Polymer materials like polyamide are commonly used for sutures, medical tubes, syringes or even mitral valves as well as for various packaging purposes . In all these applications polymers with antimicrobial surface properties are of eminent interest, as they minimize the risk of device related and polymer associated contaminations and infections . Moreover, in addition to the antimicrobial properties, the modified polymer materials should also feature a high biocompatibility, which is often challenging to realize.
In the last years, nanometals like silver have been the subject of widespread research activities and became prominent also for use in medical applications . The microbiocidal activity goes back to the presence (release) of free metal ions [4, 5]. The phenomena that some metal ions are able to damage pathogens, including bacteria and fungi, is commonly known as the oligodynamic effect . Beside silver or copper also zinc ions may have an oligodynamic (antimicrobial) effect [7–10]. Zinc is well known as a trace metal and an essential dietary mineral, which is required to support human biochemical processes and health in contrast to silver. However, in high concentrations, it has been shown to possess antibacterial properties .
To make use of the effect of zinc as an antimicrobial agent, the formation of a suitable compound form is required that fulfills several demands because it is well known that the antimicrobial activity of potential metal-compounds is a direct consequence of the solubilization and metal ion release properties which are governed by the size, morphology and surface chemistry of the particulate agents [11, 12]. In this context it could be shown, that if particulate additives are reduced from a micrometer to a nanometer size, the resultant properties could change drastically [13, 14]. The effect of particle size and dispersion on the biological activity has been demonstrated not only for metallic (e.g., silver) [4, 11, 15], but also for organic antimicrobials (e.g., Triclosan)  and pharmaceutical drugs . Concerning the latter, it could be shown that a molecular dispersion (solid solution) of usually poorly soluble drugs in a given matrix leads to improved bioavailability compared to common particulate (crystalline) disperse systems . A main reason for the increase in bioactivity and bioavailability is supposed to be a substantial increase of the specific surface area (SSA)/bulk volume (VMatrix) ratio with decreasing particle size of the biological active agent. Thus, finely distributed metal-particles can exhibit an increased release rate of corresponding biological active ions [4, 15]. Moreover, a fine distribution – of nanometer or molecular level – is also associated with an increase of atoms or terminal groups of the addition agent present on the surface of the modified bulk material. Consequently, the surface features determine the forces of dynamic interactions that operate between the polymeric nanocomposite material and biological molecules and cellular compartments at the solid-liquid interface with characteristic decay lengths  as schematically illustrated in Figure 1A.
In contrast to the possible benefits of utilizing nano-sized antimicrobial agents, the issue of nanoparticle toxicity or possible bioadverse effects of nanomaterials has recently raised concerns . In general, these problems are often a direct outcome of several things like the use of unsuited surfactants, carriers, uncontrolled release properties (e.g., burst release) and/or elementary leaching of nanoparticular addition agents. To overcome these drawbacks and ensure a nano-sized dispersion, the bioactive component may be encapsulated into the interior of a hyperbranched polymer (hbp) [21–23]. Their ability to act as hosts for the entrapment of guests has been already successfully approved for dye molecules , drugs , DNA  and metal ions as well as metal nanoparticles [22, 27, 28]. The utilization of these special kinds of branched macromolecule architectures as carrier allows the tailoring of important properties like solubility and surface chemistry due to a high density of exo-presented terminal groups [22, 28, 29]. For hyperbranched polyethylenimine (hb-PEI) it is confirmed that the partial modification of terminal groups with hydrophobic moieties leads to amphiphilic core-shell architectures which behave like unimolecular inverse micelles  capable of trapping guests inside the hydrophilic hb-PEI-core [22, 24, 31]. Further, the release of entrapped guests can be controlled by the core-shell-structure properties .
The potential of amphiphilic zinc-loaded hyperbranched polyethylenimine (hb-PEIm-Zn2+) as surface active addition agent for the antimicrobial functionalization of polymeric materials for medical device applications has not been investigated yet. In the study presented, the morphology and antimicrobial main features of polymer composites in the form of cast films (CF) and injection molded plates (IMP) containing the organic-inorganic hb-PEIm-Zn2+-hybrids were investigated. This was exemplified using the thermoplastic polymer polyamide (PA) as the matrix material. The mixing of PA and hb-PEIm-Zn2+ was primarily done by hot melt extrusion. Beside the already mentioned assessment of antibacterial properties, the investigations comprise the analysis of the alteration of the bacterial surface adhesion characteristic by the material functionalization as a model for polymer-associated infections and as the initial step in the development of bacterial biofilms on the surface of a polymer. Furthermore, special regard was also directed to the issue of biocompatibility (in-vitro cytotoxicity and sensitization potential) and hemocompatibility. The latter is an important attribute for materials in direct contact with human blood. Materials with quite good cyto- and hemocompatibility imply great potential for intra- and extra-corporeal medical device applications.
Results and discussion
Characterization of the hb-PEIm-Zn2+-hybrid
Amphiphilic molecule structures are potentially useful as carriers for drugs because of their ability to build micelle-like configurations which can entrap the active component . This has been shown in depth for micelles based on linear block copolymer amphiphiles . But the assemblies are very dynamic structures that can be damaged/collapsing due to shear forces or other external influences. Opposite to this, amphiphilic hyperbranched polymer architectures can impart unique advantages, such as increased stability, and are also suited for the noncovalent encapsulation of guests like drugs or metal ions [21, 25, 29]. The three-dimensional structure of the prepared amphiphilic core-shell-architectures based on high molecular weight hyperbranched polyethylenimine (hb-PEI) provides large cavities for intra-molecular uptake of guests. In this context, solution experiments have shown that while the zinc ion precursor is insoluble in trichloromethane it slowly dissolves in the inverse micelle solution formed in the presence of the hb-PEIm carrier polymer. The self-assembling into discrete zinc-hybrid aggregates is triggered by the entrapment of the zinc ions due to the chelating properties of the hb-PEI-core as illustrated in Figure 1B. The chelating ability of branched polymeric amines is notably higher than that of linear PEI . Thus, branched polymeric amines like hb-PEI are generally able to coordinate more readily to metal ions compared to their linear analogues because the electron-donating nitrogen atoms are already situated in sites from which they can directly interact with metal ions making long range chain reorganization unnecessary . Hence, the amphiphilic modified hb-PEIm is an effective carrier which can entrap zinc ions up to a high loading level (molar ratio of Zn/Nhb-PEI=1:6) as proved by AAS measurements (cf. Table 1). Although higher loads are theoretically conceivable, it is assumed that steric hindrance and strong neighboring interactions lead to thermodynamically unstable ion accumulation and therefore finally flocculation occurs (Zn/Nhb-PEI>1:6). The stock solutions were turbid.
To get more insights into the supramolecular structure of the hb-PEIm-Zn2+-hybrid (Zn/N=1:6) it was further analyzed by FT-IR, TEM and WAXS. The changes in the spectroscopic IR-features confirm that zinc ions undergo complexation by the N-atoms of the backbone of the hb-PEI core of the amphiphilic carrier polymer as schematically illustrated in Figure 1B (for more details see supporting information). As shown in Figure 1C and D, investigations by TEM emphasizes a distinct molecular disperse distribution of zinc ions inside the amphiphilic hb-PEIm carrier polymer. Assuming that zinc-complexation would fail then the unsolved/uncomplexed inorganic zinc precursor would be existent and observable as distinct separated clusters, which is not the case. This is in accordance with the results of the supramolecular structure analysis of the hb-PEIm-Zn2+-hybrid by means of WAXS as shown in Figure 1E. The specific patterns of the polycrystalline zinc precursor completely disappeared. The discrete broad peak at 2 Theta 21.6° in the diffraction patterns of hb-PEIm and hb-PEIm-Zn2+ can be assigned to the long chain crystallization of the attached hydrophobic moieties forming the outer shell of the carrier and hybrid, respectively (cf. scheme in Figure 1B). As can be further seen from Figure 1E, there is a thermal induced transition of the morphology of the hb-PEIm-Zn2+-hybrid. After heating above the melting point (Tm) of hb-PEIm-Zn2+ (hb-PEIm-Zn2+(a.m.)), only an amorphous halo is observable due to inhibition of recrystallization. This change of the structural characteristic of the zinc-hybrid is in harmony with investigations by DSC (see supporting information). Uninfluenced by this effect, the hb-PEIm-Zn2+-hybrid possesses a high thermal stability as affirmed by the DTG-signal in the thermal gravimetrical powder analysis (cf. Table 1). Hence, it can be expected that it withstands the rigid thermal conditions of hot melt processing, which is usually utilized to shape medical devices made of thermoplastic polymers like syringes, tubes and valves or packaging materials.
Characterization of the PA composites with hb-PEIm-Zn2+-hybrid
The hot melt shaped samples (cast films CF and injection molded plates IMP), which contain of up to 5 wt% of the hb-PEIm-Zn2+-hybrid, were characterized by means of TEM, WAXS and DSC-measurements (for detailed data see supporting information). Additionally, the time and concentration dependency of the zinc ion release properties of the zinc-doped samples was analyzed.
From the results of the conducted TEM, WAXS and DSC analyses it can be assumed that the incorporation of the hb-PEIm-Zn2+-hybrid via hot melt extrusion leads to a dispersion of the zinc-hybrid in its amorphous form, which is conform to the already discussed polymorphism of the hb-PEIm-Zn2+-hybrid (cf. Figure 1E). The amorphous nature of the hb-PEIm-Zn2+-hybrid facilitates an almost molecular disperse distribution (solid solution) [17, 18]. Thus, there were no individual particulate conformations in the bulk of the PA matrix observed by TEM investigations. That means the PA molecules interact strongly with the hb-PEIm-Zn2+-hybrids preventing aggregation into discrete particulate clusters enabling a quasi-solution (complete miscibility). Independent of the incorporation of the zinc-hybrid, an alteration of the polymorphic form between the CF and IMP samples occurred (for details see WAXS patterns in supporting information). The differences in matrix morphology can be primarily attributed to a specific alteration in processing (cooling) conditions which determine the crystal packing of the PA matrix molecules. In case of fast cooling (preparation of the CF samples) crystallization is repressed, on the other hand, under high pressure and lower cooling rate (preparation of the IMP samples) crystallization takes place, so that the PA6 can be obtained in its semi-crystalline monoclinic α-form . The (metastable) nearly amorphous constitution of the PA CF composites could be also verified by DSC measurements (see supporting information).
With respect to an antimicrobial functionalization of the material surfaces, a release of zinc ions from the hb-PEIm-Zn2+-hybrids is crucial. Such a relationship between the release of metal ions and the antibacterial efficacy has been shown in depth for polymer composites with silver [4, 15, 36]. Figure 2A,B illustrates the time and concentration dependency of the zinc ion release for the different PA composite shapes (CF and IMP). It is obvious that the zinc ion release is proportional to the content of hb-PEIm-Zn2+ in the PA bulk. That means the release is directly linked to the entrapped zinc ions in hb-PEIm as reservoir. Furthermore, it could be observed that the total amount of zinc ions released increases with immersion time, but the increase is higher in the first 48 h than afterwards. However, an initial abrupt (burst) release does not happen. In this context it is well known that the release behavior depends on the chemical similarity between the drug and the polymer matrix . From the release profiles a high chemical compatibility and a sustained release for extended periods (>14 d) can be expected. Moreover, from the slope of the curves between 48 h and 336 h in Figure 2A,B a release rate, normalized to the sample surface area, of approx. 1.2 and 4.4 μg L–1 cm–2 d–1 can be calculated for the IMP and CF with 5 wt% of hb-PEIm-Zn2+-hybrid, respectively. In order to explain the different release rates it is necessary to consider both sample shape and matrix morphology. With decrease of sample thickness the surface/volume ratio drastically increases. Associated with that a higher amount of hb-PEIm-Zn2+-macromolecules are situated close to the material surface. Besides, the almost amorphous bulk morphology in case of the CF material further enhances the effective release of zinc ions because of a higher water adsorption/diffusion rate which is associated with the crystallinity of the PA6 matrix . A correlation between the release kinetics and the water uptake of PA6 was also documented for antimicrobial silver-doped composites [15, 39]. Usually it will take several days until the bulk of the sample is completely soaked with water. As a consequence, it can be concluded that during short term contact with water particularly the amount of hb-PEIm-Zn2+-hybrids near to the surface region contributes to the zinc ion release.
Assessment of the antimicrobial material properties
As already pointed out and exemplarily displayed by the results in Figure 2A,B, the biological active features of the prepared composites are connected with a release of zinc ions via the surface of the matrix polymer and thus with the depletion of the effect over time. In this context, a material composition with uncontrolled or burst release properties as well as elementary leaching of the antimicrobial agent may lead to a positive result in short term antimicrobial tests (e.g., agar diffusion test), but it would be uneligible for applications with long term antimicrobial and biocompatibility demands. Hence, the antibacterial properties of the PA composites (IMP) were investigated as a function of the concentration of hb-PEIm-Zn2+-hybrid and the alteration due to 336 h of water storage. From Figure 2C it becomes obvious that even the lower release rate of zinc ions in case of the IMP renders the PA composite surfaces antibacterial. The bacteria cells (S. aureus) were eliminated completely within 24 h by contact with the unaged test specimens. Upon encounter with water for 336 h, only in case of the lowest content (1 wt% of hb-PEIm-Zn2+) a significant decline of the antimicrobial activity could be observed which can originate in a depletion of releasable zinc ions. Apart from that the antibacterial effect remains constant which allows expecting a sustained release of bioactive (oligodynamic) zinc ions even for extended periods (>14 d). These results lead to the conclusion that to assure an utmost reduction of the bacteria growth in contact with the material surface for a long time a concentration of about 2.5 wt% or higher of the hb-PEIm-Zn2+-hybrid should be appropriate.
To get further insight into the antimicrobial efficacy, PA composites (CF, IMP) with 5 wt% of the hb-PEIm-Zn2+-hybrid were tested against a set of additional germs of practically relevance including Gram-negative bacteria strains, fungal strains and an alga. The microbiological test results are summarized in Figure 2D. Whereas the unmodified PA composites (CF, IMP references) generally did not show any intrinsic antimicrobial effect (data not shown), the zinc-doped composites exhibited a strong antibacterial activity against Staphylococcus aureus, Klebsiella pneumoniae, and Salmonella typhimurium. A complete inhibition of the bacterial growth was observed for these bacterial strains during a contact time of 24 h. Furthermore, the zinc-doped CF and IMP composite materials also possessed a significant antimycotic effect against Candida albicans and even the growth of the alga Prototheca wickerhamii was significantly reduced. In contrast, under the used experimental conditions none of the zinc-doped materials was active against the Gram-negative bacterium Pseudomonas aeruginosa and spores of the fungus Aspergillus fumigatus. These findings indicate by the way that the antimicrobial effect does not arise from a toxic effect of the dendritic carrier polymer because the terminal amino-groups of neat dendritic polymers are commonly very toxic, particularly for Gram-negative bacteria like Pseudomonas aeruginosa . It can be assumed that the detected activity is specific for the released zinc ions and their oligodynamic action.
However, from the practical point of view, problems of hygiene and infections in the medical field are not solely associated with individual pathogenic germs but often also with the formation of a biofilm on the corresponding surfaces. Against this background the influence of the antimicrobial material modification on surface bioadhesion was explored.
Assessment of the initial bioadhesion and biofilm inhibition material properties
Particular for long term demands, the deposition and settlement of bacteria on material surfaces as well as the growth of biofilms must be taken into account because this frequently leads to serious problems during the application of medical devices [2, 41]. Therefore, the attachment of the bacterial strains Staphylococcus aureus and Staphylococcus epidermidis on the surface of the PA composites (IMP) with 5 wt% of the hb-PEIm-Zn2+-hybrid was studied under simulated physiological conditions. As a model system of practical significance a mixed culture of both bacterial strains was used for the examination as both microorganisms are commonly associated with biofilms on medical devices . As is well known, the bioadhesion, as the initial step in biofilm formation process, is not only influenced by the surface chemistry, but is also strongly dependent on the contact/flow conditions. Considering this aspect, the surface bioadhesion behavior was investigated under static as well as under dynamic contact conditions.
In agreement with the microscopic observations presented in Figure 3A–H a significant difference in initial bacterial surface adhesion after 24 h of incubation can be detected between the neat PA (IMP reference) and the zinc-doped PA composite material under both, static and dynamic conditions of contact. On the unmodified PA surfaces a high bacterial adhesion occurred. Thus, a relatively dense and homogeneous layer of adhered microorganisms is observable under static conditions (Figure 3A,B) and a more clustered structure of adhered microorganisms under the utilized dynamic conditions (Figure 3E,F). The initial adherence was in general remarkably higher under static conditions compared to that under dynamic ones (Figure 3I). In contrast to these observations, bacteria cells attached in general only sporadically to the surface of the zinc-doped PA composites as could be seen in the CLSM (Figure 3C,G) and SEM pictures (Figure 3D,H). This is in accordance with a substantial change in the quantity of adhered bacterial cells as displayed by the data in Figure 3I. The degree of bacterial deposition decreased considerably due to the incorporation of the antimicrobial hb-PEIm-Zn2+-hybrid in the polymeric matrix material. Furthermore, the numerical analysis of the fluorescence microscopic vitality staining highlights an antibacterial surface activity of the PA composites with hb-PEIm-Zn2+-hybrid (Figure 3J). This demonstrates, in accordance with the assessment of the antimicrobial surface properties according to the cover film test method, the efficacy of the hb-PEIm-Zn2+-hybrid as antimicrobial surface modifier even under these more complex and physiological relevant in-vitro test conditions.
Altogether the results validate that the modification of the PA (IMP) with the hb-PEIm-Zn2+-hybrid extensively alters the surface properties which lead to a significant reduction of bacterial cell adhesion. The outcome of the reduction of bacteria attachment in relation to the total number of adhered bacteria and in connection with an antimicrobial surface effect is only slightly different for the static and the dynamic scenario. In particular, the hampered bioadhesion can be considered as a consequence of a changed surface chemistry due to the amphiphilic and/or cationic nature of the hb-PEIm (carrier polymer), similar to the non-covalent forces involved in bioadhesion discussed in the literature [43, 44], and less as a result of the antibacterial surface properties caused by the release of bioactive zinc ions. However, since the initial adhesion of bacteria cells and the subsequent development of a biofilm may be two distinct phenomena , the findings are indicative for a significant effect of the hb-PEIm-Zn2+-hybrid regarding biofilm inhibition properties but cannot be fully extrapolated to biofilm formation under various physiological conditions.
Evaluation of the material biocompatibility
Often materials with biocide properties do not meet the demands to deliver both (i) a broad and long-lasting antimicrobial efficacy as well as (ii) a high biocompatibility in contact with biological interfaces (e.g., skin or tissue cells) at the same time and dosage level. Hence, beside the examination of antimicrobial effectiveness it is also of eminent importance to prove that the material modification has no harmful bioadverse effects, e.g., causing cytotoxic implications. The assessment of the biocompatibility is therefore the key from the perspective of evidence of an antimicrobial specific effect and of a safe use of the corresponding antimicrobial agent/polymeric material in medical device applications. A potential cell toxic effect of the antimicrobial hb-PEIm-Zn2+-hybrid was analyzed by the ATP-bioluminescence assay (DIN EN ISO 10993-5: 2009) utilizing HaCaT cells. The examinations were conducted with PA composites (CF, IMP) with the highest application concentration of 5 wt% hb-PEIm-Zn2+ versus the corresponding unmodified CF and IMP materials as references and the pure growth medium as the negative control (100% standard).
As already shown in Figure 4, the total amount of zinc ions released is depending on material shape and the corresponding alteration of the surface to volume ratio (CF>IMP), which has to be taken into account with respect to the conducted biocompatibility tests utilizing extracts of the IMP and CF samples, respectively. However, as the results in Figure 4A display, the addition of the hb-PEIm-Zn2+-hybrid does not lead to any toxic outcomes. Neither the 24 h nor the 72 h extracts of the zinc-doped polymer composites exhibited a cytotoxic effect at an extraction ratio of ≤20 mg: 1 mL. Even the higher zinc amount released in case of the CF composites (cf. Figure 2B) is below a toxic concentration. In this respect, it can be assumed that particularly the hydrophobic shell of the polymeric carrier hb-PEIm inhibits a simple wash-out (leaching) and an easy extraction of the hb-PEIm-Zn2+-hybrids from the bulk material and thus plays an important role regarding the good biocompatibility which was found. Further, also a potential release of pro-inflammatory cytokines by human HaCaT-keratinocytes was investigated but there was no increase of interleukin 6 (IL6) and 8 (IL8) release detectable in the cell culture supernatant after 24 h or 48 h of incubation with the respective material extracts. Moreover, an enhanced liberation of sulfidoleucotrienes, which are inflammatory mediators and related to allergen stimulation , was not determined in response to incubation of human leukocytes with the 72 h extracts of the PA composites (CF, IMP) with 5 wt% hb-PEIm-Zn2+-hybrid.
In an additional experiment we analyzed the cytotoxic potential of material extracts from PA composites (CF) containing an equal amount of hb-PEIm-Zn2+ or ZnPT utilizing L929 mouse fibroblasts. The biocidal zinc-compound ZnPT is commonly used in anti-dandruff shampoos and antifouling coatings. As can be seen by comparison of the data in Figure 4B, whereas the ZnPT introduces an high cytotoxicity the hb-PEIm-Zn2+-hybrid leads to no cytotoxic effect related to the viability of the L929 cells at an extraction ratio of ≤6 cm2: 1 mL. The comparison clearly indicates a substantial good biocompatibility of the hb-PEIm-Zn2+-hybrid.
The presented in vitro toxicological measurements are predictive of in vivo disease outcomes but cannot be fully extrapolated to a direct contact with human tissue. Nonetheless, a significant elemental leaching and wash-out of hb-PEIm-Zn2+ causing toxic and irritant effects to keratinocytes in the epidermis can be nearly excluded based on the results obtained. These findings are especially impressive for the antimicrobial modified CF composites assayed because of the significantly higher release of zinc ions compared to the IMP material as discussed earlier (cf. Figure 2A,B). That means the hb-PEIm-Zn2+-hybrid is a promising antimicrobial modifier for plastic materials in various medical and particularly dermatological applications.
Evaluation of the material hemocompatibility
The so far presented analytical results show, that the biophysicochemical properties of the developed hb-PEIm-Zn2+-hybrid and the PA composites containing them combine both, a high antimicrobial surface effect and a good biocompatibility. Beyond this, especially with regard to implant and wound care applications, the hemocompatibility of the material (surface) is also a very important issue. Blood compatibility is still a great challenge in the development of biocompatible polymer compositions . It is possible that materials themselves or addition agents may be thrombogenic or generate an immediate material-related inflammatory response. In this context the measurement of thrombin generation during blood coagulation is an important diagnostic tool for hemocompatibility testing . However, only few fundamental mechanisms of hemostatic action of different materials that interact with biological systems are elucidated by now .
To gain a deeper understanding of the material property relationships with regard to hemocompatibility, the impact of the functionalization of the PA bulk material by the antimicrobial hb-PEIm-Zn2+-hybrid on human whole blood was evaluated. In order to give a full picture of the physiologically complex blood-material interactions, CF composites with 5 wt% hb-PEIm-Zn2+ were used to analyze the outcome of a direct blood contact and a 72 h extract of the corresponding IMP material (extraction ratio ≤1 mg: 1 mL) was used to analyze the outcome of an indirect blood contact. As the whole blood clotting results presented in Figure 4C,D show, the biocidal hb-PEIm-Zn2+-hybrid does not provoke a significantly adverse effect on the thrombin generation. As revealed by the comparison of the measurements with CF and IMP samples, the slight increase of the blood coagulation kinetics (reduced lag time) and the total amount of thrombin generated goes mainly back to the direct blood-material surface contact (see also CF samples in Table 2). Further experiments also showed no significant effect of the antimicrobial material modification on prothrombin time (PT) and activated partial thromboplastin time (aPTT), respectively (Table 2).
Moreover, it is well known that the adsorption of plasma proteins on material surfaces can strongly effect the blood clot formation (thrombogenicity) . The tendency toward thrombogenicity of materials can be quantitatively expressed by the blood clotting index (BCI). The data in Figure 4E,F confirm that the incorporation of the antimicrobial hb-PEIm-Zn2+-hybrid (and the zinc-ion release) does not affect the hemocompatibility of the composite materials. The development of the BCI over time for the zinc-doped samples and the neat PA samples is within the same range of experimental accuracy. In accordance with the findings in the whole blood thrombin generation tests, the accelerated clot formation in whole blood compared to the control in the case of the CF samples could be again attributed to a direct material contact and is not an outcome of the modification of the surface chemistry by the hb-PEIm-Zn2+-hybrid. That means the surface of the CF itself triggers a thrombus formation.
Additionally, synthetic materials in contact with blood can also mediate an immediate and complex object-related inflammatory reaction of the immune system. The release of polymorphonuclear (PMN) elastase into the blood plasma from neutrophil granulocytes and/or the activation of the complement system can be an indication for an inflammatory response. Beside the chemistry, material properties inciting in vivo inflammatory potency are also governed by size, surface and morphology . In the conducted measurements, no significant change of the PMN elastase release could be observed neither after incubation of human whole blood with the zinc-doped CF composite samples or IMP material extracts nor after incubation with the zinc-free references (Table 2). In accordance, no activation of the plasma cascade system occurred (Figure 5). The C5 complement convertase activity induced by a direct contact of the PA composites with/without the antimicrobial hb-PEIm-Zn2+-hybrid is much less than the effect which unfunctionalized materials like medical stainless steel (MS) show. Materials like MS, which are commonly used for surgical instruments or orthopedic implants, are known to activate the complement system . Moreover, even materials like LDPE (low-density polyethylene), which are e.g., used for medical tubing applications, and PDMS (polydimethylsiloxane), which is a silicon-based organic polymer e.g., found in breast implants, elicited a significant activation of the C5 complement convertase. All the more impressive is that the addition of the hb-PEIm-Zn2+-hybrid was found to be inactive with regard to C5 complement convertase activation. That means the hb-PEIm-Zn2+-hybrid seems to have a low potential to induce an immune response. Moreover, no hemolytic effects were observed for the same materials (see also supporting information), which indicates once again the good hemocompatibility of the hb-PEIm-Zn2+-hybrid as an antimicrobial addition agent.
A novel antimicrobial organic-inorganic hybrid based on amphiphilic dendritic hyperbranched polyethylenimine with zinc was prepared and successfully used as surface modifier for PA (cast films and injection molded plates). As we could show, the utilized amphiphilic hyperbranched core-shell carrier (hb-PEIm), together with the amorphous nature of the final hb-PEIm-Zn2+-hybrid enables a quasi-solid solution (molecular dispersion) in the bulk of the matrix polymer and that in turn leads to a pronounced surface exposure of the hybrid agent. This seems to be the key for the remarkable biological surface features. Moreover, the eluation tests give evidence for the conclusion, that the effective release of bioactive zinc ions is determined by the amphiphilic core-shell carrier itself, the sample shape and the morphology of the semi-crystalline polymer bulk matrix. As a consequence of the oligodynamic effect of the released zinc ions a formidable antimicrobial activity of the PA composites with the hb-PEIm-Zn2+-hybrid against a broad range of germs of clinical relevance including the bacteria strains Staphylococcus aureus, Salmonella typhimurium and Klebsiella pneumoniae, the yeast Candida albicans as well as the alga Prototheca wickerhamii could be found. The estimated zinc ion release properties indicate that the particular bactericidal efficacy of the hb-PEIm-Zn2+-hybrid containing composites may be kept for long term if an appropriate loading level is applied. Furthermore, the observations in cultured keratinocytes and fibroblasts reflect an overall remarkable good biocompatibility of the hb-PEIm-Zn2+-hybrid and the corresponding polymer composites (CF, IMP), respectively. Additionally, no indications were found in-vitro that the antimicrobial hb-PEIm-Zn2+-hybrid may induce an inflammatory response. It also exhibited no hemolytic properties and did not activate the complement system, neither at a direct nor at an indirect material contact with human blood or plasma in vitro. Thus no evidence is seen that would restrict the use of the hb-PEIm-Zn2+-hybrid as an antimicrobial modifier for medical device applications even of such devices in contact with blood. Besides, under both, static and dynamic conditions of contact, a significant reduced attachment of the bacteria strains Staphylococcus aureus and Staphylococcus epidermidis was observed in the presence of the hb-PEIm-Zn2+-hybrid. Thus, it can be concluded that the zinc-hybrid agent even clearly contributes to a reduction of bacterial surface adhesion. This could counteract the formation of a biofilm.
Finally, it can be stated that the novel hb-PEIm-Zn2+-hybrid is suitable as advanced additive agent for the creation of antimicrobial polymers in a wide range of applications, especially medical devices, which possesses a high antimicrobial efficacy and good cyto-/hemocompatibility as well as and reduced bacterial surface adhesion.
Materials and methods
Preparation of the amphiphilic hb-PEIm-Zn2+-hybrids
Water-free hyperbranched polyethylenimine (hb-PEI) of high molecular weight (25,000 g mol–1, BASF) was used as received and modified to a nominal value of 50 mol-% of the terminal groups with hexadecanoic acid (p.a., Merck) according to the melt reaction procedure described in the literature . The received product (hb-PEIm) was dissolved in trichloromethane (CHCl3, ph.eur., Roth) and the clear yellowish solution was filtered for purification before further use. The zinc loading was carried out by subsequent addition of the zinc precursor (ph.eur., Applichem) to the inverse micelle solution. The slowly dissolution of the zinc-precursor and the self-assembling of the hb-PEIm-Zn2+-hybrids were accomplished at room temperature under stirring for up to 24 h. In this manner, stock solutions of zinc-hybrids with a molar ratio of zinc ions (Zn2+) to nitrogen atoms (N) of the hb-PEIm between 1:16 and 1:2 were prepared. Above the uptake/complexation capacity of the hb-PEIm carrier the solutions were saturated and clouded. These solutions were not applied for further preparations.
Afterwards, CHCl3 was distilled off and the remaining precipitates (slurries) were first freeze-dried and then dried in vacuum until a constant mass was achieved. The received light yellow hb-PEIm-Zn2+ (Zn:N=1:x) solid was finely ground before further use. For the preparation of the PA composites, only the hb-PEIm-Zn2+ (Zn:N=1:6) was used, which was the zinc-hybrid with the highest stable zinc loading concentration.
Preparation of the zinc-doped composites, cast films and injection molding plates
The used PA (PA6, Ravamid R200S, Ravago) has a melting point of 220°C and a moisture absorption capacity between 2.8 wt% (saturation at humid air) and approx. 10 wt% (saturation at water storage) in its semi-crystalline monoclinic α-form. Firstly a masterbatch resin (MB) containing 10 wt% of the hb-PEIm-Zn2+-hybrid was prepared by hot melt extrusion of the thoroughly pre-dried components utilizing a synchronous twin screw extruder (ZSK-25, Coperion, Werner&Pfeiderer, through-put: 10 kg/h, barrel heat zone temperature Tmax: 240°C). In a second step, the dehumidified MB was diluted with pure PA to vary the zinc content. Therefore, the MB was extruded again (synchronous twin screw extruder, Rheomex PTW 16/25, through-put: 1 kg/h, barrel heat zone temperature Tmax: 220°C) and the resulting pre-dried resin was shaped into a cast film (CF) (Randcastle single screw extrusion system ½”, barrel heat zone temperature Tmax: 220°C, cooling role Tchill: 20°C). Otherwise the pre-dried MB was mixed with an appropriate amount of the pure PA resin and charged in the plasticizer unit in an injection molding machine (Battenfeld HM 110/350 H/130 V, barrel heat zone temperature Tmax: 220°C mold temperature: 80°C) for shaping plates (IMPs). In this manner CF samples (thickness of approx. 100 μm) and IMP samples (60 mm×60 mm×2 mm) containing 1.0, 2.5 or 5.0 wt% of the zinc-hybrid were prepared.
Without preparing a MB first, but otherwise in the same manner CF samples with zinc pyrithione powder (ZnPT, Janssen Pharmaceutica N.V.) were prepared.
Morphological and physical characterization of the hb-PEIm-Zn2+-hybrid and the composite materials
Characterizing of the molecular structure of the hb-PEIm-carrier polymer and the hb-PEIm-Zn2+-hybrids was done by infrared absorption spectroscopy (FT-IR) using the KBr pellet technique (Digilab FTS 2000), transmission electron microscopy (TEM JEM 2010), thermogravimetric analyses using a heating rate of 10 K/min under nitrogen atmosphere (TGA/SDTA 8051, Mettler-Toledo), differential scanning calorimetry (DSC) using a heating rate of 5 K/min or 10 K/min under nitrogen atmosphere (DSC 204, Netzsch) and wide angle X-ray scattering (WAXS) using transmission mode (D8 Advance, Bruker AXS, CuKα with λ=1.54184 Å). The amount of zinc of the hb-PEIm-Zn2+-hybrids and the zinc concentrations of aqueous elutes (pH 5, acetic acid) made by immersion of the zinc-doped sample forms (CF, IMP) for a certain period of time were determined by atomic absorption spectroscopy (AAS, Analyst Perkin-Elmer800). All elution experiments were performed in triplicate.
Assessment of antimicrobial efficacy
The antibacterial properties of the samples were monitored using the cover film test method according to the DIN EN ISO 22196: 2007 utilizing the bacterial strains Staphylococcus aureus ATCC 6538, Klebsiella pneumoniae ATCC 4352, Salmonella typhimurium Ta 100, and Pseudomonas aeruginosa DSM 1117. The fungicidal and algicidal properties were analyzed using the alga Prototheca wickerhamii DSM 10639, the yeast Candida albicans DSM 1386 and the fungi Aspergillus fumigatus DSM. All experimental germs were pre-incubated in Caso bouillon and Sabouraud bouillon, respectively, for 18 h at 37°C and diluted with 1/20 nutrient broth to a working solution of 2.5–10×105 cfu/mL. Steam sterilized (121°C/15 min) samples were incubated for 24 h at 37°C in a wet chamber. Thereafter the surviving germs in the suspension were determined by plate count technique. The antimicrobial efficacy of the sample was calculated as the cfu (colony forming units) reduction in [Δlog] relative to the internal control (the neat cover film material). For a better comparison the Δlog values were converted into the reduction of microbial growth in [%]. The results presented in this paper are the mean values of at least three independent determinations.
Monitoring of bacterial adhesion and initial biofilm formation
Specimens (IMP) with 5 wt% of hb-PEIm-Zn2+ were utilized for studying bioadhesion behavior. The samples were thoroughly cleaned by rinsing with Millipore-Q water and steam sterilized prior to use. As test organism the bacterial strains Staphylococcus aureus ATCC 12600 and Staphylococcus epidermidis PCM 2479 were used. The germs were pre-incubated in Tryptic Soy Broth (TSB: 17.0 g/L Casein, 3.0 g/L Soybean Meal, 5.0 g/L NaCl, 2.5 g/L KH2PO4, 2.5 g/L Dextrose) for 18 h at 37°C under gentle stirring. Thereafter the suspensions were centrifuged and the bacteria were collected and washed with PBS followed by the next centrifugation step. Finally, the bacteria were resuspended in equal parts in growth medium (5 g/L pepton, 3 g/L meat extract, Fluka) to a working concentration of 108 cfu/mL.
To investigate the bioadhesion behavior under static conditions the samples were incubated with the bacterial suspension in a 24-well culture plate for 24 h at 37°C. Thereafter the samples were removed and carefully washed with PBS to remove non-adhered bacteria cells before microscopic image analysis was conducted.
To investigate the bioadhesion behavior under dynamic conditions the material specimens were incubated with the bacterial suspension for 24 h at 37°C in a laboratory bioreactor coupled with a flow through chamber system as previously described in detail . The bacterial suspension was circulating in the system with a volumetric flow rate of 3 mL/min (streaming rate in the center of the chamber channel: 4.4 mm/s). Subsequently, a rinsing cycle with PBS was performed to remove non-adhered bacteria cells from the surface of the sample material prior to image analyses.
For image analyses of the surface adhesion of bacteria cells confocal laser scanning microscopy (CLSM, Leica) and scanning electron microscopy (SEM, Zeiss) were used. Stacks were recorded with z sections of 0.8 μm (excitation: 488 nm, emission: 500–530 nm and 620–700 nm) and the CLSM data were processed with the Volocity software package (V5.1). Furthermore, fluorescence microscopic evaluation and cell viability assay were applied to quantify dead and live bacteria cells (BacLight™ – Live/Dead® staining kit, Molecular Probes).
To quantify the total number of bacteria cells attached to the sample surface after incubation an additional wash step with PBS in an ultrasonic bath was conducted to achieve a complete removal of the adhered bacteria cells. The complete detachment was checked by CLSM analysis.
The results presented in this paper are the mean values of at least three independent determinations (test series with separately cultured strains).
Assessment of biocompatibility
To determine potential cytotoxic effects of the PA composites with hb-PEIm-Zn2+ in-vitro measurements according to the DIN EN ISO 10993-5: 2009 were conducted. For this purpose, extracts of specimens (CF and IMP with 5 wt% of hb-PEIm-Zn2+) were prepared by incubating pieces of the materials with Dulbecco’s modified Eagle’s medium (DMEM; PromoCell, Germany) at 37°C for 24 h and 72 h (extraction ratio 20 mg: 1 mL), respectively.
Human HaCaT-keratinocytes (kindly provided by Prof. Dr. N.E. Fusenig, German Cancer Research Center, Heidelberg) were cultured in DMEM, supplemented with 10% fetal calf serum (PromoCell, Germany) and 1% antibiotic-antimycotic solution (Gibco BRL, Germany) at 37°C in a humidified atmosphere, containing 5% CO2. L929 mouse fibroblasts were used in the comparison experiments with extracts of PA composites (CF) containing hb-PEIm-Zn2+ and ZnPT (extraction ratio 6 cm2: 1 mL), respectively.
For experiments, the cells were detached with trypsin-EDTA solution (Gibco BRL, Germany), seeded in 96-well cell culture plates (ca. 4 104 cells/cm2) and incubated with the sample extracts in different concentrations for 24 h or 48 h at 37°C and 5% CO2. Thereafter the determination of cell proliferation was carried out on the basis of a luminometric ATP assay (ATPlite, Perkin Elmer Inc., MA, USA) by measurement of the bioluminescence using a microplate luminometer (LUMIstar Galaxy, BMG Labtech, Germany). The cellular ATP content was then determined on the basis of a standard curve. The proliferation of the cells in the pure growth medium served as the (negative) control.
To predict the pro-inflammatory potential cell culture supernatants from the cytotoxicity assays were collected and the content of human interleukins IL-6 and IL-8 were quantified by use of an enzyme immunoassay (eBioscience, Germany) as previously described in detail .
The production of sulfidoleucotrienes in vitro was determined using a cellular antigen stimulation test as described elsewhere . Release of sulfidoleucotrienes by human leucocytes was determined by specific ELISA (CAST®-2000 ELISA, Bühlmann Laboratories AG, Switzerland) utilizing a microplate photometer FLUOstar Galaxy (BMG Lab Technologies GmbH, Germany).
Assessment of hemocompatibility
To investigate possible adverse effects by an indirect material contact sample extracts (IMP with 5 wt% of hb-PEIm-Zn2+) were prepared by incubating pieces of the material (5 mm×5 mm) with physiological sodium chloride solution (0.9% NaCl; Fresenius Kabi Deutschland GmbH) at 37°C for 24 h and 72 h, respectively (extraction ratio 20 mg: 1 mL).
To investigate possible adverse effects by a direct material contact samples from the CF material (5 wt% of hb-PEIm-Zn2+) were cut out using punch biopsies with different diameters.
For all experiments, human whole blood was obtained from healthy, unmedicated human volunteers and anticoagulated with sodium citrate (0.1 mL/mL blood). Human normal plasma (Coagulation Control N, Haemochrom Diagnostica, Germany) was employed in the hemostasis assays.
For all measurements of fluorescence intensity/optical density (OD) in the hemocompatibility assays a SPECTROstar Omega (BMG Labtech GmbH, Germany) was used.
Real time measurement of thrombin generation in plasma were run according to the Thrombin Generation Assay (TGA, Haemochrom Diagnostica, Germany) with CF samples (d=4 mm) and IMP sample extracts (extraction ratios assayed: 0.01, 0.1 and 1.0 mg: 1 mL). Empty wells served as control. After incubation with 80 μL human normal plasma the coagulation cascade was started by addition of 100 μL substrate solution, which contained CaCl2. The changes in fluorescence intensity (ex/em: 360 nm/460 nm) related to the formation of thrombin was measured over time (thrombogram). The thrombin concentration was then determined on the basis of a standard curve.
Prothrombin time (PT) and activated partial thromboplastin time (aPTT) were measured by mechanical endpoint determination. Therefore, CF samples (d=8 mm) and IMP sample extracts (extraction ratios: 0.01, 0.1 and 1 mg: 1 mL) were incubated with 200 μL human normal plasma in a cuvette in the coagulation analyzer MC1 (Greiner Biochemica GmbH, Germany) for 2 min at 37°C. In the case of PT measurements, coagulation was started by addition of 400 μL high-sensitivity reagent (MCQuick-Liquid) consisting of tissue thromboplastin and CaCl2 (0.02 M). In the case of aPTT determination, 200 μL aPTT-reagent (Greiner Biochemica GmbH, Flacht, Germany) were added before incubation. Coagulation was then started by addition of 200 μL CaCl2 (0.02 M) and aPTT was measured as described in the literature .
For validation of the thrombogenic potential, CF samples (d=6 mm) and IMP sample extracts (extraction ratio: 1 mg: 1 mL) were analyzed in direct contact with human whole blood in a thrombogenicity assay. Therefore, the samples were equilibrated in test tubes at 37°C. Human citrate blood, either recalcified with 0.2 M CaCl2 (at a ratio of 10:1) or mixed with 0.9% NaCl solution (non-activated blood), was used. Subsequently, the corresponding samples were incubated with 60 μL of the activated or non-activated blood at 37°C for 10, 20, 30, 40, 50, or 60 min. Thereafter the free erythrocytes (not captured in a fibrin clot) were lysed with 1.8 mL dest. H2O for 10 min at 37°C. After centrifugation at 2500×g for 2 min, the amount of released (free) hemoglobin in the supernatant was measured at 540 nm. and the blood clotting index (BCI) was calculated as described in the literature .
To specify the influence on the release of PMN-elastase from neutrophil granulocytes, CF samples and IMP sample extracts (extraction ratio: 1 mg: 1 mL) were equilibrated at 37°C with human whole blood. Empty test tubes were used as negative controls and 0.1% Triton X-100 was defined as positive control. The corresponding samples were incubated with whole human citrate-blood (60 μL) for 5 and 30 min., respectively. Thereafter the plasma was obtained by centrifugation at 2.000×g for 2 min and the amount of elastase in the plasma was determined using a PMN-elastase kit (Milenia Biotec GmbH, Germany). The specific OD was measured at 450 nm with a reference measurement at 620 nm and the PMN-elastase content was evaluated on the basis of a 4-parameter-fit with lin-log coordinates for OD versus concentration.
For predicting the potential of the antimicrobial material modification to activate the complement system, the C5 complement convertase activity was measured using the Complement Convertase Assay (CCA, HaemoScan, Netherlands). Therefore, CF samples (0.5 cm×1 cm) were incubated with 650 μL human plasma at room temperature. Thereafter the samples were washed with 1.6 mL buffer to remove unbound complement proteins. Subsequently, the samples were incubated with 500 μL substrate solution for 24 h. 225 μL of the supernatant were transferred to clear 96-well-plates and OD was measured at 405 nm. The activity of the complement convertase was expressed as the specific OD of the supernatant after 24 h of incubation normalized to the surface area of the material specimens.
For evaluation of hemolytic effects, erythrocytes were incubated with sample material. Therefore, citrate whole blood was centrifuged at 2.500×g for 8 min. The cell pellet was washed three times with PBS before the erythrocytes were resuspended in 11 mL PBS. Afterwards 100 μL of the erythrocyte suspension were incubated with the CF samples (d=8 mm) or the IMP sample extracts (extraction ratio: 1 mg: 1 mL) in 1 mL PBS for 3 h at 37°C on a shaker (200 rpm). For complete lysis (positive control) 100 μL of the erythrocytes suspension were incubated with 1 mL dest. H2O. 100 μL of an erythrocyte suspension incubated in 1 mL PBS served as negative control. Supernatants were collected by centrifugation at 16,000×g for 15 min. 225 μL of the supernatant were transferred to clear 96-well-plates and the specific OD at 540 nm was measured. The hemolysis was calculated according to Equation 1.
All data presented in this paper are mean values of at least three determinations. One-way analysis of variance was carried out to determine statistical significances (Microsoft® Excel 2000). Differences were considered statistically significant at a level of p<0.05.
The authors would like to thank the Federal Ministry of Economics and Technology for the financial support of this work (BMWi, project no. MF100065).
Wintermantel E, Ha SW. Medizintechnik mit biokompatiblen Werkstoffen und Verfahren, 3rd ed. Berlin Heidelberg New York: Springer, 2002:150.Google Scholar
Katsikogianni M, Missirlis YF. Concise review of mechanisms of bacterial adhesion to biomaterials and of techniques used in estimating bacteria-material interactions. Eur Cells Mater 2004;8:37–57.Google Scholar
Salz D, Mahltig B, Baalmann A, Wark M, Jaeger N. Metal clusters in plasma polymer matrices, Part III: optical properties and redox behaviour of Cu clusters. Phys Chem Chem Phys 2000;2:3105–10.CrossrefGoogle Scholar
Lansdown AB, Sampson B, Laupattarakasem P, Vuttivirojana A. Silver aids healing in the sterile skin wound: experimental studies in the laboratory rat. Br J Dermatol 1997;157:728–35.CrossrefGoogle Scholar
von Nägeli C. On the oligodynamic phenomenon in living cells. Denkschriften der Schweiz Naturforsch Ges 1983;33:174–82.Google Scholar
Karuna Sagar I, Nagesha CN, Gundu Rao P. Effectiveness of zinc sulphadiazine, zinc sulphate and silver sulphadiazine in the topical therapy of pseudomonas infection in burns. Burns 1978;6:131–3.Google Scholar
Atmaca S, Gül K, Çiçek R. The effect of zinc on microbial growth. Turk J Med Sci 1998;28:595–7.Google Scholar
Varkey AJ. Antibacterial properties of some metals and alloys in combating coliforms in contaminated water. Sci Res Essays 2010;5:3834–9.Google Scholar
Dal Lago V, de Oliveira LF, de Almeida Gonçalves K, Kobarg J, Borba Cardoso M. Size-selective silver nanoparticles: future of biomedical devices with enhanced bactericidal properties. J Mater Chem 2011;21:12267–73.CrossrefGoogle Scholar
Ravishankar Rai V, Jamuna Bai A. Nanoparticles and their potential application as antimicrobials. In: Méndez-Vilas A. Editor. Science against microbial pathogens: communicating current research and technological advances, Badajoz: Formatex 2011:197–209.Google Scholar
Amin RM. Nanotechnology in controlling infectious disease. In: Hunter RJ, Preedy VR, editors. Nanomedicine in health and disease, St. Helier Jersey British Channel: Science Publishers Islands, 2011:167–83.Google Scholar
Allahverdiyev AM, Abamor ES, Bagirova M, Rafailovich M. Antimicrobial effects of TiO2 and Ag2O nanoparticles against drug-resistant bacteria and leishmania parasites. Future Microbiol 2011;6:933–40.CrossrefPubMedGoogle Scholar
Zhang H, Wang D, Butler R, Campbell NL, Long J, Tan B, et al. Formation and enhanced biocidal activity of water-dispersable organic nanoparticles. Nat Nanotechnol 2008;3:506–11.PubMedCrossrefGoogle Scholar
Koltzenburg S. Formulation of problem drugs – and they are all problem drugs. In: Reintjes T, Editor. Solubilizer Compendium – Solubility Enhancement with BASF Pharma Polymers, Lampertheim: BASF, 2011:9–26.Google Scholar
Haag R, Krämer M, Stumbé JF, Krause S, Komp A, Prokhorova S. Dendritic polymers as multifunctional supports and nanocarriers for drugs. Polym Prepr 2002;43:328–32.Google Scholar
Aymonier C, Schlotterbeck U, Antonietti L, Zacharias P, Thomann R, Tiller JC, et al. Hybrids of silver nanoparticles with amphiphilic hyperbranched macromolecules exhibiting antimicrobial properties. Chem Commun 2002;24:3018–9.CrossrefGoogle Scholar
Monticelli O, Russo S, Campagna R, Voit B. Preparation and characterization of blends based on polyamide 6 and hyperbranched aramids as palladium nanoparticle supports. Polymer 2005;46:3597–606.CrossrefGoogle Scholar
Kannan RM, Perumal OP, Kannan S. Dendrimers and hyperbranched polymers for drug delivery. In: Labhasetwar V, Leslie-Pelecky DL, editors. Biomedical applications of nanotechnology. Hoboken, NJ, USA: John Wiley & Sons, Inc., 2007:105–29.Google Scholar
Li Y, Cui L, Li Q, Jia L, Xu Y, Fang Q, et al. Novel symmetric amphiphilic dendritic poly(L-lysine)-b-poly(L-lactide)-b-dendritic poly(L-lysine) with high plasmid DNA binding affinity as a biodegradable gene carrier. Biomacromolecules 2007;8:1409–16.CrossrefGoogle Scholar
Antonietti L, Aymonier C, Schlotterbeck U, Garamus VM, Maksimova T, Richtering W, et al. Core-shell-structured highly branched poly(ethylenimine amide)s: synthesis and structure. Macromolecules 2005;38:5914–20.CrossrefGoogle Scholar
Vrignaud S, Anton N, Gayet P, Benoit JP, Saulnier P. Reverse micelle-loaded lipid nanocarriers: a novel drug delivery system for the sustained release of doxorubicin hydrochloride. Eur J Pharm Biopharm 2011;79:197–204.CrossrefPubMedGoogle Scholar
Calabretta MK, Kumar A, McDermott AM, Cai C. Antibacterial activities of poly(amidoamine) dendrimers terminated with amino and poly(ethylene glycol) groups. Biomacromolecules 2007;8:1807–11.PubMedCrossrefGoogle Scholar
Cramton S, Gerke C, Schnell N, Nichols W, Gotz F. The intercellular adhesion (ica) locus is present in Staphylococcus aureus and is required for biofilm formation. Infect Immun 1999;67: 5427–33.PubMedGoogle Scholar
Wollina U, Hipler UC, Seeling H, Oelschläger H. Investigations of interactions of chlormezanone racemate and its enantiomers on human keratinocytes and human leucocytes in vitro. Skin Pharmacol Physiol 2005;18:132–8.PubMedCrossrefGoogle Scholar
Nowak G, Lange U, Wiesenburg A, Bucha E. Measurement of maximum thrombin generation capacity in blood and plasma using the thrombin generation assay (THROGA). Semin Thromb Hemost 2007;33:508–14.CrossrefPubMedGoogle Scholar
Raghavachari M, Tsai H, Kottke-Marchant K, Marchant RE. Surface dependent structures of von Willebrand factor observed by AFM under aqueous conditions. Colloids Surf B: Biointerfaces 2000;19:315–24.CrossrefGoogle Scholar
Sokolov A, Hellerud BC, Johannessen EA, Mollnes TE. Inflammatory response induced by candidate biomaterials of an implantable microfabricated sensor. J Biomed Mater Res A 2012;100:1142–50.CrossrefPubMedGoogle Scholar
Frant M, Stenstad P, Johnsen H, Doelling K, Rothe U, Schmid R, et al. Anti-infective surfaces based on tetraether lipids for peritoneal dialysis catheter systems. Mat-wiss u Werkstofftech 2006;37:538–45.Google Scholar
Shih MF, Shau MD, Changa MJ, Chiou SK, Changa JK, Cherng JY. Platelet adsorption and hemolytic properties of liquid crystal/composite polymers. Int J Pharm 2006;327:117–25.PubMedCrossrefGoogle Scholar
The online version of this article (DOI: 10.1515/bnm-2014-0009) offers supplementary material, available to authorized users.
About the article
Published Online: 2014-10-08
Published in Print: 2014-09-01