A role of the Arabidopsis polyprenol reductase 1 in brassinosteroid biosynthesis

Overexpression of Arabidopsis POLYPRENOL REDUCTASE 1 (PPRD1) partially rescued the phenotype of the de-etiolated2-1 (det2-1) brassinosteroid biosynthesis mutant and increased its brassinosteroid contents. A loss-of-function pprd1 mutation enhanced the det2-1 short-root phenotype and further reduced its brassinosteroid levels, suggesting that PPRD1 plays a role in brassinosteroid biosynthesis. Brassinosteroids (BRs) are essential plant steroids that regulate various growth and developmental processes (Nolan et al. , 2020). Brassinosteroid biosynthesis involves multiple cytochrome P450 enzymes and a single non-CYP450 enzyme DET2 (Zebosi et al. , 2024), which was named after its dark “de-etiolated” mutant phenotype (Chory et al. , 1991). DET2, an Arabidopsis homolog of the mammalian steroid 5α-reductases (SRD5As), catalyzes the reduction of the Δ4, 5 double bond of 3-oxo-Δ4, 5 sterols (Li et al. , 1996; Fujioka et al. , 1997) (Figure S1). Unlike other severely dwarfed BR-deficient mutants, det2-1 is a relatively weak mutant retaining ~10% of the wild-type BR levels, leading to the hypothesis that det2-1 might possess a BR-biosynthetic “non-DET2” SRD5A (Fujioka et al. , 1997), which was later supported by biochemical evidence (Rosati et al. , 2005). However, the molecular identity of this “non-DET2” SRD5A remains unknown, even 30 years after the DET2 cloning. Genome sequencing analysis revealed multiple SRD5A homologs in nearly all sequenced plant species (Ali et al. , 2024). Nevertheless, no “non-DET2” SRD5A has yet been linked to BR biosynthesis. To determine whether the weak det2-1 phenotypes were due to residual activity of the mutant det2-1 enzyme (carrying the Glu204-Lys mutation), we generated a null det2-c mutation in det2-1 via the CRISPR/Cas9 approach. The resulting det2-c mutant was phenotypically identical to det2-1 (Figure S2), indicating that the ~10% residual BR content in det2-1 was unlikely to have originated from the mutant det2-1 enzyme. This finding strongly suggested the presence of an alternative “non-DET2” SRD5A enzyme in BR biosynthesis, providing a rationale to investigate other candidate enzymes. The Arabidopsis genome encodes a total of nine SRD5A enzymes (Figure S3). At5g16010 was recently named as DET2L for DET2-like protein due to its highest sequence similarity to DET2 (Ali et al. , 2024), while At1g72590 (polyprenol reductase 1/PPRD1), At2g16530 (PPRD2), and At3g43840 (PPRD3; a truncated polypeptide of only 84 amino acids) are homologs of the human type 3 SRD5A (SRD5A3) (Jozwiak et al. , 2015). SRD5A3 reduces the α-isoprene unit of polyprenal in dolichol biosynthesis essential for protein N-glycosylation (Wilson et al. , 2024) (Figure S1), but its role in steroid metabolism remains controversial. At3g55360 is a well studied trans-2, 3-enol CoA reductase (ECR; also known as CER10/ECERIFERUM 10) involved in the biosynthesis of very-long-chain fatty acids (Zheng et al. , 2005) (Figure S1). The biochemical and physiological functions of At1g73650, At1g18180, and At2g46890 remain largely unexplored. We employed an overexpression strategy to identify an Arabidopsis SRD5A capable of rescuing the det2-1 mutation. Although DET2L shares the highest sequence similarity with DET2, overexpression of a DET2L-GFP fusion transgene in det2-1 failed to restore the BR-related growth defects (Figure S4A, B), whereas DET2-GFP overexpression fully complemented the det2-1 phenotypes. Additionally, two CRISPR/Cas9-generated det2l mutations did not exacerbate the det2-1 dwarfism (Figure S4C–F). These findings suggested that DET2L does not participate in BR biosynthesis. We suspected that DET2L's inability to compensate for the DET2 function was likely to have stemmed from the absence of two conserved catalytic residues, Glu57 and Tyr90 in DET2, which are replaced by Tyr73 and Phe105, respectively, in DET2L (Figure S5). We also overexpressed PPRD1/2-FLAG transgenes in det2-1. Remarkably, only PPRD1 − -->---> but not PPRD2 − -->---> substantially rescued the det2-1 phenotypes (Figure S6A, C, D, G). Further analysis showed that PPRD1 overexpression (PPRD1-OE) enhanced BR signaling in a dosage-dependent manner, increasing root length and hypocotyl height of light-grown and dark-grown seedlings, respectively (Figure 1A–E). PPRD1-OE also elevated the transcript abundance of a BR-induced gene IAA19 while reducing the expression of two BR-repressed genes, CPD and DWF4 (Figure 1F). Importantly, BR quantification revealed that PPRD1-OE significantly increased the content of castasterone (CS), the immediate precursor of brassinolide (BL), the most bioactive BR (Figures 1G, S6). In contrast, PPRD2-OE only had a marginal effect on det2-1 (Figure S7B–G), suggesting their distinct substrate preference despite their high sequence identity and similar 3D structures (Figure S8). Although a previous study had reported PPRD1 localization at the plasma membrane (Jozwiak et al. , 2015), our confocal microscopy and subcellular fractionation analyses revealed that both DET2 and PPRD1 localized to the ER membrane (Figure S9), further supporting the functional substitution of DET2 by PPRD1. PPRD1 overexpression rescues the det2-1 mutation, while pprd1 mutations enhance the short-root phenotype of the det2-1 mutant (A–C) Photographs of soil-grown 3-week-old plants (A), 10-d-old Petri dish-grown seedlings (B), and 5-d-old dark-grown seedlings (C) of the indicated genotypes. (D, E) Quantitative analyses of root length of light-grown seedlings (D) and hypocotyl length of dark-grown seedlings (E) of the indicated genotypes. Each bar represents the mean length of 30 seedlings, with error bars indicating ± SD. (F) Quantitative real-time reverse transcription polymerase chain reaction analysis of three BR-responsive genes. (G) Quantification of the CS content in whole seedlings. (H, I) Photographs of 8-d-old seedlings of the indicated genotypes grown on normal half-strength Murashige and Skoog (MS) agar (H) or brassinolide (BL) -containing half-strength MS medium (I). (J, K) Quantitative analysis of the root length of Arabidopsis seedlings grown on normal half-strength MS medium (J) or BL-containing half-strength MS medium (K). Each bar represents the average root length of ~30 seedlings, with the error bar indicating ± SE. (L) Measurement of CS levels in roots. Scale bars in (A–C, H, and I) represent 10 mm. Statistical analyses were performed using Student's t-test: ns, P > 0. 05; *P ≤ 0. 05; **P ≤ 0. 01; ***P ≤ 0. 001; and ****P ≤ 0. 0001. To further explore the role of PPRD1 in BR biosynthesis, we generated pprd1 mutants on both DET2+ and det2-1 backgrounds using the CRISPR/Cas9 approach. We found that pprd1 mutations had a minimal impact on rosette size on either the DET2+ or det2-1 background (Figure S10). However, consistent with the highest PPRD1 expression detected in roots (Figure S11), the pprd1 det2-1 double mutant exhibited significantly shorter roots compared with the det2-1 single mutant (Figure 1H–J). Notably, null pprd1 mutations did not affect root growth in the wild-type background (Figure 1H), and exogenous BL treatment fully rescued the pprd1-caused short-root phenotype (Figure 1K, L). BR quantification further revealed a significant reduction in CS levels in pprd1 det2-1 compared with det2-1 (Figures 1L, S6), confirming PPRD1's role in BR biosynthesis. The lack of an enhanced dwarf phenotype in the pprd1 det2-1 double mutant suggested the involvement of additional “non-DET2” SRD5A (s) functioning in shoot tissues. We therefore tested if ECR/CER10 might serve this role; however, overexpression of an ECR-FLAG transgene in det2-1 failed to restore BR-deficient phenotypes (Figure S12), effectively ruling out its contribution toward BR biosynthesis. Relative weak morphological phenotypes associated with null or severe mutations of DET2 orthologs have been observed in several other plant species, suggesting the presence of “non-DET2” SRD5As that compensate for DET2 loss. While PPRD1 is unique to Arabidopsis thaliana, many plant species possess multiple PPRD2 isoforms (Ali et al. , 2024), raising the possibility that some of these duplicates might also substitute for DET2 function. Structural comparison between PPRD1 and PPRD2s (Figure S8), site-directed mutagenesis of substrate entry gate or binding pocket, or the generation of chimeric PPRD enzymes, could provide valuable insight into their substrate specificity and potential multifunctionality. Given that both types 1 and 2 of human SRD5As fully rescue the Arabidopsis det2-1 mutation, this system could also be used to test whether human SRD5A3 could reduce the Δ4, 5-double bond of 3-oxo-Δ4, 5 steroids. Overall, our study not only revealed a novel physiological function of the Arabidopsis PPRD1 in BR biosynthesis but also opens up new avenues for exploring alternative steroid reductases involved in steroid metabolism. This work is dedicated to the late Professor Joanne Chory, who discovered the Arabidopsis det2-1 mutant and led the early DET2 studies. We thank Professor Qi-Jun Chen of China Agricultural University for the CRISPR/Cas9 plasmids. This work was partly supported by grants from the National Natural Science Foundation (NSFC31970187 to L. L. and NSFC31870253 to J. L. ) and funds from the South China Agricultural University, Hong Kong Baptist University, and the Hong Kong Jockey Club Charities Trust. All authors of this work declare no conflicts of interest. L. L. and J. L. conceived the research plans; L. L. , J. Z. , and J. M. supervised the project, discussed experimental design, and analyzed the results with H. W. and J. L. ; H. W. performed the majority of the work for this study with material and technical help from S. L. , W. Liu, and W. Li; H. W. , J. L. , and J. Z. prepared the manuscript. All authors read and approved its content. Additional Supporting Information may be found online in the supporting information tab for this article: http: //onlinelibrary. wiley. com/doi/10. 1111/jipb. 70022/suppinfo Figure S1. Comparison of the chemical reactions catalyzed by SRD5As Figure S2. A CRISPR/Cas9-generated det2-c null mutant is morphologically indistinguishable from the det2-1 mutant Figure S3. A simple phylogeny analysis of the DET2 and its homologs in Arabidopsis and human Figure S4. Overexpression or loss-of-function mutation of DET2L had little effect on the det2-1 mutant Figure S5. Structural comparison between DET2 and DET2L proteins Figure S6. Quantification of CS contents Figure S7. Overexpression of PPRD1 but not PPRD2 rescued the det2-1 mutation Figure S8. Sequence and structural comparison between PPRD1 and PPRD2 Figure S9. Both DET2 and PPRD1 are ER-localized membrane proteins Figure S10. Creation of CRISPR/Cas9 mutants of PPRD1 on the wild-type and det2-1 mutant background Figure S11. Quantitative reverse transcription polymerase chain reaction (RT-PCR) analysis of PPRD1 expression in different tissues Figure S12. Overexpression of ECR fails to rescue the det2-1 mutation Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.