The haematopoietic phosphatase PTPN22 is a key regulator in balancing immune responses between self-reactivity and tolerance. PTPN22 downregulates T cell signaling and harbors the non-HLA genetic variation most strongly associated with autoimmune disease in humans, the single nucleotide polymorphism R620W. The effect of this mutation is currently controversial due to confounding results in mouse and human models. The polymorphism is linked to increased susceptibility to autoimmunity in both human and mouse models, although the latter does depend on genetic background. However, mouse data clearly shows that the polymorphism has a loss-of-function effect on T cell signalling, whereas studies in human models largely demonstrate a gain-of-function effect for R620W. A confounding issue in human studies is that they depend on comparison of T cells from distinct individuals, on protein over-expression, or on RNA interference, techniques for which it is difficult to control for genetic and environmental variables, changes in stoichiometry, and off-target effects or incomplete knockdown, respectively. We aimed to create isogenic human cell lines with mutations in PTPN22 at the genomic level to alleviate the complications inherent in analysing human data. In addition to autoimmune pathogenesis, we are interested in the role of PTPN22 in a cancer setting. Because PTPN22 has a strong suppressive effect on T cell responses to weak affinity antigen, which encompass most tumour antigens, we postulated that knocking out PTPN22 may better enable T cells to kill tumour cells. Furthermore, we have shown that PTPN22 knockout (KO) leads to increased IL-2 expression in mouse T cells, and that this effect is protective against TGF-β mediated suppression, a common driver of T cell inhibition in the tumour microenvironment. T cell transfer experiments in mice showed that PTPN22 KO T cells are indeed more effective at reducing tumour size. Based on these findings, we aim to determine whether PTPN22 KO in human cells confers a similar effect on signaling. To investigate the effects of PTPN22 KO on human T cell signaling, we used CRISPR gene-editing to target PTPN22 in a Jurkat cell line. By combining this technique with lentiviral transduction of a specific T cell receptor, we generated human cell lines which are genetically identical, save for specific alterations to PTPN22, and which can be stimulated with strong or weak cognate antigen. We found that PTPN22 KO Jurkat cells develop an enhanced activation phenotype upon stimulation, including increased IL-2 expression. Additionally, PTPN22 KO Jurkat cells show enhanced Erk signalling following stimulation with weak affinity antigen, but this difference is lost as stimulus strength increases. CRISPR technology has presented the opportunity to create novel models of PTPN22 signalling in the context of human T cell lines. The data from these lines suggests that, unlike the R620W mutation, complete loss of PTPN22 has a comparable effect in human and mouse T cells. In conjunction with our previous findings, these results suggest that knocking out PTPN22 may lead to signalling alterations that improve adoptive T cell cancer therapy.

This study consists of two pieces of work investigating haematological malignancies; Acute Lymphoblastic Leukaemia (ALL) and Diffuse Large B Cell Lymphoma (DLBCL). Firstly, Pre-B ALL represents the most common paediatric malignancy and despite increasingly improved outcomes for patients, ~ 20% of all patients diagnosed with ALL relapse. Activating mutations in the RAS pathway are common (~50%) and result in hyperactivation of the MAPK pathway. I identified Erk negative feedback control via DUSP6 to be crucial for NRASG12D-mediated pre-B cell transformation and investigated its potential as a therapeutic target. I showed that a small molecule inhibitor of DUSP6 (BCI) selectively induced cell death in patient-derived pre-B ALL cells; with a higher sensitivity observed in relapse pre-B ALL. I also discovered that a high level of Erk activity is required for proliferation of normal pre-B cells, but dispensable in leukemic pre-B ALL cells. In addition, I found that human B cell malignancies can be grouped into three categories that fundamentally differ in their ability to control Erk signalling strength. Secondly, DLBCL is the most common haematological malignancy and although potentially curable with chemotherapy, 40% of patients still succumb from their disease. Recent exome sequencing studies have identified hundreds of genetic alterations but, for most, their contribution to disease, or their importance as therapeutic targets, remains uncertain. I optimised a novel approach to screen the functional importance of these mutations. This was achieved by reconstituting non-malignant, primary, human germinal centre B cells (GC B cells) with combinations of wildtype and mutant genes to recapitulate the genetic events of DLBCL. When injected into immunodeficient mice, these oncogene-transduced GC B cells gave rise to tumours that closely resemble human DLBCL, reinforcing the biological relevance of this system. To screen potential tumour suppressor mutations in this system in a high throughput fashion, I developed a lymphoma-focused CRISPR library of 692 genes recurrently altered in B cell lymphomas. These experiments identified GNA13 as an unexpectedly potent tumour suppressor in human GC B cells and provided new understanding to its mechanism of action. These findings provide novel understanding of the complexity of oncogenic mechanisms in human B cell malignancies.

Der RAF-MEK-ERK-Signalweg steuert grundlegende, oftmals entgegengesetzte zelluläre Prozesse wie die Proliferation und Apoptose von Zellen. Die Dauer des vermittelten Signals wurde als entscheidener Faktor für die Steuerung dieser Prozesse identifiziert. Es ist jedoch nicht eindeutig geklärt, wie die verschiedenen früh und spät reagierenden Genexpressionsmodule kurze und lange Signale unterscheiden können und durch welche kinetischen Merkmale ihre Antwortzeit bestimmt wird. In der vorliegenden Arbeit wurden sowohl Proteinphosphorylierungsdaten als auch Genexpressionsdaten aus HEK293-Zellen gewonnen, die ein induzierbares Konstrukt des Proto-Onkogens RAF tragen. Hierbei wurde ein neues Genexpressionsmodul identifiziert, dass sich aus sofort induzierten aber spät antwortenden Genen zusammensetzt. Es unterscheidet sich in der Genexpressionsdynamik und Genfunktion von anderen Modulen, und wurde mit Hilfe mathematischer Modellierung experimenteller Daten identifiziert. Es wurde festgestellt, dass diese Gene aufgrund von langen Halbwertszeiten der vermitteltenden mRNA in der Lage sind spät auf das eingehende Signal zu reagieren und die Dauer des Signals in die Amplitude der Genantwort zu übersetzen. Trotz der langsamen Akkumulation und damit späten Antwortzeit, konnte aufgrund einer GC-reichen Promoterstruktur zunächst vermutet und mit Hilfe eines Markerverfahrens bestätigt werden, dass die Transkription dieser Gene instantan mit Beginn der ERK-Aktivierung startet. Eine vergleichende Analyse zeigte, dass das Prinzip der Signaldauer-Entschlüsselung in PC12-Zellen und MCF7-Zellen, zwei paradigmatischen Zellsystemen für die ERK-Signaldauer, konserviert ist. Insgesamt deuten die Ergebnisse der Untersuchung darauf hin, dass das neu identifizierte Genexpressionsmodul der Entschlüsselung der ERK-Signaldauer dient und das mRNA Halbwertszeiten sowohl hierfür, als auch für die zeitliche Abfolge der Genantwort eine entscheidende Rolle spielen. / The RAF-MEK-ERK signalling pathway controls fundamental, often opposing cellular processes such as proliferation and apoptosis. Signal duration has been identified to play a decisive role in these cell fate decisions. However, it remains unclear how the different early and late responding gene expression modules can discriminate short and long signals and what features govern their timing. Both protein phosphorylation and gene expression time course data was obtained from HEK293 cells carrying an inducible construct of the proto-oncogene RAF. A new gene expression module of immediate-late genes (ILGs) distinct in gene expression dynamics and function was identified by mathematical modelling. It was found that mRNA longevity enables these ILGs to respond late and thus translate ERK signal duration into response amplitude. Despite their late response, their GC-rich promoter structure suggested and metabolic labelling with 4SU confirmed that transcription of ILGs is induced immediately. A comparative analysis showed that the principle of duration decoding is conserved in PC12 cells and MCF7 cells, two paradigm cell systems for ERK signal duration. Altogether, the findings of this study indicate that ILGs decode ERK signal duration and that both decoding capacity and gene expression timing are governed by mRNA half-life.