In the Heat of the Moment: ZTL-Mediated Protein Quality Control at High Temperatures

Sometimes, bad things happen to good proteins. During maturation, folding may go awry and expose hydrophobic residues normally buried in their core. Exposure to high temperatures, or oxidative stress, which is kind of a big deal for photosynthetic organisms, can also lead to protein misfolding. Hydrophobic surfaces come together and form insoluble protein aggregates that must be removed in order to avoid cellular toxicity.

Heat shock proteins (HSPs) are molecular chaperones that assist in protein folding. Among them, HSP90 ensures proper folding of the auxin receptor, disease resistance proteins, and the F-box protein ZEITLUPE (ZTL), a central component of the circadian system in Arabidopsis thaliana. New findings by Gil et al. (2017) add a circadian dimension to plant responses to heat stress and show that ZTL mediates polyubiquitination of aggregated proteins, marking them for degradation by the 26S proteasome.

The Arabidopsis HSP90 gene family is represented by seven members, of which four encode cytosolic proteins. In animals, HSP90 sequesters heat shock factors (HSFs) in the cytosol as inactive monomers. Protein misfolding caused by exposure to heat stress acts as an irresistible magnet to HSP90; HSFs are then released and free to induce the expression of other HSPs to help cells face heat stress. In Arabidopsis, HSP90 activity decreases after heat exposure, resulting in activation of HSFs (Yamada et al., 2007).

HSP90 decides on the fate of misfolded proteins, which can go one of three ways: (1) The protein is successfully renatured, (2) the protein cannot be refolded and is targeted for degradation via the proteasome, or (3) the misfolded protein is part of a protein aggregate that HSP90 cannot handle, and the whole protein blob is sent to lysosomes for disposal. A connection between HSP90 and the 26S proteasome was first drawn in yeast, where assembly of the proteasome is HSP90-dependent in vitro and in vivo (Imai et al., 2003). A direct role for HSP90 in targeting protein for degradation would demand that HSP90 interact with F-box proteins: This was demonstrated in 2011, between HSP90 and the clock protein ZTL (Kim et al., 2011). Seedlings treated with the HSP90 inhibitor geldanamycin have lower ZTL levels, resulting in higher levels of the ZTL target TIMING OF CAB2 1 (TOC1) and a long circadian period, similar to RNAi lines with a reduction in cytosolic HSP90. These experiments were performed at 23°C, so clearly HSP90 is critical for proper maturation of ZTL outside of a possible role in heat responses, but this does not preclude a role for ZTL in heat stress responses through its interaction with HSP90. Gil et al. (2017) now provide evidence that ZTL is an important player in thermotolerance.

The authors found that the ztl mutant shows reduced basal and acquired thermotolerance, while ZTL-overexpressing plants are more resistant (see figure). This is dependent on time of day, coinciding with the presence of ZTL protein. Misfolded proteins that fail to refold or be marked for degradation are deposited into insoluble aggregates. In the absence of ZTL, more proteins end up in these aggregates, and polyubiquitination is lower and slower than in wild-type seedlings. In contrast to the HSP90-HSF interaction in yeast, the HSP90-ZTL interaction is strengthened by heat shock, the consequence of which is greater amounts of TOC1 and PRR5 (two circadian targets of ZTL-mediated degradation) at high temperatures.

ZTL is important for protein quality control at high temperature. Seedlings were exposed to heat stress (40°C for 4.5 h in the middle of the day) and allowed to recover at 23°C for 5 d. Loss of ZTL function is associated with poor tolerance of heat stress, while overexpression of ZTL results in better performance. (Adapted from Gil et al. [2017], Figure 1.)

Does heat stress have any effect on the clock itself? Clear rhythms were observed in both the wild type and ztl mutant at 23°C and 28°C, but not at 40°C. Rhythms were still detectable at 35°C, but the phase of the clock had changed, indicative of phase resetting by the large temperature step up from 23°C to 35°C. In other words, protein misfolding and aggregation become obvious at 35°C and higher, conditions when circadian rhythms have largely broken down, so that the roles of ZTL in circadian function and thermotolerance happen in nonoverlapping conditions. However, ZTL does play a role in the strength of rhythmic behavior at higher temperatures, as the amplitude of rhythms in the ztl mutant gets progressively weaker and faster than in the wild type as temperatures rise.

A number of questions remain. For example, the scaffold protein GIGANTEA (GI) participates in the maturation of ZTL, together with HSP90. Are gi mutants more sensitive to heat shock, in a ZTL-dependent manner? Do other F-box proteins interact with HSP90 to selectively degrade targets under specific conditions? From a circadian perspective, the observation that the phase of the clock shifts so dramatically when shifted from 23°C to 35°C is fascinating. Finally, do plants maintain circadian rhythms at all temperatures or, as this article suggests, is there a breakdown in rhythmicity at extreme temperatures? Cold temperatures were previously reported to stop the clock at 4°C in Arabidopsis (Bieniawska et al., 2008). Only time will tell…