*2.4. Phospho-Mimetic T4*/*5D and T158*/*159D Recombinant Proteins Lack FMN*

GOX activity requires FMN as a cofactor [30] and therefore a change in FMN content could explain the altered activities of our phospho-mimetic recombinant GOX proteins (Tables 2 and 3). By measuring the ratio of the absorbance between protein (at 280 nm) and FMN (at 450 nm) (A280nm/450nm), it was possible to determine the presence of FMN in the purified GOX proteins used to measure enzymatic activities. An example of absorption spectra of WT and T158/159 mutated GOX proteins highlights the typical spectra associated with FMN that was missing in the GOXT158/159D forms thereby indicating an absence of cofactor (Figure 1). Similar absorption spectra were carried out for each recombinant protein and used to calculate the A280nm/450nm ratios given in Table 4. It can be seen that all FNR-containing GOXWT proteins had a similar low A280nm/450nm ratio of between 8 and 9. On the other hand, *At*GOX1T158D, *At*GOX2T158D and *Zm*GO1T159D exhibited higher A280nm/450nm values of 23, 20.5 and 19.5, respectively (Table 4) since they all lacked a significant typical FMN absorption signature (Figure 1), thus indicating an absence (or an extremely reduced amount) of cofactor.

**Figure 1.** Absorption spectra of recombinant *At*GOX1, *At*GOX2 and *Zm*GO1 and their T158/T159 phospho-site mutated forms.

In this way, *At*GOX1T4D, *At*GOX2T4D and *Zm*GO1T5D were seen also to lack FMN since they had high A280nm/450nm ratios of 28.3, 36.5 and 25.7, respectively (Table 4). Therefore, there appeared to be a good correlation between FMN content and GOX activity for these mutated forms. However, this was not always the case. *At*GOX1T158V, *At*GOX2T158V and *Zm*GO1T159V as well as *At*GOX1T265D, *At*GOX2T265D and *Zm*GO1T266D appeared to have a normal FMN content (Table 4) even though they exhibited a strong decrease of kcat (Table 2). This was also seen for *At*GOX2S212A, *At*GOX2S212D, *Zm*GO1S213A and *Zm*GO1S231D which showed normal A280nm/450nm ratios (Table 4) but reduced kcat values (Table 2).

**Table 4.** A280/450nm ratios of recombinant *At*GOX1, *At*GOX2 and *Zm*GO1 proteins.


Mean values ± SD from three independent experiments. Statistical significance was determined by a Student's *t*-test. Values in bold and marked by an asterisk were significantly different compared to the corresponding WT protein (*p* < 0.05).

#### **3. Discussion**

Regulation of the photorespiratory cycle is still poorly understood even if phosphoproteomics studies have indicated that all photorespiratory enzymes except glycerate kinase can be phosphorylated [24]. In this context, GOX phosphorylation could be important since several phospho-sites have been reported and the phosphorylated residues have been conserved during the evolution of land plants (Figure S2). In this study, we explored the role of GOX phosphorylation by analyzing the kinetic parameters of both phospho-dead (to control the importance of the original residue) and phospho-mimetic recombinant Arabidopsis (C3-plant) and maize (C4-plant) photorespiratory GOX proteins.
