 And what I'm going to talk about today, you know, COVID, as Matt already mentioned, has kept us all out of the field, which has been difficult for some of us, but it has got me getting back to my Russ life cycles and sort of life history strategies as a way to try to explain some of the success, I view success in rust punch, I just in terms of species. And so what I'm going to deviate a bit from the prior talks, because there's not a lot of experimental work here. So I'm just going to delve right into the realm of mass speculation. But let me take you on a ride here about what I think is going on here in the rust life cycle and how this translates into success for this lineage. Now I won't be labor this point because I'm sure most of you are familiar with rust fungi. But of course, biologically speaking, this is a really fascinating group, at least to me, there's good practical reasons for studying these. But they these this group of fungi has many characteristics has evolved a lot of sort of almost unique or unusual features that we don't really see elsewhere in the fungi. And this includes, of course, strict obligate by a trophism to the extent where these can't be manipulated in pure culture, for example. As far as we know, some species do have the largest genomes and fungi were getting in the range of gigs here for some of these things. Alternation of generations is of course very rare in fungi where you have different spore of thali and gametathali that are produced at different time points in the life cycle. And most of these species also alternate on different hosts, a term heteroetius, this is term we use for that. So they produce the gametathalis on one host and the spore thalis on another. And then in addition to this, we have these four pre programmed stages or leading to five different types of developmental spores that all function differently and look quite different from each other into this into this mix. And then because of some of these complexities and other reasons, one of the difficulties with working with this group is that, of course, our complete life cycle data are missing for many, many species. And so trying to make broad inferences about this group when you're lacking these types of data can be somewhat prohibitive. Just to remind you, our rust fungi, what we termed rust fungi belong to a single order, pexinealis. And we're talking about 8000 species of known described rust fungi here. And at least when I did these calculations several years ago, this may have changed a little bit due to an influx of new species descriptions lately. But five or 10 years ago, this did comprise about 25% or a quarter of our known obsidium mycota. So it's a huge group of fungi. In fact, our second largest order after Gary Kaley's in terms of species, and certainly the largest group of plant pathogens. And so this leads us to kind of what we think of as the central conundrum for your denologists, which is trying to reconcile these really complex life cycles with the obvious success in terms of species of the group. And I'm a real fan of John Bonner, and I'm sorry, Don, to bring up Princeton here at a Harvard symposium. But Bonner did a lot of work with life cycles and a lot of thinking about life cycles. And in Bonner's view, you know, an organism, we can't think of organisms, usually we think of them statically, you know, as the adult in plants or animals, one stage. But Bonner, you know, we need to consider the entire life cycle when we talk about organisms and certainly an organism, an individual, is sort of the unfolding or the expression of a genome over ontology, right? It's not a single time point. And if we look at life history trait through life cycles across the tree of life, there are different patterns that repeat themselves. And of course, to me, the most interesting of these is that pattern of complexity, the complex life cycles, and how you define a complex life cycle really depends on what group you're working with. Parasitologists like to think of these as parasites that alternate, just require alternation between different hosts. It's been broadened out a bit to encompass any organism that has some abrupt autogenic change that encompasses morphology, physiology, and behavior somewhere during its development. But whatever definition you use, Russ Fungi certainly fit in this. But we don't have a lot of really good studies still about complex life cycles and how these may have evolved despite the fact that they are quite prevalent, again, throughout the tree of life, or at least in eukaryotic life. Most of the work is done with animal pathogens such as those things in the ampicomplexa or parasitic ailments. And here's a typical Plasmodium life cycle here. But even when you start talking about complex life cycles, there's grades of complexity there. And the Russ Fungi fit into what's usually the rarest type that has developed, which is something that we've referred to as the haplodiplonic cycle, where you're talking about multicellular, diploid, and diploid, for Fungi, of course, but multicellular, gimme the thali, and spore a thali. So this is actually quite rare, the development of this. Again, Russ Fungi are the best expression of this type of life cycle. And I'm going to assume that most of you remember something about your basic Russ life cycle. So I won't be labor the point here, but just again, understand that this is a template for a temperate heterogeneous macrocyclic Russ. So it's expressing off five spore stages. And these different gimme the thali and spore a thali are produced on different hosts. There's, at least for a temperate Russ, a very seasonal component. So we see usually the city of spore production in the spring, and tealia spores being the overwintering crop of gul. And then what's interesting to me, one of the interesting things about this life cycle is we certainly have the production of mea spores here, but we have this increase of mea spore stage here. So the sperm agonia, we normally think of the city of Mycete, you undergo myosas, you produce these four haploid gametes, and then they've got to go out and find mates. But with the Russ, there's an amplification of those gametes in this sperm agonial stage. And that means that basically this product of a myotic event has the opportunity to mate with multiple, multiple individuals, theoretically. So we have this amplification here. And then the other important amplification stage is with these mito spores. So the urodinia spores themselves, in a heteroaceous species, this is the only spore stage that is capable of re-infecting the same host from which it was produced. And so in epidemiology, this becomes extraordinarily important. And most of what we know about Russ at the molecular level is work done to look at the cues and expression in urodinia spores. If urodinia spore infection, if a single urodinia spore cycle gets started early enough in the season, one spore can turn into millions of spores. So what's interesting here is you have amplification of any new gametic products or any new fertilization. But you also have the generation of lots of mitotic mutation, which can lead to interesting cases here where you might have adaptation also being driven in this stage from this increase in the mitotic cycle. Okay, so that's, as I said, is your template Russ cycle, life cycle. But where we've worked these things out, we know that that is not the only life cycle that's possible in the Russ fungi. Certainly, there are macrocyclic, if we sort of flatten this life cycle out, we have, as I've just shown you the heteroseous ones, but we do have odysseous ones, which produce all these stages on the same host. And then we have lots of variations of this. We have, for example, demicyclic Russ, which skip this mitotic reinforcement stage so they don't have the production of urodinia spores. We have microcyclic Russ fungi, which skip both of these mitotically produced spore stages and just do something that we would think of as more of a traditional Basidiomyc type of pattern, where you have production of telia, which terminate to produce you Basidiom, Basidiospores, and those made either through spermagonia or not in complete the cycle. And then we even have a variation of the microcyclic called the endocyclic. If you look through the literature, you'll see that people used to call these cyclical through Isha, but in fact, this Isham is functioning like a telium. In other words, it's producing telius spores. So myosis is occurring here. So it's really just a variation of microcyclic where the telia and the telius spores morphologically resemble Isha but function for overwintering and myosis. So in urodinology, there are all sorts of debates about the course of evolution of these different life cycles. There's one train of thought that the microcyclic stage came first and then these other spore stages were added. There's another train of thought where the ground plan was this heterogeneous macrocyclic cycle and that these others are derived. And so as y'all know, when you're lacking fossil evidence and experimental approaches, the best way to sort of test these different hypotheses is to just do an evolutionary reconstruction or phylogenetic reconstruction. And this is very old work that we did just to resolve the pachyneomycotina, the suborder to which the rest belong to. And here we were able to resolve actually the sister order to the pachynealis. And these are a very small order fungi, interesting little guys called the platyglomialis. And what's really nice about platyglomialis, although there's few species, there are two genera, yoconarsomniola, that are parasites, obligate parasites, more or less a very, very difficult to grow and culture of ferns. And these have more or less a typical, in comparison to a rust fungus, a microcyclic type of life cycle, but a typical basidiomycete life cycle where they produce basidiospores, they undergo mating, and then they produce a fertile layer in these little fruiting bodies here of these probacidia that look very much like ateliospore. And these probacidia actually function like ateliospore. So under the red cues, these will undergo myosis, terminate to produce basidia and start the life cycle all over again. So certainly the sister group were capable of parasitism, form these probacidia and had something analogous to a microcyclic life cycle. But the next step would be, of course, to see what's actually going on down at the base of the rust fungi. And so this is, again, very early attempt to resolve the rust tree of life, using exemplars from all across the tree of life. And this area up here in blue is largely unresolved, but we did manage to achieve some decent resolution down here at the base. And the first thing you'll notice down here at the base is this really interesting fungus, Roger Pedersonia Toria, which appears to be our earliest extant rust, at least the earliest extant rust that we know of. And this thing is only found on a very relictual host that's confined to a couple of populations in the Pacific Northwest right now, was probably much more broadly distributed in the past. And as far as we know, Roger Pedersonia Toria only produces the gametothalus, never found a sporthalic state for this one. We look at the next group here, which includes things like our coffee rust pathogen, Himelia vastatrix, and this entire group shaded in yellow here, this is a group of fungi for which we only know the opposite, we only know the sporthalus. We have no idea what the gametothalus is, or if it exists. These do occasionally form basidiospores under the right environmental conditions, and those basidiospores do not reinfect the host that produce them. So we're not looking at something audacious, we're looking at something that's persisting as far as we know in this sporthalic state. And then the sister group to those Himelias is this family microinicariaceae. It's not a very speciose family, but the important thing here is that several of these species are known to have complete heterogeneous macrocyclic life cycles with an alternation of hosts. And these alternates, in fact, one of these collections Matt sent to me from Chile, but they alternate on notofagus and podocarp, so they seem confined primarily to the southern hemisphere. So if we put this pattern, including our outgroups all together, what we're inferring from these data is that the ancestor to the rust really most likely was host alternating, and it was likely heterogeneous. So it produced separate gametothalus and sporthalus, and those were probably produced on the different hosts. And we may be looking at here at extinction of alternate hosts in some of these lineages to explain why we've never been able to find the other Dallas type. Obligate bio-trophy was probably also a feature of these, it certainly seems to be a feature of Roger Petersonia. So this brings us back to the central paradox. If that is the basal plan, if that's the life history strategy, how do we explain this tremendous diversity in terms of species? If you define success for lineages as speciation, then this is a very successful lineage. And again, most of the thinking about these complex life cycles has been done with the animal parasites, but they've not been able to effectively resolve this question. There's an inherent bias, almost an axiom in parasitology that parasites are sort of an endpoint in evolution. It's a secondary strategy and it's not a successful strategy, yet we see the opposite here. Now in parasitology, they've tried to explain success of complex life cycles in terms usually of equating mass or growth rate with success. It doesn't make a lot of sense with fungi, certainly doesn't make much sense with rest fungi to make these kinds of equations because obviously if you're going to grow to a size where you consume your host before it can reproduce, then you're not going to be successful at speciating. And the other idea that's often invoked is this idea of upward incorporation, so it's a jump in trophic levels. The idea here being if you're a parasite of a host that is preyed on by a predator, then it makes sense for you to jump a trophic level into the predator so that you in turn aren't preyed upon. So this is sort of a type of host jumping whereas this would be maximizing fitness to your host, so a sort of co-evolutionary invocation there. And indeed in plant pathology, this is the way that we tend to think about pathogen, plant pathogen evolution either as being a function of host jumps or host switching shown here in green or as a function of co-evolution with your host such as the red queen hypothesis, etc. So in your denology, of course, we have batted both of these ideas around multiple, multiple times and reached no successful conclusion about which process really explains evolution in the rust. And so this is where I turned to my postdoc at the time, Andy Wilson, and got him to do some reconciliation analyses to test both of these hypotheses, co-evolution and host jumping within a rust framework. And we did something a little bit different here. We selected, we wanted to select from across the rust tree of life, but also select primarily known heteroecious macrocyclic fungi so that we could dissect out the both hosts. So the gametothalus stage from the sporethalus stage. And of course we made our phylogenies for the rust, for the sporethalus host and for the gametothalus host and then tried to reconcile these individually. And there's a lot of really interesting things going on here with this work. But the one thing that I want to draw your attention to is that there is a very, very strong signal of co-diversification when we reconcile the gametothalus host with the rust phylogeny. But when we try to reconcile the sporethalus host, we just don't see a strong signal at all of co-diversification there. And so the model we came up with, what we think is happening here, is that actually both of these selective pressures are in play, these forces are in play, but they're on play at different parts of the life cycle. So we have biological speciation, which we can see in these co-diversification or specialization, excuse me, acting on the gametothalus. And we have biogenic radiation or series of host jumps actually acting on the sporethalus. And it makes sense biologically because we know that this critical stage of fertilization, this has a very temporal and spatial component in rust biology. And so this has to have some reflection in host specificity here to have successful fertilizations. Any mutations that would allow a broader use of host here would have to be compensated for at the same time and place in another compatible individual. Whereas, of course, we have this induction of the mitotic massive sporulating increased stage here that allows immediate mutations that allow a jump to another host to become reinforced and reflect later in the life cycle. So being mindful of time here, this is sort of our game plan. So how do you get these other derived life cycles? Because we've seen that's not the only plan or strategy in the rust fungi. We have these other life history strategies that have evolved. And to look at this, we go back to Trenjal, who was a great erudinologist at the turn of the last century, who thought a lot about life cycles in rust. And Trenjal had made an observation, what he called correlated species. And that is that in several instances he could find cases where two different rust species infected the same host. But when they did this, one of those species was utilizing that host just for this sporothalus. And the gametothalus would be produced on a different host. Whereas the other host was microcyclic, reduced microcyclic or endocyclic in form, and just went through this reduced life cycle on that what would be considered the sporothalus host for the correlated hetero species. And so to test whether we could actually see at a granular level, whether this type of correlated species patterns could be found in the rust, this is just a visual representation of what we were looking at. We actually chose the genus Trenjalia aptly enough. And here's an example of one of Trenjal's correlated species where you have prunispinosa, which is host alternating on a renunculaceae, and then Trenjalia fusca, which is actually using the same host, but to produce atelial spores. And this is work I did with Marcus Scholler, where we actually found, of course, multiple instances of these correlated species pairs within this one genus Trenjalia. And in fact, we find other instances of this throughout the rust tree. So the model Marcus and I proposed then was sort of, we have these normal forces, actually not so normal, but we have co-evolution and some species jumps occurring in time along this axis for speciation. But we also show that this is also the starting point for other types of speciation in this direction, for instance, forming the microcyclic species. And we hypothesize that you might be able to find some evidence if this model was true, then you would expect to see also some cases where this split had occurred. And the original sporethalus host was the only host that these things would be completing their life cycle on. And in order to test that hypothesis, we've really needed a much better resolved tree of life. And this is something that was just completed this year after about 15 years of trying to select good exemplars from across the rust tree of life. These analyses were done by former postdoc Alistair McTaggart. But basically, we were able to, with some confidence, resolve this tree of life, at least in the deeper nodes for the rust fundae, looking across all rust. And we made a deliberate attempt in this analysis to both include type species for different genera, just to resolve a lot of nomenclatural problems, but also to include a variety of different life cycle types within this tree. And to see if we can see any evidence for that opposite for speciation on the sporethalus host, I want to dive in right here to this lineage, the Ravanelli NE. So this is one of our sub orders. It's been a little bit tricky to resolve, especially the Ravanelli AC, for reasons that I don't have time to get into. But what's interesting about the Ravanelli NE, we've resolved four families here. And the two earliest diverging families are heteromacro. So there are typical, and we've already seen that with transgelia, where we've got species in here that are host alternating and macrocyclic. But then we have this revised family for Copseraceae. And if we look at the actual species now that we can confidently assign to the Copseraceae, all of these are only known from Euridinia, Euridinia and Telia. Occasionally produced besidio spores under the right conditions. Those besidio spores do not reinfect the original host. And then we get into the Ravanelli AC. And what's interesting here is these are all the leases species we know full life cycles for are actually audacious and macrocyclic. So they produce all five spores stages on the same host. So what do I think is happening here? Again, looking back at a transgelia-like model where we have this temperate macrocyclic rust and a rupture that separates whether it's extinction or some sort of ecological disruption, but it separates the sporethalus host and the gametothalus host from the sporethalus host. And so what happens to these lineages through time? While we same with transgelia, of course, we can have the evolution of microcyclic or endocyclic species. But now we think we see evidence in the Ravanelli AC of this also occurring here, where we do have evolution of species from the sporethalus stage, such as we've just seen in the Copseraceae. And that's given enough evolutionary time. What we think of is these nonfunctional besidio spores can eventually become adapted to reinfect this original host. And once that hump is overcome, we see immediate proliferation recapitulation of these other spore states. And this can best be visualized if we just look at the data for Ravanelli, which has been extraordinarily difficult to resolve. And this is a splits tree view. And if you look at this thing, you see that there is no simple bifurcating pattern of the evolution here. This is a massive radiation of this particular Ravanelli lineage. And we think, or at least hypothesize that it's finally through time the ability of these spore stages to be recovered that drove this particular radiation. Now, the big question now is, do we see the same thing here? Do we see any evidence of radiation or continued speciation from these microcyclic forms? And in short, the microcyclic endocyclic rust are basically scattered a couple endpoints here and lots and lots of them here in the crown radiation of rust. But we don't see evolution of lineages of microcyclic endocyclic rust. And this brings us back to this Urodinius spore stage and the importance of this stage in driving a lot of this evolution in this particular group. So the model now, we believe, I believe that we've seen this happening again in modern pairs. This, we do see some evolution or some evidence for this type of pathway happening as well, deeper in the rust tree of life. So we have radiations occurring in different directions. And then of course, we also have the radiations from the host jumps just from any individual here. And so where do we go from here? How do we test these ideas a little bit further? I'm out of time, but I will say this is looking with Sebastian Du Plessis. Genomes is obviously the next place we can go again with absence of good fossil record data or experimental methods to test this. And when we, there's very few genomic data for rust fungi and it's mostly confined to the crown of the rusts and even less transcriptomic data. But certainly we see evidence that there are common effectors when we look, parse out all the transcriptomic data from different rusts. Effectors are the thing that most people concentrate on. There's thousands of effectors in rust genomes. But when we look at the actual expression of them for the few species that we actually have expression data for different parts of the life cycle, of course, the expression patterns are different. And there are certain families of effectors expressed here on the spore thalus host and different families expressed here on the gameta thalus host. And so we were beginning to get a picture where we have this massive genomic toolkit, but there is specialization even in the effector trans, effector expression patterns of which of these are utilized in different points in the life cycle. So lots of funding through the years to help support these, these studies. And of course, I would be remiss if I didn't mention the members of my lab, even though the vast majority of these don't work on rust, they've still been instrumental in helping my thinking and keeping me on my toes through the years. And so I will end there. Thank you for your attention.