Small Heat Shock Proteins Are Novel Common Determinants of Alcohol and Nicotine Sensitivity in Caenorhabditis elegans

Abstract

Addiction to drugs is strongly determined by multiple genetic factors. Alcohol and nicotine produce distinct pharmacological effects within the nervous system through discrete molecular targets; yet, data from family and twin analyses support the existence of common genetic factors for addiction in general. The mechanisms underlying addiction, however, are poorly described and common genetic factors for alcohol and nicotine remain unidentified. We investigated the role that the heat shock transcription factor, HSF-1, and its downstream effectors played as common genetic modulators of sensitivity to addictive substances. Using Caenorhabditis elegans, an exemplary model organism with substance dose-dependent responses similar to mammals, we demonstrate that HSF-1 altered sensitivity to both alcohol and nicotine. Using a combination of a targeted RNAi screen of downstream factors and transgenic approaches we identified that these effects were contingent upon the constitutive neuronal expression of HSP-16.48, a small heat shock protein (HSP) homolog of human α-crystallin. Furthermore we demonstrated that the function of HSP-16.48 in drug sensitivity surprisingly was independent of chaperone activity during the heat shock stress response. Instead we identified a distinct domain within the N-terminal region of the HSP-16.48 protein that specified its function in comparison to related small HSPs. Our findings establish and characterize a novel genetic determinant underlying sensitivity to diverse addictive substances.

Keywords: HSF1; alcohol; alpha crystallin; heat shock proteins; nicotine.

Figure 1

The heat shock transcription factor, HSF-1, is a common modulator of alcohol and nicotine sensitivity. (A) Locomotion rate was determined by quantifying thrashing over a range of external ethanol concentrations after a 10-min exposure. The hsf-1 sy441 mutant allele increased ethanol sensitivity in comparison to wild type (Bristol N2). Exposure to a preconditioning heat shock (HS) reduced ethanol sensitivity in Bristol N2 worms, but increased ethanol sensitivity further for hsf-1 sy441 worms. P < 0.001 (two-way analysis of variance); N = 20 for each strain at each ethanol concentration. (B) Locomotion rate was determined by quantifying thrashing after a 10-min exposure to 400 mM of external ethanol. Transgenic rescue of hsf-1 (sy441) worms with wild-type hsf-1 driven by its endogenous promoter (hsf-1;Ex[P_hsf-1::hsf-1]) fully restored ethanol sensitivity to a level statistically equivalent to Bristol N2 wild types. Overexpression of hsf-1 in Bristol N2 wild type (wild-type;Ex[P_hsf-1::hsf-1]) did not alter ethanol sensitivity in comparison to hsf-1 sy441 or N2 worms. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 50 for each strain. (C) Locomotion rate was determined by quantifying body bends over a range of external nicotine concentrations after a 14-min exposure. The hsf-1 sy441 mutant allele had a reduced nicotine sensitivity in comparison to wild type (Bristol N2). Exposure to a preconditioning HS enhanced nicotine sensitivity in Bristol N2 worms, but had no effect on hsf-1 sy441 worms. *P < 0.05 (two-way analysis of variance); N = 20 for each strain at each ethanol concentration. (D) Locomotion rate was determined by quantifying body bends following a 14-min exposure to 1.5 µM of external nicotine. Transgenic rescue of hsf-1 (sy441) worms with hsf-1 driven by its endogenous promoter (hsf-1;Ex[P_hsf-1::hsf-1]) restored the nicotine stimulatory phenotype, significantly increasing sensitivity. Overexpression of hsf-1 in Bristol N2 (wild-type;Ex[P_hsf-1::hsf-1]) did not alter nicotine sensitivity in comparison to Bristol N2. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 20 for each strain.

Figure 2

Modulation of drug sensitivity requires neuronal HSF-1 expression. (A) In comparison to hsf-1 sy441, transgenic rescue with either the hsf-1 promoter (P_hsf-1) or a panneuronal (P_rab-3) promoter restored ethanol sensitivity equal to Bristol N2. Expression in muscle (P_myo-3) or the intestine (P_vha-6) did not rescue the hsf-1 (sy441) phenotype. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 30 for each strain. (B) In comparison to hsf-1 sy441, transgenic rescue with the panneuronal (P_rab-3) or the body-wall muscle (P_myo-3) promoters restored nicotine concentration statistically equivalent to Bristol N2, whereas rescue with the hsf-1 (P_hsf-1) promoter increased nicotine sensitivity further. Expression in the intestine (P_vha-6) had no effect. n.s., nonsignificant. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 20 for each strain.

Figure 3

RNAi of HSP-16.48 phenocopies HSF-1 in the modulation of drug sensitivity. (A) RNAi knockdown of the small HSP, hsp-16.48, statistically phenocopied the ethanol sensitivity phenotype of hsf-1 RNAi knockdown in comparison to empty vector control or knockdown of the endoplasmic reticulum chaperone hsp-3. Note that hsp-16.48/hsp-16.49 is a gene duplication and, as such, RNAi should affect expression of both genes. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 20 for each RNAi target. (B) RNAi knockdown of the small HSP, hsp-16.48, phenocopied the nicotine sensitivity phenotype of hsf-1 RNAi knockdown in comparison to empty vector control or knockdown of the endoplasmic reticulum chaperone hsp-3. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 20 for each RNAi target.

Figure 4

Neuronal HSP-16.48 overexpression alters drug sensitivity in both N2 and hsf-1 sy441. Loss-of-function mutations of individual heat shock protein (HSP) genes were analyzed. Note that, due to genetic proximity, hsp-16.1/hsp-16.48 and hsp-16.11/hsp-16.49 mutant alleles affect two heat shock protein genes. (A) In comparison to Bristol N2 and hsf-1 sy441 worms, the ethanol phenotype of hsf-1 sy441 was phenocopied by the hsp-16.1/hsp-16.48 ok577 mutant. *P < 0.05 (analysis of variance with Tukey post hoc comparisons); N = 10 for each strain. (B) In comparison to Bristol N2 and hsf-1 sy441 worms, the nicotine phenotype of hsf-1 sy441 was not phenocopied by either the hsp-16.1/hsp-16.48 ok577 or the hsp-16.11/hsp-16.49 tm1221 mutants. N = 20 for each strain. (C) Panneuronal expression of hsp-16.48 is sufficient to restore ethanol sensitivity of hsf-1 sy441 worms to a statistically equivalent level to Bristol N2 worms. n.s., nonsignificant. Panneuronal expression of hsp-16.48 reduces ethanol sensitivity of Bristol N2 worms to a significantly greater level than for hsf-1 sy441 worms. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 30 for each. (E) Panneuronal expression of hsp-16.48 is sufficient to restore nicotine sensitivity of hsf-1 sy441 worms to a statistically equivalent level to Bristol N2 worms. Panneuronal expression of hsp-16.48 increases nicotine sensitivity to a significantly greater level than for hsf-1 sy441 worms. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 20 for each strain.

Figure 5

The function of HSP-16.48 in drug sensitivity is unrelated to the heat shock stress response. Exposure to ethanol or nicotine do not activate the HS stress pathway at the concentrations used in this study. Quantitative PCR of hsp-16.48 (A), hsp-16.1 (B), and hsp-70 (C) shows an upregulation in transcript expression in response to heat shock (+HS), but not to ethanol or nicotine exposure. There was also no effect on transcript expression in response to a longer ethanol or nicotine exposure that matched the HS protocol (1-hr exposure, 1-hr recovery). (D) Photographs of worms expressing green fluorescent protein (GFP) under the control of the hsp-16.48 promoter. (Top left) GFP expression was visualized at low levels without heat shock preconditioning. The cell that most reliably expressed GFP at visual levels under basal conditions was preliminarily identified as the DB2 motorneuron (indicated by arrow). (Top right) A large increase in GFP expression occurred in most cells following exposure to our heat shock protocol, indicating the activation of the heat shock stress response. Exposure of worms to the concentrations of alcohol (bottom left) or nicotine (bottom right) used in this study did not increase GFP expression, even after long-term (16 hr) exposure. Bar, 0.1 mm. (E) Western blot of Phsp-16.48::GFP. Protein expression is upregulated in response to HS, but not acute ethanol or nicotine exposure or longer exposure matching the HS protocol. (F) Transgenic overexpression in Bristol N2 worms of hsp-16.1, but not hsp-16.48 or the HSP-70 chaperone hsp-1, induces resistance to a subsequent temperature stress. Surviving worms of the indicated strains were quantified following exposure to 33° for 16–18 hr. In comparison to Bristol N2 worms, only Bristol N2;Ex[P_hsf-1::hsp-16.1] demonstrated an improvement in survival. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 3 (of 30 animals each per strain).

Figure 6

HSP-16.48 function requires an essential seven amino acid section of its N-terminal domain. (A) Schematic demonstrating the truncation or mutations used in structure–function analysis of HSP-16.48. The position of the seven amino acids (Δ38–44) in the deletion mutant are indicated in red. The position of the Glu118 amino acid is indicated in yellow. (B) The function of HSP-16.48 in ethanol sensitivity is determined by the inclusion of amino acids 36–44 within its N terminus. In comparison to Bristol N2 worms, truncation of either the N terminus (ΔN), both the N and C termini (ΔNC), or the seven amino acids (Δ38–44) only block the effects of HSP-16.48 overexpression on ethanol sensitivity. Fusing the N terminus of HSP-16.48 to the crystallin domain and C terminus of HSP-16.1 (HSP-16.1 chimera) converts the phenotype of HSP-16.1 into HSP-16.48. An E118N mutation that blocks oligomerization of crystallin proteins has no effect on HSP-16.48 function. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 30 for each strain. (C) The function of HSP-16.48 in nicotine sensitivity is determined by both amino acids 36–44 within the N terminus plus its C-terminal domain. In comparison to Bristol N2 worms, truncation of the N terminus (ΔN), the C terminus (ΔC), both the N and C termini (ΔNC), or seven amino acids (Δ38–44) only block the effects of HSP-16.48 overexpression on nicotine sensitivity. The HSP-16.1 chimera could not convert the phenotype of HSP-16.1 into HSP-16.48. The E118N mutation had no effect on HSP-16.48 function. *P < 0.05 (one-way analysis of variance with Tukey post hoc comparisons); N = 35 for each strain.