Heat stress (HS) is one of the most devastating environmental stresses that a plant can face during its life cycle. At the cellular level, HS impacts membrane fluidity, microtubule organization and activity, and the general stability of enzymes participating in a variety of physiological processes [1
]. Consequently, high temperatures negatively affect the growth of vegetative and floral organs, induce flower abortion, and cause deviations from physiological developmental transitions, including gametophytic defects [4
At the molecular level, HS causes the accumulation of misfolded proteins, which is a condition referred to as proteotoxicity. This hampers the functionality and stability of structural, enzymatic, and regulatory proteins [8
]. Protection and recovery from HS depend on the activation of a complex network of molecular responses, which are collectively called the HS response (HSR). A major feature of HSR is the transcriptional upregulation of hundreds of genes coding for proteins with a variety of biological functions, including reactive oxygen species scavengers, hormone metabolism, transcription and translation, signaling, and protein fate [9
]. Among them, a special focus has been placed on heat shock proteins (Hsps), acting as molecular chaperones, for their essential role in the maintenance of protein homeostasis under both physiological and stress conditions [15
The majority of HS-induced genes are controlled by members of the HS transcription factor (Hsf) family [18
]. Plants encode for a large number of Hsfs, which, based on the structure, domain, and functional peculiarities are categorized in classes A, B, and C [20
]. In plants, HsfA1 members are considered as essential regulators of the initial response and basal thermotolerance; these are required for the upregulation of stress-induced Hsfs, which further contribute to stress response maintenance and stimulation [11
]. Class B Hsfs are mostly involved in the repression of HSR during attenuation after heat stress, but in some cases, a co-activator function with class-A Hsfs has been reported as well [27
]. The function of the class C Hsfs remains to be explored.
The misregulation of Hsf and chaperone networks cause deviations from physiological growth and development both in vegetative and reproductive tissues [11
]. Therefore, Hsfs are embedded in a regulatory network involving different layers of control mechanisms that adjust their activity based on the cellular demands. Such a post-translational mechanism has been exemplified on the level of interaction of Hsf with chaperones, co-chaperones, and other associated factors that eventually control the activity, stability, and nucleocytoplasmic equilibrium of Hsfs [33
Heat shock binding protein (HSBP) is a conserved eukaryotic protein that primarily acts as a negative regulator of Hsfs via interaction with the oligomerization region of Hsfs [39
]. The HSBP monomer has an α-helical structure that can form trimers and hexamers via coiled-coil interactions [39
]. The nuclear translocation of HSBP upon stress and during recovery from HS is related to the inactivation of Hsfs [41
]. It has been proposed that HSBP modulates the attenuation phase of HSR. Arabidopsis thaliana
HSBP, Zea mays
EMP2/HSBP1, and Oryza sativa
HSBP1 and HSBP2 are induced by HS, but also show enhanced expression under physiological conditions in siliques, embryos, and panicles, respectively [41
]. Maize EMP2/HSBP1 functions in the early stages of kernel development under physiological conditions, which is consistent with a developmental role [42
]. Interestingly, HSBP only interacts with class A Hsfs, and the different HSBPs show a specificity for different HsfA members. Maize EMP1/HSBP1 interacts with HsfA2e, HsfA3, HsfA4d, and HsfA5, while HSBP2 binds to HsfA2c and HsfA4a [43
]. Consistent with a distinct interaction profile that pairs OsHSBP1 and OsHSBP2 with different Hsfs, transgenic lines show differences in the transcript regulation of Hsf-dependent genes such as Hsps [44
We explored the capacity of ethyl methanesulfonate (EMS)-induced mutations in the HSBP coding gene to have a positive impact on the thermotolerance of tomato plants. Tomato is an economically and dietary important crop world-wide, and has long served as a model plant for flesh fruit development [45
]. Considering that tomato is cultivated in areas that are and will be heavily affected by global warming (e.g., the Mediterranean basin) the identification or development of heat-resilient germplasm is of utmost importance for farmers and consumers. Through a TILLING screening performed on a Red Setter mutant population, we identified a mutation in the tomato HSBP1
gene. This mutation causes a Met to Ile substitution in one of the helical heptad repeats. The effect of the mutation on the HSBP activity on important Hsfs was examined in tomato protoplasts, while the impact of the mutation on plant performance under high temperatures was examined by monitoring the physiological parameters of thermotolerance. We observed a reduction of the negative regulation of the HSBP mutant on Hsf activity resulting in plants with increased thermotolerance in the absence of significant phenotype alterations under non-stressed conditions. Thereby, the HSBP1
mutant line that is identified can be further used for the genetic improvement of thermotolerance in tomato.
2. Materials and Methods
2.1. Plant Material and Stress Treatment
Phenotyping was performed in a greenhouse under controlled conditions via a Scanalyzer 3D platform (LemnaTec Gmbh) in 2-L pots, containing 1.5 kg of soil (50:50 peat moss and river sand). Before sowing, 30 units of nitrogen, 40 units of anhydrous phosphate, and 30 units of potassium oxide were added to the substrate mixture. Growth conditions were 25/20 °C day/night temperature, 65%, relative humidity with a 16 h per day photoperiod. Plants were irrigated with 100 ml of water every 3 days during the analyses. For leaf surface temperature monitoring, plants were either kept at 25 °C or exposed to a gradual temperature increase from 8:00 to 13:00, at which point the temperature reached 36 °C, remained there for 1 hour, and then gradually declined to 25 °C until 18:00. The temperature was recorded on either the youngest fully emerged leaf, or the third oldest leaf of each plant using a PAM 2500 (Walz, Germany).
2.2. TILLING Screening
For the identification of induced point mutation in HSBP1
, a TILLING platform based on Red Setter cultivar was used [46
]. DNA amplification was performed using nested PCR with gene-specific primers (HSBP1-For-ext: GGCCCTTTAAAGAACTCTCTCTG, HSBP1-Rev-ext: ATAGGCGGGTGTAGGGTTCT, HSBP1-For-int: TTGGTTCAATTTTCATGCACTT, HSBP1-Rev-int: AAAAAGGCTATAAATTTTCTATTATTGC. Internal primers were 5’-end labeled with IRDye 700 and IRDye 800 dye (LI-COR, Lincoln, NE, USA), respectively. The PCR amplifications were carried out according to the experimental conditions described previously [47
]. Mutation detection was performed by using Endonuclease ENDO I [48
] and LI-COR 4300 DNA Analyzer (LI-COR, Lincoln, NE, USA). Adobe Photoshop software was used for image analysis (Adobe Systems Inc., San Jose, CA, USA). Mutation was validated by Sanger sequencing, and its position was defined at nt 761 from the first nucleotide of the amplicon generated by primers HSBP1-For-ext/HSBP1-Rev-ext. Prediction of the impact of amino acid change on protein function was done using SIFT software [49
2.3. Genotyping of Mutant Plants
The genotyping of plants was performed as previously described [50
]. M3 seeds of families containing the HSBP1
mutant allele (HSBP1m
) were grown in a greenhouse under standard conditions and confirmed by Sanger sequencing. Homozygous plants for G761A HSBP1
mutation were identified and backcrossed to the Red Setter parent line. BC1F1 plants were selfed, and BC1F2 progenies were genotyped for G761A HSBP1
mutation. Using homozygous BC1F2 mutant plants, a further selfing was adopted to obtain BC1F3 seed stocks. BC1F3 progeny carrying the wild-type HSBP1
allele were used as control plants and are referred to here as HSBP1wt
2.4. Seedling Thermotolerance
Four-day-old seedlings were germinated in the dark at 25 °C on wet paper towels in sealed petri dishes, and were exposed to 25 °C, 39 °C, 42 °C, or 45 °C for 1 hour in a water bath. Thermotolerance was evaluated by measuring the hypocotyl length for the following 7 days.
2.5. Image Based Phenotyping: Data Acquisition and Processing
Phenotyping through image analysis was performed with a Scanalyzer 3D System (LemnaTec GmbH). Plants from each genotype were divided into two groups: control (non-stressed) and heat-shocked (1 hour at 39 °C) daily for four days. Visible light images of the plants were captured immediately after heat shock treatment.
The imaging involving three mutually orthogonal vantage points was used to evaluate the morphometric parameters of the plant, such as height and biomass [51
]. The digital biovolume was calculated from the three orthogonal images of the same plant according to the formula [52
The color classes that were chosen were determined experimentally for each experiment by examining the hue histogram. Here, only yellow and dark green are shown. The number of fruits was recorded in full maturity from the second truss. In addition, the seed number for each fruit from this truss was also determined.
2.6. Expression Constructs
The expression constructs of HSBP1
wild-type and HSBPm
genes were cloned either with an N-terminal green fluorescence protein (GFP) or HA-tag, using the appropriate primers (Table S1
). Plasmids encoding for HsfA1a, HsfA2, and HsfB1 as well as PGmHsp17-CI
::GUS have been described elsewhere [20
]. For the repressor assay, the pRT103 GUS vector was used with three inserted heat stress element (HSE) oligonucleotides downstream of the TATA box of the CaMV35S promoter. Therefore, the high constitutive activity of the CaMV35S of the vector was reduced in the presence of HSE-binding factors [53
2.7. Protoplast Preparation and GUS Reporter Assays
Mesophyll protoplasts from sterile grown tomato plants cv. Moneymaker
were isolated and transformed as previously described [22
]. Fifty thousand protoplasts were transformed with a total of 10 μg of plasmid DNA per sample consisting of 0.5 μg of each Hsf or HSBP1-expressing plasmid and 1 μg of the reporter plasmid DNA construct. The total amount of plasmids was complemented to 10 μg with a pRT-Neo mock plasmid. Following transformation, protoplasts were incubated for 6 hours at 25 °C, and β-glucuronidase (GUS) activities were determined as described previously [53
]. Alternatively, protoplasts were exposed to the indicated temperature in a water bath, collected by centrifugation, and then snap frozen in liquid nitrogen.
2.8. RNA Extraction and Transcript Analysis
Total RNA was extracted using the E.Z.N.A. Plant RNA Kit (Omega Bio-Tek, Norcross, GA, USA) following the manufacturer’s instructions. cDNA was synthesized using 1 μg of total RNA with Revert Aid reverse transcriptase (Thermo Scientific) following the manufacturer’s protocol. The expression of HSP70-1
, and HsfA2
genes was determined using quantitative real-time PCR (qRT-PCR) on a StepOnePlus (Thermo Fisher Scientific). The reaction (10 μl) consisted of gene primers (Table S1
Green FastMix Low ROX™
(Quanta Biosciencies), and the template. Thermal cycling conditions were 95 °C/3 min followed by 95 °C/15 s, 60 °C/30 s, and 72 °C/30 s for 40 cycles. Gene primers were designed using PRIMER3 (www-genome.wi.mit.edu/cgi-bin/primer/primer3.cgi/
). Data were analyzed by standard methods [55
] and presented as relative levels of gene expression using the EF1α (Solyc06g005060) and Actin (Solyc11g005330) genes as internal standards.
The subcellular localization of GFP-tagged HSBP1 and HSBP1m proteins was performed under a Leica SP5 confocal laser-scanning microscope. GFP was excited at 488 nm and mCherry at 561 nm. ENP1-mCherry was used as a nuclear marker protein [56
]. Fluorescence emission was measured at 490–548 nm (GFP) and 570–656 nm (mCherry).
2.10. Orthology Search and in Silico Structure Prediction
Orthologous genes to At-HSBP1
were identified via OrthoDB [57
]. The secondary structure was predicted and visualized by I-TASSER based on the models with lower C scores [58
]. CCBuilder with an implemented BUDE algorithm was used for the calculation of force fields scores for interaction energies based on an analysis of the wild-type MSESIISKIDEMGNRIDELE or mutant MSESIISKIDEIGNRIDELE peptide, using the following parameters: trimer oligomeric state, radius 5.1, Pitch 226, and interface angle 24 [59
2.11. Transcriptome Data
Information on the expression of selected genes was obtained by the TOMEXPRESS database. The comparison of expression profiles between Sl-HSBP1 and Hsfs or Hsps was done by Pearson correlation analysis based on transcript levels across 106 individual samples of different organs, tissues, and tomato genotypes at different developmental stages.