Parietal pericardial tissues were obtained from five patients with dilated cardiomyopathy and two heart transplant donors. The patients’ pericardial tissues constituted the case group, while the donors’ pericardial tissues served as the normal control group. Two patient samples and one normal sample were used for transmission electron microscopy (TEM), while the remaining five patient samples and one normal sample were used for histological analysis and scanning electron microscopy (SEM). All human pericardial samples were collected from patients who underwent heart transplantation at the Affiliated Zhengzhou Seventh People’s Hospital of Xinxiang Medical University. Written informed consent was obtained from all participants, and the study was approved by the Ethics Committee of Zhengzhou Seventh People’s Hospital in accordance with the Declaration of Helsinki (Approval No.: 2024[ky-008]). Tissue blocks were excised from the edges of the pericardial incision.

The patient cohort consisted of three males and two females aged 49–64 years (mean age: 56.8 years). The two donors were healthy males aged 32 and 38 years, respectively. The disease duration ranged from 10 to 16 years, with a mean of 12.8 years. Heart failure in all patients was induced by primary (hereditary) dilated cardiomyopathy. The ultrasound diagnosis met the criteria for heart failure. According to the New York Heart Association (NYHA) classification of cardiac function, four cases were classified as class IV and one case as class III (see Table 1).

Fresh pericardial tissue samples (1 cm $\times$ 2 cm) from five patients and one normal control were collected and fixed in 4% paraformaldehyde for 48 hours. The samples were then processed using standard paraffin embedding techniques and sectioned at a thickness of 5 µm. Sections were stained with hematoxylin and eosin (H&E) and mounted with neutral resin. The prepared slides were examined and analyzed under an optical microscope for morphological evaluation.

Fresh pericardial tissues from two patients and one normal control were cut into 1 mm $\times$ 1 mm sections and immediately fixed in 4% glutaraldehyde prepared in phosphate-buffered saline (PBS). The samples were refrigerated at 4 °C for 48 hours. Following primary fixation, the tissue sections were subjected to standard procedures, including secondary fixation, dehydration, and embedding, followed by semi-thin sectioning. Mesothelial cells were localized using Azan staining. Ultra-thin sections were then prepared and stained at the selected sites. Observations and analyses were performed using a Hitachi HT7800 transmission electron microscope (HT7800; Hitachi High-Tech Corporation, Tokyo, Japan).

Human pericardial tissues measuring 2 cm $\times$ 2 cm were fixed in 2.5% glutaraldehyde for 24 hours. The samples were then rinsed with 0.1 mol/L PBS (pH 7.2) for 1 hour, followed by dehydration through a graded ethanol series. After dehydration, the tissues were dried using a critical point dryer. The dried samples were mounted on stubs with conductive adhesive tape and sputter-coated with gold for 1 minute using an ion sputter coater. The specimens were then observed and photographed using a Zeiss EVO-10 scanning electron microscope (EVO-10; Carl Zeiss AG, Oberkochen, Baden-Württemberg, Germany).

Fresh pericardial tissues measuring 1 cm $\times$ 2 cm were cryo-embedded and sectioned at a thickness of 8 µm. The sections were fixed in cold acetone at 4 °C for 10 minutes, followed by washing with PBS. After blocking with 10% goat serum at 37 °C for 30 minutes, sections were incubated overnight at 4 °C with the primary antibody (rabbit anti-$\beta$-tubulin antibody, Affinity, 1:500). Following PBS washing, sections were incubated at room temperature for 1 hour with the fluorescent secondary antibody (Cy3-conjugated goat anti-rabbit IgG, Beyotime, 1:500), and washed again with PBS. Subsequently, IF488-WGA working solution (Ruibao Biotechnology, 1:250) was applied, followed by DAPI staining for 10 minutes. After final PBS washing, sections were photographed under a fluorescence microscope. Ten visual fields containing mesothelial cells were randomly selected from each section, and the positive area (expressed as a percentage of the total area) was calculated for each field. Data are expressed as mean $\pm$ standard deviation (SD), and comparisons between groups were performed using a t-test. A value of p 0.05 was considered statistically significant.

After extraction of the donor and recipient hearts, 10 mL of pericardial fluid was immediately collected, and 10 inflammatory factors were measured using flow cytometry, including interleukin-10 (IL-10), interleukin-12p70 (IL-12p70), interleukin-17 (IL-17), interleukin- 1 beta (IL-1$\beta$), interleukin-6 (IL-6), interleukin-8 (IL-8), interferon-alpha (IFN-$\alpha$), interferon-gamma (IFN-$\gamma$), tumor necrosis factor-alpha (TNF-$\alpha$), and interleukin-5 (IL-5). The results were analyzed using Prism 7.00 statistical software (7.00; GraphPad Software, Inc., San Diego, CA, USA). A t-test was used to compare data between the two groups, and the results are expressed as the mean $\pm$ SD ($\overline{X}$ $\pm$ SD). A p value 0.05 was considered statistically significant.

Under both light and electron microscopy, there were no significant differences in the tissue or cellular morphology of the parietal pericardium between patients and normal controls. The main structural features were as follows: The mesothelial cells of the serous pericardium were classified as flat, oval, or short columnar in shape. These cells were typically arranged in a single layer, although in some regions, two to four overlapping cells were observed, forming a multilayered arrangement. Most cells exhibited a brush border on the surface facing the pericardial cavity. The nuclei were centrally located and their shapes corresponded to the overall morphology of the cells, appearing flat, round, or oval. The region containing the nucleus was often slightly elevated compared with the surrounding cell surface. The basal surface of the mesothelial cells was supported by a thin basement membrane, beneath which an occasional layer of flattened fibroblasts was present. Below the basement membrane lay the fibrous pericardium, characterized by abundant collagen fiber bundles and a sparse distribution of fibroblasts (Fig. 1A,C; Fig. 2C,a; Fig. 3A).

Fig. 1.

Classification and arrangement of human parietal serous cells. (A) Three types of arrangements: a single layer of flat cells (i.e., squamous epithelial cells), a single layer of columnar cells, and a multilayered arrangement of cells (indicated by arrows), scale bar = 50 µm. (B) Scanning electron microscopy image of the serous cell surface. Arrows in (a) indicate long spindle-shaped cells; arrows in (b) indicate polygonal cells; arrows in (c) indicate oval cells. scale bar = 20 µm. (C) Transmission electron microscopy image of serous cells. (a) Flat and oval mesothelial cells in normal tissue, scale bar = 150 nm. (b) Numerous microvilli are observed on the surface of patient serous cells (arrows), scale bar = 200 nm. (c) Serous cells contain numerous Golgi apparatus (arrows), scale bar = 200 nm. (D) Schematic representation of different mesothelial cell morphologies and their spatial relationships. (a) Columnar mesothelial cells rich in cilia on the cell surface. (b) Elliptical mesothelial cells without cilia on the cell surface. (c) Flat cells without cilia. Three different forms of cells have a complete basement membrane at the base, with a layer of flat fiber cells dispersed below the basement membrane.

Fig. 2.

Surface ultrastructural morphology and structure of human parietal serous cells. (A) Serous cells at the cardiac base were observed under scanning electron microscopy. These cells lack cilia but display abundant surface protrusions. (a) Within the dashed outline is a complete mesothelium with multiple protrusions. The arrow indicates the cellular protrusion, scale bar = 10 µm. (b) A displays long protrusions spanning one or several cells (indicated by arrows), scale bar = 50 µm. (B) Connections between serous cells and pathological alterations of cilia, scale bar = 200 nm. (a,b) show normal serous cells with fewer cilia, while (c) shows patient-derived cells with increased cilia. (C) Basement membrane (arrow). (a) shows an intact, smooth basement membrane in the normal sample, scale bar = 2 µm; (b) shows a disrupted basement membrane in the patient sample, scale bar = 2 µm; (c) shows mesothelial cells from the patient penetrating into the basement membrane, leading to discontinuity, scale bar = 5 µm. a, b, and c all show a large number of disordered cilia on the cell surface. D, Desmosomes (white arrows); TJ, tight junctions (red arrows); ICS, intermediate junctions (white arrows).

Fig. 3.

Pathological changes of cilia in serosal cells. (A) The surface of serosal cells shows a brush border under light microscopy (hematoxylin and eosin [H&E] staining, indicated by arrows), scale bar = 5 µm. (B) Clustered cilia on the cell surface were observed by scanning electron microscopy (arrows), scale bar = 50 µm. (C) Ciliary processes of varying lengths were observed by scanning electron microscopy, scale bar = 10 µm. (D) Normal cell cilia, scale bar = 100 µm. (E) Cilia of patient-derived cells, arrow indicates the cross-section of cilia, scale bar = 50 µm. (F,G) Microtubule-containing cilia (arrows) and microtubule-deficient cilia, which are generally thicker and have a blurred internal matrix, F, scale bar = 30 µm (high mag), scale bar = 50 µm (low mag), G, scale bar = 20 µm. (H,I) Bifurcated cilia (arrows), scale bar = 25 µm. (J) Cilia exhibiting increased stromal density and edema, scale bar = 20 µm. (D–J) Observed under transmission electron microscopy.

Scanning electron microscopy revealed that the serous cells of the parietal layer of the pericardium in both the case group and the normal group exhibited three distinct surface morphologies: spindle-shaped, elliptical, and polygonal (Fig. 1B). Spindle-shaped cells were arranged in parallel, whereas oval-shaped cells were often scattered among other cell types. Polygonal cells, mostly hexagonal, extended small projections that made contact with neighboring cells. The surfaces of polygonal cells displayed numerous finger-like protrusions, while the surfaces of spindle-shaped and oval cells were smooth and devoid of such structures (Fig. 1A,D; Fig. 2A,a,b). Cells with smooth surfaces exhibited central prominences and peripheral extensions of larger cellular projections, resulting in diverse morphologies that often resembled starfish or twisted shapes. These larger protrusions facilitated connections between adjacent cells, with some extensions spanning over one or more cells to establish contact with distant cells (Fig. 2A). Finger-like cilia were present on the surfaces of all three cell types, although their numbers varied (Fig. 2B). Therefore, pericardial serosal cells can be categorized into ciliated and non-ciliated cells. Moreover, the cilia exhibited uneven lengths and irregular arrangements (Fig. 2C; Fig. 3E–J).

Mesothelial cells displayed tight junctions, intermediate junctions, and desmosomal structures along their lateral surfaces (Fig. 2B). In regions lacking specialized junctional complexes, protrusions were observed (Fig. 3C). A uniformly textured basement membrane was present beneath the basal side of the mesothelial cells, under which smaller fibroblasts were identified—consistent with observations under light microscopy (Fig. 1C,a,b). The contact between the basement membrane and mesothelial cells was either smooth or interdigitated in a staggered pattern (Fig. 2C,c). Within the mesothelial cells, an abundant Golgi apparatus and vesicular structures were observed, whereas mitochondria appeared relatively underdeveloped (Fig. 1C,c). These findings provide detailed insights into the ultrastructural features of mesothelial cells in the serous pericardium, elucidating their morphology, junctional complexes, and organelle characteristics.

Overall, no significant differences were observed in the morphological characteristics of serous cells between normal individuals and heart failure patients; the primary pathological changes were confined to the cilia.

Under the light microscope, a brush-like border was observed on the surface of both normal and patient serosal cells (Fig. 3A). Scanning electron microscopy revealed clustered cilia on the cell surface, which varied in length (Fig. 3B,C). Transmission electron microscopy showed that mesothelial cells were mainly flat or oval in shape. Since the cells were examined in cross-section, short cylindrical cells were not observed. Most mesothelial cells displayed uniformly fine protrusions on their surfaces (Fig. 1C,D; Fig. 2B,C; Fig. 3D). In cross-sections of these protrusions, microtubules were arranged circumferentially beneath the cell membrane, with two or three parallel microtubules located at the center of the protrusion (arrows in Fig. 3F–H). However, only a few sections showed typical microtubule structures; most lacked visible microtubules. In longitudinal sections of the protrusions, no obvious microtubules were detected. In both transverse and longitudinal sections, the internal matrix of cilia without microtubules appeared blurred (Fig. 3E–J). Overall, the patient’s cell processes contained few or absent microtubules. Some cilia branched at their bases (arrows in Fig. 3H,I), whereas other cells exhibited smooth surfaces with few or no cilia. Notably, only a small portion of the cilia on the patient’s cell surface showed clearly defined microstructures, while most lacked visible internal organization. Cilia with unclear internal structures appeared thicker, suggesting interstitial expansion, tubulin dissolution, and edema (Fig. 3E–J).

Compared with the normal group, mesothelial cells from patients exhibited an increased number of cilia, which were longer and more sparsely distributed (Fig. 1C, a,b). Electron microscopy revealed that many cilia became thicker, with blurred microtubules and signs of edema. In some regions, numerous serosal cells lacked cilia altogether. Cells without cilia displayed more rough surface protrusions, particularly longer intercellular extensions that enhanced connections among adjacent or neighboring cells.

Immunofluorescence staining revealed strong positive expression of $\beta$-tubulin in the parietal serous cells of patients, whereas normal parietal serous cells showed weak positive or even negative expression. Under high magnification, $\beta$-tubulin was evenly distributed throughout the cytoplasm, with occasional visible cross-sections of microtubules. A significant difference in expression intensity was observed between the two groups (p 0.01). WGA staining was applied to visualize the cell membrane, allowing clear localization of $\beta$-tubulin within the cytoplasm. It is noteworthy that a small amount of $\beta$-tubulin was also detected in various cells of the fibrous pericardium (Fig. 4).

Fig. 4.

Differential expression of $\beta$-tubulin between normal and patient serous cells (arrows). (A) Red fluorescence indicates $\beta$-tubulin, stained by immunofluorescence; green fluorescence marks the cell membrane, stained with WGA, scale bar = 50 µm. (a) Normal parietal serous cells. Arrows indicate serous cell membrane; (b) Serous cells from patients with dilated cardiomyopathy; (c) Fibrous pericardium (EP). (B) High-magnification views of images (A,a) and (A,b), scale bar = 25 µm. (C) Quantitative comparison of $\beta$-tubulin expression in serous cells between normal and dilated cardiomyopathy (DC) groups. N, normal. **p 0.01.

Compared with normal pericardial fluid, the levels of IL-10, IL-6, IL-8, and TNF-$\alpha$ were significantly increased (p 0.05 or p 0.01), whereas the levels of IL-12p70, IL-17, IL-1$\beta$, IL-5, IFN-$\alpha$, and IFN-$\gamma$ showed no significant differences (p $>$ 0.05). The results are presented in Table 2.

The limitation of this study lies in the small sample size used, as obtaining fresh human hearts is difficult, and there is no comparison with heart failure induced by hypertrophic cardiomyopathy and restrictive cardiomyopathy. In addition, if a large animal model of heart failure is prepared and some intervention factors are applied before observing the morphological changes of cilia in pericardial mesothelial cells, it will provide more powerful evidence for further elucidating the mechanism of cilia pathological changes.