Background

Mangroves are the only woody halophytes living at the confluence of land and sea along tropical and subtropical tidal wetlands [3]. The Intergovernmental Panel on Climate Change predicted that the global mean temperature will increase by 1.5 ℃ or more over the next 20 years. If the temperature rises by more than 4.0 ℃, mangroves would extend to Nan**g (32°37΄ N), Jiangsu Province [4]. From 1980 to 2009, the average temperature in the southern coastal area of China increased by 0.5 ℃ every 10 years, of which the magnitude showed a gradually increasing trend with latitude (http://www.cma.gov.cn). Consequently, some attempts have been made to transplant mangroves in higher latitude, such as Wenzhou (27°56′ N), Taizhou (28°41΄ N), Zhoushan (29°93΄ N), and Shanghai (30°53΄ N) [4,5,6,7]. Global warming facilitates the range extensions of mangroves to higher latitude, whereas the accompanying climate instability, such as extreme cold event could lead to physiological damage, mortality, and/or range contraction [8, 9]. In January 2016, a successive historic freezing event (minimum -5.5 °C) occurred in Yueqing Bay (28°20′ N), Wenzhou, which caused severe damage or even complete loss of introduced mangroves. Therefore, it is of great significance to characterize and understand how mangroves adapt and acclimate to freezing temperature at higher latitude [4, 6, 9].

Kandelia obovata, a member of the genus Kandelia in the family Rhizophoraceae, is the most cold-tolerant mangrove species [2, 9]. Moreover, due to its specific viviparous phenomena, beautiful shape, and unique floral pattern, K. obovata is an excellent coastal wetland landscape tree [Effect of exogenous ABA on freezing tolerance of K. obovata under natural frost condition

Exogenous ABA is being investigated as a novel strategy to improve plants defense against cold stress. Huang et al. [19] found that foliar application of ABA could reduce membrane lipid peroxidation and alleviate cell membrane injury through promoting Pro synthesis in sugarcane seedlings. In anthers, Sharma and Nayyar [15] found that sucrose degradation and transport is regulated by ABA and accumulates in higher amounts under cold stress. Our study showed that exogenous ABA has a promotive effect on the osmolytes, especially for SP and Pro (Fig. 1). Approximately 8,000 chilling-induced genes were observed in Arabidopsis, particularly those involved in protein biosynthesis [36]. Gilmour et al. [37] reported that over-expression of CBF3 in Arabidopsis results in multiple biochemical changes that ultimately increase the concentration of Pro and SS. In the present study, up-regulated DEGs responding to ABA treatment under freezing stress were found to be enriched in nitrogen metabolism, arginine and Pro metabolism, and peroxisome (Fig. 3e). Moreover, we also observed that ABA triggered the expression of P5CS and P5CR, but inhibited the expression of ProDH (Supplementary Fig. S5), which might account for the increase in the Pro content [38]. Pro has also been proposed to function as a molecular chaperone preventing protein aggregation, stabilizing M4 lactate dehydrogenase [39], protecting nitratereductase [40], and stabilizing ribonucleases and proteases [41] in response to environmental stress. Additionally, Pro accumulation can provide a way to buffer cytosolic pH, balance cell redox status as a ROS scavenger, and store carbon and nitrogen [38]. In chickpea, Kaur et al. [42] reported that the chilling stress injury measured as oxidative stress, electrolyte leakage, loss of chlorophyll and decrease in leaf water content was mitigated significantly after foliar application of Pro. Kumar and Yadav [43] confirmed the protective effects of exogenous Pro to cold stress in Camellia sinensis through inhibiting lipid peroxidation as well as by activating or protecting some antioxidants and glyoxalase pathway enzymes. In the bamboo, Liu et al. [44] reported that the alleviation in chilling injury might be caused by enhanced enzyme activities related to Pro metabolism. Therefore, we speculated that Pro accumulation played adaptive roles in K. obovata cold hardiness. Improvement of freezing tolerance of K. obovata via engineering Pro metabolism is an existing possibility and should be explored more extensively [25]. Notably, the SP and Pro sharply decreased with the duration of low temperature, whereas SS continuously increased, and ultimately the contents of all osmolytes returned to the original values as temperature recovered (P > 0.05), implying K. obovata did not suffer obvious frost damage after exogenous ABA usage. These results also indicated there might be a sequentially synergistic effect of osmolytes, of which SP and Pro worked immediately facing the freezing stress, and SS acted during the whole cold event.

Exogenous application of ABA could enhance the antioxidant capacity, whose effect might vary positively depending upon the ABA concentration (Fig. 1). Specifically, there were no remarkable changes for SOD and CAT during the cold event under ABA 100 mg L−1 (P > 0.05), significantly higher than those without ABA spraying initially facing the freezing stress. For POD, the decreasing amplitude was mitigated and the effect was proportional to the ABA concentration. Finally, the POD content was 1.87 times that of CK. These results suggested that exogenous ABA played an active role of radical scavenging performance in K. obovata response to cold stress, and more prominently in POD compared with SOD and CAT. Consistently, Rubio et al. [45] also found the combined effect of ABA and low-temperature treatments on the expression of CBF/DREB1 transcription factors VvCBF2, VvCBF3, VvCBF4 and VvCBF6, antioxidant and dehydrin genes, and the acquisition of freezing tolerance. As shown in Fig. 3e, DEGs enriching in peroxisome responding to cold stress after foliar application of ABA, and consequently, there were no significant varieties in MDA contents during the cold event. Those observations resonated with the results of Sandhu et al. [46], and Huang et al. [20], who reported that exogenous ABA triggered the antioxidant defense, and ultimately maintained cell membrane stability and normal function under cold stress.

Conclusions

To the best of our knowledge, this is by far the northernmost and lowest temperature field study on mangrove cold resistance. This study provides a foundation for a better understanding of the response in the osmolytes, enzymatic antioxidants, and transcriptome profiling of K. obovata under natural frost conditions at ~ 32o N, as well as the acquisition of cold-resistance capability responding to exogenous ABA. Specifically, SS played a more important role than Pro in enhancing tolerance to freezing stress. For enzymatic antioxidants, POD and CAT work collaboratively to remove hydrogen peroxide, of which CAT was more important. Transcriptome analysis further indicated that phenylpropanoid metabolism, especially the flavonoid biosynthesis, played a vital role in the cold resistance of K. obovata. Exogenous ABA application effectively alleviated the adverse effects of freezing stress on K. obovata by increasing the contents of osmotic adjustment substances and enhancing the activities of antioxidant enzyme, especially the Pro and POD. In addition, our findings also offered a sound theoretical foundation for expanding mangroves plantations in higher latitudes, as well as the development coastal landscape.

Materials and methods

Plant materials and processing

In 2014, we set up seedling garden of K. obovata of the Shupaisha wetland, which located in Wenzhou City, Zhejiang Province, China (27°56′ N, 120°51′ E). In early April 2019, ~ 120 three-year-old K. obovata seedlings of the similar size were chosen introduced to Qidong (31°59′ N, 121°46′ E), Jiangsu province, and ~ 93% seedlings survived over the summer (Fig. 5). The voucher specimen was deposited in the Zhejiang Institute of Subtropical Crops, Zhejiang Academy of Agricultural Sciences, China. The field planting area was demarcated in fifteen subplots, of which each subplot had 6 seedlings with spacing 0.5 m × 0.5 m.

Fig. 5
figure 5

Location of the original area (27o56′ N, Shupaisha Island, Wenzhou, Zhejiang Province) and introduced area (31o59′ N, Haifu Town, Qidong, Jiangsu Province) of K. obovata on the coast of south-eastern China and layout of the experimental treatment

Full size image

Cavanaugh et al. [47] identified a temperature-related ecological threshold of -4 °C for mangroves. On December 1–3, 2019, the first frost event (minimum -5.5 °C) occurred at Qidong and gave us a chance to investigate the effects and responses of K. obovata when exposed to the natural extremely cold event. At 12:00 a.m. December 1st (10.2 °C), leaves were sampled from each treatment with three subplots for a total of 9 replicate seedlings as CK. ABA (Macklin, China) was dissolved in 98% ethanol (0.1%, v/v) and “Tween-80” was used as the develo** agent (0.1%, v/v). In our preliminary experiment, the treatment of ABA 30 mg L−1 remarkably improved the cold resistance of K. obovata, whose overwintering retention rate increased from 22.6% to 55.0% from 2018 to 2019. Therefore, five different concentrations of ABA (0, 5, 25, 50, and 100 mg L−1) were set to treat K. obovata. Then at 2:00 p.m. December 1st (11.5 °C) the ABA solution was sprayed with a hand-held sprayer until run off. Finally, leaves were sampled from the same seedling at different stages of the cold event, including 6:00 a.m. on December 2nd (-4.1 °C), 12:00 p.m. on December 2nd (-5.3 °C), 12:00 a.m. on December 3rd (9.7 °C). Then they were immediately washed with distilled water, frozen in liquid nitrogen, and kept at -80℃ until required. Fortunately, there was no obvious morphological damage during this cold event and all seedlings survived in the following spring as it was a warm winter. Consistently, our previous experiment also showed that slight morphological freezing damage occurred in potted eight-month-old seedlings after 12 h subjected to -5.5 °C in the manual climatic box (Fig. S1).

Physiological and biochemical analyses

The soluble sugar (SS), soluble protein (SP), proline (Pro), superoxide dismutase (SOD), peroxidase (POD), catalase (CAT), and malondialdehyde (MDA) were measured to reveal the varieties of physiological and biochemical status. About 0.1 g of leaf samples were heated in boiling water for 30 min, then centrifuged at 5,000 rpm, and measured SS concentration of the supernatant using the anthrone colorimetric method. For the Pro detection, approximately 0.1 g of leaf samples were homogenized in 3% aqueous sulphosalicylic acid, heated at 100 ℃ for 30 min and filtered. Then the filtrate was mixed with acid-ninhydrin and glacial aceticacid (1:1:1, v/v/v) in a water bath at 100 ℃ for 1 h. Finally, the reaction mixture was extracted with toluene and the absorbance was determined at 520 nm. For the SP, SOD, POD, and CAT detection, approximately 0.1 g of leaf samples was taken into a mortar, and 1 mL ice-cold sodiumphosphate buffer solution (50 mmol L−1, pH 7.0) mixed with ethylenediaminetetraacetic acid (1.0 mmol L−1) and 2% polyvinylpyrrolidone was added in an ice bath. Then the homogenate was centrifuged at 5,000 rpm, 4℃ for 10 min, and the supernatant was collected. The SP content was measured with the coomassie brilliant blue staining method described by Bradford [48]. The SOD activity was measured following the photoreduction of nitroblue tetrazolium assay [49], while CAT and POD activity were determined according to Wang et al. [6]. For the MDA detection, ~ 0.5 g of leaf samples were placed in 5 mL 10% trichloroacetic acid and centrifuged at 5,000 rpm. Then the supernatant was mixed with 2 mL of 0.67% thiobarbituric acid, heated at 100 ℃ for 30 min, centrifuged at 5,000 rpm, and finally the absorbances at 450, 532, and 600 nm were recorded, respectively. The MDA content was calculated based on the following formula: C (µmol L−1) = 6.452 × (A532 − A600) − 0.559 × A450. Each experiment contained three biological and technical replicates.

RNA extraction, library preparation, and sequencing

A total of 24 independent RNA-Seq libraries from the K. obovata leaves of eight groups, with three biological replicates for each group, were constructed and sequenced: CK (10.2 ℃), CK (-4.1 ℃), CK (-5.3 ℃), CK (9.7 ℃), ABA-100 (10.2 ℃), ABA-100 (-4.1 ℃), ABA-100 (-5.3 ℃), and ABA-100 (9.7 ℃). RNA extraction method was provided in the Supplementary data. In addition, reverse transcription, library construction, and sequencing were performed at Shanghai Majorbio Bio-pharm Biotechnology Co., Ltd. (Shanghai, China) and specific operation procedures were provided in the Supplementary data.

De novo assembly and annotation

The raw paired-end reads were trimmed and quality controlled by SeqPrep (https://github.com/jstjohn/SeqPrep) and Sickle (https://github.com/najoshi/sickle) with default parameters. Then clean data from K. obovata samples was used to do De novo assembly with Trinity (http://trinityrnaseq.sourceforge.net/) [23]. All the assembled transcripts were searched against the (National Center for Biotechnology Information) NCBI protein non-redundant (NR), Clusters of Orthologous Genes (COG), and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases using BLASTX to identify the proteins that had the highest sequence similarity with the given transcripts to retrieve their function annotations and a typical cut-off E-values less than 1.0 × 10–5 was set. Blast2GO software (http://www.blast2go.com/b2ghome) [50] was used to get gene ontology (GO) annotations for describing biological process, cellular component, and molecular function. Metabolic pathway analysis was performed using the KEGG (http://www.genome.jp/kegg/) [51].

Differential expression analysis and functional enrichment

The expression level of each transcript was calculated according to the transcripts permillion reads (TPM) method to identify differential expression genes (DEGs). RSEM (http://deweylab.biostat.wisc.edu/rsem/) was used to quantify gene abundance [52]. Essentially, differential expression analysis was performed using the DESeq2 with FDR (P-value after adjusting for false discovery rate) ≤ 0.05 and |log2 fold change|> 1 considered to be significant [53]. The Venn diagram was constructed to explore DEGs related to cold tolerance using software available online (http://bioinformatics.psb.ugent.be/webtools/Venn/). The functional-enrichment analysis was performed to identify which DEGs was significantly enriched in GO terms and metabolic pathways at Bonferroni-corrected P ≤ 0.05 compared with the whole-transcriptome background. Moreover, GO functional enrichment and KEGG pathway analyses were also carried out by Goatools (https://github.com/tanghaibao/Goatools) and KOBAS (http://kobas.cbi.pku.edu.cn/home.do) [https://www.metaboanalyst.ca). After discarding undetectable or relative low expression genes (TPM < 10), weighted gene co-expression network analysis (WGCNA) package in R was used to generate co-expression network modules. The topological overlap-based dissimilarity measure was used to hierarchically cluster all the coding sequences [55]. To make the network show an approximate scale-free topology, using an unsigned type of topological overlap matrix (TOM), the soft threshold power β of six was chosen (model fitting index R2 > 0.8), a minimal module size of 30, and a branch merge cut height of 0.25. The module eigengene (the first principal component of a given module) value was calculated and used to evaluate the association of modules with SS, SP, Pro, SOD, CAT, POD, and MDA. Moreover, network visualization analysis for modules based on WGCNA was conducted to find the top 5 hub genes.