Term placentas (n = 10) from pregnant Duroc sows were collected from University of São Paulo, Pirassununga campus, Pirassununga, São Paulo State, Brazil with the institution’s deed of donation and authorization for use. The samples were initially separated, then pooled to ensure randomness and capture the inherent variability across specimens for the subsequent assays. The samples were washed, kept in ice and transported to the Laboratory of Tissue Engineering of the Department of Surgery of the School of Veterinary Medicine and Animal Science from the University of São Paulo. This research was performed according to institutional ethics committee regulations of the University of São Paulo (CEUAx protocol no. 9963170523). The experimental design is summarized in Figure 1.

The normality of the data was first assessed using the Shapiro-Wilk test. The Mann-Whitney test was applied for quantitative data, with results presented as mean ± standard deviation (SD). Principal component analysis (PCA) was performed for spectroscopic assays, followed by one-way ANOVA to compare the means. Statistical significance was set at p < 0.05, and Tukey’s post hoc test was applied where appropriate. All statistical analyses were conducted using GraphPad Prism 8.0 (GraphPad Software, Inc., San Diego, CA, United States).

Structurally, the porcine placenta is classified as diffuse epitheliochorial, with the allantochorion, or Chorioallantoic membrane (CAM), representing the most abundant, thick, and vascularized portion of the fetal membranes. It is also in direct contact with the maternal endometrium (Klisch and Schraner, 2021). Given these characteristics, the CAM was selected for the production of porcine placental scaffolds due to its higher extracellular matrix content, as well as its highly vascularized microenvironment, which lies at the interface of maternal-fetal communication (Figure 2A). Macroscopically, the decellularization process was confirmed by the gradual discoloration of the tissue as the decellularizing solutions progressed, ultimately resulting in complete whitening of the tissue (Figure 2B). Quantification of total genomic DNA demonstrated a 97.3% reduction in DNA content compared to native samples, indicating the effectiveness of the proposed protocol in removing nucleic acids (Figure 2C). Furthermore, both H&E and DAPI staining revealed the absence of nuclei in the decellularized samples, corroborating the DNA quantification results (Figures 2D–K). H&E staining also showed the preservation of key structural features of the porcine CAM, suggesting that the structural framework was largely intact following decellularization, with no clear signs of extracellular matrix degradation compared to control samples (Figures 2D–G).

The preservation of the key components of the placental extracellular matrix (ECM) ensures that the produced scaffold retains the three-dimensional structure of the native tissue, as well as the molecular composition of the tissue microenvironment, facilitating proper interaction with cellular populations (Almeida et al., 2022b). To assess this, a morphological and ultrastructural characterization was performed on both the native placental tissue and the acellular membranes, to evaluate the integrity of the primary extracellular matrix proteins. Initially, histological analyses (Figure 3) were conducted to evaluate and quantify, through specific staining, total glycosaminoglycans (Figures 3A–E), elastic fibers (Figures 3F–J), and total collagen (Figures 3K–O). In terms of total glycosaminoglycans (GAGs), the semiquantitative analysis of the area stained with Alcian Blue (Figure 3E) showed no significant differences between the native samples (82.628 ± 1.938) and decellularized samples (84.490 ± 5.926). Regarding the elastic fibers, which are responsible for the elasticity and flexibility of the placental tissue and are highlighted in dark purple by Weigert’s Fuchsin-Resorcin staining (Figures 3F–J), no significant differences were observed between the native (28.77 ± 5.400) and the decellularized membranes (26.44 ± 4.270).

Considering that collagen is the most abundant component present in the porcine chorioallantoic membrane, a more in-depth analysis was conducted to assess the preservation of both the composition and the three-dimensional structure of these components. The morphometric analysis, carried out by quantifying the area occupied by total collagen stained blue in Masson’s trichrome staining (Figures 3K–O), revealed no significant changes between the native samples (83.28 ± 22.540) and decellularized samples (85.73 ± 13.50). To evaluate the integrity of both mature and immature collagen fibers, Picrosirius Red staining was performed, which differentiates thick, more mature collagen fibers with reddish and yellowish tones, while thinner, more immature collagen fibers stain in greenish hues (Figures 4A,B). The quantitative analysis also showed no significant changes in either fiber type, suggesting that the heterogeneous framework of collagen fibers withstood the decellularization process (Figures 4C,D). For the thicker collagen fibers, the morphometric distribution values for the native samples were 4.177 ± 0.5314, while for the decellularized samples, they were 4.080 ± 0.497. For the thinner collagen fibers, the morphometric distribution values for the native samples were 2.810 ± 0.4613, while for the decellularized samples, they were 3.373 ± 0.682.

The three-dimensional structure of the placental scaffolds was further evaluated using scanning electron microscopy to assess the integrity of the placental membranes following the decellularization process (Figures 4E–T). In the native tissue, a high density of chorionic villi is observed, accompanied by an intricate vascular network responsible for maternal-fetal exchanges. These structures are predominantly supported by the basal membrane, which is in direct contact with the placental mesenchyme, the area with the highest concentration of extracellular matrix in the allantochorion (Figures 4E–L). These chorionic and vascular structures are no longer present in the decellularized samples, as the detergents effectively removed the cellular components, leaving only the fiber framework. A smooth surface is visible at lower magnification (Figure 4M), indicating the absence of villi. At higher magnification, a complex and intertwined network of fine and thick fibers forms a well-organized mesh (Figures 4N–T). No evidence of fiber degradation or disorganization is seen, suggesting the preservation of the three-dimensional placental connective tissue microarchitecture. These ultrastructural observations, associated with the histological findings, strongly support the conclusion that the placental scaffolds retained their structural integrity after decellularization.

To complement the morphological data, immunohistochemical analyses of three key extracellular matrix proteins, elastin, fibronectin, and laminin, were performed (Figure 5). The morphometric analysis of the area occupied by elastin (Figure 5E), the primary protein constituent of elastic fibers, showed no significant differences between the native samples (40.82 ± 12.68) and decellularized samples (36.30 ± 14.26). In the allantochorion, elastin was predominantly localized in blood vessels, with a smaller presence in the placental stroma (Figures 5A–D). Regarding fibronectin, a glycoprotein essential for cell adhesion and communication between the intra- and extracellular microenvironments, the morphometric analysis of the area occupied (Figure 5J) also revealed no significant differences between the native samples (62.67 ± 13.38) and decellularized samples (56.79 ± 12.15). The distribution of fibronectin in the stroma was similar in both groups, suggesting that this critical protein remained largely intact after the decellularization process (Figures 5F–I). The morphometric analysis of the area occupied by laminin (Figure 5O), another adhesive glycoprotein found in higher amounts in the basal membrane, also showed no significant differences between the two groups. However, a slight decrease was observed in the decellularized samples (38.83 ± 6.257) compared to the native samples (44.42 ± 5.925). Considering that laminin consists of multiple subunits, and this analysis focused on the most abundant subunit, α2 (Sainio and Järveläinen, 2020), the decellularization process may have affected each subunit differently, potentially accounting for the slight decrease. Nevertheless, the distribution pattern of laminin in the basal membrane of the chorionic villi and blood vessels remained intact, suggesting that both the structural and compositional integrity of the scaffolds were preserved (Figures 5K–N).

In previous studies of our group, vibrational spectroscopy using FTIR-ATR and Raman techniques was employed to characterize and assess the physicochemical alterations in native and decellularized samples of porcine ovaries, fallopian tubes, and uterine horns (Almeida et al., 2023; Almeida et al., 2024a; Almeida et al., 2024b). The bands corresponding to amides are particularly prominent in the FTIR-ATR spectra (Figure 6), as they are directly associated with the collagen content of the material. Specifically, Amide I can serve as a marker for the secondary structure of proteins, while Amide II may be linked to the hydration of protein structures and could serve as an indicator of collagen self-assembly (Sohutskay et al., 2020; Wang et al., 2014). Regarding Amide III, it may correlate with the triple helical structures of collagen, particularly when comparing the vibrational ratio of this mode with the band attributed to CH₂ bond deformation centered at 1,450 cm⁻¹ (Wang et al., 2014). Amides A and B are also potentially related to the protein structures that constitute the collagen molecule, with bands centered around 3,280 cm⁻¹ and 2,962 cm⁻¹, respectively. Additionally, FTIR analysis reveals bands around 1,080 cm⁻¹, which indicate the presence of covalently bound sulfated glycosaminoglycans (GAGs) attached to the core proteins (Nara et al., 2015; Wang et al., 2014).

To estimate the approximate content of each component, the areas under each vibrational mode were quantified through integration, after defining the regions corresponding to functional groups and drawing a baseline between the extreme points of each band (Figure 6B). The content of Amide I, Amide II, proteoglycans, and triple helical collagen structures (as indicated by the ratio between Amide III and 1,450 cm⁻¹) did not show statistically significant differences, suggesting the absence of alterations in the extracellular matrix components detected by FTIR.

In addition to the FTIR data, Raman spectroscopy (Figure 6C) was also utilized to investigate the placental ECM components in both native and decellularized membranes. Among the bands identified, those associated with Amide I, centered at 1,665 cm⁻¹, and Amide III, located around 1,245 cm⁻¹, are particularly notable (Arezoo et al., 2021; Bergholt et al., 2019; Nara et al., 2015; Nguyen et al., 2013; Timchenko et al., 2017; Zhang et al., 2011). The band at 1,450 cm⁻¹ corresponds to hydrocarbon vibrations typically associated with elastin and/or collagen (Arezoo et al., 2021), while the band at 1,062 cm⁻¹ is linked to glycosaminoglycans (GAGs) (Arezoo et al., 2021; Bergholt et al., 2019; Nguyen et al., 2013). In the region around 1,005 cm⁻¹, the spectrum corresponds to the aromatic rings of phenylalanine, while the band at 875 cm⁻¹ is attributed to proline and hydroxyproline (Emami et al., 2021; Nguyen et al., 2013; Timchenko et al., 2017; Villaret et al., 2019). At 752 cm⁻¹, the tryptophan spectrum is observed; however, this band may overlap with the elastin spectrum, which is typically identified in the range of 722–725 cm⁻¹ (Ingle et al., 2025; Pielesz et al., 2019).

Principal component analysis (PCA) was conducted on the Raman spectra to identify spectral differences among the samples. Figure 6D displays the score plot, with PC1 accounting for the largest proportion of variance in the data (57.8%), followed by PC2 (30.9%). This analysis suggests a trend of differentiation between the samples, indicating potential spectral differences between the native and decellularized samples. However, the absence of a complete separation between the spectra implies that similar functional groups may still contribute to the data, possibly unaffected by the decellularization process. Figure 6E presents the loading plot for component PC1. The Amide III band significantly contributes to the spectral differentiation, followed by the CH₂ and CH₃ chains, as well as the regions associated with proline, hydroxyproline, and elastin. However, when the regions of the bands are marked and their respective areas integrated, as shown in Figure 6F, including Amide I, Amide III, phenylalanine, proline, hydroxyproline, GAGs, and elastin, with potential contributions from tryptophan, only the latter band displayed statistically significant differences. Thus, the Raman spectroscopy results align with the FTIR data. Finally, the reduction observed in the elastin-associated band was 36.9% when comparing the decellularized samples to the native samples.

The biomechanical performance analysis of the acellular placental biomembranes produced (Figure 7) is intrinsically linked to the structure of the extracellular matrix and directly impacts its functionality (Urbanczyk et al., 2020). By positioning the test specimens in the universal mechanical testing equipment until rupture (Figures 7C–F), three variables related to the mechanical properties of the tissues were evaluated: maximum pulling force (Figure 7G), maximum elongation (Figure 7H), and stiffness (Figure 7I). No statistical difference was observed between native and decellularized samples in the three parameters analyzed; however, with regard to stiffness, there was a slight decrease in the decellularized samples, although this change was not statistically significant. For maximum pulling force, the native samples had an average of 1.928 ± 0.4088 N, while the decellularized samples had an average of 2.075 ± 0.3811 N. For the maximum elongation parameter, the native samples showed an average of 5.006 ± 2.090 mm, and the decellularized samples had an average of 4.460 ± 2.232 mm. Regarding stiffness, the native membranes had an average of 0.4275 ± 0.2279, while the decellularized membranes showed an average of 0.3320 ± 0.07014. Additionally, a swelling test was conducted to assess the fluid retention capacity of the acellular membranes (Figure 7J). The results demonstrated that the volume of the samples significantly increased over 48 h, with the mass of the samples rising by more than 972% compared to the initial mass at 48 h. This indicates a high fluid retention capacity, which may be associated with the high porosity of the material.

To evaluate the biomembrane’s cytocompatibility, 2 cell types were employed to assess both their adhesion and interaction with the scaffolds, as well as cellular viability over a specified period (Figure 8). SEM analysis revealed that canine YS cells could adhere to porcine placental matrices (Figures 8A–F), exhibiting numerous extensions and a high cellular density, indicating a suitable cell-matrix interaction. Regarding viability, no statistically significant differences were observed between cells cultured in 2D on plastic and those cultured in 3D on the scaffolds, indicating the absence of cytotoxicity (Figure 8G). For L929 fibroblasts, cell interaction with the extracellular matrix was also evident through the presence of extensions, along with a notable abundance of rounded structures resembling extracellular vesicles, suggesting active cellular secretion (Figures 8H–M). Furthermore, the results demonstrated high cellular viability in fibroblasts cultured on the scaffolds, with slightly higher rates than those cultured in conventional 2D systems (Figure 8N). These findings suggest that the biomembranes can support the culture of various cell types across different species, indicating their potential for interspecies translational applications.

The biomaterials’ immunogenicity is a crucial factor for their use in preclinical models, as the tissue response to these materials should not incite an exaggerated inflammatory reaction or cause damage to the surrounding tissue (Almeida et al., 2024). To evaluate the biocompatibility of placental matrices, we used a subcutaneous scaffold implant model in immunocompetent Wistar rats (Figures 9, 10). A clinical evaluation was performed throughout the experimental period, revealing no signs of edema, redness, or infection, suggesting that the scaffold did not induce any excessive inflammatory response. Histopathological analysis using H&E staining at 7 days post-implantation revealed that the scaffold remained intact, surrounded by a local inflammatory response with a substantial recruitment of inflammatory cells (Figure 9G). This was expected due to the presence of a foreign body and the trauma caused by the surgical procedure to implant the scaffold (Figure 9B). However, by day 14 post-implantation, the inflammatory response had diminished, with no further leukocytic infiltration observed, as compared to day 7 (Figure 9D). No signs of vacuolization, edema, or tissue degradation were observed around the implant, and the scaffold showed signs of color loss, indicating biomaterial degradation and integration with the host tissue. This observation was further supported by the analysis at 30 days post-implantation, where only degraded remnants of the scaffold were left in the subcutaneous tissue, demonstrating that the biomembrane had been degraded and absorbed by the surrounding tissue without signs of immune rejection or tissue necrosis (Figure 9F).

These findings were further complemented by histological analysis using Masson’s trichrome staining, highlighting the total collagen content in the sample and the surrounding tissue (Figure 10). It is evident that as the subcutaneous implantation period progresses, the scaffold gradually adopts a lighter blue hue, indicating its degradation by the host tissue cells (Figures 10C,G,K). This observation is reinforced by the analysis of the predominant collagen fiber types in the scaffold through polarized light microscopy of Picrosirius Red staining (Figure 10). At 7 days post-implantation, the placental scaffolds display a reddish coloration, which indicates the presence of dense, mature, and still-structured collagen (Figure 10D). By day 14, a significant reduction in birefringence is observed, with the red coloration of the fibers shifting to yellowish and greenish types (Figure 10H), indicating continued degradation of the scaffold. At 30 days, the collagen fibers remaining from the scaffold are almost undetectable, deeply embedded within the adjacent tissue, which further corroborates the previous histopathological findings (Figure 10L). These results suggest that acellular porcine placental biomembranes are both biocompatible and biodegradable, as there is no evidence of an exaggerated immune response, and the near-complete disappearance of the scaffold indicates its absorption by the host’s adjacent tissue.