Unveiling the native architecture of adult cardiac tissue using the 3D-NaissI method

Abstract

Accurately imaging adult cardiac tissue in its native state is essential for regenerative medicine and understanding heart disease. Current fluorescence methods encounter challenges with tissue fixation. Here, we introduce the 3D-NaissI (3D-Native Tissue Imaging) method, which enables rapid, cost-effective imaging of fresh cardiac tissue samples in their closest native state, and has been extended to other tissues. We validated the efficacy of 3D-NaissI in preserving cardiac tissue integrity using small biopsies under hypothermic conditions in phosphate-buffered saline, offering unparalleled resolution in confocal microscopy for imaging fluorescent small molecules and antibodies. Compared to conventional histology, 3D-NaissI preserves cardiac tissue architecture and native protein epitopes, facilitating the use of a wide range of commercial antibodies without unmasking strategies. We successfully identified specific cardiac protein expression patterns in cardiomyocytes (CMs) from rodents and humans, including for the first time ACE2 localization in the lateral membrane/T-Tubules and SGTL2 in the sarcoplasmic reticulum. These findings shed light on COVID-19-related cardiac complications and suggest novel explanations for therapeutic benefits of iSGLT2 in HFpEF patients. Additionally, we challenge the notion of "connexin-43 lateralization" in heart pathology, suggesting it may be an artifact of cardiac fixation, as 3D-NaissI clearly revealed native connexin-43 expression at the lateral membrane of healthy CMs. We also discovered previously undocumented periodic ring-like 3D structures formed by pericytes that cover the lateral surfaces of CMs. These structures, positive for laminin-2, delineate a specific spatial architecture of laminin-2 receptors on the CM surface, underscoring the pivotal role of pericytes in CM function. Lastly, 3D-NaissI facilitates the mapping of native human protein expression in fresh cardiac autopsies, offering insights into both pathological and non-pathological contexts. Therefore, 3D-NaissI provides unparalleled insights into native cardiac tissue biology and holds the promise of advancing our understanding of physiology and pathophysiology, surpassing standard histology in both resolution and accuracy.

Keywords: ACE2; Cardiac tissue; Cardiomyocytes; Claudin-5; Connexin-43; Fluorescence confocal imaging; Pericytes; SGTL2.

Fig. 1

Schematic workflow of 3D-NaissI method. a Procedure from beating heart retrieval to biopsy preparation for confocal imaging. b Confocal imaging of fluorescent staining in fresh cardiac biopsies and z-stack acquisitions

Fig. 2

Optimizing experimental conditions to preserve stability, integrity, and viability of fresh cardiac tissue biopsies using the 3D-NaissI method. a Stability-integrity of fresh cardiac biopsies from mice qualitatively evaluated by confocal imaging following 48 h of fluorescent staining with cell surface probe (wheat germ agglutinin-WGA), cytoskeletal actin (SPY™-actin) and nuclei (DAPI) at 4, 22 or 37 °C and in PBS, HBSS or a cardioplegic solution. Images illustrate cardiomyocyte (CM) cytoarchitecture under the different conditions and are representative of 3–5 independent experiments (5 mice). b Stability-integrity of fresh cardiac biopsies assessed by quantification of the inter-CM space (indicative of the cardiac tissue cohesion) in fluorescent-WGA-stained fresh biopsies at 4 or 22 °C in PBS, HBSS or a cardioplegic solution. Data are mean ± s.d. n = 5 mice (5–10 biopsies/mouse, 1 image /biopsy), one-way ANOVA, Tukey post-hoc test. c Mitochondria viability in fresh cardiac biopsies from mice qualitatively evaluated by confocal imaging following 48 h of fluorescent staining with cell surface probe (WGA)/Mitochondria membrane potential probe (Mitotracker) and nuclei (DAPI) at 4, 22 or 37 °C and in PBS, HBSS or a cardioplegic solution. Images are representative of 3 independent experiments (3 mice). d Negative control for mitotracker probe staining using paraformaldehyde (PFA)-fixed, non-living cardiac biopsies. e Viability of fresh cardiac biopsies from mice assessed by measuring β-adrenergic activity of the tissue. Upper panel: cAMP production quantified in biopsies stimulated or not (basal) with 10 µM isoproterenol (ISO) or 10 µM carvedilol alone or in combination for 30 min at room temperature. Data represent the mean ± s.e.m. of 4 different biopsies from one mouse and are expressed as cAMP concentration (nM)/mg heart. Lower panel: cAMP production quantified in fresh cardiac biopsies collected immediately (Day 0), or 2 (Day 2), or 4 (Day 4) days after collection and stimulated or not (basal) with 10 µM isoproterenol (ISO) for 30 min at room temperature. Data are mean ± s.e.m., n = 3 mice (3–4 biopsies/mouse), one-way ANOVA, Holm-Sidak’s post-hoc test. (* P < 0.05; **** P < 0.0001; ns, not statistically significant)

Fig. 3

Performance of 3D-NaissI compared to conventional histological techniques. a Representative confocal images of fluorescent wheat germ agglutinin (WGA) and DAPI-staining in cardiac tissue (left ventricles) from fresh cardiac biopsies (3D-NaissI method), PFA-fixed-tissue or Formalin-Fixed Paraffin-Embedded (FFPE) tissue. b Quantification of inter-CM space (indicative of the cardiac tissue cohesion) and CM area in fluorescent-WGA-stained cardiac tissue using fresh cardiac biopsies (3D-NAissI; PBS, 4 °C), PFA-fixed tissue or FFPE-tissue. Data are mean ± s.d., n = 3 mice (7–10 biopsies/mouse, 1 image /biopsy), one-way ANOVA, Tukey post-hoc test

Fig. 4

Immunofluorescent staining of fresh cardiac biopsies using 3D-NaissI method. a Schematic illustrating the hypothetical antibody penetrance in fresh cardiac biopsies, highlighting staining differences between whole and transected cardiomyocytes. b Representative images (3 independent experiments) display XY, XZ, YZ projections and associated Z-stacks, illustrating fluorescent staining of a whole cardiomyocyte (left panel) or a sectioned cardiomyocyte (right panel) for wheat germ agglutinin (WGA), α-actinin, DAPI in fresh cardiac biopsies (3D-NaissI, 2-month old male mice). Both cases exhibit WGA-rod shape surface staining, with a notable absence of α-actinin staining is observed exclusively in the whole cardiomyocyte. c Representative images (3 independent experiments) of fluorescent staining for WGA, α-actinin, DAPI in the cardiac tissue from 2-month-old male mice, highlighting the contrast in antibody penetrance between fresh biopsies (3D-NAissI) with whole cardiomyocytes and PFA-fixed- tissue with permeabilized cardiomyocytes

Fig. 5

Artifacts of “connexin 43 lateralization” in formalin-fixed paraffin-embedded (FFPE)-cardiac tissues revealed through 3D-NaissI. a, c Representative images of connexin-43 immunostaining in the presence of fluorescent wheat germ agglutinin (WGA) in fresh-(3D-NaissI) or FFPE- cardiac tissue from 2-month old male X-linked muscular dystrophy (MDX) mice and their respective controls (C57BL6J OlaHsd). Zoomed-in images (right panels) illustrate the localization of connexin-43 at the lateral membrane or the intercalated disc of the cardiomyocyte. b, d Representative images and corresponding quantification of connexin-43 fluorescence intensity along the manual dotted trace (lateral membrane and intercalated disk) performed on confocal images of (b) dystrophin/Connexin-43/DAPI- or (d) Connexin-43/wheat germ agglutinin (WGA)-stained Formalin-Fixed Paraffin-Embedded (FFPE) cardiac tissue or fresh cardiac biopsies (3D-NaissI) from 2-month-old male (b) control or (d) MDX mice. e Corresponding quantification of the inter-cardiomyocyte (CM) space in fluorescent-WGA-stained biopsies from a,c. Data are mean ± s.d. (e, FFPE) n = 2 mice /group; (e, 3D-NaissI) n = 1 mice/group (5–10 biopsies/mouse, 1 image /biopsy)

Fig. 6

Visualization of the native expression pattern of Claudin-5 using the 3D-NaissI method. Representative confocal images (3 independent experiments) of Claudin-5 immunostaining in both fresh (3D-NaissI) or formalin-fixed paraffin-embedded (FFPE) cardiac tissue from 2-month-old male mice, in the presence of wheat germ agglutinin (WGA) and DAPI (upper panels: transversal view; lower panels: longitudinal view). Antibodies targeting either an intracellular epitope (a, Acris antibody) or an extracellular epitope (b, Novus antibody) of Claudin-5 were used

Fig. 7

Visualization of native ACE2 expression in cardiac tissue though the advanced 3D-NaissI Method. Representative images (3–5 independent experiments) of: a ACE2 immunostaining with fluorescent wheat germ agglutinin (WGA) and DAPI in fresh cardiac tissue (3D-NaissI) from 2-month-old male mice, rats or 19-year-old-man. Zoomed-in images (right panels) highlight ACE2 localization within cardiomyocytes. b ACE2-Isolectin B4 (IsoB4)-DAPI co-staining in fresh cardiac tissue from 2-month old male rats. Zoomed-in image (right panels) highlights ACE2 localization within vascular cells (Iso-B4 positive cells). c (Left panels) ACE2-Caveolin-3 (CAV3) co-staining in fresh cardiac tissue from 2-month old male mice. Zoomed-in images emphasize ACE2 positive co-localization with Caveolin-3 within cardiomyocytes (arrows, yellow dots on the merge). (Right panel) Corresponding quantification of Mander’s coefficient for co-localization between ACE2 and CAV3. Data are mean ± s.d. n = 3 mice (3–5 biopsies/mouse, 8–15 CMs/mouse). d Representative ACE2 expression quantified by western blot in 25 and 50 µg lysates from purified adult cardiomyocytes from 2-month-old male mouse

Fig. 8

Visualization of native SGTL2 expression in cardiac tissue though the advanced 3D-NaissI Method. Representative images (3–4 independent experiments) of: a SGTL2 immunostaining with fluorescent wheat germ agglutinin (WGA) and DAPI in fresh cardiac tissue (3D-NaissI) from 2-month-old male mice (3 independent experiments) or a 20-year-old woman. Zoomed-in images (right panels) highlight the punctuated and laminar SGTL2 staining within cardiomyocytes and beneath the plasma membrane (WGA-positive). b SGTL2-Calsequestrin (CASQ)-WGA-DAPI co-staining in fresh cardiac tissue from 2-month-old male mice. Zoomed-in image (right panels) emphasizes SGTL2 positive co-localization with Calsequestrin (yellow dots, arrows) within cardiomyocytes. c Representative SGTL2 expression quantified by western blot in lysates from purified adult cardiomyocytes of 2-month-old male mouse, with or without SGLT2 immunoprecipitation. Untransfected HEK293T cells, which endogenously express SGLT2 (immature form ~ 73 kDa and higher-order oligomers), or HEK293T cells transiently overexpressing different levels of SGLT2, were used as positive controls, as indicated

Fig. 9

Unveiling a novel 3D architectural structure in fresh cardiac tissue through advanced 3D-NaissI imaging. Representative images three-dimensional confocal imaging (3–5 independent experiments) of: a Wheat germ agglutinin (WGA)-DAPI-stained fresh cardiac biopsies from 2-month-old male mice, rat or a 26-year-old-man, revealing unidentified WGA-positive tube-like structures. These structures intricately surround cardiomyocytes along their longitudinal axis, as depicted by α-actinin positive staining (b). These novel 3D-structures are: c Isolectin B4 (IsoB4)-positive, d Neural/glial antigen 2 (NG2)-positive, e Laminin-2-positive and f meticulously organized along the α-dystroglycan positive-staining at the surface of cardiomyocytes. g Schematic illustrating a novel model of pericyte architecture in the cardiac tissue