Natural product P57 induces hypothermia through targeting pyridoxal kinase

Abstract

Induction of hypothermia during hibernation/torpor enables certain mammals to survive under extreme environmental conditions. However, pharmacological induction of hypothermia in most mammals remains a huge challenge. Here we show that a natural product P57 promptly induces hypothermia and decreases energy expenditure in mice. Mechanistically, P57 inhibits the kinase activity of pyridoxal kinase (PDXK), a key metabolic enzyme of vitamin B6 catalyzing phosphorylation of pyridoxal (PL), resulting in the accumulation of PL in hypothalamus to cause hypothermia. The hypothermia induced by P57 is significantly blunted in the mice with knockout of PDXK in the preoptic area (POA) of hypothalamus. We further found that P57 and PL have consistent effects on gene expression regulation in hypothalamus, and they may activate medial preoptic area (MPA) neurons in POA to induce hypothermia. Taken together, our findings demonstrate that P57 has a potential application in therapeutic hypothermia through regulation of vitamin B6 metabolism and PDXK serves as a previously unknown target of P57 in thermoregulation. In addition, P57 may serve as a chemical probe for exploring the neuron circuitry related to hypothermia state in mice.

Fig. 1. P57 induces a reversible hypothermia in mice and has a neuroprotective effect in mouse MCAO model.

a Chemical structure of P57. b Core temperature of P57-treated mice. P57 or the vehicle control was injected intraperitoneally into mice at 0 min (arrow). n = 5 mice, significant differences between treatments were calculated using two-way analysis of variance (ANOVA). c Oxygen consumption level of P57-treated mice. Volume of oxygen (VO₂) was measured every 24 min, n = 3 mice, two-way ANOVA. d Carbon dioxide output level of P57-treated mice. Volume of Carbon dioxide (VCO₂) was measured every 24 minutes, n = 3 mice, two-way ANOVA. e Respiratory exchange ratio of P57-treated mice. Respiratory exchange ratio is the ratio of VCO₂ to VO₂, n = 3 mice, two-way ANOVA. f Food intake of P57-treated mice in metabolic chambers. Food intake over 72 hours was measured. n = 3 mice, student’s t test. g Core temperature of mice treated with P57 four times. P57 or the vehicle control was injected intraperitoneally into mice at 0, 3, 6, and 9 h (arrow). n = 3 mice, two-way ANOVA. h–j P57 treated-mice after middle cerebral artery occlusion (MCAO). P57 (25.0 mg/kg) or the vehicle was injected intraperitoneally into mouse 2 h after MCAO. Mice were sacrificed at 24 h after drug injection and brain were quickly removed for TTC staining. Core temperature of mice in MCAO + P57+ Warming group maintained by the heating pad. n = 2 mice for sham group, n = 5 mice for others. h Statistic infarct volumes of brain slices from mouse model of MCAO. student’s t test. i Rectal temperature of mice in MCAO model. student’s t test. j Representative images of TTC staining of brain slices in MCAO model treated with P57. Experiments in Fig.1 were performed under ambient temperatures at 22–24 °C. Student’s t test used was two-sided. All error bars are presented as mean values  ±  s.e.m. *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001. Source data are provided as a Source Data file.

Fig. 2. P57 binds with PDXK and inhibits its enzyme activity.

a Chemical structure of biotinylated P57 probe. b, c Affinity chromatography experiment with mouse brain lysate b or purified protein PDXK c using P57-Bio (2.0 μM) probe in the absence (middle lane) and presence (right lane) of P57 (20.0 μM). b Western blot was performed to measure the binding amount of PDXK with P57-Bio. c Silver staining was performed to measure the binding amount of PDXK with P57-Bio. d Dose–response curve for the binding interaction between P57 and PDXK using MST. The concentration of purified protein PDXK is kept constant, while the concentration of P57 varies from 0 to 100 µM. The binding curve yields a dissociation constant K_d = 15.8 ± 2.9 µM. e A surface plasmon resonance (SPR) assay characterizing the binding between P57 and purified protein PDXK. Color lines, model fits of SPR data from different concentrations of P57. The calculated dissociation constant K_d = 3.5 µM. f Kinetic analysis of PDXK inhibition by P57 with respect to the substrate pyridoxal (PL) shows that P57 acts as a competitive inhibitor. g A surface rendering of predicted binding mode of P57 (stick rendering) at the active site of human PDXK (cyan). Interactions between P57 and surrounding residues of PDXK, critical residues are rendered in stick representation (C, cyan; N, blue; O, red; H, gray), hydrogen-bond interactions are shown as black dashed lines. h 2D-dimensional interaction schemes of predicted binding pose of P57 in ATP/PL active site generated by Maestro. Magenta lines with arrow denote hydrogen bonds. Source data are provided as a Source Data file.

Fig. 3. P57 mainly targets PDXK in hypothalamus to induces hypothermia.

a LC-MS detection of PL in hypothalamus (left) and brain tissues (right). Hypothalamus and brain tissue were sampled at 0 min, 30 min and 90 min after treated with P57 (25.0 mg/kg), n = 12 mice, student’s t test. b Schematic showing the bilateral injection of drug into the POA. Coronal brain sections showing location of cannula implanted in POA. Scale bars, 2 mm. c Rectal temperature of mice treated with P57 in the POA. 15 mg/mL or 30 mg/mL P57 (0.6 mg/kg or 1.2 mg/kg,) was injected into POA bilaterally at 0 min (arrow) (400 nL for each laterally). n = 3 mice, two-way ANOVA. d Schematic showing the bilateral injection of virus into the POA. Coronal brain sections showing EGFP (upper right) or Cre-EGFP (lower right) expressed in POA. Scale bars, 500 μm. e, f Core temperature of P57-treated mice with conditional knockout PDXK or not in POA. Virus were injected into hypothalamus bilaterally and let the mice recover for one month, then P57 (25 mg/kg) was injected intraperitoneally, n = 3 mice, two-way ANOVA. g Core temperature of mice with conditional knockout PDXK or not in POA. Mice were treated with cold exposure (10 °C) for 24 h, n = 3 mice. h Core temperature of PL-treated mice. PL was injected intraperitoneally, n = 6 mice, two-way ANOVA. i Core temperature of mice treated with PL in the POA. 100 mg/mL or 200 mg/mL PL (5.0 mg/kg or 10.0 mg/kg) was injected into POA bilaterally (500 nL for each laterally), n = 4 mice, student’s t test. j Core temperature of mice treated with combination of PL (200.0 mg/kg) and P57 (12.5 mg/kg). PL or P57 was injected intraperitoneally into C57BL/6 J male mice, n = 5 mice, two-way ANOVA. Student’s t test used was two-sided. All error bars are presented as mean values  ±  s.e.m. * P < 0.05, ** P < 0.01, *** P < 0.001 and ns, not significant. Source data are provided as a Source Data file.

Fig. 4. P57 and PL play a similar effect on hypothalamus neurons.

a Bar plot showing the number of DEGs affected by P57 and PL across each cluster in hypothalamic neuron subtypes. b Scatter plot shows the correlation of the number of DEGs affected by P57 and PL. Each point represents a cluster, the value means the number of DEGs in each cluster. The linear best fit line is shown, with 95% confidence intervals (shaded areas) and the Pearson correlation coefficient (R) and p-value (P) were calculated (two-sided test). c Heatmap of shared DEGs in hypothalamus with P57 and PL treatment. d The relative ratio of unique and overlapped DEGs between P57 and PL pair in each cluster. The burlywood and skyblue represents the number of specific DEGs of PL and P57 respectively, and the red stands for the number of overlapped DEGs. The heatmap on the left ranks each cluster by the Jaccard Similarity in DEGs between P57/PL and control group.

Fig. 5. The schematic illustration of P57 regulating body temperature.

The natural product P57 crossed the blood-brain barrier and entered the brain, inhibiting PDXK enzyme activity, leading to the accumulation of PL, and eventually causing the mouse core body temperature to drop.