Fig 1: Hyperglycemia positively correlated with low serum APN levels and low APN levels positively correlated with delayed retinal vascularization in premature infants (gestational age at birth: 23–30 weeks, birth weight: 348–1,716 g) Mean weekly APN levels and 95% confidence interval for mean plasma glucose concentration in relation to APN level tertiles (T). At postnatal weeks 1 and 2, infants in the Tlow APN had significantly higher mean plasma glucose concentrations than infants in the Thigh. APN Tlow 3.86 (range: 1.79–5.97) μg ml−1 (red), APN Tintermediate 8.63 (range: 6.43–11.54) μg ml−1 (yellow), and APN THigh 17.50 (11.84–36.00) μg ml−1 (green). n = 50, **P < 0.01; ***P < 0.001. t‐test.Mean weekly APN levels and 95% confidence interval in infants with normal (blue) versus delayed (red) vascularization. n = 44, ***P < 0.001. t‐test.Schematics of retinal vascularization in premature versus term infants. Preterm (red) versus term (blue) infants had delayed retinal vascularization. Circles in red represent zones of retina in preterm infants.
Fig 2: Recombinant APN treatment restored retinal function in hyperglycemia‐associated retinopathy AElectroretinogram (ERG) assessment of retinal function in control and HAR mice showing photoreceptor response amplitude (RmP3), sensitivity (S), and post‐receptor response amplitudes (V m), sensitivity (σ), and retinal sensitivity (S m). ERG parameters are presented as the log change from control mice. Representative ERG waveforms (left). n = 10–12 per group. *P < 0.05, ***P < 0.001. ANOVA.BCorrelation of post‐receptor sensitivity deficits (Logσ) with the sum of deficits in photoreceptor sensitivity (LogSrod) and saturated amplitude (LogRrod). n = 10–12 per group. Pearson R test.CMeasurement of outer nuclear layer thickness in WT control and HAR retinas at P30. Representative images of H&E‐stained retinal cross sections (left). Scale bar, 50 μm. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; ONL, outer nuclear layer. n = 6 per group. Unpaired t‐test.D, EHyperglycemia was induced with 25 mg kg−1 STZ (i.p., daily from P2 to P12) in WT mice. The mouse pups received msAPN (0.6 μg g−1) or vehicle treatment daily from P7 to P39. Littermate controls were used. At P40, rod (D) and cone (E) ERG were conducted. Cone ERG was reflected by post‐receptor responses (V m). n = 7 to 12 per group. **P < 0.001. ANOVA (D) or t‐test (E).Data information: Data presented as mean ± SEM (A, C–E).
Fig 3: Activation of the APN pathway improved photoreceptor mitochondrial activity (related to Fig 4) Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) of WT control and HAR P10 retinas. n = 5–6 animals (24–32 punches) per group.APN deficiency did not affect retinal metabolic enzyme mRNA in deep retinal blood vessels. qRT–PCR of Hk1 and Cs in LCM deep blood vessels in Apn −/− versus WT HAR mice. n = 3 pooled retinas/group. Unpaired t‐test.Mitochondrial and cytoskeleton morphology in 661W cells. Cell was stained with MitoTracker for mitochondria (red), F‐actin for cytoskeleton (green), and DAPI for nucleus (blue). Images were taken under 630× magnification. Scale bar, 10 μm.Mitochondrial morphology (MitoTracker, red) under hyperglycemic condition, in the presence of AdipoRon or vehicle. Images were taken under 630× magnification. Scale bar, 10 μm. The number of cells within each category of mitochondrial morphology (tubular, intermediate, and fragmented) was counted and represented as a percentage of the total number of cells. n = 30 images/group. Kruskal–Wallis test.qRT–PCR of mitochondrial fusion (Mfn1, Mfn2) and fission (Fis1, Mff) proteins in AdipoRon‐ versus vehicle‐treated 661W cells. n = 7–9/group. Unpaired t‐test (Mfn1, Mfn2, Mff); Mann–Whitney test (Fis1).Extracellular acidification rate (ECAR) of WT retinas treated with msAPN or vehicle (PBS). n = 6 animals (12 punches) per group. ANOVA, *P < 0.05, **P < 0.01.ECAR of 661W cells treated with AdipoRon or vehicle. n = 10 per group. ANOVA, *P < 0.05, **P < 0.01 (0.5 μM AdipoRon versus vehicle); † P < 0.05, †† P < 0.01 (0.05 μM AdipoRon versus vehicle).Deep retinal vascular formation in WT HAR mice co‐treated with msAPN (0.6 μg g−1) and Compound C (AMPK inhibitor, 2 μg g−1) or vehicle from P7 to P9. n = 4–8 per group. Scale bar: 50 μm. Unpaired t‐test.Data information: Data presented as mean ± SEM (A, B, D–H).
Fig 4: Photoreceptor‐derived VEGFA was modulated by the APN pathway but not contributed to normal deep blood vessel growth (related to Fig 4) Vegfa mRNA expression in retinal neuronal layers (ONL, INL, GCL) of Apn −/− versus WT P10 HAR retinas (LCM and qRT–PCR). n = 3 pooled retinas/group. ANOVA. n.s., not significant. Data presented as mean ± SEM.Representative images of deep retinal vascular network and quantification in AAV2‐hRK‐Vegfa‐GFP versus AAV2‐hRK‐GFP (control) virus subretinally injected in WT mice. Retinas were examined at P10. Scale bar, 50 μm. n = 6–7 per group. Unpaired t‐test. Data presented as mean ± SEM.
Fig 5: APN pathway activation modulated photoreceptor mitochondrial function in hyperglycemic mice Quantitative proteomic analysis of key glucose metabolic enzymes and mitochondrial metabolic enzymes in hyperglycemic (HAR) WT and Apn −/− mice at P10. n = 6 mice/group. Blue: maximum of levels; orange: minimum of levels.Seahorse XFe96 Flux Analyzer oxygen consumption rate (OCR) of WT HAR P10 retinas treated with msAPN or vehicle (PBS). Basal: initial OCR without treatment; max, maximal OCR: after addition of FCCP (carbonyl cyanide‐4‐trifluoromethoxy phenylhydrazone). Treatment with rotenone and antimycin A (RAA) reveals non‐mitochondrial respiration. ANOVA, *P < 0.05. n = 7–8 per group.mRNA expression of metabolic enzyme (Hk1, Cs) in retinal neuronal layers (ONL, INL, GCL) of Apn −/− versus WT HAR P10 retinas (LCM and qRT–PCR). n = 3 pooled retinas/group. ANOVA. n.s., not significant.Western blot for p‐AMPK/AMPK in 661W cells treated with AdipoRon or vehicle. n = 4 per group. Unpaired t‐test.qRT–PCR of key metabolic enzymes (Hk1, Cs) in AdipoRon‐ versus vehicle‐treated 661W cells. n = 9 per group. Unpaired t‐test. n.s., not significant.OCR of 661W cells treated with AdipoRon or vehicle. Leak indicates OCR independent of ATP production with proton leak induced with oligomycin (oligo) treatment. n = 10 per group. ANOVA. *P < 0.05, **P < 0.01.qRT–PCR of Pdgfb in AdipoRon‐ or vehicle‐treated 661W cells with and without oligomycin. n = 12 replicates/group. ANOVA.qRT–PCR of Pdgfb in msAPN‐ or vehicle‐treated WT HAR retinas. n = 12 replicates/group. Unpaired t‐test. Pdgfb expression in photoreceptors (ONL from LCM) from Apn −/− versus WT HAR P10 mice (LCM and qRT–PCR). n = 4–5 replicates from three pooled retinas/group. Unpaired t‐test.qRT–PCR of Pdgfb in hRK‐shPdgfb‐transfected 661W cells. n = 9 per group. Unpaired t‐test.Representative images of deep retinal vascular network and quantification in AAV2‐hRK‐Pdgfb‐GFP versus AAV2‐hRK‐GFP (control) virus subretinally injected in WT mice. Retinas were examined at P10. Scale bar, 50 μm. n = 8–12 retinas/group. Unpaired t‐test.Data information: Data presented as mean ± SEM. See also Figs EV3 and EV4.
Supplier Page from Enzo Life Sciences, Inc. for Adiponectin (mouse), (recombinant)