Collagenase@magnetite: proteolytic composite for magnetically targeted minimally invasive surgery

Abstract

Magnetic proteolytic composites attract great scientific attention as a potential controllable catalytic system for medical applications. The use in medical practice opens new horizons for targeted tissue lysis, while providing prospects for minimally invasive surgery and treatment of pathologies, such as adhesive disease. This article reports the first production of a fully biocompatible magnetically targeted composite based on the collagenase G enzyme entrapped within a sol–gel magnetite matrix. The composite material consists entirely of components approved by the FDA for parenteral administration. The entrapped enzyme demonstrates outstanding thermal stability: the denaturation temperature is higher by 18 °C and the activity is retained even at a temperature of 65 °C. The enzymatic activity of the composites is studied in details on different biosubstrates, such as collagen, gelatin, and collagen tissue samples of animal origin during in vitro tests. Also, the possibility of magnetic targeting is shown. The nanocomposite behaves itself as an effective proteolytic system with prolonged action and could be magnetically separated and reused without activity loss at least 4 times. Due to magnetic targeting, the efficiency of proteolysis could be enhanced up to 20 times.

---

Introduction

Proteolytic enzymes are widely used in medicine. They are used to treat cancer tumors,^1–3 find their application in cosmetology,^4,5 but also as a rehabilitation means after surgery.^6,7 Collagenase is one of the most promising enzymes in this regard due to its narrow specificity in the cleavage of the peptide bonds between amino acids in the molecules of collagen, the main structural protein in the human body. Collagenase is widely applied in medicine for the treatment of Dupuytren's contracture,^8–11 cleaning Schwann cells,¹² as well as for isolating islet pancreatic cells or in forensic-medical examination and biotechnology for homogenizing tissues.^13–16 However, similarly to the other enzyme species upon oral administration, collagenase is quickly degraded in the gastrointestinal tract,¹⁷ whereas parenteral administration of collagenase rapidly results in its inactivation by the immune system and its absorption by the reticuloendothelial cells.¹⁸ Moreover, the lack of a targeting property causes a number of side effects, such as non-selective cleavage of the extracellular matrix, leading to inflammatory reactions and tissue damage.^19,20

One of the most promising methods for increasing the selectivity and stability of enzymes is the production of magnetically targetable composite materials. There are number of studies dealing with different approaches in designing various magnetic proteolytic systems, but basically their composition is the same, consisting of three main components: a proteolytic enzyme, a magnetic core and a binding agent, usually a sol–gel silica matrix^21–23 or an organic polymer.^24,25 Such composite systems with proteolytic enzymes immobilized on the magnetic nanoparticles demonstrate high enzymatic activity and are essentially promising for industrial applications,²⁶ e.g., in agriculture for water purification from nematodes,²⁷ in leather processing and currying.²⁸ Attempts to produce therapeutic magnetically targeted proteolytic systems however are faced with problems mainly associated with the strict regulations for substances approved for parenteral use, since the composition of such systems often include materials and components which do not have an approval for subcutaneous or parenteral administration²⁹ by the leading health organizations, such as FDA and EMA. Thus, the problem of creating magnetically targeted proteolytic systems with real therapeutic prospects still persists.

A solution to this problem would be the usage of a so-called entrapment procedure, where the enzyme molecules are being trapped inside a porous inorganic matrix in the course of an irreversible sol–gel transition. In this way, the total number of the used components is reduced to only two: a therapeutic enzyme and a suitable magnetic inorganic oxide matrix made of maghemite or magnetite nanoparticles which are the only magnetic nanoparticles with an appropriate legal status, being used in medicine as a MRI contrast agents.³⁰

In our previous investigations we have demonstrated the possibility of creation of such composite materials via direct entrapment of enzymes into magnetite matrixes employing a sol–gel transition process.^31–33 In this way, several model enzymes were entrapped into the nanoporous matrixes as a result of a controlled polycondensation of magnetite nanoparticles sol with corresponding enzyme molecules. This procedure resulted in an enzyme@ferria nanocomposites with uniformly distributed enzymes over the matrix. The entrapped enzymes are physically trapped in the nanopores of the material and unable to “escape” the composite structure. However, due to the mass transfer, such entrapped enzymes are able to promote enzymatic reactions while remaining within the matrix.³² Potential substrates for entrapped enzymes can be both small molecules or large macromolecules, such as proteins like a plasminogen,³³ giving an opportunity to use such composite material in order to produce magnetically-targetable therapeutic enzymatic systems. Moreover, it was shown that enzymes in such nanocomposites are protected from aggressive conditions and inactivating agents by a nanocage and appear to be more stable than free one, showing a prolonged profile of action.^32,33

In this article we apply the explained approach of nanocomposite enzyme@ferria synthesis³² in the development of more specific biocompatible proteolytic nanocomposites (ColG@ferria), based on a collagenase G enzyme (ColG) and a magnetite sol–gel matrix. We show that a product is a composite with a prolonged profile of action and a possibility for magnetic targeting. In the following sections we describe the synthetic procedures, the composition and the structure of the obtained material, together with the analysis of the data on its stability at elevated temperatures and examine its activity towards a number of collagen-containing substrates, biological objects and animal tissues. Based on experimental observations and in silico calculations, possible organization and structure of the composite together with the mechanism of action are proposed. As it will be described below, the proteolytic activity of the composite is present even when enzyme release is absent, offering a possibility for creation of proteolytic composites with prolonged effect suitable for therapeutic application.

Results and discussion

Synthesis and characterization of collagenase@magnetite composites

As it was mentioned in the Introduction section, one of the main issues in the creation of magnetic therapeutic nanocomposites is to use magnetic nanomaterials approved for intravenous injection. At present those are: maghemite and magnetite. The latter one has a stronger magnetic moment, thus it is preferable for magnetic targeting. In order to produce magnetic proteolytic composites, enzyme molecules are coupled with the magnetic component in a variety of ways, but almost all of them implement additional coupling agents such as organic linkers, polymers or silica coatings.^21–25 This however results in a complicated synthetic procedures and also interfere with the legal status of the resulting composites.

To overcome this problem but also to ensure injectability of the obtained materials, the nanocomposite presented in this work was obtained (for details see Experimental part) by the sol–gel co-condensation of the enzyme collagenase G (ColG) and a magnetite sol in a manner described in our earlier work.³² The result is a two-component composite made only of ColG and magnetite nanoparticles (ColG@ferria). The formation mechanism of proteolytic composites is briefly described as follows. In the first step the positively charged magnetite nanoparticles (+30 mV at neutral pH level – for details see ESI, Fig. 1S†) electrostatically interacts with negatively charged collagenase molecules (IP of collagenase is at pH = 5.9).³⁴ After drying the resulting mixture, a rigid magnetite xerogel matrix with entrapped enzyme molecules is formed due to irreversible sol–gel transition (Fig. 1a–c). The formed matrix possesses well-developed porous nanoarchitecture with a high specific surface area (120 m² g⁻¹ for a magnetite matrix and 115 m² g⁻¹ for a 10% wt collagenase@magnetete composite see ESI, Fig. 2S†) and an average pore diameter of 8 nm. The matrix is composed of highly crystalline magnetite nanoparticles with narrow particle size distribution and an average diameter of 10 nm (Fig. 1b and c) and a crystal phase of magnetite, according to XRD (Fig. 1c). After condensation, the obtained composite material may be easily mechanically milled in a mortar, dispersed in saline solution and filtered through a suitable filter with pore dimensions meeting the criteria for the maximum particle size suitable for parenteral injection (Fig. 2a–c).^35,36 The resulting particles have narrow size distribution with major fraction of 1 μm (Fig. 2b) and remaining stable at least for an hour when dispersed in saline, solution. In human plasma this stability is higher, reaching at least 24 hours due to higher viscosity and the presence of albumin proteins which additionally stabilize the ColG@ferria composite. The stability could be further increased by optimizing the structure and the composition for more specific tasks, but for this conceptual work this was enough. Composite nanoparticles possess developed external structure (Fig. 2c) characterized by a high level of porosity, which is an important factor for the preservation of the enzymatic activity of the enzyme in the entrapped state. An EDX analysis (Fig. 2d and e) show uniform distribution of signals originating from iron (constituting the magnetite nanoparticles – Fig. 2d) and nitrogen (originating from the ColG molecules – Fig. 2e), indicating a homogeneous distribution of the enzyme in the nanoparticle.

Fig. 1 Characterisation of the sol–gel magnetite matrix. HR-SEM image of the magnetite matrix (a) HR-TEM image of the magnetite matrix, crystal lattice presented in the insert. (b), XRD spectra of the magnetite matrix (compared to JCPDS file no. 19-0629: red lines) (c).

Fig. 2 Characterization of ColG@ferria composite nanoparticles. SEM image of ColG@ferria composite nanoparticles after mechanical milling and filtration (a); hydrodynamic diameter distribution of the produced ColG@ferria NPs (b); SEM image of ColG@ferria aggregate (c) and element distribution by EDX of iron (d) and nitrogen (e).

Depending on the mass fraction of the ColG in the ColG@ferria composite, the enzyme entrapment may be either full or there may be a partial release of the ColG from the composite. In order to measure enzyme release, adsorption spectra of the supernatant solution were measured over a period of time for the amount of a free ColG to be determined. As it can be seen from the release studies (cf. Fig. 3a), when the mass fraction of the entrapped ColG is below 10% wt (corresponding to 90 CDU mg⁻¹), only 2% of total ColG quantity is detached from the composite in the course of 12 hours. According to the curve presented in Fig. 3a, these 2% of molecules could be referred to the residual non-entrapped ColG, which is only weakly interacting with the composite surface. At higher mass fractions of ColG, a major release of the enzyme is observed (Fig. 3a). For 12.5% wt ColG@ferria composite, a release of up to 22% of the total content is observed, which corresponds to a residual content of ColG in the composite material at a level of 9.5% wt. This could be due to the electrostatic repulsion of ColG molecules and the composite matrix (see Fig. 3b). As can be observed, the alternations of the enzyme mass fraction in the composite changes the zeta potential of the composite nanoparticles (by sign and value) and also results in an alternation of the isoelectric point with pH (Fig. 3c). The charge of the ColG@ferria composite particle becomes similar to the free ColG molecule, for mass fractions of the enzyme above 5%. However, in the range between 5 and 10% the electrostatic repulsion of ColG molecules located at the outer layers is smaller than the attracting hydrophobic and van der Waals interactions of the enzyme with the matrix walls, resulting in a stability and neglecting leach. Nevertheless, as the percentage rises above 10%, the enzyme molecules are repelled from the surface of the composite nanoparticles, resulting in a sustainable leach of the least blocked enzyme molecules from the composite. Based on this results it could be concluded that in order to create a long-lasting magnetic proteolytic agent, 10% wt ColG@ferria should be used, as its composition remains stable over long period of time. Thus, a composite with such a composition (washed thoroughly to ensure removal of a free enzyme) was used in our further experiments. Because ColG molecules are at fully entrapped state, additional benefits of the composite, such as thermal stability (already reported in our previous papers^32,37 for other systems), were investigated and are discussed in the next section.

Fig. 3 Characterisation of ColG@ferria composites with different mass fraction of ColG. Release profile of collagenase from the collagenase@ferria composite as a function of mass fraction of enzyme in the composite (a), zeta potential of composite nanoparticles as a function of mass fraction of enzyme in the composite (b), zeta potential of 10% col@ferria composite as a function of pH (c).

Thermal stability of ColG@ferria

A distinctive feature of enzyme entrapped into a xerogel matrix is the template nano-organization of produced composites with respect to enzymes used, resulting in a particular stabilization of protein molecules.^32,33,37 This phenomena appears for the entrapped enzymes only and marks the distinction from a simple adsorption or other types of enzyme immobilization. To evaluate the stability of the entrapped ColG, the denaturation temperature measurements were conducted using DSC (Differential Scanning Calorimetry). The relative enzyme activity of free and entrapped ColG was evaluated at various temperatures, but also after incubation at elevated temperature for different periods of time.

According to DSC curves (Fig. 4a), denaturation temperature for the entrapped ColG is increased by 18 °C, compared to 61 °C for free collagenase, and amounts to 79 °C. Such high temperature shift can be explained by the tight interaction between the nanocage walls and the enzyme molecules, which prevents them to undergo a conformational change and protects them from further denaturation. An increase in denaturation temperature allows carrying out enzymatic reactions under conditions unachievable for a free enzyme and significantly mitigating the product storage conditions. The thermal stabilization of entrapped ColG was tested by analyzing ColG activity at various temperatures (Fig. 4b), using a low molecular weight substrate (see Experimental sections for details). The curves of the relative enzymatic activity reveal that the maximum activity for a free ColG is observed at temperature of 40 °C and further increase of the temperature up to 65 °C leads to decrease of the activity to zero due to a complete denaturation of the enzyme, which is also accompanied by its coagulation. Completely different thermal stability picture is observed for ColG@ferria composite nanoparticles. For composite nanoparticles, the maximum of the activity is shifted by 10 °C to higher temperatures (compared to the free enzyme) and amounts to 50 °C, so that at the temperature of free enzyme denaturation, the relative activity of the composite material remains at 51% of its maximum value. The thermal stability of ColG@ferria composite is also confirmed by testing its activity after an incubation for a varying period of time at 50 °C (Fig. 4c). Comparison of the activity data indicates that the incubation of free collagenase for 4 hours reduces its relative activity to 11% of its original value. On the other hand, the relative activity of 10% wt ColG@ferria composite, under similar conditions, is almost 6 times higher and amounts to 65% of its initial value. This outstanding thermal stability of the entrapped enzyme confirms a high degree of protection against aggressive factors and high stability of the composite as a whole. Enzymatic activity described in this section relates to a low-molecular chromogenic substrate. In order to get closer to the real processes however, further investigations on more complex systems were required. These results will be described in the next section.

Fig. 4 Thermal stability of the entrapped ColG. DSC curves of ColG@ferria composite and a free ColG (a), relative activity of ColG@ferria composite and free ColG at different temperatures (b), relative activity of ColG@ferria and free ColG after incubation at 50 °C for a period of time (c).

Bioactivity of the ColG@ferria composite

The main question to be answered is: will the entrapped enzyme be capable to effectively perform collagen fibers lysis under sterically hindered conditions? Collagen fibrils and collagen fibers are very complex objects to cleave, since they have a dense and a complex structure: typically one collagen fibril consists of 5 tropocollagen molecules, each of which in turn consists of three collagen subunits or strands.³⁸ For the cleavage of such biological rope, a strict orientation of collagenase molecule relative to the substrate is required.

In order to evaluate ColG@ferria bioactivity composite NPs was tested using model biological objects, such as collagen from calf skin, stained gelatin samples and pectoral fascia obtained from Gallus domesticus.

In the first set of experiments ColG activity was tested on collagen from calf skin. For that purpose, the collagen was dissolved in water and the turbid collagen solution was then treated by preliminary rinsed 10% ColG@ferria composite nanoparticles. In the course of collagen cleavage to low-molecular fragments, solution transmittance was measured to evaluate the kinetic of the process. This collagen cleavage experiments revealed the composite material to be an active proteolytic agent with an activity comparable to the free ColG. The rate of collagen cleavage by the magnetic composite in solution is about 2 times slower than that of a free collagenase (Fig. 5a), (8 days for magnetic composite versus 4 days for free collagenase). Though the ColG@ferria composite is less active proteolytic, the composite material can be separated from the reaction mixture due to its magnetic properties using an external magnetic field and be reused. Experiments show that the activity of the composite material remains essentially at the same level at least after four cycles of use (Fig. 5b), totally decreasing approximately by 10%, which represents good stability of the composite material and prolonged profile of action characteristic for the entrapped enzymatic systems, when the activity is not related to the release of the enzyme.

Fig. 5 Bioactivity of 10% wt ColG@ferria composite. Dynamics of a collagen lysis process by ColG@ferria and free ColG (a); collagen lysis by reused ColG@ferria composite. After each cycle composite was magnetically separated and reused (b), dynamics of a gelatin sample lysis process by ColG@ferria and free ColG (c).

The next step was the cleavage of more complex solids, and the first such model object was a sample of a stained gelatin (see Experimental section for details). Gelatin is a derivative of collagen, capable of yielding semi-solid objects similar to biological tissues.³⁹ For simplicity of studying lysis, the gelatin samples were stained with eosin dye. During the sample lysis process, gelatin framework structure was cleaved, leading to the release of eosin molecules and a change in the intensity of the solution color. Experimental data reveal (Fig. 5c) the ability of the magnetic composite to cleave solid gelatin substrate, despite the fact that the reaction occurs in a heterogeneous environment. The ColG@ferria composite shows almost 2.5 times lower activity compared to the free ColG (complete decomposition of a gelatin sample by free collagenase amounts to 4.1 hours versus 10 hours for the proteolytic composite), which can be explained by lower mobility of proteolytic particles and steric hindrance for the entrapped ColG molecules. Although lysis reaction in this case proceeded in a more heterogeneous environment (solid nanoparticles dispersed over solid gelatin) compared to the liquid collagen dispersed system, the process was much faster due to the fact that gelatin consists of partially hydrolyzed collagen and its structure is less organized and interconnected.

A distinctive feature of the magnetic composite is the possibility of magnetic targeting and magnetic concentration at a given point. This is shown by performing a tests on a glass slide (Fig. 6a–d), ColG@ferria composite could be easily concentrated at a desired location using an external magnetic field by placing neodymium magnet, which allows to perform the lysis of the substrate in a selected area. Magnetic localization accelerates lysis rate proportionally to reduction of the area under effect. Full lysis of 12 cm² gelatin sample on the glass strip took 245 min for free ColG, 620 min for equivalent quantity of ColG@ferria, and 65 min for localized lysis by ColG@ferria composite when been magnetically concentrated at the area of 1 cm². SEM images clearly show the process of lysing the gelatin sample by the ColG@ferria composite. The composite material literally melts the model sample, gradually destroying it in its location (Fig. 6c–d). Magnetic targeting feature is important for achieving the point impact, e.g., in selective tissue lysis or selective lysis of the substrate upon cultivating cells.

Fig. 6 Magnetically targeted lysis of gelatin sample by 10% wt ColG@ferria. Optical microscope visualization of magnetically targeted gelatin sample lysis by ColG@ferria composite before (a) and after 2 hours of exposure (b), SEM visualization of magnetically targeted lysis process after 10 minutes of exposure (c), zoomed area of gelatin sample treated with ColG@ferria composite (d).

Efficiency of the magnetically targeted ColG@ferria composite was also revealed for genuine collagen tissue of animal origin: pectoral fascia, which is closest in its composition and structure to the collagen material commissures of adhesive disease.⁴⁰ This pathological condition often occurs in inflammatory processes or after physical damage of fascia and is characterized by an uncontrolled growth of collagen tissue, leading to abnormal connections between organs that prevent their normal movements and can impair with some organ dysfunctions,⁴¹ with surgery as the only way to the treatment of this condition. In this case, the possibility of minimally invasive surgery by means of a magnetically localized composite material would be particularly useful. In our experiments, biological tissue was placed on a glass slide and subjected to magnetically targeted cleavage similarly to an experiment with gelatin sample. Controlled lysis of the genuine animal tissue (Fig. 7a–c) occurs for a longer period of time than that of a gelatin layer of the same thickness (12 hours versus 65 min for gelatin), due to more complex tissue structure. Despite this, magnetic targeting of the composite allowed for the selective removal of collagen fibers at a place where the ColG@ferria composite was concentrated, proving the proposed concept.

Fig. 7 Lysis of pectoral fascia with magnetically targeted ColG@ferria composite NPs after 0 hours (a), 6 hours (b), and 12 hours (c) of exposure.

It can be concluded that ColG@ferria composite indeed acts as an active proteolytic nanocomposite on substrates of different origin. The last question to be answered is: how should the composite material structure be organized, in order to be able to perform the lysis process of such complex substrates? The speculative answer to this question, based on all the data collected so far, is given in the next section.

ColG@ferria structure organization

As it was mentioned earlier, the mechanism of ColG action is very complicated and consists of two steps in which the N-terminal activator domain cooperates with the peptidase domain in recognition and processing of both collagen triple helices and microfibrils.³⁸ The first stage is the interaction of a collagenase with collagen molecules followed by a change in a collagenase conformation, after which activator domain and proteinase domain approach each other, and the interaction occurs at the bottom of the saddle and by an alternative four-helix bundle arrangement at the saddle seat. Only in this closed state is the activator HEAT repeats able to interact with triple-helical collagen and initiate the unwinding of the triple-helix α-chains, which are processively cleaved.³⁸

Structural organization of the ColG molecule has a pronounced charge distribution on the surface (Fig. 8a). In the fragment of the molecule responsible for proteolytic activity (activator domain and proteinase domain) there is a distinct area for docking of collagen, which has the same sign charge as the used magnetite matrix, which should lead to electrostatic repulsion of the section from magnetite nanomatrix. The remaining fragments of the molecule and its overall charge are oppositely charged with respect to a magnetite matrix, which should result in electrostatic attraction of ColG molecule. Taking into account the mechanism of ColG action, the data on enzyme release from the composite vs. mass fraction of the enzyme (Fig. 3a), and data on zeta potential of the composite particles vs. mass fraction of the enzyme (Fig. 3b), one can schematically depict the composite structure, as suggested in Fig. 8b. Apparently, there is an electrostatic interaction between the C-terminal collagenase domain and magnetic matrix nanopores, wherein the proteinase domain is electrostatically repelled from the matrix and guided to the surface of the composite particles sticking out in the shape of a “claw”, which is not subject to strong steric hindrances and capable of lysing collagen fibers. Such nanoarchitecture remains relatively stable over time, and the resultant magnetic nanocomposite can be magnetically separated and used repeatedly.

Fig. 8 Structure of the ColG and ColG@ferria composite. Surface charge plot of the ColG molecule visualised with HyperChem 8.0.8 (a) supposed schematic representation of ColG@ferria composite based on obtained experimental data (b).

Conclusions

This paper presents a new class of collagenase-based magnetically targeted biocompatible proteolytic systems obtained by the entrapment method. Nanoarchitecture of the obtained material is able to cleave collagen molecules in the form of collagen fibers and collagen tissues while providing prolonged action and reusability. Due to the peculiarities of the entrapment process, the enzyme appears to be stabilized against aggressive factors, such as temperature. It has 18 °C higher denaturation temperature as compared to free enzyme and shows high levels of thermal stability. The activity maximum is shifted by 10 °C to the value of 50 °C, with an activity at 65 °C being 51% of the maximal, compared to the completely denatured and with zero activity free enzyme at this temperature. Due to the synthesis conditions and used materials the composite nanomaterial is completely biocompatible and suitable for parenteral administration. This class of proteolytic systems has high prospects for application in industry and medicine, fighting pathologies associated with collagen tissue abnormalities, such as adhesive disease and Dupuytren's contracture, and opens the door to a new era in medicine – minimally invasive nanosurgery.

Experimental section

Materials

Chemicals. The hydrosol was prepared from iron(II) chloride tetrahydrate, iron(III) chloride hexahydrate and ammonia; Tris buffer (pH 7.4), collagenase G, collagen from calf skin type III and gelatin from porcine skin were all obtained from Sigma-Aldrich. Chromogenic substrate (for-Ala-Phe-Lys-pNA) and saline were obtained from “Kvik” LTD Company.

Preparation of ferria hydrosol. Pure ferria hydrosol was prepared ultrasonically from iron(II) chloride tetrahydrate and iron(III) chloride hexahydrate as described.⁴² The mass fraction of in the resulting sol of magnetite nanoparticles with an average particle size of 10 nm was 2 wt%.

Preparation of proteolytic composites. 330 μL of freshly prepared hydrosol was mixed with 20 to 100 μL of collagenase G solution (900 CDU mL⁻¹) and stirred for 5 minutes. The resulting mixture was dried in a desiccator under reduced pressure. The produced composite material was mechanically crushed, suspended in 1 mL of deionized water and passed through a 1 μm filter to filter out all particles larger than 1 μm. The resulting particles present in the filtrate were used in the experiments. To evaluate the release of the enzyme, the same amount of the crushed composite was transferred into a quartz cuvette and treated with 2 mL of saline solution while absorption spectrum was measured at 195 nm over time at temperature of 37 °C.

Preparation of gelatin jelly. 0.2 g of gelatin was dissolved in 1 mL of distilled water at a temperature of 40 °C for 5 min. 100 μL of a 1% solution of eosin dye was added, stirred and allowed to cool to the room temperature and kept at 4 °C for 2 hours to yield dense homogeneous structure. Resulting stained jelly was washed with water to remove excess of the dye and used in the experiments.

Collagen model of animal origin. Pectoral fascia obtained from Gallus domesticus was used as model genuine collagen of animal origin. A 5 μm-thick tissue sample was removed from the body during dissection procedure and analyzed without additional processing. All experiments were approved by institutional animal ethical committee (Institute of Cytology, RAS, Russia no. 0110) and are in agreement with the guidelines for the proper use of animals for biomedical research.

Methods

Evaluating thermal stability of collagenase and ColG@ferria. To evaluate the thermal stability of the composite at various temperatures, a 150 μL of 10 wt% ColG@ferria colloid (equal to 67.5 CDU) composite was mixed with 1.5 mL of Tris buffer and incubated at 20–65 °C for 10 minutes, after which 0.5 mL of a 3 mg mL⁻¹ solution of chromogenic substrate (for-Ala-Phe-Lys-pNA) was added and incubated for 120 seconds at the same temperature measuring the absorbance at 405 nm. Testing for possible activity of the rinsing solutions was carried out with 1.5 mL of Tris-buffer and 0.5 mL of the chromogenic substrate solution. For comparison, 75 μL of collagenase solution (900 CDU mL⁻¹) and 1.5 mL of Tris buffer were taken and treated similarly to the bioactive composite.

To evaluate the thermal stability of the composite at elevated temperatures, 150 μL of 10 wt % ColG@ferria composite was mixed with 1.5 mL of Tris buffer and incubated at 50 °C for 1–4 hours, allowed to cool to room temperature, mixed with 0.5 mL of the chromogenic substrate at concentration of 3 mg mL⁻¹ and the change in optical density was measured over time at 405 nm at 37 °C. For comparison, 75 μL of collagenase solution (900 U mL⁻¹) and 1.5 mL of Tris buffer were taken and treated similarly to the bioactive composite.

Studying activity of the ColG@ferria composite on collagen solution. 10 mg of collagen was dissolved in 3 mL of Tris buffer in a cuvette for 15 minutes, the resulting turbid solution was treated with 250 μL of a 10% wt ColG@ferria prepared as described above (an activity of 450 CDU mL⁻¹), and incubated at 37 °C, measuring the absorption spectrum at 415 nm every 24 hours. For comparison, 10 mg of collagen in 3 mL of Tris-buffer and 125 μL of collagenase solution (activity 900 CDU mL⁻¹) were taken and treated similarly to the bioactive composite.

Proteolytic activity of ColG@ferria composites on gelatin samples. Proteolytic activities of ColG@ferria were evaluated by measuring the adsorption spectra of the solution while cleaving stained gelatin sample. 0.5 g of stained gelatin sample produced as described above was placed in 3 mL of purified water treated with 150 μL of 10% wt ColG@ferria composite (equal to 65.7 CDU). Changes in the optical density of the solution at 520 nm in a kinetic mode were measured for 10 hours at 37 °C. For comparison, 75 μL of collagenase solution (900 U mL⁻¹) was taken and treated similarly to the bioactive composite.

Proteolytic activity of ColG@ferria composites on gelatin films. 100 μL of 20% gelatin stained with eosin was applied on a glass slide, forming a 2 mm thick layer and allowing to stand at 4 °C for 2 hours to form dense homogeneous structure. The resulting sample was treated with 150 μL of magnetic composite sol with an activity of 450 U mL⁻¹, and the magnetic composite was concentrated with an external magnetic field by manipulating with neodymium magnet. Activity of proteolytic system was evaluated using a LOMO MIKMED optical microscope with an ×10 lens at a temperature of 25 °C. Analogous experiments were performed for the magnetite matrix without the enzyme, for comparison reasons.

Studying proteolytic activity of magnetite nanoparticles using SEM. The studies were performed in a similar way to those carried out with an optical microscope using as a sample tip for SEM as a substrate. After incubation for 10 min at 37 °C the sample was placed in a vacuum desiccator and dried over silica. SEM research was carried out after coating a gold layer with thickness of 10 nm.

Degradation of pectoral fascia. The 5 μm thick fascia was separated from the pectoralis major and placed on a glass slide, treated with 150 μL of ColG@ferria sol with an activity of 450 CDU mL⁻¹. The magnetic composite was concentrated with an external magnetic field and incubated at 37 °C for 12 hours at a humidity of 100% to prevent drying. The result of fascia degradation was studied using a LOMO MIKMED microscope with an ×10 lens.

Characterization methods. Specific surface area, pore volume and pore size distribution were investigated using Quantachrome Nova 1200e by nitrogen adsorption at 77 K and analyzed by the BET and BJH equations. Prior to analysis, all samples were degassed at room temperature for 48 hours. The samples for transmission electron microscopy (TEM) were obtained by dispersing a small probe in ethanol to form a homogeneous suspension. Then, a suspension drop was coated on a copper mesh covered with carbon for a TEM analysis (FEI TECNAI G2 F20, at an operating voltage of 200 kV). To analyze samples using high-resolution scanning electron microscopy (SEM), the obtained ground xerogel was deposited on a metal tip and investigated without additional spraying using a Magellan 400L ultra-high resolution electron microscope. For a SEM analysis of gelatin degradation by proteolytic composites, the samples were dried in vacuo for 1 h and examined using a Tescan VEGA 3 electron microscope. Optical microscopy was done on a LOMO Biolam M-1 microscope with an ×10 lens. Hydrodynamic diameter was measured by the DLS technique on Photocor Compact Z. Spectrophotometrical measurements of enzymatic activity were carried out using an Agilent Cary HP 8454 Diode Array spectrophotometer with TEC. DSC curves were obtained with a 204 F1 Phoenix NETZSCH apparatus, and a heating rate of 10 °C min⁻¹ was used from 30 °C to 150 °C under nitrogen.

Acknowledgements

Electronic microscope measurements were performed in Nanocharacterization Department of Hebrew University, Jerusalem. This work was supported by Russian Foundation for Basic Research, grant No. 16-13-00041.

References

1. R. Weissleder, C. H. Tung, U. Mahmood and A. Bogdanov, Nat. Biotechnol., 1999, 17, 375–378.
2. L. Eikenes, Ø. S. Bruland, C. Brekken and C. de Lange Davies, Cancer Res., 2004, 64, 4768–4773.
3. T. T. Goodman, P. L. Olive and S. H. Pun, Int. J. Nanomed., 2007, 2, 265–274.
4. G. Secchi, Clin. Dermatol., 2006, 26, 321–325.
5. A. M. Pendzhiev, Pharm. Chem. J., 2002, 36, 315–317.
6. S. K. Nair, I. K. Bhat and A. L. Aurora, Arch. Surg., 1974, 108, 849–853.
7. R. F. Diegelmann and M. C. Evans, Front. Biosci., 2004, 9, 283–289.
8. S. Coleman, D. Gilpin, F. T. Kaplan, A. Houston, G. J. Kaufman, B. M. Cohen, N. Jones and J. P. Tursi, J. Hand Surg., 2014, 39, 57–64.
9. L. C. Hurst, M. A. Badalamente, V. R. Hentz, R. N. Hotchkiss, F. T. Kaplan, R. A. Meals, T. M. Smith and J. Rodzvilla, N. Engl. J. Med., 2009, 361, 968–979.
10. T. M. Skirven, A. Bachoura, S. M. Jacoby, R. W. Culp and A. L. Osterman, J. Hand Surg., 2013, 38, 684–689.
11. D. Warwick, M. Arner, G. Pajardi, B. Reichert, Z. Szabo, E. H. Masmejean, J. Fores, D. S. Chapman, R. A. Gerber, F. Huard, A. Seghouani and P. P. Szczypa, J. Hand Surg., 2015, 40, 124–132.
12. Y. Q. Jin, W. Liu, T. H. Hong and Y. Cao, J. Neurosci. Methods, 2008, 170, 140–148.
13. P. Moldéus, J. Högberg and S. Orrenius, Methods Enzymol., 1978, 52, 60–71.
14. C. F. Markarian, G. Z. Frey, M. D. Silveira, A. R. Milani, P. B. Ely, A. P. Horn and M. Camassola, Biotechnol. Lett., 2014, 36, 693–702.
15. S. J. Huang, R. H. Fu, W. C. Shyu, S. P. Liu, G. P. Jong, Y. W. Chiu and S. Z. Lin, Cell Transplant., 2013, 22, 701–709.
16. P. Salehinejad, N. B. Alitheen, A. M. Ali, A. R. Omar, M. Mohit, E. Janzamin and S. N. Nematollahi-Mahani, In Vitro Cell. Dev. Biol.: Anim., 2012, 48, 75–83.
17. M. Ashford and J. Fell, J. Drug Targeting, 1994, 2, 241–257.
18. K. G. Wakim and G. A. Fleisher, J. Lab. Clin. Med., 1963, 61, 107–119.
19. D. Negrini, O. Tenstad, A. Passi and H. Wiig, Am. J. Physiol.: Lung Cell. Mol. Physiol., 2006, 290, L470–L477.
20. P. Libby, Arterioscler., Thromb., Vasc. Biol., 2012, 32, 2045–2051.
21. M. R. Villegas, A. Baeza and M. Vallet-Regí, ACS Appl. Mater. Interfaces, 2015, 7, 24075–24081.
22. B. Zhang, Z. Luo, J. Liu, X. Ding, J. Li and K. Cai, J. Controlled Release, 2014, 192, 192–201.
23. H. Chen, Z. Zhen, W. Tang, T. Todd, Y. J. Chuang, L. Wang and J. Xie, Theranostics, 2013, 3, 650.
24. K. Kim, J. Lam, S. Lu, P. P. Spicer, A. Lueckgen, Y. Tabata and F. K. Kasper, J. Controlled Release, 2013, 168, 166–178.
25. A. Fujita, H. Kawakita, K. Saito, K. Sugita, M. Tamada and T. Sugo, Biotechnol. Prog., 2003, 19, 1365–1367.
26. D. Li, W. Y. Teoh, J. J. Gooding, C. Selomulya and R. Amal, Adv. Funct. Mater., 2010, 20, 1767–1777.
27. A. Butt, A. Farrukh, A. Ghaffar, H. Duran, Z. Oluz, H. ur Rehman and B. Yameen, RSC Adv., 2015, 5, 77682–77688.
28. A. Zaks, Curr. Opin. Chem. Biol., 2001, 5, 130–136.
29. T. M. Allen and P. R. Cullis, Science, 2004, 303, 1818–1822.
30. L. X. Tiefenauer, A. Tschirky, G. Kühne and R. Y. Andres, Magn. Reson. Imaging, 1996, 14, 391–402.
31. A. S. Drozdov, K. V. Volodina, V. V. Vinogradov and V. V. Vinogradov, RSC Adv., 2015, 5, 82992–82997.
32. A. S. Drozdov, O. E. Shapovalova, V. Ivanovski, D. Avnir and V. V. Vinogradov, Chem. Mater., 2016, 28, 2248–2253.
33. A. S. Drozdov, V. V. Vinogradov, I. P. Dudanov and V. V. Vinogradov, Sci. Rep., 2016, 28119.
34. R. Sugasawara and E. Harper, Biochemistry, 1984, 23, 5175–5181.
35. Particulate contamination: sub-visible particles, European Pharmacopoeia, Directorate for the Quality of Medicines & HealthCare of the Council of Europe (EDQM), Council of Europe, Strasbourg, 8th edn, 2014, ch. 2.9.19, pp. 321–323.
36. R. H. Müller, Int. J. Pharm., 1998, 160, 229–237.
37. V. V. Vinogradov and D. Avnir, J. Mater. Chem. B, 2014, 2, 2868–2873.
38. U. Eckhard, E. Schönauer, D. Nüss and H. Brandstetter, Nat. Struct. Mol. Biol., 2011, 18, 1109–1114.
39. R. T. Jones, Pharm. Capsules, 2004, 23–60.
40. A. Nagler, O. Genina, I. Lavelin, M. Ohana and M. Pines, Am. J. Obstet. Gynecol., 1999, 180, 558–563.
41. C. L. Kowalczyk and M. P. Diamond, Peritoneal Adhesions, Springer, Berlin Heidelberg, 1997, pp. 315–324.
42. A. S. Drozdov, V. Ivanovski, D. Avnir and V. V. Vinogradov, J. Colloid Interface Sci., 2016, 468, 307–312.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra14711a

---