2.1 Mice

Wnt1-Cre mice, and C57BL/6-Gt (ROSA)^{26Sortm1(EYFP)Cos/}J (Rosa^YFP/YFP) mice were provided by Drs. H. Sucov (Southern California University), H. Enomoto (RIKEN Kobe). C57BL/6-Gt (ROSA)26Sor^{tm1(HBEGF)Awai/}J (Rosa^iDTR/iDTR) mice and C57BL/6 mice were obtained from the Jackson Laboratory and CLEA Japan, respectively. All mice were maintained at the Institute of Laboratory Animals, Mie University; all experimental procedures were approved by the Institutional Animal Care and Use Committee of Mie University (approval number: 24–48) and were performed according to the Mie University guidelines for laboratory animals.

2.2 Genotyping

Genomic DNA samples were prepared from tail tips, and PCR was performed. Primers used for PCR are listed in Table S1.

2.3 TBI

Wnt1-Cre/+; Rosa^YFP/YFP or Wnt1-Cre/+; Rosa^YFP/DTR mice were subjected to lethal irradiation of 9.5 Gray. Three days later, all mice were sacrificed and the thymus was isolated. Single-cell suspension was prepared and the effect of irradiation on the thymus size, the number and percentages of thymocytes, and the gene expression levels of thymic stromal cells were assessed by histology, immunohistochemistry, flow cytometry, and quantitative RT-PCR analysis.

2.4 Adult Thymic Organ Culture (ATOC) and diphtheria toxin (DT) Administration

Thymi were harvested from lethally irradiated (9.5 Gray) 8–10-week-old Wnt1-Cre/+; Rosa^YFP/YFP, Wnt1-Cre/+; Rosa^YFP/DTR, Rosa^YFP/DTR(+/+^YFP/DTR) mice. For in vitro depletion of YFP⁺ NC-derived mesenchymal cells, thymus lobes were incubated on a Whatman NucleporeTM Track-Etched membrane (GE Healthcare) [39] in the presence of 100 ng/mL DT (Sigma-Aldrich) with RPMI-1640 medium containing 10% fetal bovine serum (FBS), 1× nonessential amino acid (Sigma-Aldrich), 1 mM pyruvic acid (Gibco), 2 mM L-glutamic acid (Gibco) and 10 mM HEPES (pH8.0), 10⁻⁷ M 2-mercaptoethanol (2-ME; WAKO) at 37°C with 5% CO₂ for 4 days [38, 39]. Before culture and 4 days after culture, YFP⁺ cell reduction was examined under a fluorescent microscope. Flow cytometry was used to analyze YFP⁺ cell reduction 4 days after culture. All thymus cultures regardless of controls were performed in the presence of DT.

2.5 Addition of Periostin and/or Flt3L Protein in Culture

Thymi from lethally irradiated 8–10-week-old Wnt1-Cre/+; Rosa^YFP/YFP, Wnt1-Cre/+; Rosa^YFP/DTR, and Rosa^YFP/DTR(+/+^YFP/DTR) mice were isolated and cultured as described above. Additionally, 10 ng/mL of recombinant mouse Periostin/OSF-2 (2955-F2-050; R&D) and/or recombinant mouse Flt-3 Ligand/Flt3L (427-FL-005; R&D) were added to the culture medium [38].

2.6 Thymus Transplantation Under the Renal Capsule

Adult thymi isolated from lethally irradiated (9.5 Gray) 8–10-week-old Wnt1-Cre/Rosa^YFP/DTR mice and Rosa^YFP/DTR(+/+^YFP/DTR) or C57BL/6 mice (CD45.2) were transplanted under renal capsules in an anesthetized C57BL/6 (CD45.1) mice [38]. Thymi were isolated from the renal capsules at early (11–12 days) and late (21 days) stages after transplantation and T-cell reconstitution was examined.

2.7 Flow Cytometric Analysis and Cell Sorting

Thymocytes were isolated from the thymi. For stromal cells, thymi were minced and digested with 0.1 mg/mL collagenase D (Roche), 2.4 U/mL Dispase (Gibco), and 0.1 mg/mL DNase I (Wako, Fuji film) in 10% FBS (Gibco)/HBSS (Nissui) and incubated for 2 h at 37°C. Single-cell suspensions were blocked with Rat IgG and stained with each Ab in Table S2. For lineage-negative markers, PE-conjugated Abs were used in Table S2, and for cell sorting. EasySep Mouse PE positive Selection Kit II (STEMCELL Technologies) was used to selectively remove hematopoietic cells to enrich thymic stromal cells for efficient sorting and cells were stained with each Ab in Table S2. Propidium iodide (PI; Dojindo) was used for the exclusion of PI⁺ dead cells. FACS Canto II and FACS Aria (BD Biosciences) were used as the cell analyzer and cell sorter; all data were analyzed with FlowJo software (Tree Star).

2.8 Histochemistry and Immunohistochemistry

The thymus was fixed in 4% paraformaldehyde and embedded in Super Cryo-embedding Medium (Section-Lab). Frozen sections (8–10 µm thick) were prepared with a cryostat (CM3050S; Leica) using Kawamoto's film methods [40]. For immunohistochemistry, sections were blocked with 1% bovine serum albumin (Sigma) and stained with each Ab in Table S3. Images were captured using microscopy (BZ-X700; Keyence). For estimation and counting, the number of images in each section was analyzed using Image J (version 1.49.1).

2.9 Quantitative RT-PCR

Total RNA was isolated from cells using TRIzol reagent (Invitrogen) and reverse transcription (cDNA synthesis) was performed using ReverTra Ace (Toyobo) and oligo (dT) primers (Toyobo). First-strand synthesized cDNA was used for three real-time quantitative polymerase chain reactions (RT-qPCR) using PowerUp SYBR 4 Green Master Mix (Applied Biosystems) in QuantStudio3 (Thermo Fisher Scientific). Amplification was performed using two-step PCR (40 cycles for 3 s at 95°C and 30 s at 60°C) followed by melting curve analysis. Gene-specific primers are listed in the legend of Figure S3. Relative mRNA expression levels were normalized against the level of Hprt. The experimental results were acquired from three independent experiments, in which each gene was assayed in triplicate.

2.10 Statistical Analysis

Data are expressed as means ± SD. Each experiment was repeated at least twice, with similar results being obtained each time, unless otherwise indicated. Statistical analysis was done with Microsoft Excel (2021), Prism 8 (GraphPad), and EZR (Saitama Medical Center, Jichi Medical University) [41]. Statistical significance when comparing two groups was evaluated using a two-tailed unpaired t-test. For manifold condition comparisons, one-way analysis of variance (ANOVA) was used followed by Tukey's multiple comparison test.

3.1 Thymic NC-Derived Mesenchymal Cells Are Less Susceptible to TBI

To test the hypothesis that NC-derived mesenchymal cells are involved in thymic regeneration, we first subjected Wnt1-Cre/+; Rosa^YFP/YFP (hereafter abbreviated Wnt1-Cre/+^YFP/YFP) mice to a single dose of lethal TBI (Figure 1A). Subsequently, we examined the effect of TBI on the thymus, T cells, TECs, and mesenchymal cells 3 days after irradiation. Following irradiation, the size of the thymus was reduced (Figure 1B), and the percentages and numbers of CD4⁺/CD8⁺ T cells markedly decreased (Figure 1C). Moreover, the percentages and numbers of TECs (CD45⁻/Ter119⁻/CD31⁻/EpCAM1⁺/MHC-II⁺) including CD45⁻/Ter119⁻/CD31⁻/EpCAM1⁺/MHC-II⁺/UEA-1⁻/Ly51⁺ cells and CD45⁻/Ter119⁻/CD31⁻/EpCAM1⁺/MHC-II⁺/UEA-1⁺/Ly51⁻ cells were significantly reduced with an increase in CD45⁻/Ter119⁻/CD31⁻/EpCAM1⁺/MHC-II⁺/UEA-1⁺/Ly51⁺ cells (Figure 1D). Thymic mesenchymal cells were separated into total thymic mesenchymal cells (CD45⁻/Ter119⁻/CD31⁻/ EpCAM1⁻), YFP⁺ NC-derived mesenchymal cells (hereafter abbreviated YFP⁺-mesenchymal/mesenchyme) (CD45⁻/Ter119⁻/CD31⁻/ EpCAM1⁻/YFP⁺) and YFP⁻ non-NC-derived mesenchymal cells (hereafter abbreviated YFP⁻ mesenchymal/mesenchyme) (CD45⁻/Ter119⁻/CD31⁻/EpCAM1⁻/YFP⁻). Interestingly, the total number of thymic mesenchymal cells and YFP⁻ mesenchymal cells decreased following irradiation; however, the number and percentage of YFP⁺-mesenchymal cells were not affected by irradiation (Figure 1E–G). To verify that these YFP⁺ cells were indeed mesenchymal cells, we analyzed the expression of specific mesenchymal markers. The YFP⁺ cells expressed PDGFRα, β, Sca1, and gp38, but not CD31 (Figure S1). These results suggest that YFP⁺-mesenchymal cells are less susceptible to TBI and may play a significant role in thymic regeneration post-TBI.

3.2 Depletion of Thymic NC-Derived Mesenchymal Cells Impairs Reconstitution of DN1-Like ETPs After Transplantation

Informed by the results above, we investigated the role of YFP⁺-mesenchymal cells in thymic regeneration. To this end, we employed a DT depletion system and engineered Wnt1-Cre/+^YFP/YFP mice to express DTR to generate Wnt1-Cre/+^YFP/DTR mice (Figure 2A). Using Wnt1-Cre/+^YFP/YFP as the control, we depleted YFP⁺-mesenchymal cells by administering DT into a culture medium containing thymus from the lethally irradiated Wnt1-Cre/+^YFP/YFP and Wnt1-Cre/+^YFP/DTR mice and cultured as previously described [38]. Four days after culture, a micrograph of YFP expression and the area occupied by YFP⁺ cells in Wnt1-Cre/+^YFP/DTR (Wnt1-Cre/+^Y/D) mice markedly decreased (Figure 2B,C). Furthermore, the viability of cells postirradiation, both before and after ATOC, did not undergo significant changes although there was a slight reduction in both percentage and numbers (Figure 2D).

To substantiate this finding, we proceeded to analyze the proportion of YFP⁺ cells after culture. Interestingly, although YFP⁺-mesenchymal cells showed a notable reduction following DT administration, the depletion had no apparent effect on thymic epithelial cells and YFP⁻ mesenchymal cells (Figure 2E), similar to our previous findings [38]. To ascertain whether YFP⁺-mesenchymal cells play a vital role in thymic regeneration post-TBI, we transplanted the thymus (donor) depleted of YFP⁺-mesenchymal cells under the renal capsule of C57BL/6 (host) mice (Figure 3A). We previously reported that NC depletion affected the number of ETPs 20 days after transplantation [38]. Subsequently, we evaluated reconstitution at both early (11–12 days) (Figure 3B,C) and late (21 days) (Figure 3D,E) stages of transplantation. In order to distinguish host-reconstituted hematopoietic cells and those from the donor, we utilized CD45.1 (host) and CD45.2 (donor) mice. Intriguingly, reconstitution of lineage-negative cells (B220⁻/Gr-1⁻/Mac-1⁻/Ter119⁻/CD4⁻/CD8⁻) and DN1-like ETPs, characterized phenotypically as Lin⁻/c-kit⁺/CD25⁻ fraction, was significantly reduced in the absence of YFP⁺-mesenchymal cells as evidenced in Wnt1-Cre/+^Y/D mice compared with the Rosa^YFP/DTR(+/+^YFP/DTR) (+/+^Y/D) (wild-type) mice at both early and late stages of transplantation (Figure 3B–D). However, there was no difference in the percentages of CD4⁺/CD8⁺ T cells 21 days after transplantation, although the number of thymocytes and CD4⁺/CD8⁺ T cells was reduced (Figure 3E). While the ablation of YFP⁺-mesenchymal cells resulted in impaired reconstitution in both early and late stages of transplantation, abundance of DN1-like ETPs during the early phase of transplantation was greatly affected (Figure 3B,D). These findings demonstrate that YFP⁺-mesenchymal cells are important for the abundance of ETPs for thymic reconstitution.

3.3 Thymic NC-Derived Mesenchymal Cells Induce the Upregulation of Hematopoietic Factors After Irradiation

Although YFP⁺-mesenchymal cells exhibited decreased susceptibility to TBI, the impact of TBI on their functional properties remained unknown. As irradiation was previously shown to modulate the functional properties of cells within the thymic stroma [24], we next examined the effect of TBI on specific genes pivotal to T-cell development. Consequently, Wnt1-Cre/+^YFP/YFP mice were subjected to TBI. Three days following TBI, thymi were isolated, and single cells were prepared. The sorted cells were first categorized into three distinct fractions: the CD31-negative thymic stromal cell fraction (CD45⁻/Ter119⁻/CD31⁻), thymic epithelial cell fraction (CD45⁻/Ter119⁻/CD31⁻/EpCAM1⁺), and thymic YFP⁺-mesenchymal cell fraction (CD45⁻/ Ter119⁻/CD31⁻/EpCAM1⁻/YFP⁺) for subsequent gene expression analysis (Figure S2). Following irradiation, the relative gene expression of Flt3L showed a marked increase of 3.5-fold in thymic stromal cells, while it decreased by half in the thymic epithelium, and increased by 2.5-fold in YFP⁺-mesenchyme (Figure 4A). Comparatively, Periostin expression surged by 8-fold in the thymic stromal cells, remained unchanged in the thymic epithelium, and increased by sixfold in YFP⁺-mesenchyme (Figure 4B). Correspondingly, following irradiation, the relative expression of four genes such as Flt3L, Periostin, Ccl25, and Ccl19 were upregulated in thymic stromal cells but were downregulated in the thymic epithelium (Figure 4A–D). On the other hand, the relative expression of Dll1, Il7, Cxcl12, Scf, and Bmp4 were increased in the thymic epithelium and NC-derived mesenchyme following irradiation (Figure 4E–I). These results indicate that irradiation impacts the gene expression profile within the thymic stroma, with most genes showing upregulation in YFP⁺-mesenchyme. Furthermore, a comparison of gene expression between stromal cells, TECs, and YFP⁺ NC-derived mesenchymal cells, using Hprt as a reference gene, showed that transcript levels of Flt3L and Periostin were higher in YFP⁺ NC-derived mesenchymal cells than in TECs, both before and after irradiation. Notably, the expression of Flt3L, Periostin, and Ccl25 was particularly elevated after irradiation (Figure S3).

Next, since thymic mesenchymal cells are NC and non-NC-derived, the gene expression profile of four selected genes Flt3L, Periostin, Ccl19, and Ccl25 between YFP⁺ NC-derived mesenchymal cells, YFP⁻ non-NC-derived mesenchymal cells, and TECs were examined. Even though the relative fold expression of Flt3L, Periostin, and Ccl19 were significantly increased in both YFP⁺ NC-derived mesenchymal cells and YFP⁻ non-NC-derived mesenchymal cells after irradiation, the relative fold expression of Flt3L, Periostin, and Ccl25 but not Ccl19 was significantly higher in YFP⁺ NC-derived mesenchymal cells than in YFP⁻ non-NC-derived mesenchymal cells after irradiation, although transcript levels were not compared (Figure S4). This suggests a potential compensatory role of NC-derived mesenchyme specifically in mitigating thymic epithelial damage induced by irradiation. Although most genes displayed an upregulation in YFP⁺-mesenchyme, Periostin, and Flt3L were particularly pronounced following irradiation and localization with YFP⁺ mesenchyme at the perivascular region (Figure 4A,B,K,L). Considering the significance of Flt3L and Periostin in regulating the abundance of ETPs and promoting tissue remodeling and regeneration, respectively, it is plausible that the depletion of YFP⁺-mesenchymal cells had an impact on the expression levels of these molecules. This phenomenon may have contributed to the observed impaired reconstitution following transplantation. Taken together, these findings indicate that YFP⁺-mesenchymal cells are important in producing molecules such as Flt3L and Periostin that may enhance the abundance of DN1-like ETPs for thymic reconstitution post-TBI.

3.4 Depletion of Thymic NC-Derived Mesenchymal Cells Reduces Flt3L and Periostin Expression in the Thymus

To determine whether the ablation of YFP⁺-mesenchymal cells affected the expression of Flt3L and Periostin in the thymus, we irradiated Wnt1-Cre/+^YFP/YFP (control) and Wnt1-Cre/+^YFP/DTR mice and depleted YFP⁺-mesenchymal cells in culture in the presence of DT for 4 days. Following YFP⁺-mesenchymal cell depletion, the thymus was stained with antibodies against Flt3L and Periostin. Even though some Flt3L⁺ and Periostin⁺ expression was discernible postdepletion of YFP⁺-mesenchymal cells, the area of Flt3L⁺/YFP⁺-mesenchymal cells and Periostin⁺/YFP⁺-mesenchymal cells were markedly reduced (Figure S5A,B). This result suggests that the depletion of YFP⁺-mesenchymal cells induced a downregulation of expression of Periostin and Flt3L, which may have potentially contributed to the impaired reconstitution and reduced abundance of DN1-like ETPs after transplantation.

3.5 Periostin and Flt3L Enhance Abundance of ETPs for Thymic Reconstitution After Irradiation

It is unclear whether the direct association of YFP⁺-mesenchymal cells with Periostin or Flt3L regulates the abundance of ETPs for thymic reconstitution after irradiation. The receptor for Flt3L (Flt3/Flk2) is known to be expressed on T-cell progenitors, which are a subset of ETPs considered to embody the most immature intra-thymic progenitors [8, 42]. Considering the aforementioned findings of YFP⁺-mesenchymal cells upregulating the expression of Periostin and Flt3L after irradiation, co-localizing with Periostin and Flt3L at the thymic perivascular region, and the reduced expression of Periostin and Flt3L after the depletion of YFP⁺-mesenchymal cells, we investigated whether the addition of Periostin and/or Flt3L in culture would ameliorate the effect induced by depleting YFP⁺ NC-derived cells. Consequently, thymi from irradiated wild-type (+/+^Y/D), and Wnt1-Cre/+^Y/D mice were cultured in the presence of DT, and Periostin and/or Flt3L. Four days after culture, the thymus was transplanted under the renal capsules, and reconstitution was evaluated at 11 days following transplantation (Figure 5A). The percentages and numbers of lineage-negative cells and DN1-like ETPs (Lin⁻/c-Kit⁺/CD25⁻), in the presence of DT and Periostin and Flt3L, were significantly increased compared with the control with only DT (Figure 5B). Additionally, the percentages and numbers of CD4⁺/CD8⁺ T cells were notably higher in the presence of DT and Periostin and Flt3L compared with the presence of DT alone (Figure 5C).

Furthermore, the addition of Periostin or Flt3L alone increased both the percentages and numbers of lineage-negative cells and DN1-like ETPs at 11 days after transplantation (Figure 6A–C), as well as the numbers of thymocytes and the percentages and numbers of CD4⁺/CD8⁺ T cells at 21 days posttransplantation (Figure 6D–G). Nevertheless, there appears to be no difference in the percentages of Flt3L compared with the DT-treated Wnt1-Cre/+^Y/D (DT+Wnt1-Cre/+^Y/D) mice (Figure 6F). These results indicate that both Periostin and Flt3L regulate the abundance of ETPs and CD4⁺/CD8⁺ T cells for thymic reconstitution. However, it is worth mentioning that although the addition of a single factor in the absence of NC-derived mesenchymal cells promoted the abundance of DN1-like ETPs and CD4⁺/CD8⁺ T cells, Periostin alone showed a marked increase in the numbers of DN1-like ETPs and CD4⁺/CD8⁺ T cells compared with Flt3L. Taken together, these results suggest that either Periostin or Flt3L is important for the abundance of DN1-like ETPs and CD4⁺/CD8⁺ T cells for thymic reconstitution after irradiation.

4.1 Data Limitations and Future Perspectives

An interesting finding of this study is that NC-derived mesenchymal cells are less sensitive to irradiation compared with non-NC-derived mesenchymal cells and TECs, and they produce factors that enhance thymus regeneration. While NC-derived mesenchyme secretes several factors that promote thymic regeneration after irradiation-induced damage, particularly Flt3L and Periostin, the specific mechanisms driving the expression of Flt3L and Periostin expression in these cells were not investigated in this present study. It is possible that the NC origin of these mesenchymal cells endowed them with developmental plasticity, enabling the expression of these molecules, along with others, in response to irradiation for thymus recovery. Future studies might focus on uncovering the unique mechanisms regulating the expression of Flt3L and Periostin and the function of other upregulated molecules in NC-derived mesenchyme, which are critical for thymus regeneration following irradiation-induced damage.