Figure 1.
Position dependence of maternal blood to placentas along the uterine horn.
(A) Experimental scheme for multi-modal functional imaging of pregnant mice: pregnant female ICR mice (E17.5) were analyzed using MRI, intravital fluorescence microscopy, and ex vivo fluorescence analysis of the maternal blood volume in the placenta (PBVm). Note fetuses (F) and their placentas (white arrow heads). (B) Data for a pregnant mouse carrying 5 fetuses in one uterine horn. Position dependence of maternal bi-directional perfusion was detected by MRI (|BD-ASL|). (C) Ex vivo fluorescence and corresponding PBVm values for the placentas in one uterine horn, in the same pregnant mouse as in A. (D) Correlation between PBVm and |BD-ASL| (n = 11 dams, 86 placentas/fetuses; r = 0.62, P<0.0001). (E) Correlation between PBVm and fetal body weight (n = 11 dams, 86 placentas/fetuses; r = 0.64, P<0.0001).
Figure 2.
Assessment of the arterial blood supply to the placenta via BD-ASL MRI and intravital fluorescence microscopy.
Two methods were used to explore the pattern of transfer of arterial blood to the placentas along the uterine horns in pregnant mice at late gestation (E17.5): 1) Bi-directional ASL methodology (Panels A–C); and 2) Intravital fluorescence microscopy imaging of the uterine arterial blood supply subsequent to intravenous administration of FITC-dextran to mice having undergone surgical arterial ligations of either the uterine branch of the ovarian artery, or the uterine artery (Panels D–F). (A–C) Placental saturation transfer maps obtained by BD-ASL MRI of an ICR pregnant mouse (E17.5). For placentas positioned closer to the cervix (Panel A: L1, R1), mainly negative BD-ASL contrast voxels (blue) were observed, consistent with the predominant contribution of maternal blood flow through the uterine artery. In placentas closer to the ovary (Panel C: L5–7), the BD-ASL contrast was mainly of positive voxels (red), implying that placentas in this part of the uterine horn are supplied through blood mainly from the uterine branch of the ovarian artery. Placentas located in the central region of the uterine horn (Panel B, L3–4) had a dispersive pattern of BD-ASL values with both negative and positive voxels, consistent with a dual supply from both the uterine artery and the uterine branch of the ovarian artery, respectively. (D) Intravital fluorescence microscopy image of the arterial blood supply to an intact uterine horn (a snapshot from Movie S1). (E) Intravital fluorescence microscopy image of the arterial blood supply to a uterine horn following ligation of the uterine artery (a snapshot from Movie S2). (F) Intravital fluorescence microscopy image of the arterial blood supply to a uterine horn following ligation of the uterine branch of the ovarian artery (a snapshot from Movie S3).
Figure 3.
Contribution of arterial blood supply to the placenta.
Arterial ligations were performed on ICR pregnant mice (E17.5) for either the uterine branch of the ovarian artery (n = 10 mice, 5 mice on each side; 82 placentas/fetuses); or in the uterine artery (n = 10 mice, 5 mice on each side; 92 placentas/fetuses). (A, B) Relative fluorescence signal in placentas along two uterine horns upon ligation of the left (A) or the right (B) uterine artery (each graph presents data from one animal. orange: ligated uterine horn; green: non-ligated uterine horn). (C, D) Relative fluorescence signal in placentas along two uterine horns upon ligation of the left (C) or right (D) ovarian artery (each graph presents data from one animal). (E, F) Histological sections of the placentas along two uterine horns upon ligation of the right uterine artery (E; same animal as in panel B) or right ovarian artery (F; same animal as in panel (D) Left) H&E section. Right) fluorescence microscopy (right; Green = FITC-dextran; Blue = DAPI; inset fluorescent image of the ex vivo placenta).
Figure 4.
Arterial delivery of maternal blood to the placentas is dependent on litter size and on the position of the placenta along the uterine horn.
(A) Mean fluorescence signal in placentas of a uterine horn after ligation of the left uterine a. (artery), right uterine a., left ovarian a., or right ovarian a. (n = 5 animal/group). For each animal, values were normalized (%) to the mean fluorescence signal of the placentas in the contralateral, non-ligated horn (mean ± SEM; a,b significant differences: P<0.05). (B) Mean fluorescence signal in placentas of a uterine horn after unilateral ligation of the uterine or ovarian arteries (n = 10 animal/group, left and right sides combined). For each animal, values were normalized (%) to the mean fluorescence signal of the placentas in the contralateral non-ligated horn (mean ± SEM; a,b significant differences: P<0.05). (C) Overall relative contributions of the uterine and ovarian arteries to the mean fluorescence signal in placentas in a uterine horn. Data are presented as Box-and-whisker plots, with the median, the 25% and 75% percentile ranges (box depth), and the maximum and minimum (T-bars). Significant differences (a, b; P<0.05). (D) Relative contributions of the uterine and ovarian arteries to the fluorescence signal in placentas located in different positions along the horn. Data were extracted only from symmetric pregnancies that contained 4 fetuses in each horn (n = 5 dams). (E) Relative contributions of the uterine and ovarian arteries to the fluorescence signal in placentas located in different positions along the horn. Data were extracted only from symmetric pregnancies that contained 6 fetuses in each horn (n = 4 dams).
Figure 5.
Intra-uterine neighbor effect: reduced |BD-ASL| and PBVm in placenta positioned adjacent to a pathological/dead fetus.
(A) A representative example from one animal; placental |BD-ASL| (upper panel), and placental fluorescence signal with the corresponding PBVm values (lower panel) of placentas along a uterine horn that contains one pathological fetus (#1, marked in red). The upper image presents the pathological fetus and its placenta adjacent to a normal fetus and placenta (fetus #3). Note the reductions in |BD-ASL| and PBVm values in the placenta of the pathological fetus, as well as in the adjacent placenta (#2). (B) Mean |BD-ASL| values in placentas of pathological/dead fetuses, fetuses positioned adjacent to pathological/dead fetus, and normal fetuses located far from a pathological/dead fetus. Note that mean |BD-ASL| value is significantly lower for those fetuses that have a pathological/dead neighbor. Different letters above bars indicate significant differences (mean ± SEM; a,b, P<0.05) (C) Fluorescence values and corresponding PBVm values in placentas of pathological/dead fetuses, fetuses located near a pathological/dead fetus, and normal fetuses located distant from a pathological/dead fetus. Note that mean fluorescence value and PBVm values are significantly lower for those fetuses that have a pathological/dead neighbor (mean ± SEM; a,b, P<0.05).
Figure 6.
Placenta of Akt1+/+ positioned adjacent to an Akt1−/− placenta/fetus showed increased |BD-ASL| and PBVm.
(A) |BD-ASL| values in placentas of fetuses of different Akt1 genotypes (mean ± SEM; a, b, P<0.05). (B) Fluorescence values and PBVm values in placentas of fetuses of different Akt1 genotypes (mean ± SEM; a, b, P<0.05). (C) |BD-ASL| values in Akt1+/+ placentas located near an Akt1−/− fetus, as compared to Akt1+/+ placentas not located near an Akt1−/− fetus (mean ± SEM; P = 0.0712). (D) Fluorescence values and PBVm values in Akt1+/+ placentas located near a Akt1−/− fetus, as compared to Akt1+/+ placentas not located near an Akt1−/− fetus (mean ± SEM; a, b, P<0.05).
Figure 7.
Electrical circuit modeling of the hemodynamics of maternal arterial supply in the mouse pregnancy.
(A) Diagram of mouse uterine horns and their arterial blood vessels in the gestation period after placentas have formed (E10.5 to term). Our results demonstrate that each placenta can be perfused from two maternal arteries. F Fetus, K Kidney, Ov Ovary, Pl Placenta, Ut Uterine. (B) Schematic diagram of mouse placenta. Dec Decidua, Sp Spongiothrophoblast, TGC Throphoblast Giant Cells, Lab Labyrinth. (C) Numerical simulation of blood flow in multi-fetus pregnancy modeled as an electrical circuit. Bi-directional blood flow in each of two uterine horns was modeled by the respective currents: Iua(l/r) for the left or right uterine arteries and Ioa(l/r) for the left or right uterine branch of the ovarian artery. The balance between Iua(l/r) and Ioa(l/r) was set at 3∶1 using resistors placed into the circuit near the battery. Resistance to flow along the uterine branch of the ovarian artery and the uterine artery was modeled by a series of identical resistors (one per implantation site). (D) Placenta modeled as an electrical circuit. Resistance to flow into the placenta via the spiral arteries was modeled by low-value resistors, which were connected separately to a second resistor to ground, simulating exchange within the placenta itself and the flow back to ground (i.e., clearance of blood through the maternal veins). Diffusion across the placenta was modeled by the insertion of a large resistor (20× the value of the dual resistors to ground representing flow into the placenta), between the dual arterial input resistors. Total maternal blood supply to each placenta Ip,i was therefore derived from Ip,oa,i+Ip,ua,i. Arrows indicate the direction of flow.
Figure 8.
Numerical simulation of placental function in the pregnant mouse reveals a possible hemodynamic basis for the observed ‘position dependence’ and ‘neighbor effects’.
(A) Simulation performed for the case of five fetuses in a single uterine horn and a hypothetical case, in which the resistance along the uterine and ovarian arteries was very low. Note the flattening of the U-shaped pattern in the latter case. (B) Simulation of the effect of increasing litter size with a fixed battery voltage VH (no cardiac output compensation). (C) Simulating a case of nine ‘fetuses/placentas’ with adjustment of VH (i.e., adjustment of cardiac output) to maintain the same average Ip as in the case of five fetuses per uterine horn. (D) Ligation simulations of the uterine artery showing Ip of placentas in the non-ligated and ligated horns, normalized to the average Ip of the non-ligated horn. (E) Ligation simulations of the uterine branch of the ovarian artery showing Ip of placentas in the non-ligated and ligated horns, normalized to the average Ip of the non-ligated horn. (F) Simulation of the placenta of a fetus that spontaneously died during mid pregnancy. The current flowing in normal placentas located closest to the dead fetus is compared to placentas located distant from the dead fetus. The dead fetus was simulated by reduced placental resistance compared to normal placentas. (G) Simulation of Akt1−/− placenta; the average Ip,Akt1+/+ of Akt1+/+ having a Akt1−/− neighbor compared with Akt1+/+ placentas having only Akt1+/+ neighbors. The Akt1−/− placenta was simulated as increased resistance compared to Akt1+/+ placentas.