The Warburg Effect Mediator Pyruvate Kinase M2 Expression and Regulation in the Retina

Abstract

The tumor form of pyruvate kinase M2 (PKM2) undergoes tyrosine phosphorylation and gives rise to the Warburg effect. The Warburg effect defines a pro-oncogenic metabolism switch such that cancer cells take up more glucose than normal tissue and favor incomplete oxidation of glucose, even in the presence of oxygen. Retinal photoreceptors are highly metabolic and their energy consumption is equivalent to that of a multiplying tumor cell. In the present study, we found that PKM2 is the predominant isoform in both rod- and cone-dominant retina, and that it undergoes a light-dependent tyrosine phosphorylation. We also discovered that PKM2 phosphorylation is signaled through photobleaching of rhodopsin. Our findings suggest that phosphoinositide 3-kinase activation promotes PKM2 phosphorylation. Light and tyrosine phosphorylation appear to regulate PKM2 to provide a metabolic advantage to photoreceptor cells, thereby promoting cell survival.

Figure 1. Immunofluorescence analysis of PKM2 and PKM1 in mouse retinas.

Prefer-fixed sections of dark- (A,B,F,G) and light-adapted (C,D,H,I) mouse retinas were stained for PKM2 (A–D), PKM1 (F–I), arrestin (B,D,G,I) and DAPI (B,D,E,G,I,J). Panels (B,D,G,I) represent the merged images of either PKM2 or PKM1 and arrestin, whereas panels (E,J) represent the omission of primary antibody. ROS, rod outer segments; RIS, rod inner segments; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bar 50 μm.

Figure 2. Immunofluorescence analysis of PKM2 and PKM1 in cone-dominant retinas.

Prefer-fixed sections of dark- (A,D) and light-adapted (B,E) Nrl^−/− mouse retinas were subjected to immunofluorescence with anti-PKM2 (A,B) and anti-PKM1 (D,E). Panels (C,F) represent the omission of primary antibody. POS, photoreceptor outer segments; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bar 50 μm.

Figure 3. Light-dependent PKM2 phosphorylation in rod- and cone-dominant retinas.

Prefer-fixed sections of dark- (A,B) and light-adapted (C,D) mouse retinas were subjected to immunofluorescence with anti-p-PKM2 (Y105) (A,C), and anti-arrestin (B,D) antibodies. Panels (B,D) represent the merged images of p-PKM2 and arrestin, whereas panel (E) represents the omission of primary antibody. ROS, rod outer segments; RIS, rod inner segments; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Retinal lysates from dark- and light-adapted mice were subjected to immunoblot analysis with anti-pPKM2 (F) and anti-PKM2 (G) antibodies. Densitometric analysis of pPKM2 was performed in the linear range of detection, and absolute values were then normalized to PKM2 (H). Data are mean + SEM, n = 4. *p < 0.05. Pyruvate kinase activity was measured from dark- and light-adapted mouse retinas with an LDH-coupled enzyme assay. Data are mean ± SD, n = 3, *p < 0.05. Prefer-fixed sections of dark- (J) and light-adapted (K) Nrl^−/− mouse retinas were subjected to immunofluorescence with anti-pPKM2 (J,K). Panel (L) represents the omission of primary antibody. POS, photoreceptor outer segments; OPL, outer plexiform; IPL, inner plexiform layer. Scale bar 50 μm. Full-length blots are presented in the Supplementary Information.

Figure 4. Rhodopsin activation regulates the phosphorylation of PKM2.

Prefer-fixed sections of dark- (A,B,F,G) and light-adapted (C,D,H,I) Rpe65^−/− mouse retinas were subjected to immunofluorescence with anti-pPKM2 (A–D), anti-PKM2 (F–I), and arrestin (B,D,G,I) antibodies. Panels (E,J) represent the omission of primary antibody. ROS, rod outer segments; RIS, rod inner segments; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bar 50 μm.

Figure 5. Biochemical characterization of PKM2 and PKM1 isoforms on isolated photoreceptor cells from dark- and light-adapted Balb/c mice.

Retinal homogenates from dark- and light-adapted Balb/c mice were subjected to OptiPrep™ (8–40%) density gradient centrifugation (A). Fractions of inner segments and intact photoreceptors (B) were collected from the top to the bottom of the gradients. A ten-microliter sample (one-microliter for rhodopsin) was subjected to immunoblot analysis (C) with opsin, α-3 Na/K ATPase, pPKM2, PKM2, and PKM1 antibodies. Full-length blots are presented in the Supplementary Information.

Figure 6. Biochemical characterization of PKM2 and PKM1 isoforms on isolated cone photoreceptor cells from Nrl^−/− mice.

Retinal homogenates from Nrl^−/− mice were subjected to OptiPrep™ (8–40%) density gradient centrifugation, and fractions of inner segments and intact photoreceptors (A) were collected from the top to the bottom of the gradients. A ten-microliter sample (three-microliter for M-opsin) was subjected to immunoblot analysis (B) with M-opsin, α-3 Na/K ATPase, pPKM2, PKM2, and PKM1 antibodies. Full-length blots are presented in the Supplementary Information.

Figure 7. Biochemical characterization of PKM2 and PKM1 isoforms on isolated photoreceptor cells from dark- and light-adapted Rpe65^−/− mice.

Retinal homogenates from dark- and light-adapted Rpe65^−/− mice were subjected to OptiPrep (8–40%) density gradient centrifugation, and fractions were collected from top to the bottom of the gradients. A ten-microliter sample (one-microliter for opsin) was subjected to immunoblot analysis with opsin, pPKM2, PKM2, and PKM1 antibodies. Full-length blots are presented in the Supplementary Information.

Figure 8. Light-induced PKM2 phosphorylation is PI3K-dependent.

Ex vivo mouse retinal explants prepared in the dark were incubated in DMSO or PI3K-inhibitor LY294002 for 10 min prior to exposure to room light for 30 min. Retinal proteins were immunoblotted with anti-pAkt, anti-Akt, anti-p-PKM2, and anti-PKM2 (A) antibodies. Densitometric analysis of pAkt/Akt and pPKM2/PKM2 (B). Data are mean + SEM, n = 4. *p < 0.05. Nrl^−/− (NRL) and Nrl^−/−/cone-p85α^−/− (NRL-p85) mouse retinas were used to measure NADPH levels (C). Data are mean + SEM, n = 4. *p < 0.02. Equal amounts of retinal mRNA from three independent one-month-old Nrl^−/− and Nrl^−/−/cones-p85α^−/− mice were used for real-time (RT)-PCR and normalized by β-actin levels (D). The mRNA levels were averaged for 6PGD, G6PD, Glut1, HIF-1α, HK-II, ME1, and PKM2. The data are mean ± SD, n = 3. *P < 0.05. Decreased phosphorylation of PKM2 in cones lacking the p85α subunit of PI3K. Prefer-fixed sections of dark- (E) and light-adapted (F) Nrl^−/−/cone-p85α^−/− mouse retinas were subjected to immunofluorescence with anti-pPKM2. Panel G represents the omission of primary antibody. POS, photoreceptor outer segments; OPL, outer plexiform layer; IPL, inner plexiform layer. Scale bar 50 μm. Full-length blots are presented in the Supplementary Information.