**4. Discussion**

The polyelectrolyte complex in the aqueous medium interacts and self-assembles by electrostatic interactions. Therefore, the development of new stable formulations, constituting polyelectrolyte, requires an in-depth characterisation of the interactions to obtain biodegradable and biocompatible products with unique properties, without losing the polymers inherent characteristics. In the present work, efficacies analytical approaches were set up to characterize the HA/Ch complex and to evaluate its stability toward enzymes.

The interaction between HA and Ch was affected by the structural peculiarity of each polysaccharide. With regards to HA, the chemical structure was well-defined, meanwhile for Ch, the DS and DA were variable. Generally, the proportion of glucosamine residues in chitosan had a significant influence on their various properties. In particular, the distribution of N-acetyl groups along the biopolymer chain might influence the solubility of the polymer and the inter-chain interactions, due to the hydrogen bonds and the hydrophobic character of the acetyl group. Moreover, the derivatization of chitosan with alkyl groups induced high mobility side chains, causing them to be highly hydrated. It follows that the DS and DA of Ch are important parameters for the property of the polysaccharide in solution and for the interaction with HA. A schematic representation of how Ch and HA interact to form a complex is given in Figure S1.

A quantitative <sup>1</sup>H NMR method was applied to determine the DA and DS of Ch. The Ch sample with DA and the DS values of 7.5% and 59%, respectively, is soluble in water at neutral pH and it presents a low viscosity, as compared to the same molecular weight of HA, as shown by the HP-SEC-TDA results (Table 1) and the viscosity curves (Figure 6a). HA was approximately 6 times more viscous in terms of intrinsic viscosity than Ch, and the hydrodynamic radius for the negative charge polysaccharide was double, as compared to Ch, although the molecular weights were quite similar. This behavior was confirmed by the DOSY NMR experiment (Table 3). Indeed, the average D values of HA were lower than those of Ch, indicating a lower mobility due to higher viscosity. When the two polysaccharides were mixed, anaverage molecular weight distribution between HA and Ch was obtained, which maintained a high intrinsic viscosity value (Figure 2, Table 2).

The homogeneity of the complex was demonstrated by the Zp measurements (Table 5) —one Zp value was observed and through the comparison of the results, the HA/Ch complex Zp was about –23 mV, lower than that obtained for the HA alone (−34 mV), indicating that the negative charges of HA were partially reduced in the presence of Ch, due to the phenomena of secondary interaction between the polysaccharides. The strict interaction between HA and Ch was also demonstrated by DOSY:the D values of HA and Ch in the complex were similar (Table 3).

It clearly appears that by mixing the two polysaccharides, an interesting modulation of the viscoelastic properties could be easily obtained (Figure 6b). The mechanical spectra of the solutions prepared using only HA or Ch represent two different behaviors—the HA solution had a relevant viscoelastic character, whereas the Ch sample presented a more evident liquid/elastic character. Finally, the HA/Ch solution showed the highest values of both elastic and viscous moduli, with a shift toward higher frequency of the crossover point (Table 4).

SEM analysis on the Ch/HA composite supported the data that suggest an intimate mixing of the components, revealing characteristic features of the individual polysaccharide, such as lamellae, fibers, and pores.

We demonstrated that HA and Ch, solubilized together, can generate secondary interactions that modified the chemico-physical properties of polysaccharides. To verify if these phenomena could affect the biocompatibility, the stability toward enzymatic actions was evaluated, comparing the results obtained from a single component preparation and the complex of HA/Ch. While the HA was fully degraded by Hyaluronidase and was not hydrolyzed by Lysozyme, Ch was mildly degraded by Hyaluronidase, and was fully hydrolyzed by Lysozyme. Using the Hyaluronidase and Lysozyme complex with each polysaccharide, the Mw values were the same for those obtained with the specific enzymes. This meant that there was no cooperative effect in the degradation process. Analyzing the HA/Ch complex, HA hydrolysis was not modified by Ch, wheras Ch appeared more resistant to enzyme hydrolysis in presence of HA. Our hypothesis suggests that the penetration of the enzymes to the network HA/Ch, and in particular, the availability of Ch molecules to

the active site of the enzymes were slightly hindered; however, but the hydrolysis processes was maintained.
