Introduction

Cold stress is one of the major environmental factors that profoundly influence plant growth, development, and overall productivity. To avoid the deleterious effects of cold stress, plants undergo a crucial process known as cold acclimation, which enables plants to acquire freezing tolerance upon prior exposure to non-freezing cold temperatures. Cold acclimation encompasses a spectrum of intricate molecular and physiological changes within plants. These adaptations include alterations in plant membranes, the accumulation of cytoplasmic Ca2+ ions, elevation of reactive oxygen species (ROS) levels, and activation of ROS scavenging systems (Shi et al. 2018; Ding and Yang 2022; Wi et al. 2022; Kidokoro et al. 2022). Additionally, cold acclimation involves adjustments in protein and sugar synthesis, modulation of transcription factor activity, and dynamic changes in the expression of cold-responsive genes (Shi et al. 2018; Ding and Yang 2022; Wi et al. 2022; Kidokoro et al. 2022).

Cold stress elicits a robust induction of genes associated with cold stress response and freezing tolerance. Among these, the C-repeat/DRE binding factor (CBF) transcription factors, a highly conserved family in plants, act as key regulators orchestrating cold-responsive gene expression (Shi et al. 2018; Wi et al. 2022; Kidokoro et al. 2022). In Arabidopsis, three CBF genes CBF1, CBF2, and CBF3, which are tandemly arranged on chromosome 4, are rapidly induced by cold stress. Upon activation, CBFs initiate a comprehensive cold-responsive transcriptional cascade, leading to the robust expression of various cold-inducible genes such as responsive to dehydration (RD) and cold-regulated (COR) genes. This sequential gene activation ultimately contributes to the enhancement of freezing tolerance. Notably, Arabidopsis mutants with mutations in all three CBF genes exhibit a marked reduction in freezing tolerance (Jia et al. 2016; Zhao et al. 2016; Zhao and Zhu 2016), underscoring the pivotal role of CBF genes as the primary regulators of cold-inducible gene expression during the cold acclimation process.

The cold-inducible transcription of CBF genes is controlled by several transcription factors, such as inducer of CBF expression 1 (ICE1), calmodulin-binding transcription activator (CAMTA) family, and several circadian clock components including circadian clock associated 1 (CCA1), late elongated hypocotyl (LHY), and reveilles (Ding and Yang 2022; Wi et al. 2022; Kidokoro et al. 2022). ICE1, a well-characterized transcriptional activator belonging to the MYC-like basic helix-loop-helix (bHLH) transcription factor family, plays a crucial role in the cold activation of the CBF genes through direct binding to the CBF gene promoters (Chinnusamy et al. 2003; Lee et al. 2005; Dong et al. 2006). ICE overexpression enhances the cold induction of CBF gene expression, which is notably impaired in the loss-of-function ice1 mutant. Post-translational modifications tightly control ICE1 expression at the protein level. Phosphorylation of ICE1 by the cold-activated open stomata1 (OST1), a member of the SNF1-related protein kinase family, enhances the ICE1 protein stability and facilitates its binding to the CBF gene promoter, thereby positively regulating freezing tolerance (Ding et al. 2015; Zhan et al. 2015). Sumoylation of ICE1, mediated by the SUMO E3 ligase SAP AND MIZ1 domain-containing ligase 1 (SIZ1), also contributes to ICE1 stabilization, promoting CBF gene expression under cold stress (Miura et al. 2007). Conversely, the E3 ubiquitin ligase high expression of osmotically-responsive genes 1 (HOS1) mediates the ubiquitination and subsequent degradation of ICE1 in response to cold stress (Dong et al. 2006; Miura et al. 2011). Supporting this, overexpression of HOS1 suppresses the expression of CBF genes and their targets, reducing resistance to freezing stress, while the loss-of-function hos1 mutant exhibits enhanced cold-responsive gene expression (Ishitani et al. 1998; Lee et al. 2001; Dong et al. 2006).

Since the identification of the Arabidopsis ice1-1 allele in mutant screening (Chinnusamy et al. 2003), the accumulated knowledge in ICE1-mediated cold acclimation research has shaped the current consensus that ICE1 serves as a pivotal regulator in the CBF signaling pathway during cold acclimation. However, recent studies have presented evidence challenging this understanding. They demonstrated that the repression of CBF genes in the ice1-1 mutant is not caused by a mutation in the ICE1 gene, but rather by DNA methylation-mediated gene silencing induced by T-DNA insertion. Furthermore, these studies revealed that both ICE1 and its homolog gene ICE2 have little role in the cold regulation of CBF gene expression (Kidokoro et al. 2020; Kim et al. 2020; Thomashow and Torii 2020). If these findings hold true, it raises questions not only about the cold regulation of CBF gene expression by ICE1 but also about the various regulatory mechanisms governing ICE1 protein activity under cold stress. Despite these emerging controversies, the ICE1-CBF regulatory model remains prominently featured in recent literature on cold acclimatization (Ding and Yang 2022; Wi et al. 2022; Wang et al. 2023b; Zheng et al. 2023), suggesting the urgent need for the comprehensive revalidation of the ICE1-CBF regulatory model.

Dysfunction of ICE1 is known to result in developmental defects, including compromised germination and abnormal stomatal development (Kanaoka et al. 2008; Denay et al. 2014; MacGregor et al. 2019), suggesting a potential indirect effect of these developmental defects on cold-inducible CBF gene expression. To revalidate the ICE1-CBF regulator model, it is crucial to establish growth conditions where the ice1 mutant exhibits developmental characteristics similar to wild-type (WT) plants. Previous reports have demonstrated that severely reduced seedling growth in the ice1-2 mutant can be restored to WT levels through glucose supplementation (Liang and Yang 2015). In addition, ICE1 is stabilized by light but undergoes degradation in the dark via the ubiquitination-proteasome pathway mediated by the E3 ubiquitin ligase constitutive photomorphogenic1 (COP1) (Lee et al. 2017). It is therefore anticipated that germinating Arabidopsis plants on glucose-containing media and incubating them under short-day (SD) conditions, preferably with reduced light effects, would substantially alleviate the growth differential between WT plants and ice1 mutants.

In this study, Arabidopsis plants were germinated on media containing 1% glucose and incubated under SD conditions before exposure to 4 ℃. Our modified growth conditions resulted in the ice1-2 mutants displaying growth patterns comparable to those of WT plants. Using WT plants and ice1-2 mutants showing similar growth, we examined expression profiles of cold-responsive genes under cold stress. As with the findings by Kidokoro et al (2020), our results confirmed no discernible differences in the cold induction of CBF genes and their downstream targets between WT plants and ice1-2 mutants. Furthermore, the cold-induced destabilization of ICE, a process known to be facilitated by HOS1, was not observed in two distinct types of ICE-expressing transgenic plants. Correspondingly, hos1 mutants exhibited expression patterns of cold-responsive genes similar to those of WT plants. All data generated in this study indicate that the HOS1-ICE1 regulatory module is non-functional for the transcriptional activation of CBF genes and their targets under cold stress.

Materials and Methods

Plant Materials and Growth Conditions

All Arabidopsis thaliana lines used in this study were in Columbia (Col-0) background, and the Arabidopsis seeds of ice1-2 (SALK_003155), hos1-3 (SALK_069312), and hos1-5 (SAIL_1211_D02) were obtained from the Arabidopsis Biological Resource Center (ABRC). The proICE1::MYC-ICE1 ice1-2, 35S::MYC-ICE1, and 35S::GFP-ICE1 transgenic lines have been reported previously (Lee et al. 2015, 2017). Seeds were sterilized with 75% ethanol with 0.1% Triton X-100 and washed in 70% ethanol. Sterilized seeds were cold-stratified at 4 ℃ in darkness for 3 days and allowed to germinate on 1/2 × Murashige and Skoog (MS) agar under SD (8 h light/16 h dark) or LD (16 h light/8 h dark) conditions with cool white light illumination at a light intensity of 120 μmol photons m−2 s−1. Plants were grown in a controlled culture room set at 22 ℃ with a relative humidity of 60% under SD or LD conditions.

To synchronize the growth of ice1-2 mutants with Col-0 plants, both plants were germinated on MS agar supplemented with 1% glucose and grown for 10 days at 22 ℃ under SD conditions before being transferred to soil. The ice1-2 mutants, exhibiting growth as closely as possible to Col-0, were transferred into the soil alongside Col-0 plants and continued to grow at 22 ℃ under SD conditions until the onset of cold treatment. For the assays on the effects of cold stress on gene expression, 14- or 21-day-old seedlings were transferred to a cold chamber set at 4℃ at the zeitgeber time (ZT) 4 and incubated for the indicated time durations under continuous light conditions.

Gene Expression Analysis

Total RNA was extracted using TRI Reagent (ThermoFisher Scientific) and cDNA was synthesized from 2 µg of total RNA using RevertAid First Strand cDNA Synthesis kit (ThermoFisher Scientific) according to the manufacturer’s recommendations. cDNAs were diluted to 60 μL with H2O and 1 L of diluted cDNA was used for PCR amplification. Quantitative PCR reactions (qPCR) were performed in 96-well blocks using TOPreal SYBR Green qPCR PreMIX with low ROX (Enzynomic) in a final volume of 20 µL. Gene expression levels were normalized relative to the eukaryotic translation initiation factor 4A1 (eIF4A) gene (At3g13920). All qPCR reactions were conducted in biological triplicates. The comparative ΔΔCT approach was used to evaluate the relative amounts of each amplified product in the samples (Livak and Schmittgen 2001). The threshold cycle (CT) for each reaction was automatically determined by the system’s default parameters. Primers used for RT-qPCR are listed in Supplementary Table 1.

Immunoblot Analysis

Approximately 40 mg of the seedlings grown on MS-agar plates were used for the immunoblot analysis. Ten-day-old seedlings grown at 22 ℃ under long-day (LD) conditions were shifted to either 22℃ or 4℃ at ZT4 and then incubated under continuous light conditions for the indicated time duration. Whole seedlings were collected and processed by grinding them in liquid nitrogen. To extract the proteins, the ground tissue was boiled in 2 × SDS loading buffer, separated through SDS-PAGE, and then transferred onto PVDF membranes (Bio-Rad). To detect ICE1 protein levels in transgenic plants, anti-MYC (05-724, Sigma-Aldrich) and Anti-GFP (ab32146, Abcam) antibodies were used along with anti-mouse antibody conjugated to horse radish peroxidase (HRP) (ADI-SAB-100-J, Enzo) and anti-rabbit antibody conjugated to HRP (SC-2357, Santa Cruz Biotechnology), respectively. SuperSignal ECL western blotting substrate (34580, Thermo Scientific) was used for membrane development and protein signals were detected using the ImageQuant LAS 500 (GE HeathCare). To acquire loading controls, the membranes were stained with Bio-Safe Coomassie G-250 stain (1610786, Bio-Rad) or the presence of Tubulin was detected using anti-α-Tubulin antibody.

Quantification and Statistical Analysis

Data for quantification analysis were presented as mean ± standard error of the mean (SEM). Statistical analyses were performed using Prism (GraphPad) software; one-way ANOVA and Tukey’s multiple comparisons tests were used to determine the significant differences among multiple groups. Significant differences were indicated as *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Results

Establishment of Growth Conditions for Synchronizing Plant Growth

Plants with growth defects tend to be highly susceptible to stress. For this reason, establishment of growth conditions where the ice1 mutants exhibit comparable growth to WT plants prior to cold treatment is crucial for determining whether the cold regulation of CBF gene expression is mediated by ICE1.

ICE1 protein is destabilized in the dark via the COP1-mediated ubiquitination-proteasome pathway (Lee et al. 2017). Building on these previous observations, we hypothesized that the impact of functional ICE1 protein would be diminished with shorter day length, consequently reducing the growth difference between WT Col-0 plants and ice1 mutants. To test this possibility, we conducted a straightforward comparison of ICE1 protein levels in ICE1 transgenic plants grown under LD and SD conditions. Under LD conditions, ICE1 protein levels remained high during the day, with a slight decrease at night. In contrast, under SD conditions, ICE1 proteins were detected low at basal levels with no accumulation during the day (Fig. S1). These observations support the notion that SD conditions likely attenuate the effects of the ICE1 protein.

As previously reported, ice1-2 mutants, when grown on sugar-deficient MS media, exhibit markedly reduced germination and post-germination growth, with their growth defects being rescued by sugar supplementation (Liang and Yang 2015). We examine the effects of supplementing sugar molecules on media to enhance the growth of ice1-2 mutants under SD conditions. While sucrose treatments showed no significant effect, the addition of glucose markedly improved both germination and post-germination growth in ice1-2 mutants. Notably, on MS media containing 1% glucose, the growth defects observed in ice1-2 were significantly restored. However, ice1-2 mutants did not grow well under LD conditions even on MS media 1% glucose. (Figs. S2–S4).

Based on the outcomes of these experiments, we established growth conditions to ensure similar growth of Col-0 and ice1-2 before cold treatment. Both Col-0 plants and ice1-2 mutants were germinated and grown for 10 days on MS media supplemented with 1% glucose. Subsequently, Col-0 plants and ice1-2 mutants displaying a WT-like growth phenotype were transferred to soil and further grown until the initiation of cold treatments. It is important to note that all phases of plant growth occurred under SD conditions. Under these modified growth conditions, the ice1-2 mutant exhibited growth patterns comparable to those of Col-0 plants (Fig. 1A). To validate that Col-0 plants and ice1-2 mutants were at comparable growth stages, we examined the expression levels of squamosa promoter binding protein-like (SPL) genes targeted by microRNA156 (miR156), a miRNA known to mediate plant developmental age signaling (Fornara and Coupland 2009; Wang et al. 2009; Wu et al. 2009). Across all growth stages examined, no differences were observed between Col-0 plants and ice1-2 mutants in the gene expression patterns of SPL3 and SPL9, two representative targets of miR156 (Fig. 1B). Collectively, these data indicate that the ice1-2 mutants can be grown similarly to Col-0 plants under modified growth conditions.

Fig. 1
figure 1

Establishment of growth conditions devoid of differences in growth between Col-0 plant and ice1-2 mutant. A Time-dependent growth patterns of Col-0 and ice1-2 under modified growth conditions. All plants were grown under SD conditions. Col-0 and ice1-2 grown for 10 days on MS agar containing 1% glucose were transferred to soil and incubated for indicated time duration for leaf number measurement (left panel). The total leaf numbers of 16 plants were measured for each plant genotype with age and data for quantification were presented as mean ± standard error of mean (SEM) (right panel). The white scale bar represents 1 cm. B RT-qPCR analysis of SPL3, SPL9, and MIR156A gene expression in Col-0 and ice1-2. Aerial parts of seedlings grown for 12, 17, 21, 28 days under modified growth conditions were collected at the zeitgeber time (ZT) four time point for total RNA isolation. RT-qPCR was conducted in biological triplicates and data for quantification were presented as mean ± SEM. Statistically significant differences are indicated (ns not significant).

Full size image

No Differences between Col-0 and ice1-2 in the Expression of CBF and COR Genes Under Cold Stress

Arabidopsis harbors a group of genes, namely CBF1, CBF2 and CBF3, which are rapidly upregulated upon exposure to cold temperatures. This upregulation of CBF genes is subsequently accompanied by the transcriptional activation of CBF-targeted COR and RD genes, ultimately enhancing cold tolerance (Shi et al. 2018; Ding and Yang 2022; Wi et al. 2022; Kidokoro et al. 2022). Employing Col-0 plants and ice1-2 mutants grown under our modified growth conditions, we investigated the effect of the ice1 mutation on the cold induction of CBF genes and their downstream targets. Remarkably, the expression patterns of the three CBF genes under cold stress exhibited no discernible differences between Col-0 and ice1-2 (Fig. 2A). Consistent with these findings, the cold-induced expression patterns of CBF target genes COR15A, COR47, and RD29A in ice1-2 mutants were similar to those observed in Col-0 plants (Fig. 2B), suggesting that ICE1 has little role in the cold regulation of CBF-mediated signaling pathways in Arabidopsis. RT-qPCR data on cold-induction of CBF gene expression in 35S::GFP-ICE1 transgenic plants also support this (Fig. S5).

Fig. 2
figure 2

Expression patterns of cold-responsive genes in Col-0 plant and ice1-2 mutant. Plants grown under modified growth conditions for 14 or 21 days were exposed to cold temperature (4 ℃) under continuous light conditions and their aerial parts were collected 3 or 24 h later after cold exposure for total RNA extraction. The mRNA levels of mRNAs of CBF1, CBF2, and CBF3 genes (A) and three CBF target genes (B) were measured by RT-qPCR. RT-qPCR was conducted in biological triplicates and data for quantification were presented as mean ± SEM. Statistically significant differences are indicated (****, P < 0.0001; ns not significant).

Full size image

In our modified experimental conditions, CBF gene expression were not highly induced by cold treatments. To make sure that our cold treatments were done properly, we examined the cold-induced CBF3 expression using whole Col-0 seedlings grown on MS media under LD conditions. As expected, RT-qPCR data showed that the expression of CBF3 gene increased hundreds of-fold in cold-treated Col-0 plants (Fig. S6), supporting that there is nothing wrong with our cold treatments. However, when plants were grown in soil, and therefore only the aerial parts were harvested for total RNA extraction, only a 5-to 30-fold increase in CBF3 gene expression was measured in cold-treated Col-0 plants (Fig. S6). These results indicate that the huge increase in CBF gene expression by cold treatment is mainly in the roots, and that cold induction of the CBF gene in the aerial part is nevertheless clearly observed under our modified conditions.

Cold Stress Has No Effect on ICE1 Protein Abundance

Cold-induced ubiquitination and subsequent degradation of ICE1 protein has been well defined as a major proteolytic control of ICE1 protein activity to explain changes in CBF gene expression during cold acclimation process (Dong et al. 2006; Miura et al. 2011). Our results suggest that ICE1 is not involved in the cold regulation of CBF gene expression, as described previously (Kidokoro et al. 2020; Kim et al. 2020). Please note that in most previous studies, the cold-induced degradation of ICE1 is the result of using ICE1-overexpressing transgenic plants (Dong et al. 2006; Ding et al. 2015). Therefore, changes in ICE1 protein levels under cold stress need to be reconfirmed using transgenic lines expressing ICE1 protein under the control of its native gene promoter.

Immunoblot assays were conducted to examine the cold regulation of ICE1 protein abundance, employing two distinct types of ICE1 transgenic plants, proICE1::MYC-ICE1 ice1-2 and 35S::GFP-ICE1 (Lee et al. 2017). Unexpectedly, the immunoblot data revealed that ICE1 protein levels remained constantly stable regardless of exposure to cold stress (Fig. 3A, Fig. S8). Notably, there was no observed cold-induced destabilization of ICE1 protein in either transgenic plant. To address the possibility of a problem with ICE1 proteolysis in transgenic plants, we examined whether ICE protein undergoes degradation in darkness. Consistent with previous studies (Lee et al. 2017), ICE1 protein levels decreased in darkness and the dark-induced degradation of ICE1 was evident in both distinct types of transgenic plants (Fig. 3B), confirming the absence of problems in ICE1 proteolysis. Additional immunoblot assays with 35S::MYC-ICE1 transgenic plants also support that ICE1 protein abundance is not affected by cold stress (Fig. S8).

Fig. 3
figure 3

ICE1 protein accumulation in response to cold and darkness. Two kinds of ICE1 transgenic plants proICE1::MYC-ICE1 ice1-2 and 35S::GFP-ICE1 were used for immunoblot analysis of ICE1 protein abundance with anti-MYC or anti-GFP antibodies, respectively. A part of the membrane stained with Coomassie brilliant blue (C) was added as loading control. A No effect of cold on ICE1 protein abundance. Ten-day-old seedlings grown at 22 ℃ under long-day (LD) conditions were shifted to either 22 ℃ or 4 ℃ at ZT4 and then incubated under continuous light conditions for the indicated time duration. Whole seedlings were collected for total protein extraction. B Reduction of ICE1 protein levels in darkness. Ten-day-old seedlings grown at 22℃ under LD conditions were shifted to either continuous light or dark conditions at ZT4 and then incubated for the indicated time duration. Whole seedlings were collected for total protein extraction.

Full size image

HOS1 is Not Involved in the Cold Regulation of the CBF Signaling Pathway

Contrary to previous understanding, our observations indicate that ICE1 plays a limited role in the cold regulation of gene expression within the CBF signaling pathway and is not subject to degradation under cold stress. Given that the E3 ubiquitin ligase HOS1 contributed to the cold regulation of the CBF signaling pathway through mediating the cold-induced degradation of ICE1, it needs to reevaluate the role of HOS1 in the cold regulation of CBF regulon.

To address this, we analyzed the gene expression patterns of the CBF target genes in two distinct hos1 alleles, hos1-3 and hos1-5. Similar to the ice1-2 mutant, both hos1 mutants displayed comparable expression levels of CBF1, CBF2, and CBF3 genes under cold stress as observed in Col-0 (Fig. 4). This suggests that HOS1 may not be linked with the ICE1-CBF regulatory model during the cold acclimation process.

Fig. 4
figure 4

Expression patterns of cold-responsive genes in Col-0 plant and hos1 mutants. Plants grown at 22 ℃ under LD conditions for 10 days were shifted to 4 ℃ at ZT4 and then incubated for 12 h. Whole seedlings were collected for total RNA extraction. The mRNA levels of CBF1, CBF2, and CBF3 genes were measured by RT-qPCR. RT-qPCR was conducted in biological triplicates and data for quantification were presented as mean ± SEM. Statistically significant differences are indicated (*, P < 0.05; ****, P < 0.0001; ns not significant).

Full size image

Discussion

Despite the widely acknowledged role of the ICE1 transcription factor in regulating CBF gene and their targets for over a decade, recent controversies have emerged regarding the actual function of ICE1 in the context of cold stress response (Kidokoro et al. 2020; Kim et al. 2020; Thomashow and Torii 2020). To address this ambiguity regarding the involvement of ICE1 in the cold regulation of the CBF regulon, our primary objective was to establish plant growth conditions minimizing the growth defects observed in the ice1-2 mutant, thereby enabling it to exhibit growth similar to WT plants. We achieved this by modifying the plant growth conditions, germinating seeds on MS media supplemented with 1% glucose and growing them under SD conditions. Under these modified conditions, we successfully incubated the ice1-2 mutants to exhibit growth patterns closely resembling those of WT plants before subjecting them to cold treatment.

Consistent with the observations reported by Kidokoro et al (2020), our results revealed no discernible differences in the cold induction of CBF regulon. Intriguingly, under our experimental conditions, the cold-induced destabilization of ICE1, a process known to be facilitated by HOS1, was not evident. This was further supported by the similar expression patterns of CBF genes observed in hos1 mutants and WT plants. All data from our study strongly challenge the long-standing consensus derived from decades of plant research, suggesting that the HOS1-ICE1 regulatory module has little functionality in the transcriptional activation of CBF regulon under cold stress. Nevertheless, considering previous studies that have shown the direct regulation of CBF expression by ICE1 under cold stress in maize and rice (Jiang et al. 2022; Zhang et al. 2017), as well as the cold degradation of OsICE1 by OsHOS1 in rice (Zhang et al. 2017), it is crucial to establish with certainty the involvement of HOS1 in the destabilization of ICE1 through careful experimental designs.

It is widely recognized that the loss-of-function of the ICE1 gene diminishes cold tolerance, while its overexpression enhances freezing tolerance (Chinnusamy et al. 2003; Lee et al. 2005; Dong et al. 2006). However, if ICE1 does not play a role in the cold regulation of the CBF regulon, how does it contribute to the acquisition of plant tolerance to cold and freezing temperatures? Recent investigations shed light on this intriguing question. In maize, ZmICE1 has been identified as a repressor of key glutamate/asparagine biosynthesis-related asparagine synthetase genes, blocking a surge in mitochondrial reactive oxygen species and thereby augmenting cold tolerance (Jiang et al. 2022). In potatoes, under cold stress, StICE1 directly stimulates the upregulation of low temperature inducible 6A gene expression, preserving cell membrane stability and enhancing cold tolerance (Wang et al. 2023a). These studies suggest that, irrespective of cold regulation of CBF regulon, ICE1 may play a crucial role in the acquisition of plant cold tolerance by modulating the expression of genes involved in metabolic adjustments.

To gain a comprehensive understanding of how ICE1 contributes to cold tolerance in Arabidopsis, it is crucial to combine ChIP-sequencing data on ICE1 binding to target genes with transcriptome analysis for the precise identification of ICE1 target genes. In addition, given the existing ChIP-seq data showing that all three CBF genes are targets of ICE (Tang et al. 2020), an approach to explain why CBF gene expression is not affected by ice1 mutations despite ICE1 binding to CBF genes would also be needed. Our modified plant growth conditions, which enable us to grow ice1 mutants similarly to WT plants, will be very helpful in precisely identifying ICE1 target genes that are misregulated by the ice1 mutation in RNA sequencing.