All experimental procedures included in this project were carried out in conformity with European Union Directive 2010/63/EU and were approved by Hospital Gregorio Marañón’s Research Committee and Ethical Committee for Animal Experimentation (ES280790000087).

C57BL6/J mice were obtained from Jackson Laboratory for this experiment. Transgenic mice with an EGFP gene in their ROSA26 locus blocked by LoxP-flanked STOP fragment (B6; 129-Gt(ROSA)26Sortm2Sho/J, #004077) were crossed with transgenic Sox2-Cre mice (B6.Cg-Edil3Tg(Sox2-cre)1Amc/J, #008454). This resulted in offspring with EGFP expression in all cells derived from the epiblast during embryogenesis, as well as any tissue with constitutive Sox2 expression (ROSA26-EGFP-Sox2).

Male ROSA26-EGFP-Sox2 mice were then crossed with WT mice to produce heterozygous offspring with similar Sox2-triggered EGFP expression in their organism (Figure 1).

For the functional study, male C57BL6/J mice were modified to express DTR in all their cells.

While a cross between DTR/DTR males and WT/WT females was attempted to guarantee WT/DTR genotype offspring (and thus full microchimerism clearance upon DT inoculation), breeding was unsuccessful, and heterozygotic males were used instead. Male heterozygotic WT/DTR mice were crossbred with female WT/WT mice, resulting in offspring with WT/WT and WT/DTR genotypes, which would likewise produce microchimerisms displaying both genotypes. Cells expressing DTR die after exposure to Diphtheria Toxin (DT), allowing WT/DTR genotype cells to be selectively eliminated through DT inoculation (200 ng, intraperitoneal injection).

This injection was carried out on female breeders 3 days after giving birth, in order to clear any WT/DTR microchimerisms present in their organism. Further administrations were performed at post-partum day 6 and day 9 in order to maximize FMc depletion, and the myocardial infarction model was carried out at post-partum day 10, 7 days after first DT injection, following previous reports of DTR + cell clearance on mouse heart (Cha et al., 2003; Akazawa et al., 2004).

The myocardial infarction experimental model consisted on the ligation of the Left Anterior Descending coronary artery (LAD-ligation), as previously described (Villalba-Orero et al., 2021). Briefly, a week after DT injection, mice were anesthetized, intubated, and mechanically ventilated for the procedure. A left thoracotomy was performed between the third and fourth ribs, and the left descending artery was permanently ligated after pericardiectomy to induce a myocardial infarction.

Transthoracic echocardiography was performed at three time points: before the LAD-ligation (baseline), and at days 3 and 28 after the LAD-ligation. This process was performed by an expert operator using a high-frequency ultrasound system (Vevo 2100, Visualsonics Inc.), in blinded conditions.

The parasternal long axis was analyzed at three levels (basal, mid, and apical) and all measurements were averaged over three consecutive cardiac cycles. LVEF and FS were calculated using the modified Quinone method, using the following formulas: LVEF = LVIDed 2 − LVIDes 2 / LVIDed 2

FS = (LVIDed−LVIDes)/LVIDes.

Where LVIDed is left ventricular internal diameter at end diastole and LVIDes is left ventricular internal diameter at end systole.

2D-guided M-mode of parasternal short-axis short (middle) was used to measure ventricle wall thickness.

For the infarct size measurements, hearts were cut into 5 µm-thick slices along the transverse axis from apex to base with a microtome. Sections were stained using Masson’s trichrome (MTC) to study fibrosis, and midline infarct length was measured as previously described (Takagawa et al., 2007). Three measurements from three different slices of the same sample were averaged to obtain the % length value for each sample, and these values were averaged across all samples in the same group (N = 5 each).

For flow cytometry assays, blood samples (250 µL per animal) were collected under anesthesia on gestation day 20 and 30 days post-partum (N = 1) and stored in MiniCollect tubes with EDTA to prevent coagulation. An erythrocyte lysis protocol was carried out using Red Blood Cell Lysis solution (RBCL, cat#130-094-183, Miltenyi Biotec). Briefly, 100 µL of blood were mixed with 900 µL of RBCL and incubated for 10 min at room temperature (RT). Samples were centrifuged at 300 g for 5 min, and the supernatant was removed, obtaining an erythrocyte-free pellet.

Samples were analyzed using a MACSQuant Analyzer 16 flow cytometer (Miltenyi Biotec) and images were processed with the Kaluza software. Blood samples were analyzed without the use of antibodies, with direct detection of the intrinsic fluorescence showcased by the transgenic mice model used in this study.

For the myocardial sample assays, hearts were extracted from the animals (N = 2) after sacrifice through cervical dislocation and cut into three pieces each. No perfusion was included, and the samples were rinsed thoroughly with sterile phosphate-buffered saline (PBS 1x) afterwards to remove traces of blood and subjected to mechanical digestion with a syringe over a 40 μm cell strainer. FCM buffer (2% Fetal Bovine Serum and 0.1% sodium azide in PBS 1x) was added throughout the process. Samples were centrifuged for 5 min at 4°C at 1500RPM and resuspended, then stored at 4°C until they were analyzed with the flow cytometer. Myocardial cell samples were analyzed without the use of antibodies.

The statistical significance of the differences between group means was assessed through unpaired Student’s t-test or one-way ANOVA, as appropriate. In cases where ANOVA indicated inter-region variability, a multiple comparison was performed using the Tukey test. Data are presented as mean ± SD, and a p < 0.05 was considered statistically significant.

To determine the presence of fetal microchimerisms, WT female mice were crossed with ROSA26-EGFP-Sox2 male mice, which express EGFP in all cells, including sperm. To verify the presence of fetal microchimerisms in maternal blood, blood samples were extracted from the WT female breeders at day 20 of pregnancy and 1 month post-partum, and the presence of EGFP + cells was analyzed using flow cytometry (Figure 2). Any fluorescence detected in blood would correspond to cells of fetal origin, as WT animals did not showcase any baseline fluorescence (Supplementary Figure S1). A slight increase in positive fluorescence cells (1.28%, compared to the baseline 0.08%) was detected at day 20, which would correspond to a moderate number of fetal cells having crossed into the maternal systemic circulation with the potential to form microchimerisms; this number was reduced at 30 days post-partum (0.41%), although the value still remained elevated compared to the baseline in control WT females. This observation would point to microchimerisms being present in the maternal circulation throughout pregnancy and remaining for at least 1 month after delivery, in accordance with previous studies which found even longer periods of permanence (Bianchi et al., 1996; Bayes-Genis et al., 2005; Kamper-Jørgensen et al., 2014).

To determine the presence of microchimerisms in the heart, cells from myocardial samples were analyzed by flow cytometry (Figure 3). WT and ROSA26-EGFP-Sox2 mice were used as negative and positive controls, respectively. As expected, the WT mouse showed no EGFP + cells in the heart, while most cells in ROSA26-EGFP-Sox2 mice were EGFP+ (Figures 3A, B). Hearts from WT female mice crossed with ROSA26-EGFP-Sox2 male mice showed 1.6%–1.7% EGFP + cells (Figures 3C, D), indicating that fetal microchimerisms migrated and engrafted to the heart. EGFP + cells were also observed in maternal heart tissue samples through confocal microscopy, while there was no such fluorescence in the base WT animals (Supplementary Figures S2, S3).

Since we had detected fetal microchimerisms in the heart, we next investigated whether these microchimerisms had any effect on cardiac function following myocardial infarction. For this purpose, WT female animals were crossed with male transgenic animals expressing diphtheria toxin receptor (DTR). Fetal microchimerisms present in maternal tissue were eliminated by treating the mothers with diphtheria toxin (DT), which induced the death of cells expressing the DTR (i.e., fetal microchimerisms).

DT was administered 3, 6, and 9 days post-partum to remove any cells expressing DTR. Myocardial infarction was induced through permanent ligation of the left descending coronary artery 10 days post-partum, and its presence was confirmed 3 days after surgery by echocardiography as a strong reduction in left ventricular ejection fraction (LVEF).

DT treatment resulted in improved cardiac contraction, as evidenced by increased LVEF and stroke volume (SV) values 28 days post-MI in microchimerism-depleted mice compared to untreated mice. This effect following DT administration was not observed in WT mothers that had been crossed with WT males, suggesting that it was specific and caused by the depletion of DTR-expressing fetal microchimerisms in the maternal cardiac tissue (Figures 4A–D; Table 1). DT-treated animals also showcased a smaller average infarct length compared to its control group after histological study (Figure 5).

While previous research from Kara et al. had pointed at fetal microchimerisms not only being present in maternal heart tissue but also specifically homing in on injured areas, there was no clear evidence of any effects on tissue repair, beneficial or detrimental. The capability of differentiation into cardiac cell lineages was demonstrated in vitro but not in vivo, where chimerisms just formed disorganized and undifferentiated structures and remained stuck. The result of this not only failed to support the hypothesis of fetal cells being able to aid in maternal injury repair and provide benefits, but it directly opposed such a notion. Indeed, the deletion of microchimerisms in female breeders undergoing myocardial infarction was associated with improved contractile capacity compared to mice in which microchimerisms were retained.

The reported lack of proper differentiation and structure in vivo may be the cause of this pernicious effect. Microchimerisms may be unable to mount a proper response to the damage and complete repairs to the tissue, while obstructing the mother’s own response. This situation would be more likely to occur in the natural environment, as opposed to previous reports where the specific positive selection for cells with the desired traits and the right environments may have ensured a higher efficiency and likelihood of an effective response. On the other hand, microchimerism-depleted females would rely exclusively on their own cells and regenerative processes, which would have a higher success rate and recover a higher degree of function after injury.

It should be noted that this depletion was not complete, as some fetal microchimerisms remained in the hearts because they did not express the DTR receptors; as such, one hypothesis would suggest that the process of cell death experienced by the DTR + microchimerism population could be triggering a proliferative and reparative response of the DTR-population in turn, which would prime them for a more efficient reparative action after MI, compared to the more extensive but inactive population of fetal microchimerisms present in non-depleted maternal hearts.

It is also worth pointing out that this study did not include any significant characterization of the cell populations involved in these dynamics; the overall size range of the cells detected through FC would potentially match a myocytic profile, as opposed to other cell types such as lymphocytes, but this doesn’t constitute definitive evidence of any specific cell type being favored by the fetal microchimerism population in the maternal heart. Likewise, the possibility of contamination of the heart samples from circulating cells present in traces of blood cannot be completely ruled out, with the potential skewing of the results regarding density and type of FMc present in cardiac tissue. Thus, it would be interesting to include that kind of effort in any future research on this topic, to better understand what is actually happening with the fetal microchimeric cells and the nature of whatever specific activity they are carrying out in a pathological context. For example, the immune system is involved in the response to MI, and as such it is possible that cells of fetal origin presenting an immune system phenotype could be taking part in that response and even resulting in such a negative effect as it may be inferred from the results of this study.