Human FOXN1-deficiency is associated with αβ double-negative and FoxP3+ T-cell expansions that are distinctly modulated upon thymic transplantation

Abstract

Forkhead box N1 (FOXN1) is a transcription factor crucial for thymic epithelium development and prevention of its involution. Investigation of a patient with a rare homozygous FOXN1 mutation (R255X), leading to alopecia universalis and thymus aplasia, unexpectedly revealed non-maternal circulating T-cells, and, strikingly, large numbers of aberrant double-negative αβ T-cells (CD4negCD8neg, DN) and regulatory-like T-cells. These data raise the possibility that a thymic rudiment persisted, allowing T-cell development, albeit with disturbances in positive/negative selection, as suggested by DN and FoxP3+ cell expansions. Although regulatory-like T-cell numbers normalized following HLA-mismatched thymic transplantation, the αβDN subset persisted 5 years post-transplantation. Involution of thymus allograft likely occurred 3 years post-transplantation based on sj/βTREC ratio, which estimates intrathymic precursor T-cell divisions and, consequently, thymic explant output. Nevertheless, functional immune-competence was sustained, providing new insights for the design of immunological reconstitution strategies based on thymic transplantation, with potential applications in other clinical settings.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. Peripheral T-cells prior to thymus transplantation in a patient with R255X FOXN1 mutation.

At day 166 of life there were 10.4% (660 cells/µl) CD4 T-cells, 9.7% (612 cells/µl) CD8 T-cells; and 10.7% (676 cells/µl) DNαβ T-cells in the peripheral blood. CD4+ (A), CD8+ (B), and DNαβ T-cells (C) exhibited a memory-effector activated phenotype and were able to produce significant amounts of IL-2, IFN-γ and IL-4. (D) Percentages of cycling cells, as assessed by Ki-67 expression, within CD8, CD4, DNαβ and γδ T-cells. (E) Dot-plots showing that: FoxP3 expression is restricted to CD3+ cells; that within total FoxP3+ cells (328 cells/µl) there were 67% CD4+, 17% DNαβ and 12% cells co-expressing CD4 and CD8 (DP, DP represented 3% of total αβ cells and 26% of them were FoxP3+) as illustrated by the expression of CD4 and CD8 within total αβ cells (grey) and within FoxP3+ cells (black); the concomitant expression of FoxP3 with other Treg markers (CTLA-4, CD39), or Ki-67 within total CD4+ T-cells; and the lack of cytokine production by FoxP3+ cells. (F) Vβ distribution within FoxP3+ and FoxP3− CD4 T-cells performed at day 254 of life. Graphs show the proportion of FoxP3+ CD4 T-cells and FoxP3− CD4 T-cells belonging to a given Vβ family as assessed by flow cytometry. Dot-plots show analysis after gating on the respective populations; numbers within each quadrant represent the proportion of cells expressing the respective molecules. Cytokine production was assessed after 4 hours stimulation of PBMC with PMA plus Ionomycin in the presence of Brefeldin A.

Figure 2. Immunological reconstitution and Treg recovery after thymic transplantation in a patient with R255X FOXN1 mutation.

(A) Kinetics of the frequency of naïve cells (CD45RA+CD27+) within CD4 and CD8 T-cell subsets; histogram shows CD31 expression within naïve CD4 T-cells at 58 months post-transplant, a marker associated with recent thymic emigrants. (B) Longitudinal quantification of sjTREC (left) and sj/βTREC ratio (right) quantified in total PBMC. (C) Assessment of TCR repertoire by spectratyping analysis of the CDR3 Vβ regions of purified naïve CD4 T-cells at 59 months post-transplant. (D) Representative dot-plots of the longitudinal analysis of the frequency of cells expressing FoxP3 and/or CD25 within total CD4 T-cells (upper dot-plots), and the phenotype of circulating FoxP3+ cells at 36 months post-transplant after successive gates on CD4+ and FoxP3+ T-cells (lower dot-plots); numbers inside quadrants represent the frequency of cells expressing the indicated molecules. (E) Graph shows the proportion of circulating FoxP3+ and FoxP3− CD4 T-cells belonging to each of the Vβ families at 48 months post-transplantation analysed by flow cytometry.

Figure 3. Reduced but functional CD8+ T-cell compartment after fully-mismatched HLA class I thymus transplantation.

(A) Circulating CD8+ αβ T-cells analysed 58 months post-transplantation (5% of total lymphocytes; 83cells/µl) by flow cytometry in terms of: naïve/memory/effector phenotype assessed by CD45RA and CD27 expression; levels of activation markers (CD38 and HLA-DR); IFN-γ and IL-2 production upon PMA+ionomycin stimulation; ex vivo frequency of cycling cells (Ki-67+); and ex vivo frequency of apoptotic cells assessed by annexin V staining. (B) TCR repertoire diversity evaluated by spectratype of CDR3 Vβ regions of purified CD8+ T-cells at 59 months post-transplant. (C) Graph shows the coordinate increase in the proportion of terminally differentiated effector cells defined as CD45RA+CD27neg cells within total CD8+ T-cells during acute varicella infection and its decrease in parallel with its clinical resolution.

Figure 4. Persistence of DNαβ T-cells in a patient with R255X FOXN1 mutation after thymic transplantation.

(A) Absolute counts and proportion of DN, CD8+, CD4+ cells within circulating αβ+ T-cells 59 months post-transplantation. Analysis of DNαβ T-cells revealed: (B) a relatively undifferentiated memory phenotype with increased expression of mucosal homing molecules; (C) a skewed repertoire as assessed by spectratype of CDR3 Vβ regions; (D) reduced ability to produce IFN-γ, IL-4, or IL-17 but high IL-2 production upon PMA+ionomycin stimulation; (E) no terminal-effector differentiation according to CD27, CD57 and perforin expression (histogram compares perforin levels within DNαβ, CD8 and CD4 T-cells); (F) reduced levels of the activation markers CD38 and HLA-DR and absence of Treg-associated markers despite the increased levels CD25 expression, Numbers inside dot-plots represent frequency of cells expressing the indicated molecules acquired with FACSCalibur flow cytometer.

Figure 5. Ability of expanded circulating DNαβ T-cells to respond to IL-7 and IL-2.

Representative flow cytometric analysis of freshly isolated PBMC from a patient with R255X FOXN1 mutation 59 months after thymic transplantation illustrating (A) the preserved expression of IL-7Rα and increased CD25 expression within DNαβ T-cells; and (B) the high levels of Bcl-2 expression within DNαβ in comparison with CD8 and CD4 T-cells. (C) Up-regulation of p-STAT5 upon 15 min stimulation of PBMC with IL-7 (50 ng/ml), IL-2 (100 U/mL) or IL-15 (25 ng/ml), bars represent p-STAT5 MFI within gated DNαβ, CD8 and CD4 T-cells. Freshly isolated PBMC were cultured for 5-day in the presence of IL-7 (10 ng/ml), IL-2 (10 U/mL), or IL-15 (12.5 ng/ml) or anti-CD3 plus anti-CD28 stimulation and graphs represent the fold change of CD25MFI with respect to medium (D), and the frequency of Ki-67+ cells (E), within gated DNαβ, CD8 and CD4 T-cells. (F) Representative analysis of freshly isolated PBMC illustrating the low levels cycling cells (Ki-67+) despite the increased CD25 expression within gated DNαβ T-cells. Numbers inside dot-plots represent frequency of cells expressing the indicated molecules acquired with FACSCanto (p-STAT5 and 5-day cultures) or FACSCalibur flow cytometers.