ABSCISIC ACID-INSENSITIVE 5 (ABI5) Orchestrates Seasonal Growth Cessation and Wood Formation Inhibition in Populus tomentosa

Abstract

Perennial trees in temperate regions precisely coordinate the timing of seasonal growth cessation and dormancy with environmental cues, primarily photoperiod. While the roles of abscisic acid (ABA) in dormancy regulation are well-established, its function in growth cessation remains less defined. ABSCISIC ACID-INSENSITIVE 5 (ABI5) is a basic leucine zipper transcription factor that plays a central role in ABA-mediated development and abiotic stress responses, yet its roles in photoperiodic regulation of growth cessation and its coordination with radial stem growth remain unknown. Here, we demonstrate that in poplar (Populus tomentosa) trees, exogenous ABA application exacerbated short-day (SD)-induced growth inhibition, accelerated bud set, and strongly suppressed secondary xylem formation. We identified a Populus ABI5 homolog, PtoABI5, whose expression is induced by both ABA and SDs. Overexpression of PtoABI5 phenocopied and enhanced SD responses, leading to premature growth cessation and a pronounced inhibition of cambial division and wood formation under SDs. Conversely, PtoABI5 suppressed the expression of the GA biosynthesis gene, while it enhanced the expression of GA catabolic genes. Exogenous GA application partially rescued both the apical growth defects and the impaired secondary xylem development in PtoABI5-overexpressing plants. Our findings establish PtoABI5 as a central integrator, linking ABA and GA signaling pathways to coordinately arrest shoot apical growth and seasonal wood formation, thereby fine-tuning the seasonal growth cycle in perennial trees.

1. Introduction

Perennial trees in boreal and temperate latitudes orchestrate a precise seasonal growth cycle to survive the recurring challenge of winter. This adaptive strategy involves a coordinated transition from active vegetative growth in spring and summer to a state of dormancy in autumn and winter, a process synchronized primarily by the reliable environmental cue of photoperiod [1,2]. As days shorten in autumn, trees such as those in the genus Populus initiate a well-defined sequence of developmental events. Short-day (SD) conditions first trigger growth cessation at the shoot apex, culminating in bud set, the transformation of the shoot apical meristem into a dormant bud enclosed by protective scales derived from stipules [3]. Initially, this bud is in a state of ecodormancy, where growth arrest is imposed by external signals and can be reversed by returning to long-day (LD) conditions [4]. Prolonged exposure to SDs then drives the bud into a deeper physiological state of endodormancy, which requires a sustained period of chilling to be released, thereby preventing premature bud break during transient warm spells in winter [5]. Although growth cessation and the establishment of endodormancy are temporally linked and critical for winter survival, they have often been studied as distinct physiological phases. However, growing evidence suggests an overlap and interaction in their underlying regulatory networks, indicating shared hormonal and genetic control points [6,7].

The molecular framework for photoperiodic control of growth cessation in trees has been largely deciphered, revealing a pathway conserved yet distinct from the flowering network in annual plants [8]. The core of this pathway in Populus centers on FLOWERING LOCUS T2 (FT2), a key integrator of seasonal signals [9]. Under growth-permissive LDs, FT2 is highly expressed exclusively in mature leaves, and its encoded protein acts as a long-distance mobile regulator that promotes vegetative growth [10]. SD-induced downregulation of FT2 leads to the suppression of its downstream targets, including LAP1 (an ortholog of Arabidopsis APETALA1) and AIL1 (an ortholog of Arabidopsis AINTEGUMENTA), ultimately inducing growth cessation and bud set [11,12]. In addition, branching regulator BRANCHED1 (BRC1) interacts with FT2 to execute growth arrest programs under SD conditions [13]. While this FT2-centered pathway elegantly explains the photoperiodic gating of apical meristem activity, it primarily represents a signaling and transcriptional framework. The phytohormonal signals that modulate the sensitivity, amplitude, and spatial coordination of this response remain less clearly defined.

Among plant hormones, gibberellins (GAs) and abscisic acid (ABA) are established key players with opposing roles in growth-dormancy transitions [14]. GAs are potent promoters of shoot elongation and are consistently implicated in maintaining the active growth state. They act as antagonists of dormancy, in part by regulating symplastic connectivity through plasmodesmata; GA application can prevent SD-induced dormancy, while GA deficiency or inhibition promotes it [5,15,16]. Notably, GA (unlike FT2) also acts locally in the shoot apex, downstream of FT2, in seasonal growth control [16]. Conversely, ABA is classically associated with stress responses and dormancy. SDs elevate ABA levels by activating NINE-CIS-EPOXYCAROTENOID DIOXYGENASE (NCED) genes, encoding key enzymes for ABA biosynthesis [17,18]. Its levels often increase in buds during endodormancy establishment and maintenance, and it is considered a key hormone in enforcing and sustaining the dormant state [18]. ABI1 encodes a type 2C protein phosphatase that negatively regulates ABA signaling. Compared to wild-type plants, the abi1-1 mutant shows a reduced response to ABA and significantly lower plasmodesmata (PD) closure frequency under short-day (SD) conditions. Importantly, although growth is arrested, abi1-1 plants fail to transition into endodormancy [19]. Thus, ABA appears to regulate dormancy establishment and maintenance, rather than its induction, by modulating callose metabolism-related genes and thereby influencing PD closure. However, a significant gap exists in our understanding of ABA’s role in the initial phase of SD-induced growth cessation, prior to dormancy establishment. Moreover, a critical and often overlooked aspect of seasonal adaptation is the synchronous arrest of not only primary (apical) growth but also secondary (radial) growth driven by the vascular cambium.

Secondary growth, responsible for wood formation, exhibits pronounced seasonality, resulting in the characteristic annual growth rings [20]. Cambial activity ceases in autumn and resumes in spring, yet the molecular mechanisms that couple this cyclical activity with the photoperiodically controlled arrest of the shoot apex are poorly understood. This represents a significant knowledge gap, given the ecological and economic importance of wood formation. Transcriptomic studies indicate that SD conditions alter the expression of genes involved in the metabolism and signaling of both GA and ABA in cambial region tissues [21,22], hinting at their potential role as coordinators of simultaneous growth arrest in different meristems [23]. The antagonistic relationship between ABA and GA is a cornerstone of hormonal crosstalk regulating developmental transitions, such as in seed germination [24]. Therefore, it is plausible that a similar antagonistic module, responsive to photoperiod, operates to coordinate the autumn cessation of both primary and secondary growth in trees.

ABA signaling converges on a suite of transcription factors, with ABSCISIC ACID-INSENSITIVE 5 (ABI5) being a major bZIP-type effector in Arabidopsis. ABI5 is central to ABA-mediated inhibition of germination and seedling growth and is a key node for hormonal integration. For example, ABI5 antagonizes GA signaling by repressing GA biosynthesis and interacting with DELLA proteins [24,25]. While the role of ABI5 in annual plant stress and seed biology is well-documented, its function in perennial trees, particularly in the context of seasonal growth cycles integrating photoperiod, ABA, and GA, remains completely unexplored. Whether an ABI5-mediated module exists to translate photoperiod-modulated ABA signals into coordinated growth arrest across meristems is an open question.

Here, we hypothesized that ABA, acting through ABA-responsive transcription factor ABI5, serves as a critical hormonal amplifier and integrator, synchronizing the photoperiod-induced cessation of both apical and cambial growth by modulating the antagonistic GA pathway. Using Populus tomentosa as a model, we combined physiological treatments, molecular genetics, and histological analyses to test this hypothesis. We demonstrate that exogenous ABA application potently enhances SD-induced growth cessation and, crucially, strongly inhibits cambial activity and secondary xylem formation. We identified and characterized PtoABI5 as a central SD- and ABA-inducible transcription factor. Functional analysis via overexpression reveals that PtoABI5 is sufficient to accelerate apical growth cessation and to potently suppress cambial division and xylem development under SDs. Mechanistically, we show that PtoABI5 downregulates GA biosynthetic genes and that the growth defects in PtoABI5-overexpressing plants can be rescued by exogenous GA application. Our work uncovers the ABA-ABI5-GA regulatory module as a key mechanism for integrating photoperiodic cues to synchronize the autumn cessation of primary and secondary growth in perennial trees, bridging a major gap between apical meristem and vascular cambium seasonal biology.

2. Results

2.1. Exogenous ABA Application Exacerbates Short-Day-Induced Growth Cessation in Populus Trees

While ABA is well-established in dormancy regulation, its specific role in growth cessation remains less defined. To investigate the involvement of ABA in this process, wild-type (WT) P. tomentosa plants were grown under long-day (LD: 16 h light/8 h dark) conditions for approximately four weeks and then transferred to short-day (SD: 12 h light/12 h dark) conditions for up to 30 days. Plants under SDs were additionally treated with 100 µM ABA or a mock solution. SD exposure inhibited growth, and exogenous ABA significantly enhanced this growth-inhibitory effect of SDs, leading to a greater reduction in plant height (Figure 1A,B). Whereas SD alone induced growth cessation in apical buds at 21 days, SD combined with ABA accelerated this process, triggering growth cessation at 14 days and bud set at 21 days (Figure 1C). Accordingly, SD exposure reduced the number of newly formed leaves, an effect that was further amplified by ABA application (Figure 1D). Consistent with this phenotypic observation, transcript levels of the vegetative growth marker FT2 in mature leaves (10th from the top) were significantly suppressed by SD and showed an even greater reduction under SD+ABA conditions (Figure 1E). Similarly, expression of LAP1 and AIL1 in apical buds, downstream mediators of the FT2-mediated pathway, was downregulated by SD and further inhibited by ABA treatment under SDs (Figure 1F). Thus, our findings establish that ABA plays a crucial role in promoting growth cessation, in addition to its well-known functions in dormancy establishment.

2.2. Short Days and ABA Application Inhibit Cambial Activity and Secondary Xylem Formation

To examine the impact of ABA on cambial dynamics and secondary growth, we performed histological analyses on stem cross-sections from poplars grown under LD, SD, and SD+ABA conditions. SD conditions inhibited stem radial growth, and ABA application under SDs potently inhibited secondary xylem formation, evident as a marked decrease in both the number of secondary xylem cell layers and the vessel lumen area (Figure 2A–C). SD conditions alone also suppressed xylem development, which was attributable to impaired cambial proliferation (Figure 2D). This inhibition was significantly stronger in the SD+ABA treatment, which resulted in fewer cambial cell layers per file compared to SD alone (Figure 2E).

WUSCHEL-RELATED HOMEOBOX4 (WOX4) plays a central role in regulating cambial activity. It integrates auxin and upstream TRACHEARY ELEMENT DIFFERENTIATION INHIBITORY FACTOR/PHLOEM INTERCALATED WITH XYLEM (TDIF/PXY) signaling to promote cambium cell division in both trees and herbaceous species [26,27]. In Populus, auxin preferentially accumulates in cambium cells, where it controls cambial activity by activating WOX4 expression and directly induces the expression of HB7 and HB8, two genes encoding homologous HD-ZIP III transcription factors that promote secondary xylem formation [28,29]. Consistent with SD-induced phenotypic changes, key genes related to cambial division (WOX4a/b) and xylem differentiation (HB8) were downregulated by SDs and further suppressed by exogenous ABA (Figure 2F). Previous studies have shown that ABA integrates photoperiodic cues to control cambium growth through auxin signaling. Specifically, PIN-FORMED5 (PIN5), which encodes an auxin transporter located on the endoplasmic reticulum membrane, affects cytosolic auxin concentration in cambium cells and thereby restricts radial growth in Populus [22]. Additionally, the auxin-responsive cell expansion genes EXPA1.1 and EXPA1.2 participate in normal xylem development [30]. Using these genes as markers, we observed that PIN5a and PIN5b were upregulated by SDs, consistent with prior findings, and PIN5b expression was further increased by ABA. In contrast, EXPA1.1 and EXPA1.2 expression was suppressed by SDs and further downregulated by ABA (Figure 2G). Considering that SD promotes ABA biosynthesis and accumulation in stems [22], these findings collectively support a crucial role for ABA in SD-mediated inhibition of cambial activity and secondary xylem development.

2.3. Identification and Characterization of ABI5 Homolog in Populus

The ABI5 transcription factor is a central mediator of ABA signaling involved in various developmental and stress responses [31]. To investigate the roles of ABI5 homologs in perennial trees, we performed a BLAST search using the phytozome database and retrieved protein sequences of ABI5 homologs from various species. Phylogenetic analysis led to the identification of one ABI5 ortholog in the P. tomentosa genome, designated PtoABI5 (Figure 3A and Figure S1A). PtoABI5 was constitutively expressed across multiple tissues, including young and mature leaves, apical buds, and stems (Figure S1B).

To test its responsiveness to ABA, WT plants were treated with 100 µM ABA for 6 h, followed by qPCR expression analysis of PtoABI5 in different tissues. Results showed that PtoABI5 expression was induced by ABA in stems, mature leaves, and apical buds (Figure 3B). Given the primary role of photoperiod in controlling growth cessation in Populus trees, we also examined the expression responsiveness of PtoABI5 to SD exposure in these tissues. Similar to ABA treatment, PtoABI5 transcript levels were significantly elevated under SD conditions (Figure 3C). These results suggest that PtoABI5 could function as a node integrating ABA signaling and photoperiodic regulation of growth cessation.

2.4. Overexpression of PtoABI5 Promotes Growth Cessation Under Short Days

To investigate the involvement of PtoABI5 in photoperiodic responses, we generated transgenic P. tomentosa plants overexpressing PtoABI5. Three transgenic lines with expression gradients were selected and propagated for further study (Figure 3D). Under growth-permissive LDs, PtoABI5-OE lines exhibited dose-dependent suppression of vegetative growth, characterized by reduced plant height, a phenotype reminiscent of SD-treated WT plants (Figure 3E). When WT and PtoABI5-OE plants grown in LDs for four weeks were transferred to SD conditions. PtoABI5-OE plants displayed a further reduction in plant height, indicating exacerbated growth inhibition (Figure 3F). Compared to WT, PtoABI5-OE lines (particularly L9 and L10) showed earlier apical bud growth cessation at two weeks in SDs, accelerated bud set, and produced fewer new leaves (Figure 3G,H).

We next examined expression of vegetative marker genes in PtoABI5-OE (L10) to elucidate PtoABI5’s role in promoting growth arrest. Intriguingly, while expression of FT2 in leaves and LAP1 in apical buds was not affected by PtoABI5 overexpression under LDs, it was significantly more suppressed in these transgenic lines under SDs compared to WT (Figure 3I,J). AIL1 expression was suppressed in apical buds of PtoABI5-OE plants under both LD and SD conditions (Figure 3J). These observations suggest that PtoABI5 integrates ABA signaling to trigger growth cessation under SDs.

2.5. Overexpression of PtoABI5 Enhances SD-Mediated Suppression of Cambial Activity and Secondary Xylem Development

Considering that PtoABI5 expression is induced by SD cues via enhanced ABA accumulation, we investigated whether enhanced PtoABI5 expression negatively impacts cambial activity during secondary growth. Cross-sectional analysis of a marked internode revealed that PtoABI5-OE lines had fewer secondary xylem cell layers and a smaller vessel lumen area under SDs compared to WT (Figure 4A–C). Consistently, cambial cell layers per file were also reduced in the overexpression lines, and SD-induced inhibition of cambial division was stronger in PtoABI5-OE plants (Figure 4D), indicating that PtoABI5 negatively regulates cambial activity and xylem development.

Aligning with the known inhibitory effects of SDs on cambium growth, expression levels of cambium division and differentiation genes (WOX4a, WOX4b, HB8) and auxin-responsive cell expansion genes (EXPA1.1/1.2) were suppressed in stems of PtoABI5-OE plants under LDs, and further reduced under SDs. In contrast, expression of auxin efflux transporters PIN5a and PIN5b was elevated in PtoABI5-OE plants under both LD and SD conditions (Figure 4E,F). Collectively, these results demonstrate that PtoABI5 integrates ABA and photoperiod signals to coordinately repress both apical and lateral (cambial) growth.

2.6. Exogenous GA Application Partially Rescues Growth Cessation in PtoABI5-OE Poplars

Given the well-documented antagonism between ABA and GA in seasonal growth regulation [6], and evidence that ABI5 represses GA biosynthesis in Arabidopsis [32], we hypothesized that PtoABI5 might negatively regulate GA metabolism in Populus. We found that the GA biosynthesis gene GA20ox5 was downregulated by SD exposure in WT plants and in LD-grown PtoABI5-OE plants. Conversely, GA catabolism genes GA2ox4 and GA2ox5 were upregulated in SD-treated WT plants and PtoABI5-OE plants in both stems and mature leaves (Figure 5A and Figure S2). These results suggest that PtoABI5 mediates SD cues to impede GA accumulation, which is essential for normal growth for apical meristem and vascular meristem. Additionally, qPCR analysis showed that ABA biosynthesis genes NCED1 and NCED3 were expressed at higher levels in PtoABI5-OE leaves than in WT, indicating that PtoABI5 acts at the nexus of ABA-GA signaling in controlling growth cessation.

As high concentrations of bioactive GA under SDs were sufficient to sustain shoot elongation growth under SDs, and exogenous GA application can prevent SD-induced dormancy in poplars [15,16,33], we tested whether GA could mitigate the PtoABI5-mediated phenotype. Four-week-old WT and PtoABI5-OE plants were transferred from LD to SD conditions, with PtoABI5-OE plants subjected to 50 µM GA via foliar spray and soil irrigation. Exogenous GA partially restored the plant height of PtoABI5-OE lines (L9, L10) (Figure 5B,C). Furthermore, GA treatment delayed SD-induced bud set in PtoABI5-OE plants, resulting in new leaf formation comparable to WT plants in SDs (Figure 5E). These findings indicate that GA acts downstream of the ABI5-mediated pathway to sustain growth. Correspondingly, in mature leaves of SD-treated plants, the suppressed FT2 transcript level in PtoABI5-OE plants was largely elevated by GA application (Figure 5F). Expression of LAP1 and AIL1 in apical buds was also restored by GA in the PtoABI5-OE background (Figure 5G).

2.7. Exogenous GA Application Partially Restores the Impaired Cambial Activity and Xylem Development in PtoABI5-OE Plants

GA preferentially accumulates in developing xylem and plays a crucial role in regulating cambium activity and xylem differentiation [20,34]. To investigate whether GA could rescue the inhibition of secondary growth in PtoABI5-OE plants, we sampled the marked internode (2nd below the apical bud prior to three weeks of SD treatment) for cross-sectional analysis. Under SD conditions, mock-treated PtoABI5-OE plants demonstrated inhibited secondary xylem formation compared to WT plants. GA application to PtoABI5-OE plants restored secondary xylem formation, leading to comparable xylem cell layers to WT stems under SDs (Figure 6A,B). Vessel lumen area was also significantly restored by GA application in PtoABI5-OE plants compared to mock controls (Figure 6C). Moreover, cambial division activity was similarly enhanced by GA treatment (Figure 6D). Accordingly, GA application upregulated the expression of key genes involved in cambium division (WOX4a/b), xylem development (HB8), and cell expansion (EXPA1.1/1.2) in stems of PtoABI5-OE plants. In contrast, expression of PIN5a and PIN5b, negative regulators of cambial activity, was downregulated by GA treatment (Figure 6E,F). These results confirm that GA antagonizes the ABI5-mediated ABA pathway to promote cambial activity and secondary xylem formation.

3. Discussion

The timely cessation of growth in autumn is a critical adaptive strategy for perennial trees in temperate regions, ensuring survival through winter [6,7]. This process requires the precise coordination of multiple developmental programs, including apical growth arrest, bud formation, dormancy establishment, and the slowdown of stem radial growth [3,35]. While photoperiod is the primary trigger and the function of ABA in dormancy is well-established [19], its role in the photoperiodic regulation of growth cessation is less defined. Our study advances the understanding of ABA’s role by demonstrating its involvement in synchronizing apical growth arrest with the inhibition of cambial activity. We further identify the Populus ABA-responsive transcription factor PtoABI5 as a central integrator, linking short-day perception to ABA signaling to orchestrate both apical growth cessation and stem radial growth, thereby coupling primary and secondary growth dynamics in the seasonal cycle.

3.1. ABA Coordinates the Initiation of Growth Cessation with Radial Growth Inhibition

In perennial trees, SDs primarily suppress the FT2-LAP1-AIL1 regulatory cascade and GA levels, resulting in growth cessation and bud set [11,12,16]. ABA is recognized for its roles in bud dormancy maintenance and stress acclimation [14]. Mechanistically, ABA promotes plasmodesmata (PD) closure by inducing CALLOSE SYNTHASE 1 (CALS1), thereby blocking growth-promoting signals from entering the meristem [19]. However, its role in the initial phase of growth cessation has been ambiguous. The poplar abi1-1 plants with impaired ABA responses could not enter the endodormancy phase, but showed comparable growth cessation responses in WT plants [19]. Additionally, inhibiting ABA biosynthesis with fluridone application in poplars had no effect on bud set under SDs [23], while ABA treatment can induce or accelerate growth cessation and dormancy in apple [2,36], reflecting the ambiguous roles of ABA in regulating growth-dormancy transitions.

To clarify ABA’s role in SD-induced growth cessation, we examined its effects on the initial phase. Our findings demonstrate that exogenous ABA application potently promotes SD responses, leading to earlier growth cessation, accelerated bud set, and stronger suppression of leaf production and key gene expression (e.g., FT2, LAP1, AIL1) compared to SD treatment alone (Figure 1). These findings are consistent with reported promotive effects of ABA on growth cessation and dormancy induction in apple trees [36]. Crucially, we extended the role of ABA beyond the shoot apex to the vascular cambium during secondary growth. ABA markedly enhanced the SD-induced inhibition of cambial cell division and secondary xylem formation (Figure 2). This dual effect on apical and lateral meristems suggests that ABA acts as a broad-spectrum growth inhibitor that reinforces the photoperiodic signal, ensuring a synchronized shutdown of growth across the plant body as autumn progresses. This coordinated inhibition likely serves an ecological function, preventing vulnerable new tissues from being formed when conditions become unfavorable, thereby optimizing resource allocation for winter hardening. In Arabidopsis, ABA plays a crucial role in regulating secondary cell wall (SCW) deposition and lignification [37]. Further studies are needed to elucidate the molecular mechanisms underlying ABA-mediated regulation of cambial activity and wood formation in trees.

3.2. PtoABI5 Functions as a Key Integrator of Photoperiod and ABA Signaling

ABI5 is a basic leucine zipper transcription factor with a well-established role in ABA-mediated seed physiology. Beyond ABA signaling, ABI5 has also been implicated in GA and light signaling during seed germination [25]. In this study, we identified a single, functionally conserved ABI5 homolog in P. tomentosa. PtoABI5 expression is induced not only by ABA but also by SD conditions (Figure 3A–C), indicating that it acts at the intersection of hormonal and environmental signaling pathways.

The functional significance of PtoABI5 was confirmed through transgenic analysis. Overexpression of PtoABI5 resulted in dose-dependent growth suppression (Figure 3E). Notably, the observed growth inhibition in PtoABI5-OE plants under LDs, which occurred without changes in FT2 or LAP1 expression, underscores the existence of FT2-independent pathways for growth control. In this context, PtoABI5 likely acts through its conserved function in antagonizing GA signaling to impose a growth restraint independent of the FT2 pathway. Under SDs, PtoABI5-OE plants exhibited a hypersensitive response, undergoing growth cessation and bud set more rapidly and severely than controls (Figure 3G,H). Since ABA post-translationally activates ABI5 through mechanisms such as SnRK2-mediated phosphorylation and stabilization [38], both the abundance and activity of PtoABI5 are probably maximized under SDs. This probably leads to the strong suppression of downstream targets like FT2 and LAP1. These findings suggest that PtoABI5 is not merely an ABA-response gene but a critical component that modulates the sensitivity of the photoperiodic growth cessation pathway.

The integrative role of PtoABI5 was also evident in secondary growth regulation. PtoABI5-OE plants displayed significantly enhanced inhibition of cambial activity and xylem development under SDs (Figure 4). Downregulation of key cambial regulators (WOX4a/b, HB8) and cell expansion genes (EXPA1s) in response to ABA and in PtoABI5-OE plants links this transcription factor to the transcriptional network controlling wood formation. This establishes a molecular link through which SD signals, possibly through its effect on PtoABI5 expression and/or ABA accumulation, can directly repress the cambial transcriptional program via ABI5. Therefore, PtoABI5 emerges as a master coordinator, translating combined photoperiodic and ABA cues into a unified transcriptional output that arrests both the shoot apical meristem and the vascular cambium.

It is noteworthy that ABI5 physically interacts with ELONGATED HYPOCOTYLE 5 (HY5), a positive regulator of light signaling, to fine-tune stress and developmental response in Arabidopsis [39]. In Populus, HY5a integrates photoperiodic cues to suppress SD-induced growth cessation and bud set downstream of phytochrome PHYB2 [40]. Specifically, HY5a directly activates expression of the vegetative marker FT2 while repressing the circadian oscillator LHY2, a known repressor of FT2 expression [40]. Notably, in Arabidopsis, HY5 mediates ABA response during seed germination and early seedling growth by binding to and activating ABI5 expression [41,42]. In contrast, HY5 and ABI5 play antagonistic roles in growth cessation, reflecting divergent adaptative strategies between annual herbaceous plants and woody perennials. They may even form a protein complex or co-regulate common downstream targets. Whether and how Populus HY5 directly regulates ABI5 expression, or physically interacts with ABI5 to modulate downstream pathways, remains an interesting future direction.

3.3. ABI5-Mediated Growth Arrest Involves Antagonism of Gibberellin Signaling

The antagonism between ABA and GA is a recurring theme in plant development [23]. Under SD conditions, elevated ABA levels induce expression of SHORT VEGETATIVE PHASE-LIKE (SVL), which directly activates expression of NCED3 and ABA receptors RCAR1/2, forming a positive feedback loop with ABA [43]. SVL acts as a central dormancy inducer by suppressing the growth-promoting FT/GA pathway while promoting CALS1 expression to mediate plasmodesmatal closure [43,44].

In contrast to ABA, GAs act as growth-promoting hormones that stimulate cambial division during secondary growth and sustain shoot growth even under growth-limiting SDs [15,16,34]. Our study places ABI5 within this antagonistic framework in the context of perennial growth cycles. We found that PtoABI5 overexpression negatively regulates GA accumulation by suppressing the GA biosynthesis gene GA20ox5 while upregulating catabolism genes GA2ox4 and GA2ox5 (Figure 5). Notably, ABA biosynthesis genes NCED1 and NCED3 were upregulated in PtoABI5-OE leaves under SDs (Figure S3), suggesting that PtoABI5 inhibits growth by repressing the GA pathway while activating the ABA pathway. These results are consistent with reports in Arabidopsis that ABI5 directly targets and downregulates GA biosynthesis genes GA3ox1 and GA3ox2, while upregulating ABA biosynthesis genes NCED6 and NCED9, thereby controlling both hormone levels [32]. The functional relevance of this hormonal interaction was demonstrated by the partial rescue of both apical and cambial growth defects in PtoABI5-OE plants upon exogenous GA application (Figure 5 and Figure 6). GA treatment restored expression of growth-promoting genes (FT2, LAP1, AIL1, WOX4s, HB8), sustaining shoot primary and secondary growth under SD conditions. Our results demonstrate that PtoABI5 overexpression leads to the concurrent arrest of both apical and cambial growth. However, the regulatory mechanisms and sensitivity in these distinct tissues may not be identical. Apical meristem arrest is closely associated with the photoperiod-regulated FT2 pathway, whereas cambial activity is governed by a multifaceted hormonal network, prominently featuring GA and auxin. Notably, GA serves as a central integrator, jointly regulating bud dormancy and wood development. By downregulating GA biosynthetic genes (GA20ox) and upregulating GA catabolic genes (GA2ox), PtoABI5 action depletes GA levels that are essential for both processes. Therefore, ABI5 can integrate the control of bud and stem development primarily through modulating the GA pathway. Future studies employing tissue-specific transcriptomic and hormone profiling approaches will be valuable to further dissect ABI5’s distinct direct targets in shoot apices versus the vascular cambium.

Collectively, these findings support a model in which SD conditions and associated ABA accumulation upregulate PtoABI5, which in turn represses GA biosynthesis. It is plausible that reduced bioactive GA levels promote growth cessation and dormancy, potentially via stabilization of DELLA proteins, known repressors of growth and integrators of multiple signals [45]. Supporting this, the Populus DELLA protein PtoRGL2 has been shown to promote short-day-induced growth cessation and dormancy, as well as to inhibit stem radial growth [33]. Our observation that GA acts downstream of PtoABI5 to sustain growth under SDs reinforces this sequential relationship. Thus, PtoABI5 serves as a molecular switch that shifts the hormonal balance from a GA-promoted growth state to an ABA-mediated arrested state. Further studies are needed to identify the direct downstream targets of ABI5 involved in coordinating apical bud and vascular cambium growth.

4. Materials and Methods

4.1. Plant Material Growth Conditions

Wild-type (WT) Populus tomentosa Carr. was used as the genetic background. For clonal propagation, shoot segments bearing 2–3 leaves were excised from tissue-cultured plantlets and cultivated on woody plant medium under sterile conditions. Plantlets were initially grown in small pots (9 cm diameter) for 3 weeks and then transplanted into larger pots (25 cm diameter, 20 cm height). Plants were maintained in a controlled glasshouse at 24 °C under long-day (LD; 16 h light/8 h dark) conditions for 4 weeks. Growth conditions were maintained with 10,000 lux supplemental light (provided by 20 W T8-type LED lamps) and 60% relative humidity.

4.2. Short-Day and Phytohormone Treatments

WT and transgenic plants were maintained under long-day (LD) conditions or transferred to short-day (SD; 12 h light/12 h dark) conditions for up to three weeks. After transfer to SD, shoot apices were examined every three days for growth cessation and bud set and were photographed, as previously described [46]. Independent transgenic lines per genotype (4–6 representative plants each) were analyzed. For hormone treatments, abscisic acid (ABA) (Solarbio, A8060, Beijing, China) and gibberellin GA₃ (Solarbio, G8040, Beijing, China) were dissolved in anhydrous ethanol to prepare 10 mM stock solutions, stored at −20 °C. Working solutions were prepared by diluting the ABA stock to 100 μM and the GA₃ stock to 50 μM using pure water, and stored at 4 °C. An equal concentration of anhydrous ethanol solution served as the mock control. To test whether ABA promotes SD-induced growth arrest, WT P. tomentosa plants under SD were sprayed with 100 μM ABA every two days for about three weeks, with mock treatment as control. To examine ABA regulation of ABI5 transcription, WT plants were hydroponically treated with 100 μM ABA for 6 h, after which tissues were flash-frozen in liquid nitrogen and stored at −80 °C. To determine whether GA₃ rescues premature growth arrest in PtoABI5-OE lines (L9 and L10) under SD, plants were sprayed with 50 μM GA₃ every two days for about three weeks, alongside mock-treated controls.

4.3. Identification of ABI5 Genes in Populus Species

To identify ABI5 homologs, the protein sequence of Arabidopsis thaliana ABI5 was used as a query in a BLASTP search against the protein databases of P. trichocarpa (http://plants.ensembl.org/Populus_trichocarpa/Info/Index, accessed on 15 January 2026). Candidate sequences were further verified by domain analysis to confirm the presence of complete ABI5 conserved domains. A phylogenetic tree was reconstructed from the aligned amino acid sequences using the Neighbor-Joining method in MEGA 7.0 with 1000 bootstrap replicates. Multiple sequence alignments of ABI5 proteins from different plant species were performed using DNAMAN v9.0.

4.4. Plasmid Construction and Plant Transformation

For overexpression constructs, the coding sequences of target genes were amplified from a P. tomentosa cDNA library and cloned into the pCXSN-FLAG vector downstream of the CaMV35S promoter. All constructs were introduced into Agrobacterium tumefaciens strain GV3101. Agrobacterium-mediated transformation of P. tomentosa was performed as described previously [47]. Putative transgenic plants were selected on hygromycin, and positive lines were confirmed by PCR genotyping and quantitative RT-PCR. Primers are listed in Table S1.

4.5. RNA Extraction

For gene expression analysis, the following tissues were collected separately from WT and PtoABI5-OE plants under LD or after ~3 weeks of SD treatment: a marked internode, the 10th fully expanded leaf from the apex, and the shoot apex (including unexpanded leaves or leaf primordia for non-dormant apices). For tissue-specific expression profiling, apical bud, young leaves (3rd–4th from apex), mature leaves (10–11th), young stems (3rd and 4th internodes), mature stems (10–11th internodes), lateral buds (buds from internodes 1–12), and root tips were collected. All samples were immediately frozen in liquid nitrogen and stored at −80 °C. Total RNA was extracted using the Biospin Plant Total RNA Extraction Kit (BioFlux, Beijing, China).

4.6. qRT-PCR Analysis

1 μg of total RNA was reverse-transcribed into cDNA using HiScript III RT SuperMix for qPCR (Vazyme, R433-01, Nanjing, China). RT-qPCR was performed with Universal SYBR qPCR Master Mix (Vazyme, Nanjing, China) on a qTOWER3G IVD Real-Time PCR machine (Analytik Jena AG, Jena, Germany). The Ubiquitin gene was used as an internal reference. Relative expression was calculated using the 2^−ΔΔCt method. Primers used are listed in Table S1.

4.7. Histochemical Staining and Microscopy

After 3 weeks of SD treatment, we harvested the first internode above the marked one from both WT and PtoABI5-OE. Stem sections (80 μm) were prepared using a VT1000 Vibratome (Leica, Wetzlar, Germany), stained with 0.1% (v/v) toluidine blue, and observed under a microscope (Olympus, Tokyo, Japan) and Digital Scanning Microscopy Imaging System (Precipoint M8, Freising, Germany). The number of xylem cells and cambial layers was counted, with at least 240 measurements per line. Vessel lumen area was measured using ImageJ (v1.4.3.67), with no fewer than 500 data collected per analysis.

4.8. Statistical Analyses

All data are presented as mean ± standard deviation (SD). Significant differences (p < 0.05) among multiple groups were analyzed using one-way ANOVA with the Least Significance Difference (LSD) test using the SPSS software (IBM SPSS Statistics 20). Significant differences (* p < 0.05; ** p < 0.01; *** p < 0.001) between two groups were analyzed using Student’s t-test using the GraphPad Prism software (v8.3.0).

5. Conclusions

This study establishes that ABA signaling, mediated by the transcription factor PtoABI5, is a critical component of the photoperiodic response pathway that synchronizes seasonal growth cessation in Populus. We demonstrate that ABA acts as a potent enhancer of the short-day (SD) signal, coordinating the arrest of both primary (apical) and secondary (cambial) growth. First, exogenous ABA application exacerbated SD-induced phenotypes, accelerating bud set and significantly inhibiting cambial activity and secondary xylem formation. Second, we identified PtoABI5 as an integrator, whose expression is induced by both ABA and SDs. Genetic analyses confirmed its essential role, as overexpressing PtoABI5 inhibited growth under long days and rendered plants hypersensitive to SDs, leading to a pronounced, coordinated suppression of shoot elongation and wood formation. Third, we delineated a mechanistic link to gibberellin metabolism, showing that PtoABI5 represses the GA biosynthesis gene GA20ox5 and catabolism genes GA2ox4 and GA2ox5. Exogenous GA can partially rescue the growth defects in PtoABI5-overexpressing plants.

Collectively, these findings support a model in which short days elevate PtoABI5 expression, which, in turn, amplifies the photoperiodic signal for growth arrest. PtoABI5 executes this function through a tissue-specific mechanism: (1) In leaves, it represses FT2 expression and reduces gibberellin (GA) levels, thereby downregulating the LAP1-AIL1 pathway to promote apical growth cessation and bud set. (2) In stems, it inhibits GA metabolism and modulates auxin homeostasis, together suppressing the transcriptional network that maintains cambial proliferation and xylem differentiation. Thus, PtoABI5 acts as a central integrator that connects photoperiod cues with hormonal signals (ABA and GA) to ensure a coordinated and timely cessation of growth across distinct meristematic tissues (Figure S4).