Extranuclear function of Arabidopsis HOOKLESS1 regulates pleiotropic developmental processes in a non-cell-autonomous manner

Abstract

Introduction

A number of plant mutants exhibit highly pleiotropic phenotypes, often involved in the synthesis, inactivation, or signal transduction of plant hormones. Downstream signaling components of plant hormones, such as transcription factors, are often quite diverse across tissues, which might lead to their highly pleiotropic function (Wolters & Jürgens, 2009). While c. 10 plant hormones have been identified to date, it is predicted that a number of substances with plant hormone-like activity remain unidentified. For example, the plant-specific cytochrome P450 family CYP78A is implicated in the synthesis of novel bioactive substances (Anastasiou et al., 2007). This assertion is supported by observations that CYP78A5/KLUH acts in a non-cell autonomous manner, and the cyp78a5 single and cyp78a5 cyp78a7 double mutant exhibits pronounced pleiotropic phenotypes, including short plastochron length, early flowering, reduced apical dominance, reduced organ size, sterility, and fasciation (Anastasiou et al., 2007; Adamski et al., 2009; Eriksson et al., 2010; Poretska et al., 2020; Nobusawa et al., 2021). The function of CYP78A5 appears to be independent of known plant hormones (Wang et al., 2008).

In order to avoid the risk of damage during germination in the soil, a predominant proportion of eudicot plants develop an apical hook in the etiolated seedling under dark conditions. Several plant hormones, including auxin, ethylene, gibberellin, brassinosteroid, and jasmonic acid (JA), are known to be involved in hook formation (Béziat & Kleine-Vehn, 2018; Wang & Guo, 2019). Arabidopsis HOOKLESS1 (HLS1) is recognized as a crucial regulator of hook formation. For instance, the null mutant of hls1 exhibits a completely hookless phenotype even in the presence of ethylene, which strongly promotes hook formation (Guzmán & Ecker, 1990). However, HLS1, in fact, exhibits highly pleiotropic functions influencing various plant development processes beyond apical hook formation, such as leaf number (plastochron length), leaf senescence, disease susceptibility, flowering time, and thermomorphogenesis (Li et al., 2004; Liao et al., 2016; Jin et al., 2020). Furthermore, the quadruple mutant of HLS1 and its paralogs HLS1-LIKE HOMOLOG (HLH) 1–3 exhibits various phenotypes not observed in the hls1 single mutant, such as dwarfism and frequent increases in cotyledon number (Chang et al., 2013).

HLS1 shows significant homology to the family of N-acetyltransferases, although its specific substrate remains arguable. Liao et al. (2016) showed through ChIP assay that HLS1 is enriched at the transcription start site (TSS) and coding sequence (CDS) of the WRKY33 and ABA INSENSIGTIVE 5 genes, and that the acetylation levels of histone H3 are reduced in the hls1 mutant, suggesting that HLS1 is a histone acetyltransferase. However, they could not detect histone acetyltransferase activity of HLS1 in vitro. On the other hand, Huang et al. (2024) identified AUTOPHAGY-RELATED PROTEIN18a (ATG18a), one of the key factors in autophagy, as a substrate for acetylation by HLS1 during nutrient deficiency-induced senescence. However, they found that autophagy is not involved in hook formation, suggesting that ATG18a does not serve as a universal substrate for all the phenomena involving HLS1 in plant development.

HLS1 has been reported to interact with various proteins. Lyu et al. (2019) reported that HLS1 is activated through oligomerization in the nucleus and inactivated by binding to PHYTOCHROME B (PhyB) under light conditions. Guo et al. (2023) showed that HLS1 binds the ethylene signal transduction factor ETHYLENE INSENSITIVE 2 (EIN2), Wang et al. (2023a) reported that HLS1 binds to the flowering-promoting transcription factor CONSTANS, and Xiong et al. (2023) demonstrated that HLS1 binds to a small ubiquitin-like modifier E3 ligase. Additionally, as mentioned above, Huang et al. (2024) showed that HLS1 also binds to ATG18a. These diverse HLS1 interactors could potentially account for the pleiotropic functions of HLS1. However, these interactions form a highly intricate network, and a comprehensive and unified understanding of the HLS1 function remains to be established.

Besides the hls1 mutant, a number of other mutants exhibiting a hookless phenotype have been identified. The transcription factors EIN3, ETHYLENE INSENSITIVE LIKE 1, and PHYTOCHROME INTERACTING FACTORs (PIFs) play a crucial role in the regulation of hook formation by collaboratively enhancing the expression of HLS1. Notably, the multiple mutant of these factors shows a complete hookless phenotype (An et al., 2012; Shi et al., 2018; Zhang et al., 2018). In addition, mutants involved in the synthesis and signaling of auxin or brassinosteroids, as well as those of regulator controlling photomorphogenesis, such as CONSTITUTIVE PHOTOMORPHOGENIC1 (COP1), have also been identified (Mazzella et al., 2014). Notably, the altered meristem program1 (amp1)/cop2 mutant also exhibits a hookless phenotype (Lehman et al., 1996; Raz & Ecker, 1999). AMP1 encodes a putative carboxypeptidase that has been implicated in the regulation of translation via miRNA (Li et al., 2013). However, the molecular function of AMP1 is unclear. Interestingly, in addition to the hookless phenotype, amp1 exhibits a range of pleiotropic phenotypes that are also observed in hls1, such as shortened plastochron, early flowering, and early leaf senescence (Chaudhury et al., 1993; Kong et al., 2015; Poretska et al., 2020; Nobusawa et al., 2022).

In this study, we conducted a detailed phenotypic analysis of the hls1 and hls1 hlh1 double mutants, confirming the extensive pleiotropy of HLS1/HLH1 functions. We revealed that HLS1 is often expressed in tissues distinct from those exhibiting the phenotypic manifestations, indicating that HLS1 functions in a non-cell autonomous manner. Our research revealed that HLS1 exhibits a diffusive localization pattern in the cytoplasm and nucleus, and that the presence of HLS1 outside the nucleus is sufficient to complement the hls1 phenotypes. Genetic analysis revealed that HLS1 and AMP1 act in the same pathway in the plastochron regulation. Furthermore, HLS1 was not considered to directly participate in the synthesis or inactivation of known plant hormones. Our findings suggest that HLS1 functions as an acetyltransferase metabolizing substrates extranuclearly and potentially intranuclearly, regulating pleiotropic phenomena of plant development in a non-cell-autonomous manner, and acting in a novel network distinct from those of the known plant hormones.