Generation of Functionally Competent Human Mast Cells from Limited Blood Volumes

Abstract

Mast cells (MCs) are innate immune cells that are derived from CD34⁺ hematopoietic stem/progenitor cells (HSPCs) and mature in peripheral tissues such as skin and mucosa. Mature human MCs can be generated from peripheral blood, but this process requires substantial blood volumes as HSPC frequencies are typically very low. The aim of this study was to validate a new in-house-developed protocol for the generation of MCs from less than 20 mL of peripheral blood. To this end, we used a magnetic bead-based procedure to isolate ‘untouched’ HSPCs from 14 to 16 mL peripheral blood (PB). In total, 12 cultures were set up with blood from seven healthy donors, wherein HSPCs were first expanded for 4 weeks, followed by another 8 weeks of culture in MC maturation-inducing medium. Flowcytometric analysis, histochemical staining, and degranulation assays were used to assess their phenotypic and functional features. Our data show comparable expression of cytoplasmic granules and cell-surface expression of MRGPRX2, FcεR1α, and CD117 in 8/12 blood-derived MCs (PB-MCs) and buffy coat-derived HSPCs (BC-MCs). PB-MCs responded to classic stimulating agents like IgE/anti-IgE and C48/80. Hence, our novel MC generation protocol yields functionally competent MCs with no compromise in their maturation or activation potential despite 12 weeks of in vitro culture.

1. Introduction

Mast cells are tissue-resident innate immune cells that are typically found in connective tissue throughout the body. Their presence is most abundant at sites where “the host meets the environment”, which is the skin and mucosae of respiratory- and gastrointestinal tracts [1]. MCs express a substantial number of cell-surface receptors, including numerous cytokine/chemokine receptors, complement receptors, and pattern recognition receptors, as well as the stem cell factor (SCF) receptor (also known as KIT receptor or CD117), which is crucial for MC survival. Furthermore, MCs express—among others—the high affinity Immunoglobulin E (IgE) receptor (FcεRI) and the receptor of multiple endogenous and exogenous ligands Mas-related G-protein-coupled receptor member X2 (MRGPRX2) [2,3,4,5]. While FcεRI reacts specifically to crosslinking of dimeric IgE, MRGPRX2 activation can be induced by multiple ligands, including C48/80 [6]; both ways of MC activation lead to the release of inflammation-promoting mediators such as histamine, tryptase, and β-hexosaminidase [2,4]. Accordingly, tissue-resident MCs are important regulators of tissue homeostasis, as they regulate functions of other immune and non-immune cells and critically contribute to the first-line defense against invading pathogens, including detoxification of venoms as well as tissue repair responses [7,8]. Furthermore, MCs play a critical role in both IgE-mediated allergic reactions and IgE-independent hypersensitivity reactions [5,9,10,11], but also in numerous other pathological conditions, including cancer, autoimmune disease, cardiovascular disease, neurodegenerative disease, and tissue fibrosis [1,2,3,12,13,14,15]. More specifically, MCs seem capable of disrupting the blood–brain barrier and thereby promoting immune cell infiltration and neuroinflammation. They also contribute to synovial inflammation, as seen in rheumatoid arthritis and tissue damage associated with systemic lupus erythematosus or type 1 diabetes. [16]. Furthermore, mast cells can be found—alongside other immune cells and fibroblasts—in skin and other solid tumors wherein MC mediators have an impact on tumor growth and metastasis [17]. In addition, there is also a group of primary MC disorders that comprises mastocytosis and monoclonal and idiopathic MC activation syndrome [12].

To better understand how human MCs contribute to physiological processes, as well as their dysfunctional behavior in human MC disorders, there is a major need for in vitro models to study (aberrant) differentiation, tissue accumulation, activation and degranulation and, additionally, new drugs or compound screening. Currently, existing MC-based in vitro models are severely hampered by difficulties of obtaining substantial numbers of appropriately matured cells. To circumvent these issues, mature MCs have been isolated from different human tissues (e.g., skin, foreskin, intestines, and lymph vessels), which is also challenging due to the limited numbers of cells that can be retrieved [18,19,20]. Low yields, along with limited proliferative activity and lifespan of mature tissue-derived MCs, make this approach less suitable for large-scale experiments. Alternatively, different human MC lines, for instance, HMC-1 and LAD2, have also been used to explore degranulation and proliferation mechanisms, potential therapeutic intervention, and interaction with other cell types [21,22,23,24,25,26,27]. However, the use of these cell lines is not ideal either, as this requires cellular manipulation to optimize certain responses in vitro, as we have shown for the HMC-1 line [22]. More importantly, the use of human MC lines does not take into account the genetic background against which experimental conditions are tested, given that these cell lines are by no means representative of the genetic diversity present within the healthy and diseased human population of interest.

To study the biological properties of human MCs, primary MCs are mostly generated from CD34⁺ HSPCs isolated from, respectively, the leukocyte fraction of whole blood donations (‘buffy coat’), umbilical cord blood (UCB), PB, or liquid bone marrow (BM) taps of healthy donors [28,29,30,31,32,33,34,35,36]. However, this approach comes with significant limitations. Unlike UCB or BM samples, HSPC frequencies in peripheral blood are usually very low. Hence, in vitro generation of patient-derived MCs requires substantial amounts of starting material (>50 mL of PB) or even a leukapheresis sample [28,37]. This is particularly an issue when studying MC function in pediatric cases or in patients for whom the collection of BM samples is not a part of routine diagnostics. Second, like available MC cell lines, healthy-donor HSPCs lack a relevant germline configuration, and these samples do not contain HSPC clones expressing relevant somatic mutations, like, for example, the KITp.D816V mutation. Thirdly, most protocols described in the literature apply a positive selection approach to obtain highly purified HPC, which involves antibody targeting and magnetic column-based or flowcytometric sorting of CD34⁺ cells while discarding the remaining material. Although this approach might result in high HSPC purity, it will result in lower absolute HPC yield, which may negatively affect the success rate of MC generation in cases where only small volumes of PB are available. Positive selection may also affect downstream molecular characteristics and immunophenotypic behavior, which could result in a selective loss of distinct progenitor subsets [38,39,40,41].

Here, we present a novel in-house-developed culture method for generating mature and functionally competent human MCs from 14 to 20 mL of PB. We demonstrate that these PB-MCs show equal morphological, immunophenotypical (MRGPRX2, CD117, and FcεR1a expression), and functional (IgE and non-IgE mediated degranulation) characteristics as BC-MCs.

2. Results

2.1. Morphologic and Phenotypic Characterization of Human Blood-Derived MC

Supplementary Table S1 lists details on culture media used for the generation of MCs from isolated HSPCs. After 3 days in culture in 96 round-bottom wells, HSPCs exhibited a diameter of approximately 10–15 µm, lacked metachromatic granules, and displayed a large nucleus with minimal cytoplasm. At this stage, the cells were typically surrounded by numerous erythrocytes and other adherently growing non-lineage cells (Figure 1A). Following the initial expansion phase in 96-well round-bottom wells for up to 14 days and subsequent transfer to 6-well plates, the number of erythrocytes and adherent cells slowly declined while the number of MC-like cells steadily increased for up to 8 weeks of culture.

Figure 1.

Microscopic evaluation of mature MCs generated from 14 to 20 mL of peripheral blood. (A) Schematic overview of culture set-up and microscopic evaluation at indicated time points; (B) samples taken from different MC cultures were stained with toluidine blue to evaluate the granular content of cultured cells. Top images show representative cells obtained after 3 days of expansion. Center image shows PB-MCs at 3 months of culture; bottom images show MCs grown from buffy coat-derived HSPCs (BC-MCs). Cells were analyzed at 100× magnification. Note the (variable) granular content of both PB-MCs and BC-MCs as compared to cells present early after culture initiation. The differences in staining intensities observed between PB-MCs and BC-MCs are most likely attributable to manually adjusted variations in light exposure during image capture. The Figure was created with BioRender.com.

After ≥2 months of culture in 6-well plates, there were no erythrocytes or adherent cells left in the cultures, and remaining MCs displayed diameters ranging from 25 to 30 µm. MCs exhibited substantial cytoplasmic accumulation of granules, though there was some variation in granular content (Figure 1B, Supplementary Figure S2). The nucleus was centrally located, and the cytoplasm became moderate to abundant in volume. Additionally, mature MCs displayed a distinct peripheral border. Morphologically, PB-MCs and BC-MCs exhibited similar morphology, suggesting that the differentiation process yields comparable cell phenotypes regardless of HSPC source.

2.2. PB-MCs and BC-MCs Express Comparable Cell-Surface Markers

To further characterize the phenotype of PB-MCs, we assessed the expression of various MC-specific cell-surface receptors following 3 months of in vitro differentiation (Figure 2A). Antibody details are provided in Supplementary Table S2. PB-MCs exhibited high expression of CD117, a hallmark marker of MC progenitors and mature MC, as well as FcεRIα, the α-chain of the high-affinity IgE receptor. In addition, MRGPRX2, a G-protein-coupled receptor also involved in mast cell activation, was expressed, although some inter-donor variation was observed. Despite this variability, the overall expression levels of CD117, FcεRIα, and MRGPRX2 on PB-MCs were comparable to those observed in control BC-MCs, which is indicative of correct MC differentiation. Additional flow cytometry analysis was conducted to evaluate the expression of CD34, CD38, CD45 (early hematopoietic markers), CD13, CD33 (myeloid markers), CD25 and CD30 (associated with systemic mastocytosis), HLA-DR (antigen presentation), CD9 and CD11b (adhesion molecules), CD69 (activation marker), CD40 (TNF receptor family), and CD71 (transferrin receptor). These data illustrate the broader receptor landscape and further confirm comparable mast cell identity of MCs generated from either PB or BC (Figure 2B).

Figure 2.

Cell-surface expression of MC-specific and other hematopoietic markers on CD117⁺, MCs generated from HSPCs isolated from peripheral blood (PB), or buffy coat (BC). (A) Live MC gating was performed as shown in Supplementary Figure S1. Mean fluorescent intensity (difference between unstained cells and cells with antibody bound) was calculated using FlowJo 10.8.1 software. Additional markers shown in (B) for 12 weeks of cultured PB-MCs (orange bars) or BC-MCs (gray bars). Error bars indicate SEM of datasets generated on cultured cells from 3 different buffy-coat or peripheral blood donors. The Figure was created with GraphPad Prism 9.2.0 software.

2.3. PB-MCs Display Robust Degranulation Potential in Response to Various Stimuli

The degranulation capacity of in vitro-generated MCs was assessed by measuring β-hexosaminidase release upon stimulation with IgE/anti-IgE or C48/80; non-specific activation was induced by the addition of calcium ionophore A23187 [42]. Both PB-MCs and BC-MCs demonstrated significant increases in β-hexosaminidase release across all three stimulatory conditions, indicating successful activation of degranulation pathways (Figure 3A). Activation-induced release of β-hexosaminidase was paralleled by significant upregulation of CD63 across different conditions (Figure 3B), confirming no difference in functional competence between PB-MCs and BC-MCs.

Figure 3.

PB-MCs display substantial release of β-hexosaminidase and CD63 upregulation after exposure to classic MC activators IgE/anti-IgE or C48/80. (A) Degranulation induced by various stimuli and (B) CD63 upregulation after successfully induced degranulation. PM-MCs generated from 3 different donors (orange, red, and blue bars) and control BC-MCs (green bars) were sensitized overnight with human-IgE and stimulated with anti-IgE, C48/80, or A23187, as explained in Section 4. Unstimulated MCs (NC) served as controls to correct for background degranulation. Each bar shows the mean ± standard deviation of 6 technical replicates per donor and stimulation condition. Gating for the data shown in (B) is explained in Supplementary Figure S1. The Figure was created with GraphPad Prism 9.2.0 software.

3. Discussion

Human MCs are pivotal players in immune responses and tissue homeostasis, but research on these cells has been constrained by difficulties in obtaining and maintaining sufficient numbers of primary MCs for in vitro research. In this study, we present a novel method for generating mature and functionally competent human MCs from HSPCs present in small volumes of PB using a lineage cell depletion strategy that yields ‘untouched’ HSPCs and some non-lineage cells like erythrocytes and ‘stromal’ cells; the latter are lost over time. Our novel method supports a model wherein morphological transformation from a HSPC-enriched population into mature MC-like cells occurs. Importantly, culture media and growth factors used for MC generation are not different than those used in established protocols. PB-MCs exhibit comparable morphological and functional characteristics to BC-MCs, as evidenced by immunophenotyping analysis before and after exposure to classic MC activators. These findings are consistent with previous studies showing that human MCs derived from different sources exhibit similar cell-surface receptor profiles and morphology upon differentiation in vitro [1,33,35,43]. Although PB-MCs demonstrated robust degranulation in response to IgE/anti-IgE, C48/80, and A23187—as evidenced by β-hexosaminidase release—we observed some donor-dependent variability in degranulation (Figure 3), possibly influenced by donor-specific genetic factors. It should be noted, however, that functional MCs secrete many different types of inflammation-promoting compounds, including lipid mediators and cytokines, which were not quantified in this study. Another important observation to mention is that 8/12 independently initiated PB-MC cultures from HSPCs isolated from seven different donors resulted in sufficient numbers of MCs for characterization. Cell expansion rates in three successful MC cultures are displayed in Supplementary Figure S2. Thus, although our protocol seems capable of generating functionally competent MCs, the success rate is not 100%. This may be due to inter-donor or inter-sample variation in the frequency of MC-committed progenitor cells present at the start of HSPC culture. Another unknown variable is the number of contaminating cell types that pass through the magnetic column along with ‘untouched’ CD34⁺ HSPCs. These include the adherently growing cells visible in the images shown in Figure 1A. Furthermore, the magnetic bead cocktail used for negative enrichment of CD34⁺ HSPCs contains CD123 beads, which are meant to remove mature CD123^high dendritic cells and basophils. While CD123 (the IL3 receptor α chain) is also broadly expressed by different myeloid-committed precursor cells [44], neither IL-3 [45] nor the combination of IL-3 plus IL-6 [46] promotes the survival of mast cell progenitors. Perhaps inter-donor variability in CD123 expression by CD34⁺ HSPCs could explain why one-third of cultures did not yield sufficient MC numbers for characterization. Due to low cell yields, we were, unfortunately, unable to assess CD123 levels at the start of all HSPC cultures from PB. Another striking observation is the residual expression of CD34 on mature MCs (Figure 2B). In mice, CD34 is highly expressed by committed MC progenitors, as well as by their in vitro- or in vivo-differentiated offspring [47]. Functional loss of CD34 leads to enhanced MC aggregation. In the human setting, on the contrary, CD34 is lost during MC differentiation [45]. More research is needed to unravel the biological relevance of this observation.

4. Materials and Methods

4.1. Sample Collection and Processing

Venous blood samples from adult healthy donors were drawn in sodium-heparin tubes. Buffy coats (BC, 50 mL) were prepared during routine processing of whole blood donations by Sanquin Blood Supply, Amsterdam, The Netherlands. PB and BC samples were kept at RT until the isolation of mononuclear cells (MNC) by standard Ficoll Paque density gradient separation.

4.2. Isolation and Ex Vivo Differentiation of HSPC

MNC fractions were first enriched for CD34⁺ HSPCs through depletion of lineage-committed hematopoietic cells using the human lineage cell depletion kit (Miltenyi Biotec, Bergisch Gladbach, Germany). This kit contains a ready-to-use cocktail of biotin-conjugated antibodies against T cells and NK cells (CD2⁺, CD3⁺, CD16⁺, and CD56⁺), B cells (CD19⁺), dendritic cells (CD56⁺, CD123⁺), monocytes (CD14⁺, CD15⁺, CD16⁺, and CD56⁺), macrophages (CD11b⁺, CD14⁺), granulocytes (CD15⁺, CD16⁺), and erythroid cells (CD235a⁺). Labeled cells were subsequently labeled with anti-biotin coated magnetic beads, and depletion of lineage cells was conducted on an automated magnetic cell separator (AutoMACS) according to the manufacturer’s instructions (Miltenyi Biotec). The procedure for generating control MCs was adapted from the procedure described by Folkerts et al. [48]. Briefly, the enriched HSPC fraction was cultured for 4 weeks in polystyrene 96 round-bottom microplates (Corning via Sigma-Aldrich, Amsterdam, The Netherlands) containing MC medium I. Cultures wherein cell density was not increased on day 14 of HSPC expansion were considered unsuccessful. A representative image thereof is shown in Figure 1A. After successful HSPC expansion, the culture medium was gradually replaced by MC medium II, a less nutrient-rich medium that induces MC maturation. Cells were cultured in 6-well plates or tissue culture flasks for up to 12 weeks to allow full differentiation into functionally competent MCs. Detailed information on culture media used can be found in the Supplementary Table S1. In addition, Supplementary Figure S3 shows the decision tree that was used for evaluation and the changes made during the cultures.

4.3. Phenotypic and Functional Characterization of In Vitro-Generated MCs

After 12 weeks of culture, 5 × 10⁵ PB-MCs or BC-MCs were plated onto SHI-FIX coverslips (Everest Biotech, Bicester, UK) in a dome-shaped droplet and allowed to adhere for 1 h at 37 °C. Adherent cells were fixed with Mota’s fixative [49] for 15 min, rinsed with deionized water, and air-dried before staining with 60 g/L toluidine blue (Sigma-Aldrich) for 1 h. The coverslip was mounted on a microscope slide using Kaiser’s glycerin (Merck via Sigma-Aldrich) gelatine and allowed to solidify for at least 30 min before analysis at 100× magnification using an Axiovert microscope equipped with an AxioCAM MR5 camera (Zeiss via Adamas Instrumenten NV, Rhenen, The Netherlands) to assess their granule content.

Degranulation potential was assessed after the addition of 2 µg/mL of human IgE (Merck via Sigma-Aldrich) to MCs cultured overnight in complete MC medium II without SCF. Following IgE sensitization, cells were washed with Tyrode’s Salts buffer (Sigma-Aldrich) supplemented with 1.5 mM CaCl₂ and incubated for 45 min of incubation at 37 °C with the following stimuli: 50 µg/mL compound 48/80 (C48/80, Merck via Sigma-Aldrich) 1 µg/mL mouse anti-human IgE (clone G7-26, BD Biosciences, Drachten, The Netherlands) or 1 µM calcium ionophore A23187 (Merck via Sigma-Aldrich). Unstimulated MCs were used to correct for background degranulation. To measure the level of β-hexosaminidase released by activated MCs, the cells were spun down to collect the supernatant. The cells were subsequently lysed by adding 1% Triton^TM X-100 buffer (Sigma-Aldrich) and vigorous pipetting. Lysed cells and supernatant were both mixed in a 1:1 ratio with freshly prepared 4-methyl umbelliferyl-N-acetyl-β-D-glucosaminide (4-MUG) (1:1 ratio) for 50–60 min. The reaction was stopped by adding 8.2 g/L glycine solution, pH 10.7 (Merck via Sigma-Aldrich), and fluorescence intensity (FI) was measured at 360/452 nm using a GloMax Explorer Multimode Microplate Reader (Promega, Leiden, The Netherlands). Data were corrected for background fluorescence (452 nm), and the β-hexosaminidase release (supernatant) relative to the total β-hexosaminidase content (lysed cells) was calculated using the following formula:

$$\mathrm{\%}\mathrm{\beta }-\mathrm{h}\mathrm{e}\mathrm{x}\mathrm{o}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{d}\mathrm{a}\mathrm{s}\mathrm{e} \mathrm{r}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{s}\mathrm{e} = {\displaystyle \frac{2\times \mathrm{\Delta }\mathrm{s}\mathrm{u}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{t}(=\mathrm{F}\mathrm{I} \mathrm{s}\mathrm{u}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{t}-\mathrm{F}\mathrm{I} \mathrm{u}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d} \mathrm{c}\mathrm{o}\mathrm{n}\mathrm{d}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n})}{\mathrm{\Delta }\mathrm{s}\mathrm{u}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{t}+(4\times \mathrm{\Delta } \mathrm{c}\mathrm{e}\mathrm{l}\mathrm{l} \mathrm{l}\mathrm{y}\mathrm{s}\mathrm{a}\mathrm{t}\mathrm{e})}}$$

To study the cell receptor expression profile of activated MCs, cells were resuspended at 2 × 10⁵ cells in PBS containing 0.5% (w/v) bovine serum albumin, 2 mM EDTA, and 0.05% sodium azide (FACS buffer) and stained with appropriately diluted CD117-, MRGPRX2-, FcεR1α-, and CD63-specific antibodies (Supplementary Table S2). After 30 min incubation on ice, cells were washed in FACS buffer and analyzed on a BD FACS Lyric flow cytometer or a FACSCanto flow cytometer (BD Biosciences). The viability dye 7-AAD was added shortly before data acquisition. Data analysis was conducted using Infinicyt™ Software (CE-IVD version 2.0.6) or FlowJo Software version 10.8.1 (FlowJo LLC, Ashland, OR, USA), after first removing doublets and debris using FSC-H/FSC-A and FSC/SSC plots. MCs were identified by gating on CD117⁺/CD34⁺ populations. The expression of various markers was determined, with the CD117-population serving as a background control, as detailed in Supplementary Figure S1.

5. Conclusions

We present the first morphological and functional data of MCs generated according to a new in-house-developed MC differentiation protocol. Based on the expression of classic cell-surface receptors like FcεRI and MRGPRX2, which can be activated by classic MC-activating compounds, we believe that these MC-like cells are fully competent. Our approach seems to be a significant improvement over traditional MC-generating protocols that rely on a positive selection of CD34⁺ HSPCs; the latter protocols are often associated with reduced HSPC yields and potential loss of crucial progenitor populations. A more explicit comparison, including transcriptomic analysis, is, however, needed to determine the full biological properties of these alternatively generated MCs, as compared to MCs derived from CD34⁺ HSPCs isolated via existing protocols [48]. We further aim to search for (sample-intrinsic) factors that influence the success rate of our MC protocol, and to study whether the addition of replication-deficient stromal cells can shorten the time span needed for the generation of functional MCs. In the case that these follow-up studies indeed show that our MC protocol generates comparable cells as MCs generated via the gold standard method, our protocol may open new avenues for studying the role of MC biology in a variety of diseases, including allergic and atopic disorders, cancer, cardiovascular disease, and autoimmune disorders. Such studies will focus on addressing different potential MC phenotypes associated with specific pathological conditions, as well as setting up in vitro models to study MC function in tissue organoids containing, for instance, tumor cells, stromal cells, and other immune cells that can interact with MCs.

Abbreviations

The following abbreviations are used in this manuscript:

HSPC	Hematopoietic stem and progenitor cells
PB	Peripheral blood
BC	Buffy coat
BM	Bone marrow
SCF	Stem cell factor
MRGPRX2	Mas-related G-protein-coupled receptor member X2
MNC	Mononuclear cells
MC	Mast cells
UCB	Umbilical cord blood

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/ijms27041793/s1.

Author Contributions

W.A.D., P.M.v.H. and P.L.A.v.D. supervised the study. S.J.v.d.M., M.d.T. and V.H.J.v.d.V. generated and analyzed data. S.J.v.d.M., W.A.D. and A.G.S.v.H. drafted the manuscript. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of the Erasmus MC on 6 April 2021 (MEC 2021-0251).

Informed Consent Statement

Healthy donors recruited at Erasmus MC provided written informed consent for the use of surplus blood cells for scientific research according to protocol MEC 2021-0251. Informed consent for supplying pseudonymized buffy coats generated from whole blood donations at Sanquin Blood Supply was organized according to Sanquin’s guidelines.

Data Availability Statement

The original contributions presented in this study are included in the article and Supplementary Materials. Further inquiries can be directed to the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding Statement

This research received no external funding.