Interactions between the Microbiota and the Host Immune System in Hematopoietic Stem Cell Transplantation

Abstract

The commensal human microbiota and the human immune system form an inseparable unit. Reciprocal effects ensure their homeostasis. The microbiota is crucial for immune system maturation and contributes directly and indirectly to innate and adaptive immune defense mechanisms. In turn, the immune system tolerates and selectively spares commensal microbial colonizers, and shapes the community structure. Immune-microbial interactions are crucial in the healthy host. Upon acute or chronic disease, especially when involving infections and inflammation, changes in bacterial diversity and in the residential microbial community structure have become evident. In a number of conditions including cancer, the state of the microbiota is associated with clinical outcomes and the patients’ recovery from disease might to a certain extent depend on it. This attracts attention to the microbiota as a promising resource for future clinical predictive and prognostic markers, and even as a target for therapeutic modulation. However, the underlying mechanisms that play a role in homeostatic and disturbed immune-microbial interactions in the human host are not completely understood. A concept that is of interest for studying the role of the microbiota constitutes hematopoietic stem cell transplantation (HSCT). It involves the treatment of patients with e.g. hematologic malignancies, such as leukemia, through the infusion of a donor’s (allogeneic) or their own (autologous) stem cells. Preceding HSCT, patients undergo preparative immunosuppressive and chemotherapeutic regimens, and antimicrobial prophylaxis. These treatments introduce immuno-compromisation and changes in the commensal microbiota. The aim of this thesis was to characterize interactions of the microbiota and the human immune system in the context of HSCT. For cohorts of allogeneic and autologous HSCT patients, we implemented longitudinal study designs spanning the time from before, at the time of, and after transplantation. By using 16S rRNA gene profiling, we assessed temporal dynamics of the gut, oral, and nasal microbiota. We determined changes in bacterial alpha diversity, proportional changes on taxonomic family level, and temporal abundance trajectories of specific phylogenetically closely related sequence variants. In multivariate multi-table analyses, we examined whether changes in the microbial community structure are associated with immune cell counts, immune markers and clinical outcomes after transplantation. We accounted for patient characteristics, such as age and sex, underlying disease, transplantation modalities, such as stem cell source, and treatment regimens. Moreover, we assessed in a machine learning approach whether microbial abundances prior to transplantation can be used to predict the development of graftversus-host disease (GvHD) after allogeneic HSCT. In the first study, we investigated host-microbial associations in children undergoing allogeneic HSCT. We focused on dynamics in the gut microbiota around the time of transplantation and during the first month post-HSCT. We found that high abundances of the Lactobacillaceae family were associated with the occurrence of moderate to severe acute GvHD and high mortality. These concerned patients also had high plasma levels of the antimicrobial peptide human beta-defensin 2 (hBD2), which we assessed for the first time in the HSCT context. High abundances of the families Ruminococcaceae and Lachnospiraceae were associated with rapid reconstitution of NK and B cells. Moreover, treatment with specific antibiotics were associated with high Enterobacteriaceae abundances. In the second study, we were interested in studying long-term microbial abundance dynamics in relation to pediatric allogeneic HSCT and therefore monitored the microbiota over a period of one year. We also examined the microbiota at additional mucosal sites, namely the oral and nasal cavities. With the aid of phylogenetic trees, we identified taxa that discriminated the microbial community structure between time points. These taxa included for instance Enterococcus spp. and Blautia spp. in the gut. While the temporal trajectory of Enterococcus spp. showed an expansion after transplantation, Blautia spp. decreased in abundance. Similarly, abundances of Actinomycetaceae in the oral cavity and Corynebacteriaceae in the nasal cavity decreased post-transplant. Common to all was the regain of pretransplant like abundance levels after approximately three months. Furthermore, pre-transplant abundance of specific taxa at all three body sites, e.g. intestinal Parabacteroides distasonis, in this cohort showed potential to predict subsequent acute GvHD development. We contextualized these findings by integrating microbial abundances with the above named host parameters in multivariate analyses. This revealed similarities in host-microbial associations at the gut, oral, and nasal sites. The third study comprised investigations within a cohort of adult autologous HSCT recipients. Previous research including our first two studies focused primarily on allogeneic HSCT. Here, we revealed associations of gut, oral, and nasal microbial abundances with host immune parameters in autologous HSCT patients. For instance, the abundances of oral Prevotella spp. correlated with counts of T lymphocyte subpopulations. Since the findings in this study are based on a rather small group of patients, further research is needed to validate our observations. This thesis extends the knowledge about interactions of the microbiota and the host immune system in hematopoietic stem cell transplantation. We extend the knowledge about commensal microbial shifts after HSCT by characterizing not only patterns of disturbance early after transplantation, but also the reconstitution of the microbiota at late follow-up time points. The ‘three months’ time point might provide a crucial temporal threshold for evaluating whether microbial homeostasis has been regained. Microbial associations with specific immune markers, such as hBD2, were assessed, which are novel within the HSCT context. We suggest further targeted research to evaluate the clinical relevance of these markers. The correlation between abundances of specific commensal taxa and adaptive and innate immune cell reconstitution after HSCT constitutes an interesting novel finding that is in line with the role of the microbiota in immune system maturation in the health host. This emphasizes the importance of microbial homeostasis in HSCT patients and warrant careful considerations with regards to antimicrobial treatment in these patients. We revealed host-microbial associations shared across body sites, e.g. certain immune cell counts that correlate with microbial abundances of gut, oral, and nasal taxa. Since the feasibility of collecting oral and nasal swab samples is higher compared with fecal sample collection, these findings might open opportunities for future clinical routine screening of the microbiota, and thereby to take advantage of its possible value in guiding personalized treatment strategies. We showed that acute GvHD post-transplantation might be predicted from microbial abundances pre-HSCT in our cohort, and propose that the predictive potential of specific taxa is worth validating in further cohorts. Specific microbial markers might in the future aid with identifying patients at risk of adverse outcomes and allow for early preventive interventions. Our initial findings in a small cohort of autologous HSCT patients highlight the relevance of investigating the role of the microbiota in this patient group, which is also of importance as a reference for allo-HSCT cohorts concerning donor-related effects.