Science Buddies: "Ask an Expert"

I'm trying to make artificial liver bile using a composition list that I found from a book on Medical Physiology. It states that bile is made up of:
water
bile salts (bile acids)
bilirubin
cholesterol
fatty acids
lecithin
and some electrolytes. Such as, Na(1+), K(1+), Ca(1+), Cl(1-), HCO3(1-)

Some problems that I'm encountering with this are that (1) I'm trying to find some kind of scientific recipe that can essentially give me the moles/amounts of these composites. I do have the amounts of each of these parts of bile, but I'd like to see how scientists go about using them. (2) Bile acids are mostly made up of cholic acid, and chenodeoxycholic acid. Thing is, bile acids are kept at "intestinal pH", so if I do manage to mix all of my things up, how can I make sure that the pH can be similar to that in the intestine. (3) I'm aware that by mixing these things together, some of them might have reactions which can change the pH of the whole mixture, and cholesterol can be easily affected by that. Would I need some kind of buffer in this experiment?

Thanks.

Hi,

This paper may give you a little more information of the composition of bile:

http://www.pubmedcentral.nih.gov/pagere ... dex=1#page

The physiologic pH you are looking for appears to be between 7.1 and 8.5 (slightly basic):

http://www3.interscience.wiley.com/jour ... 1&SRETRY=0

http://books.google.com/books?id=eMRJYO ... &ct=result

It is possible you might have to adjust the pH.

Best of luck on the project!

Barrett Tomlinson

Hi Barrett,

Thanks for the links! I looked at the information in them. The second link that you mentioned does not load... Other than my research on bile, I'm having a few more problems.

(1) I'm not quite sure where I could get bile salts (which are mostly made up of cholic acid and chenodeoxycholic acid, I've learnt), bilirubin (which is made up of decomposed hemoglobin molecules, and so it's hard to obtain, if at all), cholesterol, lecithin, or the fatty acid (which I'm assuming must be a simple chain of fatty acid, but I don't know which one...)
The electrolytes can easily be provided by my school, and I know I can maintain the acidity between 7.1 and 8.5 (which is the level that the bile salts should be at). It's just the other things that I can't obtain easily, which are the key ingredients too! Are there any alternatives to the above ingredients? Like, are there other chemicals that might have the same functions or similar structures?

(2) I also learnt recently that the sensitive materials, like cholesterol, should be added last... Where can I find out which ones are sensitive, and then in which order should I mix the sensitive ones in?

I'm running into a lot of problems... Please help me!!

Thanks again,
A-red

Hi A-red,

This is an interesting project, and I'm sorry to see you running into so many problems.

One question I have is, what is your ultimate goal? Are you just trying to make bile, or are you planning to use the bile to conduct further experiments? If it's the latter, perhaps you can obtain bile samples from a laboratory to use for your experiments. Could you give us a little more insight into your ultimate objectives?

Best wishes,
Heather

Hi Heather,

Actually, I'm making the bile for a kind of experiment later on. I'm going to be adding it to oil, to watch how it physically breaks it down. Then I'm going to add acetaminophen (Tylenol) to the oil and bile, and see how the function is affected. Then I will add echinacea instead of Tylenol and observe and compare the effects that has on the ability of my 'bile' to break the fat down. I talked this over with my Biology teacher too, and he suggests that I should combine the ingredients that I are accessible to me (I can get cholesterol by the way). Then I can try to maintain the pH levels between 7.1 and 8.5 using NaOH and HCl, depending on whether I want to increase or decrease the levels.

I'm essentially trying to make something that mimics the function of bile. I know this bile isn't going to be as good as natural bile... which is why I'm a little concerned. But I'm figuring that since my project is a study, whatever I can accomplish in the tests I described would be beneficial to my whole project. But if on the other hand, I can't get the bile to break down anything then it shouldn't affect my project. Because my main focus is the research that I'm putting into drug metabolism and the function that bile plays in that.

The science fair is actually next Friday (Feb 13) so I'm getting a little panicky.... I'll try to see if my prepared bile can mimic the natural functions, and if it doesn't then that's okay, because I can fall back on my research, which is quite extensive thankfully. If I can make it to the next level, I'll get a bit more time inbetween to either make better bile, or get some samples from a laboratory like you suggested.

I'd be really grateful for whatever help I can still get in terms of preparing bile, or any research resources that might aide me in my study.

Thanks,
A-red

Hi,

I'd like to give y'all a quick update on my project. While I was trying to prepare bile today at school (which was honestly turning out to be a much more complicated task than I'd thought), I went to up this teacher for some help. And guess what? She just happened to have DRIED BILE! Just like you suggested getting samples, Heather!! I am so excited now, because I can actually go on to my tests now, instead of focussing on making the bile!

So, I now have to figure out how much water I should add to it so that I get a good, natural consistency. There's another tiny problem with that... she could only give me 0.62 grams, which was all that was left of the 3 grams she initially had. So what I need to do is figure out what maximum amount of water I can add to the dried bile, so that I can get a bit more to work with.

My new plan is as follows: (1) find out what maximum amount of water I can add to the dried bile, (2) divide the bile solutions into four separate samples: one as the control sample to observe its function with the fat I use, and three for the different kinds and mixtures of drugs that I'm going to be putting in, (3) find out how much acetaminophen or echinacea would be naturally metabolized by the amount of bile that I can prepare, (4) when I'm taking my observations, I'll time the reactions, and then also check the pH levels after the fat has disintegrated enough that it's not visible. I'm guessing that acetaminophen will make the solution more acidic (it's an acid after all).

So that's what I'm doing tomorrow (Sat). I'm going to find answers to the questions I listed above, and if you have any ideas, I'd be extremely grateful!

Thanks and if I run into any other problems, I'll let you know!

A-red

Hey there,

Update: I figured out how much water I need to add to the sample of dried bile that I have. Yay! And I've typed up my procedure and everything. But I'm just having a bit of trouble right now with the amount of Tylonel that I should use... I have those Regular Strength 325mg tablets. I'm sure I need to grind up one tablet, or two maximum, and then... what? I know I'd have to dissolve the Tylenol in some water (I'm using distilled water for all of my tests by the way), but I'm not sure how much Tylenol would end up in the liver when one takes a regular dose (325mg)? And then in the liver, how much of it would go into the bile? Does all of it go there? Probably not, because the Tylenol would go into the blood stream to help the nerves, like when we have a headache or a muscle cramp. How can I figure that out?

Thanks,
A-red

Hi A-red,

Thanks for the update! I'm glad to hear that you were able to figure out how much water to add to the bile to conduct your experiment.

I'm a bit confused about your next step. I thought that you would be looking at the effects of Tylenol on the ability of bile to break up oil. I know you've done a lot of research on your topic, but to help you focus your efforts, let's review how bile works:

Bile is an "amphipathic" molecule (having both polar and nonpolar parts) that is used to aid in lipid (fat) digestion. It is produced in the liver, then stored and concentrated in the gall bladder. When fats from a meal move from the stomach to the beginning of the small intestine (called the duodenum), the gall bladder sends bile to the duodenum. There, the bile acts to break up the fats into smaller pieces so that the digestive enzymes can better break down those fats. I like to think of it working the way soap works. When you get oil on your hands, it doesn't come off with just water. But if you put soap on your hands, the soap molecules break the oil into smaller pieces that you can then wash away with water. Here is the Wikipedia article on bile: http://en.wikipedia.org/wiki/Bile

My point is that I don't think you should be worried about how much Tylenol makes it to the liver. Instead, you would want to estimate how much of the Tylenol makes it to the duodenum of the small intestine. Some of the Tylenol may be absorbed across the stomach lining (into the blood), so you could reduce the amount compared to an entire pill. However, most absorption occurs in the small intestine, so I don't think using one pill's worth will be that inaccurate for your experiment.

I hope that helps. Please continue to keep us posted, and let us know if you have further questions.

Best wishes,
Heather

Hi Heather!!

Thanks soooooooooooooo much!! That helps a LOT!!! I was about to conduct my experiment just now, and decided to check on my post before I did so. I was getting worried too... But you totally saved my day!! Alright!!

Okay, so I'll do my experiment, using almost all of the Tylenol pill and then I'll record my observations and everything. I'll let you know if anything else comes up!

Cheers,
A-red

Hi there,

Thanks for all of the help y'all gave me with my project, because guess what? I'm going into my city's District Science Fair!!!

My project in the end was really good. I had been able to do my experiment with the bile, and I got really great results. See when I added 3 drops of cooking oil, as a fat, in the centre of a pure bile sample, it disintegrated gradually but quickly, within 10 seconds. But then when I added ground-up Tylenol to the bile, and put in 3 drops of oil, it simply floated on top as a big blob! The same happened when I added Echinacea to the bile, and when I added Tylenol and Echinacea together.

After talking with the judges about my project, and getting really good feedback, I have a few ideas about how I can make my project more valid for answering my question. My overall question is whether Tylenol and Echinacea actually damage our liver if taken together. So in my next experiments I'm going to be testing why the Tylenol and Echinacea stopped the bile from performing its functions. Is it because acetaminophen's polarity counteracted with the bile's? Is acetaminophen in the same chemical structure by the time it gets to the liver, or has it been transformed into one of its metabolites by then? When food is digested, does any of it go into the bile at all? Questions like that. And I mentioned these questions to my judges, and, fortunately, they were impressed by the fact that I'm not stopping where I am right now!!

So first off, I'm going get more bile (at least 20g I hope!!!), so that I can conduct more trials. Obviously I can't just pass off my results so far as being the final thing... So any idea where I could buy more bile, or dried bile? And if you have any suggestions on what else I could work on to prove or disprove that Tylenol and Echinacea together are harmful for the liver, I would love to hear them!!!

Cheers,
A-red

Hi,

I'm wondering if there is a way for me to measure the amount of bilirubin in dead liver cells?? I have designed an experiment in which I'm testing the affects of acetaminophen and echinacea on cow liver cells under a microscope. Is there any way at all that would allow me to measure toxic substances in the cells, via staining etc?

Also, I've worked a lot on understanding metabolism of drugs (especially acetaminophen and echinacea) and I have also studied how either or both of these drugs can affect bile's function, as well as a liver cell's structure (cell lining, nucleus, etc.). I don't have just one experiment that sums it all up... because my question can't be answered that way. If you recall my question was "How and why does the interaction of Echinacea and Acetaminophen damage the liver?" I decided on just testing as many things as I can, and then concluding with a possible answer to that question, which would initially be a hypothesis for an experiment that I want to do, if I had the equipment and the permission to use living samples. Is that okay? Would judges mark me down for just giving them a hypothesis as the end result of my study, in place of a conclusion?

Thanks,
A-red

A-Red - Congratulations on the success of the project and on moving to the next level! I am working to get an expert to address your questions here (and I know you also have a new thread, too).

Amy
Science Buddies

Congratulations. My wife was a pharmaceutical toxicologist and did her thesis on using liver tissue culture to perform initial screening of substances. From what I remember, if you lyse the liver cells and then centrifuge them, any bilirubin will end up in the solute. You can then normalize the pH and run this through the same testing you would do with blood plasma to detect bilirubin. You need to do a little research on your specific bilirubin testing method to determine if there is anything left over from they lysing process that could interfere with the assay so there might be some other pre-processing needed to remove the interfering chemical.

The typical way toxins are measured is via High Pressure Liquid Chromatography and Gas Chromatography sometime used in tandem with HPLC feeding GC.

The only thing that you can see with a microscope are any leasions or other changes in the liver tissue itself. The toxins would be way too small to see. Even if you used an electon microscope, you might see shadows of the elements in the chemicals but you wouldn't be able to figure out what the chemicals were.

Hi there,

I researched a bit on the centrifugation idea that you gave me Craig, and I've decided to go along that route. But I realized quickly that separating bilirubin or albumin (a protein that binds to bilirubin for the excretory process) from a batch of liver cells can get very complicated. Instead, I'm planning on simply centrifuging samples of liver tissue with and without acetaminophen and/or Echinacea, and comparing the amounts of the separations on a gradient. I'm going to do this by measuring the amounts in the test tube with a ruler, because I think there would be apparent lines between the separations.

What I'm having some trouble with now is how I'm going to prepare the liver tissue for centrifugation... That is, how should I lyse the liver cells (soap detergent and alcohol??)? I'm guessing I'd have to break down the liver tissue in a blender, and then lyse the cells. I'm also confused as to how the tissue might be centrifuged.... because I'm sure I'd need some kind of solution first, and if it's all solid after the blending process, should I add water to this?

Please reply as soon as you can, because I'm planning on doing my experiment tomorrow!!

Thanks,
A-red

Hi,

I just want to make sure you understand the urgency of my question... the science fair is this Thursday, and I've been trying to patch up my areas of confusion with no luck. This isn't an integral part of my project, but it is definitely going to give it a boost, seeing as it is a study.

Thanks again,
A-red

I think you are missed something significant. In order to test whether some chemical/drug affects liver metabolism or function, you have to have viable liver cells (ones that are currently alive and functioning). This means you have to work with live animals which have livers which is ususally outside the bounds of what can be done for Science Fair projects without special permission and supervision.

When you asked about how this might be done earlier, I thought you had completed your science fair project and you were asking purely for your own knowledge and not to actually pursue doing this in the context of your Science Fair project.

Sorry if I misled you.

Hi,

Thanks for your reply!! Yes, I did realize that I'd need to work with functioning live liver cells in order to test how the function of the liver might change with acetaminophen and Echinacea. And of course I can't actually use a working liver for my Science Fair project... But I figured that if I can get significant results even with dead liver cells, it would help me in making a point. Of course, like I said before, my project is a study so this particular test wouldn't really affect the credibility of my other results. I have mentioned this to the previous judges, and they didn't have further objections....

So, I think I'm going to use a mortar and pestle to break up the cells, based on what my bio teacher has told me. And then add soap detergent as well to further aid the breakdown process. So my question is: should I add water to the grinded liver cells, in case I don't end up with a solution for centrifugation?

Thanks,
A-red

Hi,

I just want to thank everyone at Science Buddies for their wonderful help with my project. At the Regional Science Fair, I ended up winning a Bronze medal in the Health Sciences category!! So your help really paid off! I'm going to keep on working on my topic though, because I haven't collected valid results just yet, so you will all be hearing from me again soon.

Thanks again,
A-red

A-red - Congratulations! It's wonderful to hear of your bronze medal in the regional competition. That's great! I'm glad to hear that the help you received from Science Buddies Ask an Expert experts was important to your project, and it's wonderful to hear that you are continuing your research. Please let us know how it goes!

Amy
Science Buddies

Hi!

Like I promised last year, I'm back with some updates on my project. Over the summer, I got the chance to test the effect of Tylenol and Echinacea on HepG2 liver carcinoma cells, and I noticed some really interesting growth trends in the populations. I have been presenting my results in science fairs again, and have successfully progressed to the regional fair again! But, I feel like I can probably improve my statistical analysis of the results before the fair, and it would be great if you could help me out here if you can.

To quickly update you on what I did: I basically grew seven cell cultures (control, high and low dosage Echinacea, high and low dosage Tylenol, high and low dosage combinations) of which I had three trials each. Over a 10-day period, I counted the number of cells in each Petri dish every 2-3 days (No, I didn't individually count each of the millions of cells! Instead I took out measured samples from the dishes and counted the cells visually in a counting chamber under a microscope, which is essentially a grid for easy counting. Then I just plugged the number into a formula to estimate the total number of cells in the Petri dish - pretty simple).

I found that the cell cultures treated to medium and high dosage Echinacea grew at a very natural rate (exponential), whereas the growth trend of the cells with medium and high Tylenol was slightly set off from the natural curve. Successfully affirming my hypothesis (that the combination would offset the natural growth of HepG2 cells), the cells in my combination treatments were clearly not growing exponentially. Sadly, I couldn't gather data from the high dosage combination treatment, because my cells somehow got contaminated. But the medium dosage treatment certainly shows that the two drugs are reacting and can be adverse for the growth of liver cells.

Now, here's the part where I need some help and advice: as you may have already noticed, the way that I've formed my conclusion is by comparing the growth trends of my cells with the exponential trend. I chose to do this, because cells naturally grow at an exponential rate, and I thought it would be a good tool in assessing whether my cells grew naturally. Can there be another method of statistically analyzing my results? I know you probably need to see the data or graphs in order to do that, but I'm feeling insecure about posting my results on here.... If someone could give me their email address maybe, I could send you my whole report. Or if there's another way of helping me, please let me know!

Thanks!
A-red

Hi,

Unfortunately, all communications between experts and students have to take place on the forums. That said, I think I may be able to help you without seeing your data.

I'm not sure how much math/statistics you have had, so I will start out by just giving you a general procedure and then you can tell me what steps you need help with.

First, you will need to log-transform your growth variable so that you end up with an approximately linear relationship between time and growth. This means that you need to take the natural logarithm of your cell numbers and plot it against time, checking to see if it's approximately linear. If it is, great, we can go to the next step. If not, post back and let me know what it looks like.

Next, you need to do something called an ANCOVA. You will need to have access to a statistics program. If you don't have access to a statistics program, let me know--we can kind of 'cheat' and do a similar analysis using simple linear regression, which we ought to be able to do online for free. In your ANCOVA, the natural logarithm of cell number will be your dependent variable, time will be your covariate, treatment will be a random factor, and dosage will be a fixed factor.

I'll stop there, because there's not much point in going further if you don't have access to a statistics program. Let me know what you have, and we'll work with that.

Hi,

Sorry I didn't get back to you sooner.... I've looked at what you've suggested and I should be able to do this. Just have to find a program to do ANCOVA though. My graphing calculator is able to calculate ANOVA... would it make a huge difference if I used this one? And what are ANCOVA and ANOVA anyway? I've searched online and in some stats books my math teacher gave me, but I still don't really know!

AND, I wanted to inform all of you that I have been selected to attend the national science fair from my regional science fair!!!

Thanks for your help!! I'm going to improve my work with the ANCOVA (or ANOVA depending) and then we'll see what happens! So please, can you help me a little bit with understanding the statistical methods you mentioned?

Cheers,
A-red

Of course.

ANOVA stands for 'ANalysis Of VAriance.' ANCOVA stands for 'ANalysis of CO-VAriance.' Unfortunately, they are two very different things, so you will not be able to use the ANOVA function on your graphing calculator. An ANOVA simply tells you whether there are statistical differences in the means of different groups. An ANCOVA tells you whether or not there are statistical differences in the means of different groups AFTER controlling for some other continuous variable.

For example, you would do an ANOVA to see if there were differences in male and female human heights (i.e. if the mean male height was different from the mean female height). However, you can imagine that if the males and females were many different ages, this could affect their height too. So, you might then do an ANCOVA looking at the effects of gender on height after controlling for the effects of age.

Can you ask your science teacher if you can get access to a program that does statistics? If not, we'll try something different.

Wow thanks Melissa! Sorry I didn't back to you sooner... it's been crazy at school.

This is really making sense now, and I can see how the ANCOVA fits my data better. I'm still not sure if I have access to a program for calculating ANCOVA, but I will check with my teachers tomorrow and let you know. In the meantime, I think I will most probably need to use the 'cheating' method, as you put it! Anyway, I'll come back with more details tomorrow, and then we can finally get this done. Thanks again!

Hey I talked to my teachers, and I can't find a program to do ANCOVA. I might, just might, be able to find something, but it's too late to rely on that possibility. Could you please tell me about the simple linear regression? This is starting to get urgent, and it'll be great if I can hear from you today or early tomorrow. Thanks!

Okay! Just so you know, I'm in Europe--hence the weird reply schedule.

Here's what you do: use this web page: http://home.ubalt.edu/ntsbarsh/Business ... ession.htm to calculate a simple linear regression between log-transformed cell counts (y) and time (x) for each of your different groups (e.g. high echinacea, low Tylenol, control, etc.). Use a confidence level of 0.95.

For each regression, you want to write down the slope and its standard error (don't worry about all the rest). Again, I'm not sure how much math you've had, but a linear regression gives you an equation in the form y=mx+b, where m is the slope and b is the intercept. You are interested in seeing if the slopes are different for different treatments.

The way you can do this is to calculate a 95% confidence interval for the slope. You do this by adding and subtracting 2 standard errors from the estimate for the slope. For example, if your slope was 1.0 and your standard error was 0.2, the 95% confidence interval would be (1.0-0.4 to 1.0+0.4), or (0.6-1.4).

Once you have calculated all of these confidence intervals, you can compare them. Any two confidence intervals that do not overlap are statistically significantly different (kind of...as I said, this is 'cheating', and it's not quite valid to do it this way for various complicated reasons). In other words, if you had two confidence intervals from 0.6-1.4 and from 1.2-1.8, these two slopes would not be different because some of the numbers overlap (both slopes COULD be between 1.2 and 1.4). However, if your intervals were 0.6-1.4 and 1.5.1.9, you could say that those slopes are different and that the two treatments caused the liver cells to grow differently.

Keep in mind that you will need to tell the judges that you realize you should have done an ANCOVA but that you couldn't, so you compared the slopes instead.

Anyway, hope that helps--if you have more questions, just let me know.