*2.10. Statistical Analysis*

Data were analyzed using GraphPad Prism software (GraphPad Software 5.01, San Diego, CA, USA) by one-way ANOVA followed by Bonferroni post hoc test. In all tests, n (sample size) represents biological replicates (donors). Results are presented as mean ± SD. Level of significance was set at 0.05 and is indicated as \**p* < 0.05, \*\**p* < 0.01, or \*\*\**p* < 0.001.

#### **3. Results and Discussion**

All tests were performed with reflected and placental amnion samples, as we and others already have shown significant differences between reflected and placental amnion [16–19].

#### *3.1. Cellular Viability Strongly Decreased in Floating hAM Samples*

In order to test hAM viability in vitro, biopsies kept under floating conditions for 21 days were tested for cell viability with the EZ4U assay. Indeed, cell viability significantly decreased in reflected and placental amnion compared to fresh biopsies at day 0 (Figure 1A), confirming that standard cell culture conditions do not sufficiently maintain cell viability. This is in line with previous studies, where loss of viability was already observed [9,10,24]. Interestingly, in these studies, cells in non-distended hAMs remained viable under osteogenic [9], chondrogenic [24] or Schwann cell-like cell differentiation conditions [10]. This could mean that under differentiation conditions, cells may receive signals that sustain cell viability.

To test our hypothesis (whether mechanical tissue distention could prolong cellular viability and preserve mitochondrial function in in vitro culture), we incubated hAM biopsies for 21 days under two different conditions. For distended samples, the hAM was mounted on CellCrown™ inserts in order to expose the hAM to tensile strength (Figure 1B). For non-distended biopsies, we kept the biopsies "floating" in culture medium (Figure 1C). In contrast to floating biopsies, biopsies mounted on CellCrown™ inserts for 21 days showed no loss of cellular viability (Figure 1A). It has to be taken into account that in floating samples, cells still have close cell–cell contact, however, no mechanical tension on the tissue is present.

#### *3.2. Mitochondrial Membrane Potential Drastically Decreased in Floating hAM Samples*

In order to see if mitochondria play a role in this strong decrease of cell viability, mitochondrial membrane potential was visualized after 21 days. Membrane potential was slightly decreased in distended reflected amnion (Figure 2B,G) compared to fresh hAM at day 0 (Figure 2A,G). Distended placental amnion (Figure 2E,G) at day 21 did not show any significant decrease compared to fresh biopsies at day 0 (Figure 2D,G). In contrast, mitochondrial membrane potential was drastically reduced in both reflected amnion (Figure 2C,G) and placental amnion (Figure 2F,G) under floating conditions after 21 days, compared to day 0 (Figure 2A,D,G) or distended biopsies (Figure 2B,E,G). Such changes in mitochondrial membrane potential can be connected to loss of cell viability [28].

**Figure 2.** Mitochondrial membrane potential of reflected (**A**,**B**,**C**) and placental (**D**,**E**,**F**) amnion. Mitochondrial membrane potential (red) was stained with tetramethylrhodamin-methylester (TMRM; 500 nM) at day 0 (**A**,**D**) and at day 21 in biopsies cultivated while mechanically stretched (DIS; **B**,**E**) or kept floating (FLO; **C**,**F**). Imaging was performed with an inverted confocal microscope (LSM510, Zeiss, excitation/emission 543 nm/585 nm). Image analysis was performed with Zeiss ZEN2009 software (**G**). Mean ± standard deviation (SD), n = 3 (donors); representative images of one donor. Scale bar: 100 μm. DIS: distended biopsies; FLO: floating biopsies. Level of significance is indicated as \* *p* < 0.05, \*\*\* *p* < 0.001

#### *3.3. Mitochondrial Respiration and ATP Concentration Strongly Decreased in Floating hAM Samples*

Since mitochondrial membrane potential does not necessarily reflect mitochondrial activity, we determined mitochondrial ROUTINE respiration, a measure for total mitochondrial oxygen consumption. In fresh biopsies (day 0), significantly higher ROUTINE respiration was detected in placental compared to reflected amnion (Figure 3A). This is in line with a previous publication of our group [18]. No difference was observed between day 0 and distended reflected amnion at day 21. Placental amnion biopsies showed an approximately 30% lower ROUTINE respiration at day 21 compared to day 0. This is insofar interesting, as it has been shown that regarding mitochondrial activity, placental amnion is also much more responsive to changes in the microenvironment compared to reflected amnion [29]. In that study, responsiveness of placental amnion to inhibition of ATP synthase was much more pronounced. Furthermore, human amniotic mesenchymal stromal cells of placental amnion were found to be more susceptible to changes in oxygen tension [30]. As expected, ROUTINE respiration of both regions (reflected and placental) dramatically decreased in floating biopsies after 21 days compared to day 0. ROUTINE respiration was also significantly lower in floating compared to distended biopsies in reflected amnion. This effect was again even more pronounced in placental amnion (Figure 3A). The reasons for the repeatedly observed differences between reflected and placental amnion are still not known. We believe that the different anatomical locations of the hAM (one covering the placenta, the other opposite it) influence its properties. It is likely that during pregnancy, the two different amniotic sub-regions have different biological functions. Of note, regarding mechanical forces, the rupture of membranes at term takes place in the zone of altered morphology, an area within the reflected amnion [31].

The drastic decrease of mitochondrial oxygen consumption in floating samples of both amniotic regions, together with the massive loss of mitochondrial membrane potential, indicate severe mitochondrial dysfunction. We assume that this dysfunction is the prerequisite for the loss of cellular viability displayed in Figure 1.

**Figure 3.** Measurement of mitochondrial activity. (**A**) Mitochondrial ROUTINE respiration (total mitochondrial oxygen consumption) was measured with high resolution respirometry (Oxygraph 2k, Oroboros Instruments) in reflected and placental amnion. Oxygen consumption was determined in fresh biopsies (day 0) and at day 21 in biopsies cultivated while mechanically stretched (DIS) or kept floating (FLO). Measurement of ATP levels. (**B**) ATP levels were determined using the ATP Bioluminescence Assay Kit CLS II (Roche) in fresh biopsies (day 0) and at day 21 in biopsies cultivated while mechanically stretched (DIS) or kept floating (FLO) in reflected and placental amnion. Mean ± SD, n = 4 (donors). P: placental amnion; DIS: distended biopsies; FLO: floating biopsies. Level of significance is indicated as \**p* < 0.05, \*\**p* < 0.01, \*\*\**p* < 0.001.

Next, we wanted to see whether the changes in mitochondrial respiration also have an effect on ATP levels (Figure 3B). The results showed a similar pattern to the results of ROUTINE respiration. ATP concentrations at day 0 were also higher in placental amnion compared to reflected amnion, which is in line with a previous publication [29]. Distended biopsies after 21 days showed lower ATP concentrations compared to day 0 in both regions. Biopsies kept floating for 21 days showed a strong decrease in ATP concentrations in both regions (reflected and placental) compared to day 0. Floating biopsies of placental amnion showed lower ATP concentrations compared to distended biopsies. These data again indicate mitochondrial dysfunction.

Taken together, floating biopsies seem to have lost most of their viability, whereas in distended biopsies, viability could be sustained to a large extent. The question arose whether the loss of cellular viability was due to apoptosis or necrosis. The very low levels of ATP contradict apoptosis, since induction of apoptosis requires ATP [32,33]. We, however, measured ATP concentrations at a time point at which most of the cells had already lost their viability, meaning that the process of cell death had already been executed. In order to investigate whether apoptosis was involved in the initiation of cell death at an earlier time point, we shortened the incubation period and incubated non-distended and distended amnion biopsies for only 14 days.

#### *3.4. Caspase 3 is Strongly Upregulated in Floating hAM Samples*

We first subjected paraffin-embedded hAM biopsies to immunohistochemistry staining with anti-cleaved caspase 3 antibodies. As expected, in fresh biopsies (day 0), no caspase 3 positive cells were detected in either amniotic region (Figure 4A,D,G). After 14 days, in distended biopsies, scattered expression of caspase 3 was found in reflected (Figure 4B,G) and placental samples (Figure 4E,G). In contrast, in floating biopsies, expression of caspase 3 drastically increased in both regions (Figure 4C,F,G), compared to fresh biopsies (Figure 4A,D,G), and distended biopsies (Figure 4B,E,G), indicating the occurrence of apoptotic cell death [34].

**Figure 4.** Caspase 3 immunohistochemical staining. Histological sections of reflected (**A**,**B**,**C**) and placental (**D**,**E**,**F**) amnion at day 0 (**A**,**D**) and at day 14 in biopsies cultivated while mechanically stretched (**B**,**E**) or kept floating (**C**,**F**) were stained for the expression of apoptosis marker caspase 3 (brown). Image analysis was performed with ImageJ software (**G**). Mean ± SD, n = 3 (donors); representative images of one donor. Scale bar: 100 μm. DIS: distended biopsies; FLO: floating biopsies. Level of significance is indicated as \*\*\**p* < 0.001.

#### *3.5. Severe Loss of Mitochondrial Internal Structure under Floating Conditions*

In order to see if mitochondria are involved in the process of cell death, we investigated mitochondrial morphology using transmission electron microscopy. In fresh biopsies (day 0) of the hAM, mitochondria displayed well developed cristae within a strongly contrasted matrix in reflected (RA, Figure 5A) and placental (P, Figure 5D) hAM. At day 14, in distended samples, mitochondria of the placental region retained their cristae, however, the mitochondrial matrix appeared less dense (Figure 5E). Most of the mitochondrial cristae in distended reflected amnion were lost (Figure 5B). Floating samples showed mitochondria with severe loss of cristae and overall integrity in both regions (Figure 5C,F). These internal structural changes can be an indication for the onset of apoptosis [35,36].

#### *3.6. Mitochondria-Linked Apoptotic Gene Expression was Upregulated within Seven Days*

In order to confirm that apoptosis was mitochondria-linked, we compared BAX and BCL-2 levels. Since we already observed strong upregulation in caspase 3 expression and changes in mitochondrial ultrastructure on day 14, we shortened the incubation time to seven days. Indeed, amnion samples showed higher gene expression of BAX and BCL-2 at day 7 in distended and floating biopsies in both regions compared to day 0 (Figure 6A,B). Moreover, floating placental amnion on day 7 showed significantly higher BAX expression compared to distended biopsies (Figure 6A). Calculating the BAX/BCL-2 ratio revealed that BAX levels in distended samples were higher compared to day 0, but this was only significant in reflected amnion (Figure 6C). As expected, the most pronounced effects were observed between day 0 and floating biopsies of day 7 for both regions (Figure 6C). Additionally, floating placental amnion had a higher BAX/BCL-2 ratio compared to distended biopsies. Although the increase in the BAX/BCL-2 ratio in distended samples also points to the onset of apoptosis, the low number of caspase 3 positive cells (Figure 4) showed that apoptosis was only partly initiated in distended samples. In floating samples, results of BAX/BCL-2 gene expression and caspase 3 immunohistochemistry indicate that the lack of tissue distention initiated mitochondria-mediated apoptosis. The results are in line with studies showing that the mitochondria-linked pathway via BAX is involved in the initiation of apoptosis in hAM in vivo [37,38], a crucial step that leads to the rupture of the membranes at term. Interestingly, it was also shown that chronic stretching of isolated human amniotic epithelial cells (hAECs) cultivated on flexible bottom cell culture plates increased the

expression of pre-B cell colony-enhancing factor. This protected isolated hAECs from apoptosis [39], suggesting that distention can prolong cellular life span in vitro.

**Figure 5.** Changes in mitochondrial morphology were analyzed with transmission electron microscopy (Tecnai 20, FEI Europe, Eindhoven, Netherlands) in reflected (RA) (**A**,**B**,**C**) and placental (P) (**D**,**E**,**F**) amnion at day 0 (**A**,**D**) and at day 14 in biopsies cultivated while mechanically stretched (**B**,**E**) or kept floating (**C**,**F**), n = 2 (donors). Scale bar: 200 nm. Arrows indicate mitochondria. DIS: distended biopsies; FLO: floating biopsies.

**Figure 6.** Gene expression of B-cell lymphoma 2-associated X protein (BAX) (**A**) and B-cell lymphoma (BCL)-2 (**B**) was determined by qPCR in fresh biopsies (day 0) and at day 7 in biopsies cultivated while mechanically stretched (DIS) or kept floating (FLO) in reflected and placental amnion. The results were normalized on the geometric mean of 4 different reference genes. The BAX/BCL-2 ratio is shown in (**C**). Mean ± SD, n = 3 (donors). DIS: distended biopsies; FLO: floating biopsies; Cq value: cycle of quantification value. Level of significance is indicated as \**p* < 0.05, \*\**p* < 0.01, \*\*\**p* < 0.001.

#### **4. Limitations of the Study**

A limitation of this study is that no specific set of inhibitors was used for this pathway.
