Statistics 24

All values are expressed as the mean ± standard deviation (SD). Homoscedasticity of data were verified by Levene test, then, data were examined using one-way analysis of variance followed by Tukey's test or Games-Howel test for unpaired results (including controls). All statistical analyses were conducted using SPSS Statistics 24 software. Differences were considered to be significant for values of P ≤ 0.05.

To verify the role of NOS in NO^⋅ production by beewolf eggs, we used an inhibition assay (Willmot et al., 2005 (link)). Since L-arginine is the substrate for NO^⋅ production by NOS, we injected either an inhibiting L-arginine analog or, for controls, a non-inhibiting enantiomer into freshly laid beewolf eggs. Chemicals were dissolved in 0.1 mol/l phosphate buffer pH 7.4. Using a microinjector (see above) eggs were injected with about 0.2 µl of 1.5 mol/l solutions of (1) the competitive inhibitor Nω-nitro-L-arginine methylester (L-NAME, Sigma-Aldrich, USA) (experimental group, N = 14), or (2) the non-inhibiting Nω-nitro-D-arginine methylester (D-NAME, Sigma-Aldrich, USA) (control group 1, N = 9) or (3) not injected at all (N = 14, control group 2). Each egg of the three groups was placed individually in the lid of a reaction tube with an iodide-starch solution as described above and incubated for 24 hr at 28°C. Then NO^⋅ production was assessed by measuring absorbance of the solution with a photometer (Implen Nanophotometer) at 590 nm. Statistical comparison of the groups was conducted using Mann-Whitney U-tests with correction after Holm (Holm, 1979 ) (SPSS Statistics 24).

The experimental data obtained in the thermal inactivation of commercial lipases were fitted to modeled theoretical curves and the kinetic parameters were determined using a non-linear regression analysis by SPSS Statistics 24 statistical analysis system software package.
Physical and statistical criteria were considered to establish the best fit of the models studied. Physical criterion is the absence of negative estimated parameters [33 (link)]. The statistical criteria include the coefficient of determination (r²), chi-square (χ²) (Equation (9)), and the standard error of means (SEM) (Equation (10)).
$${\chi }^2=\frac{\sum {\left(y_{exp}-y_{pred}\right)}^2}{\left(m-p\right)}$$
$$SEM=\frac{\sum {\left(y_{exp}-y_{pred}\right)}^2}{\sqrt{m}}$$
where m is the number of observations, and p is the number of parameters. y_exp and y_pred are the experimental and predicted residual enzyme activities, respectively. The model with the lowest χ² and SEM, and higher r² for the residual activity is the best choice from a statistical point of view.