The glycine-rich domain of GRP7 plays a crucial role in binding long RNAs and facilitating phase separation

Abstract

Microscale thermophoresis (MST) is a well-established method to quantify protein-RNA interactions. In this study, we employed MST to analyze the RNA binding properties of glycine-rich RNA binding protein 7 (GRP7), which is known to have multiple biological functions related to its ability to bind different types of RNA. However, the exact mechanism of GRP7's RNA binding is not fully understood. While the RNA-recognition motif of GRP7 is known to be involved in RNA binding, the glycine-rich region (known as arginine-glycine-glycine-domain or RGG-domain) also influences this interaction. To investigate to which extend the RGG-domain of GRP7 is involved in RNA binding, mutation studies on putative RNA interacting or modulating sites were performed. In addition to MST experiments, we examined liquid-liquid phase separation of GRP7 and its mutants, both with and without RNA. Furthermore, we systemically investigated factors that might affect RNA binding selectivity of GRP7 by testing RNAs of different sizes, structures, and modifications. Consequently, our study revealed that GRP7 exhibits a high affinity for a variety of RNAs, indicating a lack of pronounced selectivity. Moreover, we established that the RGG-domain plays a crucial role in binding longer RNAs and promoting phase separation.

Figure 1

Analysis of the RNA binding of AtGRP7 via microscale thermophoresis (MST). (a) Thermophoresis traces of three individual measurements (1–3, displayed in blue (1), green (2) and red (3)) with a titration series of 16 of AtGRP7 and a constant concentration of AtCOR15A. The x-axis displays the time (s) of the thermophoresis experiment, 0 s as start of the thermophoresis. The y-axis displays the relative fluorescence measured in the heated area. The blue area marks F_cold while the red area marks F_hot, both used to calculate the binding curve. (b) Binding curves of three individual measurements (1–3, displayed in blue (1), green (2) and red (3)) of AtGRP7 and AtCOR15A. The x-axis displays the Ligand concentration (M, logarithmic). The y-axis displays the fraction bound, from 0 (no RNAs bound) to 1 (all RNAs bound). (c) K_ds of different RNAs bound by AtGRP7 shown in a bar graph. The x-axis displays the RNA measured and the y-axis the K_d in µM with a cut between 1 and 4 µM.

Figure 2

Comparison of dissociation constants (K_ds) of AtGRP7 for RNAs varying in length, with and without UTRs and Introns, as well as single and double stranded RNA. (a) K_ds of AtGRP7 for RNAs varying in length displayed in a bar graph. The length of RNAs is shown on the X-axis, the Y-axis shows the K_d in µM. K_ds were compared by one-way ANOVA and Tukeys test (p = 0.01). Similar letters indicate no significant difference. (b) Comparison of K_ds of AtGRP7 for RNAs with and without UTRs and Introns. Y-axis shows the K_d in µM. K_ds were compared by by one-way ANOVA and Tukeys test (p = 0.05). Same letters indicate no significant difference. (c) Comparison of K_ds of AtGRP7 for ssRNA and dsRNA. Y-axis shows the K_d in µM. K_ds were compared by one-way ANOVA and Tukeys test (p = 0.05). Same letters indicate no significant difference.

Figure 3

Dissociation constants (K_ds) of AtGRP7 for methylated RNA. (a–d) display the binding affinities of AtGRP7 towards four different RNAs with and without methylation. Y-axis: K_d in µM, X-axis: different RNA with different methylations. The binding affinity of AtGRP7 for methylated and unmethylated RNA was compared with a one-way ANOVA and a Tukeys test for p = 0.05. Shared letters indicate no significant difference in binding affinity.

Figure 4

Dissociation constants of AtGRP7^short for small and long RNAs. (a) AtGRP7 consist of two domains, the RNA-recognition motif (RRM) from 1 to 87 aa and the glycine-rich region (88–176 aa), while the truncated AtGRP7^short only consist of the RRM. AtGRP7^RGG consist of the RGG-domain. (b) Binding affinity of AtGRP7short for small and long RNAs, shown in a bar graph. The K_d of AtGRP7^short towards the different RNAs was compared by a one-way ANOVA with a Tukeys test (p = 0.05). (c) Comparison of AtGRP7 and AtGRP7short RNA binding affinities towards small RNAs, shown in a bar graph. The Y-axis displays the K_d in µM cut between 3 and 4 µM. The binding affinity of AtGRP7 and AtGRP7^short towards the different RNA was compared by a one-way ANOVA with a Tukeys test (p = 0.05). Shared letters indicate no significant different K_d. (d) Binding affinity of the eYFP-tagged AtGRP7^RGG for AtGRP7 3′UTR and AtGRP7 CDS RNA displayed in a bar graph. The y-axis resembles the K_d in µM and is cut between 0.1 and 0.3 µM.

Figure 5

Mutations and differences in the glycine-rich region of GRP7 and their effect on the dissociation constants (K_ds) for different RNAs. (a) The RGG-domain of AtGRP7, starting at amino acid 88, in comparison with the RGG-domain of two mutants, AtGRP7^YtoE and AtGRP7^StoE as well as to BnGRP7 and the hybrid protein AtBnGRP7. Differences are highlighted in red. (b–g) Binding affinity of AtGRP7, AtGRP7^YtoE, AtGRP7^YtoG, AtGRP7^StoE, AtGRP7^StoG, BnGRP7 and AtBnGRP7 for AtGRP7 (b), AtGRP7 pre-mRNA (c) AtGRP8 (d), AtGRP8 pre-mRNA (e), AtCOR15A (f) and AtCOR15A pre-mRNA (g). Y-axis: K_d in µM, cut between 12 and 13 µM. The binding affinities of AtGRP7 and different versions of GRP7 towards different RNAs were compared with one-way ANOVA and a Tukeys test for p = 0.05. Shared letters indicate no significant difference in binding affinity.

Figure 6

Liquid–liquid phase separation of AtGRP7 and mutants. 10 µM of eYFP-tagged AtGRP7, AtGRP7^short, AtGRP7^RGG, AtGRP7^YtoE, AtGRP7^StoE, AtGRP7^YtoG and AtGRP7^StoG and their phase separation behavior with and without 0.5 µM miRNA164. The white bar represents 5 µm.