Description

In plants, leaf senescence is a process where older leaves turn yellow due to chlorophyll degradation as they export nutrients, primarily nitrogen, to support the growth of new tissues. The timing of leaf senescence is regulated by many mechanisms since both early and late leaf senescence are detrimental (Woo et al., 2019). Early leaf senescence does not allow adequate growth to support reproduction while late leaf senescence can increase the sink strength of the leaves, rendering them less capable of nutrient export.

PEN1/SYP121 (AT3G11820) is a syntaxin of plants that is the Qa component of a plasma membrane target-SNARE complex (PEN1-SNAP33) that docks transport vesicles via binding to vesicle SNAREs (VAMP 721/722) for vesicle fusion and exocytosis (Lipka et al., 2007; Bassham and Blatt, 2008). PEN1 is important for the localized thickening of the cell wall (papillae formation) that slows the penetration of fungal hyphae, one mechanism of nonhost resistance (Collins et al., 2003). pen1 mutants were first identified based on faster fungal penetration. In addition, although pen1 mutants displayed a normal phenotype in the absence of infection, they displayed a constitutive increase in the phytohormone salicylic acid (SA) (Zhang et al., 2007). SA is a positive regulator of pattern triggered immunity (PTI) and leaf senescence (Tian et al., 2025).

PEN3/PDR8 (AT1G59870) is an ABC transporter that carries the indole-derived antimicrobial molecule, camalexin, across the plasma membrane (He et al., 2019). pen3 mutants were selected based on faster fungal penetration (Stein et al., 2006) and pen3 mutants accumulate camalexin post-infection. pen3 mutants also show leaf yellowing when grown under high light (HL) intensity (900 µmoles photons/m ² /sec). The HL-induced chlorosis is dependent on SA and is not observed when SA synthesis is inhibited by mutation in isochorismate synthase1 ( sid2 ).

We previously noted that pen1pen3 double mutants displayed early leaf senescence when grown under lower light intensity (40 µmoles photons/m ² /sec) (Crane et al., 2019). The plants grew normally for approximately six weeks, and then older leaves rapidly turned yellow. The early leaf senescence phenotype was reversed by the sid2 mutation, demonstrating that the early leaf senescence was dependent on SA. These findings suggest that PEN1 and PEN3 have an additive effect on SA levels, with elevated SA levels leading to premature leaf senescence.

GFP-tagged PEN1 was shown to physically interact with Hypersensitive Induced Reaction2 (HIR2, AT3G01290) in microsomal fractions using immunoprecipitation followed by mass spectrometry ​(Fujiwara et al., 2014)​. HIR2 contains a stomatin/prohibitin/flotillin/HflK/C (SPFH) domain​. There are four  HIR  paralogs in Arabidopsis and  HIR2  is strongly expressed in leaves. HIR2 localizes to membrane nanodomains via S-acylation (Weber et al., 2025; Daněk et al., 2026). HIR2 interacts with a wide variety of leucine-rich receptor protein kinases including FLS2 that mediates PTI and BAK1, a co-receptor for many immune receptors (Wu et al., 2025). Arabidopsis lines that overexpress HIR2 show increased PTI (Qi et al., 2011). HIR2 is likely an assembly complex that tethers proteins to plasma membrane nanodomains. Importantly, SA levels were reduced by 70% in tobacco plants with silenced NbHIR3 expression and transient overexpression of NbHIR3 increased SA levels (Li et al., 2019). In addition, a cell death inducing protein (SsXyl2) from a necrotrophic fungal pathogen interacts with HIR2 in tobacco ( Nicotiana benthamiana ) at the plasma membrane to promote cell death activity (Wang et al., 2024). Nbhir2 knockout mutants showed a reduced hypersensitive response and transient expression of both SsXyl2 and NbHIR2 resulted in larger areas of cell death.

We asked whether the early senescence phenotype of pen1pen3 was dependent on HIR2. Was it possible that PEN1 interaction with HIR2 prevents HIR2-induced PTI such that a pen1 mutant allows unrestrained HIR2 activity and early leaf senescence? To address this question, we produced pen1pen3hir2 mutants and quantified their leaf senescence phenotypes. We hypothesized that  pen1pen3hir2  mutants would display a less severe early leaf senescence phenotype when compared to  pen1pen3 . 

A T-DNA insertion for&nbsp; hir2-2 &nbsp;(SALK_124393, named hir2 in this study) was crossed to&nbsp; pen1-1pen3-1 . One triple mutant (line 5, panel A) was selected from the F ₃ generation as being homozygous for a T-DNA insertion disrupting the fourth exon of the HIR2 open reading frame (Weber et al., 2025) and for two separate point mutations that produce nonsense ( pen1-1 , Collins et al., 2003) and missense ( pen3-1 , Stein et al., 2006) null mutations (panel A). WT, pen1pen3 , hir2 and pen1pen3hir2 plants were grown under a 20:4 long day light cycle (40 µmoles photons/m ² /sec) under nutrient replete conditions (n=12 per genotype). These conditions minimize abiotic stress, which can promote leaf senescence. After six weeks of growth, both pen1pen3 and pen1pen3hir2 started to show yellowing of mature rosette leaves while WT and hir2 remained green. At seven weeks, plants were harvested, and panel B shows representative plants of each genotype. Initially, the rosette leaves for each plant were characterized as green, <50% yellow, > 50% yellow or brown. Panel C shows that WT and hir2 individuals had nearly all green rosette leaves while nearly all pen1pen3 and pen1pen3hir2 rosette leaves displayed at least some yellowing. The percentage of leaves with <50% yellow, > 50% yellow and brown was compared between pen1pen3 and pen1pen3hir2 and no significant difference was observed (One-way ANOVA, Tukey's HSD correction for multiple comparisons, p-value = 0.999). Similarly, no significant difference was observed between WT and hir2 (p-value 0.952).

After the initial rosette leaf observations, leaf 6 from each individual plant was weighed and stored at -80 ^o C prior to chlorophyll quantitation (Panel D). Compared to WT, chlorophyll levels were lower in pen1pen3 and in pen1pen3hir2 . There was no significant difference between pen1pen3 and pen1pen3hir2 or between WT and hir2 . The difference between hir2 and pen1pen3hir2 was close to significance (One-way ANOVA, Tukey's HSD correction for multiple comparisons, p-value = 0.0661).

Leaves 4 and 5 were harvested together from each individual plant and expression of SGR1 , a senescence-associated gene, was measured using real-time qPCR with ACT2 as the reference transcript (Panel D). SGR1 expression was significantly higher in pen1pen3 and in pen1pen3hir2 compared to WT and hir2 , respectively. SGR1 expression was not significantly different when pen1pen3 and pen1pen3hir2 were compared (One-way ANOVA, Tukey's HSD correction for multiple comparisons, p-value 0.1727). A decrease in SGR1 expression was noted between hir2 when compared to WT, suggesting HIR2 may have an independent positive role in leaf senescence; however, delayed leaf senescence was not seen when chlorophyll was quantified. Both pen1pen3 and pen1pen3hir2 exhibited similar early leaf senescence indicating that hir2 does not reverse the leaf senescence symptoms. These data do not support our hypothesis, and the proposed role for PEN1 in sequestering HIR2 to prevent early leaf senescence is not supported by our experimental evidence.

Although&nbsp;PEN1 has been shown to interact with HIR2, our findings demonstrate this interaction is not playing a role in early leaf senescence conferred by pen1pen3 . Elevated SA contributes to early leaf senescence in pen1pen3, and lower SA was noted in NbHIR3 -silenced tobacco lines; however, the hir2 mutation in Arabidopsis did not alleviate early leaf senescence. One caveat is the three&nbsp;additional&nbsp;Arabidopsis&nbsp; HIR &nbsp;paralogs, although PEN1 interaction with these proteins was not observed. Regardless, this potential genetic redundancy may complicate analysis. In the future, quadruple mutants ( pen1pen3hir2hir3 ) including the paralog with the next highest expression level in leaves,&nbsp; HIR3 &nbsp;(AT1G69840) , can be compared to pen1pen3 .&nbsp;In addition, SA levels in the hir2-2 mutant should be quantified.

hir2-2 (AT3G01290, SALK_124393) is a T-DNA insertion disrupting the 4 ^th exon in the open reading frame and is a knockout mutant (Weber et al., 2025). pen1-1 (AT3G11820, TAIR Accession Polymorphism:1010235864) is a nonsense mutation, C to T at chromosome 3 position 3730649 (A on reverse strand of panel A, Collins et al., 2003). pen3-1 (AT1G59870, TAIR Accession Polymorphism:1010235837) is a missense G354D, G to A at chromosome 1 position 22036079 (Stein et al., 2006). Mutant lines were obtained from the Arabidopsis Biological Resource Center (Columbus, OH). Chlorophyll quantitation, RNA extraction, cDNA synthesis and real-time qPCR have been described (Zimmerman et al., 2024). One-way ANOVA with Tukey's HSD correction was used for multiple comparisons (GraphPad Prism).

Primers used in this study:

HIR2:

SALK_124393_LP: CACTCCTGCGACTTTCTTCTC

SALK_124393_RP:&nbsp; ACCATGTCCAACACATCCTTC

SALK_LBP: GGATTTTGCCGATTTCGGAACC

PEN1

PEN1_643_F: CAGAGGTCCTGGTTCGATCG

PEN1_643_R: CTTCCCTTCTCCGCCATTGT

pen1-1_643_Seq: GCCTTCTTCAACGCGACTCC

PEN3

PEN3_721_F: GTGAGTCCAACTCTCGCCTT

PEN3_721_R: AAGGTCTGTTCGGGTTCACC

pen3-1_721_Seq: CCGTGTAACAGATTTTGGGGC

qPCR

ACT2_F: GCGACTTGACAGAGAAGAAC

ACT2_R: GAAAGAGCGGAAGAAGATGAG

SGR1_F: TGGGCAAATAGGCTATACCG

SGR1_R: CCACCGCTTATGTGACAATG