Sigmastat version 3

All statistical tests were done with SigmaStat version 3.5 (SPSS Inc., Chicago, IL, USA). After initial tests for data normality (Kolmogorov-Smirnov-test) and Equal Variance, parametric or non-parametric statistical methods were executed. For comparison of multiple ages or culturing conditions, we used one-way analysis of variance (ANOVA), followed by the Holm-Sidak method for pairwise multiple comparisons, or Kruskal-Wallis one-way ANOVA on ranks (KW-ANOVA), followed by the Dunn’s method for pairwise multiple comparisons. To test for differences between network types, independent of age or density variations, we also used Two-way ANOVA, for factors of age or density and network type. Differences between experimental sets were also tested with t-test or Mann-Whitney rank sum test (MW-RST). The proportions of BrdU labeled IN and PN neurons in different conditions were compared with the chi-square-test. A P-value of ≤ 0.05 was considered statistically significant. Asterisks in all graphs show the level of statistical significance (*P ≤ 0.05, **P ≤ 0.01, ^***P ≤ 0.001).

Baseline hemodynamic variables and body weights were analyzed using Student's t-test (SigmaStat Version 3.5, SPSS Inc., Chicago, IL). Glutamate induced changes in MAP, LSNA, and SSNA were analyzed using a two-way analysis of variance (ANOVA). When ANOVA indicated a significant interaction, differences between individual means were assessed by post-hoc Holm-Sidak test (SigmaStat). The between subjects factor of sedentary vs. active groups was grouped with n = 7 animals in each; whereas, the within subjects factors of location of injections was grouped within each animal. The injections were compared rostrocaudally at three different levels (L1, L2, and L3). At each level, the injections were collapsed at rostral, central and caudal regions during the analysis. In five of seven sedentary animals and in four of seven physically active animals we were able to complete 27 injections (9 injections at each level). Due to the presence of blood vessels on the dorsal surface of the medulla we performed 18 and 21 injections in the two remaining animals in the sedentary group and 15, 15, and 21 injections in the three remaining animals in the physically active group. For all analyses, a probability of p < 0.05 was considered statistically significant. Data are expressed as mean ± SEM.