Functional analysis of HSA32 homologs by complementation assay

The interplay between HSA32 and HSP101 has been found only in higher plants, i.e.

Arabidopsis and rice (Wu et al., 2013; Lin et al., 2014). The structural difference between the higher plant, lower plant, and prokaryotic HSA32s raised a question whether they share the same molecular function. To address this question, we introduced Arabidopsis, tomato, rice, M. polymorpha, P. patens, B. subtilis, and C. glutamicum HSA32 homologs into Arabidopsis hsa32-1 for complementation test. The transgenic lines have been generated previously for expressing the HSA32 homologous proteins from above-mentioned species except P. patens (Siou-Ying Lin, unpublished results). The maps of the

corresponding binary constructs are shown in Appendix Fig.1. To facilitate the detection of protein expression, the HSA32 homologs were all fused to 3XHA tag at the N-terminus.

The protein level of HSA32 homologs can be detected by immunoblotting using the antibody against HA-tag.

In this study, I made the construct for expression of the P. patens HSA32 protein in Arabidopsis hsa32-1. To this end, primers containing KpnI and SacI cleavage sites were designed to amplify PpHSA32 cDNA from pYC110, a plasmid containing the PpHSA32 cDNA (Wu, 2007). The PCR product was confirmed to have a correct size by gel electrophoresis and then purified (Fig. 3B). The purified product was cloned into pCR8/GW/TOPO TA vector and transformed into TOP10 E. coli. The transformed E. coli was cultured on the solid medium containing spectinomycin. The plasmids purified from eight individual colonies were digested by KpnI and SacI, which generated the digested plasmid products of expected sizes, 2847 bp and 846 bp (Fig. 3C). The result suggests that PpHSA32 was successfully cloned into pCR8/GW/TOPO TA vector. Plasmid #1 was sequenced to confirm no mutation occur in PpHSA32 and named pCR8-PpHSA32 (Fig.

3A).

PpHSA32 was then subcloned into pCAMBIA1390-3HA-AtHsa32 (Appendix Fig.

1) by replacing AtHSA32, which generated pCAMBia1390-3HA-PpHsa32 (Fig. 3D). The plasmid was transformed into TOP10 E. coli. Plasmids purified from eight colonies were

separately digested by KpnI and SacI and were confirmed the product sizes by gel electrophoresis (Fig. 3E). All eight digested plasmids showed expected product sizes.

Plasmid #1 was chosen for transformation of Arabidopsis hsa32-1.

pCAMBIA1390 derived plasmid contains hygromycin resistance gene so that seeds collected from T0 plants could be screened on the selection medium containing hygromycin B (Harrison et al., 2006). Twenty T1 seedlings that developed true leaves and showed hypocotyl elongation in the dark were selected for further seed production. T2 seeds collected from T1 plants were screened on the selection medium to isolate transgenic lines with single T-DNA insertion event. If the T1 plants carry a single T-DNA insertion event, the survival rate of T2 plants on the selection medium would be near 75%, including homozygous and heterozygous transgene. Twelve T2 plants that contain single T-DNA insertion event and express PpHSA32 under heat stress were selected to collect T3 seeds for isolating the homozygous lines. The homozygous lines that show 100%

resistance to hygromycin in T3 generation were kept for further experiments. I got seven independent homozygous lines which were named PpHSA32 #1, 2, 3, 12, 13, 14, and 18.

3.3.2 Thermotolerance assay and expression analysis of HSA32 homologs and AtHSP101 in transgenic plants

hsa32-1 shows defect under LAT condition (Wu et al., 2013), so we want to know

whether the HSA32 homologs could rescue the defect or not. Fig. 4A and C showed that only HSA32 homologs from higher plants (Arabidopsis, tomato, and rice) could complement the mutant phenotype of hsa32-1, whereas those from the other species (M.

polymorpha, P. patens, B. subtilis, and C. glutamicum) could not. The result suggests that

the angiosperm HSA32s have similar biological function.

The molecular function of HSA32 in Arabidopsis and rice is to retard the degradation of HSP101 (Wu et al., 2013; Lin et al., 2014). Immunoblotting analysis showed that the expression of Arabidopsis, tomato, and rice HSA32 homologs retarded the degradation of HSP101, while those from the other species did not (Fig. 4B, D), suggesting that the molecular function of HSA32 is conserved in Arabidopsis, tomato, and rice. The level of PpHSA32 in #14 remained relatively high after 48 h of recovery (Fig. 4B), indicating that the failure of complementation might be likely due to the functional difference between higher and lower plant HSA32s instead of the fast decay of the HSA32 homologs.

Nevertheless, the possibility of 3XHA tag disruption or low expression of MpHSA32, BsPSL, and CgPSL causing the failure of complementation cannot be excluded.

Interestingly, AtHSA32, SlHSA32, and OsHSA32 were rapidly induced after 2 h of recovery and remained high while HSA32 in wild type gradually accumulated until 24 h then gradually degraded. SlHSA32 and OsHSA32 were even expressed under normal condition. Similar to previous report (Wu et al., 2013), the transcript of HSA32 in wild

type increased 53-fold after 2 h of recovery and remained 5-fold – 6-fold after 24 – 48 h of recovery (Fig. 4E). The relative high transcript level of HSA32 in AtHSA32 complementation line before heat treatment might cause the fast induction of HSA32 after heat treatment. Although the transcript level of HSA32 in AtHSA32 complementation line before heat treatment is higher than wild type at each time point, the HSA32 protein level in control sample of AtHSA32 complementation line was lower than that in 24-h recovery sample of wild type, indicating that unknown translational or post-translational regulation mechanisms might involve in the inhibition of HSA32 accumulation under non-heat stress condition.

3.4 Functional analysis of HSP101 homologs by complementation assay