preisverleihung 2018 stiftung professor dr. max cloëtta heft nr. 46 «long-term single-cell quantification: new tools for old questions» prof. dr. timm schroeder prof. dr. johanna joyce «exploring and therapeutically exploiting the tumor microenvironment»

stiftung professor dr. max cloëtta fünfundvierzigste preisverleihung 9. november 2018 lausanne heft nr. 46 der schriftenreihe stiftung professor dr. max cloëtta schaffhauserstrasse 43, postfach, 8042 zürich telefon 044 350 44 35 telefax 044 350 44 32 e-mail cloetta@stiftung.ch www.cloetta-stiftung.ch

inhalt vorwort des präsidenten des stiftungsrates und der geschäftsführerin ......................................................................... 5 urkunde für prof. dr. timm schroeder ............................................. 11 kurzbiografie von prof. dr. timm schroeder .................................... 13 festvortrag von prof. dr. timm schroeder ........................................ 18 long-term single-cell quantification: new tools for old questions urkunde für prof. dr. johanna joyce ................................................. 45 kurzbiografie von prof. dr. johanna joyce ....................................... 47 festvortrag von prof. dr. johanna joyce ........................................... 52 exploring and therapeutically exploiting the tumor microenvironment informationen über die stiftung prof. dr. max cloëtta ..................... 90 übersicht über die bisher erschienenen publikationen der schriftenreihe stiftung professor dr. max cloëtta ............................ 92 ehrentafel der preisträger .................................................................. 103 3

vorwort prof. dr. fritjof helmchen so wie sterne in einer galaxie erst entstehen, sich dann umwandeln und schliesslich sterben, so durchlaufen zellen in unserem körper einen le- bensweg von ihrer geburt bis zu ihrem tod beziehungsweise ihrer tei- lung. das schicksal der zellen wird dabei sowohl von den molekularen prozessen und signalkaskaden innerhalb der zelle gelenkt als auch durch den austausch und die wechselwirkungen mit den umgebenden zellen. innere wie äussere faktoren bestimmen zum beispiel wie aus einer stammzelle spezialisierte zellen hervorgehen, die sich normal entwickeln und zur gesunden funktion eines gewebes beitragen, aber auch wie sich krebszellen heranbilden, im körper ausbreiten, festsetzen und zerstöre- risch verhalten. diejenigen prozesse im detail aufzuklären, die der fort- schreitenden entwicklung von zellen und ihren schicksalsentscheidun- gen zugrunde liegen, methoden zu entwickeln, die diese prozesse messbar, sichtbar und manipulierbar machen, und schliesslich die neu gewonnenen erkenntnissen in innovative therapieansätze zur behand- lung von krankheiten umzusetzen, dies sind grundpfeiler der forschung der beiden diesjährigen cloëtta-preisträger. prof. dr. timm schroeder konzentriert sich dabei in seiner forschung besonders auf die enorme «innere» komplexität der molekularen netz- werke, die die entwicklung von stammzellen steuern. durch bahnbre- chende methodische entwicklungen, die es ermöglichen, mit filmaufnah- men einzelne fluoreszenzmarkierte zellen bei ihrer teilung und der erzeugung von nachkommenschaft über lange zeiträume zu verfolgen, gelang es ihm die aktivität von entscheidenden signal- und kontrollmo- lekülen zu messen und zu analysieren. dadurch konnten wichtige neue erkenntnisse über verschiedene arten von stammzellen gewonnen wer- den. die dabei gesammelten grossen datenmengen erfordern anspruchs- volle methoden der bildverarbeitung und bioinformatik. mit prof. dr. johanna joyce wird als zweite preisträgerin eine heraus- ragende krebsforscherin ausgezeichnet, die sich insbesondere für die «äussere» komplexität der wechselwirkungen von krebszellen mit ihrer 5

brigitt küttel geschäftsführerin stiftungsrat die gemeinnützige hertie-stiftung mit sitz in berlin fokussiert ihre för- dertätigkeit auf zwei gebiete: «gehirn erforschen» und «demokratie stär- ken». eine äusserst interessante spannweite – mit einem persönlichen bezug zur stiftung prof. dr. max cloëtta. seit 2006 vergibt die hertie-stiftung rund alle zwei jahre eine senior-for- schungsprofessur, mit der sie das lebenswerk verdienter wissenschafter ehrt und diesen ermöglicht, auch über die ordentliche pensionierung hi- naus weiter tätig zu sein. als bislang einziger forscher ausserhalb deutschlands erhielt prof. ad- riano fontana im jahr 2009 eine senior-forschungsprofessur dieser ver- dienten stiftung zugesprochen, rund ein jahr nach seiner wahl in unse- ren stiftungsrat. er gehöre, so die hertie-stiftung, als internationaler experte im bereich der neuro- und infektionsimmunologie zu den 100 weltweit am häufigsten zitierten immunologen, der unter anderem 1999 mit dem deutschen aids-forschungspreis und 1997 dem hoechst marion roussel-multiple-sklerose-forschungspreis ausgezeichnet wurde. die stiftung prof. dr. max cloëtta ehrte ihn bereits 1993 mit ihrem preis. die hertie-professur von professor fontana endete im juni 2017, und ein halbes jahr später, auf ende 2017, gab er das präsidium unserer stiftung ab. wir freuen uns, dass er uns aber als mitglied des wissenschaftlichen ausschusses sein immenses wissen auch weiterhin zur verfügung stellt, und danken ihm sehr herzlich für seinen grossen, immer sorgfältigen und wertschätzenden einsatz für die stiftung. mit professor fritjof helmchen, professor für neurowissenschaften und co-direktor des instituts für hirnforschung an der universität zürich, hat auf den 1. januar 2018 ein weiterer hervorragender forscher und ehe- maliger cloëtta-preisträger (2015) das präsidium übernommen. wir freuen uns sehr auf die weitere zusammenarbeit. 7

fung auf eine ordentliche professur zu überbrücken. die stipendien wer- den alle zwei jahre vergeben, das nächste mal wieder 2020. 2018 finanzierte die stiftung prof. dr. max cloëtta folgende forschende an schweizer universitäten mit fünfjähriger unterstützungsperiode: dr. mathias hauri-hohl, 1975, universitäts-kinderspital zürich, ab- teilung stammzellentransplantation. projekt: «improving t-cell recons- titution and enhancing central tolerance mechanism in hematopoietic stem cell transplantation». unterstützungsdauer: 1.1.2016–31.5.2021 (sistierung 1.4.2018–31.8.2018) dr. alexandre theocharides, 1975, universitätsspital zürich, klinik für hämatologie. projekt: «the role of cell-extrinsic factors in hema- topoietic stem cell malignancies». unterstützungsdauer: 1.6.2015– 31.5.2020 dr. wei lynn wong, 1976, universität zürich, institut für experimen- telle immunologie. projekt: «the role of iaps and ripks in hemato- poiesis and disease, specifically in tumor formation and metastasis». unterstützungsdauer: 1.1.2016–31.12.2020 mit dreieinhalbjähriger unterstützungsdauer: dr. grégory verdeil, 1976, universität lausanne, abteilung für funda- mentale onkologie und ludwig cancer centre. projekt: «finding and characterizing new targets to overcome t cell exhaustion for immuno- therapy of cancer». unterstützungsdauer: 1.8.2017–31.1.2021 dr. britta maurer, 1976, universitätsspital zürich, klinik für rheu- matologie und zentrum für experimentelle rheumatologie. projekt: «early diagnosis in disease monitoring of systemic autoimmune disor- ders with molecular targeted imaging». unterstützungsdauer: 1.4.2018– 30.9.2021 klinische medizin plus seit 2010 vergibt die stiftung prof. dr. max cloëtta in zusammenarbeit mit der uniscientia stiftung, vaduz, stipendien «klinische medizin plus». medizinerinnen und medizinern werden während oder unmittel- 9

the cloëtta prize 2018 is awarded to professor timm schroeder born in 1970 in pretoria, south africa department of biosystems science and engineering eth zurich, basel for his ground-breaking contributions to research and technology development in the field of molecular control of mammalian stem cell fate at the interface of biology, medicine and informatics lausanne, 9th november 2018 in the name of the foundation board: the president the vice president a member 11

curriculum vitae name: languages: position: timm schroeder born 3. september 1970 in pretoria, south africa german, married man: german, english, french, japanese, latin machine: vb, vb.net, vba, lua, c++ professor of cell systems dynamics, department of biosystems science and engineering, eth zürich, mattenstr. 26, 4058 basel, switzerland timm.schroeder@bsse.ethz.ch career 2013– 2011–13 2004–11 2002–04 2000–02 full professor, cell systems dynamics department of biosystems science and engineering (d-bsse) swiss federal institute of technology (eth) zürich, basel, switzerland 2015–17/2017–: deputy-/head of department d-bsse director, research unit stem cell dynamics, helmholtz zentrum münchen – german research center for environmental health, neuherberg, germany principal investigator (tenured 2009), research group «hematopoiesis», institute of stem cell research, helmholtz zentrum münchen 2006–13: deputy director, institute of stem cell research visiting researcher, group for stem cell biology, riken center for developmental biology, kobe, japan postdoc, gsf, munich, germany 2001 and 2002: research visits in kyoto (groups tasuku honjo and shinichi nishikawa) and harvard (group dan tenen) universities, japan and usa 13

selected publications 1. coutu dl*, kokkaliaris kd, kunz l and schroeder t* (2018). multicolor quantita- tive confocal imaging cytometry. nature methods 15, 39–46. 2. coutu dl, kokkaliaris kd, kunz l and schroeder t (2017). three-dimensional map of nonhematopoietic bone and bone-marrow cells and molecules. nature biotechnology 35, 1202–1210. 3. hoppe ps, schwarzfischer m, loeffler d, kokkaliaris kd, hilsenbeck o, moritz n, endele m, filipczyk a, gambardella a, ahmed n, etzrodt m, coutu dl, rieger ma, marr c, strasser mk, schauberger b, burtscher i, ermakova o, bürger a, lickert h, ner- lov c, theis fj and schroeder t (2016). early myeloid lineage choice is not initiated by random pu.1 to gata1 protein ratios. nature 535, 299–302. 4. hilsenbeck o, schwarzfischer m, skylaki s, schauberger b, hoppe ps, loeffler d, kokkaliaris kd, hastreiter s, skylaki e, filipczyk a, strasser m, buggenthin f, feigel- man js, krumsiek j, van den berg ajj, endele m, etzrodt m, marr c, theis fj* and schroeder t* (2016). software tools for single-cell tracking and quantification of cellular and molecular properties. nature biotechnology 34, 703–706. 5. filipczyk a, marr c, hastreiter s, feigelman j, schwarzfischer m, hoppe ps, loeff- ler d, kokkaliaris kd, endele m, schauberger b, hilsenbeck o, skylaki s, hasenauer j, anastassiadis k, theis fj* and schroeder t* (2015). network plasticity of pluripotency transcription factors in embryonic stem cells. nature cell biology 17, 1235–1246. 6. filipczyk a, gkatzis k, fu j, hoppe ps, lickert h, anastassiadis k* and schroeder t* (2013). biallelic expression of nanog protein in mouse embryonic stem cells. cell stem cell 13, 12–13. 7. eilken hm, nishikawa s-i and schroeder t (2009). continuous single-cell imaging of blood generation from haemogenic endothelium. nature 457, 896–900. 8. rieger ma, hoppe ps, smejkal bm, eitelhuber ac and schroeder t (2009). hema- topoietic cytokines can instruct lineage choice. science 325, 217–218. 9. schroeder t (2008). imaging stem-cell-driven regeneration in mammals. nature 453, 345–351. 10. schroeder t (2005). tracking hematopoiesis at the single cell level. annals of the new york academy of sciences 1044, 201–209. 15

research projects. however, as discussed for some examples below, sur- prisingly many obvious questions remain without satisfying and accepted answers despite decades of intensive research. this is certainly the case in hematopoietic stem cell biology, the classical mammalian stem cell system. many of the concepts, terms and questions of stem cell research have been defined in this system since the middle of last century, but re- main unresolved controversies. during my doctoral thesis, where i worked on the possible effects of notch activation on hematopoietic progenitor cell fates, i realized that many basic conceptual questions in the field re- main disputed. i felt that the lack of adequate technologies for quantify- ing the dynamics of cellular and molecular behaviors is one important reason for this lack of satisfying answers. our blood system produces millions of cells every second of our life. as in other cell systems, the number and type of cells produced must be tightly regulated, and also adapted over time to changing needs. failure to produce the right number of the right cells at the right time and loca- tion can quickly lead to deadly diseases like anemia or leukemia. it seems obvious that the first step in analyzing the molecular control of the un- derlying cell fate choices (fig. 1) must be to know what cells actually do in health and disease, and how cell fate choices change upon molecular mutation and manipulation. figure 1: cell fate options of multipotent stem and progenitor cells. these are chosen in close reciprocal dynamic interaction with the microenvironment of individual cells. 17

as shown in this example, snapshot data allows very different assump- tions about the involved cell fate choices which are all compatible with the measured data. for example, half of the cells might have died, or none of the cells could have died. obviously, the assumed reasons e.g. for nor- mal or pathological tissue (re)generation, possible interventions in dis- ease, and the molecular machineries picked to the analyze for many years would fundamentally differ depending on whether one picks the first or second (of many possible additional) interpretation. the same not only holds true for quantifying cell fates, but also for ana- lyzing the dynamics of the molecules involved in their control. depending on how frequent and long the e.g. expression, activation or subcellular lo- cation of molecules of interest is analyzed, one will come to very different conclusions about possible dynamics like oscillations, step functions or transient peaks. however, this knowledge is crucial to understand how the molecular machineries leading to normal or diseased cell behavior are wired and implemented, and how to manipulate them for therapy. again, typical snapshot average data is too ambiguous and usually allows differ- ent competing interpretations (schroeder, nature methods 2011; etzrodt et al., cell stem cell 2014; skylaki et al., nature biotechnology 2016). figure 3: continuous single-cell molecular quantification is required to understand the molecular dynamics underlying cell fate control. 19

the purpose of this manuscript is the review of some of my own group’s work honored with the cloëtta prize, not a comprehensive review of the literature. the listed references are therefore restricted to my own publi- cations. for a more comprehensive and balanced representation of the relevant literature, i refer to the references in my listed publications. development of technologies for long-term single-cell quantification of cellular and molecular dynamics for the reasons discussed above, we develop novel technologies allow- ing the continuous long-term imaging, single-cell tracking and quantifi- cation of cells. figure 5: long-term imaging, segmentation and tracking enables the single-cell quan- tification of cellular and molecular dynamics over up to weeks. adapted from (hilsenbeck et al., nature biotechnology 2016; skylaki et al., nature biotechnology 2016). 21

japanese mentor shinichi nishikawa – it turned out that in routine day- to-day use, it was neither a match for the data volumes at hand, nor for the required usability, reliability and specific functionality. figure 6: software tools for single-cell segmentation (faster), tracking (ttt) and quan- tification (qtfy, xit) developed in the schroeder group. all our published software is open-sourced and can be downloaded at www.bsse.ethz.ch/csd/software.html. i therefore had to start programming myself, the result of which (ttt, fig. 6) (hilsenbeck et al., nature biotechnology 2016) has meanwhile been much further developed by contributions of many, and proven use- ful for many published and ongoing studies in groups on 4 continents. together with self-programmed software for microscope hardware con- trol, computer vision and machine learning for cell recognition and seg- mentation (hilsenbeck et al., bioinformatics 2017), automated cell track- ing, image correction (schwarzfischer et al., proceedings microscopic image analysis with applications in biology 2011; buggenthin et al., bmc bioinformatics 2013; peng et al., nature communications 2017) and quantification (hilsenbeck et al., nature biotechnology 2016), as well as statistical analysis of pedigree structures (stadler et al., journal of theoretical biology 2018) and machine learning for high-dimensional pattern recognition and cell fate predictions (buggenthin et al., nature methods 2017), it is now part of a continuously growing software pipe- line. this pipeline enables the long-required continuous long-term single-cell quantification of many dimensions of cellular and molecular properties, dynamics and kinship. for example, divisional history, position, interac- tion, and protein expression or activity are recorded and quantified for all observed individual cells over many generations (fig. 5). as discussed above, this is a crucial prerequisite for the improved understanding of 23

these combined solutions routinely and robustly work in my and other groups. however, they still require dedicated specialists who understand the value of the new kind of generated data to be willing to invest their time. many challenges remain, and generic one-fits-all solutions do not exist. my disappointing answer to the typical question of interested col- leagues “how do i best do these experiments” unfortunately remains (and will likely also remain in many instances in the future) “it depends”. each novel combination of specific biological question, available biological material, reporters and culture system, required imaging frequency, di- mensionality and duration, and optical properties of the observed struc- tures will typically need rounds of optimization, and specialists’ love and care in acquisition and analysis of data (skylaki et al., nature biotech- nology 2016). in many cases, automation of data analysis fails due to the lack of e.g. reliable computer vision solutions, and manual curation, error correction or even analysis and generation remain required. given the new kind of continuous single-cell quantification and kinship data, the required mathematical tools often have not yet even been developed, let alone implemented into easy to use automated software tools, and a lot of groundwork is still required in this area. nevertheless, i am convinced that quantification of behaviors over time, as opposed to states at one timepoint, will be the future also of routine screening approaches e.g. in pharmaceutical industry. indeed, we have begun to use long-term singe-cell fate quantification for mid-throughput screening for novel extracellular regulators of stem cell self-renewal ex- pressed by their niche. by observing the behavior of individual stem cells in complex co-cultures with stromal cells, and the concurrent manipula- tion of 50 different candidate genes in the stem cells’ environment, we were able to identify more novel regulators in a year than the field had in the previous 20 years of research using the same cell models (kok- kaliaris et al., blood 2016). this well demonstrates that the conclusions yielded by this continuous observation approach are typically so much more robust and allow insights which would otherwise be missed, that higher investments into the more demanding technological approach will quickly pay off. in particular for recurring problems as in high-through- put screening with standardized cells and questions to be analyzed, the relevant steps can be automated with sufficient reliability. most of the 25

figure 9: nature biotechnology cover depicting one view of the central sinus of mouse bone marrow from the voluminous open imaging data published in this issue. reprinted by permission from springer nature, nature biotechnology, three-dimensional map of non- hematopoietic bone and bone-marrow cells and molecules, daniel l coutu, konstantinos d kokkaliaris, leo kunz, timm schroeder, copyright 2017. we are now using this approach to better quantify the location e.g. of dif- ferent hematopoietic and mesenchymal and skeletal stem and progenitor cell populations, their hierarchy, and their possible interactions with their microenvironments in the bone marrow and other tissues. most recently, by incorporating antibody-based proximity ligation into the approach, we also succeeded to improve its sensitivity to the sin- gle-molecule level (kunz et al., unpublished). this now allows the pre- cise quantification of the location and concentration of many relevant molecular regulators and their interaction with other molecules, in space and simultaneously in relation to multiple cell types of interest. it will be exciting to see the individual molecular players in their specific locations and in relation to their producing and target cells, as opposed to the dif- fuse idea of an average even distribution throughout a tissue. 27

ance of detectable blood cell numbers, with subsequent migration of the early blood cells to these sites either by the circulation within blood ves- sels, or by active migration outside the vessels (tanaka et al., cell re- ports 2014). finally, another explanation was the existence of hemogenic endothelium. in this case, endothelial cells would first be generated, and a subset would later differentiate into blood (fig. 10). it would explain why nascent blood cells are found next to endothelium, and why endothe- lial and early hematopoietic cells share many molecular markers. how- ever, this explanation could also hold true for all other above-mentioned hypotheses. figure 10: possible cellular sources for the first blood cells during embryo-genesis. pos- sible relationships between endothelium and blood. left: endothelium and blood are inde- pendently created from one progenitor (hemangioblast). right: blood is generated from specialized hemogenic endothelial cells. the existence of hemogenic endothelium could be proven by continuous long-term single-cell imaging of murine endothelial to hematopoie- tic transition in mesodermal cells derived from embryonic stem cells or primary embryonic mesoderm (eilken et al., nature 2009). why was it so difficult to prove the existence of hemogenic endothelium? since this process happens within the embryo, in mammals also deep in the uterus, it could never be observed live and at the single-cell level. the available snap-shot data from e.g. fixed and sectioned embryos could not exclude the other hypotheses discussed above as the sole and sufficient explanation. the existence of hemogenic endothelium thus remained dis- puted until 2009. 29

proving attempts of generating definitive blood stem and progenitor cells from pluripotent and endothelial cells in vitro. lineage choice: controlled by cell-intrinsic stochastic switches or instructed by cell-extrinsic signals? how are lineage choice decisions made in differentiating multipotent pro- genitor cells? are they made cell-autonomously by cell-intrinsic mech- anisms or instructed by cell-extrinsic signals? this central and seemingly trivial question in hematopoietic stem and progenitor cell biology is dis- cussed since the 1950s, and two major schools of thought with opposing basic concepts remain under discussion until today. the more obvious hypothesis assumes that lineage choice is instructed by signals from the microenvironment, which activate signaling pathways controlling the mo- lecular programs inducing lineage choice, commitment and maturation (see also next chapter). however, colony assays in vivo and in vitro yield very heterogeneous lineage outputs with lineage choice frequencies which are constant only at the population level, but different and unpredictable between individual cells. this observation is hard to reconcile with the idea that lineage choice is under strict control of extracellular signals since all cells in the same culture medium should then behave the same. an alternative model of lineage choice therefore assumes cell intrinsic mechanisms which lead to different lineage choices with specific prob- abilities, respectively. in this case, the lineage choice of an individual cell is independent of its environment and cannot be predicted. however, at the population level, frequencies of a specific lineage are fixed. this is a very attractive model, since it would allow multipotent cells the required flexibility to differentiate into different cell types, while also being ro- bust against dysregulated signals from the environment which would lead to overshooting uni-lineage differentiation (and thus the lack of required other lineages). in this model, the required adaptation of lineage output of the blood system depending on the body’s need, in case of e.g. infec- tions or lower oxygen environments, would be ensured by allowing sur- vival and proliferation only of the required cell types after their lineage commitment, but not by influencing the lineage choice itself (see also se- lective model in the next chapter). 31

figure 11: the putative transcription factor network underlying cell intrinsic stochastic lineage decision making during myeloid differentiation of hematopoietic stem cells. parts adapted from (krumsiek et al., plos one 2011). we non-invasively quantified the protein numbers of two transcription factors, pu.1 and gata1, in living cells throughout their differentiation. these two proteins have long been described as lineage specific transcrip- tion factors for the monocytic/granulocytic and megakaryocytic/eryth- roid lineages of the hematopoietic system, respectively. their overexpres- sion can reprogram cells from one to the other lineage, respectively. both protein can bind each other to inhibit the other’s activity, and can auto-ac- tivate the transcription of their own genes, respectively. this wiring con- stitutes a toggle switch, where higher expression of one would lead to ever stronger expression and dampening of the stronger and weaker fac- tor, respectively. a cell initially expressing both factors (e.g. before lin- eage choice), would thus flip into a state where only one of the factors would be expressed, leading to the lineage decision driven by this factor. after years of technological optimizations, generating the required re- porter mouse lines and manually tracking, we finally were able to simul- taneously quantify the dynamics of protein expression for both transcrip- tion factors in living differentiating hematopoietic stem cells and all their 33

nature versus nurture: lineage selection or instruction by hematopoietic cytokines? can cell-extrinsic cytokine signals influence the lineage choice of multi- potent hematopoietic progenitors? related to the question discussed above, this central and obvious question in hematopoietic stem and pro- genitor cell biology was intensively discussed since the 1950s. it surpris- ingly remained without a definitive answer for more than half a century – and while billions of us dollars’ worth of cytokines were used annually for clinical therapy. it was well-known that the lineage composition of hematopoietic colo- nies is influenced by the specific microenvironment these colonies de- veloped in in vivo, or by the presence of hematopoietic cytokines in de- fined culture conditions in vitro. the types of living cells ultimately produced from multipotent blood progenitor cells can therefore be in- fluenced by cell-extrinsic signals. however, this could be explained by very different fundamental mechanisms – lineage instruction versus lin- eage selection – which would both lead to the same final experimental observations described above. one possibility is that cytokine signaling directly influences the genetic and epigenetic programs controlling lin- eage choice – lineage instruction. however, it could also be possible that cells make their lineage choice independently of signaling pathway ac- tivity (see previous chapter), and cytokine signaling would only influ- ence the survival and/or proliferation of already lineage committed cells. in this case, signaling pathways activated by cytokine signaling would only influence the molecular programs involved in cell survival or pro- liferation control, but have no influence on molecular lineage choice control. 35

individual cell during that time – which could have cell-intrinsically com- mitted to another lineage and was then killed due to the lack of its re- quired lineage specific cytokine – was impossible with previous technol- ogies. we therefore used continuous long-term observation of all individual cells produced from individual granulocyte-monocyte progenitor (gmp) cells over many days until their lineage commitment could be reliably detected. gmps were cultured under chemically defined conditions with the pres- ence of either cytokine g-csf or cytokine m-csf, leading to the pro- duction of only granulocytic or monocytic cells at the end of the cultures from the same starting gmp population, respectively. this allowed us to quantify the frequency of cell death and division events for all of the prog- eny of the initial cells. between these culture conditions, we could not find relevant differences in cell proliferation, or in cell death. importantly, the frequency of cell death events was not sufficient to explain the lack of production of granulocytic cells under m-csf conditions or mono- cytic cells under g-csf conditions. thus, it was not the selective killing of the “other lineage” cells under lineage specific cytokine culture con- ditions leading to a uni-lineage differentiation output. lineage choice must therefore have been directly instructed by signaling activity. signaling pathways activated by cytokine receptors therefore must change the molecular programs controlling lineage choice. this insight not only clarifies a long-standing dispute about a core mechanism of multipotent progenitor cell fate control. it is important also because it offers an ex- cellent experimental system to now identify the pathways and relevant molecular mechanisms underlying lineage choice. this should be easy. one would think that a simple comparison of the in- tracellular signaling pathways activated by the opposing cytokine recep- tors would identify the pathway(s) responsible for one or the other line- age choice. however, it turns out to be surprisingly difficult. despite their opposing effects on lineage choice, the receptors for g-csf and m-csf both activate many signaling pathways, and most of them overlapping and highly interconnected. add the shared confounding effects of both cytokines on cell survival, proliferation, maturation, adhesion and acti- vation, it becomes very demanding to disentangle the effect of these in- 37

there is beautiful precedence for this concept from cell lines. however, due to the technological demands, the concept has never been tested for primary hematopoietic stem and progenitor cells. it, again, requires the demanding combination of non-invasive high-frequency quantification of signaling pathway activity over hours with the long-term quantifica- tion of future cell fate choices over days – in living single cells and all of their progeny. we have therefore developed approaches for the high-fre- quency live quantification of transgenic biosensors for signaling pathway activity – simultaneously for many different pathways in primary mouse and human stem and progenitor cells. indeed, we find highly heteroge- neous signaling pathway dynamics in individual cells of purified progen- itor populations, despite stimulation with the same cytokine. it will now be interesting to link these specific dynamics to the future cell fate choices of individual cells to add another layer of molecular fate control to our understanding of hematopoietic cell fates. asymmetric cell division of hematopoietic stem and progenitor cells are hematopoietic stem and progenitor cell fates controlled by asymmet- ric cell division? this question has been under dispute for many years. in asymmetric cell division, the future asymmetric fates of two sister cells are fixed during the division of their mother cell. this could be due to e.g. the asymmetric inheritance of intracellular cell fate determinants into the two daughters, or the orientation of the division plane leading to un- equal niche access of the two sisters after division (fig. 15). it would be an attractive explanation how the number of stem cells could be kept con- stant throughout the body without the need of complex and potentially vulnerable systemic feedback mechanisms. each stem cell would, under homeostatic conditions, give rise to one daughter which would go on to differentiate and produce the different cell types of the blood system, and one daughter which replaces its mother as a stem cell, thus keeping the stem cell pool size constant. while it has beautifully been shown to ex- ists in other cell types, and some textbooks include asymmetric division even in the definition of hematopoietic stem cells, many researchers do not believe it exists in these cells. again, the reason for this long-stand- ing dispute is the lack of adequate technology. observation of either asym- metric fates or asymmetric inheritance of intracellular molecules or niche 39

titation. however, the reproducible frequency, and the clear correlations found now finally allow us to conclude that asymmetric cell division ex- ists in hematopoietic stem cells. it is an orthogonal and high-level regu- latory mechanism controlling hematopoietic stem cell fates with a lot of potential for novel insights. it will now be exciting to unravel the molec- ular mechanisms, target effector programs and possibilities for molecu- lar manipulation of this process for therapeutic intervention. where to next? after 1.5 decades of my own independent research group, we have es- tablished important and unique technologies for long-term single-cell quantifications. these approaches work, both in my own, and in other groups. but they still need expert knowledge and further optimizations. they have contributed to answering diverse long-standing questions in different cell types and molecular systems. after slow and sometimes te- dious development of technology, we are now getting faster and faster in successfully applying it to novel biological questions. in addition to quan- titative observations, the precise molecular manipulation, both through fine control of fluidics, and increasingly through fast optogenetic ap- proaches, will become important for unravelling the functional role of specific molecules in regulatory networks. with increasing numbers of well-defined culture systems for different cell and tissue types, more and more biological questions become available for long-term single-cell im- aging and quantification. the advent of organoid cultures of many solid tissues in combination with light-sheet imaging will lead to a surge of imaging data to be analyzed for the same concepts, but will also require novel custom software components. long-term in vivo single-cell imag- ing with sufficient throughput and duration remains a crucial goal in the field, but will likely require novel imaging modalities for many of the bi- ological questions at hand. improved automation, algorithms and soft- ware remain a crucial requirement, and a lot of work still has to be done in this area. after mostly working on murine systems for the first many years due to their better experimental accessibility, reproducibility and the availability of transgenic reporter systems, we have begun to increas- ingly build on the gathered experience for the analysis also of human cells. finally, i am convinced that quantification of cellular and molecu- 41

dettinger p, frank t, etzrodt m, ahmed n, reimann a, trenzinger c, loeffler d, kok- kaliaris kd, schroeder t and tay s (2018). automated microfluidic system for dynamic stimulation and tracking of single cells. anal. chem. acs.analchem8b00312. eilken hm, nishikawa s-i and schroeder t (2009). continuous single-cell imaging of blood generation from haemogenic endothelium. nature 457, 896–900. endele m, loeffler d, kokkaliaris kd, hilsenbeck o, skylaki s, hoppe ps, schambach a, stanley er and schroeder t (2017). csf-1-induced src signaling can instruct mono- cytic lineage choice. blood 129, 1691–1701. etzrodt m, endele m and schroeder t (2014). quantitative single-cell approaches to stem cell research. cell stem cell 15, 546–558. filipczyk a, marr c, hastreiter s, feigelman j, schwarzfischer m, hoppe ps, loeffler d, kokkaliaris kd, endele m, schauberger b, hilsenbeck o, skylaki s, hasenauer j, ana- stassiadis k, theis fj and schroeder t (2015). network plasticity of pluripotency tran- scription factors in embryonic stem cells. nat. cell biol. 17, 1235–1246. hilsenbeck o, schwarzfischer m, skylaki s, schauberger b, hoppe ps, loeffler d, kok- kaliaris kd, hastreiter s, skylaki e, filipczyk a, strasser m, buggenthin f, feigelman js, krumsiek j, van den berg ajj, endele m, etzrodt m, marr c, theis fj, et al. (2016). soft- ware tools for single-cell tracking and quantification of cellular and molecular properties. nat. biotechnol. 34, 703–706. hilsenbeck o, schwarzfischer m, loeffler d, dimopoulos s, hastreiter s, marr c, theis fj and schroeder t (2017). faster: a user-friendly tool for ultrafast and robust cell seg- mentation in large-scale microscopy. bioinformatics. hoppe ps, coutu dl and schroeder t (2014). single-cell technologies sharpen up mam- malian stem cell research. nat. cell biol. 16, 919–927. hoppe ps, schwarzfischer m, loeffler d, kokkaliaris kd, hilsenbeck o, moritz n, en- dele m, filipczyk a, gambardella a, ahmed n, etzrodt m, coutu dl, rieger ma, marr c, strasser mk, schauberger b, burtscher i, ermakova o, bürger a, et al. (2016). early myeloid lineage choice is not initiated by random pu.1 to gata1 protein ratios. nature 535, 299–302. kokkaliaris kd, drew e, endele m, loeffler d, hoppe ps, hilsenbeck o, schauberger b, hinzen c, skylaki s, theodorou m, kieslinger m, lemischka i, moore k and schroeder t (2016). identification of factors promoting ex vivo maintenance of mouse hematopoietic stem cells by long-term single-cell quantification. blood. krumsiek j, marr c, schroeder t and theis fj (2011). hierarchical differentiation of my- eloid progenitors is encoded in the transcription factor network. plos one 6, e22649. 43

the cloëtta prize 2018 is awarded to professor johanna a. joyce born in 1972 in london, england ludwig institute for cancer research department of oncology university of lausanne for her ground-breaking contributions to the understanding of the tumor environment and the factors that influence tumor progression and therapeutic resistance lausanne, 9th november 2018 in the name of the foundation board: the president the vice president a member 45

curriculum vitae institution: department of oncology university of lausanne ludwig institute for cancer research switzerland www.joycelab.org, johanna.joyce@unil.ch date of birth: 28th march 1972 nationality: irish education 1990–1994 1995–1999 b.a. (hons) in genetics, trinity college dublin, ireland ph.d. in biology, university of cambridge, uk employment history – research experience present present 2014–2015 2014–2015 2010–2014 2010–2014 2005–2010 2004–2010 full professor, department of oncology, university of lausanne, switzerland full member, ludwig institute for cancer research, lausanne branch, switzerland full member, memorial sloan kettering cancer center, new york, usa full professor, weill cornell graduate school of medical sciences, ny, usa associate member, memorial sloan kettering cancer center, new york, usa associate professor, weill cornell graduate school of medical sciences, ny, usa assistant professor, weill cornell graduate school of medical sciences, ny, usa assistant member, memorial sloan kettering cancer center, new york, usa 47

2016–present 2016–2017 2014–2018 2007–2011 scientific advisory board, irb institute, barcelona, spain elected chairperson of aacr tumor micro- environment working group steering committee, aacr tumor micro- environment working group elected to council of the international proteolysis society selected funding swiss bridge award (1/1/2018–31/12/2019) investigating and therapeutically targeting neutrophils in brain metastasis swiss cancer league (1/1/2017–31/12/2019) targeting tumor-associated macrophages to enhance therapeutic efficacy in gliomas cancer research uk grand challenge team award (1/5/2017–30/4/2023) imaxt: imaging and molecular annotation of xenografts and tumors breast cancer research foundation (1/10/2009–present) microenvironmental regulation of breast cancer metastasis and therapeutic response national cancer institute, r01 grant (1/7/2014–30/6/2019) investigating and targeting tams in the glioma microenvironment roche strategic alliance (1/11/2016–31/10/2018) combinatorial strategies for therapeutically targeting the glioma microenvironment alan and sandra gerry foundation (1/8/2014–7/31/2016) investigation of the microenvironmental transcriptome in breast to brain metastasis ono pharmaceuticals (1/7/2015–30/6/2016) small molecule inhibitor screen for m2 macrophage reprogramming 49

selected publications 1. quail df*, olson oc*, bhardwaj p, walsh la, akkari l, quick m, chen ic, wen- del n, ben-chetrit n, walker j, holt pr, dannenberg aj, joyce ja (2017). obesity alters the lung myeloid cell landscape to enhance breast cancer metastasis via il5 and gm-csf. nature cell biology 19: 974–987. 2. olson oc, kim h, quail df, foley ea, joyce ja (2017). tumor-associated mac- rophages suppress the cytotoxic activity of antimitotic agents. cell reports 19: 101–113. 3. quail d and joyce ja (2017). the microenvironmental landscape of brain tumors. cancer cell 31: 326–341. 4. quail df, bowman rl, akkari l, quick ml, schuhmacher aj, huse jt, holland ec, sutton jc, joyce ja (2016). the tumor microenvironment underlies acquired resistance to csf-1r inhibition in gliomas. science 352: aad3018. 5. bowman rl, klemm f, akkari l, pyonteck sm, sevenich l, quail df, dhara s, simpson k, gardner ee, iacobuzio-donahue c, brennan cw, tabar v, gutin ph, joyce ja (2016). macrophage ontogeny underlies differences in tumor-specific education in brain malignancies. cell reports 17: 2445–2459. 6. sevenich l, bowman r*, mason sd*, quail df, rapaport f, elie bt, brogi e bras- tianos p, hahn wc, holsinger l, massague j, leslie cs, joyce ja (2014). analysis of tu- mour and stroma-supplied proteolytic networks reveals a brain metastasis-promoting role for cathepsin s. nature cell biology 16: 876–888. 7. pyonteck sm*, akkari l*, schuhmacher aj*, bowman rl, sevenich l, quail d, olson oc, quick m, huse j, teijeiro v, setty m, leslie c, oei y, pedraza a, zhang j, bren- nan cw, sutton jc, holland ec, daniel d, joyce ja (2013). csf-1r inhibition alters mac- rophage polarization and blocks glioma progression. nature medicine 19: 1264–1272. 8. quail df and joyce ja (2013). microenvironmental regulation of tumor progression and metastasis. nature medicine 19: 1423–1437. 9. shree t*, olson oc*, elie bt, kester jc, garfall al, simpson k, bell-mcguinn km, zabor ec, brogi e, joyce ja (2011). macrophages and cathepsin proteases blunt chemotherapeutic response in breast cancer. genes and development 25: 2465–2479. 10. gocheva v*, wang hw*, gadea bb, shree t, hunter ke, garfall a, berman t, joyce ja (2010). il-4 induces cathepsin protease activity in tumor-associated macrophages to promote cancer growth and invasion. genes and development 24: 241–255. (full publication list and pdfs of all published articles: http://joycelab.org/publications) 51

introduction tumors contain diverse cell types and inflammatory mediators within their tme, including endothelial cells, fibroblasts, tissue-resident and peripherally-derived immune cells, among others (joyce, 2005; quail and joyce, 2013; quail and joyce, 2017c) (fig. 1). depending on the organ, there are also unique compositions of tissue-specific resident cell types and extracellular matrix molecules, which can affect tumor development in different ways (joyce and pollard, 2009; quail and joyce, 2017a). in- deed, tumor progression is not only dictated by genetic alterations within the cancer cells, but also by whether the surrounding niche is permissive to growth at each stage of disease. thus, a full mechanistic understand- ing of both tumor cell-intrinsic and -extrinsic mediators of malignant pro- gression is critical to optimize therapeutic strategies against cancer. interactions between tumor cells and the associated stroma and cells of the immune system profoundly influences cancer initiation, progression and patient prognosis. the link between chronic inflammation and tum- origenesis was first proposed by rudolf virchow in 1863 following his seminal observation that infiltrating leukocytes are a hallmark of tumors (virchow, 1863). another clinical investigator of that era, stephen paget, specifically recognized the importance of the microenvironment in de- termining the organ-tropism of metastasis, leading to his seminal “seed and soil” hypothesis published in 1889 (paget, 1889). paget stated that “when a plant goes to seed, its seeds are carried in all directions; but they can only live and grow if they fall in congenial soil”. however, despite these and other key findings dating back to the late 19th century, for many subsequent decades the tme was overlooked, as researchers predomi- nantly focused on identifying the genetic drivers of cancer. in recent years the tme field has exploded, with a plethora of studies contributing to a molecular and cellular understanding of the importance and complexity of the tme, further complicating the already challeng- ing task of understanding and treating cancer. thus, while cancer was long viewed as a heterogeneous disease driven by dna mutations and genomic alterations in tumor cells, it is now evident that tumors are sim- ilarly diverse by nature of their microenvironmental composition. more- over, in response to evolving environmental conditions and oncogenic 53

figure 1: multiple stromal cell types converge to support a tumorigenic primary niche. after circumventing cell-intrinsic mechanisms of apoptosis, tumor cells are subject to elim- ination pressures by the immune system. tumor cell-specific antigens have a role during this process, which are recognized by cytotoxic immune cells, leading to their destruction. fibroblasts and macrophages within the tumor microenvironment (tme) also contribute to a growth-suppressive state; however, these cells may later become educated by the tumor to acquire pro-tumorigenic functions. for instance, tumor-associated macrophages (tams) support diverse phenotypes within the primary tumor, including growth, angiogenesis and invasion, by secreting a plethora of pro-tumorigenic proteases, cytokines and growth fac- tors. as tumors grow, immune-suppressor cells, including myeloid-derived suppressor cells (mdscs) and regulatory t cells are mobilized into the circulation in response to activated cytokine axes that are induced by tumorigenesis, and infiltrate the growing tumor to dis- rupt immune surveillance through multiple mechanisms, including, but not limited to, dis- ruption of antigen presentation by dendritic cells, inhibition of t and b cell proliferation and activation, or inhibition of natural killer (nk) cell cytotoxicity. cancer-associated fi- broblasts (cafs), which become activated by tumor-derived factors, secrete extracellular matrix (ecm) proteins and basement membrane components, regulate differentiation, mod- ulate immune responses and contribute to deregulated homeostasis. in addition to cellular contributions, several extracellular properties contribute to tumor progression, including low oxygen tension, high interstitial fluid pressure and changes in specific components of the ecm. from quail and joyce, nature medicine (2013). 55

of matrix-degrading enzymes, specifically cathepsin proteases and hep- aranase (gocheva et al., 2010b; hunter et al., 2014), and identified the upstream cytokines responsible for their induction (yan et al., 2016) (fig- ure 2, and discussed further below). figure 2: model of reciprocal interactions between cancer cells, tumor-associated mac- rophages and additional cells in the tumor microenvironment that enhance malignant pro- gression. il-4 and other th2 cytokines are produced by cancer cells and t cells in the tme, leading to an increase in protease activity in tumor-associated macrophages (tams) that promotes several hallmarks of cancer, including angiogenesis, invasion and metas- tasis. schematic depicts data compiled from: gocheva, wang et al, genes dev (2010); pyonteck, akkari, schuhmacher et al, nat med (2013); akkari et al, genes dev (2014); hunter et al, oncogene (2014); sevenich et al, nat cell biol (2014); yan, wang, bowman and joyce, cell reports (2016); quail et al, science (2016). through our exploration of the molecular differences between tams and their normal counterparts, we also became intrigued as to whether tis- sue-resident macrophages, such as microglia in the brain, differ from pe- ripherally-recruited macrophages in terms of gene expression, epigenetic regulation and tumorigenic functions. to address this question, robert bowman, a graduate student in my lab, used complementary cell line- age-tracing genetic models to selectively distinguish resident microglia (mg) from bone marrow-derived macrophages (bmdms) recruited from the periphery (bowman et al., 2016). using this strategy, he investigated the epigenetic and transcriptomic regulatory machinery underlying dif- 57

therapeutic targeting of tams in cancer to evaluate the therapeutic potential of targeting tams, we have taken a number of different approaches, including pharmacological inhibition or genetic ablation of colony stimulating factor-1 (csf-1) signaling (quail and joyce, 2017b), which is a key mediator of macrophage survival and differentiation (noy and pollard, 2014). we began by investigating the consequences of csf-1 deletion using a pancreatic neuroendocrine can- cer mouse model (rip1-tag2), as we had simultaneously revealed key functions for tam-supplied cathepsins in rip1-tag2 mice (see below). stephanie pyonteck, a graduate student in my lab, found that csf-1 de- letion led to tam depletion in the pancreas and a substantial reduction in cumulative tumor burden (pyonteck et al., 2012). interestingly, she de- termined that this resulted from a significant decrease in the initial angi- ogenic switching of progenitor lesions and subsequent development of tumors, rather than an evident effect on tumor growth. this study thus revealed important functions for tams at the earliest stages of tumor in- itiation, thereby expanding the repertoire of tam functions beyond the promotion of advanced malignancy and metastasis. in collaboration with our colleague laura tang, at mskcc in new york, we also analyzed a cohort of tissue samples from human pancreatic neuroendocrine tumors (pannets). we found that elevated tam number in the pancreas in- creased with tumor grade, as in the mouse model, and importantly showed that this can be prognostic for pannet patients that develop liver metas- tases (pyonteck et al., 2012) (figure 4). 59

another tumor type in which tams are highly abundant and associated with aggressive disease are gliomas that arise in the brain. we and others have found that tumor-associated macrophages and microglia together can comprise up to 30% of the total tumor mass in glioblastomas, thus representing the most abundant non-cancerous cell type in this high-grade malignancy (hussain et al., 2006; komohara et al., 2008; bowman et al., 2016; quail and joyce, 2017a). most therapeutic approaches directly tar- geting tumor cells in glioblastoma have failed. we therefore proposed an alternative strategy: to target tams in the brain tme. stephanie pyon- teck, leila akkari, alberto schuhmacher and several other key lab mem- bers teamed up to work on this exciting project. we used an inhibitor of the csf-1 receptor, csf-1r, to target tams in mouse glioblastoma mod- els developed by our collaborator eric holland, who was also then at mskcc. treatment with this selective inhibitor (blz945 from novar- tis) regressed established high-grade tumors, even after just one week of treatment (pyonteck et al., 2013). we found that csf-1r inhibition mark- edly increased tumor cell apoptosis and phagocytosis in vivo, while de- creasing proliferation and glioma malignancy, and significantly extend- ing survival (figure 5). figure 5: csf-1r inhibition specifically targets macrophages, improves survival and de- creases glioma malignancy in the transgenic pdgf-driven glioma (pdg) mouse model. symptom-free survival curves are shown for pdg mice treated in an early intervention trial with a csf-1r inhibitor blz945 (red) or vehicle (black), demonstrating a dramatic in- crease in survival following csf-1r inhibition. from pyonteck, akkari, schuhmacher et al, nature medicine (2013). 61

to explore the underlying mechanisms. interestingly, upon isolation and transplantation of tumor cells from recurrent gliomas into naïve animals, daniela found that sensitivity to csf-1r inhibition was re-established, indicating that the resistance was in fact driven by the microenvironment. through rna-sequencing of glioma cells and tams purified from treated tumors, and ex vivo cell culture assays, we discovered an elevation in pi3k pathway activity in recurrent glioblastoma following csf-1r in- hibition, which was driven by macrophage-derived igf-1 and tumor cell igf-1r (quail et al., 2016) (figure 6). consequently, combining igf-1r or pi3k blockade with continuous csf-1r inhibition in recurrent tum- ors dramatically extended overall survival. by contrast, monotherapy with igf-1r or pi3k inhibitors in rebound or treatment-naïve tumors was minimally effective, indicating the necessity of combination therapy to expose pi3k signaling dependency in recurrent disease. mechanistically, daniela found that t cell-derived il4 led to macrophage activation in re- current tumors, and elevated stat6 and nfat signaling upstream of igf-1 induction. similarly, inhibition of any of these pathways in vivo was also sufficient to significantly extend survival when combined with csf-1r inhibition (quail et al., 2016). given that pi3k signaling is ab- errantly activated in a substantial proportion of glioma patients, includ- ing through mutations in pten and other pi3k pathway components, it is possible that this pathway could also contribute to intrinsic resistance to csf-1r inhibition. our findings thus revealed the importance of con- tinuous bidirectional feedback between cancer cells and the tme, and support the notion that although immune and stromal cells are less sus- ceptible to genetic mutation than are cancer cells, a tumor can nonethe- less acquire a resistant phenotype by exploiting its extracellular environ- ment (quail et al., 2016; quail and joyce, 2017b). 63

cysteine cathepsin proteases and heparanase, was a prominent feature of tams in multiple tmes (gocheva et al., 2010b; hunter et al., 2014). in earlier experiments, dating back to my postdoc in doug hanahan’s lab at ucsf, we had found that expression of a subset of 6 of 11 cathepsin family members were progressively upregulated during pancreatic can- cer progression (joyce et al., 2004) in the rip1-tag2 mouse model in- troduced above. cathepsins are typically lysosomal enzymes, which are critical for terminal protein degradation (olson and joyce, 2015). to ex- plore whether they might have extra-lysosomal functions in cancer, we used activity-based probes developed by matthew bogyo, a chemical bi- ologist and long-standing collaborator now at stanford university. we successfully imaged global cathepsin activity in vivo in several mouse models of cancer and found a specific increase in tams within the tme of pannets, breast cancer, and lung metastases (joyce et al., 2004; gocheva et al., 2010b) (figure 7). figure 7: increase in cathepsin activity in tams during pancreatic islet tumor develop- ment in the rip1-tag2 mouse model. cathepsins are highly activated in infiltrating mac- rophages during tumor progression at the angiogenic islet and tumor stages of rt2 tum- origenesis. mice were injected with the cathepsin activity-based probe (cath-abp), and the resulting tissues stained with a f4/80 antibody to visualize macrophages. normal (n) tag+ islets were analyzed at 4–7 wks of age, hyperplastic (h) islets at 8 wks, angiogenic (a) islets at 10 wks, and tumors (t) at 13.5 wks of age. the percentage of f4/80+ cells that were cath-abp+ was determined by image analysis and is indicated in the representative image for each stage. macrophages present in the normal adjacent exocrine (e) pancreas did not show high levels of cathepsin activity. arrows represent the invasive tumor front. adapted from gocheva, wang et al., genes and development (2010). 65

figure 8: stat3 and stat6 signaling pathways synergize to promote cathepsin secretion from macrophages via ire1-alpha activation. we found that the th2 cytokine il-4 synergizes with il-6 and il-10 in macrophages to promote pancreatic neuroendocrine tumor growth and invasion. this synergy depends on stat3 and stat6 interaction to activate ire1-alpha, leading to a pronounced secretion of cathepsin proteases and induction of unfolded protein response-related genes. from yan, wang, bowman and joyce, cell reports (2016). cathepsin proteases are potent regulators of multiple hallmarks of cancer to gain insights into the mechanisms by which cathepsins regulate differ- ent hallmark capabilities of cancer, we devised a comprehensive genetic strategy to delete individual cathepsins (alone and in combination) and de- termine the consequences for tumor proliferation, angiogenesis, and inva- sion. using the rip1-tag2 model we first sought to identify the key tu- mor-promoting family members from the six that we found upregulated (b, c, h, l, s, and z) from whole genome expression analyses (joyce et al., 2004). there were several compelling reasons for undertaking this ge- netic analysis. first, to fully understand how cathepsins promote tumori- genesis it was critical to determine how each family member individually regulates tumor growth, invasion and angiogenesis. second, from a trans- lational perspective, when using pan-family inhibitors, there is the possi- bility of undesirable effects if some family members are actually tumor suppressors (lopez-otin and matrisian, 2007). thus, identifying the tu- mor-promoting proteases, and developing selective inhibitors that only tar- get these enzymes is critical; as we have also addressed pharmacologically (sadaghiani et al., 2007; elie et al., 2010; sevenich et al., 2014). 67

to unravel the molecular mechanisms by which cathepsins promote the different hallmarks of cancer summarized above we employed both can- didate-based strategies and unbiased proteomic screens to reveal their substrates within the tme (figure 9). for example, leny gocheva iden- tified cleavage of the cell adhesion protein e-cadherin by the pro-inva- sive cathepsins b, l and s as a key mechanism that contributes to tumor invasion (gocheva et al., 2006). lisa sevenich discovered a brain metas- tasis-promoting function for cathepsin s via shedding of the junctional adhesion molecule, jam-b, which facilitates extravasation of tumor cells into the brain across the blood-brain barrier (bbb) (sevenich et al., 2014). leila akkari, with leny gocheva and other lab members, found that cathepsin z promotes cancer cell invasion and proliferation through a unique rgd-binding motif in its pro-domain, which promotes attach- ment via integrins to different ecm components, in a fak/src-depend- ent manner (akkari et al., 2014). moreover, we demonstrated that for this cathepsin family member, its tumor-promoting functions are actually in- dependent of its enzymatic activity, and instead rely on ecm-mediated signaling. in collaboration with the vlodavsky lab in israel, cathepsin l was identified as the major protease responsible for activation of the key matrix-degrading enzyme, heparanase (abboud-jarrous et al., 2008). in an interesting convergence, we had previously shown that inhibition of heparanase (joyce et al., 2005) disrupts several of the same tumorigenic pathways as pan-cathepsin inhibitors (joyce et al., 2004; bell-mcguinn et al., 2007; elie et al., 2010). karen hunter, a graduate student in my lab, thus investigated how genetic modulation of heparanase levels reg- ulates tumor progression using heparanase knockout and heparanase-over- expressing mice in the rip1-tag2 model, and thereby identified critical roles for heparanase in promoting lymphangiogenesis and tumor inva- sion (hunter et al., 2014). 69

in addition to these targeted candidate approaches, which were each very fruitful in identifying bona fide cathepsin substrates, we also collaborated with the lab of chris overall in vancouver to perform unbiased proteom- ics screens in vivo (prudova et al., 2016). by applying 8-plex itraq ter- minal amine isotopic labeling of substrates (tails), a systems-level n-terminome degradomics approach, we identified cleavage sites for in vivo substrates of cathepsins b, h, l, s, and z within the tme by taking advantage of the different cathepsin knockouts we had generated in the rip1-tag2 background (table 1). we validated several of the substrates using independent experimental approaches, including the glycolytic en- zyme pyruvate kinase m2 associated with the warburg effect, the er chaperone grp78, and the oncoprotein prothymosin-alpha. collectively, our studies over the past decade have revealed novel, unex- pected roles for cathepsin proteases as critical processing and activation enzymes, functioning as “master regulators” at the apex of multiple pro- tease networks, thereby greatly expanding their functions in cancer be- yond simple matrix degradation (mason and joyce, 2011; sevenich and joyce, 2014; olson and joyce, 2015). in parallel with the findings discussed here, we have also had many suc- cessful collaborations with colleagues exploring the roles of tumor-asso- ciated macrophages and myeloid cells in numerous diverse contexts. this includes investigating tam metabolism within the tme with carlos-car- mona fontaine and joao xavier; exploring how the senescence-associ- ated secretory program in the liver tme impacts macrophage polariza- tion with amaia lujambio, leila akkari and scott lowe; targeting tams in thyroid cancer with mabel ryder and jim fagin; working with matej krajcovic and mike overholtzer on phagocytosis, lysosome fission and nutrient uptake; rich bakst and rich wong on the promotion of perineu- ral cancer invasion by inflammatory monocytes; imaging of tams in breast cancer with avigdor leftin, nir ben-chetrit and jason koutcher, and lipid flux in macrophages with prakrit jena and dan heller; and in- jury-related brain inflammation with nduka amankulor and eric hol- land (amankulor et al., 2009; carmona-fontaine et al., 2013; krajcovic et al., 2013; lujambio et al., 2013; ryder et al., 2013; bakst et al., 2017; carmona-fontaine et al., 2017; jena et al., 2017; leftin et al., 2017), among other studies. 71

figure 10: cathepsin proteases and therapeutic resistance. adaptive upregulation of cathepsins can occur through the increased recruitment of cathepsin-high tams in response to chemotherapeutic agents such as paclitaxel or etoposide. these adaptive increases in intratumoural cathepsin activity levels blunt therapeutic efficacy, which can accordingly be improved by cathepsin inhibition. schematic from olson and joyce, nat rev cancer (2015), depicting data from shree, olson et al, genes dev (2011). we recently extended this initial finding (figure 10) by incorporating live cell imaging to investigate precisely how tams impact taxol-induced alterations in the mitotic arrest of cancer cells, through a collaboration with emily foley at mskcc, that was led by oakley olson in my lab. oakley found that macrophages suppress the duration of taxol-induced mitotic arrest in breast cancer cells and promote earlier mitotic slippage (olson et al., 2017a). this correlated with a decrease in the phosphoryl- ated form of histone h2ax (γh2ax), decreased p53 activation, and re- duced cancer cell death in interphase. he found that acute and specific depletion of major histocompatibility complex class ii (mhcii)-low tams increased taxol-induced dna damage and apoptosis in cancer cells, leading to greater efficacy in preclinical intervention trials. oak- ley’s mechanistic investigations also revealed the importance of the mapk/erk kinase (mek) pathway in this protective effect (figure 11), and mek inhibition blocked the protective capacity of tams and phen- ocopied the effects of tam depletion on taxol treatment in vivo (olson et al., 2017a). thus, we found that tams suppress the cytotoxic effects of taxol, in part through cell non-autonomous modulation of mitotic ar- rest in cancer cells, and consequently targeting tam-cancer cell interac- tions potentiates taxol efficacy (shree et al., 2011; olson et al., 2017a). 73

microenvironmental regulation of metastasis cancer cells in an aggressive primary tumor are adept at exploiting their local tissue environment. by contrast, when metastatic cells leave these favorable surroundings, they must possess or acquire traits that will allow them to survive and colonize foreign, potentially hostile tissue environ- ments (figure 12). the obstacles that metastasizing tumor cells encoun- ter vary from organ to organ, and are highly influenced by cells of the tme (joyce and pollard, 2009; quail and joyce, 2013). indeed, dissem- ination can occur to multiple organs, yet metastatic tumors typically grow in only one or a few sites, indicating critical roles for the microenviron- ment in this process, as already appreciated by paget in the late 19th cen- tury (paget, 1889). figure 12: initiation of secondary outgrowth in metastatic niches. dormant micrometasta- ses are held in check by several mechanisms including tumor mass dormancy, or angiogenic dormancy, when proliferation is balanced by apoptosis because of a lack of vasculature and limited supply of nutrients and oxygen. multiple tme cell types contribute to the re-estab- lishment of vascularity at the secondary site, including myeloid and endothelial cell progen- itors and tams. in addition, tumor cells can enter immune-induced dormancy whereby im- munogenic cells are cleared, and cells that are able to survive enter a state of equilibrium. immune suppressor cells are recruited to tumors in response to this process and contribute to the establishment of an immunosuppressive state within secondary tissues. once microme- tastases overcome dormancy, they become receptive to signals and cell types within the tme to further support their expansion. for example, tams are abundant in metastases of multi- ple cancer types and support different tumorigenic processes to allow for outgrowth, includ- ing vascularization, impaired immunogenicity and enhanced survival in overt metastases. platelets, and components of the coagulation system are also important mediators of meta- static outgrowth, as they interfere with the ability of natural killer (nk) cells to destroy mi- crometastases and support clot formation, which in turn causes the recruitment of myeloid suppressor cells. from quail and joyce, nature medicine (2013). 75

beyond the local tme, an inflammatory systemic environment can also affect disease outcome, by perturbing homeostasis within multiple tis- sues throughout the body. this becomes particularly important during metastasis, where systemic alterations can modify the tissue landscape of distant organs and support tumor cell colonization by establishing a pre-metastatic niche (mcallister and weinberg, 2014). indeed, chronic inflammation can significantly increase cancer risk and disease progres- sion (quail and joyce, 2013). investigation into how the systemic envi- ronment affects tumor biology is therefore critical for an integrated un- derstanding of cancer. as such, we have begun to explore how the systemic microenvironment modulates tumorigenesis and metastasis. we first chose to assess the clinically relevant case of obesity-associated chronic inflam- mation, as it can disrupt homeostasis within tissue microenvironments (olson et al., 2017b). given the correlation between obesity and increased relative risk of death from breast cancer, we focused on determining whether obesity-associated inflammation promotes metastatic progres- sion (quail et al., 2017). in this study, daniela quail and oakley olson together showed that obe- sity causes lung neutrophilia in otherwise-normal individuals (quail et al., 2017). they found this occurred independently of diet content; rather it was directly related to increased adiposity and the production of il5 by adipose tissue. they found that il5 increases csf2 (gm-csf) expres- sion by il5r+ monocytes, and enhances neutrophil trafficking to lung (figure 13). furthermore, elevated serum gm-csf promotes myelopoie- sis, leading to an expansion of peripheral neutrophils. in mouse models, obesity-associated lung neutrophilia enhanced breast cancer metastasis to this organ, and depletion of gr1+ neutrophils in obese animals reversed this effect (olson et al., 2017b; quail et al., 2017). we also found that gm-csf is predominantly expressed in the lungs of obese mice, and that gm-csf blockade in vivo reverses the pro-metastatic effects of obesity. interestingly, weight loss was equally effective at reversing all these phe- nomena in mice, including breast-to-lung metastasis. in collaboration with andrew dannenberg, peter holt and colleagues in their labs in new york, we had the opportunity to analyze human serum from morbidly obese individuals who had undergone a 10 % weight loss following diet restriction. this weight loss was associated with reduced 77

conclusions and perspectives we have been fortunate to gain important insights into many of the ques- tions posed when i initiated my lab’s research program over a decade ago, as highlighted by the representative studies discussed here. we have iden- tified several mechanisms of tam education, elucidated processes by which tams, neutrophils and other immune cells promote tumorigenesis, discovered that tams have potent protective functions in blocking thera- peutic efficacy, identified the tme as a major mediator of resistance to tam therapies, and have helped to illuminate the interplay between the tme and cancer cells during different stages of the metastatic process. going forward, we are focusing much of our efforts in my lab on under- standing and therapeutically targeting brain malignancies and metastatic disease, both from the perspective of the tme (figure 14). glioblastomas and brain metastases are among the most lethal of cancers, with an aver- age lifespan of a year or less following diagnosis. given this dismal pa- tient prognosis, we became very interested in studying these particular brain malignancies several years ago, and in investigating both the simi- larities and differences between primary and metastatic brain cancers, which may have important implications for understanding differential im- munotherapy efficacy, for example. while we have been able to make sev- eral important insights into the brain tme from our recent studies (pyon- teck et al., 2013; bowman and joyce, 2014; sevenich et al., 2014; bowman et al., 2016; quail et al., 2016; yan et al., 2017), as a field we have much to discover and understand about the unique and particularly challenging microenvironment of brain cancers (quail and joyce, 2017a). 79

gression. although as a field it is widely recognized that there are cancer cell-intrinsic differences in tumor evolution and response to therapy by virtue of different molecular subtypes, appropriate dissection of the dif- ferent tme determinants of therapeutic response is still in its infancy, and largely untapped clinically. going forward, it will therefore be crit- ical to determine the many differences in microenvironmental composi- tion between distinct tumor subtypes in order to achieve a comprehen- sive understanding of tumor biology, including consideration of matrix stiffness, tumor-stromal interactions, and immune cell landscapes. moreover, it will be important to globally address how all aspects of the tme are affected by both standard of care therapy and new investigational therapies across all brain tumors and their respective molecular subtypes. from a practical perspective, we need to engage actively with medicinal chemists to improve drug delivery into the brain; a perennial challenge for all brain-targeted therapies, including those directed against the tme. we need to understand how the generally immunosuppressive environment of the brain is further exacerbated in the context of brain cancers in order to devise therapies to overcome this. finally, if we cannot take a “one size fits all” approach for targeting the tme in different brain malignancies, we will need to determine where the vulnerable points are to attack at a more personalized level. given the current advances being made in the immunotherapy and tme fields, however, we can also expect an exciting and illuminating time ahead for basic research and clinical translation in brain cancers, and for microenvironment biology as a whole. acknowledgements i am deeply grateful and honored by the max cloëtta foundation and board members for recognizing our research with this distinguished prize. i would like to first thank all the past and present members of my lab. without their dedication, creativity, hard work, self-motivation, collegi- ality, enthusiasm and love of science, i would not be here receiving this honor. it has been a true privilege to work with these young scientists, to discuss and debate science with them, to help nurture their passion for discovery, and to see them move on to successful independent positions around the world. 81

ship and a brilliant sense of humor. finally, i am forever grateful to my extraordinary parents, maureen and michael, for their unparalleled love, generosity and support, which was invaluable in fostering confidence, in- dependence, and the joy of discovery. references abboud-jarrous, g., atzmon, r., peretz, t., palermo, c., gadea, b.b., joyce, j.a., and vlo- davsky, i. (2008). cathepsin l is responsible for processing and activation of prohepara- nase through multiple cleavages of a linker segment. journal of biological chemistry 283, 18167–18176. akkari, l., gocheva, v., kester, j.c., hunter, k.e., quick, m.l., sevenich, l., wang, h.w., peters, c., tang, l.h., klimstra, d.s., reinheckel, t., and joyce, j.a. (2014). distinct func- tions of macrophage-derived and cancer cell-derived cathepsin z combine to promote tumor malignancy via interactions with the extracellular matrix. genes and development 28, 2134–2150. akkari, l., gocheva, v., quick, m.l., kester, j.c., spencer, a.k., garfall, a.l., bowman, r.l., and joyce, j.a. (2016). combined deletion of cathepsin protease family members re- veals compensatory mechanisms in cancer. genes and development 30, 220–232. amankulor, n.m., hambardzumyan, d., pyonteck, s.m., becher, o.j., joyce, j.a., and holland, e.c. (2009). sonic hedgehog pathway activation is induced by acute brain injury and regulated by injury-related inflammation. journal of neuroscience 29, 10299–10308. bakst, r.l., xiong, h., chen, c.h., deborde, s., lyubchik, a., zhou, y., he, s., mcna- mara, w., lee, s.y., olson, o.c., leiner, i.m., marcadis, a., keith, j., al-ahmadie, h., katabi, n., gil, z., vakiani, a., joyce, j.a., pamer, e., and wong, r.j. (2017). inflamma- tory monocytes promote perineural invasion via ccl2-mediated recruitment and cathep- sin b expression. cancer research 77, 6400–6414. bell-mcguinn, k.m., garfall, a.l., bogyo, m., hanahan, d., and joyce, j.a. (2007). in- hibition of cysteine cathepsin protease activity enhances chemotherapy regimens by de- creasing tumor growth and invasiveness in a mouse model of multistage cancer. cancer research 67, 7378–7385. bingle, l., brown, n.j., and lewis, c.e. (2002). the role of tumour-associated macrophages in tumour progression: implications for new anticancer therapies. journal of pathology 196, 254–265. 83

hunter, k.e., palermo, c., kester, j.c., simpson, k., li, j.p., tang, l.h., klimstra, d.s., vlodavsky, i., and joyce, j.a. (2014). heparanase promotes lymphangiogenesis and tumor invasion in pancreatic neuroendocrine tumors. oncogene 33, 1799–1808. hussain, s.f., yang, d., suki, d., aldape, k., grimm, e., and heimberger, a.b. (2006). the role of human glioma-infiltrating microglia/macrophages in mediating antitumor im- mune responses. neuro oncology 8, 261–279. jena, p.v., roxbury, d., galassi, t.v., akkari, l., horoszko, c.p., iaea, d.b., budhathoki-up- rety, j., pipalia, n., haka, a.s., harvey, j.d., mittal, j., maxfield, f.r., joyce, j.a. and hel- ler, d.a. (2017). a carbon nanotube optical reporter maps endolysosomal lipid flux. acs nano 11, 10689–10703. joyce, j.a., baruch, a., chehade, k., meyer-morse, n., giraudo, e., tsai, f.y., green- baum, d.c., hager, j.h., bogyo, m., and hanahan, d. (2004). cathepsin cysteine proteases are effectors of invasive growth and angiogenesis during multistage tumorigenesis. can- cer cell 5, 443–453. joyce, j.a. (2005). therapeutic targeting of the tumor microenvironment. cancer cell 7, 513–520. joyce, j.a., freeman, c., meyer-morse, n., parish, c.r., and hanahan, d. (2005). a func- tional heparan sulfate mimetic implicates both heparanase and heparan sulfate in tumor an- giogenesis and invasion in a mouse model of multistage cancer. oncogene 24, 4037–4051. joyce, j.a., and pollard, j.w. (2009). microenvironmental regulation of metastasis. nature reviews cancer 9, 239–252. klemm, f., and joyce, j.a. (2015). microenvironmental regulation of therapeutic response in cancer. trends in cell biology 25, 198–213. komohara, y., ohnishi, k., kuratsu, j., and takeya, m. (2008). possible involvement of the m2 anti-inflammatory macrophage phenotype in growth of human gliomas. journal of pathology 216, 15–24. krajcovic, m., krishna, s., akkari, l., joyce, j.a., and overholtzer, m. (2013). mtor reg- ulates phagosome and entotic vacuole fission. mol biol cell 24, 3736–3745. lavin, y., mortha, a., rahman, a., and merad, m. (2015). regulation of macrophage de- velopment and function in peripheral tissues. nature reviews immunology 15, 731–744. lavin, y., winter, d., blecher-gonen, r., david, e., keren-shaul, h., merad, m., jung, s., and amit, i. (2014). tissue-resident macrophage enhancer landscapes are shaped by the local microenvironment. cell 159, 1312–1326. 85

pyonteck, s.m., akkari, l., schuhmacher, a.j., bowman, r.l., sevenich, l., quail, d.f., olson, o.c., quick, m.l., huse, j.t., teijeiro, v., setty, m., leslie, c., oei, y., pedraza, a., zhang, j., brennan, c.w., sutton, j.c., holland, e.c., daniel, d. and joyce, j.a. (2013). csf-1r inhibition alters macrophage polarization and blocks glioma progression. nature medicine 19, 1264–1272. pyonteck, s.m., gadea, b.b., wang, h.w., gocheva, v., hunter, k.e., tang, l.h., and joyce, j.a. (2012). deficiency of the macrophage growth factor csf-1 disrupts pancreatic neuroendocrine tumor development. oncogene 31, 1459–1467. quail, d.f., bowman, r.l., akkari, l., quick, m.l., schuhmacher, a.j., huse, j.t., hol- land, e.c., sutton, j.c., and joyce, j.a. (2016). the tumor microenvironment underlies ac- quired resistance to csf-1r inhibition in gliomas. science 352, aad3018. quail, d.f., and joyce, j.a. (2013). microenvironmental regulation of tumor progression and metastasis. nature medicine 19, 1423–1437. quail, d.f., and joyce, j.a. (2017a). the microenvironmental landscape of brain tumors. cancer cell 31, 326–341. quail, d.f., and joyce, j.a. (2017b). molecular pathways: deciphering mechanisms of re- sistance to macrophage-targeted therapies. clinical cancer research 23, 876–884. quail, d.f., and joyce, j.a. (2017c). multifaceted effects of the microenvironment on tu- mour progression. nature reviews cancer https://www.nature.com/nrc/posters/microen- vironment/micro_poster.pdf quail, d.f., olson, o.c., bhardwaj, p., walsh, l.a., akkari, l., quick, m.l., chen, i.c., wendel, n., ben-chetrit, n., walker, j., holt, p.r., dannenberg, a.j., and joyce, j.a. (2017). obesity alters the lung myeloid cell landscape to enhance breast cancer metastasis through il5 and gm-csf. nature cell biology 19, 974–987. ruffell, b., and coussens, l.m. (2015). macrophages and therapeutic resistance in cancer. cancer cell 27, 462–472. ryder, m., gild, m., hohl, t.m., pamer, e., knauf, j., ghossein, r., joyce, j.a., and fagin, j.a. (2013). genetic and pharmacological targeting of csf-1/csf-1r inhibits tumor-as- sociated macrophages and impairs braf-induced thyroid cancer progression. plos one 8, e54302. sadaghiani, a.m., verhelst, s.h., gocheva, v., hill, k., majerova, e., stinson, s., joyce, j.a., and bogyo, m. (2007). design, synthesis, and evaluation of in vivo potency and se- lectivity of epoxysuccinyl-based inhibitors of papain-family cysteine proteases. chemis- try and biology 14, 499–511. 87

89

stiftungsrat im jahr 2018 setzt sich der stiftungsrat wie folgt zusammen: prof. dr. fritjof helmchen universität zürich institut für hirnforschung präsident des stiftungs- rates und des medi- zinischen ausschusses dr. hans bollmann** rechtsanwalt, zürich prof. dr. hugues abriel* prof. dr. gabriela finke* universität bern institut biochemie und molekularmedizin universitäts-kinderspital beider basel universität basel, departement biomedizin vizepräsident des stiftungsrates mitglied des stiftungsrates mitglied des stiftungsrates prof. dr. adriano fontana* universität zürich mitglied des stiftungsrates peter ochsner prof. dr. walter reith* institut für experimen- telle immunologie banque cantonale vaudoise, verwaltungs- ratsmitglied universität genf departement pathologie und immunologie mitglied des stiftungsrates mitglied des stiftungsrates hans georg syz-witmer** maerki baumann & co. ag, zürich mitglied des stiftungsrates prof. dr. bernard thorens* universität lausanne mitglied des stiftungsrates zentrum für integrative genforschung martin wipfli rechtsanwalt, zürich mitglied des stiftungsrates *mitglieder des medizinischen ausschusses **mitglieder des vermögensverwaltungs-ausschusses 91

heft nr. 6: preisverleihung 1978 festvortrag des preisträgers dr. susumu tonegawa: «cloning of immunoglobulin genes» heft nr. 7: preisverleihung 1979 festvorträge der beiden preisträger professor dr. theodor koller: «die strukturelle organisation des genetischen apparates höherer organismen» professor dr. jean-pierre kraehenbuehl: «hormonal control of the differentiation of the mammary gland» heft nr. 8: preisverleihung 1980 festvorträge der beiden preisträger professor dr. edward w. flückiger: «neue aspekte der mutterkorn-pharmakologie» pd dr. albert burger: «ein vierteljahrhundert nach entdeckung des aktivsten schilddrüsenhormons (trijodothyronin)» heft nr. 9: preisverleihung 1981 festvorträge der beiden preisträger professor dr. rolf zinkernagel: «‹selbst›-erkennung in der immunologie» professor dr. peter a. cerutti: «a role for active oxygen species in human carcinogenesis?» 93

heft nr. 14: preisverleihung 1986 festvorträge der beiden preisträger professor dr. ueli schibler: «mechanismen der gewebespezifischen regulation von genen» professor dr. walter schaffner: «enhancer-sequenzen und die regulation der gen-transkription» heft nr. 15 preisverleihung 1987 festvorträge der beiden preisträger professor dr. jacques louis: «rôle des lymphocytes t spécifiques sur l’évolution des lésions induites par leishmania major, un parasite vivant dans les macrophages de leur hôte» professor dr. joachim h. seelig: «magnetische resonanz–ein ‹magnetisches auge› zur unter- suchung der lebensvorgänge in vivo» heft nr. 16: preisverleihung 1988 festvorträge der beiden preisträger professor dr. jean-dominique vassalli: «protéolyse et migrations cellulaires: multiples facettes du contrôle d’une cascade enzymatique» pd dr. hans hengartner: «über die immunologische toleranz» heft nr. 17: preisverleihung 1989 festvorträge der beiden preisträger professor dr. heini murer: «parathormon: auf dem weg zum verständnis seiner wirkung im proximalen tubulus» dr. hugh robson macdonald: «selection of the t cell antigen receptor repertoire during development» 95

professor dr. michel aguet: «neue erkenntnisse zur biologie und molekularbiologie der interferone» heft nr. 22: preisverleihung 1994 festvorträge der beiden preisträger professor dr. rer. nat. hans rudolf brenner: «die regulation der expression synaptischer transmitter- rezeptoren während der synapsenbildung» daniel pablo lew, professeur ordinaire de médecine: «signal transduction and ion channels involved in the activation of human neutrophils» heft nr. 23: preisverleihung 1995 festvorträge der beiden preisträger professor dr. jürg reichen: «die sinusoidale endothelzelle und leberfunktion» dr. george thomas jr.: «the p70s6k signal transduction pathway, s6, phosphorylation and translational control» heft nr. 24: preisverleihung 1996 festvorträge der beiden preisträger dr. lukas c. kühn: «molekulare mechanismen zur steuerung des eisenstoffwechsels» professor dr. peter sonderegger: «wegweiser- und sensormoleküle beim gezielten wachstum der nervenzellausläufer» 97

heft nr. 29: preisverleihung 2001 festbeiträge der beiden preisträger professor dr. isabel roditi: «the surface coat of african trypanosomes» dr. thierry calandra: «innate immune responses to bacterial infections: a paradigm for exploring the pathogenesis of septic shock» heft nr. 30: preisverleihung 2002 festbeiträge der beiden preisträger professor dr. bernard thorens: «impaired glucose sensing as initiator of metabolic dysfunctions» professor dr. andrea superti-furga: «molecular pathology of skeletal development» heft nr. 31: preisverleihung 2003 festbeiträge der beiden preisträger professor dr. michael nip hall: «tor signalling: from bench to bedside» pd dr. bernhard moser: «chemokines: role in immune cell traffic» heft nr. 32: preisverleihung 2004 festbeiträge der beiden preisträger professor dr. amalio telenti: «adaption, co-evolution, and human susceptibility to hiv-1 infection» professor dr. radek c. skoda: «the control of normal and aberrant megakaryopoiesis by thrombopoietin and its receptor, c-mpl» 99

heft nr. 37: preisverleihung 2009 festbeiträge der beiden preisträger professor dr. margot thome-miazza: «molecular mechanisms controlling lymphocyte proliferation and survival» professor dr. walter reith: «regulation of antigen presentation in the immune system» heft nr. 38: preisverleihung 2010 festbeiträge der beiden preisträger professor dr. christan lüscher: «sucht: die dunkle seite des lernens» professor dr. burkhard becher: «cytokine networks: the language of the immune system» heft nr. 39: preisverleihung 2011 festbeitrag der preisträgerin professorin dr. petra s. hüppi: «from cortex to classroom» heft nr. 40: preisverleihung 2012 festbeitrag des preisträgers professor dr. olaf blanke: «brain mechanisms of bodily self-consciousness and subjectivity: review and outlook» heft nr. 41: preisverleihung 2013 festbeitrag der preisträger prof. dr. andreas papassotiropoulos und prof. dr. dominique j.- f. de quervain «genetics of human memory; from gene hunting to drug discovery» 101

ehrentafel der preisträger prof. dr. jean-pierre kraehenbuehl pd dr. albert burger prof. dr. peter a. cerutti 1974 dr. urs a. meyer 1975 pd dr. hans bürgi 1976 dr. rui de sousa 1977 prof. dr. franz oesch 1978 dr. susumu tonegawa 1979 prof. dr. theodor koller 1980 prof. dr. edward w. flückiger 1981 prof. dr. rolf m. zinkernagel 1982 pd dr. jürgen zapf 1983 prof. dr. peter böhlen 1984 prof. dr. heidi diggelmann prof. dr. jean-françois borel 1985 prof. dr. hans thoenen dr. roberto montesano 1986 prof. dr. walter schaffner 1987 prof. dr. jacques louis 1988 prof. dr. jean-dominique vassalli 1989 prof. dr. heini murer dr. hugh robson macdonald prof. dr. joachim h. seelig pd dr. jean-michel dayer pd dr. claes b. wollheim pd dr. hans hengartner prof. dr. ueli schibler 103

ehrentafel der preisträger prof. dr. dominique muller prof. dr. michael o. hengartner 2003 prof. dr. michael nip hall pd dr. bernhard moser 2004 prof. dr. amalio telenti prof. dr. radek c. skoda 2005 prof. dr. urs emanuel albrecht 2006 prof. dr. adrian merlo 2007 prof. dr. françois mach prof. dr. nouria hernandez 2008 prof. dr. darius moradpour 2009 prof. dr. margot thome-miazza 2010 prof. dr. christian lüscher 2011 prof. dr. petra s. hüppi 2012 prof. dr. olaf blanke 2013 prof. dr. andreas papassotiropoulos prof. dr. dominique j.-f. de quervain 2014 prof. dr. marc y. donath 2015 prof. dr. dominique soldati-favre 2016 prof. dr. michel gilliet prof. dr. andreas lüthi prof. dr. henrik kaessmann prof. dr. sabine werner prof. dr. walter reith prof. dr. burkhard becher prof. dr. fritjof helmchen 105