A novel tetratricopeptide-repeat protein, TTP1, forms complexes with glutamyl-tRNA reductase and protochlorophyllide oxidoreductase during tetrapyrrole biosynthesis

Abstract The biosynthesis of the tetrapyrrole end-products chlorophyll and heme depends on a multifaceted control mechanism that acts primarily at the post-translational level upon the rate-limiting step of 5-aminolevulinic acid synthesis and upon light-dependent protochlorophyllide oxidoreductase (POR). These regulatory processes require auxiliary factors that modulate the activity, stability, complex formation, and subplastidal localization of the relevant proteins. Together, they ensure optimal metabolic flow during the day and at night. As an Arabidopsis homolog of the POR-interacting tetratricopeptide-repeat protein (Pitt) first reported in Synechocystis, we characterize tetrapyrrole biosynthesis-regulating tetratricopeptide-repeat protein1 (TTP1). TTP1 is a plastid-localized, membrane-bound factor that interacts with POR, the Mg protoporphyrin monomethylester cyclase CHL27, glutamyl-tRNA reductase (GluTR), GluTR-binding protein, and FLUORESCENCE IN BLUE LIGHT. Lack of TTP1 leads to accumulation of GluTR, enhanced 5-aminolevulinic acid synthesis and lower levels of POR. Knockout mutants show enhanced sensitivity to reactive oxygen species and a slower greening of etiolated seedlings. Based on our studies, the interaction of TTP1 with GluTR and POR does not directly inhibit their enzymatic activity and contribute to the control of 5-aminolevulinic acid synthesis. Instead, we propose that TTP1 sequesters a fraction of these proteins on the thylakoid membrane, and contributes to their stability.

Fig. S2: PChlide accumulation after prolonged dark phases.The accumulation of PChlide was measured in 3-week-old plants grown under short-day conditions (A; SD, 10 h light/14 h dark, 120 µmoles photon m -2 s -1 , 21 °C, 60 % humidity) or continuous light (B; 24 h light, 120 µmoles photon m -2 s -1 , 21 °C, 60 % humidity) followed by 16 h of dark incubation.The PChlide concentrations are quantified relative to fresh weight (FW) in (A) and (B).Statistical significance was calculated by one-way ANOVA (Tukey post-hoc test, p < 0.05, n = 3).(B) The accumulation of PChlide was measured in CL-grown seedlings which were subjected to an extended dark phase of 16 h.Statistical significance was calculated by Kruskal Wallis Test (Bonferroni post-hoc test, p < 0.05, n ≥ 3).(C) Three-week-old seedlings, grown under CL were subjected to 16 h of dark incubation.The abundance of TBS proteins was analyzed by immunoblotting and RBCL was used as loading control.The relative intensities of the chemiluminescent signals of GluTR1 and PORB were quantified with Image J.   The indicated proteins were fused to the N-terminal (pNUB) or C-terminal (pDHB3.1)half of ubiquitin, and the different TBS enzymes and auxiliary factors were analyzed for interaction with TTP1.Successful co-transformation of both constructs was confirmed by growth on Leu-and Trp-deficient (-lt) media.In the case of an interaction, cells were able to grow on minimal media without His, Ura, Leu and Trp (hult).To reduce the incidence of false negative results, 20 mM 3-aminotriazole (3AT) was added.The empty pDHB3.1 vector was used as the negative control.Table S1: List of chloroplast localized TPR-proteins and comparison of the identified TPR-motif to the TPR-motif of Pitt.Overall, 22 TPR proteins with a putative localization in the chloroplast were identified.The localization is based on in-silico predictions, made by either ChloroP, Cell-PLoc 2.0 or LOCALIZER (marked with an asterisk, https://localizer.csiro.au/).The TPR motifs were verified by TPRpred (https://toolkit.tuebingen.mpg.de/tools/tprpred).The calculated P-value and probability for this prediction is given.The lengths of the identified TPR domain and the numbers of motifs were also calculated by TPRpred.The degrees of identity and similarity of the predicted amino-acid sequences of the TPR domains to the known TPR domain of Pitt were calculated with Needleman-Wunsch Global Alignment (Needleman and Wunsch, 1970).The presence of a transmembrane domain (TM) was predicted by using the TMHMM 2.0 server (https://services.healthtech.dtu.dk/service.php?TMHMM-2.0)

Fig. S3 :
Fig. S3: Abundance and activity of the photosynthetic complexes in two ttp1 lines.(A) Thylakoids of five-week-old Arabidopsis seedlings were isolated, solubilized with 1 % of ndodecyl-beta-maltoside and separated on a 4-12 % blue native polyacrylamide gel.For each sample an equivalent amount of Chl was loaded.(B) Photosynthetic parameters were analyzed by Pulse-amplitude modulation after a 15 min dark adaptation of the used leaves.(C) 77K chlorophyll florescence was measured in three biological and three technical replicates.The measurements were normalized to the fluorescence at 720 nm and the averaged values were plotted.

Fig. S4 :
Fig. S4: Phenotype of the ttp1-3 knockout mutant and two overexpression lines under different light conditions.All investigated lines were grown for three weeks under the indicated light conditions, including continuous light (CL), long day (LD), short day (SD) and fluctuating light (FL).

Fig. S5 :
Fig.S5: Split-ubiquitin analysis.The indicated proteins were fused to the N-terminal (pNUB) or C-terminal (pDHB3.1)half of ubiquitin, and the different TBS enzymes and auxiliary factors were analyzed for interaction with TTP1.Successful co-transformation of both constructs was confirmed by growth on Leu-and Trp-deficient (-lt) media.In the case of an interaction, cells were able to grow on minimal media without His, Ura, Leu and Trp (hult).To reduce the incidence of false negative results, 20 mM 3-aminotriazole (3AT) was added.The empty pDHB3.1 vector was used as the negative control.

Fig. S6 :
Fig. S6: Statistics of the data shown in Fig. 6.The statistics for data of figure 6 are summarized.
Fig. S7: Detection of reactive oxygen species in ttp1 knockout and knockdown lines.All analyses were performed on four-week-old Arabidopsis plants grown under SD conditions (10 h light / 14 h darkness, 120 µmoles photons m -2 s -1 , 21°C, 60% humidity).(A) Representative images of leaves stained with nitro-blue tetrazolium chloride or 3,3′-diaminobenzidine to detect superoxide anions or hydrogen peroxide, respectively, are shown.(B) Relative expression levels of the indicated ROS marker genes, which were grouped according to Woodson et al. 2015.The relative expression levels are given below the Table.Statistical significance was calculated by two-way ANOVA analysis (Tukey post-hoc test, p < 0.05, n = 4).