Fig 1.
paqr-1(et52) is a gof allele and an R109A substitution also acts as a gof paqr-1 allele.
(A) Cartoon representation of the PAQR-1 and PAQR-2 proteins indicating the transmembrane domains (TM), localization motif (LM) and the position of the R109C substitution encoded by the et52 allele in the cytoplasmic N terminus. The percentages of sequence identity between the various domains of PAQR-1 and PAQR-2 are indicated. (B-C) Western blot detection of HA-tagged proteins expressed from the CRISPR/Cas9-modified endogenous paqr-1(syb1401), paqr-1(syb364) and paqr-2(syb1401) alleles now encoding HA::PAQR-1, HA::PAQR-1(R109C) and HA::PAQR-2, respectively. (D) Confocal images of transgenic adult C. elegans expressing Ppaqr-1::PAQR-1WT::GFP and Ppaqr-1::PAQR-1R109C::GFP, with frequencies of transgenic worms expressing the reporter in gonad sheath cells or intestine indicated (n = 50). Arrowheads indicate enrichment of the reporter on plasma membranes. Expression in body muscle and other tissues was also observed but at much lower frequencies. (E) The glucose intolerance of paqr-2 mutant worms is abrogated by PAQR-1(R109C) but not wild-type PAQR-1 when provided as transgenes. Transgenic worms were identified by the presence of the GFP transformation marker myo-2::GFP, and the outlines of worms are indicated by dashed lines in the left panels. The right-side graph shows worm length 72 hours after placing L1s on normal plates or plates containing 20 mM glucose. (F) Expression of PAQR-1(R109A) also abrogates the glucose intolerance phenotype of paqr-2 mutant worms; the graph shows the length of worms placed as L1s on the indicated media and measured 72 hours later. The dashed lines in E and F indicate the approximate size of the L1 larvae at the start of the experiment. Significant differences compared to the paqr-2 genotype are indicated where: *p<0.05, **p<0.01 and ***p<0.001 (ns: not significant).
Fig 2.
The paqr-1(et52) allele suppresses several paqr-2 mutant phenotypes.
(A-B) Length of worms placed as L1s on the indicated media/temperature and measured 72 or 144 hours later. Note that the gof paqr-1(et52) allele but not the null paqr-1(tm3262) allele suppresses the glucose and cold intolerance of the paqr-2(tm3410) null mutant. (C) Tail tip phenotype scored on 1-day old adults. Note that the gof paqr-1(et52) suppresses the tail tip defect of the paqr-2(tm3410) mutant; note that 100% of the paqr-2 mutant worms show the withered tail tip phenotype and that this is completely suppressed by the paqr-1(et52) mutation. (D-E) The gof paqr-1(et52) allele suppresses the excess SFA and PUFA depletion in the PEs of the paqr-2(tm3410) null mutant; both the null paqr-1(tm3262) and gof paqr-1(et52) have normal levels of SFAs and PUFAs among PEs. (F-G) The double null paqr-1(tm3262);paqr-2(tm3410) mutant has a more severe excess SFA and PUFA depletion in PEs than the single null paqr-2(tm3410) mutant. (H-I) FRAP measurements showing that the gof paqr-1(et52) suppresses the low membrane fluidity found in paqr-2(tm3410) mutants grown on 20 mM glucose (GLU) or fed E. coli pre-loaded with 2 mM palmitic acid (PA). The dashed line in A indicates the approximate size of the L1 larvae at the start of the experiment. Significant differences compared to the paqr-2 genotype are indicated where: * p<0.05, **p<0.01 and ***p<0.001 (ns: not significant).
Fig 3.
The gof paqr-1(et52) allele acts via the same pathway as paqr-2 but independently from iglr-2.
(A) Simplified PAQR-2/IGLR-2 membrane homeostasis pathway. (B) Length of worms placed as L1s on the indicated media/RNAi treatment and measured 72 hours later. Note that the ability of paqr-1(et52) to suppress the poor growth defect of the null paqr-2(tm3410) mutant is dependent on the desaturases fat-5, -6, and -7, as well as the sbp-1 and mdt-15 transcription factors. Note also that these RNAi experiments are performed using the HT115 E. coli strain, which in itself is poorly tolerated by the paqr-2(tm3410) mutant[18]. (C-D) Images and length of worms placed as L1s on the indicated media and measured 72 hours later. Note that the gof paqr-1(et52) allele suppresses the lethality of the double null mutant paqr-2(tm3410); nhr-49(gk405) on normal growth media but not on plates containing 20 mM glucose. (E-F) Images and length of worms placed as L1s on the indicated media and measured 72 hours later. Note how the gof paqr-1(et52) allele suppresses the glucose intolerance of the double null mutant paqr-2(tm3410) iglr-2(et34) but not that of the single null mutant iglr-2(et34). (G) Length of worms placed as L1s on NGM media and measured after 144 hours cultivation at 15°C. Note how the gof paqr-1(et52) allele suppresses the growth intolerance of both the double null mutant paqr-2(tm3410) iglr-2(et34) and the single null mutant iglr-2(et34). The dashed lines in B, D, F and G indicate the approximate size of the L1 larvae at the start of the experiment. Significant differences compared to the paqr-2 genotype are indicated where: * p<0.05, **p<0.01 and ***p<0.001 (ns: not significant).
Fig 4.
Domain swapping experiments indicate that the intracellular domains of PAQR-2 and IGLR-2 are likely regulatory.
(A) Cartoon representation of the constructs used in this structure-function study, all driven by the paqr-2 promoter. (B), Western blots showing expression of the tested constructs; two independent transgenic lines were generated for each construct, labelled A and B. (C-F) Length of worms placed as L1s on NGM media and measured after 144 hours cultivation at 15°C. Note that pQC20.2 is an excellent suppressor of the paqr-2(tm3410) iglr-2(et34) double null mutant (panel C) but that pQC20.4 is not (panel D), which suggests that the transmembrane domains of PAQR-2 require IGLR-2 for activity. Crosses in D indicate likely lethality since it was impossible to obtain a strain with that genotype. (G) Main interpretations from the data in (C-F). The dashed lines in C-F indicate the approximate size of the L1 larvae at the start of the experiment. Significant differences compared to the paqr-2 genotype are indicated where: * p<0.05, **p<0.01 and ***p<0.001 (ns: not significant).
Fig 5.
Overexpression of AdipoR1 or AdipoR2 compensates for the absence of AdipoR2 in mammalian cells.
(A) Western blot detection of VSV-AdipoR1-FLAG transiently expressed in HEK293 cells, with GAPDH used as a loading control. (B) Snapshots from a typical FRAP experiment, with the bleached areas (large circle) and reference area (small circle) indicated before and after bleaching. (C) Thalf from FRAP experiments showing that the addition of 400 μM palmitic acid causes membrane rigidification and that this effect is attenuated by AdipoR1 overexpression (R1OE). (D) Western blot detection of HA-AdipoR2-MYC transiently expressed in HEK293 cells, with GAPDH used as a loading control. (E) Thalf from FRAP experiments showing that the addition of 400 μM palmitic acid causes membrane rigidification and that this effect is attenuated by AdipoR2 overexpression (R2OE). (F) Efficiency of the AdipoR2 siRNA knockdown normalized to non-target siRNA (NT siRNA) and quantified using QPCR. (G) Thalf from FRAP experiments showing that the addition of overexpression of AdipoR1 (R1OE) prevents membrane rigidification in cells where AdipoR2 has been silenced by siRNA. (H) FRAP curves for the AdipoR2 siRNA-treated samples from G; note the improved fluorescence recovery rate when AdipoR1 is overexpressed (R1OE). Significant differences are indicated where: * p<0.05 and ***p<0.001.
Fig 6.
Speculative mechanism of PAQR-1/PAQR-2 regulation.
We propose that the cytoplasmic domain (orange/red circles) can regulate access to the catalytic site (Rx) located within the cavity formed by the seven transmembrane domains of PAQR-1/2. In PAQR-1(+), the regulatory domain partially blocks access, resulting in a small rate of conversion of the substrate (S) to signaling product (P). The R109C mutation causes a conformational change in the cytoplasmic regulatory domain, providing greater access to the catalytic site, which results in more substrates being converted to signaling products, which explains the gain-of-function nature of this mutation. The activity of PAQR-1 is not affected by membrane composition, but is rather dose-dependent. In fluid/thin membranes, the PAQR-2 and IGLR-2 proteins readily diffuse within membranes and do not form stable complexes, though they may occasionally interact with each other; the PAQR-2 cytoplasmic inhibitory domain blocks access to the active site under these conditions. In rigid/thick membranes, the PAQR-2 protein causes local deformations of the lipid bilayer that stabilize its interaction with IGLR-2, forming energetically favored signaling nexus; the PAQR-2 cytoplasmic inhibitory domain is displaced through its interaction with IGLR-2, freeing access to the catalytic site. The model attempts to explain the regulation of the hydrolase activity (e.g. ceramidase) that likely provides a ligand for downstream targets; PAQR-1 and PAQR-2 may have additional common interaction partners (not depicted) that explain why PAQR-2 can act as an inhibitor of PAQR-1(R109C) when IGLR-2 is absent.