*3.1. NMR Spectroscopy Study*

The NMR spectra of the HA sodium salt were obtained as reported in Materials and Methods. The observed chemical shifts were aligned with the literature data [8] and the integral of the signals (3H δ 2.08 CH3(CO-NH); 2H δ 4.5–4.62) showed the correct ratio of protons resonances. In total 16H (3 + 11 + 2). For PLA (3H δ 1.4–1.8 CH3;1H δ 5.2–5.4 CH, with a correct ratio between the resonance integrals 3H:1H). A typical spectrum of HA-PLA is reported in Figure 2.

δ δ

Δ

δ δ δ

δ δ

δ

δ

**Figure 2.** NMR spectrum for HA-PLA DAC® sterilized HA sodium salt–PLA at a concentration of 7.5 mg/mL in D2O-PBS obtained as reported in Materials and Methods.

The samples of HA-PLA DAC® sterilized were preliminary studied in D2O as to reduce the large water resonance dissolving at a 7.5 mg/mL concentration. The NMR resonance assignment was achieved by literature data distinguishing between HA and HA-Na salt. A mixture of D2O-DMSO was also examined. In fact, in the NMR spectra, a marked dependence of the PLA resonances upon the different solvent has been observed. In the solvent mixtures DMSO-d6–D2O (DMSO-d6:D2O = 9:1) or (DMSO-d6:D2O = 8:2) an increase in intensity of the PLA resonances with respect to the HA-Na resonances has been observed at the same concentration, with an increase of the intensity of the HA resonances upon the increase of the D2O ratio (reported in Supplementary Materials, Figure S1).

During this study, the possibility of micellar aggregation in solvents was also considered. In fact, in the mixtures of water and DMSO at different ratios, it is easy to detect a probable formation of micelles. PLA is much more soluble in the non-aqueous system while HA strongly prefers the aqueous one. This effect has been already reported with copolymers like HA-PLA [17,33].

The stability of the presence of the conjugate between HA and PLA needed to be proved after sterilization. In fact, there was the possibility that the spectrum of Figure 2 would result as a simple additive overlap between the NMR spectrum of the HA and PLA. In our case, the DOSY experiment of HA-PLA after sterilization showed a unique diffusive front and, thus, confirmed the presence of a unique product (HA-PLA) chemically linked together. The observed diffusive front was 8.91 × 10−<sup>11</sup> m<sup>2</sup> s −1 .

Moreover, DOSY experiments reported in Figure 3 for sterilized HA-PLA DAC® obtained as reported in Materials and Methods revealed that no differences are present in hydrodynamic volumes. This indicates there is only one population with the same diffusion coefficient and a unique molecular weight for the HA-PLA copolymer. Particularly the sharp shape of the peak on the left side of the DOSY representation indicates that nearly a unique species is present in the solution for the observed diffusive front. A weak diffusion is visible at 0 ppm probably due to a trace of the silyl moiety of DSS bound to the macromolecule, due to hydrophobicity of PLA moiety.

−

−

− − **Figure 3.** Diffusion ordered NMR (DOSY) experiment of the sample HA-PLA DAC® in D2O-PBS after sterilization. The front of diffusion indicated a Diffusion coefficient D = 8.91 × 10−<sup>11</sup> m<sup>2</sup> s −1 . A weak resonance is reported at 0 ppm probably due to the silyl moiety of DSS bound to the macromolecule.

Thus following the recommendations [29] of using standards with sufficient molecular homogeneity to calibrate the MW, a series of DOSY experiments were conducted on separate solutions of HA fragments with different MWs (13 kDa, 50 kDa, 208 kDa), to check the linearity of hydrodynamic radii and the MWs in hyaluronans. The results are reported in Figure 4 with a final log-log plot in Figure 5 reporting the good correlation found. An investigation was also performed to ascertain if an interference of macromolecules of different sizes in the measure of diffusion could occur. These results are reported in more detail in Supplementary Materials and in a dedicated section of the Discussion.

**Figure 4.** DOSY experiments of three different HA polymer standards in D2O with different MWs. (**a**) 13 kDa; (**b**) 50 kDa; (**c**) 208 kDa. To avoid confusion due to possible interferences, the experiments were run on single polymer samples. The diffusion coefficient obtained are used for the calibration curve in the log-log plot of the Figure 5.

**Figure 5.** Calibration curve of the diffusion coefficient of fragments of HA with different MWs: 13 kDa; 50 kDa; 208 kDa. The diffusion of the trace of protonated water molecule (1% in D2O) is also included with a correlation of 0.999.

The calibration curve of standards is reported in the following Figure 5, with a correlation coefficient of 0.999. The inspection of the results of the calibration curve indicated that the experiment is able to distinguish between fragments with MW lower than 13 kDa (from 5.0 to 13 kDa), in line with the expected range of MWs in the degradation experiments. Furthermore, the DOSY results included the resonance of protons of residual water, whose diffusion value displayed a good alignment with those of the fragments of HA observed. A more detailed discussion of the diffusion results is reported in the related paragraph in the Discussion section.

Moreover, a detailed study of the possible interferences between these types of polymers was carried out using a mix of the 13 kDa with the 50 kDa ones and of the 13 kDa with the 208 kDa standard samples. In addition, these results are reported in Supplementary Materials and in the related paragraph in the Discussion section.
