A Sister of PIN1 gene in tomato (Solanum lycopersicum) defines organ initiation patterns by maintaining epidermal auxin flux

The spatiotemporal localization of the plant hormone auxin acts as a positional cue during early leaf and flower organogenesis. One of the main contributors to auxin localization is the auxin efflux carrier PIN-FORMED1 (PIN1). Phylogenetic analysis has revealed that PIN1 genes are split into two sister clades; PIN1 and the relatively uncharacterized Sister-Of-PIN1 (SoPIN1). In this paper we identify entire-2 as a loss-of-function SlSoPIN1a (Solyc10g078370) mutant in Solanum lycopersicum. The entire-2 plants are unable to specify proper leaf initiation leading to a frequent switch from the wild type spiral phyllotactic pattern to distichous and decussate patterns. Leaves in entire-2 are large and less complex and the leaflets display spatial deformities in lamina expansion, vascular development, and margin specification. During sympodial growth in entire-2 the specification of organ position and identity is greatly affected resulting in variable branching patterns on the main sympodial and inflorescence axes. To understand how SlSoPIN1a functions in establishing proper auxin maxima we used the auxin signaling reporter DR5::Venus to visualize differences in auxin localization between entire-2 and wild type. DR5::Venus visualization shows a widening of auxin localization which spreads to subepidermal tissue layers during early leaf and flower organogenesis, showing that SoPIN1 functions to focus auxin signaling to the epidermal layer. The striking spatial deformities observed in entire-2 help provide a mechanistic framework for explaining the function of the SoPIN1 clade in angiosperm species. Author Summary The plant hormone auxin acts as a positional signal in most plant developmental processes. The PIN-FORMED family of auxin transporters are the main contributors to auxin localization, especially PIN-FORMED1, which has been studied extensively in plant model species Arabidopsis thaliana. Members of the PIN-FORMED gene family have been found in all plant species, but there is a scarcity of mutants described outside Arabidopsis thaliana. Using Solanum lycopersicum (tomato) as a system, this study identifies a loss of function mutant from the Sister-Of-PIN1 clade in the SlSoPIN1a gene. The characterization of this mutant reveals the role of SlSoPIN1a in establishing position of organ initiation during shoot and reproductive development, including a role in establishing proper spiral phyllotaxy. We use an auxin visualization technique to conclude SlSoPIN1a functions in specifying auxin presence in proper cell layers to establish organ and tissue positioning. This work gives further evolutionary context to how PIN-FORMED genes act to establish organogenesis in the plant kingdom.


121
S. lycopersicum has also been used specifically to understand auxin directed developmental 122 mechanisms such as SlSoPIN1 protein localization in developing organs [6,39-41] and effects 123 of auxin application on organogenesis [3,4,6]. Although S. lycopersicum is used extensively as a   to S. lycopersicum [36,37]. These reports suggested that a duplication event in the SoPIN1 150 clade in S. lycopersicum occurred roughly sometime after the divergence between S. 7 lycopersicum and Mimulas guttatus [36,37]. To gain a more precise understanding of the history 152 of the SoPIN1 clade, we performed phylogenetic analysis on PIN1 and SoPIN1 genes sampling 153 Solanaceae more extensively by including Capsicum annuum, Nicotiana

165
thaliana and the closely related species C. rubella ( Figure 1A and  lines, sf, e-2, and div were characterized as having leaf phenotypes (S1 Figure

194
Organ patterning varied considerably between e-2 individuals, and the most consistent 195 phenotype was a deviation from the wild type spiral phyllotactic pattern. This aspect of the 196 phenotype was subsequently used for further co-segregation and complementation analyses. 197 198

C-T 490 co-segregates with the e-2 phenotype 199
We tested if C-T 490 was responsible for the e-2 phenotype by asking if C-T 490 co-200 segregates with the phyllotaxy phenotype. We crossed e-2 with the wild type background, self-9 pollinated the heterozygous progeny, and scored for presence and absence of the spiral 202 phyllotactic pattern in the F2 population (n = 112   Figure 4K). In order to quantify leaf shape explicitly, shape differences were derived from 293 characterization of the terminal leaflet of mutant and wild type. Overall, e-2 leaflets have a 294 significantly larger area (Welch Two Sample t-test, p = < 2.20 e-16) ( Figure 4L) and have higher 295 circularity, a measure of leaflet serration and lobing (Welch Two Sample t-test, p=2.391e-13), 296 than wild type individuals (S1 Table). We used Elliptical Fourier analysis [50,51] to quantify 297 differences in shape through measurement of leaflet outlines. Using Principal Component

301
The outline shape varied extensively between e-2 leaflets, as lobing and serrations were placed 302 13 seemingly randomly ( Figure 4K) resulting in the average e-2 outline as a near circular shape 303 compared to wild type, which usually consists of three prominent lobes ( Figure 4O). As seen in 304 other loss of function sopin1 and pin1 mutants [1,11,45], lateral organ morphogenesis is often

SlPIN1a functions in meristem maturation and organ specification 469
Of all the PINs, PIN1 appears to be the most important for reproductive development     observed in e-2/slsospin1a [45]. In A. thaliana, leaf development can be split into three 547 developmental stages which are morphologically distinct [27]. In atpin1, the leaf phenotype 548 clearly varies in the three leaf stages[70], as leaf vasculature and shape gets progressively 549 impaired with age, until there is a complete loss of lateral organ development after bolting [27].

550
There are two developmental stages in S. lycopersicum leaf formation, juvenile leaves, and 551 adult leaves before and after sympodial growth. In e-2/soslpin1a phyllotaxy abnormalities occur

Phylogenetic analysis 578
We used only sequences from species from fully sequenced genomes to ensure proper         33 807 S1

808
Measurements were made on binary images made from terminal leaflets of leaves 1 -4 of six 809 week old plants. Numbers presented as mean ± standard deviation. * indicates the differences 810 between wild type and e-2 are significant at P < 0.05. Circularity score of 1 signifies a perfect 811 circle, while the closer to zero, the leaves are progressively more lobed. All measurements were 812 made in ImageJ.