Next Article in Journal
Coculture with Neural Stem Cells May Shift the Transcription Profile of Glioblastoma Multiforme towards Cancer-Specific Stemness
Next Article in Special Issue
Morphological and Chemical Investigation of Ovarian Structures in a Bovine Model by Contrast-Enhanced X-ray Imaging and Microscopy
Previous Article in Journal
Detection and Genotypic Analysis of Anaplasma bovis and A. phagocytophilum in Horse Blood and Lung Tissue
Previous Article in Special Issue
Automatic Evaluation for Bioengineering of Human Artificial Ovary: A Model for Fertility Preservation for Prepubertal Female Patients with a Malignant Tumor
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJMS
Volume 24
Issue 4
10.3390/ijms24043240
ijms-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Definition and Quantification of Three-Dimensional Imaging Targets to Phenotype Pre-Eclampsia Subtypes: An Exploratory Study

by
Sammy Hermans
1,
Jacob Pilon
1,
Dennis Eschweiler
2,
Johannes Stegmaier
2,
Carmen A. H. Severens–Rijvers
3,
Salwan Al-Nasiry
4,
Marc van Zandvoort
5,6 and
Dimitrios Kapsokalyvas
1,7,*
1
Department of Genetics and Cell Biology, Maastricht University, 6200 MD Maastricht, The Netherlands
2
Institute of Imaging and Computer Vision, RWTH Aachen University, 52074 Aachen, Germany
3
Pathology, GROW, Maastricht University Medical Centre (MUMC), 6229 HX Maastricht, The Netherlands
4
Obstetrics and Gynaecology, GROW, Maastricht University Medical Centre (MUMC), 6229 HX Maastricht, The Netherlands
5
Department of Genetics and Cell Biology, GROW, CARIM, MHeNS, Maastricht University, 6200 MD Maastricht, The Netherlands
6
Institute for Molecular Cardiovascular Research IMCAR, University Hospital RWTH Aachen, 52074 Aachen, Germany
7
Interdisciplinary Centre for Clinical Research IZKF, University Hospital RWTH Aachen, 52074 Aachen, Germany
*
Author to whom correspondence should be addressed.
Int. J. Mol. Sci. 2023, 24(4), 3240; https://doi.org/10.3390/ijms24043240
Submission received: 21 December 2022 / Revised: 27 January 2023 / Accepted: 31 January 2023 / Published: 7 February 2023
(This article belongs to the Special Issue Advanced Techniques in Reproductive Medicine Research 2.0)
Browse Figures
Review Reports Versions Notes

Abstract

:
Pre-eclampsia is a severe placenta-related complication of pregnancy with limited early diagnostic and therapeutic options. Aetiological knowledge is controversial, and there is no universal consensus on what constitutes the early and late phenotypes of pre-eclampsia. Phenotyping of native placental three-dimensional (3D) morphology offers a novel approach to improve our understanding of the structural placental abnormalities in pre-eclampsia. Healthy and pre-eclamptic placental tissues were imaged with multiphoton microscopy (MPM). Imaging based on inherent signal (collagen, and cytoplasm) and fluorescent staining (nuclei, and blood vessels) enabled the visualization of placental villous tissue with subcellular resolution. Images were analysed with a combination of open source (FIJI, VMTK, Stardist, MATLAB, DBSCAN), and commercially (MATLAB) available software. Trophoblast organization, 3D-villous tree structure, syncytial knots, fibrosis, and 3D-vascular networks were identified as quantifiable imaging targets. Preliminary data indicate increased syncytial knot density with characteristic elongated shape, higher occurrence of paddle-like villous sprouts, abnormal villous volume-to-surface ratio, and decreased vascular density in pre-eclampsia compared to control placentas. The preliminary data presented indicate the potential of quantifying 3D microscopic images for identifying different morphological features and phenotyping pre-eclampsia in placental villous tissue.

1. Introduction

Pre-eclampsia (PE) is a systemic vascular disorder affecting 4–5% of pregnant women globally [1]. PE is a leading cause of maternal and foetal morbidity and mortality, and confers a life-long risk to maternal and infant cardiovascular health [2,3,4,5]. Classically, PE manifests as new-onset hypertension and proteinuria after 20 weeks of gestation [6]. Nevertheless, symptoms, severity, time of onset, and outcome can vary significantly, and early diagnosis remains challenging [6,7].
The wide range of clinical phenotypes makes PE more fitting to be a syndrome instead of a single disease. In particular, the distinction of early- and late-onset PE (EO-PE and LO-PE) is frequently applied, and both are likely to have distinct aetiologies [8,9]. Nonetheless, the exact definition of EO-PE and LO-PE is controversial. The current stratification only focusses on time of onset of clinical symptoms, gestational age at time of birth, or a combination of these, without referring to underlying causes [6,9,10].
Quantifying altered placental morphology in different clinical presentations of PE is key to distinguish the clinical syndrome into distinct pathological phenotypes. Currently, 2D histological evaluation is the gold standard for placental investigation [11,12,13], yet none of the histological findings are specific for PE, i.e., are also observed in other placental syndromes such as intra-uterine growth restriction (IUGR) [14,15,16].
More advanced imaging methods such as confocal microscopy have been used for imaging terminal villous vasculature [17,18,19,20]. However, the villous tree is built up of stem villi (of which the diameter can be up to 3000 µm), intermediate villi, and terminal villi [21]. Multiphoton microscopy (MPM) is a tomographic (3D imaging) technique with subcellular resolution. It is based on multiphoton excited fluorescence and features deep tissue penetration (up to 1 mm) [22]. Therefore, thick tissue slabs (containing different components of the villous tree) can be visualized, closely to their native form. Moreover, MPM is efficient in exciting autofluorescence signals and suitable for the label-free imaging of cytoplasm and connective tissue, even in vivo [23].
MPM has been applied to investigate placental membrane architecture [24,25]. However, to our knowledge, the phenotype quantification of the villous tree and vasculature of healthy and PE placentas with MPM has not been described yet. The aim of this study was to develop a MPM placental imaging protocol and to demonstrate quantifiable imaging targets. Quantification of features such as syncytial knots and vasculature characteristics could aid in a more unequivocal characterization of different PE subtypes and improve disease stratification.

2. Results

2.1. Imaging Targets in Placental Villous Tissue

2.1.1. Trophoblast and Fibrotic Regions: Disorganised Trophoblast in Pre-Eclamptic Placenta

H&E-stained placental chorion illustrated the nuclear accumulations and the villous tree components (Figure 1A,B). As frequently reported in the literature [13,26], EO-PE placenta was characterised by increased syncytial knotting (Figure 1B) in comparison to term control tissue (Figure 1A). As depicted in Figure 1, MPM imaging of unstained tissue (Figure 1C,D and Figure S1) and stained tissue (Figure 1E–H) provided significantly more detailed information about the structural organisation of the trophoblast. Term control placenta was characterized by a structured trophoblastic layer (Figure 1C). Nuclei were closely arranged to each other, and cytoplasmic space was small (inset Figure 1C). The trophoblast of the EO-PE case was disorganised, and numerous nuclear accumulations were present (Figure 1D, Figures S1B and S2A). Nuclei were unevenly distributed, with larger-appearing cytoplasmic spaces between nuclei (inset Figure 1D). Additionally, excessive collagen accumulation, i.e., fibrosis, was especially observed in the stem and intermediate villi of the EO-PE placenta (Figure S2A,B).

2.1.2. Nuclear 3D Architecture: Syncytial Knots Have Distinct 3D Organisations in PE

Distinct nuclear aggregate shapes on the villi were identified (Figure 1E–H). The term “knots” is often used to describe all nuclear aggregates seen in villi. However, nuclear aggregates can be subdivided into (apoptotic) knots, bridges and sprouts. Knots, mostly found towards term, are condensed nuclear aggregates protruding slightly from the villous surface [27]. Bridges are aggregated nuclei that connect two villi and are thought to provide structural support. Sprouts, mainly common in young placentas, are the initiating points for new developing villi [28]. The sprouts (Figure 1F), bridges (Figure 1G), and knots (Figure 1E,H and Figure S3) could be easily distinguished from one another with 3D information. Prominent nuclear accumulations in term placenta were frequently bridges. Although bridges can be distinguished on 2D slices (Figure 1A), their interpretation is imaging-depth dependent, and they are easily misinterpreted as knots without additional 3D information (Figure S3).
EO-PE placentas were characterized by the presence of wave-like syncytial knots (WLKs) (Figure 1E). Although the WLKs were most dominantly present on stem villi (Figure 1H), this pattern was also observed on intermediate villi (Figure 3A lower red arrow) and in villi from a hypertensive pregnancy (Figure S1D). The LO-PE placenta had numerous large, surface-extending knots (Figure 1H). Knots in term control placenta were frequently small, regional, and mildly protruding (Figure 2E). Additionally, knots in term control placenta were observed in close relation with so-called vasculosyncytial membranes (Figure S3), which represent major areas of feto-maternal exchange [29].

2.1.3. Villous Morphology and Vasculature

Villous 3D shapes were highly varying (Figure 3A–D). Depending on achievable imaging depth, linkages of terminal villi, intermediate villi, and stem villi could be visualised. The investigated EO-PE placenta had large paddle-like villous endings (Figure 3A) characteristic of PE placenta. These paddle-terminals were not uncommon and frequently present on villous tips (Figure S7). Term control placenta had shorter but highly condensed villi (Figure 3B), while branching appeared excessive in an IUGR case (Figure 3C). Notable is that the LO-PE placenta had highly varying regions with limited branching (Figure 3D) and highly branched villi (Video S1). Segmentation procedures of intermediate and terminal villi are illustrated in Figures S8–S11. Visual inspection of images (Figure 3, Figures S11 and S12) shows that vasculature morphology mainly follows the villous morphology, especially in terminal villi.

2.2. Quantification

2.2.1. Quantification of Nuclei and Their Architecture: MODEL Performance

Nuclei in placental chorion are arranged closely to each other and have varying intensities (i.e., knots are brighter and deeper in tissue, signal-to-noise ratio decreases). This makes intensity-based segmentation very challenging. The machine-based learning approach Stardist [30] was employed to segment nuclei. An overview of the developed imaging methodology and quantification is given in Figure S4.
Representative images of placental nuclei of term control and EO-PE are illustrated in Figure 2 and Figure S5. Knots appeared bright and very dense (2D images Figure 2A,E). Corresponding 2D images (Figure 2B,F) and 3D images (Figure 2C,G) segmented with Stardist demonstrate the labels of all individual nuclei identified. These images were used to calculate nuclear density. The classification of knots performed with KnotMiner is presented in Figure 2D,H.
Although non-significant, EO-PE placentas tended towards higher nuclear density compared to preterm control placenta (Figure 2I). Knot fraction (the villous volume occupied by knots) tended to be higher in EO-PE compared to preterm control placenta (Figure 2J). While significance of LO-PE could not be tested (single value), IUGR knot fraction differed non-significantly from term control. LO-PE and IUGR placenta both tended towards increased knot fraction compared to term control placenta. Notably, preterm placenta did not have a lower knot fraction than term control placenta (Figure 2J).
Knot shape was on average slightly (non-significantly) flatter and (significantly) less elongated in term control and preterm placenta (Figure 2K) compared to IUGR and EO-PE placenta, respectively. Individual knot shape quantification is represented in Figure S6.

2.2.2. 3D Native Villous Tree Organisation: Altered Surface–Volume Relations in Pre-Eclampsia

Quantitative results indicated that villous surface area increased disproportionally to volume in the investigated PE placentas (Figure 3E). Diffusion surface area and villous volume had a positive linear relation in term control placenta, whereas in PE a linear relation between surface area and volume was absent (Figure 3E,F).
The SA/vol ratio provides a measure for the ease of diffusive transport and branching in a biological mechanism to increase the SA/vol ratio. No significant difference was found between the SA/vol ratio of chorionic villi of control and PE or IUGR placenta (Figure 3G). However, the villous SA/vol ratio tended to be highest in LO-PE placenta, which could suggest increased villous branching compared to term placenta (Figure 3G). Notably, the villous SA/vol ratio was lowest in the term IUGR placenta (Figure 3G), while villi appeared highly branched (Figure 3C).

2.2.3. 3D Placental Microvasculature: A Reduced Vascular Fraction in Pre-Eclampsia

For vasculature visualization, staining was performed with UEA-1 lectin [31], which is a direct and less complex method compared to immunostaining [17,18,19,20]. Segmentation was performed with Labkit (Figures S8 and S11), and network extraction and quantification with VMTK (Figure S12). The vasculature of term placenta (Figure 4A) appeared highly torturous and branched, whereas LO-PE placenta illustrated a vascular network with longer segments and decreased branching (Figure 4B). Vessel diameters varied throughout a single placental vascular network (Figure S13F,G). Diameters in term control placenta remained steady along a 10 µm equilibrium (Figure S13G). In the LO-PE case, diameters varied highly and could become very small (Figure S13F).
Vascular fraction measured the percentage of villous volume occupied by vascular structures. Our results demonstrate a lower vascular fraction for the EO-PE placenta compared to (pre-)term control placenta (Figure 4C). Moreover, the vascular fraction in EO-PE was comparable to that of a second-trimester placenta (Figure 4C). LO-PE and EO-PE vascular fraction did not differ significantly, although LO-PE tended towards higher vascular fraction than EO-PE. Vascular content in the LO-PE placenta was significantly lower than in term control placenta when parametric statistical tests were used (Figure 4C).
Network properties were quantitatively investigated by examining branchpoint density (see Figure S12 for details on the procedure). Our preliminary data illustrated term placenta vascular networks to have the lowest average branchpoint density (Figure 4D). The highest tortuosity (although not significant) was observed in the IUGR placenta without PE (Figure 4E).

3. Discussion

The clinical classification of the main phenotypes of PE, EO-PE and LO-PE, is based on the gestational age. These phenotypes are believed to have different pathophysiologies. Imaging evidence of these differences have been demonstrated with various methods; however, 3D images with subcellular resolution are not as common. In this study, we explored the possibilities of extracting such information with MPM. The goal was to visualize placental tissue with subcellular resolution in 3D, identify imaging targets, and quantify them. We employed MPM because of its ability to image deeper into tissue, and we used a combination of staining and autofluorescence to visualize nuclei, vasculature, fibrillar collagen, and villi. We developed quantification algorithms with the goal of defining quantifiable characteristics that could be used for the classification of different pre-eclampsia phenotypes. Below, we discuss the significance of focusing on the quantification of such features and related challenges.

3.1. Trophoblast and Knots

Hypoxia (and associated trophoblast damage) is central in current pathophysiological hypotheses of PE [32]. In PE, trophoblast turnover is no longer balanced. Apoptosis is increased [33] and differentiation and fusion of the so-called cytotrophoblast to form the syncytium is decreased [34], while cytotrophoblast proliferation is reported to be increased or unaltered [35,36,37]. Altered trophoblast function provides a potential clarification for the observed structural abnormalities in PE (Figure 1C,D).
Moreover, placental hypoxia has been found to promote the aggregation of syncytial knots [38]. Increased syncytial knotting is a clear hallmark of pre-eclamptic placentas [13,26]. Distinct 3D conformations (Figure 1E–H) of knots are potentially related to different pathological mechanisms (i.e., being apoptotic or a mechanism to sequester affected nuclei and create regions for diffusional exchange). WLKs in particular were dominantly present on the investigated EO-PE placenta (Figure 1E). Nonetheless, wave-like apoptotic shedding is not specific for PE and has been described in stem villi from cases with severe IUGR without PE [26,39]. The quantification of knots (shape/volume) could potentially aid in the subdivision of pathological, apoptotic, or non-pathological (including sprouts and bridges) knots accordingly. To our knowledge, different shapes of knots have been described [26], but quantitative data regarding relative incidences of alternative knot shapes is lacking. The current developed methodology provides an approach for quantitatively investigating knots and their shapes.

3.2. Placental Villous Tree Morphology: Exchange Surface and Volume

Villous branching establishes an increased diffusional surface, and the villous SA/Vol reflects ease of nutrient exchange. Similar to our findings (Figure 3E), a disproportional relationship between surface area and volume has been reported in PE placentas based on a stereological approach [40]. Some stereological estimates only observed altered exchange surface areas in IUGR cases, and not for PE alone [41,42]. Stereology only provides estimates of 3D quantitative parameters, while quantification of MPM images represents the intact 3D villi. A 3D characterization by MPM will therefore aid in more accurate quantification of villous trees to confirm or reject the abovementioned findings.

3.3. Placental Villous 3D Microvasculature: Volume and Network Quantification

Terminal villous shapes are influenced by vascular development, and abnormal vascular development is linked to PE [10,43,44]. A vascular casting study demonstrated a reduction of total vasculature in severe PE [43]. Moreover, a higher frequency of avascular villi were described in EO-PE compared to term control [44], providing a potential explanation for reduced vascular fraction (Figure 4C). Moreover, vessel diameters are hypothesized to be influenced by placental hypoxia [10]. Therefore, vessel diameter patterns (Figure S13) provide an interesting imaging target. MPM has enabled the investigation of intact 3D microvascular networks and network quantification. In addition to branchpoint density and tortuosity (Figure 4D,E), network quantification can provide numerous quantitative morphological descriptors, such as vessel length and vessel diameter (Figure S13). According to current hypotheses (Figure S14), LO-PE exhibits increased branching, i.e., predominant vascular development by branching angiogenesis. Conversely, EO-PE placental development is hypothesised to be characterized by a predominance of non-branching angiogenesis [10,45].
During normal development, non-branching angiogenesis is the dominant process of vascular growth later in gestation [46]. Therefore, a younger gestational age is a potential confounding variable, which indicates the importance of including age-matched control placenta to study vascular network properties in PE placenta. Branching differences between EO-PE and pre-term placentas were observed; however, the collection of such samples is more challenging.
The tortuous nature of the placental vessels makes them longer, increases their surface area and slows down blood flow to increase the time for diffusional exchange [18]. High tortuosity, as observed in a IUGR case (Figure 4E), could be an adaptational response for improving diffusion to compensate for low villous SA/vol (Figure 3G). However, excessive tortuosity in vessels is also linked to pathological processes such as hypertension or weakened artery wall stiffness [47]. Vascular twisting would thus be an interesting feature to investigate further regarding placental disease.

3.4. Fibrosis

Finally, fibrosis, which is a prominent feature of PE, was identified as a quantitative marker (Figures S2 and S10). Since the current analysis focused on intermediate and terminal villi only, fibrosis was not further analysed. Future work focusing on stem villi, where fibrosis was mostly observed, will include the quantification of fibrosis.

3.5. Strengths and Limitations

The imaging of the thick tissue (i.e., no need to make thin sections) allowed the imaging of the nearly intact 3D morphology of placental terminal villi. This supported the accurate identification of knots, bridges and sprouts, and allowed for the assessment of the intact 3D organization of the villous tree and intact microvascular networks. Classical histopathology is only able to give a single-plane 2D representation of these complex 3D structures (Figure 1A,B), and therefore, MPM offers a better alternative. Imaging of even larger fields of view by image stitching (2 × 2,5 mm2) is possible, as demonstrated in Video S1, although it is more time consuming (3 h).
The MPM penetration depth of unprocessed placenta is limited to approximately 200 μm. Consequently, images of small (and frequently incomplete) networks cause risk for sampling bias. This could be addressed by optically clearing the tissue [48], a process that homogenizes the tissue’s refraction index, to allow imaging down to some millimetres. Combining image stitching and optical clearing would increase the imaged volume several fold. Larger coverage would reduce sampling bias and improve image quantification. However, image acquisition time would increase correspondingly. Acquisition speed could be improved with Light sheet microscopy (LSM) [49]; however, most common LSM microscopes are limited in resolution and penetration depth compared to MPM.
The limited sample size only allowed for the generation of preliminary results, but our goal was to develop and demonstrate the method and its potential. Future efforts should focus on investigating variation between placentas with larger sample size. From our preliminary data, it appears that there are large differences between different phenotypes. This information can be used for substantiated effect size determination. Power calculations using this substantiated effect size could be further used to calculate the number of samples needed to extract statistically significant conclusions. Additionally, obtaining a complete overview of clinical information (including doppler ultrasound findings [50]) is crucial for future research, so that observations can be correlated accordingly. Since we are generating a large set of microscopical quantifiable parameters in combination with the clinical information acquired, it would be advantageous to employ machine learning methods to differentiate between different phenotypes in the future.

4. Methods

4.1. Tissue Collection and Study Population

A total of 11 placental biopsies were collected from Maastricht University Medical Centre, the Netherlands, between November 2020 and May 2021. All samples were taken from the inner two thirds of the placental disc. Inclusion and exclusion criteria are as follows: PE was diagnosed if blood pressure was ≥140/90 mmHg on two separate occasions accompanied by proteinuria (>300 mg/24 h) after 20 weeks of pregnancy. In absence of proteinuria women, were still eligible in case of multisystem failure. Depending on onset of symptoms (< or >34 w) and time of birth (<or >37 w), placentas were subdivided into early-onset (n = 2) and late-onset PE (n = 1), respectively. IUGR was suspected if birth weight was below the 10th percentile. Uncomplicated pregnancies were defined as normotensive pregnancies from mothers with no morbidities, without need for additional care, and with delivery ≥ 37 w. Patients with comorbid conditions (e.g., diabetes, asthma) were excluded. C-section and natural birth placenta were both used. A total of 3 placental biopsies from uncomplicated pregnancies, 1 IUGR, and 1 PE sample were collected directly after birth. A total of 6 placental samples (n = 2 PE) were collected from the Pathology department of MUMC (archive) and received on formalin. The archived placentas were stored in anatomically correct order after examination, so the inner two thirds of the placental disc could be easily identified and biopsied. Collection and usage of the placentas was approved by the Medical Ethics Committee Academic Hospital Maastricht and Maastricht University (METC, 16-4-047). Clinical characteristics of all placental samples are summarized in Table 1.

4.2. Tissue Dissection and Fixation

Placental villous tissue (~20 cm3) was sectioned from a central region of the chorion. Samples that were collected directly after birth were stored up to 24 h (depending on the collection time) at 4 °C and frequently washed with phosphate buffered saline (PBS) to remove blood. High blood content can limit imaging depth. Storage did not affect villous and vasculature morphology based on brightfield microscopic observations. Subsequently, it was fixed (1 h) in 4% paraformaldehyde (PFA) and stored in PBS at 4 °C. To increase the number of samples analysed, we also collected formalin fixed samples received from the archive, which were maximally 3 months old. Upon receipt, they were washed 3× with PBS and stored at 4 °C in PBS. Samples from the archive could contain higher blood content, which could limit imaging depth, but imaging more than 150 μm was always possible. Formalin fixation, contrary to organic solvents, does not dehydrate tissue, thus better maintaining the morphology of the tissue. All samples were treated equally in terms of analysis.

4.3. Nuclear and Vascular Fluorescent Dyes

Placental villous tissue slabs (~10 mm3) were fluorescently labelled to visualize nuclei and vascular structures. Tissue was permeabilized in 0.1% triton, 1% BSA (20 min), and incubated with 10 µg/mL Ulex Europaeus Agglutinin I (UEA I), DyLight® 594 (Vector laboratories, CA, USA, cat#DL-1067-1, emmax = 618 nm) for 1–3 h, in the dark at room temperature. After washing with TBS, nuclei were labelled with 5 µM SYTO 13 green (Invitrogen, cat#S7575, emmax = 509 nm) for 30 min in the dark at room temperature.

4.4. Mounting and MPM Imaging

For imaging, samples were mounted on a 50 mm glass bottom petri dish (MatTek, Ashland, MA, USA) filled with PBS. MPM was performed with a Leica TCS SP 5 (Leica Microsystems, Wetzlar, Germany) multiphoton microscope. Excitation was at 780 nm. Fluorescence was collected with a 20x Leica HCX APO L, NA 1.0 water immersion objective or a 25× NA1.05 Olympus objective. Alternatively, the Leica Stellaris 8 Dive (Leica Microsystems, Wetzlar, Germany) employing similar settings was used.
Autofluorescence and second harmonic generation (SHG) from unstained samples was detected by an external detector with bandpass filter 460–525 nm, and a forward detector with a 380–420 nm bandpass filter, correspondingly. Signal of stained samples was detected using 4 channels. SHG was collected as described before. Internal detectors with detection bandwidths 444–496 nm, 508–559 nm, and 586–650 nm recorded autofluorescence, nuclear stain, and vascular stain, respectively.
All images were acquired in 8-bit in bidirectional scanning mode and at least 1024 × 1024 pixels. Typical field of view was 736 × 736 μm2 and imaging depth was around 200 μm and varied between stacks and samples ± 50 μm. Z-step size ranged from 1.5 µm to 5µm. A typical voxel size was 0.7 × 0.7 × 1.5 μm3. Image stitching was performed with the Leica Stellaris Dive microscope equipped with motorized stage.

4.5. Image processing and Quantitative Evaluation

4.5.1. Segmentation and Volume Quantification of Placental Villous Tissue and Vessels

Images were pre-processed, analysed, and viewed with FIJI [51]. Pre-processing consisted of de-noising with a 3D median filter, sigma 1. Subsequently, placental villi (autofluorescence), vessels, and SHG signals were segmented with the FIJI Labkit plugin [52]. Volume and surface areas of vessels and villi were quantified with 3D ImageJ suite [53] (Figures S9 and S10). Binary images were imported into 3D ROImanager [53] to obtain surface area and volume measurements. Villous vascular fraction was calculated by dividing the vascular and villous volume and was converted to percentage. The surface-area-to-volume ratio (SA/Vol) was quantified as a measure for effective diffusion area.

4.5.2. Quantifying Vascular Network Characteristics

Binarized vascular images (exported via FIJI to .nrrd files) were imported into 3Dslicer [54] to quantify vascular network characteristics. The VMTK [55] module of a 3D slicer was used to extract a network model from the vasculature and to provide automatic quantification. More detailed procedures are described in Supplementary Methods.

4.5.3. Nuclear Segmentation and Quantification

Stardist 3D [30] employs a neuronal network to separate densely packed nuclei in 3D image stacks. The model was trained with 5 manually annotated image stacks, and thereafter used to segment placental nuclei.
Quantitative analysis of predicted label maps was performed in FIJI with 3D ROI manager [53]. Nuclear density was defined as the number of nuclei relative to the villous volume (#nuclei/villous volume in mm3).
Segmentation of syncytial knots was performed with KnotMiner, available at https://github.com/stegmaierj/KnotMiner (20 December 2022). KnotMiner was built as an extension package for the MATLAB toolbox SciXMiner [56]. The volume of knots was normalised to the volume of the corresponding villus to determine knot fraction. Knot shape was described by elongation index (EI = intermediate principal axis/longest principal axis) and flatness index (FI = short principal axis/intermediate principal axis).

4.6. Statistical Analysis

Data are given as mean ±SD. GraphPad prism 5.0 was used for statistical analysis and graph representation. ANOVA with Bonferroni post-test was used to assess significant differences in parameters that were normally distributed. In the case of data that were assumed not normally distributed, Kruskal–Wallis with Dunn’s post-test was employed. If normality could not be statistically determined (by Shapiro–Wilk test) because of too small a sample size, the assumption of normality was made. The p values of <0.05 were considered statistically significant. The number images analysed for each case is indicated in Table S1, in the Supplementary Information. All images were treated with the same weight [57,58,59,60,61].

5. Conclusions

Our results demonstrated MPM to be a suitable method to visualize placental villous tissue, and the processing methodology developed has the potential to quantifiably differentiate placental morphologies. To understand quantification results, age-matched healthy controls would be necessary. This methodology could be further utilised to phenotype PE subtypes, and ultimately aid in unravelling the multi-etiological nature of pre-eclampsia. In the future, stratification based on 3D placental morphology could aid in the search for early diagnostic markers and (preventative) treatment options to improve maternal and foetal outcomes.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ijms24043240/s1.

Author Contributions

Conceptualization, S.A.-N., M.v.Z. and D.K.; methodology, D.K., C.A.H.S.–R. and S.A.-N.; software, S.H., D.E. and J.S.; formal analysis, S.H., J.P. and D.K.; investigation, S.H., J.P. and D.K.; resources, C.A.H.S.–R., S.A.-N., M.v.Z. and D.K.; data curation, S.H., J.P. and D.K.; writing—original draft preparation, S.H. and D.K.; writing—review and editing, S.H., J.S., C.A.H.S.–R., M.v.Z., S.A.-N. and D.K.; supervision, S.A.-N., M.v.Z. and D.K.; project administration, S.A.-N., M.v.Z. and D.K.; funding acquisition, S.A.-N., M.v.Z. and D.K. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by funding from the German Research Foundation DFG (DE, Grant No STE 2802/2-1). SH received funding through InterUM BV.

Institutional Review Board Statement

The Medical Ethics Committee Academic Hospital Maastricht and Maastricht University (METC AZM/UMC) approved the collection and usage of placental material on 17 August 2020, with reference number METC16-4-047, considering anonymous and retrospective data collection.

Informed Consent Statement

Patient consent was waived because this research is done on rest material. Patients in MUMC+ are aware, by means of patient information flyers, of the possibility that their rest materials can be used for scientific research. Patients can object to this use for scientific research if they wish not to participate. Objection is checked and in case a patient objects, their material is not used. This procedure is approved by the local ethics committee (METC 16-4-047).

Data Availability Statement

The data supporting these findings can be found at the Department of Genetics and cell biology, Maastricht University.

Acknowledgments

The authors wish to thank the clinical staff at the Department of Obstetrics and Gynaecology, MUMC+, for collecting placental samples for this study.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Abalos, E.; Cuesta, C.; Grosso, A.L.; Chou, D.; Say, L. Global and regional estimates of preeclampsia and eclampsia: A systematic review. Eur. J. Obstet. Gynecol. Reprod. Biol. 2013, 170, 1–7. [Google Scholar] [CrossRef]
  2. Rosenberg, K.R.; Trevathan, W.R. An anthropological perspective on the evolutionary context of preeclampsia in humans. J. Reprod. Immunol. 2007, 76, 91–97. [Google Scholar] [CrossRef]
  3. Duley, L. The Global Impact of Pre-eclampsia and Eclampsia. Semin. Perinatol. 2009, 33, 130–137. [Google Scholar] [CrossRef] [PubMed]
  4. Hakim, J.; Senterman, M.K.; Hakim, A.M. Preeclampsia Is a Biomarker for Vascular Disease in Both Mother and Child: The Need for a Medical Alert System. Int. J. Pediatr. 2013, 2013, 953150. [Google Scholar] [CrossRef] [PubMed]
  5. Davis, E.F.; Lazdam, M.; Lewandowski, A.J.; Worton, S.A.; Kelly, B.; Kenworthy, Y.; Adwani, S.; Wilkinson, A.R.; McCormick, K.; Sargent, I.; et al. Cardiovascular risk factors in children and young adults born to preeclamptic pregnancies: A systematic review. Pediatrics 2012, 129, e1552–e1561. [Google Scholar] [CrossRef]
  6. Phipps, E.A.; Thadhani, R.; Benzing, T.; Karumanchi, S.A. Pre-eclampsia: Pathogenesis, novel diagnostics and therapies. Nat. Rev. Nephrol. 2019, 15, 275–289. [Google Scholar] [CrossRef] [PubMed]
  7. Capriglione, S.; Plotti, F.; Terranova, C.; Gulino, F.A.; Di Guardo, F.; Lopez, S.; Scaletta, G.; Angioli, R. Preeclampsia and the challenge of early prediction: Reality or utopia? State of art and critical review of literature. J. Matern. Neonatal Med. 2020, 33, 677–686. [Google Scholar] [CrossRef]
  8. Raymond, D.; Peterson, E. A critical review of early-onset and late-onset preeclampsia. Obstet. Gynecol. Surv. 2011, 66, 497–506. [Google Scholar] [CrossRef]
  9. Staff, A.C. The two-stage placental model of preeclampsia: An update. J. Reprod. Immunol. 2019, 134–135, 1–10. [Google Scholar] [CrossRef]
  10. Severens-Rijvers, C.A.H. Placental Syndrome: Early Pregnancy Adaptation and Placental Development. Ph.D. Thesis, Maastricht University, Maastricht, The Netherlands, 2018. [Google Scholar]
  11. Vangrieken, P.; Vanterpool, S.F.; van Schooten, F.J.; Al-Nasiry, S.; Andriessen, P.; Degreef, E.; Alfer, J.; Kramer, B.W.; von Rango, U. Histological villous maturation in placentas of complicated pregnancies. Histol. Histopathol. 2020, 35, 849–862. [Google Scholar]
  12. Corrêa, R.R.M.; Gilio, D.B.; Cavellani, C.L.; Paschoini, M.C.; Oliveira, F.A.; Peres, L.C.; Reis, M.A.; Teixeira, V.P.A.; Castro, E.C.C. Placental morphometrical and histopathology changes in the different clinical presentations of Hypertensive Syndromes in Pregnancy. Arch. Gynecol. Obstet. 2008, 277, 201–206. [Google Scholar] [CrossRef]
  13. Fogarty, N.M.E.; Ferguson-Smith, A.C.; Burton, G.J. Syncytial knots (Tenney-parker changes) in the human placenta: Evidence of loss of transcriptional activity and oxidative damage. Am. J. Pathol. 2013, 183, 144–152. [Google Scholar] [CrossRef] [PubMed]
  14. Askar, E.; Selim, S.; Sibai, H. Histological changes of human placenta in early intrauterine growth restriction with and without preeclampsia. J. Med. Histol. 2019, 3, 65–76. [Google Scholar] [CrossRef]
  15. Brosens, I.; Pijnenborg, R.; Vercruysse, L.; Romero, R. The “great Obstetrical Syndromes” are associated with disorders of deep placentation. Am. J. Obstet. Gynecol. 2011, 204, 193–201. [Google Scholar] [CrossRef]
  16. Ridder, A.; Giorgione, V.; Khalil, A.; Thilaganathan, B. Preeclampsia: The relationship between uterine artery blood flow and trophoblast function. Int. J. Mol. Sci. 2019, 20, 3263. [Google Scholar] [CrossRef] [PubMed]
  17. Sargent, J.A.; Roberts, V.H.J.; Gaffney, J.E.; Frias, A.E. Clarification and confocal imaging of the nonhuman primate placental micro-anatomy. Biotechniques 2019, 66, 79–84. [Google Scholar] [CrossRef]
  18. Mayo, R.P.; Abbas, Y.; Charnock-Jones, D.S.; Burton, G.J.; Marom, G. Three-dimensional morphological analysis of placental terminal villi. Interface Focus 2019, 9, 20190037. [Google Scholar] [CrossRef]
  19. Merz, G.; Schwenk, V.; Shah, R.; Salafia, C.; Necaise, P.; Joyce, M.; Villani, T.; Johnson, M.; Crider, N. Three-dimensional rendering and analysis of immunolabeled, clarified human placental villous vascular networks. J. Vis. Exp. 2018, 2018, 57099. [Google Scholar]
  20. Resta, L.; Capobianco, C.; Marzullo, A.; Piscitelli, D.; Sanguedolce, F.P.; Schena, F.; Gesualdo, L. Confocal Laser Scanning Microscope Study of Terminal Villi Vessels in Normal Term and Pre-eclamptic Placentas. Placenta 2006, 27, 735–739. [Google Scholar] [CrossRef]
  21. Baergen, R.N. Manual of Pathology of the Human Placenta; Springer Science & Business Media: Berlin/Heidelberg, Germany, 2011. [Google Scholar]
  22. Helmchen, F.; Denk, W. Deep tissue two-photon microscopy. Nat. Methods 2005, 2, 932–940. [Google Scholar] [CrossRef]
  23. Kapsokalyvas, D.; Cicchi, R.; Bruscino, N.; Alfieri, D.; Prignano, F.; Massi, D.; Lotti, T.; Pavone, F.S.S.F.S. In-vivo imaging of psoriatic lesions with polarization multispectral dermoscopy and multiphoton microscopy. Biomed. Opt. Express 2014, 5, 2405–2419. [Google Scholar] [CrossRef] [PubMed]
  24. Richardson, L.; Vargas, G.; Brown, T.; Ochoa, L.; Trivedi, J.; Kacerovský, M.; Lappas, M.; Menon, R. Redefining 3Dimensional placental membrane microarchitecture using multiphoton microscopy and optical clearing. Placenta 2017, 53, 66–75. [Google Scholar] [CrossRef] [PubMed]
  25. Richardson, L.S.; Vargas, G.; Brown, T.; Ochoa, L.; Sheller-Miller, S.; Saade, G.; Taylor, R.N.; Menon, R. Discovery and Characterization of Human Amniochorionic Membrane Microfractures. Am. J. Pathol. 2017, 187, 2821–2830. [Google Scholar] [CrossRef]
  26. Kaufmann, P.; Huppertz, B. Tenney–parker changes and apoptotic versus necrotic shedding of trophoblast in normal pregnancy and pre-eclampsia. In Pre-Eclampsia: Etiology and Clinical Practice; Cambridge University Press: Cambridge, UK, 2007; pp. 152–163. [Google Scholar]
  27. Coleman, S.J.; Gerza, L.; Jones, C.J.P.; Sibley, C.P.; Aplin, J.; Heazell, A.E.P. Syncytial nuclear aggregates in normal placenta show increased nuclear condensation, but apoptosis and cytoskeletal redistribution are uncommon. Placenta 2013, 34, 449–455. [Google Scholar] [CrossRef] [PubMed]
  28. Mayhew, T.M. Turnover of human villous trophoblast in normal pregnancy: What do we know and what do we need to know? Placenta 2014, 35, 229–240. [Google Scholar] [CrossRef] [PubMed]
  29. Sankar, K.D.; Bhanu, P.S.; Kiran, S.; Ramakrishna, B.A.; Shanthi, V. Vasculosyncytial membrane in relation to syncytial knots complicates the placenta in preeclampsia: A histomorphometrical study. Anat. Cell Biol. 2012, 45, 86–91. [Google Scholar] [CrossRef]
  30. Weigert, M.; Schmidt, U.; Haase, R.; Sugawara, K.; Myers, G. Star-convex polyhedra for 3D object detection and segmentation in microscopy. In Proceedings of the 2020 IEEE Winter Conference on Applications of Computer Vision WACV 2020, Snowmass Village, CO, USA, 1–5 March 2020; pp. 3655–3662. [Google Scholar]
  31. Tatsuzuki, A.; Ezaki, T.; Makino, Y.; Matsuda, Y.; Ohta, H. Characterization of the sugar chain expression of normal term human placental villi using lectin histochemistry combined with immunohistochemistry. Arch. Histol. Cytol. 2009, 72, 35–49. [Google Scholar] [CrossRef]
  32. Huppertz, B. The Critical Role of Abnormal Trophoblast Development in the Etiology of Preeclampsia. Curr. Pharm. Biotechnol. 2018, 19, 771–780. [Google Scholar] [CrossRef]
  33. Tomas, S.Z.; Prusac, I.K.; Roje, D.; Tadin, I. Trophoblast apoptosis in placentas from pregnancies complicated by preeclampsia. Gynecol. Obstet. Investig. 2011, 71, 250–255. [Google Scholar] [CrossRef]
  34. Vargas, A.; Toufaily, C.; LeBellego, F.; Rassart, É.; Lafond, J.; Barbeau, B. Reduced expression of both syncytin 1 and syncytin 2 correlates with severity of preeclampsia. Reprod. Sci. 2011, 18, 1085–1091. [Google Scholar] [CrossRef]
  35. Arnholdt, H.; Meisel, F.; Fandrey, K.; Löhrs, U. Proliferation of villous trophoblast of the human placenta in normal and abnormal pregnancies. Virchows Arch. B Cell Pathol. Incl. Mol. Pathol. 1991, 60, 365–372. [Google Scholar] [CrossRef]
  36. Can, M.; Guven, B.; Bektas, S.; Arikan, I. Oxidative stress and apoptosis in preeclampsia. Tissue Cell 2014, 46, 477–481. [Google Scholar] [CrossRef] [PubMed]
  37. Prusac, I.K.; Tomas, S.Z.; Roje, D. Apoptosis, proliferation and Fas ligand expression in placental trophoblast from pregnancies complicated by HELLP syndrome or pre-eclampsia. Acta Obstet. Gynecol. Scand. 2011, 90, 1157–1163. [Google Scholar] [CrossRef]
  38. Heazell, A.E.P.; Moll, S.J.; Jones, C.J.P.; Baker, P.N.; Crocker, I.P. Formation of Syncytial Knots is Increased by Hyperoxia, Hypoxia and Reactive Oxygen Species. Placenta 2007, 28, S33–S40. [Google Scholar] [CrossRef]
  39. Fitzgerald, B.; Kingdom, J.; Keating, S. Distal villous hypoplasia. Diagn. Histopathol. 2012, 18, 195–200. [Google Scholar] [CrossRef]
  40. Mayhew, T.M. Patterns of villous and intervillous space growth in human placentas from normal and abnormal pregnancies. Eur. J. Obstet. Gynecol. Reprod. Biol. 1996, 68, 75–82. [Google Scholar] [CrossRef] [PubMed]
  41. Mayhew, T.M.; Manwani, R.; Ohadike, C.; Wijesekara, J.; Baker, P.N. The Placenta in Pre-eclampsia and Intrauterine Growth Restriction: Studies on Exchange Surface Areas, Diffusion Distances and Villous Membrane Diffusive Conductances. Placenta 2007, 28, 233–238. [Google Scholar] [CrossRef] [PubMed]
  42. Mayhew, T.M.; Ohadike, C.; Baker, P.N.; Crocker, I.P.; Mitchell, C.; Ong, S.S. Stereological investigation of placental morphology in pregnancies complicated by pre-eclampsia with and without intrauterine growth restriction. Placenta 2003, 24, 219–226. [Google Scholar] [CrossRef]
  43. Yin, G.; Chen, M.; Li, J.; Zhao, X.; Yang, S.; Li, X.; Yuan, Z.; Wu, A. Vascular corrosion casting of normal and pre-eclamptic placentas. Exp. Ther. Med. 2017, 14, 5535–5539. [Google Scholar] [CrossRef]
  44. Devisme, L.; Merlot, B.; Ego, A.; Houfflin-Debarge, V.; Deruelle, P.; Subtil, D. A case-control study of placental lesions associated with pre-eclampsia. Int. J. Gynecol. Obstet. 2013, 120, 165–168. [Google Scholar] [CrossRef]
  45. Mayhew, T.M.; Charnock-Jones, D.S.; Kaufmann, P. Aspects of human fetoplacental vasculogenesis and angiogenesis. III. Changes in complicated pregnancies. Placenta 2004, 25, 127–139. [Google Scholar] [CrossRef] [PubMed]
  46. Kaufmann, P.; Mayhew, T.M.; Charnock-Jones, D.S. Aspects of human fetoplacental vasculogenesis and angiogenesis. II. Changes during normal pregnancy. Placenta 2004, 25, 114–126. [Google Scholar] [CrossRef] [PubMed]
  47. Han, H.-C. Twisted blood vessels: Symptoms, etiology and biomechanical mechanisms. J. Vasc. Res. 2012, 49, 185–197. [Google Scholar] [CrossRef] [PubMed]
  48. Richardson, D.S.; Lichtman, J.W. Clarifying Tissue Clearing. Cell 2015, 162, 246–257. [Google Scholar] [CrossRef]
  49. Power, R.M.; Huisken, J. A guide to light-sheet fluorescence microscopy for multiscale imaging. Nat. Methods 2017, 14, 360–373. [Google Scholar] [CrossRef]
  50. Orabona, R.; Donzelli, C.M.; Falchetti, M.; Santoro, A.; Valcamonico, A.; Frusca, T. Placental histological patterns and uterine artery Doppler velocimetry in pregnancies complicated by early or late pre-eclampsia. Ultrasound Obstet. Gynecol. 2016, 47, 580–585. [Google Scholar] [CrossRef]
  51. Schindelin, J.; Arganda-Carreras, I.; Frise, E.; Kaynig, V.; Longair, M.; Pietzsch, T.; Preibisch, S.; Rueden, C.; Saalfeld, S.; Schmid, B.; et al. Fiji: An open-source platform for biological-image analysis. Nat. Methods 2012, 9, 676–682. [Google Scholar] [CrossRef]
  52. Arzt, M.; Deschamps, J.; Schmied, C.; Pietzsch, T.; Schmidt, D.; Tomancak, P.; Haase, R.; Jug, F. LABKIT: Labeling and Segmentation Toolkit for Big Image Data. Front. Comput. Sci. 2022, 4, 10. [Google Scholar] [CrossRef]
  53. Ollion, J.; Cochennec, J.; Loll, F.; Escudé, C.; Boudier, T. TANGO: A generic tool for high-throughput 3D image analysis for studying nuclear organization. Bioinformatics 2013, 29, 1840–1841. [Google Scholar] [CrossRef]
  54. Kikinis, R.; Pieper, S.D.; Vosburgh, K.G. 3D Slicer: A Platform for Subject-Specific Image Analysis, Visualization, and Clinical Support. In Intraoperative Imaging Image-Guided Therapy; Springer: Berlin/Heidelberg, Germany, 2014; pp. 277–289. [Google Scholar]
  55. Antiga, L.; Piccinelli, M.; Botti, L.; Ene-Iordache, B.; Remuzzi, A.; Steinman, D.A. An image-based modeling framework for patient-specific computational hemodynamics. Med. Biol. Eng. Comput. 2008, 46, 1097–1112. [Google Scholar] [CrossRef]
  56. Mikut, R.; Bartschat, A.; Doneit, W.; Ordiano, J.Á.G.; Schott, B.; Stegmaier, J.; Waczowicz, S.; Reischl, M. The MATLAB Toolbox SciXMiner: User’s Manual and Programmer’s Guide. arXiv 2017, arXiv:1704.03298. [Google Scholar]
  57. Hypertension in pregnancy. Report of the American College of Obstetricians and Gynecologists’ Task Force on Hypertension in Pregnancy. Obstet Gynecol. 2013, 122, 1122–1131.
  58. Ives, C.W.; Sinkey, R.; Rajapreyar, I.; Tita, A.T.; Oparil, S. Preeclampsia—Pathophysiology and Clinical Presentations: JACC State-of-the-Art Review. J. Am. Coll. Cardiol. 2020, 76, 1690–1702. [Google Scholar] [CrossRef] [PubMed]
  59. Ester, M.; Kriegel, H.-P.; Sander, J.; Xu, X. (Eds.) A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise; Kdd: Washington, DC, USA, 1996; pp. 226–231. [Google Scholar]
  60. Zingg, T. Beitrag zur Schotteranalyse; ETH Zurich: Zürich, Switzerland, 1935. [Google Scholar]
  61. Ahmed, A.; Dunk, C.; Ahmad, S.; Khaliq, A. Regulation of placental vascular endothelial growth factor (VEGF) and placenta growth factor (PIGF) and soluble Flt-1 by oxygen—A review. Placenta 2000, 21 (Suppl. A), S16–S24. [Google Scholar] [CrossRef] [PubMed]
Ijms 24 03240 g001 550
Figure 1. H&E-stained villous tissue of (A) term control and (B) pre-eclampsia placenta. Villous tissue is visible in pink, and white regions between villi represent the intervillous space. Green arrows point to stem villi (diameter 100–3000 µm) which contain large and centrally located arteries. The blue and dark-red arrow points to an immature intermediate villous or mature intermediate villous, respectively. Purple arrows indicate terminal villi (diameter 30–80 µm). Regions with fibrin depositions are indicated with squares. Nuclear clusters (bridges (white arrow inset (A), knots (red arrow) or even sprouts (yellow arrow inset (B))) are apparent on term control placenta and excessively present on PE placenta (insets (A,B)). (C,D) Placental tissue imaged with MPM without fluorescent labelling. (C) The epithelial trophoblast layer in term control placenta is organised, while (D) EO-PE placenta had disorganised epithelial trophoblast. (E–H) Placental tissue stained for blood vessels (red) and nuclei (blue) to visualise the different 3D shapes of nuclear clusters. (E) Wave-like knots on a stem villous from EO-PE placenta. (F) The 2nd trimester placenta illustrating a typical sprout and its 3D conformation. (G) Nuclear accumulations that form a bridge typically seen in term control placenta. (H) LO-PE placenta with various knot shapes, ranging from elongated surface extending shapes to small round protruding knots. Green—villous tissue, blue—nuclei, cyan—collagen, red—vessels. Scale bars in (E–H)—100 μm.
Figure 1. H&E-stained villous tissue of (A) term control and (B) pre-eclampsia placenta. Villous tissue is visible in pink, and white regions between villi represent the intervillous space. Green arrows point to stem villi (diameter 100–3000 µm) which contain large and centrally located arteries. The blue and dark-red arrow points to an immature intermediate villous or mature intermediate villous, respectively. Purple arrows indicate terminal villi (diameter 30–80 µm). Regions with fibrin depositions are indicated with squares. Nuclear clusters (bridges (white arrow inset (A), knots (red arrow) or even sprouts (yellow arrow inset (B))) are apparent on term control placenta and excessively present on PE placenta (insets (A,B)). (C,D) Placental tissue imaged with MPM without fluorescent labelling. (C) The epithelial trophoblast layer in term control placenta is organised, while (D) EO-PE placenta had disorganised epithelial trophoblast. (E–H) Placental tissue stained for blood vessels (red) and nuclei (blue) to visualise the different 3D shapes of nuclear clusters. (E) Wave-like knots on a stem villous from EO-PE placenta. (F) The 2nd trimester placenta illustrating a typical sprout and its 3D conformation. (G) Nuclear accumulations that form a bridge typically seen in term control placenta. (H) LO-PE placenta with various knot shapes, ranging from elongated surface extending shapes to small round protruding knots. Green—villous tissue, blue—nuclei, cyan—collagen, red—vessels. Scale bars in (E–H)—100 μm.
Ijms 24 03240 g001
Ijms 24 03240 g002 550
Figure 2. Nuclei segmentation with Stardist and knot segmentation with KnotMiner. (A) EO-PE, 2D section and (B) corresponding segmentation result. Red arrows point to syncytial knots. (C) A 3D raw image of nuclei and the corresponding 3D segmentation result. Larger images are available in Figure S5. (D) Segmentation of knots with KnotMiner, illustrated in the maximum intensity projection of the image. (E) A 2D visualisation of nuclei in term control placenta. The white arrow points to a nuclear bridge (can be recognised by dense nuclei connecting two villi, but nuclei are not as bright as compared to knots). (F) Segmentation of nuclei in term control placenta. (G) The 3D segmentation results of nuclei in term control placenta. (H) Knot segmentation in term control placenta by KnotMiner. (I) Nuclear density (ratio of nuclei over villous volume). EO-PE n = 2 placentas (images: 5), LO-PE: n = 1 placenta, (images: 2), preterm: n = 2 (images 3), term control: n = 2 (images 3), IUGR: n = 1 (images: 2). (J) Knot fraction (percentage of total knot volume over villous volume). (K) average knot morphology described by flatness and elongation index. Flatness index was not significantly different for the presented clinical cases. Samples in (J) and (K); EO-PE: n = 1 placenta (images:4), LO-PE: n = 1 (images: 1), preterm: n = 2 (images: 3), term control: n = 2 (images: 3), IUGR: n = 1 (images: 2). Results are presented as means ± standard deviation. Scale bars in (A,B,D–F): 50 μm. Scales bars in (C), and (G): 200 μm, * p < 0.05.
Figure 2. Nuclei segmentation with Stardist and knot segmentation with KnotMiner. (A) EO-PE, 2D section and (B) corresponding segmentation result. Red arrows point to syncytial knots. (C) A 3D raw image of nuclei and the corresponding 3D segmentation result. Larger images are available in Figure S5. (D) Segmentation of knots with KnotMiner, illustrated in the maximum intensity projection of the image. (E) A 2D visualisation of nuclei in term control placenta. The white arrow points to a nuclear bridge (can be recognised by dense nuclei connecting two villi, but nuclei are not as bright as compared to knots). (F) Segmentation of nuclei in term control placenta. (G) The 3D segmentation results of nuclei in term control placenta. (H) Knot segmentation in term control placenta by KnotMiner. (I) Nuclear density (ratio of nuclei over villous volume). EO-PE n = 2 placentas (images: 5), LO-PE: n = 1 placenta, (images: 2), preterm: n = 2 (images 3), term control: n = 2 (images 3), IUGR: n = 1 (images: 2). (J) Knot fraction (percentage of total knot volume over villous volume). (K) average knot morphology described by flatness and elongation index. Flatness index was not significantly different for the presented clinical cases. Samples in (J) and (K); EO-PE: n = 1 placenta (images:4), LO-PE: n = 1 (images: 1), preterm: n = 2 (images: 3), term control: n = 2 (images: 3), IUGR: n = 1 (images: 2). Results are presented as means ± standard deviation. Scale bars in (A,B,D–F): 50 μm. Scales bars in (C), and (G): 200 μm, * p < 0.05.
Ijms 24 03240 g002
Ijms 24 03240 g003 550
Figure 3. MPM imaging of villous morphology and surface and volume quantification. (A) EO-PE placenta had long slender villi. The inset indicates a paddle-like villous ending which was not uncommon for EO-PE. (B) Term control terminal villi stem branching off. Villi appear more condensed, more branched and shorter compared to EO-PE. (C) IUGR placenta displayed highly branched villi. (D) LO-PE illustrating numerous stem villi. (E) Villous volume and surface area of pre-eclampsia placenta was quantified and non-linearly related. PE data was pooled together (n = 12 images of 2 placentas). (F) Villous surface area and volume were linearly related in quantified term control villi (n = 11 images, of 3 different placentas). (G) Ratio of villous surface area over volume. No clear trend is observed. Green—villous outline, blue—nuclei, red—vascular stain. Red-arrows—(wave-like knots. Yellow arrows—sprouts. Results are presented as means ± standard deviation. Scale bars—200 μm.
Figure 3. MPM imaging of villous morphology and surface and volume quantification. (A) EO-PE placenta had long slender villi. The inset indicates a paddle-like villous ending which was not uncommon for EO-PE. (B) Term control terminal villi stem branching off. Villi appear more condensed, more branched and shorter compared to EO-PE. (C) IUGR placenta displayed highly branched villi. (D) LO-PE illustrating numerous stem villi. (E) Villous volume and surface area of pre-eclampsia placenta was quantified and non-linearly related. PE data was pooled together (n = 12 images of 2 placentas). (F) Villous surface area and volume were linearly related in quantified term control villi (n = 11 images, of 3 different placentas). (G) Ratio of villous surface area over volume. No clear trend is observed. Green—villous outline, blue—nuclei, red—vascular stain. Red-arrows—(wave-like knots. Yellow arrows—sprouts. Results are presented as means ± standard deviation. Scale bars—200 μm.
Ijms 24 03240 g003
Ijms 24 03240 g004 550
Figure 4. Placental 3D vasculature and network quantification parameters. Segmentation of a connected vascular network from (A) term control, (B) LO-PE terminal and intermediate villi. (C) The vascular fraction was calculated as vessel volume divided by corresponding villous volume of each image. EO-PE n = 1 placenta (images: 8), LO-PE: n = 1 (images: 5), preterm: n = 1 (images: 3), term control: n = 3 (images: 9), IUGR: n = 1 (images: 3). (D) Branchpoint density (#branchpoints/network length (µm)). (E) Tortuosity of vessel segments. Tortuosity is defined as the ratio between the length of the curved vessel segment and the line distance in between two branchpoints. Samples in (D,E); EO-PE n = 1 placenta (images: 3), LO-PE: n = 1 (images: 4), preterm: n = 1 (images: 2), term control: n = 2 (images: 3), IUGR: n = 1 (images: 2). ** p < 0.01, *** p < 0.001. Results are presented as means ± standard deviation. Scale bars: 200 μm.
Figure 4. Placental 3D vasculature and network quantification parameters. Segmentation of a connected vascular network from (A) term control, (B) LO-PE terminal and intermediate villi. (C) The vascular fraction was calculated as vessel volume divided by corresponding villous volume of each image. EO-PE n = 1 placenta (images: 8), LO-PE: n = 1 (images: 5), preterm: n = 1 (images: 3), term control: n = 3 (images: 9), IUGR: n = 1 (images: 3). (D) Branchpoint density (#branchpoints/network length (µm)). (E) Tortuosity of vessel segments. Tortuosity is defined as the ratio between the length of the curved vessel segment and the line distance in between two branchpoints. Samples in (D,E); EO-PE n = 1 placenta (images: 3), LO-PE: n = 1 (images: 4), preterm: n = 1 (images: 2), term control: n = 2 (images: 3), IUGR: n = 1 (images: 2). ** p < 0.01, *** p < 0.001. Results are presented as means ± standard deviation. Scale bars: 200 μm.
Ijms 24 03240 g004
Table
Table 1. Clinical characteristics of included patients/placental biopsies. EO-PE—early onset pre-eclampsia, LO-PE—late-onset pre-eclampsia, GA—Gestational age, Gravida/para—number of pregnancies/number of births. SGA/IUGR—Small for Gestational age/intra-uterine growth restriction.
Table 1. Clinical characteristics of included patients/placental biopsies. EO-PE—early onset pre-eclampsia, LO-PE—late-onset pre-eclampsia, GA—Gestational age, Gravida/para—number of pregnancies/number of births. SGA/IUGR—Small for Gestational age/intra-uterine growth restriction.
Placenta GroupMaternal Age (Years)Gravida/ParaGAPlacental Weight (g)Birthweight (g)Collection Method
1Term control301/040 w 2 dNA3750Fresh
2Term control382/140 w 4 dNA4040Fresh
3Term control332/139 w 2 dNA3150Fresh
4EO-PE331/030 w 1 d190 incomplete1080Fresh
5EO-PE261/032 w 3 d2331400Archive
6LO-PE262/037 w 5 d4212058Archive
7SGA/IUGR363/140 w 0 dNA3065Fresh
8Preterm control202/032 w 5 d3282089Archive
9Preterm control212/131 w 5 d362,52134Archive
10Maternal hypertension + IUGR261/037 w 5 d2821895Archive
112nd trimester placenta 212/122 w 6 d73532Archive
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Hermans, S.; Pilon, J.; Eschweiler, D.; Stegmaier, J.; Severens–Rijvers, C.A.H.; Al-Nasiry, S.; van Zandvoort, M.; Kapsokalyvas, D. Definition and Quantification of Three-Dimensional Imaging Targets to Phenotype Pre-Eclampsia Subtypes: An Exploratory Study. Int. J. Mol. Sci. 2023, 24, 3240. https://doi.org/10.3390/ijms24043240

AMA Style

Hermans S, Pilon J, Eschweiler D, Stegmaier J, Severens–Rijvers CAH, Al-Nasiry S, van Zandvoort M, Kapsokalyvas D. Definition and Quantification of Three-Dimensional Imaging Targets to Phenotype Pre-Eclampsia Subtypes: An Exploratory Study. International Journal of Molecular Sciences. 2023; 24(4):3240. https://doi.org/10.3390/ijms24043240

Chicago/Turabian Style

Hermans, Sammy, Jacob Pilon, Dennis Eschweiler, Johannes Stegmaier, Carmen A. H. Severens–Rijvers, Salwan Al-Nasiry, Marc van Zandvoort, and Dimitrios Kapsokalyvas. 2023. "Definition and Quantification of Three-Dimensional Imaging Targets to Phenotype Pre-Eclampsia Subtypes: An Exploratory Study" International Journal of Molecular Sciences 24, no. 4: 3240. https://doi.org/10.3390/ijms24043240

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop