Analysis of HIF-1α level and sarcomeric distribution after hypoxia and hypoxia/re-oxygenation

The study aimed to investigate the effect of hypoxia and hypoxia/re-oxygenation (H/R) on the function of cardiomyocytes cultured on 3D structures. The response of two types of human cardiac cells, a commercially available line of human cardiomyocytes - HCM and a line of human cardiomyocytes obtained by differentiating induced pluripotent stem cells - iPSC-CMs, was studied. The function of iPSC-CMs has not been fully investigated so far. However, thanks to their phenotype and spontaneous contraction, they can be a promising cell model in cardiovascular disease studies. Nanofibrous mats affect the cardiac cell morphology and stimulate growth and differentiation; therefore, they may also affect the response of cells to hypoxia. The study used polycaprolactone and polyurethane nanofibers with diameters of 509 ± 178 nm and 452 ± 151 nm, respectively. The nanofibers were oriented parallel to each other, and their Young’s modulus was 48.6 ± 3.6 MPa (for PCL) and 60.3 ± 8.9 MPa (for PU). Our previous research described the characterization of nanofiber’s morphology in detail [34, 37]. Nanofibrous mats were utilized as a substrate that can influence cardiac cell response under hypoxia, and the results were compared with the controls (cultures on polystyrene (PS)). We established three types of cultures on PS plates: the cells cultured under (a) normal oxygenation (21% O2) conditions as normoxia control, (b) under hypoxia for 6 h as hypoxia control, and under hypoxia for 6 h and re-oxygenation for 24 h as H/R control.

The results of human cardiomyocyte staining of F-actin and HIF-1α after hypoxia and re-oxygenation are shown in Fig. 1. A parallel arrangement cytoskeleton structure characterized HCM cells cultured on polystyrene plate (PS) under normoxia. The cells cultured on PS under hypoxia conditions for 6 h showed disrupted actin cytoskeleton structure and increased levels of HIF-1α factor (3.5-fold higher than in normoxia control). The cells cultured on PCL and PU nanofibrous mats subjected to 6 h of hypoxia also showed disrupted actin cytoskeleton structure and increased HIF-1α (for cultures grown on PCL scaffolds is 6.8 higher, for PU nanofibers is 4.3 than for normoxia). Despite disrupting the cytoskeleton structure, parallel orientation was evident in the cells cultured on PU nanofibrous mats. The cytoskeletal structure of the cells cultured on PCL nanofibers subjected to 6 h of hypoxia was more disturbed than observed for the other culture types. These results suggest that HCM cells, after hypoxia cultured on PCL, show greater sensitivity to hypoxic conditions than cultures grown on PU nanofibrous mats and PS.

Some literature reports indicate that the influence on damaged cardiac cells has delivered oxygen back to CMs after hypoxia [7,8,9]. Therefore, the influence of re-oxygenation on human cardiomyocyte function in 3D models was studied. After 6 h of hypoxia and 24 h of re-oxygenation (Fig. 1B), the cells had similar disruption of the cytoskeletal structure as the cells maintained 6 h under hypoxia. The most significant differences between each surface were observed for the HIF-1α. In control, H/R cells, HIF-1α gene expression decreased, while an increase was observed for HCMs cultured on PCL and PU nanofibrous mats. The protein level was 2.9-fold (for PCL) and 2.5-fold (for PU) higher than normoxia control. Hypoxia and re-oxygenation caused a decrease of HIF-1α for culture performed on PS plate (0.6-fold of 30 h normoxia control). This may indicate that nanofibrous mats can reduce the oxygen transfer rate into cells, affecting the high expression of HIF-1α. Slight disruption of F-actin filaments was observed in cells cultured on PU and PS nanofiber mats. In contrast, cells cultured on PCL nanofiber mats were characterized by significant disruption of the cytoskeletal structure, which may indicate a higher degree of cell damage. In Fig. 1C and D, the cells grown on nanofibrous mats under hypoxia and H/R show significantly higher HIFα expression than normoxia control. However, for H/R cultures, the expression is significantly lower than for cultures after hypoxia. Moreover, for the cells cultured on PCL nanofibrous mats, the disrupted structure of F-actin filaments persists for cultures after hypoxia and H/R. In contrast, the cellular structure returns to parallel arrangements for cultures grown on PS. HCMs grown on PU nanofibers remain aligned after hypoxia or hypoxia with re-oxygenation.

Fig. 1

Immunofluorescence staining of F-actin (red), HIF-1α (green), and nucleus (blue) in HCM cells cultured on a polystyrene plate (control) and PCL and PU nanofibrous mats under normoxia (21% O2), 6 h hypoxia (1% O2) (A), and 6 h hypoxia + 24 h re-oxygenation. (B) The expression of HIF-1α of HCM cells (C) after hypoxia and (D) after hypoxia with re-oxygenation. *- p < 0.05- statistically significant differences compared to control normoxia, *with line-p < 0.05 statistically significant differences between groups. n > 3. Scale bar 50 μm

Morphological and physiological terms of iPSC-CMs are more similar to fetal than adult cardiomyocytes. Fetal cells rely on anaerobic metabolism, based on glycolysis, to a greater resistance to hypoxia than mature cells. For this reason, the cells adapt to anaerobic conditions and may be less sensitive to the negative effects of oxygen deprivation [38]. Based on the above, iPSC-CMs were used as a promising cell model in our studies.

The results of iPSC-CM staining of F-actin and HIF-1α are shown in Fig. 2A and B. In the case of cells cultured on a polystyrene plate under normoxia, a parallel arrangement of F-actin filaments was observed, while in the case of cells subjected to hypoxia, disrupted sarcomeric fibers were noticed. This may indicate cell damage in response to hypoxia. A disrupted sarcomeric distribution also characterized the cells cultured on PCL and PU nanofibrous mats under hypoxia. The cytoskeletal structure of the cells cultured on nanofibrous mats made of PCL was characterized by a more parallel orientation than actin fibers building cells cultured on PU nanofibers. HIF-1α protein level decrease after 6 h of hypoxia was noted for both nanofibrous mats and PS after hypoxia compared to pluripotent stem cell-derived cardiomyocytes cultured in normoxia. It equaled 0.9-fold, 0.6-fold, and 0.3-fold of 6 h normoxia control for PS, PU, and PCL, respectively.

It was noticed that hypoxia and re-oxygenation also disrupted the actin microfilaments of iPSC-CM cells. The cells cultured on PS (control) and PCL nanofibrous mats after H/R showed more parallel arrangement than cells cultured on PU nanofibrous mats. Moreover, an increase of the HIF-1α was observed for the cells after H/R for PS and PCL nanofibrous mats compared to the cells maintained under normoxia (for PS was 1.5-fold and PCL was 1.3-fold of normoxia control). No significant differences have been noted when comparing the arrangement of actin filament in cultures after hypoxia and hypoxia with reoxygenation. In contrast, the level of HIF-1α protein changes significantly depending on the substrate and conditions (hypoxia or H/R).

In Fig. 2C and D, the HIF-1α level decreases significantly for all types of cultures after hypoxia, while for H/R, it increases significantly for cultures grown on PCL nanofibers and PS. For iPSC-CMs cultured on PU nanofibrous mats, there is also an increase in HIF-1α level after H/R, but it is significantly lower than the level of this protein in normoxia control. It seems likely that the observed increase in HIF-1a protein levels after 24-hour reoxygenation on nanofibers, relative to hypoxia without reoxygenation, may be related to the more difficult penetration of oxygen into the nanofiber structure and thus the delayed response of iPSC-CM cells to hypoxic conditions. To our knowledge, there is no explanation in the literature for this mechanism for immature cells cultured on 3D structures. However, this may be because of the impact of the physicochemical properties of the nanofibers on cell functioning.

Comparing the function of HCM and iPSC-CMs cultured on nanofibrous mats, it was noticed that actin damages after hypoxia and maintained in cultures after re-oxygenation. However, it has been noted that the disruption of actin filaments varies depending on the type of culture and nanofiber mats. HCMs cultured on PCL nanofibers showed the greatest F-actin disruption, while iPSC-CMs presented the greatest cytoskeletal disturbances for cultures on PU nanofibrous materials. Also, HCM and iPSC-CM cells differ in HIF-1α expression, which may be due to different cell physiology and their response to hypoxia and culture on 3D structures.

Fig. 2

Immunofluorescence staining of F-actin (red), HIF-1α (green), and nucleus (blue) in iPSC-CM cells cultured on a polystyrene plate (control), and PCL and PU nanofibrous mats under normoxia (21% O2), 6 h hypoxia (1% O2) (A), and hypoxia 6 h + 24 h re-oxygenation. (B) The expression of HIF-1α of iPSC-CM cells (C) after hypoxia and (D) after hypoxia with re-oxygenation. *- p < 0.05- statistically significant differences compared to control normoxia, *with line-p < 0.05 statistically significant differences between groups. n > 3. Scale bar 50 μm

Type of cell death after hypoxia and hypoxia with re-oxygenation

Degradation of F-actin indicated that cells may begin to go into apoptosis. We noticed different HIF-1α levels depending on the culture conditions (normoxia, hypoxia, or hypoxia with re-oxygenation) and the culture substrates (PS, PCL, and PU nanofibrous mats). Hypoxia and hypoxia with re-oxygenation caused a significant increase of apoptotic cells for cultures performed on nanofibrous mats (Fig. 3). For HCM, the number of apoptotic cells was 5.3%, 8.1%, and 24.0%, 21.8% for control normoxia, control hypoxia, hypoxia on PCL, and PU nanofibrous mats, respectively (Fig. 3C). The number of apoptotic HCM cells equaled 3.2%, 8.6%, 29.6%, 23% for control normoxia, control H/R, H/R on PCL, and PU nanofibrous mats, respectively (Fig. 3D).

According to the literature, iPSC-CMs are immature cells that may become less sensitive to anaerobic conditions and undergo more necrotic death during hypoxia than adult cardiomyocytes [39, 40]. iPSC-CMs cultured on PS under hypoxia caused an increase in the number of necrotic cells (10.5%) (Fig. 4A). For cultures performed on PCL and PU nanofibrous mats, it equaled 3.6%. However, we noted that iPSC-CM culture on nanofibrous mats after hypoxia and hypoxia with re-oxygenation showed a much lower percentage of necrotic cells. Culture conducted on nanofiber mats can indirectly affect the cell death pathway that the cell enters. Similarly to HCM cells, the increase of the number of apoptotic iPSC-CM cells under hypoxia was also noticed (for PS, PCL, and PU nanofibrous mats was 15.7%, 28.3%, and 25.9%, respectively) to compare with normoxia control (4.1%) (Fig. 4A and C). The number of apoptotic cells was 4.7%, 16.1%, 25.5%, and 30.1% for control normoxia, control under H/R, PCL, and PU nanofibrous mats under H/R (Fig. 4B and D). We noticed that iPSC-CM culture on nanofibrous mats after hypoxia and hypoxia with re-oxygenation induces apoptotic cell death. This is more characteristic of adult cardiomyocytes in vivo [41].

Based on the above results, HCM and iPSC-CMs cultures grown on nanofibrous mats after hypoxia and hypoxia with re-oxygenation showed a significant increase in number of apoptotic cells compared to cells cultured on polystyrene plates. Moreover, there is an increase in the number of apoptotic cells after H/R compared to cultures after hypoxia, but it is not statistically significant.

Fig. 3

Results of cell death type determination for HCM cells after 6 h of hypoxia (A, C) and hypoxia with re-oxygenation (6/24 h) (B, D) cultured on PU and PCL nanofibrous mats and PS (hypoxia or H/R control) compared to cells cultured in normoxia. *- p < 0.05- statistically significant differences with control normoxia, *with line-p < 0.05 statistically significant differences between groups. n > 3. Scale bar 100 μm

Fig. 4

Results of cell death type determination for iPSC-CMs cells after 6 h of hypoxia (A, C) or hypoxia with re-oxygenation (6/24 h) (B, D) cultured on PU and PCL nanofibers and PS (hypoxia or H/R control) compared to cells cultured in normoxia. *- p < 0.05- statistically significant differences with control normoxia, *with line-p < 0.05 statistically significant differences between groups. n > 3. Scale bar 100 μm

Analysis of gene expression under hypoxia and H/R

The changes in gene expression under hypoxia and H/R by RT-PCR analysis were carried out for five genes (Figs. 5 and 6): hypoxia-inducible factor 1-alpha (HIF-1α), mitogen-activated protein kinase kinase kinase kinase (MAP4K), Troponin T (TNNT2), calcium-ATPase type 2 (SERCA2), and sodium voltage-gated channel alpha subunit 5 (SCN5A). The genes studied were related to cell changes under hypoxia and disorders that may indicate cell dysfunction or damage. Changes in HIF-1α expression are a response of cardiomyocytes to hypoxia and are involved in the triggering of a number of pathophysiological processes [42]; MAP4K is responsible for the regulation of oxidative stress, induction of inflammation and cell death [43], TNNT2 regulates cardiomyocytes function [44], changes in the expression of genes encoding ion pumps (such as SCN5A and SERCA2) may indicate disturbances in membrane potential [45, 46]. For gene expression analysis, PU nanofibrous mats were selected. Performing RT-PCR for cells cultured on PCL nanofiber mats was difficult due to the limited amount of RNA extracted for the study.

For HCM cells, the expression of most genes decreased (HIF-1α, MAP4K, SERCA2, SCN5A) under hypoxia conditions (Fig. 4A). For all tested genes, the greater decrease in level of gene expression for cultures performed on PU nanofibrous mats than on a polystyrene plate were determined. However, only for SCN5A was it statistically significant. The level of the TNNT2 expression increased for both types of cultures under hypoxia (2.4-fold for PS and 1.3-fold for PU nanofibrous mats compared to normoxia control). For HCM cells maintained under hypoxia with re-oxygenation, the level of TNNT2 expression is higher than in controls; however, it is lower compared to cultures under hypoxia. Additionally, there is an increase in the level of MAP4K expression for cells cultured under H/R, which may indicate that re-oxygenation causes increased oxidative stress and triggers inflammation. Also, the expression of SERCA2 increased, and it is higher for cells grown on PU nanofibrous mats. The expression of the other genes (HIF-1a and SCN5A) decreased for H/R compared to the control normoxia.

Fig. 5

Gene expression analysis of HCM cells cultured under hypoxia (A) and hypoxia with re-oxygenation (B). HIF-1α encoding hypoxia-inducible factor 1-alpha, MAP4K encoding mitogen-activated protein kinase kinase kinase kinase, TNNT2 encoding troponin T, SERCA2 encoding calcium-ATPase type 2, and SCN5A encoding sodium channel protein type 5 subunit alpha. * - p < 0.05 – statistically significant differences were determined by comparison with the cells cultured on a polystyrene plate (control normoxia). *with line-p < 0.05 statistically significant differences between groups. n = 3

For iPSC-CMs, the changes in gene expressions were not significant. Figure 6A shows that the expression level of all selected genes decreases under hypoxia for cultures on nanofibrous mats and polystyrene plates compared to normoxia controls. For the HIF-1α, a significant difference in expression was noticed between cells cultured on PU nanofibrous mats and control normoxia. For the other selected genes, the changes in expression are statistically insignificant when cultures on PU nanofibrous mats are compared to normoxia and hypoxia controls. Based on the results in Fig. 6B, the expression of all study genes of cultures grown on PU nanofibrous mats and PS after hypoxia with re-oxygenation increases compared to the expression for cells under hypoxia. The expression level of genes for the test cultures (iPSC after H/R) is similar to that of the normoxia control. These results may indicate that iPSC-CMs cells are less sensitive to hypoxia and that after receiving oxygen again, cellular repair mechanisms restore cells to their pre-hypoxic state in both 2D and 3D cultures.

Fig. 6

Gene expression analysis of iPSC-CMs cells cultured under hypoxia (A) and hypoxia with reoxygenation (B). HIF-1α encoding hypoxia-inducible factor 1-alpha, MAP4K encoding mitogen-activated protein kinase kinase kinase kinase, TNNT2 encoding troponin T, SERCA2 encoding calcium-ATPase type 2, and SCN5A encoding sodium channel protein type 5 subunit alpha. * - p < 0.05 – statistically significant differences were determined by comparison with the cells cultured on a polystyrene plate (control normoxia). n = 3