**2. Results**

## *2.1. Inhibition Assay*

The inhibitory e ffect of *Ulva ohnoi* on the digestive proteases of gilthead seabream, Senegalese sole and seabass is shown in Figure 1. Dose-response inhibition curves were obtained by measuring the reduction in proteolytic activity on a standardized fish intestinal extract following incubation with di fferent volumes of *Ulva* extract. The results confirmed the presence of protease inhibitor in crude *Ulva* (CR-*Ulva*) able to inhibit up to 70% of Senegalese sole's proteolytic activity, and by 65% of protease activity in the other two fish species. It was also found that the amount of *Ulva* able to cause the inhibition of 50% digestive protease activity (IC50) ranged from 0.6 to 0.9 mg *Ulva* per unit of proteolytic activity (UA). On the other hand, autoclaved *Ulva* (heat treated; HT-*Ulva*) reduced significantly its inhibitory capacity (less than 20% inhibition in all cases) compared to CR-*Ulva* (*p* < 0.05).

**Figure 1.** Dose-response inhibition plot of *S. aurata* (**A**), *S. senegalensis* (**B**) and *D. labrax* (**C**) intestinal proteases activity following the incubation with increasing concentrations of crude (CR-*Ulva*) and heat-treated *Ulva* extracts (HT-*Ulva*). Each point represents the mean of triplicates ± SD. IC50: mg of *Ulva* needed to cause 50% protease inhibition.

Zymograms of fish digestive proteases after incubation with *Ulva* extract are shown in Figure 2. The effects of protease inhibitors were compared to a control without *Ulva* (lane 1). Noticeable reduction in the intensity of the active fractions (with proteolytic activity) was evidenced after incubation with *Ulva* (lanes 2 to 4). Differences in the inhibitory effect were observed among the three fish species. In the case of gilthead seabream, a progressive decrease in the intensity of all active proteases was observed as the *Ulva* concentration increased, whereas protease inhibition in the other fish was selective for some specific active fractions.

**Figure 2.** Substrate-SDS-PAGE (sodium dodecyl sulphate polyacrylamide gel electrophoresis) obtained after incubation of *S. aurata*, *S. senegalensis* and *D. labrax* intestinal extracts with different concentrations of *Ulva* extract. Lane 1: control without inhibitor (distilled water was used instead of *Ulva* extract); lane 2: 500 μg *Ulva* per unit of proteolytic activity (UA); lane 3: 1000 μg *Ulva* per UA; lane 4: 1500 μg *Ulva* per UA. The molecular mass markers (kDa) were phosphorylase b (94), bovine serum albumin (66), ovalbumin (45), carbonic anhydrase (30), and soybean trypsin inhibitor (20).

#### *2.2. In Vitro Casein Hydrolysis*

Proteinograms of casein hydrolyzed by fish digestive enzymes are shown in Figure 3. All the main casein fractions (34, 26 and 23 kDa), corresponding to α, β and κ subunits, were hydrolyzed after 30 min.

**Figure 3.** The time-course of in vitro proteolysis of casein by *S. aurata*, *S. senegalensis* and *D. labrax* digestive proteases in the absence of any inhibitory extract. Images show SDS-PAGE hydrolysis patterns obtained at different sampling times (0, 30, and 90 min). Numbers on the left of the electrophoresis gel stand for the molecular mass of proteins fractions (kDa).

Figure 4 shows the time-course protein hydrolysis of casein by the fish digestive proteases in the presence of *Ulva* extracts. Overall, CR-*Ulva* hampered the capacity of fish proteases to hydrolyze casein. The fate of the casein fractions throughout the in vitro assay was partial and less marked compared to the same assay carried out without *Ulva* (Figure 3). It was also observed that the inhibitory effect was dose-dependent, given that the higher CR-*Ulva* concentration in the in vitro assay, the lower the hydrolytic capacity of fish digestive enzymes against casein. The inhibitory effect of *Ulva* extracts on digestive proteases was reduced remarkably owing to thermal treatment, as revealed by the proteinograms shown in Figure 4 (lanes at the right). Thus, heat-treated seaweed extracts (HT-*Ulva*) were affected casein hydrolysis to a lesser extent than untreated *Ulva* (CR-*Ulva*), in such a way that none of the casein subunits remained after 90 min in any of the assays carried out with HT-*Ulva*. However, intraspecific and dose-dependent differences were found, and thus increasing concentrations of HT-*Ulva* also reduced casein hydrolysis, not least for seabream and seabass digestive proteases.

**Figure 4.** The time-course of in vitro proteolysis of casein by *S. aurata* (**A**), *S. senegalensis* (**B**) and *D. labrax* (**C**) digestive proteases in the presence of increasing concentrations of crude (CR-*Ulva*) and heat-treated (HT-*Ulva*) *Ulva*. The images show SDS-PAGE hydrolysis patterns obtained at different sampling times (0, 30, and 90 min). The numbers on the left of the electrophoresis gels stand for the molecular mass of the protein fractions (kDa).

The estimated values of the coefficient of protein degradation (CPD) and total amino acids released after the in vitro hydrolysis of casein are given in Table 1. The CR-*Ulva* lowered the rate of protein degradation compared to the assay performed without *Ulva* (*p* < 0.05). The lowest CPD values were observed for the highest concentrations of CR-*Ulva* in the reaction mixture (1500 μg per UA). In parallel, the amount of free amino acids released was also affected by the interaction of *Ulva* extracts with fish proteases, especially for highest concentrations of CR-Ulva. The impact of HT-*Ulva* on *S. aurata* and

*S. senegalensis* digestive protease activity was negligible, since neither the CPD nor free amino acids that were released were different from the controls. However, a certain residual inhibitory effect of HT-*Ulva* was observed on *D. labrax* enzyme extracts.

**Table 1.** The values of the estimated coefficient of protein degradation (CPD) and total amino acids released measured after 90 min of in vitro hydrolysis of casein by fish digestive proteases in the presence of different concentrations (0, 500, 1000 and 1500 μg *Ulva* per UA) of crude (CR-*Ulva*) and heat-treated (HT-*Ulva*) *Ulva ohnoi*.


The values are mean ± SD of triplicate determinations. Within each fish species, the values in the same column with different lowercase letters indicate significant differences (*p* < 0.05) owing to *Ulva* extract. a,b,c,d the superscripts indicate the treatments with significant differences.

#### *2.3. Thermal Stability of Protease Inhibitors*

The influence of a thermal treatment on the inhibitory capacity of *Ulva* on trypsin activity was evaluated. A noticeable reduction in the inhibitory effects of *Ulva* was observed as the temperature increased (Figure 5). Protease inhibitors of the *Ulva* extracts were stable at room temperature, but their inhibitory effect significantly reduced (*p* < 0.05) above 40 ◦C. The exposure of *Ulva* extract at 80 ◦C for 30 min reduced their inhibitory activity up to 50% compared to the untreated controls.

**Figure 5.** The effects of thermal treatments on the inhibitory effect of *Ulva* on trypsin activity. \* Denote significant differences among treatments (*p* < 0.05).

The existence of proteinaceous complexes consisting of *Ulva* protease inhibitors and trypsin is revealed in Figure 6. The commercial trypsin showed three protein fractions (24, 24.4, and 26 kDa, lane 1), whereas the protein pattern of the *Ulva* extract contained several protein fractions (lane 2). The incubation of trypsin with crude *Ulva* before SDS-PAGE (sodium dodecyl sulphate polyacrylamide gel electrophoresis) yielded a characteristic protein profile in which the 24 and 26 kDa fractions of trypsin disappeared. Instead, a new 31 kDa fraction appeared (lanes 3 to 5), which was not present either in trypsin or in *Ulva*. However, this 31 kDa protein fraction did not appear when trypsin was incubated with heat-treated *Ulva* (lanes 6 to 8).

**Figure 6.** SDS-PAGE carried out to detect trypsin-*Ulva* inhibitor complexes. Lane 1 = trypsin (1 mg mL−1). Lane 2 = *Ulva* (0.1 g mL−1). Lanes 3 to 5 = trypsin preincubated with CR-*Ulva* during 0, 30 and 60 min. Lanes 6 to 8 = trypsin pre-incubated with heat-treated *Ulva* during 0, 30 and 60 min.

### *2.4. Kinetic Parameter's*

The data obtained for the kinetic studies performed with trypsin and chymotrypsin are shown in Figure 7. The results demonstrated that *Ulva* protease inhibitors yielded a potential mixed type inhibition, as revealed by the decrease in Vmax as Km increased, compared to the apparent kinetic parameters obtained for both trypsin and chymotrypsin activities in the absence of *Ulva* (Table 2).

**Figure 7.** Lineweaver–Burk plots showing the inhibition of trypsin (**A**) and chymotrypsin (**B**) by *Ulva* protease inhibitors (a: 0; b: 200; c: 300; d: 400 and e: 500 μg *Ulva* per μg enzyme).

**Table 2.** Kinetic analysis of trypsin and chymotrypsin enzymes in the presence of *Ulva*.


Km: Michaelis constant; Vmax: maximum rate of reaction; pNA: para-nitroaniline.
