What Have Sequencing Approaches Taught Us About the Biology of Allogeneic Transplantation?

The increasing availability and decreasing cost of sequencing technologies and other high-dimensional analysis methods have facilitated a proliferation of high-resolution data in the field of hematopoietic stem cell transplantation (HCT). Bulk and single-cell sequencing approaches, solidifying as a standard approach in immune analysis in both pre-clinical and clinical research settings, have served as a catalyst for a wealth of new information and exciting advancements.

Dr. Gustavo de Almeida and colleagues studied T-cell populations in allogeneic HCT recipients using blood and skin samples collected after transplantation.¹  They used single-cell RNA-seq (scRNA-seq) technology to confirm the long-term (up to 2 years) persistence of host tissue-resident memory T cells (TRMs) in the skin, as well as establish their recirculating potential by identifying very small numbers of these host cells in peripheral blood. The low proliferative potential of these cells in the skin (as measured by Ki-67) suggests that they are likely to be persistent from the pre-HCT era, rather than products of a small number of host TRMs that have repopulated the skin after transplantation. The scRNA-seq approach allowed both identification of donor/host origin and assessment of functional capacity in the same specific individual cells, thus offering information not available using transcriptomics of bulk populations or flow cytometry.

Transcriptomics can also be combined with sequencing of the T-cell receptor (TCR), offering information about both T-cell specificity and functional capacity for the same unique individual cell. Dr. Ulrike Gerdemann and colleagues sought to combine these techniques in a Rhesus Macaque (RM) model of allogeneic HCT in order to understand the tissue specificity of post-transplant T cells as well as the breadth of their functional potential.²  Their work describes this novel method combining single-cell transcriptomics and TCR sequencing (RM-scTCR-Seq) and highlights that T cells purified directly ex vivo from RM liver and spleen during active graft-versus-host disease (GVHD) overlapped with highly alloreactive clones identified in vitro. These alloreactive T cells bore proinflammatory transcriptomic signatures, consistent with expectations for activated, proinflammatory alloreactive T cells. Cell populations within deep tissues are very difficult to study in the clinical setting given that sampling requires invasive procedures (e.g., endoscopy and bronchoscopy, liver biopsies) that are only performed when necessary for patient care; thus, animal models such as these provide valuable insights into the tissue-specific immune environment driving GVHD after allo-HCT.

Dr. Juliane Lohmeyer and colleagues also paired TCR and transcriptomic approaches to gain insight into post-transplant T-cell function, this time using a mouse model of GVHD after allo-HCT.³  This group specifically sought to understand the influence of regulatory T cells (Tregs) on the conventional T cell (Tcon) compartment, given that cellular therapy with Tregs remains a promising clinical strategy for the control of post-HCT alloreactivity, with marked improvement in GVHD survival in their mouse model. Mice receiving a Treg infusion in addition to their donor graft (containing bone marrow and purified CD4+ and CD8+ Tcon) were compared with control animals who received GVHD-inducing Tcon alone. The authors hypothesized that the presence of infused Tregs would constrain the clonal expansion of alloreactive donor T cells, thus explaining (at least in part) the therapeutic benefit of Treg cellular therapy. Contrary to this hypothesis though, the TCR repertoires were not altered in the Treg-treated mice; instead, the transcriptional profiles of Tcon were markedly different. Tregs induced the downregulation of proinflammatory genes such as IL18rap (encoding the IL-18 receptor accessory protein) and Th1 signature genes including Tbx21, IL12rb1, and IL12rb2 (encoding T-bet and two different subunits of the IL-12 receptor) in the Tcon compartment. The use of parallel TCR and RNA-seq technologies allows us to answer for the first time questions about the relative importance of fixed genetic features (i.e., TCRs) and flexible functional features (i.e., the transcriptome, which in turn drives protein expression).

In another important multi-omic study from 2022, led by Dr. Laetitia Dubouchet and colleagues, the authors used cutting-edge strategies to define key determinants of tolerance after transplantation.⁴  They combined multiparameter phenotyping with mass cytometry, transcriptomics, and metabolomic analysis of post-transplant peripheral blood samples collected from patients who were between one and two years post-HCT. Patients were grouped by clinical course: 1) GVHD-free and off immune suppression (IS) — the primary tolerant; 2) experienced some degree of GVHD but IS could be stopped without recurrence — the secondary tolerant; and 3) developed acute or chronic GVHD and remained dependent on IS — the nontolerant. This natural experiment allowed the authors to interrogate the differences between immune populations in the three patient categories. First, the mass cytometry profiles in which 38 phenotypic and functional markers were measured were clustered using FlowSOM, a dimensionality-reduction technique that allows the expression of all markers on all cells to be visualized in a way that is not typically possible with traditional gating strategies. This clustering revealed, for example, that primary tolerant patients had more naïve B cells and more OX-40⁺ Tregs than expected based on analysis of the donor sample. Transcriptomic pathway analysis revealed signatures uniquely associated with the primary tolerant state, though these were a mixture of apparently regulatory and stimulatory pathways, reflecting the complexity of post-transplant immune behavior, and to some extent, the challenges of interpreting bulk RNA-seq on peripheral blood samples. The additional comparison of nontolerant patients with their donors also revealed numerous signatures linked with this adverse outcome, such as the protein-level expression of CD25, CD38, and granzyme B — all of which are activation markers — as well as an increased transcriptional abundance of pathways linked with the activation of T and B cells. This rich dataset and analysis has highlighted a number of new pathways of interest and is likely to continue to be valuable to the field for years to come.

The use of novel methods, particularly those that integrate the measurements of immunologic features at the protein, metabolic, transcriptional, and genetic (e.g., TCR) level, is key for advancing our understanding of post-transplant complications and identifying pathways that should be explored as targets for mechanistic studies, and ultimately, for pharmacologic intervention. Furthermore, numerous interesting studies using sequencing and phenotyping technologies were also presented at the ASH annual meeting in December 2022, highlighting the fact that we continue to live in interesting times with respect to insights gained from high-resolution analysis methods.