Novel Triple Stimuli Responsive Interpenetrating Poly(Carboxybetaine Methacrylate)/Poly(Sulfobetaine Methacrylate) Network

The study reports the synthesis and characterization of novel triple stimuli responsive interpenetrating polymer network (IPN) based on two polyzwitterionic networks, namely of poly(carboxybetaine methacrylate) and poly(sulfobetaine methacrylate). The zwitterionic IPN hydrogel demonstrates the ability to expand or shrink in response to changes in three “biological” external stimuli such as temperature, pH, and salt concentration. The IPN hydrogel shows good mechanical stability. In addition, other important features such as non-cytotoxicity and antibiofouling activity against three widespread bacteria as P. Aeruginosa, A. Baumanii, and K. Pneumoniae are demonstrated. The in vivo behavior of the novel zwitterionic IPN hydrogel suggests that this smart material has very good potential as a biomaterial.


Introduction
Hydrogels are three-dimensional networks made from crosslinked polymers bearing in their side chains hydrophilic functional groups such as -COOH, -OH, -CONH 2 , -SO 3 H, R 4 N + , etc. These polymer networks are able to absorb and retain high water amounts, much higher than their own weight, which results in a significant increase in their volume upon swelling. The functional groups define the hydrogels' responsiveness to changes in the chemical, physical or biological environment which leads to a dramatic change in the hydrogels' size and shape. Thus, hydrogels could undergo swelling or deswelling in response to external stimuli which defines them as "smart" materials. Some of the external stimuli that could influence the hydrogels' behavior are temperature, pH, and salt concentration [1]. The hydrogels that respond to more than one external stimulus are rarely reported although they currently find an increasing number of different applications [2].
Hydrogels are soft, flexible, and usually biocompatible materials that mimic the mechanical performance of biological tissues as well as provide a good environment for cell survival and proliferation [3]. Hence, they have many biomedical applications, e.g., in tissue engineering, drug delivery, wound healing, immunotherapy, plastic surgery, etc. [3]. Smart hydrogels are also commonly used as biosensors for enzyme immobilization [4], bio actuators [5], draw agents in osmosis desalination [6], 3D and 4D printing, etc. However, the broad spectrum of hydrogel applications is somehow limited by their weak mechanical properties [7]. For this reason, many efforts are underway to improve the mechanical Gels 2023, 9, 90 3 of 20 interlacing between both zwitterionic networks in their IPN hampers the formation of these clusters, thus decreasing the amount of the physical junctions as compared to the neat PSB and PCB, respectively lowering the IPNs microhardness; and (ii) the fact that not all chain entanglements are able to carry substantial load upon the network deformation.   [23].
The chain entanglements could be divided into two groups: (i) load-bearing entanglements (designated with 1 in Scheme 1B), which are usually formed by the interlacing of two long polymer chains and this interlacing takes place far from the chain ends as well as (ii) non-load-bearing entanglements (designated with 2 in Scheme 1B), which are the entanglements between both polymers chains formed close to the chains ends. The in situ formation of the 2 nd network in the presence of the 1 st one during the IPN formation is expected to generate a less ideal network, i.e., containing more chain ends, and thus the number of the non-load-bearing entanglements is expected to be higher in the IPN obtained via the sequential method. The microhardness for such a "defects-enriched" network could be expected to be lower due to the increased number of chain ends, respectively to the number of non-load bearing entanglements. It could be concluded that the PCB/PSB IPN network microhardness is comparable to the microhardness of other polymeric materials and shows that the newly developed material is strong enough to be used as biomaterial.

Rheology
The viscoelastic properties of PCB/PSB IPN hydrogel were studied via frequency sweep of the storage (G') and loss (G'´) moduli ( Figure 1). The storage modulus of the PCB/PSB IPN (G´~ 23.5 kPa) is much higher than the loss modulus (G´´~0.5 kPa) within the entire frequency range from 0.1 to 10 Hz, indicating that the IPN hydrogel is mechanically stable. When comparing G´ modulus of the IPN hydrogel to the G´ moduli of The chain entanglements could be divided into two groups: (i) load-bearing entanglements (designated with 1 in Scheme 1B), which are usually formed by the interlacing of two long polymer chains and this interlacing takes place far from the chain ends as well as (ii) non-load-bearing entanglements (designated with 2 in Scheme 1B), which are the entanglements between both polymers chains formed close to the chains ends. The in situ formation of the 2nd network in the presence of the 1st one during the IPN formation is expected to generate a less ideal network, i.e., containing more chain ends, and thus the number of the non-load-bearing entanglements is expected to be higher in the IPN obtained via the sequential method. The microhardness for such a "defects-enriched" network could be expected to be lower due to the increased number of chain ends, respectively to the number of non-load bearing entanglements.
It could be concluded that the PCB/PSB IPN network microhardness is comparable to the microhardness of other polymeric materials and shows that the newly developed material is strong enough to be used as biomaterial.

Rheology
The viscoelastic properties of PCB/PSB IPN hydrogel were studied via frequency sweep of the storage (G') and loss (G") moduli ( Figure 1). The storage modulus of the PCB/PSB IPN (G'~23.5 kPa) is much higher than the loss modulus (G"~0.5 kPa) within the entire frequency range from 0.1 to 10 Hz, indicating that the IPN hydrogel is mechanically stable. When comparing G' modulus of the IPN hydrogel to the G' moduli of both single PSB [18] and PCB [19] hydrogels, obtained using the same or similar crosslinking agent concentrations, it appeared that the IPN hydrogels are much stronger. For example, while the 4PCB hydrogel has G'~15 kPa [19] and PSB hydrogel with comparable crosslinking density (3PSB) has G'~6 kPa [18], the PCB/PSB IPN hydrogel has much higher G' value (G'~23.5 kPa). This different rheological behavior of the IPN hydrogel could be related to the entanglements obtained via the mutual interlacing of both PSB and PCB networks into Gels 2023, 9,90 4 of 20 the IPN, which enhances the IPN's storage modulus. Thus, the PCB/PSB IPN hydrogel has improved mechanical strength as compared to its constituent single networks, which makes it very appropriate to be used in biomedical applications. both single PSB [18] and PCB [19] hydrogels, obtained using the same or similar crosslinking agent concentrations, it appeared that the IPN hydrogels are much stronger. For example, while the 4PCB hydrogel has G´~15 kPa [19] and PSB hydrogel with comparable crosslinking density (3PSB) has G´~ 6 kPa [18], the PCB/PSB IPN hydrogel has much higher G´ value (G´~ 23.5 kPa). This different rheological behavior of the IPN hydrogel could be related to the entanglements obtained via the mutual interlacing of both PSB and PCB networks into the IPN, which enhances the IPN's storage modulus. Thus, the PCB/PSB IPN hydrogel has improved mechanical strength as compared to its constituent single networks, which makes it very appropriate to be used in biomedical applications.  It should be mentioned here that the different roles that the chain entanglements in the IPN play for the microhardness and the storage modulus could originate from the fact that the former is measured for dry IPN sample, while the latter is obtained for IPN hydrogel at its equilibrium swelling ratio.

Temperature Responsiveness of PCB/PSB IPN Hydrogel
As PSB is a temperature-responsive polymer, it was interesting to check if it imparts this smart behavior also to the PCB/PSB IPN as PCB is known not to respond to temperature changes. The PCB/PSB IPN hydrogel exhibits a temperature-dependent swelling profile as demonstrated in Figure 2. Its swelling ratio (SR) increases almost twice (from 1.8 to 3.5) upon temperature increase in the studied temperature range. In a previous study, we have demonstrated that PSB hydrogels show upper critical solution temperature behavior, i.e., their swelling is enhanced upon heating, and the magnitude of this swelling enhancement is defined mainly by the crosslinking agent poly(ethylene glycol) diacrylate (PEGDA) amount: the lower the crosslinking agent concentration, the stronger the SR increase with temperature is [18]. Similarly to the observed for the PCB/PSB IPN behavior, at low temperatures (T < 20 °C) the PSB hydrogels shrink due to the formation of the dipole-dipole zip clusters between the PSB side chain zwitterionic moieties (Scheme 1A). The gradual temperature increase from 20 °C to 70 °C results in destroying the dipole-dipole zwitterionic clusters (Scheme 2), and hence the PCB/PSB IPN hydrogel expands, reaching a plateau at T > 70 °C. This plateau is defined by the constraints that the chemical crosslinking with PEGDA imparts when the entire physical network of PSB is destroyed upon heating. It should be mentioned here that the different roles that the chain entanglements in the IPN play for the microhardness and the storage modulus could originate from the fact that the former is measured for dry IPN sample, while the latter is obtained for IPN hydrogel at its equilibrium swelling ratio.

Temperature Responsiveness of PCB/PSB IPN Hydrogel
As PSB is a temperature-responsive polymer, it was interesting to check if it imparts this smart behavior also to the PCB/PSB IPN as PCB is known not to respond to temperature changes. The PCB/PSB IPN hydrogel exhibits a temperature-dependent swelling profile as demonstrated in Figure 2. Its swelling ratio (SR) increases almost twice (from 1.8 to 3.5) upon temperature increase in the studied temperature range. In a previous study, we have demonstrated that PSB hydrogels show upper critical solution temperature behavior, i.e., their swelling is enhanced upon heating, and the magnitude of this swelling enhancement is defined mainly by the crosslinking agent poly(ethylene glycol) diacrylate (PEGDA) amount: the lower the crosslinking agent concentration, the stronger the SR increase with temperature is [18]. Similarly to the observed for the PCB/PSB IPN behavior, at low temperatures (T < 20 • C) the PSB hydrogels shrink due to the formation of the dipoledipole zip clusters between the PSB side chain zwitterionic moieties (Scheme 1A). The gradual temperature increase from 20 • C to 70 • C results in destroying the dipole-dipole zwitterionic clusters (Scheme 2), and hence the PCB/PSB IPN hydrogel expands, reaching a plateau at T > 70 • C. This plateau is defined by the constraints that the chemical crosslinking with PEGDA imparts when the entire physical network of PSB is destroyed upon heating.
Thus, the PSB component of the IPN successfully imparts temperature responsiveness to the PCB/PSB IPN making it temperature responsive material.  Thus, the PSB component of the IPN successfully imparts temperature responsiveness to the PCB/PSB IPN making it temperature responsive material.

pH Responsiveness of PCB/PSB IPN Hydrogel
Due to the presence of -COOH groups in its side chains, PCB is known to be pH responsive polymer. Thus, the influence of pH on the PCB/PSB IPN hydrogel swelling was studied ( Figure 3). In an acidic medium, below pH ~ 4.5-5, PCB exists as polycation due to the positively charged quaternary groups in its side chains (-COOH groups are protonated below their pKa~4.5-5, thus transforming the zwitterionic moieties into cationic ones). Repulsive electrostatic forces occur between the positively charged amino groups in the PCB side chains, which results in a strong expansion of the PCB/PSB IPN hydrogel. Upon pH increase, -COOH groups are gradually deprotonating, which recovers the zwitterionic nature of the PCB side chains and increases the number of the newly formed dipole-dipole zip clusters. As a result, the PCB/PSB IPN hydrogel shrinks reaching a plateau at its SR at pH > 6 ( Figure 3). It is interesting to note that the single PCB Thus, the PSB component of the IPN successfully imparts temperature responsiveness to the PCB/PSB IPN making it temperature responsive material.

pH Responsiveness of PCB/PSB IPN Hydrogel
Due to the presence of -COOH groups in its side chains, PCB is known to be pH responsive polymer. Thus, the influence of pH on the PCB/PSB IPN hydrogel swelling was studied ( Figure 3). In an acidic medium, below pH ~ 4.5-5, PCB exists as polycation due to the positively charged quaternary groups in its side chains (-COOH groups are protonated below their pKa~4.5-5, thus transforming the zwitterionic moieties into cationic ones). Repulsive electrostatic forces occur between the positively charged amino groups in the PCB side chains, which results in a strong expansion of the PCB/PSB IPN hydrogel. Upon pH increase, -COOH groups are gradually deprotonating, which recovers the zwitterionic nature of the PCB side chains and increases the number of the newly formed dipole-dipole zip clusters. As a result, the PCB/PSB IPN hydrogel shrinks reaching a plateau at its SR at pH > 6 ( Figure 3). It is interesting to note that the single PCB Scheme 2. Temperature increase causes disruption of the PSB physical network and hence the hydrogel expansion.

pH Responsiveness of PCB/PSB IPN Hydrogel
Due to the presence of -COOH groups in its side chains, PCB is known to be pH responsive polymer. Thus, the influence of pH on the PCB/PSB IPN hydrogel swelling was studied ( Figure 3). In an acidic medium, below pH~4.5-5, PCB exists as polycation due to the positively charged quaternary groups in its side chains (-COOH groups are protonated below their pKa~4.5-5, thus transforming the zwitterionic moieties into cationic ones). Repulsive electrostatic forces occur between the positively charged amino groups in the PCB side chains, which results in a strong expansion of the PCB/PSB IPN hydrogel. Upon pH increase, -COOH groups are gradually deprotonating, which recovers the zwitterionic nature of the PCB side chains and increases the number of the newly formed dipole-dipole zip clusters. As a result, the PCB/PSB IPN hydrogel shrinks reaching a plateau at its SR at pH > 6 ( Figure 3). It is interesting to note that the single PCB networks show similar behavior although their SR changes stronger upon pH increase, and the plateau there is observed for pH > 9 [19]. networks show similar behavior although their SR changes stronger upon pH increase, and the plateau there is observed for pH > 9 [19]. Thus, it could be concluded that each of the PCB/PSB IPN constituents is imparting its own "smart" behavior, defining in this way its dual stimuli responsiveness. These results also confirm that the IPN approach is an effective tool to create smart materials able to respond to more than one external stimulus change.

Salt Responsive Swelling Behaviour of PCB/PSB IPN Hydrogel
Since both constituent networks, PSB and PCB, are able to respond to salt concentration changes, exhibiting antipolyelectrolyte behavior, it was expected that their IPN will also possess salt responsiveness. That is why the effect of NaCl concentration on the swelling response of the PCB/PSB IPN network in an aqueous solution was studied ( Figure 4). It is seen that the swelling ratio of the IPN hydrogel increases with NaCl concentration increase, i.e., with the ionic strength of the solution increase.   Thus, it could be concluded that each of the PCB/PSB IPN constituents is imparting its own "smart" behavior, defining in this way its dual stimuli responsiveness. These results also confirm that the IPN approach is an effective tool to create smart materials able to respond to more than one external stimulus change.

Salt Responsive Swelling Behaviour of PCB/PSB IPN Hydrogel
Since both constituent networks, PSB and PCB, are able to respond to salt concentration changes, exhibiting antipolyelectrolyte behavior, it was expected that their IPN will also possess salt responsiveness. That is why the effect of NaCl concentration on the swelling response of the PCB/PSB IPN network in an aqueous solution was studied ( Figure 4). It is seen that the swelling ratio of the IPN hydrogel increases with NaCl concentration increase, i.e., with the ionic strength of the solution increase. Thus, it could be concluded that each of the PCB/PSB IPN constituents is impart its own "smart" behavior, defining in this way its dual stimuli responsiveness. These sults also confirm that the IPN approach is an effective tool to create smart materials a to respond to more than one external stimulus change.

Salt Responsive Swelling Behaviour of PCB/PSB IPN Hydrogel
Since both constituent networks, PSB and PCB, are able to respond to salt conc tration changes, exhibiting antipolyelectrolyte behavior, it was expected that their I will also possess salt responsiveness. That is why the effect of NaCl concentration on swelling response of the PCB/PSB IPN network in an aqueous solution was stud ( Figure 4). It is seen that the swelling ratio of the IPN hydrogel increases with NaCl c centration increase, i.e., with the ionic strength of the solution increase.   PCB/PSB IPN hydrogel increases 3 times its initial weight when swollen in distilled water, while in NaCl aqueous solutions, with salt concentration ranging from 0.1 M to 5 M, its swelling ratio increases from 3.5 to 9, thus reaching 9 times the increase of its initial weight in the aqueous solution with the highest (5 M) NaCl concentration. This behavior is a result of the shielding effect that the NaCl ions play for the dipole-dipole interactions between the zwitterionic moieties in the IPN [18,19]. The increase in the salt concentration disrupts the interchain dipole-dipole interactions in a way that is akin to the way temperature does it for PSB (Scheme 2) [18]. As a consequence, the IPN hydrogel expands as the NaCl concentration in the aqueous solution increases.
It is interesting to note here that the single PSB [18] and PCB [19] hydrogels show different types of dependence of their swelling ratios with NaCl concentration increase: their SRs increase and level off at NaCl concentrations above 3 M. For the PCB/PSB IPN hydrogel, however, almost a linear dependence of its SR on the NaCl concentration increase is observed. This is a result of the chains interlacing, which occurs between both constituent networks in the IPN, which limits the SR and results in different profiles of the SR's salt concentration dependence.
The triple stimuli responsiveness of the PCB/PSB IPN hydrogel towards temperature, pH, and salt concentration was convincingly demonstrated so far. We were able to find only one report about a polymer-based triple stimuli-responsive system able to respond simultaneously to changes in temperature, pH, and ionic strength [24]. The system is based on copolymers of N-isopropylacrylamide (NIPAM) and acryloyloxyethyl trimethyl ammonium chloride (DAC), reinforced by exfoliated laponite. This material exhibits low critical solution temperature behavior with the ability to collapse upon swelling in aqueous solutions with very low salt concentrations (0.1 M), i.e., exhibiting polyelectrolyte behavior, as well as responding to pH change in a very undefined way (no clear dependence between the pH change and the SR is seen). In this respect, the newly developed PCB/PSB IPN has strongly pronounced responses in terms of its SR change with respect to the three studied stimuli, as well as changes in much wider ranges of their variation. There are also some other studies where triple stimuli-responsive systems to other biological stimuli are reported, e.g., where sensitivity to different biomolecules instead of ionic strength is shown [25,26].
One of the strongest advantages that PZI, and in particular PSB and PCB, possess is their biocompatibility. They are known to be highly biocompatible, which significantly increased the interest in PZI-based materials recently. The PZs' biocompatibility is related to the very low non-specific protein absorption that the PZIs' based materials possess, which defines also their antibiofouling activity as well as their good in vivo performance. Thus, the demonstrated improved mechanical performance as well as the triple stimuli responsiveness of the PCB/PSB IPN hydrogel has to be accompanied by a study on their biocompatibility in order to fully reveal their potential as a smart biomaterial.

Freezable and Non-Freezable Water in PCB/PSB IPN Hydrogel
The biocompatibility of the PZI-based materials is often explained by the high amount of bound water they are able to retain. Using differential scanning calorimetry (DSC) analysis, the amounts of the non-freezable (NFW) and the freezable (FW) water in PCB/PSB IPN hydrogel were determined ( Figure 5). The non-freezable water is considered to be bound by the polymer functional groups, which reduces its mobility and significantly decreases its freezing temperature, i.e., it does not freeze at 0 • C or around. Part of the water molecules in the hydrogel, however, stay "free" from such interactions, and thus they are able to freeze at 0 • C-these are designated as FW. The DSC analysis shows that the amount of FW in PCB/PSB IPN network is 10% while the NFW is~33%. For the sake of comparison, data for the same two water amounts are provided in Figure 5 for single PCB and PSB networks with the same or comparable crosslinking density. freezable water non-freezable water Figure 5. Amount of non-freezable water (NFW) and freezable water (FW) in PCB/PSB IPN hydrogel. For sake of comparison the data for single PSB and PCB networks are also provided [18,19].
It is seen in Figure 5 that the PCB/PSB IPN hydrogel has the lowest FW amount and comparable to the single PCB network NFW amount. The low amount of FW could be due to the more entangled structure of the IPN as compared to the single networks, which limit the free water molecules' diffusion. This explanation is confirmed by the equilibrium swelling ratio (ESR) values for the same samples presented in Table 2, where it is seen that the ESR value of the PCB/PSB IPN is lower than the ESR for the neat single PSB and PCB networks. Nevertheless, applying the above-outlined relation between "bound" NFW water and the biocompatibility, one could expect that the IPN hydrogel would have compared to the single PCB biocompatibility. It should be mentioned here that PCB is recognized to exhibit better in vivo performance than PSB, as reported by our [18,19] as well as others [27] recent studies.

Bacterial Film Growth Inhibition
PZs are known to exhibit antibiofouling activity, which is explained via their strong interaction with water, which lies behind the formation of a soft hydrated layer on their surface. This hydrated layer makes hard the bacteria attachment and prevents bacterial colonization, thus minimizing the risk of bacterial biofilm formation. To check if the PCB/PSB IPN hydrogel shows such inhibitory activity towards bacterial film growth, we used three commonly spread bacteria, namely P. Aeruginosa, A. Baumanii, and K. Pneumoniae. For sake of comparison the data for single PSB and PCB networks are also provided [18,19].
It is seen in Figure 5 that the PCB/PSB IPN hydrogel has the lowest FW amount and comparable to the single PCB network NFW amount. The low amount of FW could be due to the more entangled structure of the IPN as compared to the single networks, which limit the free water molecules' diffusion. This explanation is confirmed by the equilibrium swelling ratio (ESR) values for the same samples presented in Table 2, where it is seen that the ESR value of the PCB/PSB IPN is lower than the ESR for the neat single PSB and PCB networks. Nevertheless, applying the above-outlined relation between "bound" NFW water and the biocompatibility, one could expect that the IPN hydrogel would have compared to the single PCB biocompatibility. It should be mentioned here that PCB is recognized to exhibit better in vivo performance than PSB, as reported by our [18,19] as well as others [27] recent studies.

Bacterial Film Growth Inhibition
PZs are known to exhibit antibiofouling activity, which is explained via their strong interaction with water, which lies behind the formation of a soft hydrated layer on their surface. This hydrated layer makes hard the bacteria attachment and prevents bacterial colonization, thus minimizing the risk of bacterial biofilm formation. To check if the PCB/PSB IPN hydrogel shows such inhibitory activity towards bacterial film growth, we used three commonly spread bacteria, namely P. Aeruginosa, A. Baumanii, and K. Pneumoniae. P. Aeruginosa is a Gram-negative aerobic bacterium that has been identified as a pathogen for humans and plants. It could be found in the environment, e.g., in soil and water, and can cause infections in the blood, lungs (pneumonia), or other parts of the body after surgery. K. pneumoniae is a Gram-negative facultative anaerobic bacterium that although found in the normal flora of the mouth, skin, and intestines, can cause destructive changes to human and animal lungs if aspirated.
A. Baumannii is a Gram-negative bacterium that is a pathogen in humans, affecting people with compromised immune systems. It is becoming increasingly important in hospital-derived infection, as it is almost exclusively isolated from hospital environments.
The results demonstrate that PCB/PSB IPN hydrogel reduces the growth of P. Aeruginosa, K. pneumoniae and A.baumanii as compared to the control ( Figure 6). It has comparable to the neat 4PCB hydrogel antibiofouling effect towards the three bacteria and has better performance as compared to the neat 4PSB hydrogel. This is additional proof that the IPN approach is able to create novel polymeric materials by utilizing the performance of the constituent single networks. These results also confirm the expected from the FW-NFW study comparable "biological" performance of the PCB/PSB IPN with the one of the PCB single network due to the close NFW amounts they retain. P. Aeruginosa is a Gram-negative aerobic bacterium that has been identified a pathogen for humans and plants. It could be found in the environment, e.g., in soil a water, and can cause infections in the blood, lungs (pneumonia), or other parts of body after surgery.
K. pneumoniae is a Gram-negative facultative anaerobic bacterium that althou found in the normal flora of the mouth, skin, and intestines, can cause destruct changes to human and animal lungs if aspirated.
A. Baumannii is a Gram-negative bacterium that is a pathogen in humans, affect people with compromised immune systems. It is becoming increasingly important hospital-derived infection, as it is almost exclusively isolated from hospital enviro ments.
The results demonstrate that PCB/PSB IPN hydrogel reduces the growth of P. Ae ginosa, K. pneumoniae and A.baumanii as compared to the control ( Figure 6). It has com rable to the neat 4PCB hydrogel antibiofouling effect towards the three bacteria and h better performance as compared to the neat 4PSB hydrogel. This is additional proof t the IPN approach is able to create novel polymeric materials by utilizing the perf mance of the constituent single networks. These results also confirm the expected fr the FW-NFW study comparable "biological" performance of the PCB/PSB IPN with one of the PCB single network due to the close NFW amounts they retain. Biofilm formation usually starts with the formation of conditioning film, which the biomaterial surface fouled by proteins and other biological compounds present the body. The conditioning film triggers the cascade of the biofilm formation stag which include irreversible bacterial attachment, co-adhesion, and extracellular mat Biofilm formation usually starts with the formation of conditioning film, which is the biomaterial surface fouled by proteins and other biological compounds present in the body. The conditioning film triggers the cascade of the biofilm formation stages, which include irreversible bacterial attachment, co-adhesion, and extracellular matrix synthesis, followed by microbial colony formation and maturation. The result is a thick biofilm that can disperse planktonic bacteria [28]. The recent efforts in creating antibiofouling materials are devoted to the preparation of surfaces that reduce the non-specific adsorption of proteins and cells, minimizing in this way the conditioning film formation and thus preventing the biofilm formation. The antibiofouling properties are defined by the ability of the materials to form a hydration layer on their surfaces, which is an intrinsic cause for non-fouling surfaces [29]. The tightly bound water molecules form a physical as well as energetic barrier to prevent protein adsorption on the surface. Protein adsorption is known to take place through the expulsion of water molecules from both the surface and the protein in order to reduce the free energy barrier arising from the dehydration entropic effect [29]. The strength of surface hydration is defined by the physicochemical properties of the polymeric materials, e.g., surface chemistry, surface packing (i.e., film thickness, packing density, and chain conformation), etc. In addition to surface hydration, chain flexibility also plays an important role in protein resistance [29]. When protein approaches the polymer surface, the compression of the polymer chains causes steric repulsion to resist protein adsorption due to the unfavorable decrease in entropy. The best non-fouling ability of polymers can only be achieved when surface hydration and steric repulsion work together. Owing to their net charge neutrality, zwitterionic polymers exhibit ultralow protein adsorption (proteins are usually negatively or positively charged) [30]. In addition, zwitterionic polymers are more hydrophilic than poly(ethylene glycol) (PEG), which is the golden standard for non-fouling performance due to their stronger interaction with water molecules via ionic solvation, instead of the hydrogen bonding employed by PEG. Thus, the IPN approach contributes to the steric repulsion of the proteins, additionally enhancing in this way the antifouling properties of the newly developed material.

Cytotoxicity
Fast-growing human embryonic stem cells (Lep 3) were used to determine the cytotoxicity of the PCB/PSB IPN hydrogel. The Lep 3 cell viability in the presence of the PCB/PSB hydrogel is higher (~117%) as compared to the control (~100%) (Figure 7) after 24 h cultivation. Similar behavior is observed for the 4PSB sample, and thus the behavior of the IPN hydrogels could be explained by the PSB component in its composition. Notably, 4PCB hydrogel at the same time shows comparable to the control cell viability.
The results indicate that the PCB/PSB IPN hydrogel is not cytotoxic material and allows cell proliferation-this makes it a suitable candidate for biomedical applications, e.g., for tissue engineering. The blood analysis of the three studied mice groups demonstrates that the mice  The blood analysis of the three studied mice groups demonstrates that the mice from the 1st and the 3rd groups, i.e., the ones where PCB/PSB IPN hydrogels were implanted, have blood parameter counts which are within the reference range of the analyzer's calibrations. Only the mice with polycon sutures showed white blood cell (WBC) counts at the upper limits of the reference range (Table 3). Table 3. Haematological data taken from male mice, representative of the 2nd group.

Parameter
Data Range Thus, the PCB/PSB IPN hydrogel shows a very good in vivo response compared to the widely used sutures such as the polyamide ones. These results were further confirmed by the histology study.

Histology
The operative wounds were macroscopically completely healed and hairy on the 30th day of the surgery. During the whole period of study, the mice from all three groups showed good general condition and appetite. The mice from the 2nd and 3rd groups also showed a rapid return of motility in the operated legs. The histological observations showed a very slight reaction in the 1st and the 3rd groups, i.e., the mice where PCB/PSB IPN hydrogels were implanted, namely fibrous capsule formation and absence of inflammatory cells as lymphocytes, plasmocytes, polymorphonuclear leukocytes or macrophages ( Figure 8A,C). No signs of immune cell reaction as a part of the foreign body response were found at the 30th day in these groups. In contrast, the 2nd group with resolvable threads had marked infiltration with polymorphonuclear leukocytes and lymphocytes ( Figure 8B).
In all three mice groups, a fibrous capsule was formed around either the implanted IPN hydrogels or the surgical polycon node. It was more pronounced in the 1st and the 3rd groups and very thin in the 2nd group, where abundant lymphocytes were presented. On the 30th day, capsulation-linked small blood capillaries in the 1st and the 3rd groups were noticed and the nearby muscles were morphologically preserved-no inflammation, hyalinization of fibers, or atrophy was noticed. According to the observations, the local tissue application of the PCB/PSB IPN hydrogel led to normal activation and duration of the wound-healing process of skin and muscles and preserved leg motility with moderate fibrotic tissue formation. No signs of inflammation, necrosis, calcification, or other pathologies were found when PCB/PSB IPN hydrogel was implanted.
servations showed a very slight reaction in the 1st and the 3rd groups, i.e., the mice where PCB/PSB IPN hydrogels were implanted, namely fibrous capsule formation and absence of inflammatory cells as lymphocytes, plasmocytes, polymorphonuclear leukocytes or macrophages ( Figure 8A,C). No signs of immune cell reaction as a part of the foreign body response were found at the 30th day in these groups. In contrast, the 2nd group with resolvable threads had marked infiltration with polymorphonuclear leukocytes and lymphocytes ( Figure 8B). In all three mice groups, a fibrous capsule was formed around either the implanted IPN hydrogels or the surgical polycon node. It was more pronounced in the 1st and the 3rd groups and very thin in the 2nd group, where abundant lymphocytes were presented. On the 30th day, capsulation-linked small blood capillaries in the 1st and the 3rd groups were noticed and the nearby muscles were morphologically preserved-no inflammation, hyalinization of fibers, or atrophy was noticed. According to the observations, the local tissue application of the PCB/PSB IPN hydrogel led to normal activation and duration of the wound-healing process of skin and muscles and preserved leg motility with moderate fibrotic tissue formation. No signs of inflammation, necrosis, calcification, or other pathologies were found when PCB/PSB IPN hydrogel was implanted.
Thus, the in vivo studies on the PCB/PSB IPN hydrogel behavior showed blood parameters within the reference range, capsule formation, normal wound healing process and even small blood capillaries formation around the 30th day without any adverse effects on the skin and muscles around the implantation site. These experiments proved the good in vivo biocompatibility of the PCB/PSB IPN hydrogel. The in vivo studies of PSB [18] and PCB [19] single networks hydrogels showed that the PCB/PSB IPN hydrogel is more effective and has better in vivo performance as compared to them. The implantation of PSB hydrogel resulted in a slight to moderate inflammatory reaction [18], while when the PCB hydrogel was implanted, only single lymphocytes were seen [19]. In contrast, such an inflammatory reaction to the implanted PCB/PSB IPN hydrogel was absent, and they exhibited better biocompatibility. Similar conclusions could be drawn for the foreign body reaction-for PSB a few macrophages were observed in the implan- Thus, the in vivo studies on the PCB/PSB IPN hydrogel behavior showed blood parameters within the reference range, capsule formation, normal wound healing process and even small blood capillaries formation around the 30th day without any adverse effects on the skin and muscles around the implantation site. These experiments proved the good in vivo biocompatibility of the PCB/PSB IPN hydrogel. The in vivo studies of PSB [18] and PCB [19] single networks hydrogels showed that the PCB/PSB IPN hydrogel is more effective and has better in vivo performance as compared to them. The implantation of PSB hydrogel resulted in a slight to moderate inflammatory reaction [18], while when the PCB hydrogel was implanted, only single lymphocytes were seen [19]. In contrast, such an inflammatory reaction to the implanted PCB/PSB IPN hydrogel was absent, and they exhibited better biocompatibility. Similar conclusions could be drawn for the foreign body reaction-for PSB a few macrophages were observed in the implantation site [18], which were scarcely seen in the case of PCB [19] and absent in the case of the PCB/PSB IPN.

Conclusions
In this study, zwitterionic IPN based on PCB and PSB was synthesized and its smart response to changes in three different environmental parameters was demonstrated. Most of the smart materials reported so far are usually able to respond to changes in one or two external stimuli, and thus the development of material that is responsive to three "biological" parameters is unique and opens a wide area for different applications. The study has also demonstrated the potential and versatility of the IPN approach as a method for creating smart materials. The newly synthesized PCB/PSB IPN successfully inherits the smart properties of both constituent single networks and simultaneously responds to variations in temperature, pH, and salt concentration. The PCB component imparts the novel PCB/PSB IPN pH sensitivity as well as biocompatibility, demonstrated by antibiofouling activity, non-cytotoxicity, and very good in vivo performance, while the PSB component imparts temperature responsiveness. Both components, PCB and PSB define the antipolyelectrolyte behavior of the IPN; however, the IPN exhibits its own unique way of salt concentration dependence, which is linear in contrast to the single PSB and PCB salt responsiveness which SRs level off at higher salt concentrations. The PCB/PSB IPN hydrogel exhibits antibiofouling activity against some commonly spread bacteria such as P. Aeruginosa, A. Baumanii and K. Pneumoniae, it is non-cytotoxic and possesses very good in vivo biocompatibility which makes it a unique smart material that can find many applications in different areas, including medicine, pharmacy, smart materials for sensors, etc.
were diluted with physiological solution to match the turbidity of a 0.5 McFarland standard (approximate cell density 1.5 × 10 8 mL). Dry PCB/PSB IPN samples (~0.1 g), sterilized in ethylene oxide, were placed in an Eppendorf tube containing 2 mL bacterial inoculum. Bacterial inoculum without hydrogel was used as a control. All Eppendorf tubes were left at 37 • C for 24 h. Thereafter, the liquid was gently removed and the hydrogels in the tubes were washed twice with freshly prepared PSB in order to remove the loosely attached bacteria. For enumeration of bacteria, 1 mL of physiological solution was added to each Eppendorf tube followed by ultrasound sonication for 20 min. Several dilutions were completed prior to plate the bacterial suspension on Muller Hinton agar plates. After 24 h at 37 • C, the number of colonies on the agar plates was counted to determine the concentration of live bacterial cells.

Cytotoxicity
The cell viability of human embryonic fibroblasts Lep-3 when in a contact with PCB/PSB IPN was assessed by MTS test according to the manufacturer's instructions. The cells were grown as monolayer cultures in Dulbecco's modified Eagle's medium (DMEM), supplemented respectively with 10% fetal bovine serum, 100 U mL −1 penicillin, and 100 µg mL −1 streptomycin at 37 • C in a humidified CO 2 incubator (Thermo Scientific, Hepa class 100). For routine passages, the cells were detached using a mixture of 0.05% trypsin and 0.02% ethylenediamine tetra acetic acid (EDTA).
The MTS test was performed as the dry PCB/PSB networks were first swollen in sterilized distilled water for 24 h and subsequently soaked in DMEM medium for several hours. Then, they were cut into small uniform pieces and transferred into a 24-well tissue culture plate where they were sterilized by UV radiation for 2 h. A cell suspension with a density of 7.5 × 10 4 cells/well was added to the plate. The cell-seeded hydrogels were maintained at 37 • C under 5% CO 2 for 24 h, after which the 100 µL MTS reagent was added to each well. Finally, the optical density of the obtained solution was measured at 540 nm. As a control, a suspension of cells grown in non-modified medium without the presence of PCB/PSB IPN hydrogel was used. The cell viability (%) in the presence of PCB/PSB IPN hydrogels was determined relative to the cell viability found in the control. All results are reported as the mean ± standard error of the mean values obtained from three independent measurements determined using one-way analysis of variance (ANOVA).

In vivo Biocompatibility Test. Animal Design and Implantation of Biomaterials Mouse Model
The in vivo experiments were consistent with the regulations of local institutional, Bulgarian National Regulation № 20/01.11.2012 regarding laboratory animals and animal welfare and European legislation.
In the in vivo experiments were used 12 male albino laboratory mice, 4-month-old weighting about 25 g each. They were divided into 3 groups: • 1st group-mice with dorsal subcutaneous implantation of PCB/PSB IPN hydrogel with 2-3 mm in diameter • 2nd group-mice with experimentally made muscle lacerations in the region of m. biceps femoris and m. semitendinosus and surgical sutures with polycon semi-elastic surgical thread • 3rd group-mice with implantation of PCB/PSB IPN hydrogel pieces which are 2-3 mm in diameter in the region of fossa poplitea between m. biceps femoris and m. semitendinosus.
After 4 week evaluation period, all animals were humanely euthanized and materials for histology and hematology were taken.

Surgical Procedures
Incorporation of the hydrogels in PBS was carried out after sterilization with ultraviolet irradiation (UV) for 4 h and aseptical conditions. Standard surgical access through the skin-dermis and hypodermis in the dorsal region in the area of the withers (in the 1st group) and to fossa poplitea between biceps femoris and semitendinosus muscles in the 2nd and 3rd groups was made. Implants were situated in the 1st and 3rd groups and tissues were then surgically restored, in the 2nd group were made several polycone sutures and nodes for comparison of the local tissue reaction.

Hematology
Blood samples from each experimental groups were taken in sterile vacuum blood collection tubes via cardiac puncture. Hematological analyses were performed by automatic xematology analyzer "Mindray BC-2800 Vet", Mainland China. The values of the counts of routine blood parameters were checked in all groups.

Histology
Tissue samples (0.6 cm 3 in size) with incorporated implants or sutures were routinely fixed in 10% buffered formalin, dehydrated in ethanol, and embedded in paraffin.
Tissue samples (0.6 cm 3 in size) with incorporated implants or sutures were routinely fixed in 10% buffered formalin, dehydrated in ethanol, and embedded in paraffin. Tissue sections (3-5 µm thick) were stained in hematoxylin and eosin and examined by light microscope (Leica DM 5000B, Wetzlar, Germany). The sections were scored for the presence of fibrotic tissues, vascularization, inflammatory cell reactions, and foreign body response.

Informed Consent Statement: Not applicable.
Data Availability Statement: The raw/processed data required to reproduce these findings cannot be shared at this time as the data also form part of an ongoing study.