Welcome to the next section on functional vascular development assays. Here we will discuss a number of different functional vascular development assays that are being used to test the range of putative vascular disruptor compounds. That were selected to further understand the adverse outcome pathway for embryonic vascular development, and chemical specific effects on that pathway. The first assay that we will discuss, is a lower throughput human cell based tubulogenesis assay. Here, fiberglass and an endothelial cells are cocultured with stimulatory media. So media that contains growth factors that stimulate vascular tube formulation, hence the name a tubulogenesis assay. Here you can see a picture of the negative control, where there's no vessel formation. And the positive control, where there's quite a high degree of vessel formation and capillary plexus formation. We've tested the same 36 ToxCast compounds that were mentioned earlier in the lecture. That have varying signatures against the putative vascular disruptor compound adverse outcome pathway. And we've looked for concentration dependent response effects on that vessel formation. We're hoping that based on the different patterns of activities from the different chemicals against the different ToxCast assays. Paired with their results in a functional vessel developmental assay such as this one. Will allow us to identify what are those critical assay targets that may be necessary or sufficient to drive disruption of vessel formation. And what are those less critical assay targets that may not be sufficient to drive disruption of vascular development. This slide shows some preliminary results for two compounds that were mentioned earlier. One is Pyridaben the insecticide that is a known developmental toxicant in animals. You can see in the middle the virtual tissue model that predicts total disruption of capillary plexus formation. And on the right hand side, you can see the results from the human cell based vascular tube formation assay. The important thing to notice here, is that the AC50, or the concentration at which we saw 50% inhibition of tube formation. Is three orders of magnitude lower than the chemical concentration where we see cytotoxicity, or where cells start dying. This means that this compound in this particular assay system shows a very specific vascular disruptive effect. That is not simply based on general cell stress or cell death. In contrast the compound along the bottom Imazamox which is also a developmental toxicant. But not predicted to act via a vascular disruptive mechanism, had little to no activity against any of the molecular targets in the signature. Formed a well connected capillary plexus in the virtual tissue model, and was tested up to very high concentrations in the human cell based vascular tube formation assay. And showed no inhibition of tube formation and in fact showed a small stimulatory effect on vascular tube formation. Now we'll move into our next functional vascular assay. And this utilizes what's known as a small model organism platform. The small model organism that we're using here is called a zebrafish or Danio rerio. Zebrafish is a fantastic small model organism platform for understanding how chemicals might actually have effects in a whole organism. It's a biologically complex system, with an enormous amount of genetic conservation with humans. In fact 75% of the zebrafish genes have human homologues, and this certainly extends to the vascular developmental genes. A really nice thing about this small model organism model is that the embryo is transparent as it develops. Which makes it very amenable to quantitative imaging. Also, the technology has developed to the point where we have transgenic reporter lines. Which means that you can tag a specific gene, for example the VEGF receptor, which is expressed by endothelial cells with a green fluorescent protein. This allows us to map the developing vasculature across space and time. Because all of the blood vessels in the zebrafish are glowing bright green, as you can see in the middle panel here. That attribute allows for high resolution imaging and quantification of the developing vascular system, as you can see in the bottom panel. Zebrafish are also quite small and very amendable to automated high throughput screening platforms. This work was done by Tamara Tal at the US Environmental Protection Agency. And here you can see the screening strategy for the zebrafish assay that looks at vascular development in an embryonic context. First we define the space of overt toxicity for chemicals because we don't want to just kill the fish. We want to look at chemical concentrations that are lower than those that cause overt toxicity in the fish. Then we screen multiple concentrations of the chemical as the zebrafish embryo is developing. And we quantify the degree of vascular toxicity that results based on chemical exposure during embryo genesis. Here we can see the impaired angiogenesis in the zebrafish that have been exposed to another positive control compound. PTK 787 is a small molecule inhibitor of the VEGF receptor 2, the canonical receptor for the vascular endothelial growth factor pathway. You can see in the graph on the top right, as you increase the concentration of PTK 787 that the zebrafish embryo is exposed to, you get a concentration dependent decrease in intersomitic vessel link. So those are the blood vessels that run along the entire trunk of the zebrafish. And as you increase chemical exposure, those blood vessels eventually do not develop at all. And this is occurring in the absence of overt toxicity as measured by body length. This was published in Reproductive Toxicology in 2014. Another interesting way of examining the longer term consequences of early developmental exposures to vascular-disrupting chemicals, is by looking at long term effects on survival. So we can look at the morphology at five days post fertilization after exposure to the different concentrations of that reference anti angiogenic compound. And then look at how that actually impacts the survival rate of the fish after they are grown out to 30 days. And you can see that even in the absence of specific vascular malformations, such as at the very low concentration of PTK 787, 0.07 micromolar. We weren't able to necessarily see specific effects on vascular development just from looking at the morphology. But we do see a decrease in survival at 30 days across the entire population of fish. So this gives us another metric to understand how vascular disruptive compounds may be affecting the survival or the functionality of specific organisms. To summarize the results that we've seen in all of our functional assays, our human cell based tubulogenesis assay, and our zebrafish vascular development quantification assays across all 36 chemicals. We can see in this image here that if a compound has a red box around it, it was a hit in at least one of those assays. Meaning that it disrupted vascular development either in the zebrafish model or the human cell base assay or both. The green indicates that there has not been any vascular developmental disruption observed for that compound so far, and this work is still underway. So we may yet discover some vascular disruptive effects for compounds where we have yet to see an effect so far. Some interesting findings that have come out of the screening of all of these chemicals are helping us understand which of these particular assay targets represented by the different slices in the ToxPi. May be necessary or sufficient to cause disruption of vascular development, and it's potentially helping us identify data gaps. What are the molecular and cellular signaling targets where we don't have any assay information in the ToxCast screening program yet? And where do we need to further develop our technology to make sure that we have all of the key molecular and cellular signaling pathways adequately covered to be able to predict vascular toxicity in a developing organism. In summary, we've seen that quantitative adverse outcome pathways using high throughput screening dose response data allow hypothesis generation. Modeling and testing of molecular initiating events, and cellular interactions that may lead to toxicities on an organ level, an organism level, or a population level. We've examined that here with a case study on embryonic vascular development and disruption of early embryonic vascular development, leading potentially to developmental toxicity. The validation of such an adverse outcome pathway is facilitated via orthogonal assays, small model organisms such as zebrafish, and other scientifically relevant information. Here we have generated and begun to evaluate this vascular developmental screening toolbox utilizing phenotypic endpoints. In such small modeled organisms and orthogonal assays that allow us to look functionally at vascular development. And chemical effects on disruption of that vascular development. Our preliminary data on a range of chemicals with a range of different signatures, against that putative vascular disruptor compound AOP. Shows that the chemical rankings are overall well correlated among the predictive signature that's based on that AOP, and informed by the ToxCast assays. As well as zebrafish results and the in vitro human cell based tubulogenesis assays. Validated adverse outcome pathways will enable chemical prioritization, understanding out of the thousands and thousands of chemicals in our environment. Which of those should be most highly prioritized for testing, screening, and further examination. Validated AOPs also give us mechanistic understanding, why certain compounds might be causing developmental toxicity, and how. Is it via disruption of embryonic vascular development, or some other mechanism? And ultimately, once we have validated AOPs with enough coverage of the biological space and assays. Or testing systems that map to each part of that AOP, this will really facilitate high throughput risk assessments. Moving forward, we would like to leverage these diverse data streams and really improve and refine the developmental vascular toxicity signature. Some key questions are, are all of these assays relevant? And how should they be weighted? Currently, there is no weighting schema, all of the targets are assumed to be equal. But our preliminary results from the chemical screen are giving us an indication that certain targets such as the VEGFR2 receptor, or the HIF1 Alpha Hypoxic Signaling Pathway. May be more important than other targets in driving vascular disruption during embryonic growth. Also what are we missing in terms of the biology, what non ToxCast targets are needed? And where should we be developing assays and tests to query that biological space, and make sure that we have adequate coverage of embryonic vascular development? Along those lines, are there other ToxCast assays that should be included in this signature? How can we further test the signature? What other orthogonal assays should be used to test the predictions for putative vascular disruptor compounds? How do we incorporate uncertainty and variability, and ultimately how can adverse outcome pathways be put into practice in risk assessment? Thank you for listening to the lecture on modeling hazard and organ manifestation. With a particular emphasis on the case study of embryonic vascular development. We hope that after listening to this lecture, you now understand the concept of an adverse outcome pathway. Its key components and the mapping of high throughput screening assays, that you can now explain the toxicological prioritization index, or ToxPi and how it is used to prioritize chemicals. That you can define virtual tissues, and in silico testing platforms, and discuss using small model organisms and functional assays to test the predictions. There are a huge, huge number of collaborators involved in this effort to understand and tease apart the mechanisms by which chemicals might be inducing vascular toxicity in developing embryos. And therefore present a risk to humans and the environment for developmental toxicity. I would like to acknowledge the contributions of all of these individuals and thank you very much. [MUSIC]
