*2.1. Maternal Undernutrition (MUN) Model and Cross-Fostering Protocol*

We used a rat model of fetal programming induced by maternal global nutrient restriction during the second part of gestation, as previously described [21]. After observation of sperm in the vaginal smear (gestation = day 0) the dam was allocated to either MUN or the control group. Control dams were fed *ad libitum* throughout pregnancy and lactation; MUN dams were fed *ad libitum* from day 1–10 and with 50% of the averaged control daily

intake (previously calculated as 24 g/day) from day 11 to the end of gestation, returning to *ad libitum* diet during lactation. 24 h after birth the pups were sexed and weighed individually and the litter was standardized to 12 individuals, 6 males, and 6 females if possible (smaller litters were not used for this work, being used for other studies). Control and MUN litters were left with their mothers or cross-fostered.

**Cross-fostering protocol**. We used a protocol similar to [22] with some modifications. Two dams were mated at the same time, ensuring the same gestation day 0, and those with birth on the same day were cross-fostered. On postnatal day 1, the entire litter was exchanged to the foster's dam cage with bedding from the original mother to avoid rejection. The litter was standardized to 12 individuals and the pups were sexed. The rats were not individually marked to help acceptance (no rejection was detected in any of the groups).

The name of the groups was "pup type–on–mother type" (i.e., M-on-C refers to a pup from a MUN dam fostered by a control mother). Four experimental groups were studied: control-on-control (C-on-C;n=6 mothers), MUN-on-MUN (M-on-M; n = 6 mothers); control-on-MUN (C-on-M, n = 4 mothers), MUN-on-control (M-on-C;n=4 mothers).
