Allosteric inhibition of a stem cell RNA-binding protein by an intermediary metabolite
- University of Massachusetts Medical School, United States
Figures
Figure 1 with 2 supplements
MSI1 is inhibited by monounsaturated fatty acids.
(A) Pattern of MSI1 (blue) in the CNS. (B–D) EMSA and FP of MSI1 binding to RNA aptamer CCCR005 (AGCGUUAGUUAUUUAGUUCG). EMSA data (red line) were fit to the Hill equation where all shifted species …
Figure 1—figure supplement 1
(A) Coomassie-stained SDS page gel shows that recombinant MSI1 is purified to greater than 95% over a 3-column purification protocol. (B) MSI1 displays decreased affinity for an RNA aptamer upon …
Figure 1—figure supplement 2
RNA binding specificity and inhibition by specific fatty acids is conserved in MSI2.
(A) Sequence alignment shows 83% conservation between MSI1 and MSI2 within the RRM domains. Regions that correspond to α-helices, β-sheets, and intervening loops as defined by NMR spectroscopy are …
Figure 2 with 2 supplements
MSI1 inhibition is allosteric.
(A) MSI1 tryptophan fluorescence at 350 nm as a function of oleic and elaidic acid. The Kd, app, Ki, app, and Hill parameters are the average and standard deviation of three independent replicates. …
Figure 2—figure supplement 1
(A–D) The MSI1 tryptophan fluorescence spectrum changes as a function of inhibitor titration. The tryptophan fluorescence emission spectrum is strongly quenched upon oleic acid (A) and erucic acid (C…
Figure 2—figure supplement 2
(A and B) Recombinant His6-tagged MSI1 RRM1 binds RNA aptamer CCCR005 (A) and is inhibited by oleic acid (B). FP data are reported as the average and standard deviation of three independent …
Figure 3 with 1 supplement
(A and B) Model of RRM1 bound to oleic acid (yellow) calculated by Schrödinger GLIDE (Friesner et al., 2004; Friesner et al., 2006). (C) Overlay of the oleic acid bound MD simulation (gray and red) …
Figure 3—figure supplement 1
Mutational analysis supports molecular dynamics simulation and docked model predictions.
(A) MSI1 RRM1 overlays from two NMR models demonstrate the highly dynamic nature of loop 3. Arginine 53 (model 1) and arginine 61 (model 3) are predicted to interact with the caboxylate of oleic …
Figure 4
Molecular dynamics studies (A and B) represent the normalized histograms of the protein solvent accessible surface area (SASA) and radius of gyration (rgyr) illustrating the structural transition …
Figure 5 with 3 supplements
MSI-1 regulates stearoyl-CoA desaturase.
(A) HEK293T (dashed) and CG-4 (solid) cell proliferation as a function of oleic acid or stearic acid treatment (red = treated, black = untreated). The data are the average and standard deviation of …
-
Figure 5—source data 1
Lipidomics data files.
- https://doi.org/10.7554/eLife.02848.015
Figure 5—figure supplement 1
Dose response for oleic acid treatment in cell culture.
HEK293T (dashed) and CG-4 (solid) cell proliferation as a function of increasing oleic acid (red = 50 µM, grey = 5 µM, black = 0 µM). The data are the average and standard deviation of at least …
Figure 5—figure supplement 2
(A) MSI1 binds each of 7 SCD1 3′ UTR fragments containing the MSI1 consensus sequence (G/A)U1–3AGU with variable affinity. FP and F–EMSA binding experiments are reported as the average and standard …
Figure 5—figure supplement 3
Lipidomics analysis of undifferentiated and differentiated CG4 oligodendrocyte progenitor cells.
(A) Scatter plot of lipidomics data for differentiated vs undifferentiated CG4 cells. Data are reported as nMoles per million cells. Red data points indicate lipids that are significantly different …
Figure 6
Model of SCD regulation by MSI1, and subsequent downstream consequences of SCD activity changes.
https://doi.org/10.7554/eLife.02848.019Videos
Video 1
This video is a representative segment of an MD trajectory of MSI1 bound to oleic acid.
https://doi.org/10.7554/eLife.02848.013Tables
Table 1
Structure-activity relationship analysis demonstrates specificity of inhibition
| A | ||||
|---|---|---|---|---|
| Compound name | CID | Screen score | FP Ki, app (µM) | F–EMSA Ki, app (µM) |
| Chembridge 7409829 | 28425 | 0.045 | 15 ± 2.8 | 54 ± 22 |
| Aurintricarboxylic Acid (ATA) | 2259 | 0.053 | 0.23 ± 0.03 | 1.5 ± 0.14 |
| GW7647 | 3397731 | −0.028 | 6.5 ± 0.4 | 21 ± 0.8 |
| Oleic Acid | 445639 | −0.005 | 1.2 ± 0.4 | 1.4 ± 0.7 |
| B | ||||
|---|---|---|---|---|
| Compound name | Structure | Code | FP Ki, app (µM) | F–EMSA Ki, app (µM) |
| Oleic acid |  | 18:1 ω-9 | 1.2 ± 0.4 | 1.4 ± 0.7 |
| Eicosenoic acid |  | 20:1 ω-9 | 1.2 ± 0.4 | 1.7 ± 0.6 |
| Erucic acid |  | 22:1 ω-9 | 0.64 ± 0.2 | 0.82 ± 0.03 |
| Nervonic acid |  | 24:1 ω-9 | 47 ± 30 | 23 ± 8 |
| Palmitoleic acid |  | 16:1 ω-7 | 5.3 ± 0.5 | 13 ± 0.9 |
| Linoleic acid |  | 18:2 ω-6, 9 | 2.2 ± 0.2 | 1.2 ± 0.03 |
| Arachidonic acid |  | 20:4 ω-6, 9, 12, 15 | 3.0 ± 0.2 | 1.1 ± 0.3 |
| Oleoyl-CoA |  | (18:1 ω-9) | 8.1 ± 0.3 | 4.0 ± 0.2 |
| Erucyl-CoA |  | (18:1 ω-9) | 4.1 ± 0.9 | 0.62 ± 0.2 |
| Ricinoleic acid |  | (18:1 ω-9) | No inh. | 18 ± 9 |
| Oleamide |  | (18:1 ω-9) | No inh. | No inh. |
| Ethyl oleate |  | (18:1 ω-9) | No inh. | No inh. |
| 4-Methylumbelliferyl Oleate |  | (18:1 ω-9) | No inh. | No inh. |
| Elaidic acid |  | 18:1(trans) | No inh. | No inh. |
| Stearic acid |  | 18:0 | No inh. | No inh. |
| Palmitic acid |  | 16:0 | No inh. | No inh. |
| Myristic acid |  | 14:0 | No inh. | No inh. |
-
(A) Small molecule screen hits. Compound ID (CID) refers to each compound's LOPAC identification number. Screen scores were calculated by normalizing the polarization value of each compound to the no protein and no compound controls, as described in the supplemental methods. After the screen was complete, compounds that scored as hits were confirmed by FP and F–EMSA dose response experiments. Apparent inhibition constants (Ki, app) are the average and standard deviation of at least three independent experiments. (B) The code = carbon number:number of double bonds, followed by the position of the double bonds from the aliphatic end of the fatty acid. Where a fatty acid is modified, the parental fatty acid numerical code is given in parentheses for comparison purposes. FP and F-EMSA dose response results are reported as the average and standard deviation of at least three independent experiments.
