Morphological Characteristics of Young and Old Murine Hematopoietic Stem Cell Niches, as Modeled In Vitro

The hematopoietic stem cell (HSC) niche undergoes detrimental changes with age. The molecular differences between young and old niches are well studied and understood; however, young and old niches have not yet been extensively characterized in terms of morphology. In the present work, a 2D stromal model of young and old HSC niches isolated from bone marrow was investigated using light and scanning electron microscopy (SEM) to characterize cell density after one, two, or three weeks of culturing, cell shape, and cell surface morphological features. Our work is aimed at identifying morphological differences between young and old niche cells that could be used to discriminate between their respective murine HSC niches. The results show several age-specific morphological characteristics. The old niches differ from the young ones in terms of lower cell proliferating capacity, increased cell size with a flattened appearance, increased number of adipocytes, and the presence of tunneling nanotubes. In addition, proliferating cell clusters are present in the young niches but not in the old niches. Together, these characteristics could be used as a relatively simple and reliable tool to discriminate between young and old murine HSC niches and as a complementary approach to imaging with specific cellular markers.


Introduction
The stem cell niche is a specialized microenvironment with a specific anatomy and function that regulates and maintains particular kinds of stem cells [1]. The most studied stem cells are hematopoietic stem cells (HSCs), which reside in the bone marrow niche. It is widely accepted that HSCs give rise to a complete set of immune and blood progenitors and that signals from the niche environment regulate HSC fate [2]. The HSC niche is a complex physiological environment composed of three main constituents: (a) HSCs and their differentiated myeloid and lymphoid progenitors; (b) a variety of stromal cells, including osteoblasts, osteoclasts, adipocytes, endothelial cells, mesenchymal stromal cells (MSCs), endothelial and perivascular cells, and macrophages; and (c) extracellular matrix, containing various molecules such as fibronectin, laminin, collagen, and a plethora of other components, such as chemoactive compounds, signaling molecules of the nervous system, hormones, cytokines, growth factors, and other molecules. Each of them plays a role in regulating self-renewal, proliferation, HSC differentiation, and HSC population maintenance [3,4]. Due to the vital role of neuroendocrine and immune factors, some authors use the expression "HSC synapse" when speaking of the HSC niche ( Figure 1). Signals from the stroma and other niche constituent cells can reach HSCs directly via cell-cell contacts through gap junctions or adhesion receptors, as well as via paracrine factors in the extracellular matrix [5].
It is known that the hematopoietic system undergoes substantial detrimental changes with age due to a combination of intrinsic and extrinsic factors which manifest as morphological changes to the HSC niche [6]. Briefly, cellintrinsic alterations and underlying molecular mechanisms that lead to age-related defects of HSCs include (a) exhaustion of the stem cell pool, resulting in an inability to regenerate tissues, (b) telomere shortening with consequent cell senescence, (c) changes to microRNAs responsible for regulating the posttranscriptional expression of target genes, (d) DNA damage and mutations together with epigenetic changes, which can lead to the (e) loss of cell polarity and proteostasis, along with (f) a decline in mitochondrial function, accompanied by a decline in biogenesis. Lastly, (g) changes in metabolism and nutrient sensing occur, where signaling pathways involved in anabolic processes accelerate the aging of HSCs. Intrinsic cellular alterations have been extensively reviewed in several studies [7][8][9][10].
All of the intrinsic detrimental changes in aged HSCs listed above result in a decreased ability for self-renewal and myeloid-biased differentiation at the expense of lymphopoiesis [11]. Consequently, an increased number of myeloid leukemia cases and a decreased adaptive immune system are observed in the elderly [12,13].
Apart from intrinsic changes, extrinsic factors such as cytokines, growth factors, hormones, and extracellular matrix in the niche also play a role in the HSC aging process. However, knowledge regarding the specific role of these factors in aging is far less extensive [6].
The molecular mechanisms that cause detrimental effects in aging HSCs are similarly active in other constituents of the niche, resulting in decreased osteogenesis and increased adipogenesis [3], nerve degeneration [14], and impaired endothelial cell function, with consequent loss of vascular integrity and higher microvascular permeability [11]. The most apparent change in the aged stroma is increased adipogenesis, as adipocytes eventually take up the majority of bone marrow (BM) volume [3]. This change is most likely a consequence of intrinsic changes in old MSCs (e.g., changes to the genome, transcriptome, and proteome), reduced concentrations of certain hormones (sex and pituitary hormones), and other factors in the blood plasma [15,16].
Similarly, extracellular matrix components secreted mainly by niche stromal cells undergo specific agingrelated changes. For example, the matrix becomes stiffer due to decreased collagen formation and secretion and a higher mineral phosphate content [17]. Age-related alterations in the HSC niche also affect soluble mediators, such as osteopontin, CXC-chemokine ligand 12, and CCchemokine ligand 5 [10,18,19].
One morphological feature expected to enable the discrimination of cells between young and old niches is tubular membrane structures called tunneling nanotubes (TNTs). TNTs are thin, membranous cellular protrusions filled with cytoskeletal filaments (F-actin) that extend from the plasma membrane and enable intercellular connections over long distances [20]. TNT diameters range from 50 to 200 nm and can reach lengths over 100 μm, while their morphology and composition can vary between and within cell systems [20]. They can transfer cytoplasmic material and organelles between cells [21,22]. They also act as intercellular bridges that play a unique role in tissue repair and inflammation response, embryonic development, collective cell migration, injured cell recovery, cancer treatment resistance, and pathogen propagation [23]. Despite the ubiquitous observation of TNTs in cell cultures, only a few studies have described this phenomenon in vivo or in situ [24]. The significance of TNT biogenesis is unknown, but cells may need these unique structures in specific physiological circumstances to establish a means of intercellular communication not accomplished by other types of intercellular connections.
Understanding the underlying differences between young and aged hematopoietic stem cell niches is crucial for understanding the contributing factors that play a role in facilitating the aging process of HSCs. While there is a significant insight into the different molecular mechanisms that exist in young and aged stem cell niches, extensive morphological analysis of these differences has not been done before.
In this work, we isolated BM cells and cultured them as a 2D model of the niche. Light microscopy (LM) and scanning electron microscopy (SEM) were used to characterize cell shape, proportions of cells with different shapes, and their specific morphological features in 2D models of young and old murine HSC niches. We have focused on the stromal components of the bone marrow niche, composed of matrix components and stromal cells that are nonhematopoietic in nature and can be differentiated into multiple cell lineages. The aim of our work was to identify the morphological characteristics which could be used as a tool for discriminating between young and old murine HSC niches.

Materials and Methods
2.1. Animals. Young (11-12 weeks) and old (65-87 weeks) C57BL/6j mice were kept under controlled, pathogen-free conditions at the Medical Experimental Centre (MEC) in the Faculty of Medicine at the University of Ljubljana. The mice were housed in a 12 h light/dark cycle, with ad libitum access to food and water. In total, seven young and eight old mice were used, and each animal was taken as one biological repetition. The use of animals was approved by the Ethical

Isolation and Cultivation of Bone Marrow (BM) Cells as a Hematopoietic Stem Cell (HSC) Niche
Model. The mice were sacrificed using CO 2 asphyxiation followed by cervical dislocation. For BM cell isolation, we used the bone crush method. We initially tried an isolation protocol with enzymes (collagenase); however, the isolated cells were of lower viability (66%) in comparison with the bone crush method, where enzymes were not used (95.1%) [25]. Furthermore, cell suspensions isolated with collagenase showed an increased concentration of debris and changes in cell morphology. For these reasons, we decided to use the bone crush method for isolating the BM cells, which we previously optimized [26]. In short, once the mice were euthanized, the bones (tibia, femur, humerus, ilia, and vertebrae) were collected, thoroughly cleaned, and crushed with a mortar and pestle together with a cold isolation medium (RPMI-1640 medium, Gibco, Grand Island, NY; 25 mM HEPES, Gibco; 1 mM EDTA, Gibco; 1x PenStrep, Gibco). Cell suspensions were then filtrated through a 40 μm nylon cell strainer (Thermo Fisher Scientific) and centrifuged to remove remaining bone fragments. Once isolated, the cells were cultured according to the procedure used in the Murine Long-Term Culture Initiating Cell (LTC-IC) Assay. Once the pellet was obtained, the cells were resuspended in a commercial long-term culture medium (LTCM; MyeloCult™ M5300 with 10 -6 M hydrocortisone), optimized to grow adipocytes, fibroblasts, and endothelial cells, and seeded into a 12-well plate at a density of 1 − 1:5 × 10 6 cells per cm 2 [27]. Before plating, sterile cover glasses were placed into each individual well to serve as a surface for cells to attach for further SEM analysis. The cells were then cultured for 1-3 weeks in an incubator at 33°C in 5% CO 2, and every week, half of the LTCM was replaced with fresh medium in every well. The process of artificial HSC niche preparation is presented in Figure 2.  This was done to ensure that the cells developed properly and formed adherent stromal layers. Moreover, the cultures were checked to ensure the absence of contamination, which is seen as small unadhered particles in the medium.

Cell Density Assessment.
After one, two, and three weeks of culturing, cell density in the young and old niches was determined from SEM micrographs using the ImageJ software (National Institutes of Health, Bethesda, MD, USA). For each week, at least two biological repetitions (samples from two mice) were used, where the number of cells was determined on an area of at least 2.5 mm 2 .

Sample Preparation and SEM Analysis.
Cover glasses with attached cells were removed from culture plates every week for three weeks (n = 3). The cells were immediately rinsed in a fixative, as described by Hočevar et al. [28].
2.6. Statistical Analysis. All numerical data are expressed as arithmetic means + standard deviations (SD) and were statistically analyzed using the GraphPad Prism software (GraphPad Software, San Diego, CA). For comparison of cell density and distribution of different morphological cell types in the young and old murine HSC niches, ANOVA was used with Bonferroni's post hoc test. For comparison of the size of the flattened cells in the young and old HSC niches, the nonparametric Mann-Whitney U test was used. The difference between groups was considered statistically significant if the p value was less than 0.05.

Different Cell Types of HSC Niche.
Before SEM analysis, cell cultures were monitored using a light microscope to 5 Scanning validate the formation of an artificial HSC niche. The main cell types of the HSC niche were identified based on previous reports, which include adipocytes (containing lipid droplets, Figure 3(a)), fibroblast-like cells (flattened and elongated, Figure 3(b)), and endothelial cells (growing in a line, Figure 3(c)). Therefore, light microscopy results confirmed the formation of the HSC niche in vitro after four weeks of culturing.

Cell Density.
Results showed a higher density of BM stromal cells (adipocytes, fibroblast-like cells, and endothelial cells) in young niches already after one week of culturing in comparison to the old niches. However, the difference was not statistically significant. After two and three weeks of culturing, the difference in cell density became statistically significant, indicating a higher proliferating capacity of young BM cells in comparison with the old cells (Figures 4 and 5). In addition, emerging proliferating cell clusters in young niches were observed after one week of culturing ( Figure 6).

Differences in Cellular Morphological
Characteristics between Young and Old HSC Niches. Using SEM, we identified four predominant, morphologically distinct cell types in both HSC niches. These consist of flattened cells, round cells, adipocytes filled with lipid droplets (observed as round intracellular vesicles), and elongated cells with protrusions ( Figure 7). In both young and old HSC niches, flattened cells were predominant. In the young HSC niches, the proportion of flattened cells was more than 50%, while in the old HSC niches, the portion of flattened cells was around 40%. The round cells made up about 20% of the cells in the young HSC niches and slightly more, about 25%, in the old HSC niches (Figure 8). The rest of the young and old HSC niche cells were elongated with protrusions of different sizes (explained in detail below). As expected, the most significant difference between the young and old HSC niches was in the proportion of adipocytes. We estimate that less than 5% of all cells were adipocytes in the young HSC niches, but for a more accurate estimate, large numbers of cells would need to be analyzed using immunostaining and flow cytometry, for example.
We also compared the average size of flattened cells in the young and old HSC niches. In the old HSC niches, the flat cells were significantly larger (Figure 9).
The most obvious characteristic of the old niche is the increased content of adipocytes that develop from MSCs. SEM micrographs reveal multiple adipocytes in the old niches ( Figure 10). In the young niches, adipocytes are rarely seen (less than 5%).
Adipocytes show a high content of lipid droplets ( Figure 11).
As expected, a morphological characteristic of cells in the old niches, as revealed by SEM analysis, is cells with tunneling nanotubes (TNTs). These are thin cytoplasmic extensions bridging two cells that are more than 100 μm long. TNTs were only observed in the old HSC niche samples and were completely absent in the young HSC niche samples ( Figure 12).

Discussion
Our study provides evidence that 2D models of young and old murine HSC niches have distinct morphological characteristics that can be used to discriminate between them on the basis of the shape and morphological attributes of stromal cells isolated from the respective young and old bone marrows. The young niches contained proliferating cell clusters, which were absent in the old niches. A lower total number of cells characterized the old niches after two and three weeks of culturing (highlighting the higher proliferative capacity of young niche cells), along with larger sizes of flattened cells, higher number of adipocytes, and the presence of cells with TNTs. It is worth noting that in this work, we isolated BM cells and cultured them as a 2D model of the niche, in which we only managed to grow some of the niche cells (endothelial cells, adipocytes, and fibroblast cells).     Asterisk presents a significant difference between young and old niches ( * equals p < 0:05; Mann-Whitney U test).

Scanning
It was previously shown that old bone marrow (BM) cells have a lower proliferative capacity [5,8], resulting in a lower cell density in the HSC niche. In our study, this was confirmed after two and three weeks of culturing. The higher proliferative capacity of the young niches was also demonstrated by the presence of proliferating cell clusters.
After three weeks of culturing, maximum confluency was reached in the young niches, resulting in cell detachment and consequent death (seen as detached cell clusters in Figure 8). Light microscopy and SEM confirmed the presence of different cell types in the HSC niches. Our results demonstrated that the old niches had larger flattened cells,      [25][26][27].
Our results also align with prior literature describing increased quantities of adipocytes in BM with aging. This phenomenon is known as "fatty degeneration," where the adipocytes occupy almost the entire bone marrow space previously occupied by red hematopoietic marrow at a young age [29]. However, the role of adipocytes in BM is still largely unexplored [15,16]. Age-associated adipocyte hypertrophy is associated with increased lipolysis, increased secretion of the leptin polypeptide and inflammatory cytokines, and decreased secretion of anti-inflammatory cytokines and adiponectin [30,31]. It has been shown that an accumulation of adipocytes in aged BM inhibits both hematopoiesis and bone regeneration [17,29]. In this work, we have shown that the presence of adipocytes, as observed by SEM, can be used to distinguish between young and old niches.
The final characteristic we observed that enables one to discriminate between young and old niches is the presence of tunneling nanotubes (TNTs) in some stromal cells of the aged HSC niches. It is assumed that there is a positive correlation between TNT development and cellular stress, in which cells may develop TNTs as a response to stress [20]. Moreover, TNT formation in association with immunity and inflammation has been reported in several studies in recent years [21,32,33]. TNTs have also been shown to present a way of communicating between senescent cells [34]. These authors report the contribution of TNTs toward the expression of senescence markers. Senescence is related to the cellular response to different physiological states, including stress, and is characterized by the permanent exit of the cell from the cell cycle.

Conclusions
We successfully identified morphological differences between young and old HSC niches. The differences with the highest discriminative power are (1) Increased proportion of adipocytes in the old niches (more than 25%) (2) Formation of TNTs by stromal cells in the old niches that were not observed in the young niches, and (3) Larger size of flattened cells in the old niches Differences in the proportion of flattened and roundshaped cells between the young and old HSC niches were also observed but were not statistically significant. In addition, the old niches were characterized by a lower proliferative capacity, as expected.
The formation of TNTs by stromal cells in the old niches is a novel finding that has not yet been documented before and should be explored further. In addition, larger flattened cells are another morphological characteristic of old niches that indicates senescence. We conclude that the morpholog-ical features described in our study could be used as a tool to discriminate between young and old HSC niches. This is of particular interest when manipulations of stem cell niches are studied, such as seeding different combinations of cell types, adding bioactive molecules such as cytokines, or when anticipating adverse effects on a stem cell niche.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Ethical Approval
This study was conducted according to the ARRIVE guidelines (Animal Research: Reporting of In Vivo Experiments). Experiments on animal tissue were approved by the Ethical Committee for Laboratory Animals of the Republic of Slovenia (No. U34401-27/2013/13) by the Administration of the Republic of Slovenia for Food Safety, Veterinary, and Plant Protection.