h to sterilize PEG- and tiopronin-AuNPs. Although effective at preventing bacterial growth, this sterilization approach could only be applied to PEGAuNPs as the properties remained unaltered. The tiopronin-AuNPs displayed varying degrees of aggregation and coalescence of irregular shapes. 15 The studies described above demonstrate that irradiation is possibly an option for sterilization of drug-loaded nanoparticles. However, irradiation does often appear to have adverse effects on the NPs and it would therefore be necessary to individually validate each NP-drug system separately.
Formaldehyde treatment Formaldehyde is an organic solution commonly used as a disinfectant and as a fixative. There is limited information
concerning its use for sterilization of NPs for subsequent use in biomedical applications (Table 4). Formaldehyde sterilization is useful for materials that are sensitive to high temperature; however its major limitation arises from its toxicity and carcinogenicity. 21 Formaldehyde treatment of lyophilized PBCA NP powders at 60 °C showed increases in size and clumping following resuspension of the NPs. 16 Formaldehyde treatment at 60 °C for 1 h showed slight aggregation of PEG-AuNPs, while tiopronin-AuNPs showed significant aggregation and the formation of larger, irregularly shaped particles. 15 Exposure of the formaldehyde-sterilized tiopronin-AuNPs to human cell line U937 in vitro resulted in a significant decrease in cell viability. This was attributed either to the observed aggregation and changes in the shape of the nanoparticles, or to formaldehyde residues present following sterilization. Due to both these adverse effects, formaldehyde may not be recommended for sterilization of nanoparticles.
Ethylene oxide treatment Ethylene oxide (EO) treatment is widely used for the sterilization of medical devices. EO acts as a strong alkylating agent which denatures nucleic acids and functional proteins of microorganisms, thereby presenting itself as an excellent sterilization methodology. 26 In terms of its disadvantages, EO is flammable and explosive, and EO and its residues are also toxic and carcinogenic. 21,26 Not many studies have been conducted to order to investigate the effect of EO sterilization on nanoparticles (Table 4). It has been reported to be a good methodology for the decontamination of copper/low-density polyethylene (Cu/LDPE) nanocomposite T-shaped intrauterine devices (IUDs), which contain 15.0 wt% copper nanoparticles. The sterilization was performed at 50 ± 5 °C,
M.A. Vetten et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
with a relative humidity of 60%, and an EO concentration of 500 mg/L, for an exposure time period of 2 h. This treatment did not affect the internal structure, the surface functional groups, the mechanical properties, or the release of cupric ions from the device when tested 14 days after sterilization. 27 Similarly, PEG- and tiopronin-AuNPs sterilized with EO at 54 °C for 60 min showed no alteration in the structure or size distribution, assessed by TEM and UV-Vis spectroscopy, or in the coating material determined by the vibration peaks of FTIR. 15 However, EO sterilization of gentamycin-loaded collagen/PLGA microparticle composites not only resulted in aggregation and changes in the gentamycin release profiles but also produced chemical changes in the gentamycin. 24 It is possible that the same mechanism that makes EO so highly efficient for sterilization may introduce changes in NPs and their associated drugs. The addition of alkyl groups during sterilization may render the microorganisms nonviable, 26 however it may also introduce chemical and structural changes to NP-drug systems, thus removing EO as an ideal mode of sterilization of nanoparticles.
Gas plasma method Plasma can be defined as ionized gas that has properties of both gases and liquids, and has been shown to have a broad spectrum of antimicrobial effect. Although the mechanism of action of gas plasma sterilization still needs further research, it is believed that it is related to oxidation and reduction effects on microbial structures. 28 Advantages of this methodology include its use at low temperatures and its non-toxicity. 29 Effects on nanoparticles were observed in one study using this methodology (Table 4), where the effect of the gas plasma technique was investigated on PEG- and tiopronin-AuNPs. No structural changes of the tiopronin coating or the size and size distributions were observed. This same treatment, however, could cause aggregation, as confirmed by TEM, as well as alteration of the PEG coating, as confirmed by FTIR, possibly due to the highly oxidative nature of the gas plasma procedure. 15 Gas plasma sterilization may therefore not be an ideal method for sterilization of NPs because most of their bio-conjugates are prone to oxidation.
Endotoxin contamination Endotoxin contamination presents itself as a further challenge that must be addressed prior to administration of NPs to human subjects. Currently the in vitro limulus amebocyte lysate (LAL) test is the assay of choice for the determination of endotoxin contamination. The LAL assay exists in three formats, namely chromogenic, turbidity, or gel-clot assessments. A study has shown that even when formal acceptance requirements for the validity for the LAL assay are met, dramatic differences in the results between two different LAL assays for the same nanoparticle can occur. 30 Subsequently underestimation or overestimation of endotoxin concentration may be reported as a result of interference with the assay system. Nanoparticles could interfere with the reactivity of the endotoxin with the assay, with the LAL reaction
7
itself due to the catalytic properties of the NPs, or with the detection of the reaction products due to NP absorbance, luminescence or fluorescence. It is therefore suggested that when results from different LAL assays differ by more than 25%, the results should be verified by the in vivo rabbit pyrogen test. 30 Moreover, the inclusion of appropriate inhibition or enhancement controls is essential to help recognize whether negative results are due to absence of endotoxin, or inhibition of the assay. 31 Endotoxins will easily pass through a 0.22 μm membrane filter that relies exclusively on size exclusion as a retention mechanism. The pharmaceutical industry therefore makes use of several other techniques to remove endotoxin such as various types of chromatography and ultrafiltration, depending on the characteristics of the product to be sterilized. 5 For example, Polymyxin B affinity columns are very effective in removing endotoxin contamination due to their very high binding affinity for lipid A, which is the glycolipid component of lipopolysaccharide (LPS). It has however been observed that the use of Polymyxin B columns to remove LPS contamination from nanoparticles may also lead to the loss of these nanoparticles, especially those that may have high surface binding affinity to LPS. 32 Nanoparticles have a large surface area to volume ratio which enhances their reactivity and affinity to bind contaminants on their surfaces. Smaller magnetite NPs (b 50 nm) have been shown to have a higher affinity to LPS than their larger counterparts (N 100 nm), due to their larger surface area. 33 The charge of the nanoparticles may also determine how they will interact with Polymyxin B affinity columns. Endotoxins are mainly negatively charged and therefore have been shown to have especially strong binding affinity for positively charged cyclic peptide Polymyxin B. 34 Charges on the surface of nanoparticles will not only affect the binding of the endotoxin to the nanoparticle but also alter the ability of the nanoparticle to pass through the positively charged Polymyxin B column. In summary, to date, no efficient technique has been established to remove endotoxin due to the complexity of LPS and nanoparticles. Among the challenges being discussed regarding the removal of LPS contamination from nanoparticles is the extreme heterogeneity of LPS in its composition and structure, differences in size, its thermostability, and its insensitivity to pH changes. 32 The FDA recommendations of applying either high temperatures (250 °C for 45 min) or high concentrations of acids or bases would be effective in removing LPS, but these conditions would most likely also affect the properties of nanoparticles. The ultimate choice would therefore be to keep the production process contaminant free in order to eliminate the problem of removing endotoxins from the nanoparticles. Such contaminant-free processes would involve the use of ultrapure water, gloves, fume hoods, and the cleaning of the work area and glassware with PyroCLEAN™ prior to synthesis. By implementing these precautions, LPS contamination could be lowered by 16-24 fold in the production of spherical gold nanoparticles. 32
Conclusion This review has covered the sterilization methods used for various nanomaterials and the effects of the processes
8
M.A. Vetten et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx
implemented on the properties of the NPs as well as on the attached drugs. The sterilization methodologies discussed have been shown, where tested, to be effective in the removal or destruction of microbial contamination. However, they are frequently shown to alter the physicochemical properties of NPs, the stability of the attached drug, the drug-release profiles, and the toxicity of the sterilized NPs. Filtration presents itself as a preferable option for sterilization, provided that sufficient NP can be recovered from the filtrate. Irradiation appears not to affect the drug release but seems to affect the physicochemical properties of the carrier NPs. Autoclaving resulted in aggregation of the NPs in most cases. The limited information available on formaldehyde, EO and gas plasma techniques suggests that they should not be used due to either their toxicity or their ability to alter the properties of the drug-NP system. It is therefore of vital importance that the toxicity and functionality of the NPs following sterilization be assessed. From all the sterilization methodologies listed, no single process can equally be applied to all NPs and it is recommended that each NP-drug system be validated individually. Regarding the removal of endotoxins from NPs, the ideal situation would be to prevent contamination by maintaining a contaminant-free synthesis process.
Disclaimer The review reflects the views of the authors who are liable for the details and accuracy of the information herein and does not necessarily reflect the official views or policies of any institution or body. This article does not constitute standard specification nor is planned to design, construct, request for permit purposes. Any trade name used was solely for information and not for product endorsement.
References 1. Seigneuric R, Markey L, Nuyten DS, Dubernet C, Evelo CT, Finot E, et al. From nanotechnology to nanomedicine: applications to cancer research. Curr Mol Med 2010;10(7):640-52. 2. Salata O. Applications of nanoparticles in biology and medicine. J Nanobiotechnol 2004;2(1):3. 3. von Woedtke T, Kramer A. The limits of sterility assurance. GMS Krankenhhyg Interdiszip 2008;3(3) Doc19. 4. Kushwaha P. Microbial contamination: a regulatory perspective. J Pharm Res 2010;3(1):124-31. 5. Magalhaes PO, Lopes AM, Mazzola PG, Rangel-Yagui C, Penna TC, Pessoa Jr A. Methods of endotoxin removal from biological preparations: a review. J Pharm Pharm Sci 2007;10(3):388-404. 6. FDA. Bacterial endotoxins/pyrogens. Available from:, http://www.fda. gov/ICECI/Inspections/InspectionGuides/InspectionTechnicalGuides/ ucm072918.htm2009. 7. Masson V, Maurin F, Fessi H, Devissaguet JP. Influence of sterilization processes on poly(epsilon-caprolactone) nanospheres. Biomaterials 1997;18(4):327-35. 8. Özcan I, Bouchemal K, Segura-Sánchez F, Abaci Ö, Özer Ö, Güneri T, et al. Effects of sterilization techniques on the PEGylated poly(γ-benzyl-Lglutamate) (PBLG) nanoparticles. Acta Pharm Sci 2009;51:211-8.
9. Konan YN, Cerny R, Favet J, Berton M, Gurny R, Allemann E. Preparation and characterization of sterile sub-200 nm meso-tetra (4-hydroxylphenyl)porphyrin-loaded nanoparticles for photodynamic therapy. Eur J Pharm Biopharm 2003;55(1):115-24. 10. Memisoglu-Bilensoy E, Hincal AA. Sterile, injectable cyclodextrin nanoparticles: effects of gamma irradiation and autoclaving. Int J Pharm 2006;311(1–2):203-8. 11. Desai N. Challenges in development of nanoparticle-based therapeutics. AAPS J 2012;14(2):282-95. 12. Tsukada Y, Hara K, Bando Y, Huang CC, Kousaka Y, Kawashima Y, et al. Particle size control of poly(DL-lactide-co-glycolide) nanospheres for sterile applications. Int J Pharm 2009;370(1–2):196-201. 13. Fesharaki PJ, Nazari P, Shakibaie M, Rezaie S, Banoee M, Abdollahi M, et al. Biosynthesis of selenium nanoparticles using Klebsiella pneumoniae and their recovery by a simple sterilization process. Braz J Microbiol 2010;41:461-6. 14. Santos C, Gomes PS, Duarte JA, Franke RP, Almeida MM, Costa ME, et al. Relevance of the sterilization-induced effects on the properties of different hydroxyapatite nanoparticles and assessment of the osteoblastic cell response. J R Soc Interface 2012;9(77):3397-410. 15. Franca A, Pelaz B, Moros M, Sanchez-Espinel C, Hernandez A, Fernandez-Lopez C, et al. Sterilization matters: consequences of different sterilization techniques on gold nanoparticles. Small 2010;6(1):89-95. 16. Sommerfeld P, Schroeder U, Sabel BA. Sterilization of unloaded polybutylcyanoacrylate nanoparticles. Int J Pharm 1998;164(1):113-8. 17. Cavalli R, Caputo O, Carlotti ME, Trotta M, Scarnecchia C, Gasco MR. Sterilization and freeze-drying of drug-free and drug-loaded solid lipid nanoparticles. Int J Pharm 1997;148(1):47-54. 18. Zheng J, Clogston JD, Patri AK, Dobrovolskaia MA, McNeil SE. Sterilization of silver nanoparticles using standard gamma irradiation procedure affects particle integrity and biocompatibility. Nanomed Nanotechnol 2011:S:5, http://dx.doi.org/10.4172/2157-7439.S5-001. 19. Oh S, Brammer KS, Moon K-S, Bae J-M, Jin S. Influence of sterilization methods on cell behavior and functionality of osteoblasts cultured on TiO 2 nanotubes. Mater Sci Eng C 2011;31(5):873-9. 20. Gokce EH, Sandri G, Bonferoni MC, Rossi S, Ferrari F, Guneri T, et al. Cyclosporine A loaded SLNs: evaluation of cellular uptake and corneal cytotoxicity. Int J Pharm 2008;364(1):76-86. 21. Sílíndír M, Özer AY. Sterilization methods and the comparison of e-beam sterilization with gamma radiation sterilization. FABAD J Pharm Sci 2009;34:43-53. 22. Lin JJ, Hsu PY. Gamma-ray sterilization effects in silica nanoparticles/ gamma-APTES nanocomposite-based pH-sensitive polysilicon wire sensors. Sensors (Basel) 2011;11(9):8769-81. 23. Maksimenko O, Pavlov E, Toushov E, Molin A, Stukalov Y, Prudskova T, et al. Radiation sterilisation of doxorubicin bound to poly(butyl cyanoacrylate) nanoparticles. Int J Pharm 2008;356(1–2):325-32. 24. Friess W, Schlapp M. Sterilization of gentamicin containing collagen/PLGA microparticle composites. Eur J Pharm Biopharm 2006;63(2):176-87. 25. Bittner B, Mader K, Kroll C, Borchert HH, Kissel T. Tetracycline-HClloaded poly(DL-lactide-co-glycolide) microspheres prepared by a spray drying technique: influence of gamma-irradiation on radical formation and polymer degradation. J Control Release 1999;59(1):23-32. 26. Mendes GC, Brandao TR, Silva CL. Ethylene oxide sterilization of medical devices: a review. Am J Infect Control 2007;35(9):574-81. 27. Xia X, Wang Y, Cai S, Xie C, Zhu C. Will ethylene oxide sterilization influence the application of novel Cu/LDPE nanocomposite intrauterine devices? Contraception 2009;79(1):65-70. 28. Shintani H, Sakudo A, Burke P, McDonnell G. Gas plasma sterilization of microorganisms and mechanisms of action. Exp Ther Med 2010;1(5):731-8. 29. Moisan M, Barbeau J, Crevier M-C, Pelletier J, Philip N, Saoudi B. Plasma sterilization. Methods and mechanisms. Pure Appl Chem 2002;74(3):349-58. 30. Dobrovolskaia MA, Neun BW, Clogston JD, Ding H, Ljubimova J, McNeil SE. Ambiguities in applying traditional Limulus amebocyte lysate
M.A. Vetten et al / Nanomedicine: Nanotechnology, Biology, and Medicine xx (2014) xxx–xxx tests to quantify endotoxin in nanoparticle formulations. Nanomedicine (Lond) 2010;5(4):555-62. 31. Smulders S, Kaiser J-P, Zuin S, Van Landuyt KL, Golanski L, Vanoirbeek J, et al. Contamination of nanoparticles by endotoxin: evaluation of different test methods. Part Fibre Toxicol 2012;9:41. 32. Vallhov H, Qin J, Johansson SM, Ahlborg N, Muhammed MA, Scheynius A, et al. The importance of an endotoxin-free environment
9
during the production of nanoparticles used in medical applications. Nano Lett 2006;6(8):1682-6. 33. Bromberg L, Chang EP, Alvarez-Lorenzo C, Magarinos B, Concheiro A, Hatton TA. Binding of functionalized paramagnetic nanoparticles to bacterial lipopolysaccharides and DNA. Langmuir 2010;26(11):8829-35. 34. Anspach FB. Endotoxin removal by affinity sorbents. J Biochem Biophys Methods 2001;49(1–3):665-81.
