Si permeability of a deficient Lsi1 aquaporin in tobacco can be enhanced through a conserved residue substitution

Abstract Silicon (Si) is a beneficial substrate for many plants, conferring heightened resilience to environmental stress. A plant's ability to absorb Si is primarily dependent on the presence of a Si‐permeable Lsi1 (NIP2‐1) aquaporin in its roots. Structure‐function analyses of Lsi1 channels from higher plants have thus far revealed two key molecular determinants of Si permeability: (a) the amino acid motif GSGR in the aromatic/arginine selectivity filter and (b) 108 amino acids between two highly conserved NPA domains. Curiously, tobacco (Nicotiana sylvestris) stands as a rare exception as it possesses an Lsi1 (NsLsi1) with these molecular signatures but is reported as a low Si accumulator. The aim of this study was therefore to identify whether additional determinants influence Si permeability via Lsi1 channels, focusing on the role of residues that differ uniquely in NsLsi1 relative to functional Lsi1 homologs. We observed tobacco indeed absorbed Si poorly (0.1% dw), despite NsLsi1 being expressed constitutively in planta. Si influx measured in NsLsi1‐expressing Xenopus oocytes was very low (<13% that of OsLsi1 from rice (Oryza sativa) over a 3‐hr time course), which likely explains why tobacco is a low Si accumulator. Interestingly, NsLsi1P125F displayed a significant gain of function (threefold increase in Si influx relative to NsLsi1WT), which coincided with a threefold increase in plasma membrane localization in planta, as measured by transient expression of GFP constructs in Nicotiana benthamiana leaves. These findings thus reveal a novel molecular determinant of Si transport in plants and inform breeding, biotechnological, and agricultural practices to effectively utilize Si in the context of plant resilience to environmental stress.

Title: Elevated Si transport and plasma-membrane localization is conferred by a conserved residue 3 substitution in the Lsi1 aquaporin of tobacco 4 5 Authors: Devrim Coskun, Rupesh Deshmukh, Humira Sonah, S. M. Shivaraj, Rachelle Frenette-Cotton, 6 Laurence Tremblay, Paul Isenring, and Richard Bélanger 7 8 9 REVIEWER COMMENTS FROM PLANT JOURNAL AND AUTHOR RESPONSES 10 11 12 Reviewer: 1 13 14 This manuscript was a pleasure to read; the topic is scientifically very relevant. The description of the 15 results is very clear, but some points identified below could be further addressed in a revised version: 16 17 Figure 6 could be more informative regarding the localization of NsLsi1 in plasma membrane. A co-18 localization experiment with markers of the tonoplast would benefit the interpretation of the results (a 19 TIP-RFP, for instance) and the quantification of fluorescence intensity could be performed.

21
We are unsure if/how testing for co-localization with a tonoplast-intrinsic protein would 22 indeed "benefit the interpretation of the results", since, as is widely documented in the 23 literature, Lsi1 aquaporins are strictly plasma-membrane intrinsic proteins. With that said, 24 we can easily quantify fluorescence intensity in a revised version.

26
In any case, authors could discuss how a transient expression with a 35S promoter was the best 27 approach to study differences between NsLsi1WT and NsLsi1P125F targeting to the plasma membrane. 28 29 Use of the ubiquitous 35S promoter is a very popular and straightforward method to test for 30 the transient expression of genes, particularly those encoding membrane transporters. As 31 such, this was the method we selected, and quite successfully, in our opinion, since we 32 observed a clear membrane localization of the gene product, as well as differences between 33 wildtype and mutant.

35
Furthermore, this result in N. benthamiana leaf cells is not a direct evidence (as acknowledged by the 36 authors) that supports the observed increase of the transport activity, because transport experiments 37 were performed in a different model (oocytes). 38 39 As the reviewer correctly notes, we did acknowledge that the leaf transient assay provided 40 indirect evidence for the transport activity we observed (i.e. in oocytes).

42
In this regard, I do not totally agree with the sentence "Thus, this suggests the gain of function 43 observed in oocytes is likely the result of increased protein abundance at the plasma membrane" (line 44

316), and results from western blot analysis in oocytes (even if the tagged protein is non-functional) 45
could be more useful to support this conclusion. 46 47 We can revise the sentence in question to emphasize the indirect evidence the transient 48 expression analysis provides. It would read as follows: "Thus, this suggests, albeit indirectly, 49 that the gain of function observed in oocytes is possibly the result of increased protein 50 abundance at the plasma membrane". Unfortunately, the results of the Western blot analysis 51 would not be useful in this case, as not only were the tagged proteins non-functional, but 52 yielded little/no membrane expression in blots, as explained in the text.

54
So, I am afraid the title of the paper could be rephrased to highlight only the observed change in the 55 kinetic behavior of the encoded proteins, which, in my opinion, is a very interesting result. The altered 56 transport kinetics mediated by NsLsi1P125F (involving a shift in both of Vmax and Km) relative to 57 NsLsi1WT seems more related to a structural change than to an altered targeting of the protein.

59
Although we were careful not to imply that a single amino acid was responsible for the 60 phenotypes observed in tobacco, we realize that the title may be misleading. Accordingly, we 61 propose the following title change: Elevated Si transport and plasma-membrane localization 62 is conferred by a conserved residue substitution in the Lsi1 aquaporin of tobacco 63 64 Figure 3 and 5 could be merged.

66
Given that both Figures 3 and 5