*2.3. Breast Milk Collection and Measurements*

Breast milk was collected on day 12th and day 14th of lactation (the period of maximal milk yield) and was pooled, allowing the mother and pups to recover during day 13. On the day of collection, the pups were separated from the mother for 3 h; the dam was anesthetized with isoflurane inhalation anesthesia (1000 mg/g; Isoflutek, Karizoo Lab., Barcelona, Spain) and injected with 0.1–0.15 mL of oxytocin i.p. depending on rat body weight (10 UI/mL; Syntocinon, Mylan Products Ltd., Hertfordshire, UK). A 1.0–1.5 mL volume of breast milk was obtained by gentle massage of the mammary glands and collection by a glass micropipette in Eppendorf tubes. The dam was milked for a maximum of 1 h. Each breastmilk sample was divided into two aliquots, one to assess total fat and the second one to assess total proteins.

**Figure 1.** Experimental protocol diagram. MUN, maternal undernutrition during gestation; C-on-C, control pups fostered by a control mother (n = 6 dams); M-on-M, MUN rats fostered by their own mothers (n = 6 dams); C-on-M control rats fostered by a MUN mother (n = 4 dams); MUN-on-C, MUN rats fostered by a control mother (n = 4 dams).

**Total fat content.** Fat was analyzed by the Mojonnier method [23] with modifications. Briefly, a 0.5 mL volume of breast milk was mixed with 0.5 mL of ammonium hydroxide, followed by the addition of 50 μL of ethanolic phenolphthalein solution (0.5% *w*/*v*) and shaken. Thereafter, 2.5 mL of ethanol, 2.5 mL of petroleum ether, and 2.5 mL of ethyl ether were added and vigorously mixed for 30 s. Thereafter, the mix was centrifuged at 4000× *g* (3 min, RT) and the upper phase containing the fat was stored. This process was repeated three times with the remaining aqueous phase, adding the ethanol, ethyl ether, and petroleum ether. Thereafter, the fat from the 3 centrifugation was pooled, and to evaporate the ether solvent, it was placed overnight in a gravity convection oven at 50 ◦C, uncapped. Finally, the total fat content in the sample was measured by gravimetry and expressed as mg/mL of BM. A blank reaction was performed by substituting the 0.5 mL breast milk volume with H2O-Q.

**Total protein level.** Protein quantification was carried out in the defatted phase of breast milk using the Bradford method [24]. It has to be noted that centrifugated breast milk samples continue to contain proteins such as caseins. Briefly, 10 μL of BM (1:50 *v*/*v* in H2O-Q) were mixed in a microplate with a 200 μL volume of Coomassie-blue dye (1:4 *v*/*v* in H2O-Q; Bio-Rad Lab., Hercules, CA, USA). After shaking the mix for 1 min, the absorbance was measured at 595 nm in a microplate reader (Synergy HT Multimode; BioTek Instruments, Winooski, VT, USA). Total protein levels were expressed as mg of BSA/mL. BSA (Sigma-Aldrich, St. Louis, MO, USA) was used for the standard curve (range 0.1–0.5 mg/mL) and Coomassie-blue dye was substituted by H2O-Q for the blank curve.

#### *2.4. Transthoracic Echocardiography*

Transthoracic echocardiography was performed in 21 days-old rats using the VIVID q-system (GE Healthcare, Munich, Germany) equipped with a 13 MHz probe (12S-RS, GE Healthcare, Munich, Germany) as previously described [25]. Briefly, the rats were anesthetized by i.p injection with 80 mg/kg Ketamine hydrochloride (AuroMedics Pharma LLC., Dublin, Ireland) and 10 mg/kg Diazepam (Hospira, Inc., Lake Forest, IL, USA). The images were acquired with the animals in left lateral decubitus. M-mode imaging of the parasternal short-axis (papillary level) view allowed measurement of end-diastolic (mm) and end-systolic (mm) internal diameter, posterior wall thickness (mm), and interventricular septum thickness (mm) at diastole. Additionally, it was calculated the systolic

functionality as the ejection fraction (%). The pulsed-wave Doppler early-to-late transmitral peak diastolic flow velocity ratio (E/A ratio; arbitrary units) was measured to assess diastolic function (E, mitral peak early-filling velocity and A, mitral peak flow velocity at atrial contraction). The transmitral flow velocity profile was determined by positioning a sample volume at the tip of the mitral valve on the para-apical long-axis view. The E-wave deceleration time was measured as the time interval between peak E-wave velocity and the point where the descending E-wave (or its extrapolation) intercepted the zero line. Values were determined by averaging the measurements of three consecutive cardiac cycles.

#### *2.5. Aorta Structure by Confocal Microscopy*

Aorta structure was assessed in a ring cut from the PFA fixed thoracic aorta proximal to the aortic arch. The ring was mounted intact on a slide provided with a small well, to avoid compression containing Citifluor AF2 mounting medium (Aname, Madrid, Spain), as previously described [19]. The ring was visualized with a ×20 objective with a laser scanning confocal microscope (Leica TCS SP2, Leica Microsystems, Barcelona, Spain) at Excitation = 488 nm/Emission = 500–560 nm, the wavelength at which elastin can be detected by its autofluorescence allowing to detect the medial layer. Single images were captured with a ×20 air objective, ensuring the elastic lamellae were clearly visible. Quantification was performed with FIJI software [26], measuring internal and external perimeters (μm); from these measurements medial cross-sectional areas (CSAs) were calculated in μm2, assuming the sections were circular.

#### *2.6. Adipocyte Size and Browning*

PFA-fixed adipocytes from the perivascular tissue of the mesenteric bed were incubated with Mitotracker red FM (Invitrogen, ThermoFisher, Madrid, Spain) prepared in phosphate buffer saline (PBS) to stain mitochondria-positive cells [27], evaluating the percentage of beige adipocytes, since they are rich in mitochondria. Adipocytes were incubated for 60 min (1:4000 *v*/*v*, RT, darkness) and washed 3 times for 10 min in PBS. Thereafter, they were incubated for 15 min in a solution of DAPI (Invitrogen, ThermoFisher, Madrid, Spain) to stain nuclei (1:500 *v*/*v* from stock solution, RT, darkness,) followed by 3 washes in PBS for 10 min each at RT. The samples were mounted intact on a slide with a small well and were visualized with a Laser Scanning Confocal Microscope (Leica TCS SP2, Leica Microsystems, Barcelona, Spain). Five regions were randomly chosen based on DAPI wavelength and single images were captured with a ×40 objective at three wavelengths: Excitation = 405 nm/Emission = 410–475 nm to visualize nuclei, Excitation = 488 nm/Emission = 500–560 nm to visualize adipocyte size by autofluorescence [20], and Excitation = 581 nm/Emission = 644 nm to visualize mitochondria-positive cells.

The quantification of adipocyte size and proportion of mitochondria-positive cells was performed by FIJI software. The total number and area occupied by the adipocytes were quantified in the Excitation = 488 nm/Emission = 500–560 nm wavelength images (autofluorescence), and the average adipocyte size was calculated in each image. To assess Mitotracker-positive cells in perivascular WAT, the area occupied by adipocytes was measured in autofluorescent images, Mitotracker-positive areas were measured in the same region in the red wavelength images, and the percentage was calculated.

## *2.7. Statistical Analysis*

Statistical analysis was performed with R software (version 3.6.0, 2018, R Core Team, Vienna; Austria) within the R Studio interface using *rio, dplyr, compareGroups, ggpubr, devtools,* and *ggplot2* packages. Data was expressed as the median and interquartile range [Q1; Q3]. The differences in macronutrients of breast milk between control and MUN were performed by Mann-Whitney's U test. The Kruskal–Wallis test by ranks was used to test the independent experimental groups. The significant Kruskal–Wallis test was used with Dunn's test with directed pairwise comparison using C-on-C as a reference group. Second, it was considered the comparison between M-on-M versus M-on-C to determine, if

pups with prior fetal programming, the lactation could reverse the modification. A *p*-value (*p*) < 0.05 was considered significant.

#### **3. Results**

#### *3.1. Proteins and Fats in Breast Milk*

No differences between control and MUN were detected in breast milk proteins (Control = 125.0 [113.0; 129.0] mg eq. BSA/mL, n = 12 rats; MUN = 117.0 [111.0; 123.0] mg eq. BSA/mL, n = 10 rats; *p*-Value=0.539). Fat levels in MUN milk tended to be lower but did not reach statistical significance (Control = 98.5 [89.3; 113.0] mg/mL, n = 12 rats; MUN = 84.0 [66.7; 107.0] mg/mL, n = 10 rats; *p*-Value = 0.090).

## *3.2. Body Growth Gain*

From day 1 to 14 of lactation (exclusive lactation period), male rats from the M-on-M group exhibited a significantly higher increase in body weight gain compared to the C-on-C group. MUN males fostered by a C mother (M-on-C group) did not show differences compared C-on-C group, being significantly smaller compared to M-on-M males. Males from the C-on-M group had a significantly lower body weight gain compared to C-on-C (Figure 2A). By contrast, during the first lactation period, M-on-M female rats did not show differences in body weight gain compared to C-on-C or compared to M-on-C. As observed in males, control females nourished by a MUN mother (C-on-M) showed a lower body weight gain compared to C-on-C females (Figure 2A).

**Figure 2.** Body weight from day 1 to 14 of lactation (**A**) and from day 15 to 21 of lactation (**B**), and body length gain from day 1 to 14 of lactation (**C**) and from day 15 to 21 of lactation (**D**) in male and female rats from MUN and Control mothers. MUN, maternal undernutrition during gestation; C-on-C, control pups fostered by their mothers (34 males and 38 females); M-on-M, MUN rats fostered by their mothers (32 males and 40 females); C-on-M, control rats fostered by MUN mothers (21 males and 27 females); M-on-C, MUN rats fostered by a control mother (25 males and 23 females). Data show the median and interquartile range [Q1; Q3] and the showed *p*-Value (P) was extracted from Dunnett's post-hoc pairwise comparison test when the Kruskal-Wallis test was P < 0.05. Dots show outliers.

From day 15 to 21, when the pups suckle from mothers and eat by themselves, male M-on-M did not show statistical differences in body weight gain compared to C-on-C, or M-on-C males. However, M-on-M females accelerated growth during this period, as shown by the higher weight gain compared to C-on-C female rats. M-on-C rats also had higher weight gain compared to C-on-C females (Figure 2B).

From day 1 to 14, males of M-on-M length gain tended to be smaller compared to the C-on-C group but did not reach statistical significance (*p*-Value = 0.083). Cross-fostered MUN males (M-on-C) had a significantly higher length gain compared to the C-on-C group. In female rats, no differences were detected in any of the experimental groups in length gain during the exclusive lactation period (Figure 2C).

From day 15 to 21 length gain was significantly lower in M-on-M males and females compared to C-on-C control sex-matched counterparts. M-on-C males maintained a lower length gain, while M-on-C females improved the length rate, not significantly different from C-on-C females (Figure 2D).

## *3.3. Perivascular Adipocyte Size and Type*

WAT adipocyte size was assessed in perivascular fat from the mesenteric bed obtained from 21-day-old offspring. M-on-M males had significantly larger adipocytes compared to C-on-C rats (Figure 3A). Adipocytes from MUN males fostered by a control mother (Mon-C) were of similar size compared to C-on-C males and significantly smaller compared to M-on-M. Control males fostered by a MUN mother (C-on-M) also showed smaller adipocytes size compared to C-on-C males (Figure 3A).

**Figure 3.** Average size of perivascular adipocytes (**A**) and representative images obtained by confocal microscopy (**B**) of 21-day-old male and female offspring from MUN and Control mothers. MUN, maternal undernutrition during gestation; C-on-C, control pups fostered by their mothers (6 males and 6 females); M-on-M, MUN rats fostered by their mothers (6 males and 6 females); C-on-M, control rats fostered by MUN mothers (6 males and 6 females); M-on-C, MUN rats fostered by a control mother (6 males and 6 females). Data show the median and interquartile range [Q1; Q3], and the showed *p*-Value (P) was extracted from Dunnett's post-hoc pairwise comparison test when the Kruskal-Wallis test was P < 0.05. Images were acquired with a Leica TCS SP2 confocal microscope, ×40 objective at Ex = 488 nm/Em = 500–560 nm wavelength; scale bar = 50 μm. Dots shown outliers.

Regarding females, M-on-M rats had similar adipocytes compared to C-on-C. M-on-C females showed significantly smaller adipocyte size compared to C-on-C females and also compared to M-on-M. No differences were detected between control rats fostered by a MUN mother (C-on-M) and C-on-C females (Figure 3A).

We used Mitotracker, a dye that stains mitochondria, to evaluate beige-type adipocytes within perivascular WAT. Mitotracker-positive cells showed an intense stain. M-on-M males had a smaller percentage of Mitotracker-positive cells compared to C-on-C controls (Figure 4). MUN males fostered by a control dam (M-on-C) had a similar % compared to C-on-C males, being larger compared to M-on-M offspring. Control rats fostered by a MUN mother had lower Mitotracker-positive stained areas compared to C-on-C males.

As observed in males, M-on-M females had significantly lower Mitotracker positive area compared to C-on-C females. MUN rats fostered by a control mother (M-on-C) had a similar % compared to C-on-C offspring, being significantly larger compared to M-on-M rats (Figure 4).

**Figure 4.** Percentage of Mitotracker-positive adipocytes in perivascular fat (**A**) and representative images obtained by confocal microscopy (**B**) of 21-day-old male and female offspring from MUN and Control mothers. MUN, maternal undernutrition during gestation; C-on-C, control pups fostered by their mothers (6 males and 6 females); M-on-M, MUN rats fostered by their mothers (6 males and 6 females); C-on-M, control rats fostered by MUN mothers (6 males and 6 females); M-on-C, MUN rats fostered by a control mother (6 males and 6 females). Data show the median and interquartile range [Q1; Q3], and the showed *p*-Value (P) was extracted from Dunnett's post-hoc pairwise comparison test when the Kruskal-Wallis test was P < 0.05. Images were acquired with a Leica TCS SP2 confocal microscope, ×40 objective at Ex = 488 nm/Em = 500–560 nm (green images) and Ex = 581 nm/Em = 644 nm (red images); scale bar = 50 μm. Dots show outliers.

#### *3.4. Heart Structure and Function*

Heart structure and function were assessed by transthoracic echocardiography at the end of the lactation period. In male M-on-M rats, posterior wall thickness tended to be higher and interventricular septum was significantly higher compared to C-on-C males. MUN males fostered by a control mother (M-on-C) did not show differences with C-on-C males. Female pups did not show significant differences in any of the studied parameters (Figure 5a,b).

**Figure 5.** Posterior wall thickness (**A**), septum thickness (**B**), ejection fraction (**C**), and E/A ratio (**D**) of 21-day-old male and female offspring from MUN and Control mothers. MUN, maternal undernutrition during gestation; C-on-C, control pups fostered by their mothers (6 males and 6 females); M-on-M, MUN rats fostered by their mothers (6 males and 6 females); C-on-M, control rats fostered by MUN mothers (6 males and 6 females); M-on-C, MUN rats fostered by a control mother (6 males and 6 females); LV, left ventricle. Data show the median and interquartile range [Q1; Q3], and the showed *p*-Value (P) was extracted from Dunnett's post-hoc pairwise comparison test when the Kruskal-Wallis test was P < 0.05. Dots show outliers.

Systolic function, assessed as ejection fraction, was not different between groups (Figure 5C). However, diastolic function, assessed as the ratio of E/A waves, was significantly reduced in M-on-M males compared to C-on-C counterparts. No significant differences were found in MUN males fostered by a C mother (M-on-C) compared to C-on-C (Figure 5D). Female M-on-M did not show significant differences in any of these parameters compared to C-on-C. However, the E/A ratio was significantly decreased in female M-on-C compared to female M-on-M (Figure 5D).

End-systolic and end-diastolic diameters were not significantly different between C-on-C and any of the other groups (Table 1).

#### *3.5. Thoracic Aorta Structure*

In M-on-M males' internal and external diameter, and wall cross-sectional area (CSA) were significantly larger compared to C-on-C. In M-on-C males the studied parameters were not different from C-on-C males and external diameter and CSA were significantly smaller compared to M-on-M rats (Figure 6).


**Table 1.** End left ventricle dimensions of heart cycles of the offspring at day 21.

Data show median and interquartile range [Q1; Q3]. The *p*-Value was extracted from the Kruskal-Wallis test. Control (C); Maternal undernutrition (MUN).

**Figure 6.** Internal (**A**) and external diameters (**B**), and wall cross-sectional (**C**) of the thoracic aorta in 21-day-old male and female offspring from MUN and Control mothers. MUN, maternal undernutrition during gestation; C-on-C, control pups fostered by their mothers (6 males and 6 females); M-on-M, MUN rats fostered by their mothers (6 males and 6 females); C-on-M, control rats fostered by MUN mothers (6 males and 6 females); M-on-C, MUN rats fostered by a control mother (6 males and 6 females). Data show the median and interquartile range [Q1; Q3], and the showed *p*-Value (*p*) was extracted from Dunnett's post-hoc pairwise comparison test when the Kruskal-Wallis test was *p* < 0.05. Dot show outliers.

M-on-M females tended to have larger parameters compared to C-on-C, without statistical significance. However, MUN females fostered by a control mother (M-on-C) had a smaller external diameter and CSA than M-on-M rats (Figure 6).
