Here, standardized protocols are presented to assess induction of the heat shock response (HSR) in Caenorhabditis elegans using RT-qPCR at the molecular level, fluorescent reporters at the cellular level, and thermorecovery at the organismal level.
The heat shock response (HSR) is a cellular stress response induced by cytosolic protein misfolding that functions to restore protein folding homeostasis, or proteostasis. Caenorhabditis elegans occupies a unique and powerful niche for HSR research because the HSR can be assessed at the molecular, cellular, and organismal levels. Therefore, changes at the molecular level can be visualized at the cellular level and their impacts on physiology can be quantitated at the organismal level. While assays for measuring the HSR are straightforward, variations in the timing, temperature, and methodology described in the literature make it challenging to compare results across studies. Furthermore, these issues act as a barrier for anyone seeking to incorporate HSR analysis into their research. Here, a series of protocols is presented for measuring induction of the HSR in a robust and reproducible manner with RT-qPCR, fluorescent reporters, and an organismal thermorecovery assay. Additionally, we show that a widely used thermotolerance assay is not dependent on the well-established master regulator of the HSR, HSF-1, and therefore should not be used for HSR research. Finally, variations in these assays found in the literature are discussed and best practices are proposed to help standardize results across the field, ultimately facilitating neurodegenerative disease, aging, and HSR research.
The heat shock response (HSR) is a universal cellular stress response induced by cytosolic protein misfolding caused by temperature increases and other proteotoxic stresses. Activation of the HSR in Caenorhabditis elegans leads to transcriptional upregulation of heat shock genes such as hsp-70 and hsp-16.2. Many heat shock proteins (HSPs) function as molecular chaperones that restore protein folding homeostasis, or proteostasis, by directly interacting with misfolded or damaged proteins. The master regulator of the HSR is the transcription factor Heat Shock Factor 1 (HSF-1), whose activation is elegantly controlled via multiple mechanisms1.
The role of HSF-1 is not restricted to stress. HSF-1 is required for normal growth and development, as deletion of hsf-1 leads to larval arrest2. HSF-1 is also important during aging and age-related neurodegenerative diseases characterized by accumulation of protein aggregates and an inability to maintain proteostasis. Knockdown of hsf-1 causes accumulation of protein aggregates and a shortened lifespan, while overexpression of hsf-1 reduces protein aggregation and extends lifespan3,4. Therefore, regulation of HSF-1 at the molecular level has broad implications for organismal physiology and disease.
C. elegans is a powerful model organism for HSR research because the HSR can be measured at the molecular, cellular, and organismal levels4,5,6. Highlighting the power of this model, key advances in delineating the HSR pathway, such as tissue-specific differences in HSR regulation, have been discovered in C. elegans7,8. Furthermore, C. elegans is widely used for aging research and is an emerging system for modeling diseases linked to proteostasis disruption.
Although heat shock experiments with C. elegans can be quick and reproducible, there are several questions to consider before beginning. For example, which temperature should be used for induction of the HSR and how long should the worms be exposed? Is it better to use a dry incubator or a water bath? Which developmental stage should be used? Unfortunately, the methodologies used to investigate the HSR vary widely from laboratory to laboratory, causing confusion when selecting the best methodologies and making it difficult to compare results across the field.
We present robust and standardized protocols for using RT-qPCR, fluorescent reporters, and thermorecovery to measure the HSR. While these three approaches are complementary, they each have unique advantages and disadvantages. For example, RT-qPCR is the most direct and quantitative measurement of the HSR, and this assay can be easily expanded to include many different heat shock-inducible genes. However, RT-qPCR is the most expensive, can be technically difficult, and requires the use of specialized equipment. In contrast, fluorescent reporters have the advantage of measuring tissue-specific differences in HSR induction. However, they are difficult to quantitate accurately, can only measure induction above a certain threshold, and require the use of a fluorescence microscope. Additionally, the reporter strains described here are developmentally delayed compared to the standard N2 strain. Although newer reporter strains containing single-copy transgenes are available, they have not been tested here9. The third assay, thermorecovery, has the advantage of providing a physiologically relevant readout at the organismal level. However, this assay is arguably the least sensitive and most indirect. Finally, we discuss some common variations found in these assays and propose a set of best practices to facilitate research in this field.
1. Maintenance and synchronization of C. elegans
2. Fluorescent imaging of HSR reporters
3. Measurement of HSR gene expression using RT-qPCR
4. Thermorecovery assay for measuring HSR at the organismal level
Using the protocols described in this manuscript, HSR induction was measured using fluorescent reporters, RT-qPCR, and thermorecovery assays. In each case, the procedure in section 1.2 was used to generate synchronized, young adult worms that had not reached reproductive maturity.
To visualize HSR induction at the cellular level, the AM446 (hsp-70p::gfp) and CL2070 (hsp-16.2p::gfp) fluorescent reporter strains were analyzed following section 2 of the protocol. In the negative control samples without heat shock, the hsp-16.2 reporter only showed normal autofluorescence, but the hsp-70 reporter had constitutive fluorescence in the anal depressor muscle as previously reported4 (Figure 1A). After 1 h of heat shock at 33 °C, robust fluorescence was observed in both reporters; however, the pattern of expression was distinct depending on which reporter was used (Figure 1B). The hsp-70 reporter was brightest in the intestine and spermatheca, whereas the hsp-16.2 reporter was brightest in the pharynx. Additionally, the hsp-16.2 reporter had a high degree of worm-to-worm variability in the amount of induction as previously described, but the hsp-70 reporter did not13.
A commonly used variation of section 2 is to perform the heat shock in a dry incubator instead of a circulating water bath. Therefore, the difference between the two methodologies was also tested. It was found that both protocols resulted in robust induction of the two fluorescent reporters using our conditions, although a circulating water bath is recommended as a best practice (see Discussion) (Figure 1B).
To test the dependence of the reporters on the transcription factor HSF-1, feeding RNAi was used to knockdown hsf-1 before reporter induction was measured. It was found that fluorescence of both strains was severely reduced upon HSF-1 knockdown, indicating that these reporters are HSF-1-dependent as described in the literature4 (Figure 2). However, it was also observed that pharyngeal fluorescence persisted in both reporters upon hsf-1 knockdown, which is consistent with previous reports that the pharyngeal muscle is resistant to RNAi by feeding14.
To quantitate whole worm induction of the HSR at the molecular level, two endogenous HSPs were measured with RT-qPCR using section 3 of the protocol. Samples were measured in triplicate, a standard curve was generated for each of the primers, and a melt curve was analyzed for each sample for quality control. It was found that a 33 °C heat shock for 1 h resulted in more than a 2,000x increase in relative expression for two heat shock genes, hsp-70 and hsp-16.2 (Figure 3). These results show that both endogenous genes are suitable for measuring HSR induction and that a 33 °C heat shock for 1 h is sufficient to generate a substantial response. However, caution should be used in interpreting the absolute degree of heat shock induction, because the mRNA levels in the absence of heat shock are very low.
To analyze a physiological response to heat shock, an organismal thermorecovery assay was tested using section 4 of the protocol. It was found that exposure of worms to a 6 h heat shock at 33 °C led to a 20% decrease in worms with normal movement after a 48 h recovery (Figure 4A). The dependence of this assay on the HSF-1 transcription factor was tested using feeding RNAi to knockdown hsf-1 before exposing worms to the stress. It was found that knockdown of hsf-1 caused a dramatic decrease in normal movement, with >95% of worms showing jerky movement or paralysis after being prodded with a platinum wire pick.
We compared this thermorecovery assay to a widely used alternative organismal assay commonly referred to as thermotolerance. In the thermotolerance assay, worms are exposed to a continuous 35 °C temperature using a dry incubator, and the percentage of worms alive are measured at various timepoints. Using this assay, it was found that control worms continuously exposed to 35 °C died after approximately 8 h of exposure (Figure 4B). However, when the dependence of this assay on HSF-1 was tested using RNAi knockdown, it was found that inhibition of hsf-1 did not cause a decrease in thermotolerance. Similar results have been previously shown using HSF-1 mutations (see Discussion). Therefore, the use of the thermotolerance assay to measure the HSR is not recommended, and thermorecovery is the preferred method for examining the HSR at the organismal level.
Figure 1: HSR induction measured with fluorescent reporters. (A) The basal and (B) heat-inducible expression of hsp-70p::gfp and hsp-16.2p::gfp reporter strains after 1 h of heat shock at 33 °C in a water bath or incubator. Worms were raised on OP50 bacteria for 64 h, heat shocked, and then recovered at 20 °C for 8 h before imaging. For reference, the no heat-shock worms in (A) were renormalized in (B) to match the range and saturation of the heat-shocked worms. Representative images of two experimental replicates are shown. Scale bar = 250 µm. Please click here to view a larger version of this figure.
Figure 2: HSR induction measured with fluorescent reporters is dependent on HSF-1. Strains containing the hsp-70p::gfp and hsp-16.2p::gfp reporters were raised on control (L4440 empty vector) or hsf-1 RNAi plates for 64 h, exposed to a 1 h heat shock at 33 °C in a water bath, and then recovered at 20 °C for 8 h before imaging. Representative images of two experimental replicates are shown. Scale bar = 250 µm. Please click here to view a larger version of this figure.
Figure 3: HSR induction measured with RT-qPCR. N2 worms were raised on HT115 bacteria for 60 h and then heat shocked for 1 h in a 33 °C water bath. The relative mRNA levels of hsp-70 (C12C8.1) and hsp-16.2 are shown normalized to the no heat-shock control. Values plotted are the mean of four biological replicates and error bars represent ± SEM. Statistical significance was calculated using an unpaired Student's t-test. **p < 0.01. Please click here to view a larger version of this figure.
Figure 4: Thermorecovery, but not thermotolerance, is dependent on HSF-1. N2 worms were raised on control (L4440) or hsf-1 RNAi plates for 60 h and then shifted to either: (A) A 33 °C water bath for 6 h and recovered at 20 °C for 48 h before scoring for normal movement (thermorecovery), or (B) A 35 °C dry incubator and removed every 2 h until dead (thermotolerance). Each assay was done with n ≥ 30 individuals on 2 independent days. The average is shown. Please click here to view a larger version of this figure.
In the literature a wide variety of temperatures, times, and equipment have been used to assay the HSR, which has introduced unnecessary caveats and led to difficulty in comparing results between laboratories. For example, temperatures ranging anywhere from 32-37 °C and times from 15 min to several hours have been used to induce the HSR15. However, it is reported that lethality occurs as early as 3 h at 37 °C for all stages and 1.5 h for day 1 adults15. Furthermore, we show that exposure of worms to 35 °C causes lethality that is not HSF-1 dependent, making these conditions poorly suited for analysis of the HSR. In contrast, a heat shock of 33 °C for 1 h is robust enough to elicit strong induction of heat shock genes, yet mild enough to not affect worm viability. Indeed, exposure to 33 °C for as long as 6 h only causes 20% of worms to display abnormal movement. Therefore, we propose using a temperature of 33 °C and a time of 1 h as a standardized condition for RT-qPCR and fluorescent reporter assays.
Recent experiments have revealed that developmental staging of worms for HSR experiments is particularly important. It was recently shown that in C. elegans the inducibility of the HSR declines (i.e., collapses) by >50% when hermaphrodites begin egg laying5. Staging the worms correctly is critical because there are often differences in developmental timing in strains carrying mutations. If temperature-sensitive mutants are used, this will also impact results if they are not synchronized by their reproductive age. Therefore, it is recommended to carefully measure the onset of egg laying for every strain to determine when the collapse occurs. The window of time after the L4 molt and before the initiation of reproductive maturity is narrow; therefore, care must be taken so that the HSR collapse does not inadvertently cause variability in results.
In addition to developmental timing, surprisingly small changes in temperature, as little as 1 °C, can have substantial effects on the HSR. For example, thermosensory neurons in C. elegans are sensitive to temperature changes as small as ±0.05 °C16. Thus, it is imperative to use a thermometer that can accurately measure the temperature. Therefore, we propose as best practice the use of a calibrated device for temperature measurement that is precise enough to measure temperatures within ±0.1 °C. Furthermore, a thermometer with a data-logging functionality should be used to measure temperature variations across time. Many incubators are specified to have thermal variations of more than 1 °C in different parts of the incubator and across time, which can have significant effects on HSR experiments. As a best practice, we suggest using incubators that have sufficient insulation and circulation to minimize temperature fluctuations. For conducting heat shock experiments, we propose a best practice of a circulating water bath. The time it takes for an agar plate to reach a desired temperature is approximately 6-7 min in a water bath but much longer in a dry incubator15,17. However, if a circulating water bath is not available, we have shown that robust HSR induction also occurs in a dry incubator using our conditions. If a dry incubator is used, opening of the incubator for the duration of the stress should be minimized.
It is well-established that induction of heat shock genes is dependent on the master regulator of the HSR, HSF-1. Here, we present evidence that the two more indirect assays, fluorescent reporters and thermorecovery, are also dependent on HSF-1. Significantly, we found that a commonly used alternative organismal assay, thermotolerance, is not HSF-1 dependent using hsf-1 RNAi (Figure 4). Similar results have been previously reported using an hsf-1 mutant or a ttx-3 mutant, which blocks the HSR18,19,20. Together, these results indicate that the thermotolerance assay should not be used for HSR research. Furthermore, this suggests that a best practice is to test the HSF-1 dependence for any assay used to measure the HSR.
Taken together, we present a series of standardized protocols and best practices for robust and reproducible measurement of HSR induction in C. elegans. We hope that these methodologies will decrease variability in HSR experiments and increase reproducibility. Facilitating direct comparisons of HSR research between laboratories will serve to accelerate research in the HSR field. Furthermore, standardization will benefit research into aging and neurodegenerative diseases with which the HSR is intimately associated.
The authors have nothing to disclose.
This work was supported by a donation from Frank Leslie. Some strains were provided by the CGC, which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440).
18S-forward primer | TTGCGTCAACTGTGGTCGTG | ||
18S-reverse primer | CCAACAAAAAGAACCGAAGT CCTG |
||
AM446 rmIs223[phsp70::gfp; pRF4(rol-6(su1006))] | Morimoto lab | http://groups.molbiosci.northwestern.edu/morimoto/ | |
C12C8.1-forward primer | GTACTACGTACTCATGTGTCG GTATTT |
||
C12C8.1-reverse primer | ACGGGCTTTCCTTGTTTTCC | ||
CFX Connect Real-Time PCR Detection System | Bio Rad | 1855200 | |
CL2070 dvIs70 [hsp-16.2p::GFP + rol-6(su1006)] | Caenorhabditis Genetics Center (CGC) | https://cgc.umn.edu/ | |
EasyLog Thermistor Probe Data Logger with LCD | Lascar | EL-USB-TP-LCD | |
Greenough Stereo Microscope S9i Series | Leica | ||
Hard Shell 96 Well PCR Plates | Bio Rad | HSS9601 | |
hsp-16.2-forward primer | ACTTTACCACTATTTCCGTCC AGC |
||
hsp-16.2-reverse primer | CCTTGAACCGCTTCTTTCTTTG | ||
iScript cDNA Synthesis Kit | Bio Rad | 1708891 | |
iTaq Universal Sybr Green Super Mix | Bio Rad | 1725121 | |
Laser Scanning Confocal Microscope | Nikon | Eclipse 90i | |
MultiGene OptiMax Thermo Cycler | Labnet | TC9610 | |
N2 (WT) | Caenorhabditis Genetics Center (CGC) | https://cgc.umn.edu/ | |
Nanodrop Lite Spectrophotometer | Thermo Scientific | ND-LITE | |
Parafilm M Roll | Bemis | 5259-04LC | |
RapidOut DNA Removal Kit | Thermo Scientific | K2981 | |
Recirculating Heated Water Bath | Lauda Brinkmann | RE-206 | |
Traceable Platinum Ultra-Accurate Digital Thermometer | Fisher Scientific | 15-081-103 | |
TRIzol Reagent | Invitrogen | 15596026 | RNA isolation reagent |
TurboMix Attachment | Scientific Industries | SI-0564 | |
Vortex-Genie 2 | Scientific Industries | SI-0236 |