Fig 3. Menin deficiency results in disrupted metabolism in the hypothalamus.

(A) PCA of the metabolomics data revealing an obvious separation between the hypothalamus of 13-month-old ScKO mice and control mice. (B) For inverse modeling of biochemical regulation from metabolomics covariance data, a metabolic reconstruction and pathway reduction from metabonomics is performed (RECON). The metabolite variance of 2 groups is visible, which is further exploited for the calculation of the Covariance matrix (COV) and subsequently for the Jacobian matrix (JAC) using the stochastic Lyapunov matrix Eq 1 (see the METARECON strategy in Materials and methods). (C) Differential Jacobian matrix of the hypothalamus of 13-month-old ScKO mice and control mice derived from covariance data from the metabolomics datasets. All entries represent median values of 10³ calculations normalized to the square of interquartile distance. dF and dM characterize the entries of the Jacobian matrix and refer to Eq 2 (see the METARECON strategy in Materials and methods). The greatest perturbation was identified as the Jacobian entry ðf_NADH = ðf_KGA, pointing to D-serine synthesis pathway in the underlying biochemical network in panel D. (D) Simplified biochemical interaction network of D-serine synthesis pathway adjusted to the measured metabolites. (E) D-serine levels in serum from young (22-26Y) and old individuals (83-94Y) were determined by ELISA; n = 5. Data represent mean ± SEM, n.s.: not significant, *p < 0.05, **p < 0.01, ***p < 0.001. Statistical applications were analyzed by one-way ANOVA with Tukey’s post hoc analysis. The underlying data of Fig 3 can be found in S3 Information. PCA, principal component analysis.