Standardization of VPS13 proteins description in Viridiplantae to facilitate their characterization

VPS13s are well characterized proteins in yeast and mammals playing multifaceted roles in organelle biogenesis and functions (Reinisch et al., 2025). They localize at multiple organelles and membrane contact sites (Hanna et al., 2023; Reinisch et al., 2025). A large set of physiological, biochemical and structural data showed that VPS13 proteins are able to transfer lipids between membranes by a non-vesicular lipid pathway (Leonzino et al., 2021; Melia & Reinisch, 2022; Reinisch et al., 2025). In plants, the characterization of VPS13 proteins is still in its infancy. In the model plant Arabidopsis thaliana, a VPS13 protein initially called Shrubby (SHBR, At5g24740) was shown to play a role in root growth and radial patterning and to be important for plant reproduction (Koizumi & Gallagher, 2013). Another role in apomixis was identified for a VPS13 protein in Taraxacum (Dandelion) (Van Dijk et al., 2020). In 2022, T. Levine performed an extensive in silico analysis of VPS13 gene numbers and domain organization in multiple organisms including Arabidopsis thaliana, showing the general conservation of VPS13 protein organization in plants and highlighting their peculiarities (refer to subsequent text for details) (Levine, 2022). More recently, our group performed an extensive domain analysis and structural predictions of VPS13 proteins in Viridiplantae as well as a phylogenetic analysis to understand the evolution of this protein family in this clade (Leterme et al., 2023). In their work published recently in New Phytologist, Tangpranomkorn et al. undertook the functional characterization of another member of the VPS13 family (At1g48090) and showed a role in pollen germination and pollen tube elongation (Tangpranomkorn et al., 2025).

The work performed by Tangpranomkorn et al. significantly contributes to our understanding of the localization and function of VPS13 proteins in plants. However, we would like to clarify some statements made by the authors, particularly regarding the name, gene number, domain organization and phylogeny of the VPS13 proteins in plants (related to fig. 2 in Tangpranomkorn et al., 2025). In 2022, T. Levine clearly demonstrated that the A. thaliana genome encodes four distinct VPS13 proteins (Levine, 2022), and not three as stated by the authors. The different paralogs were renamed AtVPS13S (At5g24740), AtVPS13M1 (At4g17140), AtVPS13M2 (At1g48090) and AtVPS13X (At3g50380), the last one being a non-canonical VPS13 (Levine, 2022). The sequence and structural divergence of AtVPS13X compared to the other A. thaliana VPS13 proteins made its identification difficult by classical BLAST or PSI-BLAST searches and even through profile hidden Markov models, as in Tangpranomkorn et al. As a result, AtVPS13X has long been overlooked in multiple studies, including from our own group (Velayos-Baeza et al., 2004; Michaud et al., 2017; Leterme & Michaud, 2022). The formal identification of AtVPS13X in the A. thaliana proteome by T. Levine required a custom structural homology search on HHpred (Zimmermann et al., 2018) using repeating β-groove domains (refer to the subsequent text for details) that had previously been structurally defined by hand (Levine, 2022). Using AtVPS13X as a query in BLAST and PSI-BLAST searches further enabled the detection of AtVPS13X orthologs in various streptophyte species (Leterme et al., 2023). The new plant VPS13 nomenclature was proposed to avoid confusion with the metazoan VPS13 proteins already named VPS13A, B, C and D, as no clear phylogenetic relationship was found between the human and A. thaliana VPS13 proteins (Levine, 2022; Leterme et al., 2023). Human VPS13B orthologs were found in some green algae, charophytes and a bryophyte and a new paralog, named VPS13Y, was thereafter identified in green algae and in charophytes (Levine, 2022; Leterme et al., 2023). Therefore, it is important to note that AtVPS13a (At1g48090) and AtVPS13b (At4g17140) characterized in Tangpranomkorn et al. correspond to AtVPS13M2 and AtVPS13M1, respectively.

The extensive in silico analysis performed by T. Levine showed that VPS13 structure is composed of repeating β-groove (RBG) domains, which are enriched in β-sheets, folding into an elongated tunnel with a hydrophobic concave surface allowing bulk lipid transport (Levine, 2022; Neuman et al., 2022). In addition, VPS13 proteins harbor a large variety of domains inserted between the RBG repeats that regulate their intracellular localization (Levine, 2022). The domains found in almost all known VPS13 proteins, referred to as canonical domains, are: (1) a N-terminal Chorein_N domain; (2) a VPS13 adaptor binding (VAB) domain; (3) an ATG2_C domain; and (4) a Pleckstrin Homology (PH)-like domain at the C-terminus (Levine, 2022) (Fig. 1a). Most VPS13 proteins also harbor non-canonical additional domains that vary between species and paralogs and that might contribute to the regulation of VPS13 localization and function (Levine, 2022). A. thaliana VPS13M1 and M2 harbor multiple additional domains that were described and renamed in accordance with their folding (Fig. 1a) (Levine, 2022). This new nomenclature of VPS13 proteins and their structural domains constitutes a reference in this field of research and has since been extensively used by our group (Leterme et al., 2023) and others thereafter (Hanna et al., 2023; Huercano et al., 2025; Swan, 2025). Therefore, the fig. 2b of Tangpranomkorn et al. is misleading as: (1) proteins and domains are not annotated with their proper names; (2) some domains are not annotated at all (the C2 domain and the β-tripod domain in C-term in AtVPS13M1); and (3) the AtVPS13X gene is not represented.

We would also like to raise some concerns about the interpretation of the phylogeny analysis performed in this work (fig. 2a in Tangpranomkorn et al., 2025). The results presented suggest that AtVPS13M2 (At1g48090) is a land plant specific VPS13 paralog, which is not in agreement with previous work performed by our group in which we analyzed more than 400 VPS13 protein sequences from Archaeplastida, including species also analyzed by Tangpranomkorn et al. (Leterme et al., 2023). A more robust analysis using both Maximum Likelihood and Bayesian inference on a smaller dataset allowed us to trace the history of VPS13 proteins in Viridiplantae. Our results clearly showed that a VPS13M paralog was present in the most recent common ancestor of Viridiplantae and that VPS13M was then duplicated during the evolution of charophytes, a group of algae basal to land plants (Fig. 1b). We also analyzed the domain organization of VPS13M orthologs in chlorophytes and charophytes and showed that they have a similar organization than AtVPS13M1 and M2 (Fig. 1c), confirming their ancestral origin (Leterme et al., 2023). We believe that the phylogenetic analysis shown in fig. 2a of Tangpranomkorn et al. was conducted with an insufficient dataset. Consequently, the conclusion that AtVPS13M2 is a ‘land plant specific’ VPS13 is not supported by the best evidence.

To avoid further confusion in the future, we propose that the nomenclature, domain organization and phylogeny of VPS13 proteins in Viridiplantae follow the in silico analyses performed in the past (Levine, 2022; Leterme et al., 2023). This will facilitate the community's ongoing efforts to characterize this remarkable family of proteins.

None declared.

MM wrote the article with the input of SL. SL compiled the figure.

The New Phytologist Foundation remains neutral with regard to jurisdictional claims in maps and in any institutional affiliations.