Coupled Organ Chips: Read Article 07 Vernetti, L, et al. Functional Coupling of Human Microphysiology Systems: Intestine, Liver, Kidney Proximal Tubule, Blood-Brain Barrier and Skeletal Muscle. Sci. Rep. 7: 42296. 2017. PMCID: PMC5235976.
Think about how this approach could be used to study the Gut-Liver-Brain axis!
Look at the Del Rio, et al., Nutrients, 2017 to get some idea of how much work was required to confirm the prediction regarding CSF TMAO.
Post a PCRC on the Blog.
The Systems Biology of COVID-19 and the SARS-CoV-2 virus. The class will build a foundation that includes the emergence of complexity, simple biological subsystems, their reductionist and equivalent toy and organ-chip models, and the measurements required to specify model architecture and parameters. Applications to biology, physiology, medicine, chemical and biological defense, pharmacology, drug discovery, and toxicology. UGrad: PHYS 240 01 and BME 290B; Grad: PHYS 326 and BME 395C.
0) What I already knew
ReplyDeleteI was familiar with some of the main purposes/goals for organ-on-a-chips. I learned previously in this class that one of the great things about organs on a chip is that they can replace animal testing of drugs by simulating drug metabolism/kinetics on an organ chip. The article mentioned that only 70% of human toxicity was predicted from animal testing, so these chip organs can improve our accuracy in predicting effects/consequences of drugs in humans.
1) What I learned
This paper introduced me to ways in which we can "link" organs in our chip. One way is direct coupling, where the media outflow of one organ is directly based on to the next through a connecting tube. However, media, scaling, and waste management between organs is a limitation to this approach. This paper's approach is functional coupling, where each module has its own adjusted flow rate and media, allowing each organ to operate independently with its own optimal environment/conditions.
2) Questions
How will we address issues in scaling of the specific organs/systems in our GBLA on a chip? How will be address the complexity of bi-directionality between each component of our GBLA on a chip?
3) Presentation Topic
Building a functional analysis model (like Fig 2/3) of the GBLA, for a specific compound that is transported through all (perhaps a Short chain fatty acid or a B or K vitamin)
4) Thoughts
This paper was good in that it presented many of the challenges that we will most likely encounter in designing our GBLA on a chip, like the media, linking the organs, scaling, vascularizing the organs, and providing nutrients/oxygen to each component. However, my mind is still boggled by how we will "fake" nerve input to the gut, or how we will "fake" chemical secretions from liver/pancreas when the signaling between each component of our GBLA is so complex.
0) For a number of reasons, it is not clear how quickly OoCs will replace animals for toxicity. Complement, not replace. Let me know if you want to read a more detailed analysis of this issue.
Delete1) Good!
2) These need to be either considered or at least mentioned.
3) EXCELLENT - let's start building medically relevant scenarios to discuss.
4) I agree - this is the main concern we have. Our first-order approximation will be to use a microformulator to replicate the neural secretions. Second order will be to coculture neurons with other GI cells, and figure out how to stimulate the neurons.
0. We briefly went over this system in class, so I was familiar with how the system was connected. I am also fairly familiar with how the NVU device is set up.
ReplyDelete1. I learned about terfenadine and fexofenadine are. I think it’s interesting how you got around the fact that all the cells have different media compositions.
2. I don’t know that much about iPSC-derived hepatocytes, but from what I have heard about these cells, they usually need mouse fibroblasts to help them keep their cell identity and not de-differentiate. Will having other cell types link stellate cells and Kupffer cells perform a similar role as the mouse fibroblasts or will we need to have mouse fibroblasts in our system as well?
If we are differentiating stem cells into our desired cell types, how “pure” do we need our population to be in order to accurately represent the tissues/organs in our system?
3. Functional assays that we can perform on the different chips to show that the chips are working like an organ would work.
4. I think it’s interesting that you were able to find that some molecules can cross the BBB and some cannot using primary BMECs. I’m pretty biased on what BBB model I think should be used, but primary BMECs are not known for forming a very tight barrier in vitro. Did you wait until the cells tightened up before running the experiments?
0-1) Good
Delete2) Current organs on chips don't need the feeder cells. I will track down the latest liver article I have (Lansing Taylor, Pittsburgh) in a bit. They have multiple cell types per your thoughts. Good question re purity.
3) Yes. We need to design our GBA models first.
4) Yes, we waited for the cells to tighten up.
0) KNEW
ReplyDeleteI knew that guts-on-a-chip were complicated and difficult to connect. I also knew that there are sometimes off-target effects from drugs when they're not tested across multiple organs. Also, animal models aren't always great.
1) LEARNED
The organs on chips connections can be made with a single linear pass of media. Oxygenation is important, especially within the gut, where there must be an oxygenated side and an anaerobic side.
2) QUESTIONS
Within the four-organ coupling for vitamin D3 metabolites section, there was a portion that said "The liver
model metabolized vitamin D3 to the 25-(OH) vitamin D3 metabolite, consistent with in vivo findings, although
the kinetic rate was only 2.7% of in vivo rates". What can be done to incrase the kinetic rate to more similar to in vivo rates, and also, why is it so low in the gut chip?
3) PRESENTATION
I'd like to see how their intestinal model worked
4) THOUGHTS
I thought it was interesting that the gut was modeled with mature enteroctyes enteroendocrine, and goblet cells. Is that sufficient? Do we need more cell types? What was also strange to me what that even though there were somewhat high volumes sent to the MS (1-5 mL), it was more typically only 1 mL or less of media sent from one organ to another. I was just surprised that the organ connections could be made with such little fluid. The paper also mentioned that there are about 80,000 untested commonly used chemicals. Can these be studied in a "high throughput" manner, maybe with microformulators in parallel?
0) OK
Delete1) In this study, single-pass was fine. Ultimately, the interconnected organs should recirculate.
2) The cells must not have been as physiologically active as in vivo. A problem. Can be corrected for by scaling concentrations, etc. A similar problem with using any in vitro model for an in vivo human.
3) We can go over that later if you'd like.
4) 1-5 mL makes MS easier. It often can be done with as little as 1-5 uL, but its harder. We had plenty of media, so we took the easy route. This was single-pass perfusion over long times, so there was "lots" of media to go around.
Re the 80k compounds, the Tox21 and Toxcast studies (NIH/NCATS and EPA) are doing HTS screening in well plates, but the predictability is unfortunately low. Hence there is a growing interest in organoids and organs on chips.
0: I knew the basics of MPS and some of the issues like how do you connect them, the universal media problem, vascularization,and the difficulties of using PDMS. I knew the premise of a multiorgan MPS, the goal with understanding PKPD dynamics of drugs, the goal with optimization, and I know who Sarah Bluth and D. Lansing Taylor are since I met them last summer.
ReplyDelete1. The general dynamics of how terfenadine, TMA, and Vitamin D3 interacts with the four organ system. Also the tables at the end are good showing the concerns with improving the coupled MPS. I did not know fluorescent markers were a thing for MPSs. Also apart from our class, I haven't thought much about faking the effects from missing organs.
2. To what degree will we need to worry about other organs? I have an article in the box by Schroeder that discusses how other organs affect the GBA but I haven't read it yet but it could be a good source for answering the question.
3. Universal Media vs vascularization
4. Good overview of multiorgan MPSs. The tables at the end talking about limitations and future directions of MPSs is helpful.
0) OK
ReplyDelete1) OK
2) Please report on the Schroeder article
3) Beyond our class for now.
4) OK
0) What I already knew
ReplyDeleteHuman microphysiological systems (MPS) can model these interactions and are predicted to dramatically improve the efficiency of the drug development process.
1) The most important think I learned
Individual organ models can be functionally coupled to recapitulate sequential organ transport and metabolism for the refinement of the relative functional scaling of organ models, assessment of the compatibility of the materials and media, and demonstration that the coupled organ models recapitulate human in vivo ADME-TOX.
2) My most pressing question from the reading
Since most of the tests were based on a single linear pass of reagents and media, would it be feasible to induce a variable source of media flux (i.e. a differential concentration of organ metabolites) and what changes might it impose?
3) A suggestion for a class discussion
The organ metabolites
4) Any thoughts that you might have on the class or paper
This article summarized four coupled-organ-on-a-chip models and three compounds were evaluated for organspecific processing: terfenadine for pharmacokinetics (PK) and toxicity; trimethylamine (TMA) as a potentially toxic microbiome metabolite; and vitamin D3. One of the highlight would be the coupling of liver and skeletal muscle to test terfenadine toxicity which I found to be very interesting. I wonder if similar approaches can be used to construct GBA smooth muscle model?
1) What I learned
ReplyDeleteDBP helps with vitamin D3 recovery. Terfenadine toxicity is reduced by liver metabolism. TMAO can cross the blood brain barrier. New discoveries on toxicity and metabolism like these help make an organ on a chip model that closely resembles a human. Organ on a chip models can better represent humans than animal models (after reading several papers on the failures of animal models I am finally convinced that animal models are not accurate representations of humans, but what else can we use right now until organ on a chip models are accessible to everyone?)
2) Most pressing questions
How can we improve organ to organ scaling? Do we have to use trial error to determine the right concentration of media and metabolites in the organ? Additionally, how do we determine the media formula?
3) Presentation topic
Different media formulations for different organ on a chip models
4) Thoughts
Really good overview of common problems in making organ on a chip models. I am very curious about the media formulas and would like to learn more on my own. Why do different organs need different media formulas and what are the different components?
0) What I already knew
DeleteI knew the basic schematics of organ on a chip and the basic methods of how they were formed, media and microfluidics, from diagrams we looked at in class.
0) What I Already Knew:
ReplyDeleteAnimal models are far from perfect models of human biology and can fail to predict interactions. Organs on a chip have the potential to model human behavior with considerably more accuracy through communication between in vitro organ systems. And I knew related information on control of microfluidics.
1) What I Learned:
The difficulties in integrating organ-on-a-chip systems. Achieving a direct link between all the devices proves far harder than it may seem at first. The scaling of these systems challenges researchers. Functional connections serve as a way to bypass these difficulties through exterior mixing.
2) Questions:
In this system of organs the metabolism occurred in distinct linear steps, correct? Does this potentially neglect feedback from one organ to another? Or direct interactions with later organs?
Could integrating silico elements help adjust for scaling challenges? Is it possible to have a signal be amplified or diminished before entering the other systems?
3) Presentation Topic:
Cell preparation and challenges for microfluidics.
4) Thoughts:
Good to see an integrated system like what we have discussed in class. And to see methods such as mass-spec being used in such as system.