Thymic output generates a new and diverse TCR repertoire after autologous stem cell transplantation in multiple sclerosis patients

Abstract

Clinical trials have indicated that autologous hematopoietic stem cell transplantation (HSCT) can persistently suppress inflammatory disease activity in a subset of patients with severe multiple sclerosis (MS), but the mechanism has remained unclear. To understand whether the beneficial effects on the course of disease are mediated by lympho-depletive effects alone or are sustained by a regeneration of the immune repertoire, we examined the long-term immune reconstitution in patients with MS who received HSCT. After numeric recovery of leukocytes, at 2-yr follow-up there was on average a doubling of the frequency of naive CD4(+) T cells at the expense of memory T cells. Phenotypic and T cell receptor excision circle (TREC) analysis confirmed a recent thymic origin of the expanded naive T cell subset. Analysis of the T cell receptor repertoire showed the reconstitution of an overall broader clonal diversity and an extensive renewal of clonal specificities compared with pretherapy. These data are the first to demonstrate that long-term suppression of inflammatory activity in MS patients who received HSCT does not depend on persisting lymphopenia and is associated with profound qualitative immunological changes that demonstrate a de novo regeneration of the T cell compartment.

Figure 1.

Resolution of inflammatory disease activity after HSCT. Representative contrast-enhanced MRI axial scans of the brain of patient 2 at pretransplant evaluation (top) and at 2-yr follow-up (bottom) illustrate a long-lasting resolution of inflammatory disease in this patient with high pretransplant activity (arrows point at 3 out of 17 total enhancing lesions present in the brain pretherapy; no enhancing lesions were detected after therapy).

Figure 2.

Phenotypic analysis of naive and memory T cell populations. All values are the percent expression of the indicated T cell subpopulation at pretherapy and follow-up time points. (left) Subsets of CD4⁺ T cells and (right) subsets of CD8⁺ T cells, each categorized as T_naive, naive T cells; T_CM, central–memory T cells; and T_EM, effector–memory T cells. The boundaries of the boxes indicate the 25th and 75th percentiles, the lines within the boxes indicate the median, and the whiskers mark the 10th and the 90th percentiles. We observed a statistically significant (greater than twofold) increase of CD4 T_naive and a decrease of CD4 T_CM.

Figure 3.

Measures of thymic output after HSCT. (A) Phenotypic identification of RTEs in the CD4⁺ subset showed a significant increase of CD4⁺ RTEs at 2-yr follow-up. (B) TCR excision circle (TREC) analysis showed full reconstitution and increases of CD4 RTE output at 2-yr follow-up. (C) Due to extensive peripheral cell division, CD8 TREC levels only recovered up to near-pretherapy levels. Age–related interindividual variance and the small number of patients precluded a statistical evaluation of TREC data.

Figure 4.

Analysis of TCR diversity. High-resolution TCRBV CDR3 spectratyping analysis identified three basic patterns of evolution of repertoire diversity: (A) recovery of diversity from a restricted repertoire, (B) reconstitution of diversity from a normally diverse repertoire; (C) and total or partial persistence (or reemergence) of repertoire skewing. Different numbers of examples are shown to convey a correct impression of the observed frequency of each pattern. Patterns A and B were largely predominant except for patient 5, who had recurrent infections. Vertical lines are reference DNA size markers to provide the exact correspondence of CDR3 peaks across different samples. n.d., not determined (due to unavailability of 2-yr follow-up for patient 7).

Figure 5.

Extensive clonotypic renewal within the T cell repertoire. We cloned and sequenced 90 clonal TCRs from sample pairs for each of the three patterns of evolution of CDR3 length diversity previously identified by spectratyping. One representative example is shown for each pattern. Colored pie slices represent individual expanded T clones. Color and size of slices allow identification of the expanded clones and show their relative frequency, respectively. The group of nonexpanded clones (unique sequences, not shown individually for their large numbers) is shown in light gray in pretransplant samples and in white in posttransplant samples and, within each one, the proportion of single clones persisting posttransplantation is shown in dark gray. For the predominant patterns in A and B, we found a prominent increase of unique sequences, denoting increased clonal diversity also within the same CDR3 length. Pre-existing expanded clones were either undetectable or reduced after therapy. In the less frequent pattern in C, preexisting or new expanded clones constituted the majority of the examined T repertoire after therapy.

Figure 6.

Phenotypic changes of the peripheral T cell repertoire. (A) Transient increase of CD95/Fas⁺ in the CD4⁺ and CD8⁺ T cell subset in the early (peripheral) phases of immune reconstitution, suggesting an increased susceptibility to Fas/FasL-mediated apoptosis. Representative histograms from an individual patient are shown (left) as is a summary of data from all patients (right). (B) Persistent increase of CD8⁺/CD28⁻/CD57⁺ T cells during follow-up. (left) An individual patient's example of CD8⁺ gated analysis with the top histograms showing CD57 expression, and dot plots (bottom) showing CD57 (x axis) versus CD28 (y axis) expression. (right) The box plot summarizes the data from all patients. CD8⁺/CD28⁻/CD57⁺ have been characterized as a terminally differentiated subpopulation with short telomeres, defective cytolytic functions, and increased susceptibility to apoptosis.

Figure 7.

Characterization of oligoclonally expanded CD8⁺ populations. (A) In the example shown, TCR Vβ back-gating in combination with CDR3 spectratyping of CD8⁺ T cells (top inlays: black box, preHSCT; red box, 6-mo follow-up) demonstrates that oligoclonal expanded cells, representing up to 50% of all CD8⁺ cells (B), are CD57⁺/CD45RA⁺/CD45RO⁺/CD27⁻/CD95(Fas)⁺ (C), consistent with the phenotype of atypical effector–memory cells that reached replicative senescence after extensive division in vivo.