Skip to main page content
U.S. flag
See More

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 May 27;13(1):2982.
doi: 10.1038/s41467-022-30374-9.

PI(18:1/18:1) is a SCD1-derived lipokine that limits stress signaling

Maria Thürmer  1 , André Gollowitzer #  2 , Helmut Pein #  1 , Konstantin Neukirch #  2 , Elif Gelmez  1 , Lorenz Waltl  2 , Natalie Wielsch  3 , René Winkler  4 , Konstantin Löser  1 , Julia Grander  2 , Madlen Hotze  5 , Sönke Harder  6 , Annika Döding  7 , Martina Meßner  8   9 , Fabiana Troisi  1 , Maximilian Ardelt  8   9 , Hartmut Schlüter  6 , Johanna Pachmayr  8   9 , Óscar Gutiérrez-Gutiérrez  10 , Karl Lenhard Rudolph  11 , Kathrin Thedieck  5   12   13 , Ulrike Schulze-Späte  7 , Cristina González-Estévez  10   14   15 , Christian Kosan  4 , Aleš Svatoš  3 , Marcel Kwiatkowski  5 , Andreas Koeberle  16   17
Affiliations

PI(18:1/18:1) is a SCD1-derived lipokine that limits stress signaling

Maria Thürmer et al. Nat Commun. .

Abstract

Cytotoxic stress activates stress-activated kinases, initiates adaptive mechanisms, including the unfolded protein response (UPR) and autophagy, and induces programmed cell death. Fatty acid unsaturation, controlled by stearoyl-CoA desaturase (SCD)1, prevents cytotoxic stress but the mechanisms are diffuse. Here, we show that 1,2-dioleoyl-sn-glycero-3-phospho-(1'-myo-inositol) [PI(18:1/18:1)] is a SCD1-derived signaling lipid, which inhibits p38 mitogen-activated protein kinase activation, counteracts UPR, endoplasmic reticulum-associated protein degradation, and apoptosis, regulates autophagy, and maintains cell morphology and proliferation. SCD1 expression and the cellular PI(18:1/18:1) proportion decrease during the onset of cell death, thereby repressing protein phosphatase 2 A and enhancing stress signaling. This counter-regulation applies to mechanistically diverse death-inducing conditions and is found in multiple human and mouse cell lines and tissues of Scd1-defective mice. PI(18:1/18:1) ratios reflect stress tolerance in tumorigenesis, chemoresistance, infection, high-fat diet, and immune aging. Together, PI(18:1/18:1) is a lipokine that links fatty acid unsaturation with stress responses, and its depletion evokes stress signaling.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Programmed cell death decreases the cellular proportion of MUFAs in PI.
NIH-3T3 fibroblasts were treated with vehicle, TNFα (10 ng/ml), STS (0.3 µM), CHX (20 µg/ml), ETO (10 µM), TPG (2 µM), VAL (10 µM), MC (10 µM), or I3M (10 µM) or were serum starved (Serum) for the indicated period of time. a, d–f Cellular proportion of MUFAs in PI and PI(18:1/18:1) (left to right (LTR) P = 0.0000004, 0.00000007, 0.0000007) (a), PC (LTR P = 0.9999999986, 0.000002, 0.0000003) (d), PE (LTR P = 0.99991, 0.000000007, 0.99999992) (e), and PS (LTR P = 0.99992, 0.99998, 0.99997) (f); MUFAs: 16:1, 18:1. Data of (a), (d) is identical to w/o in Supplementary Figs. 13a and 16c, d. b Time-dependent changes of the cellular PI content. c Heatmap showing the time-dependent changes of the cellular PI profile (P = 0.0007). Data are given as percentage of vehicle control for each time point. Mean (c) or mean + s.e.m. (b) and single data (a, d–f) from n = 3 (a right panel, b, c, e, f), n = 4 (a left panel, d) independent experiments. ***P < 0.001 for the respective time point (b) or P values given vs. vehicle control (a, d–f); repeated measures one-way ANOVA (a, d–f) of log data (b) + Tukey HSD post hoc tests.
Fig. 2
Fig. 2. p38 MAPK activation accompanies the early cytotoxic decrease of MUFA-PI.
Fibroblasts were cultivated under diverse cytotoxic conditions for 48 h or as indicated. a Negative correlation (−0.6 > r > −1) between cellular p-p38 MAPK (Thr180/Tyr182) levels (at 48 h) and the proportions of phospholipid (PL) species are shown for the co-regulated lipid network described in Supplementary Fig. 1a. Correlations were calculated for mean p-p38 MAPK levels from three independent experiments. b Heatmap showing time-dependent changes in the activation of p38 MAPK compared to vehicle control for each time point. Representative Western blots are shown in Supplementary Fig. 2a. STS, excluded due to pan-kinase inhibition; gray color for I3M, not determined. c, d Phosphorylation and expression of p38 MAPK (P = 0.99995) (c) and JNK (P = 0.000007) (d). Western blots are representative of five (c) or three (b, d) independent experiments. Data of c is identical to w/o in Supplementary Fig. 13b, 16b and 25a. Mean (b) or mean + s.e.m. and single data (c, d) from n = 1 (b I3M for 24 h), n = 2 (b TNFα for 0.17 h, d for TPG), n = 3 (a, b, d), n = 4 (b, c for CHX, I3M at 48 h), n = 5 (b for 48 h, c) independent experiments. P values given vs. vehicle control; mixed-effects model (REML) + Tukey HSD post hoc tests of log data (c, d). e Counter-regulation of PI(18:1/18:1) ratios and p38 MAPK activation during VAL-induced cell death across cell lines. MCF-7 breast adenocarcinoma cells, HEK293 embryonic kidney cells, primary human monocytes, MM6 acute monocytic leukemia cells, HT29 colon adenocarcinoma cells, HeLa cervical carcinoma cells, HepG2 hepatoma cells, and HUVECs were treated with vehicle or VAL (10 µM) for 48 h. Percentage changes in cellular PI(18:1/18:1) ratios and p-p38 MAPK levels were calculated vs. vehicle (100%), and the difference to the vehicle control is presented. Representative Western blots are shown in Supplementary Fig. 2b. Detailed descriptions of datasets shown in panel e are given in Supplementary Note 2. P values given vs. vehicle control; two-tailed paired student t test.
Fig. 3
Fig. 3. SCD1 expression decreases during preparation for cell death.
Fibroblasts were cultivated under diverse cytotoxic conditions for 48 h (a–c, e) or as indicated (d). a Cellular proportion of non-esterified SFAs, MUFAs, and PUFAs. SFAs: 12:0, 14:0, 16:0, 18:0; MUFAs: 16:1, 18:1; PUFAs: 18:2, 20:4, 22:5, 22:6 (MUFA LTR P = 0.0146, 0.0128, 0.0023, 0.0155, 0.003; SFA LTR P = 0.0176, 0.0088, 0.0013, 0.0072, 0.0015). b Heatmap showing changes in the free fatty acid profile as compared to vehicle control. Data are given as percentage of the relative free fatty acid abundance. c Volcano plots highlighting free fatty acids that are strongly and significantly modulated by VAL or MC. Comparisons of the indicated treatment groups show the mean difference of percentage changes and the negative log10(adjusted P value). Adjusted P values given vs. vehicle control; two-tailed multiple unpaired student t tests from log data with correction for multiple comparisons using a two-stage linear step-up procedure by Benjamini, Krieger, and Yekutieli (false discovery rate 5%). d Heatmaps showing the time-dependent effect on Scd1, Actb, and Gapdh mRNA levels that were normalized to the total amount of cellular RNA and compared to vehicle control for each time point. e Protein expression of SCD1. Western blots are representative of seven independent experiments (LTR P = 0.000000002, 0.99¯). Mean (b–d) or mean + s.e.m. (a) and single data (e) from n = 2 (d for Scd1 and Gapdh at 6 h; Actb at 48 h for Serum), n = 3 (a–d), n = 6 (e for TNFα, Serum, I3M), n = 7 (e) independent experiments. *P < 0.05, **P < 0.01 or P values given vs. vehicle control; repeated measures one-way ANOVA (a) or mixed-effects model (REML) of log data (e) + Tukey HSD post hoc tests.
Fig. 4
Fig. 4. SCD1 inhibition lowers MUFA-PI levels and induces stress signaling.
Fibroblasts were treated with CAY10566 (CAY, 3 µM) (a–c, f–l), ETO (10 µM) (i) or Scd1 siRNA (d, e) for 48 h. a, e Phosphorylation of p38 MAPK. Western blots are representative of three independent experiments. Data for e is identical to Fig. 6h. b Cellular proportion of SFA-CoAs and MUFA-CoAs. SFA: 16:0-CoA; MUFAs: 16:1-CoA, 18:1-CoA (SFA P = 0.0168, MUFA P = 0.0168). c, d Cellular proportion of PI- and PC-bound MUFAs (P = 0.0000099). d Non-targeting siRNA was transfected as control (ctrl). Interconnected lines indicate data from the same independent experiment. f Protein expression of BiP. g Ratio of LC3BII/LC3BI protein levels. h PARP cleavage; cl., cleaved. Western Blots are representative of four (f, h) and five (g) independent experiments. i Annexin V and propidium iodide (PI) staining. Proportion of annexin V / PI negative cells (viable cells), annexin V positive / PI negative cells (early apoptotic cells), annexin V positive / PI positive cells (late apoptotic cells), and annexin V negative / PI positive cells (necrotic cells) as percentage of total cells. Data are identical to Fig. 6e. Cytograms are shown in Supplementary Fig. 11b. j Fibroblast morphology; scale bar, 100 µm. Phase contrast images are representative of three independent experiments. k Cell numbers. l PIP3 levels determined by ELISA. Paired data (d) or mean + s.e.m. (b, i) and single data (a, c, e–h, k, l) from n = 3 (a, d, e, ctrl), n = 4 (b, f, h, i, k, l), n = 5 (g), n = 7 (c) independent experiments and n = 9 (e, siRNA) based on three different Scd1 siRNA in three independent experiments. *P < 0.05 or P values given vs. vehicle control (a–c, f–h, k, l) or control siRNA (d, e); two-tailed paired (a–d, f–h, k, l) or two-sided unpaired student t test (e).
Fig. 5
Fig. 5. MUFA-containing PI species indicate stress in mice with Scd1 defect.
Liver, white abdominal fat, hind leg skeletal muscle, and skin were obtained from wildtype mice (wt) and mice homozygous for the Scd1ab-2J allele (Scd1ab-2J). a Protein expression of BiP and p38 MAPK, p38 MAPK phosphorylation, and the ratio of LC3BII/LC3BI protein levels in liver. b Hepatic proportion of MUFAs in PI and PC; MUFAs: 16:1, 18:1 (LTR P = 0.00000008, 0.00000003). c Heatmap showing the changes in the PI profile of liver, fat, skeletal muscle, and skin from Scd1ab-2J mice as compared to wt mice. Data are given as percentage of the relative PI abundance in wt tissues. d Hepatic proportion of PI(18:1/18:1) and PI(16:1/18:1) (LTR P = 0.00001, 0.00003). e Volcano plots highlighting PI and PC species that are regulated in murine tissues from Scd1ab-2J mice relative to wt mice. Comparisons show the mean difference of percentage changes and the negative log10(adjusted P value). Adjusted P values given vs. vehicle control; two-tailed multiple unpaired student t tests with correction for multiple comparisons using a two-stage linear step-up procedure by Benjamini, Krieger, and Yekutieli (false discovery rate 5%). Phospholipids containing two MUFAs (“MUFA/MUFA”) or one MUFA in combination with SFA or PUFA (“MUFA”) are indicated by color. Mean (c, e) or mean + s.e.m. and single data (a, b, d) from n = 6 (wt) and n = 5 (Scd1ab-2J; a for BiP, wt) mice. P values as indicated; two-tailed unpaired student t test.
Fig. 6
Fig. 6. SCD1-derived PI(18:1/18:1) suppresses stress signaling.
Fibroblasts were treated with vehicle, CAY10566 (CAY, 3 µM), ETO (10 µM), Scd1 siRNA, and/or defined phospholipid vesicles (50 µM) for 48 h (a–g) or 43 h after transfection with siRNA for 5 h (h). 16:0 (400 µM) was added 6 h before harvesting (b–d). a Phosphorylation of p38 MAPK (LTR P = 0.99996, 0.99997, 0.999996, 0.0000004). b Protein expression of BiP (LTR P = 0.0000002, 0.9999995, 0.000001). Western Blots are representative of three (a, right panel: PI(16:0/16:0) ± CAY, PI(18:0/20:4) ± CAY), four (a, left panel), five (b), six (a, right panel: PC(16:0/16:0) ± CAY) or eight (a, right panel: w/o, CAY, PI(18:1/18:1) ± CAY) independent experiments. c Subcellular distribution of BiP (red) and DAPI (blue) overlaid with phase contrast images; scale bar, 20 µm. Fluorescence images are representative of three independent experiments. d PARP cleavage; cl., cleaved. Western Blots are representative of three independent experiments. e Annexin V and propidium iodide (PI) staining. Left panel: viable, early apoptotic, late apoptotic, and necrotic cells were defined as described for Fig. 4i. Data for w/o, CAY and ETO are identical to Fig. 4i. Cytograms are shown in Supplementary Fig. 11b. Right panel: ratio of late to early apoptotic events. f Fibroblast morphology; scale bar, 100 µm. Phase contrast images are representative of three independent experiments. g Cell numbers (LTR P = 0.000004, 0.000097, 0.00009). h Scd1 was silenced in fibroblasts using three different siRNA sequences, and p38 MAPK phosphorylation was determined after 48 h. Non-targeting siRNA was transfected as control (ctrl). Western blots are representative of three independent experiments. Data for w/o (ctrl and siRNA) are identical to Fig. 4e. Detailed descriptions of datasets shown are given in Supplementary Note 7. P values given vs. vehicle control (a, b, d, e, g) or control siRNA (h) or as indicated; repeated measures two-way ANOVA (a, left panel, d), two-way mixed-effects model (REML) (a, right panel), repeated measures one-way ANOVA (e, right panel, g, left panel), or one-way mixed-effects model (REML) (g, right panel) of log data or ordinary two-way ANOVA (b, h) + Tukey HSD post hoc tests.
Fig. 7
Fig. 7. PI(18:1/18:1) pulses shape the cellular stress response.
a Cellular content of PI(18:1/18:1) directly after addition of the fusogenic phospholipid mixture with 3 nmol PI(18:1/18:1) to fibroblasts. b Time-dependent decrease of excess PI(18:1/18:1). c–i Fibroblasts were treated with vehicle or CAY10566 (CAY, 3 µM) for 48 h, and PI(18:1/18:1) (3 nmol) was incorporated into cells using fusogenic liposomes 8 h before harvesting. c Time-scale. d Phosphorylation of p38 MAPK. e Ratio of LC3BII/LC3BI protein levels. f PARP cleavage; cl., cleaved. g Fibroblast morphology; scale bar, 100 µm. Phase contrast images are representative of three independent experiments. h Protein expression of BiP. i Cell numbers. Western blots are representative of three (d, e) or four (f, h) independent experiments. Blots shown in panel d and f and e and h derive from the same membrane, respectively. Mean + s.e.m. and single data from n = 1 (b), n = 3 (a, d, e, i), n = 4 (f, h) independent experiments. Two-tailed paired student t test (a, i) of log data (d–f, h).
Fig. 8
Fig. 8. Mechanistic insights into stress regulation by PI(18:1/18:1) from quantitative proteomics.
Fibroblasts were treated with vehicle, VAL (10 µM), MC (10 µM), CAY10566 (CAY, 3 µM), CAY (3 µM) plus PI(18:1/18:1) (50 µM), or CAY (3 µM) plus PI(16:0/16:0) (50 µM) for 48 h. a Heatmap showing relative changes in protein levels. Focus is placed on proteins that (i) participate in p38 MAPK signaling, the UPR, autophagy, or programmed cell death, (ii) are up- or downregulated by VAL, MC, and CAY10566 in the same direction (≥ 20%), and iii) for which co-treatment with PI(18:1/18:1) diminishes the effect (≥10%) more pronounced than PI(16:0/16:0) (≥ 10% difference). Data of independent experiments (n = 3) were calculated as percentage of vehicle control. b Volcano plots highlighting proteins that are regulated by VAL, MC, CAY or by PI(18:1/18:1) supplementation in CAY-treated cells compared to vehicle control. Comparisons of the indicated treatment groups show the difference of mean absolute intensities of log10 data and the negative log10(adjusted P value). c Radar charts indicating the percentage changes of cellular Ppp2ca, Hspa5, and Sqstm1 in fibroblasts treated with VAL, MC, CAY, CAY + PI(16:0/16:0), or CAY + PI(18:1/18:1) relative to vehicle control. Single data (a) or mean (b, c) from n = 3 independent experiments. Adjusted P values given vs. vehicle control; two-tailed multiple unpaired student t tests from log data with correction for multiple comparisons using a two-stage linear step-up procedure by Benjamini, Krieger, and Yekutieli (false discovery rate 5%) (b, c).
Fig. 9
Fig. 9. PI(18:1/18:1) in systems with varying stress tolerance.
a, c–e, g-j Cellular proportion of PI(18:1/18:1). a IgM- and IgM+ B-cells and B-cell lymphoma from Eµ-Myc-transgenic mice. b Expression, phosphorylation, and substrate turnover of the autophagy-regulatory proteins Ulk1, S6-kinase, and 4E-BP1. Western blots are representative of two independent experiments; * non-specific bands. Each line corresponds to an individual animal. c Wild-type (‘wt’) and sorafenib-resistant (‘resistant’) Huh-7 hepatocarcinoma cells. d Resident peritoneal macrophages from young (6-8 months) and old mice (26 months). e Hematopoietic stem cells (KSLs, upper panel) and myeloid progenitor cells (MPs, lower panel) from young and aged mice. f Cellular proportion of PI(16:1/18:1) in brain from young (6-8 month) and old male and female mice (26 month). g, h Influence of starvation on planarians with impaired ER stress adaption. g Lipids were extracted from either 48 h (left panels) or 4 days (right panels) regenerating planarians. gfp dsRNA (gfp(RNAi)) was injected as control. h Relative expression in TPMs of differentially regulated genes related to p38 MAPK activation and stress signaling based on the transcriptome analysis of starved, gfp or cct3A RNAi-treated planarians at 72 h of regeneration. Significance was determined by q-value (false discovery rate (FDR)) < 0.1 for pairwise comparisons. i, j Mice were inoculated with P. gingivalis and housed for 16 weeks on normal, 16:0-enriched, or 18:1-enriched diet or first for eight weeks on 16:0-enriched and then eight weeks on 18:1-enriched diet. Mean + s.e.m. and single data from n = 2 (e, lower panel), n = 3 (e, upper panel, g, h), n = 4 (a for control, f for old female), n = 5 (a for IgM+ tumor, c, d, f, i, normal diet for non-infected mice and 16:0/18:1-treated non-infected mice, j, normal diet for non-infected mice), n = 6 (a for IgM- tumor, i, j). P values as indicated; two-tailed unpaired student t test (c, d, e, g) of log data (a, f, i, j).
Fig. 10
Fig. 10. Impact of the lipokine PI(18:1/18:1) on stress (-adaptive) signaling.
SCD1-derived PI(18:1/18:1) (i) interferes with the UPR and autophagy, (ii) prevents the activation of p38 MAPK via MKK3/6, (iii) maintains cell morphology and supports cell proliferation, (vi) diminishes the induction of programmed cell death, and (vii) its content is regulated in various stress-related processes such as tumorigenesis, chemoresistance, aging, starvation, and infection.
See this image and copyright information in PMC

References

    1. Costa-Mattioli M, Walter P. The integrated stress response: From mechanism to disease. Science. 2020;368:eaat5314. doi: 10.1126/science.aat5314. - DOI - PMC - PubMed
    1. Hayes JD, Dinkova-Kostova AT, Tew KD. Oxidative stress in cancer. Cancer Cell. 2020;38:167–197. doi: 10.1016/j.ccell.2020.06.001. - DOI - PMC - PubMed
    1. Koeberle A, Löser K, Thürmer M. Stearoyl-CoA desaturase-1 and adaptive stress signaling. Biochim Biophys. Acta. 2016;1861:1719–1726. doi: 10.1016/j.bbalip.2016.08.009. - DOI - PubMed
    1. Stockwell BR, et al. Ferroptosis: a regulated cell death nexus linking metabolism, redox biology, and disease. Cell. 2017;171:273–285. doi: 10.1016/j.cell.2017.09.021. - DOI - PMC - PubMed
    1. Johnson GL, Lapadat R. Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. Science. 2002;298:1911–1912. doi: 10.1126/science.1072682. - DOI - PubMed

Publication types