Spatiotemporal expression patterns of cytosolic AtHSP90-2 in Arabidopsis seedlings

ABSTRACT

Heat shock protein AtHSP90–2 is one of the three constitutive cytosolic HSP90s of Arabidopsis thaliana, which are highly homologous and show mild expression activation in response to stressful impacts. To characterize the functioning of AtHSP90–2, we have analyzed tissue-specificity of its expression during seedling development using a DsG transgenic line carrying a loss-of-function mutation of AtHSP90–2 via translational fusions with the β-glucuronidase reporter gene (GUS). Histochemical analysis during the first two weeks of seedling growth revealed AtHSP90–2 expression in all organs, as well as differences in its intensity between tissues and showed its dynamics. The tissue-specific AtHSP90–2-GUS expression pattern was shown to be maintained under heat shock and water deficit. The most prominent GUS staining was detected in the vascular system and hydathodes of cotyledons, and stipules. The basipetal gradient of AtHSP90–2 expression during leaf formation, its dynamics in developing stipules, and the high level of its expression in cells with active transport function suggest a special role for the gene in certain cellular processes.

KEYWORDS:

• Arabidopsis
• AtHSP90-2
• gene expression pattern
• developmental regulation
• stress

Introduction

The heat shock proteins 90 kDa (HSP90) are a widespread, highly conserved family of molecular chaperones, which is essential in eukaryotes. They represent 1–2% of the cellular proteins under physiological conditions, and their amount increases in response to stressful impacts¹. HSP90s mediate the maturation and functioning of a plethora of metastable proteins². In plants, the set of HSP90-dependent proteins includes key components of diverse signal transduction pathways, cell cycle regulation, cell differentiation, etc.^3,4.

Plant HSP90s are a multigene family with members of different subcellular compartments: cytosol/nuclear, plastid, mitochondria, and endoplasmic reticulum^5–8. In Arabidopsis thaliana, the HSP90 family consists of seven isoforms, four of which are referred as cytosolic members: stress-inducible AtHSP90–1 (At5g52640), and constitutively expressed AtHSP90–2 (At5g56030), AtHSP90–3 (At5g56010) and AtHSP90–4 (At5g56000). Taking into account the high sequence similarity of cytosolic HSP90 (88–97%) in A. thaliana,⁵, their functional specificity and correspondent regulatory mechanisms have been considered. It has been shown that the various cytosolic HSP90s differ in their responsiveness to distinct stressors and show diverse expression kinetics^9–11. Moreover, their expression is organ-specific and changes during plant development^9,12–15. In particular, AtHSP90–1 expression was determined in pollen grains and developing embryos under normal conditions and was strongly induced in all tissues under heat stress^12,13. On the other hand, the three constitutive cytosolic members - AtHSP90–2, -3 and -4 are highly homologous and showed mild expression activation in response to stressful impacts^10,11. A brief annotation of AtHSP90–2 and AtHSP90–3 on the Arabidopsis informational portal Araport (https://araport.org) describes their constitutive expression in all tissues and in abundance in root apical meristem, pollen, and tapetum⁹. Tight regulation of AtHSP90–2 activity was demonstrated during embryogenesis and seed germination¹³At5g56030.1 was described as AtHSP90–3;¹⁶. The dynamics of activation of this gene in certain tissues was shown in detail in heat-shocked seedlings. It was first detected in the root tips of germinated seeds, and then appeared in the vascular system of 5-day-old seedlings and in the shoot meristem and cotyledon veins of 7-day-old seedlings¹³. However, we did not have much information about the dynamics of AtHSP90–2 expression pattern under normal conditions. In this study, we focused on the spatial and temporal expression pattern of AtHSP90–2 in A. thaliana during early development under normal and stressful (heat shock and water deficit) conditions.

Materials and methods

Seeds of Arabidopsis thaliana (L.) Heynh. ecotype Columbia (Col-0) and the transgenic line GT_3_103910 (SM line, The JIC Gene Trap collection from the Exon Trapping Insert Consortium (EXOTIC) program) were obtained from the Nottingham Arabidopsis Stock Centre (NASC). GT_3_103910 provides a loss-of-function mutation of AtHSP90–2 via translational fusions with the β-glucuronidase reporter gene (GUS). The Ds transposon is inserted within exon: 5 in At5g56030.1 and exon: 4 in At5g56030.2, 32 bp upstream of the 3’-UTR, in forward orientation (TAIR database). The insertion was preliminary verified according to the JIC protocol (http://signal.salk.edu/database/T-DNA/SM.435.pdf). Homozygous plants were revealed by PCR using the transposon-specific primer Spm32c (5’–TACGAATAAGAGCGTCCATTTTAGAGTGA-3’) and insertion site-specific primers (SMF, gene-specific primer: 5’–ACTTGCACTAACGCCAAGTTC–3’; SMR, specific to non-coding sequence: 5’–TGTCTTGTAACCGGCGAATAC–3’).

Surface-sterilized seeds were incubated at 4°C in the dark for 2 days and plated on the 0.8% agar substrate containing 0.5× MS medium and 1% (w/v) sucrose. Seedlings were grown in sterile conditions at 22 ± 1°C under long-day (16 h light/8 h dark) and light intensity 110 μmol m⁻² s⁻¹. Seedlings were analyzed after 3, 5, 8, 10 and 12 days of growth. For heat shock (HS), plates with 8- and 12-day-old seedlings were exposed to 37°C for 0.5, 2, 8, and 24 h. For water deficit, 6-day-old seedlings grown on medium I (1×MS, 4.5% sucrose, 1% agar) were transplanted onto medium II (0.25×MS, 1% agar) to avoid osmotic shock and subjected to progressive desiccation of the agar medium for 5 days as described earlier¹⁷.

RNA extraction and RT-PCR analysis of AtHSP90–2 expression were carried out in Col-0 seedlings as described earlier¹⁷. Since AtHSP90–1, -2, -3, and -4 are highly homologous, one pair of primers to their conserved sequences was used for PCR: F: 5’–GCTTTCCAAGCTGAGAT–3’, R: 5’–ACTTCCTCCATCTTGCT–3.’ To assess gene expression of AtHSP90–2, digesting of amplified cDNA for four genes was performed accordingly¹⁸. To assess AtHSP90–2 mRNA level, equal amounts of cDNA were digested with restriction enzymes XhoI (ThermoSci) according to the manufacturer’s protocol. As a control, fragment of AtUBQ5 (At3g62250) was amplified with primers: F: 5’–AACC-CTTGAGGTTGAATCATCC–3’, R: 5’–GTCCTTCTTTCTGGTAAACGT–3.’ Analysis was conducted in two independent biological samples with two analytical replicates.

For histochemical GUS staining, GT_3_103910 seedlings were fixed with 4% (v/v) formaldehyde in 0.1 M phosphate buffer, pH 7.0, for 30 min on ice, washed in the same buffer, and submerged in staining solution containing 1 mM ferricyanide, 1 mM ferrocyanide, 0.05% (w/v) Triton X-100, 0.05% (w/v) 5-bromo-4-chloro-3-indolyl-b-glucuronide cyclohexylamine in 0.1 M phosphate buffer, pH 7.0¹⁹ in modification of²⁰. After vacuum infiltration, the staining reaction proceeded overnight at 37°C in the dark. Chlorophyll was removed by incubation in 70% ethanol, and then the tissue was rehydrated in water and mounted in 50% (v/v) glycerol for photography. The strain patterns were recorded using an Axio Vision Zeiss microscope (Germany) and a Canon 700D camera with a 100 mm Canon EF macro lens. Experiments with GUS staining were repeated three times with 8 to 20 plants examined per repetition.

Results and discussion

Given the high homology of cytosolic AtHSP90–1, -2, -3 and -4 in A. thaliana, AtHSP90–2 transcript level was determined by the restriction of amplified cDNA for the four HSP90s. The obtained results approved perceptible accumulation of AtHSP90–2 transcript under normal conditions, as well as its moderate response to temperature increase^11,13,18. The HS experiment showed that AtHSP90–2 was induced at 37°C in a time-dependent manner, with a maximum after 30 min (Figure 1a). Transient up-regulation of GUS expression with a maximum after 3-h exposition at the same temperature was earlier shown by fluorometric analysis of GUS activity in Arabidopsis plants transformed with the promoter construct of this gene¹³. One of the sources for the differences in expression kinetics may be possible involvements of additional regulatory elements outside the promoter region used in the work cited above.

Figure 1. AtHSP90–2 expression in Col-0 seedlings under heat shock (a) and dehydration (b). (a) 12-day-old seedlings were exposed to 37°C for different time periods (0.5–24 h). (b) Six-day-old seedlings were subjected to dehydration of the agar medium for 5 days. AtHSPsp90c indicates the RT-PCR amplification product of four cytosolic HSPsp90s (AtHSP90–1, -2, -3 and -4). Equal amounts of the cDNA product were digested with restriction enzyme XhoI for AtHSP90–2 (986 bp and 1086 bp are indicated with two arrows). AtUBQ5 was used as an internal control. Representative images from two independent experiments are shown. (First published as a part of the Supplementary Information in Acta Physiologiae Plantarum, 2021, 43: 58 by Springer Nature).

In the experiment with water deficit, AtHSP90–2 expression was activated within the first day, and then remained at a fairly stable level for 5 days of dehydration of the growth medium (Figure 1b). In our opinion, the increase in its transcript level at the beginning of the experiment could likely be caused by the change of growth medium¹⁷. Then, the maintenance of a constant level of AtHSP90–2 expression during the period of dehydration indicates that this gene is not involved in the response to water deficit, that is consistent with the absence of significant changes in its expression under sharp water shortage, osmotic stress, and ABA treatment^9,11.

To visualize AtHSP90–2 expression in the seedling tissue, we used the transgenic line GT_3_103910 with a transposon insertion in the third exon, 32 bp upstream of the 3’-UTR, which makes it possible to evaluate the expression of two transcript variants under the control of the chromosomal gene. In a heat-shock experiment, 8- and 12-day-old seedlings were exposed to 37°C for 2 h, giving a time lag for protein accumulation. Histochemical GUS staining revealed reporter gene activity in all organs, most prominent in cotyledons and leaves, especially in the vascular system and tips of the blades with hydathodes (Figure 2a–c), which is consistent with the results of Prasinos et al.¹³. A similar GUS expression pattern was observed in seedlings after dehydration for 5 days (Figure 2d–f).

Figure 2. Expression patterns of AtHSP90–2-GUS in seedlings under (a-c) heat shock and (d-f) dehydration. (a-c) Histochemical detection of GUS activity in (a) 8-day-old seedling, (b and c) cotyledon and the first leaf of 12-day-old seedling, correspondingly, after exposition to 37°C for 2 h. (d-f) Histochemical detection of GUS activity in (a) 8-day-old seedling after 3 days of dehydration of the agar medium, (b and c) cotyledon and the first leaf of 11-day-old seedling, correspondingly, after 5 days of dehydration. Scale bars, 0.5 mm (a), 0.1 mm (b, c).

Histological GUS analysis of seedlings grown under normal conditions was conducted at four growth stages, which reflect a range of morphological traits: 3-day-old seedlings – hypocotyl and cotyledon emergence (Figure 3a), 5-day-old seedlings – cotyledons opened fully (Figure 3g), 8-day-old seedlings – emergence of the first pair of leaves (Figure 3k), 12-day-old seedlings – the first pair of leaves formed fully (Figure 3o). Analysis of AtHSP90–2-GUS seedlings grown under normal conditions showed activity of the reporter gene in all organs. At the same time, differences in the intensity of its expression between tissues during seedling development were revealed. In 3-day-old seedlings, high levels of GUS staining were detected in the root, hypocotyl, mainly near its base and/or shoot apex, as well as in cotyledons, most prominent in the vascular system and hydathodes (Figure 3a–f). In 5-day-old seedlings, high level of AtHSP90–2-GUS expression was detected both in the epidermis and mesophyll of fully opened cotyledons (Figure 3i,j). Later, GUS activity remained strong in the mesophyll of cotyledons, but noticeably diminished in the epidermis (Figure 3r). In addition, weak staining was determined in stipules, formed beside the leaf primordia, while there was no obvious staining in the primordia themselves (Figure 3h). Later, the stipules at the base of the leaf reached bright blue color (Figure 3l,p). In 8-day-old seedlings, AtHSP90–2-GUS was expressed in cotyledons and stipules, as well as in the distal tip of not yet fully expanded leaves and trichomes on it, and weakly in the root (Figure 3k–n). In 12-day-old seedlings, strong GUS staining was detected in stipules, cotyledons, and leaves, most pronounced in the vascular system (Figure 3o–s). Such dynamics of AtHSP90–2 expression pattern in the organs and tissues of seedlings indicates its spatiotemporal regulation at the early stages of plant development under normal conditions, as well as complements a similar pattern in heat-shocked Arabidopsis plants¹³.

Figure 3. Expression patterns of AtHSP90–2-GUS during seedling development. Histochemical detection of GUS activity in 3-day-old (a – f), 5-day-old (g – j), 8-day-old ((k – n) and 12-day-old seedlings (o – s): (a, g, k, o) general view of seedlings, (b, h, l, p) shoot apex with developing stipules (pointed by arrows), (c, i, j, q, r) cotyledons, (m, s) the first leaf with trichomes, (d, e) root near its junction with the hypocotyl, (f, n) root. Scale bars, 1 mm (a, g, k, o), 0.1 mm (b-f, h-j, l-n, p-s).

It is noteworthy that the blue staining in the roots of 3-day-old seedlings was predominately localized in the nuclei (Figure 3e–f), but at the next terms, weak GUS activity was determined throughout the cytoplasm (Figure 3n). It is known that the amino acid sequences of cytosolic HSP90s of A. thaliana contain nuclear localization signal (NLS) and nuclear export sequence (NES)²¹. Earlier, prevailing nuclear localization was shown for AtHSP90–1 and AtHSP90–3 in root and hypocotyl cells of 7-day-old A. thaliana seedlings²¹. The presence and function of HSP90 in nucleus has been also described for animal cells^22,23. The nuclear localization of cytosolic HSP90s are inhibited by geldanamycin²¹, i.e. it depends on ATP. Our observations indicate that the nuclear/cytosolic distribution of AtHSP90–2 may change during plant development.

It is remarkable that AtHSP90–2-GUS expression formed a longitudinal gradient in rosette leaves. GUS staining was first detectable in the tip of emerging leaves (Figure 4a), and gradually moved downward during leaf growth (Figure 4b,c), finally spreading throughout the leaf when it matured (Figure 4d). Interestingly, this gene expression dynamics resembles the basipetal gradient of the transition from cell proliferation to postmitotic cell expansion during A. thaliana leaf growth^24,25. This transition is accompanied with great shifts in gene expression²⁵. Accordingly, our data may reflect the participation of AtHSP90–2 in maintaining the processes of expansion and differentiation of leaf cells. Supporting this assumption is the fact that HSP90 client proteins in plants include regulators of cell growth and differentiation^3,4. In addition, stress-related genes were found to be up-regulated as the leaves matured, increasing their stress tolerance²⁶.

Figure 4. Expression patterns of AtHSP90–2-GUS in the first leaves. Histochemical detection of GUS activity (a) in leaf primordia (pointed by arrows), (b, c) expanding blade, (d) mature leaf.

Overall, our results demonstrated that AtHSP90–2 expression was predominantly regulated in a tissue-specific manner during early development, which is maintained under heat shock and water deficit. Our results are consistent with the suggestion that the tissue specificity of the expression of constitutive cytosolic HSP90s are epistatic to heat-inducibility⁹, and reflect a correspondent hierarchy of regulatory mechanisms¹³. The basipetal gradient of its expression during leaf formation and its dynamics in stipules may suggest a distinctive role of the gene in functionally mature cells. Also, we can speculate that high levels of AtHSP90–2 expression in the vascular system and hydathodes of leaves, as well as in the junction between the hypocotyl and root indicate a prominent role of the gene in cells with active transport function. These observations may indicate the involvement of AtHSP90–2, as well as protein encoded by it, in certain cellular processes, that needs further investigation of the regulatory mechanisms of tissue specificity of its expression and the search for target proteins for its function in certain tissues.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the National Academy of Sciences of Ukraine [0112U000059].