Thymus engraftment and developmental progression of adoptively transferred progenitor T cells are affected by host conditioning

Abstract

Hematopoietic stem cell transplantation as treatment for patients with leukemia relies on chemotherapy and/or irradiation-based conditioning regimens to prevent graft rejection and cancer relapse. Despite advances in radiotherapy, total body irradiation for myeloablative conditioning is associated with dose and temporal heterogeneity, and how these variables impact thymus reconstitution and production of new T cells has not been fully elucidated. Here, we used in vitro–generated mouse progenitor T (proT) cells to investigate the dynamics of donor-derived cells in host thymuses under different irradiation regimens. Our results showed that proT cells are capable of homing to the thymus of both irradiated and nonirradiated mice. Notably, different irradiation protocols resulted in varying numbers of proT cells homing to the host thymus, revealing a critical time window for enhanced adoptive transfers of syngeneic and allogeneic donor cells. ProT cells given to mice, exposed to an optimal irradiation condition, showed enhanced thymus settling, slower developmental kinetics, and increased self-renewal capacity, when compared to suboptimal regimens. Our study provides valuable insights into the thymus-homing and cellular development mechanisms of transferred proT cells, as well as informing future therapeutic applications of in vitro–generated proT cells.

Introduction

Hematopoietic stem cell transplantation (HSCT) can be a life-saving therapeutic intervention for some patients with leukemia.^1,2 Allogeneic HSCT is a curative therapy for many hematologic cancers (eg acute myeloid leukemia and acute lymphoblastic leukemia), but its therapeutic effect is mediated by both high-dose chemo-radiotherapy and by the induction of the graft-versus-leukemia (GvL) effect.³ Thus, this therapy must be preceded by irradiation-based conditioning regimens given to prevent graft rejection and leukemia relapse by eradicating host T cells and reducing tumor burden.⁴ The conditioning regimens are classified by dose intensity as myeloablative conditioning, reduced intensity conditioning, or non-myeloablative conditioning.⁵ Total body irradiation (TBI) was the main modality used in myeloablative conditioning, which provided both immune suppression and leukemia eradication.⁶ The standard TBI-based conditioning included the combination of cyclophosphamide (120 mg/kg) and fractionated TBI (12 Gy), and higher irradiation dose failed to improve survival due to increased rates of organ toxicity and graft-versus-host disease (GvHD).^7,8

Nonmyeloablative and reduced-intensity conditioning regimens were also developed using a single fraction of 2 Gy TBI.^6,9 The 5-year survival of this regimen was 33% among 274 patients with acute myeloid leukemia, relying mainly on GvL.¹⁰ Overall, increased irradiation dose intensity was associated with reduced relapse risk and better leukemia control, but higher potential for GvHD and nonrelapse mortality.⁶ Nevertheless, TBI is associated with the problem of dose heterogeneity. Thus, higher precision techniques called total marrow irradiation and total marrow and lymphoid irradiation, which focus the dose to the target region while sparing the rest of the body, were developed and subsequently improved.^4,11,12 Despite the progress in radiotherapies in the past decades, the timing of irradiation prior to transplantation and its effects on the thymus-homing capability and development of transplanted cells have not been investigated.

Thymic-seeding progenitors (TSPs) are generated in the bone marrow (BM)^13,14 and enter the thymus through blood vessels at the corticomedullary junction.^15,16 After extravasation, TSPs enter a perivascular space,^17,18 becoming early thymus progenitors (ETPs) or DN1a cells, and then migrate through the epithelial basal membrane into the thymus and interact with the thymic epithelium and other cells that constitute the thymic microenvironment.¹⁹ This thymus-homing process involves multiple interactions, including VCAM-1 (vascular cell adhesion molecule-1), ICAM-1 (intercellular adhesion molecule-1), P-selectin, and chemokines.²⁰ The sequential migration of developing T cells through distinct thymic microenvironments is regulated by the ordered expression of chemokine receptors.^21,22

The homing of TSPs cells to the thymus has been mainly studied via bone marrow transplantation (BMT). After intravenous injection of BM cells in irradiated mice, a small number of donor cells could be detected in the recipient thymi within 3 to 4 hours.^17,23 The thymus-homing process of BM-derived TSPs was not continuous but intermittent in normal adult mice, exhibiting age-dependent receptivity.²⁴ Based on these observations, a thymus-BM feedback loop hypothesis was proposed: a wave of developing ETPs interacts with thymic stromal cells, which sequentially induce the release of cytokines that regulate the production, mobilization, and recruitment of TSPs from the BM.²⁵ Overall, the thymus homing of BM-derived TSPs is a gated process limited by intrathymic niches.^26,27,28

In vitro–derived proT cells are early T-lineage progenitors generated using the OP9-DL4 co-culture system.^29,30 The adoptive transfer of proT cells markedly enhanced T-cell reconstitution with a significant GvL effect but without GvHD.³¹ Preclinical studies showed that coadministration of BMT and in vitro–generated proT cells boosted thymic reconstitution in irradiated mice compared to BMT alone.^32,33 These advances suggest a promising therapeutic prospect for in vitro–generated proT cells in the future. Nevertheless, the thymus-homing characteristics of these cells and their response to host conditioning by irradiation are poorly understood.

In the present study, we generated proT cells in vitro from DsRed donor mice and implemented a series of irradiation protocols. Subsequently, we examined the dynamics of these cells in host thymus under both irradiation and nonirradiation conditions. We also assessed variations in their thymus-homing capabilities, and monitored their progression in cell numbers and development, in both syngeneic and allogeneic host mice. Our results unveiled several intriguing findings, offering valuable insights into the mechanisms underlying thymus-homing and cellular development of in vitro–generated proT cells.

Materials and methods

Mice

DsRed (B6.Cg-Tg(CAG-DsRed*MST)1Nagy/J) (stock No. 006051) mice and CB6F1/J (stock No. 100007) were purchased from The Jackson Laboratory. Young (7 to 8 weeks) cohorts of mice were used. All mice were maintained and bred at Sunnybrook Health Sciences Centre, and all animal procedures were approved by the Sunnybrook Health Sciences Centre Animal Care Committee.

Irradiation protocols

Mice (7 to 8 weeks of age) were sublethally irradiated at 5.5 Gy. Different irradiation regimens were designed as follows: 5d, irradiation was performed 5 days prior to injection of proT cells; 4d, irradiation was performed 4 days prior to injection; 3d, irradiation was performed 3 days prior to injection; 2d, irradiation was performed 2 days prior to injection; 1d, irradiation 1 day prior to injection; and 0d, mice were irradiated on the same day as the injection of proT cells. Lethal irradiation (10.5 Gy) was performed on the day of injection, and indicated as lethal 0d.

In vitro generation of proT cells and adoptive transfer

Co-cultures of lineage-negative (Lin⁻) Sca-1⁺Kit⁺ (LSK) and OP9-DL4 cells were established as previously described.³⁴ In brief, mouse BM-derived LSK cells were collected from DsRed mice by dissecting and crushing the leg bones using sterile utensils in Hanks’ balanced salt solution (HBSS), which were then filtered through a 40-μm filter to obtain a single-cell suspension. BM CD117⁺ (Kit⁺) cells were enriched using anti-CD117-MicroBeads and LS column (Miltenyi) according to manufacturer’s instructions. The CD117-enriched cells were labeled with FITC-conjugated antibodies against Lin markers (anti-B220 [RA3–6B2], anti-CD19 [1D3], anti-CD11b [M1/70], anti-Gr-1 [8C5], anti-NK1.1 [PK136], anti-CD3 [2C11], anti-CD8a [53.6-7], and anti-CD4 [GK1.5]), as well as with anti-CD117-APC (ACK2) and anti-Sca1-PE (D7) (BioLegend). The labeled cells were then sorted using cell sorter FACSAria Fusion (BD Biosciences). The collected LSK cells were cultured with OP9-DL4-7FS cells expressing the Notch ligand DLL4 as well as human cytokines IL-7, FLT3-L, and SCF. In each culture dish (15 cm²) with OP9-DL4-7FS cells (∼90% confluency), 70,000 to 100,000 LSK cells were seeded. The co-culture was maintained in α-minimum essential medium Eagle (αMEM) supplemented with 10% FBS and 1% penicillin/streptomycin (Gibco) in the presence of 1 ng/mL IL-7 (Miltenyi Biotec) and 5 ng/mL Flt-3L (Miltenyi Biotec). On day 10 after seeding, the co-culture was harvested and filtered through 40-μm cell strainers (Thermo Fisher). The harvested cells were labeled with anti-CD25-APC (PC61, BioLegend) and incubated in anti-APC-MicroBeads (Miltenyi Biotec), and subsequently enriched for CD25⁺ cells using LS column (Miltenyi Biotec) according to manufacturer’s instructions. CD25⁺ cells were resuspended in 200 μL of serum-free αMEM for intravenous injections. Recipients were either wild-type CD45.2 mice, or CB6F1/J mice. One injection contained 5 × 10⁶ CD25-enriched proT cells (∼99% purity).

Flow cytometry

Mice were sacrificed at different time points after injection of in vitro–generated proT cells. Thymuses were dissected and mashed, followed by filtering through 40-μm cell strainers in HBSS supplemented with 1% bovine serum albumin and 2 mM EDTA. A single-cell suspension was prepared from each thymus. Donor cells were identified using the fluorescence protein DsRed expression. Thymocytes in DN2, DN3, DN4, and DP stages were labeled with fluorescently conjugated antibodies CD4-APC, CD8-FITC, CD44-PE-Cy7, CD25-APC-Cy7, and CD90.2-BV711 (BioLegend) before flow cytometry was performed on FACSymphony (A5-SE) analyzer (BD Biosciences). Splenocytes were labeled with the following antibodies (BioLegend): CD19-PE-Cy7, CD11b-APC-Cy7, NK1.1-BV421, CD11c-APC, CD90.2-BV711, TCRβ-FITC, and TCRγ-FITC. Dead cells were excluded by DAPI.

Comparative analyses of donor and donor-derived cells in host thymuses

The dynamics in the numbers of donor-derived cells were analyzed using flow cytometry at different time points (days 1, 4, 7, 8, 9, 10, 11, 12, and 14 posttransfer) under both irradiation and nonirradiation conditions. To measure the thymus settling by proT cells at an early-stage posttransfer, the number of donor cells per thymus was measured using flow cytometry 24 hours posttransfer under different irradiation protocols. Additionally, the numbers of donor-derived cells per thymus were measured using flow cytometry from day 1 to day 12 posttransfer under the 0d (optimal) and 3d (suboptimal) irradiation protocols. In parallel, thymus cellularity, represented as the number of total live thymocytes per thymus, was calculated from day 0 to day 7 after irradiation. Moreover, the numbers of donor-derived cells at different developmental stages, CD4/CD8 (double negative [DN]) DN2, DN3, DN4, and double positive (DP), were measured at days 4, 5, 6, and 7 posttransfer, and the percentages of each cell type were calculated. At least 3 replicates were performed for each treatment.

5-Bromo-2′-deoxyuridine labeling

DNA synthesis of donor-derived cells was examined by intraperitoneal injection of 100 μL of 10 mg/mL 5-bromo-2′-deoxyuridine (BrdU) (BD Biosciences) in Dulbecco’s PBS. Three hours after BrdU injection, recipient mice were sacrificed for flow cytometry analysis. Thymocytes were labeled with fluorescently conjugated antibodies—BrdU-APC (BD Biosciences), TCRβ-PECy7 (BioLegend), CD4-BV421 (BioLegend), CD8-FITC (BioLegend), CD44-BB700 (BD Biosciences), CD25-APC-Cy7 (BioLegend), and CD90.2-BV711 (BioLegend)—before flow cytometry was performed on a FACSymphony (A5-SE) analyzer (BD Biosciences).

Data analysis and statistics

Flow cytometry data were analyzed using FlowJo version 10.10.0 software (TreeStar). Statistical analyses of thymocyte numbers and difference in means of donor-derived cell numbers were performed using 2-way ANOVA followed by Tukey’s multiple comparison test. Comparison of sample groups at specific time points were performed using Welch’s t–test. All data are represented as mean ± SEM in error bars. P values are indicated as *P < 0.05 and **P < 0.01.

Results

In vitro–generated proT (CD25⁺) cells show similar thymus engrafting capacity as ex vivo CD25⁺ DN thymocytes

First, we examined the homing capability of in vitro–generated proT cells by comparing it to ex vivo thymocytes isolated from adult DsRed-transgenic mice. As illustrated in the experimental schematic (Fig. 1A), LSK cells were sorted from the BM of DsRed mice and co-cultured with OP9-DL4-7FS cells to induce their differentiation into T-lineage cells.³⁴ After 10 days, the in vitro–generated CD25⁺ proT cells were enriched from the co-cultures, while ex vivo CD25⁺ CD3⁻CD4⁻CD8⁻ (DN) thymocytes were sorted from DsRed mice. These proT cells were then individually adoptively transferred into sublethally irradiated host mice, and their homing capabilities examined at 40 hours posttransfer by flow cytometry. Ex vivo– and in vitro–generated CD25⁺ cells showed a similar number of thymus-homing cells in the host thymuses (Fig. 1B), revealing that in vitro–generated proT cells possess the same engrafting capacity as ex vivo thymocytes, which supports their use as a research tool to study in vivo engraftment and growth of proT cells.

Figure 1.

Isolation and functional comparison of CD25⁺ progenitor T cells from in vitro or ex vivo sources. (A) Co-culture of BM-derived LSK cells (gate shown) on OP9-DL4-7FS cells for 10 days prior to isolation and injection. In vitro–generated T-lineage cells were MACS-enriched for CD25⁺ cells. Ex vivo thymocytes from adult mice were MACS-enriched for CD4⁻CD8⁻ cells and then sorted by flow cytometry (FACS) for CD25⁺ cells. Isolated CD25⁺ cells were then adoptively transferred by intravenous injection into sublethally irradiated (5.5 Gy) C57BL/6 host mice. Flow cytometry analysis of Lin⁻Kit⁺ (MACS-enriched) BM cells (prior to culture), in vitro–derived cells, and thymocytes are shown as indicated. (B) Comparison of the thymus engrafting capability of intravenously transferred DsRed⁺CD25⁺ progenitor T cells (ex vivo or in vitro derived) into host mice (n = 7–8; pooled from to 2 independent experiments). ns, not significant.

Comparison of thymus engraftment by in vitro–generated proT cells into irradiated or nonirradiated host mice

We next compared the homing capabilities of in vitro–generated proT cells under both irradiation and nonirradiation conditions over a period of 2 weeks after adoptive transfer, as this represents the time period for the differentiation of TSPs in the thymus.³³ The results showed that the number of donor-derived cells increased geometrically in irradiated and nonirradiated host mice during the first 9 days (Fig. 2A). While the number of donor-derived cells continued to increase in irradiated host mice until day 12, we observed a dramatic decline in nonirradiated host mice. Notably, donor-derived cells were nearly undetectable in nonirradiated mice after 2 weeks, coinciding with the expected duration of T-cell development and exit from the thymus. These results suggest that although in vitro–generated proT cells are capable of thymus homing under nonirradiation conditions, irradiation conditioning can enhance their recruitment to the thymus and extend their expansion potential in the host thymus (Fig. 2A and B). We examined for the appearance of proT cell–derived CD4⁺CD8⁺ (DP) cells in the thymus of host mice, and noted an initial increase in DP cell numbers in both irradiated and nonirradiated mice until day 9 (Fig. 2C). While DP cell numbers continued to increase in irradiated mice, a drop was seen in nonirradiated mice, implying that the formation of new DP cells was limited in the absence of irradiation.

Figure 2.

Presence of adoptively transferred in vitro–generated proT cells in the thymus and spleen of host mice. Comparison of thymus (A–B) and spleen (D–E) engraftment by donor DsRed⁺ in vitro–generated proT cells in irradiated (5.5 Gy) or nonirradiated host mice. Flow cytometric analysis of donor DsRed⁺ cells in the thymus (B) and spleen (E) of host mice, as indicated. (C) Number of donor DsRed⁺ CD4⁺CD8⁺ (DP) cells in the thymus of irradiated and nonirradiated mice. Analysis shown are pooled results from 7 independent experiments (n = 3–6), *P < 0.05, **P < 0.01.

We also examined the spleen for the presence of proT-derived donor cells to ascertain their potential for giving rise to alternative lineage fates when settling outside the thymus. Flow cytometric analysis at day 7 posttransplant revealed that proT cells differentiated into myeloid lineage (CD11b⁺ and/or CD11c⁺), and non-T lymphoid (T cell receptor–negative [TCR⁻]) lineage (NK1.1⁺ and/or CD90⁺) cells (∼50% and ∼50%, respectively) (Fig. S1A and B). Consistent with the previous characterization of Notch signaling in the generation of T-lineage cells,³⁵ we did not observe any B-lineage potential among the donor cells. Similarly, we also noted a small number (∼13%) of γδ T (TCRγδ⁺) cells in the spleen of host mice, which are likely derived from CD25⁺ DN3 cells present in the proT cell graft that have already rearranged their γδTCR chains (Fig. S1C) and can complete their γδ-lineage differentiation in the absence of continued Notch signaling, as previously reported.³⁶ A temporal kinetics analysis of proT-derived cells in the spleen of host mice showed a progressive reduction of donor cells in nonirradiated mice, becoming nearly absent by day 12, while irradiated host mice showed a persistent number of donor cells (Fig. 2D and E). These results convey the notion that proT cells are only able to undergo further αβ T-cell differentiation and cellular expansion upon thymic entry,³⁷ which is not supported in the splenic microenvironment.

Different irradiation regimens affect thymus entry, revealing a critical time window for the adoptive transfer of proT cells

The above results demonstrated that irradiation enhances the thymus engraftment by in vitro–generated proT cells; however, it remained unclear as to whether different time intervals between irradiation and adoptive cell transfer would have a further beneficial outcome. To this end, we analyzed for the presence of donor cells in the host thymus under different irradiation regimens (Fig. 3A), in which mice were exposed to sublethal irradiation (5.5 Gy) on the day of injection (0d) or 1 to 5 days prior to adoptive cell transfer (1d to 5d). Our data showed that transferred cells achieved the best performance under the 0d irradiation regimen, while 1d to 5d irradiation regimens showed a 4- to 8-fold significant reduction in the number of donor cells present in the host thymus, as compared to 0d (Fig. 3B). These results indicate that there is an apparent narrow time window after sublethal irradiation (<24 hours) for adoptively transferred proT cells to effectively enter the thymus.

Figure 3.

Definition and comparison of different irradiation regimens. (A) Irradiation regimens indicated as 5d, 4d, 3d, 2d, 1d, and 0d represent irradiation (5.5 Gy) administered on days 5 through 0 prior to the day of adoptive cell transfer injections, respectively. The red outline indicates the day of injection. (B) Comparison of the number of transferred proT cells (DsRed⁺) present in the thymus of mice under different irradiation regimens. Data are collected at 24 hours posttransfer. Of note: 5d, 4d, 3d, 2d, and 1d are not significantly different from each other; and 0d is significantly different (P < 0.001) from all other regimens.

Comparison of irradiation protocols in the recovery of thymus cellularity

Since the 0d irradiation regimen provided the best performance of transferred cells, we designated it as the benchmark, and considered all other regimens similarly suboptimal. We then compared the temporal kinetics in the number of donor-derived cells under 2 distinct irradiation protocols, 0d versus 3d (ie benchmark vs suboptimal) over a 12-day period. As noted above, a significant difference in the number of donor-derived cells is observed after 1 day, between the 2 irradiation protocols; however, this difference was not seen for days 2 to 4 after transplant (Fig. 4). In contrast, by days 5 to 7 posttransfer, a significant difference in the number of donor-derived cells is apparent, which then leads to a dramatic difference by day 12, with 0d showing a ∼10-fold higher number of donor-derived cells than the 3d regimen. These results indicate that the timing of irradiation exposure prior to the time of transplant appears to have a major impact on how well donor-derived cells expand after transfer.

Figure 4.

Dynamics in the number of donor-derived proT cells per thymus under different irradiation regimens. The number of donor-derived proT cells (DsRed⁺) present in the thymus of host mice receiving 0d or 3d irradiation regimens obtained at different time points after adoptive cell transfer, as indicated. Analysis shown are pooled results from 8 independent experiments (n = 3–5), *P < 0.05, **P < 0.01.

We considered the possibility that changes in host thymus cellularity as a consequence of irradiation would affect the expansion of immigrating proT cells. To this end, thymus cellularity was measured in nontransplanted mice from days 1 to 7 postirradiation. The results demonstrated a sharp reduction in the number of live cells, decreasing by ∼98%, at day 1 postirradiation (Fig. S2). The number of total live cells in the host thymus remained at a very low level for the next 5 days, reaching a nadir at day 4 postirradiation, with a cellularity rebound starting from day 6, consistent with previous reports.^38,39 These data help to explain the different temporal kinetics seen in the number of donor-derived cells under 0d and 3d irradiation protocols (Fig. 4), likely due to the recovery by endogenous (DN2-DN4) thymocytes, starting from day 6, that may impact on the expansion potential of 3d transferred proT cells, due to competition for access to niche and/or resources.

Faster developmental progression and decreased self-renewal of transferred proT cells in 3d versus 0d irradiation protocol

The number of donor-derived cells described above represents their total cellularity in the host thymus; we next interrogated whether their developmental progression was affected by the different irradiation regimens. To this end, we analyzed for the presence of donor cells at different stages of thymocyte differentiation (DN2-DN4 and DP). The results showed that, over a period of 7 days posttransfer, frequencies of donor cells at early stages (DN2-DN3) were greater in the host thymus of mice that received 0d as compared to 3d irradiation regimens (Fig. 5A and B). Additionally, the percentage of donor cells at later stages (DN4-DP) were lower in the host thymus of mice that received 0d as compared to 3d irradiation regimens (Fig. 5C and D). A representative flow cytometry analysis of thymocyte subsets at day 6 posttransfer is shown in Fig. 5E, which clearly shows the apparent acceleration in thymocyte differentiation by donor cells (lower %DN3 and higher %DP) in mice receiving the 3d irradiation regimen.

Figure 5.

Comparison of developmental kinetics by proT cells in the thymus of host mice after adoptive transfer under different irradiation protocols. (A–D) Percentages of different T-lineage developmental stages over a period of 7 days under 0d and 3d sublethal-irradiation protocols, as indicated. Analysis of different developmental stages were pre-gated on DsRed⁺ donor cells. (E) Phenotype of donor-derived cells in the thymus of representative host mice, on day 6 after transfer. Analysis shown are pooled results from 8 independent experiments (n = 3–6), *P < 0.05, **P < 0.01.

To better understand the effect that host cells may have on the differentiation of donor-derived proT cells, we next analyzed the recovery of radioresistant endogenous thymocytes postirradiation.⁴⁰ Notably, the thymus of mice receiving the 3d regimen showed a higher percentage of competing DN2 and DN3 cells, as compared to 0d mice, at 2 days after transplant (Fig. 6A and B). We also noted higher percentages of DP cells in the mice receiving the 3d regimen than 0d mice 4 days after transplant, as the endogenous cells would have reached this stage of development by this time point, corresponding to 7 days postirradiation, while the 0d mice needed 3 more days to reach this frequency of DP cells (Fig. 6C). Consistent with this endogenous recovery, 3d mice showed much lower percentages of CD4 and CD8 single-positive (SP) cells, as compared to 0d mice, at 4 days after transplant (Fig. 6D). These temporal kinetics follow the expected progression of DPs differentiating into SPs, which then exit thymus.

Figure 6.

Comparison of the recovery by radioresistant ETP in the thymuses of host mice after adoptive transfer under different irradiation protocols. (A–D) Percentages of different T-lineage developmental stages over a period of 7 days under 0d and 3d sublethal-irradiation protocols, as indicated. Analysis of different developmental stages were pre-gated on DsRed⁻ host cells. Analysis shown are pooled results from 8 independent experiments (n = 3–6), *P < 0.05, **P < 0.01.

We considered the possibility that the lower cell yields and accelerated differentiation by donor cells observed in mice receiving a 3d irradiation regimen could be attributable to effects on DN2 and DN3 subsets, which are known to be self-renewing cells. To investigate the role of irradiation regimens on the self-renewal potential of transferred donor cells, host mice were given a 3-hour pulse of BrdU on day 6 after transplant to label cells entering the S-phase of the cell cycle, which would identify DN3 cells undergoing self-renewal prior to undergoing V-D-J recombination and TCRβ selection-induced proliferation (Fig. 7A and B). Our results show that BrdU⁺ intracellular TCRβ⁻ DN3a donor cells are more abundant among donor DN3a cells in the thymus of 0d than 3d mice (Fig. 7C), highlighting their increased self-renewal capacity. Additionally, the proportion of BrdU⁺ DN3b and DPs among donor DN3b and DP cells, respectively, were comparable in 3d and 0d mice (Fig. 7D and E), indicating that later differentiation stages are not similarly affected by the irradiation regimens.

Figure 7.

Comparison of self-renewal and differentiation capacity of donor proT-derived cells after adoptive transfer. Schematic of BrdU staining and intracellular (i.c.) TCRβ expression of donor thymocytes from mice receiving 0d (A) and 3d (B) irradiation regimens using flow cytometry. The dot plot panels (left side) depict DN stages, pre-gated on the DsRed⁺ CD90⁺CD4⁻CD8⁻ population, of donor cells from thymus of host mice. The dot plot panels (right side) depict DN3 cells (red gate) analyzed for the BrdU incorporation and TCRβ rearrangement. DN3a undergoing self-renewal are shaded green, and DN3b cells undergoing proliferation after β-selection are shaded blue. (C–E) Percentage of BrdU⁺ cells among DN3a, DN3b, and DP cells on day 6 after adoptive transfer. (F) Number of total donor cells in the thymuses of mice that were lethally irradiated, or sublethally irradiated with 0d or 3d regimens. Thymuses from each group of mice were harvested and analyzed for number of donor-derived cells on day 1 and day 7 after transfer. (G and H) Percentages of DN3 and DP from donor cells engrafting the mice receiving 0d and 3d sublethal-irradiation regimen on day 7 after adoptive transfer. (I) Number of donor-derived DsRed⁺ DP, CD4, and CD8 cells in the thymuses of mice that were sublethally irradiated with 0d or 3d regimens. (J) Number of donor-derived cells in the thymuses of CB6F1 mice 1 and 12 days after receiving haploidentical parent-to-F1 transfer of proT cells. (K) Number of donor-derived DP, CD4, and CD8 cells in the thymuses of CB6F1 mice that received haploidentical transfer. Analysis shown are pooled results from 4 independent experiments (n = 3–5), *P < 0.05, **P < 0.01.

Next, we compared thymus-homing and differentiation of donor proT cells between 3 different conditions: sublethal 3d, 0d, and lethal 0d. Compared to 0d, the initial thymus homing and expansion were lower in 3d and lethal 0d regimens on day 1 and day 7 after transfer, respectively (Fig. 7F), likely due to competition for niches during thymus recovery (3d) and stromal cell damage (lethal 0d). While the lethal irradiation protocol would lead to less niche competition by endogenous thymocytes compared with 0d and 3d, which would result in a greater percentage of DN3 cells (Fig. 7G) and a lower percentage of DP cells (Fig. 7H), at day 7 posttransfer the total donor cell numbers were lower than 0d (Fig. 7F), likely reflecting stromal cell damage and increased systemic inflammation. Additionally, at a later time point (day 12), mice receiving the 0d regimen showed more donor-derived DP, CD4 SP, and CD8 SP cells compared to mice receiving the 3d regimen (Fig. 7I).

Although donor proT cells, which are DN2/DN3 cells prior to full αβ-TCR rearrangements, were previously shown to be host-tolerant and do not mediate GvHD,³¹ we examined whether their engraftment would be affected in the setting of allogeneic adoptive cell transfer. We injected in vitro–generated proT cells from DsRed donor (B6) mice into haploidentical recipient (CB6F1/J) mice and measured their presence on day 1 and day 12 posttransplant. Notably, a higher number of donor cells engrafted the thymus of mice receiving the 0d regimen when compared to that of 3d mice on day 1, and this advantage was also seen after 12 days (Fig. 7J). Specifically, higher numbers of donor-derived DN3 and DP cells were present in the thymus of mice receiving the 0d regimen than 3d mice (Fig. 7K).

Taken together, our findings point to the sublethal 0d regimen as the optimal protocol for the engraftment by in vitro–derived and consequently ex vivo proT cells.

Discussion

The clinical application of HSCT requires the use of conditioning regimens to ensure successful engraftment of donor cells by the host; among the different types of conditioning approaches, TBI remains as an important intervention option prior to HSCT. Here, we investigated the timing of irradiation before transplantation and tested whether it would impact thymus entry and developmental progression of donor progenitor T cells, which have not been previously addressed. Our study demonstrated that ex vivo– and in vitro–generated proT cells have equal thymus-engrafting capacity. We then focused our attention on in vitro–generated proT cells, as these have a potential therapeutic benefit, and showed that they can readily settle in the thymus under nonirradiation conditions, but sublethal irradiation led to an enhanced number of donor proT cells entering the host thymus. Notably, nonirradiated mice showed limited and accelerated developmental outcomes of donor proT cells as compared to sublethally irradiated mice. Irradiation appears to extend the duration of thymus reconstitution and improve the expansion of proT-derived thymocytes in the host mice. Our findings are consistent with the “intermittent” thymus-homing hypothesis,²⁴ whereby the nonirradiated thymus offers limited space for the incoming cells, whereas irradiation helps to vacate the niches for the adoptively transferred proT cells in the thymus of host mice.

The dose (intensity) and mode (target tissue vs TBI) of irradiation regimens, as essential components for HSCT, have been extensively investigated,^1,2,4,11,12 but the importance of the temporal interval between irradiation and adoptive transfer of proT cells has not been considered. To address this, we designed a series of irradiation protocols to examine whether the timing of when irradiation is administered affects the thymus receptivity for proT cells. We found that there is only a narrow 1-day time window for transferred proT cells to efficiently home to the thymus. This observation can be explained by the thymus-gating and the thymus-BM feedback loop hypotheses.^26,27,28 For instance, in the 0d irradiation regimen, proT cells are transferred on the same day as irradiation, when thymus niches are vacated (gate opening), with the function of thymus stromal cells (thymus-bone marrow feedback loop) remaining intact and facilitating robust thymus homing within a short period of time. This benefit appears to be diminished when using a lethal irradiation regimen, wherein thymus stromal cells are likely damaged, leading to reduced thymus receptivity as compared to the sublethal 0d regimen. In contrast, for the 1d and subsequent 2d to 5d regimens, although the gates may be opened, the combination of endogenous radio-resistant ETPs undergoing recovery²⁸ and eosinophils rushing in⁴¹ inevitably hinders the thymus-homing process of adoptively transferred proT cells.

Using the 0d irradiation regimen as the benchmark and 3d as the representative suboptimal regimen, we compared the temporal kinetics in the number of donor-derived cells engrafting syngeneic or allogeneic host mice under either protocol. We observed that in the mice receiving the 3d regimen, the initial number of transferred proT cells engrafting the thymus was lower than the 0d regimen, and this difference remained apparent at later time points. It is important to note that under the 3d irradiation regimen, the thymus has had 3 additional days to recover as compared to 0d. One obvious explanation for the initial thymus seeding difference could simply be due to the 70- to 100-fold reduction in thymic cellularity seen after day 1 to 3 post–sublethal irradiation, that is, making it a smaller sized organ for cells to migrate to. This explains the deficiency in thymus homing of proT cells transferred to mice treated with 1d, 2d, 3d, 4d, and 5d regimens. Additionally, a substantial increase in donor cell numbers was observed for the 0d regimen at days 5 to 7 posttransfer, leading to a significant and large difference between these 2 protocols by day 12 posttransfer, justifying a potential long-term benefit of the 0d regimen for both syngeneic and allogeneic hosts. This finding underscores the importance of the irradiation regimen used prior to the adoptive transfer of proT cells, and points to its therapeutic significance for future applications.

Previous studies have shown that the thymus is highly sensitive to radiation-induced damage, leading to a rapid reduction in thymocyte cellularity, primarily DP cells.⁴⁰ In addition to thymocytes, thymus epithelial cells (TECs) are also affected by irradiation, and both thymocytes and TECs undergo homeostatic recovery after thymic damage.⁴² By analyzing the dynamics in the number of total live thymocytes in response to irradiation, we found that thymocyte cellularity remained at very low levels for 5 consecutive days, followed by a significant rebound on the sixth day, which is due to radio-resistant ETPs undergoing expansion and differentiation.⁴⁰ This recovery trend in thymocyte numbers is consistent with the dynamics of donor-derived cell expansion and differentiation that occurs within the first few days under either irradiation regimen. Importantly, in mice under the 0d regimen, donor-derived proT cells would have more vacant niches available within the thymus microenvironment during the first 4 days post–adoptive transfer, which would support the enhanced expansion of donor cells.

Our findings point to differences in the differentiation kinetics of proT-derived donor cells that enter the thymus under 0d vs 3d irradiation regimens. Our results revealed accelerated developmental progression in the 3d irradiation protocol, as seen by the earlier appearance of DN4 and DP cells when compared to the 0d regimen, during the first 7 days posttransfer. Interestingly, the frequency of DN2 and DN3 cells remained higher under the 0d protocol, suggesting an enhanced self-renewal capacity, perhaps due to more access to supporting niches. In contrast, donor-derived cells entering the thymus of mice under the 3d regimen would be in effect competing for niches with the radioresistant host ETPs undergoing expansion and differentiation, diminishing their ability to self-renew and accelerating their developmental progression. In keeping with this notion, we noted that the expansion of DN3 cells was even more pronounced in the thymus of lethally irradiated mice. This implies that in the absence of internal competition by endogenous thymocytes undergoing recovery, the expansion and self-renewal of donor-derived DN3 is favored. This is consistent with the enhanced self-renewal function of endogenous DN2-DN3 cells in the donor thymus after transplant into mice lacking lymphocyte progenitors.^43,44,45

Cortical (c)TECs provide a supportive environment for the proliferation and differentiation of early progenitors, while medullary TECs play a key role in the later stages of thymocyte development, particularly in ensuring self-tolerance and preventing autoimmunity.⁴⁶ TECs are crucial in maintaining the structural integrity and functional capacity of the thymus, ensuring a continuous supply of mature, functional T cells.⁴⁷ Our data suggest that cTEC function is temporarily affected by irradiation but recovers rapidly, which has been shown to be mediated by several mechanisms.^42,48,49 In a previous BMT study, the impact of irradiation on molecules involved in thymus homing was examined using CCR7/CCR9 double knockout and selectin ligand PSGL-1^–/– BM cells.⁵⁰ This study demonstrated that PSGL-1 plays a more significant role than CCR7/CCR9 during the early stage (3 weeks) after irradiation for BMT, with CCR7/CCR9 being critical at later stages after thymic cellularity has rebounded. However, our study did not examine the molecular players involved in thymus homing; rather, the focus was on the capacity of in vitro–generated proT cells to home and engraft a host thymus. Our analysis of their early developmental progression within the thymus of differently conditioned mice should motivate additional work examining the potential interactions between the transferred proT cells and the thymic stromal environment, as part of a 0d irradiation regimen.

Considering that cTECs provide a supportive environment for the proliferation and differentiation of early progenitors,⁴⁷ we inferred that cTECs retain their functional integrity after sublethal irradiation. Nevertheless, we noted the developmental progression of proT-derived donor cells was accelerated under the 3d protocol, which could be due to changes in their self-renewal or proliferative capacity.⁵¹ Our results showed that BrdU incorporation by DN3a cells was diminished in the thymus of mice under the 3d regimen, implying that their self-renewal ability is limited when in competition with endogenous DN3 cells, which would be present concurrently in the thymus of mice under the 3d regimen.⁴⁰ On the contrary, later stages of differentiation, including DN3b and DP cells, are less affected by niche competition; hence, their proliferation in the thymus of mice under 3d regimen were similar to 0d. Nevertheless, the detailed underlying mechanisms require further investigation. Additionally, the significance of accelerated developmental progression of transferred proT cells and its correlation with thymus aging warrant further exploration.

In summary, our study showed that in vitro–generated proT cells can home to the thymus and undergo accelerated developmental progression under nonirradiation conditions, while irradiation enhances thymus settling by creating vacant niches. We identified a narrow 24-hour window for transferred cells to home to the thymus, with an optimal (0d) irradiation protocol resulting in higher overall donor cell engraftment compared to suboptimal (3d) regimen in both syngeneic and allogeneic hosts. Surprisingly, the suboptimal 3d protocol led to accelerated developmental progression, while the 0d protocol led to enhanced self-renewal. These insights into how the radiation-induced temporal changes in thymus function directly affect early stages of T-cell development help to shed light on the interactions between transferred cells and the thymic microenvironment, emphasizing the critical importance of the temporal interval between irradiation and proT cell transfer. Further investigation is needed to elucidate mechanisms driving accelerated cell development and its implications for thymus function.

Author contributions

Y.R.L. and J.C.Z.-P. conceived and designed the study. Y.R.L. and C.R.L. performed experiments. Y.R.L. analyzed the results. K.C. performed cell sorting. M.M., W.-L.L., and J.C.Z.-P. assisted with study design and data interpretation. Y.R.L. wrote the manuscript, with help and input from other authors. J.C.Z.-P. acquired funds to support the study.

Yue Ru Li (Conceptualization [Lead], Data curation [Lead], Formal analysis [Lead], Investigation [Lead], Methodology [Lead], Project administration [Lead], Validation [Lead], Writing—original draft [Lead], Writing—review & editing [Lead]), Christina R. Lee (Investigation [Supporting], Methodology [Supporting]), Mahmood Mohtashami (Conceptualization [Supporting], Investigation [Supporting], Validation [Supporting]), Kevin Conway (Methodology [Supporting]), Wan-Lin Lo (Writing—review & editing [Supporting]), and Juan Carlos Zúñiga-Pflücker (Conceptualization [Lead], Funding acquisition [Lead], Project administration [Lead], Supervision [Lead], Writing—original draft [Supporting], Writing—review & editing [Equal])

Supplementary material

Supplementary material is available at The Journal of Immunology online.

Funding

This work was supported by grants from the Canadian Institutes of Health Research (FDN154332 and PJT192050 to J.C.Z.-P.) and the Stem Cell Network (to J.C.Z.-P.).

Conflicts of interest

J.C.Z.-P. is Chief Scientific Officer and holds stock options in ProTgen Inc. The other authors have no conflicts of interest.

Data availability

No unique large data sets, reagents, or model organisms were generated. The data underlying this article will be shared on reasonable request to the corresponding author.