Team:Paris Bettencourt/Notebook/Microbiology
Introduction
The microbiology team aims to identify wine stain degrading microorganisms.
Basing this bioremediation-inspired approach on the two following articles:
- Bioremediation of phenol by alkaliphilic bacteria isolated from alkaline lake of Lonar, India
P.P. Kanekar, S.S. Sarnaik and A.S. Kelkar. Journal of applied microbiology supplement 1999. - Bacteria Subsisting on Antibiotics
Gautam Dantas, Morten O. A. Sommer, Rantimi D. Oluwasegun, George M. Church. Science 2012)
Indeed, it is highly probable that a rich-anthocyanin environment such as a vineyard, would host microbes with the desired degradation skills. In addition, the chance to find a microbe able to digest efficiently a wine stain, with its proper anthocyanin composition, in terms of anthocyanin diversity, and abundance, is theoretically increased.
To ensure the widest microorganisms diversity, we used soil samples from all around the world (Australia, Spain, Namibia, France, Croatia), particularly through collaboration with different iGEM teams.
This bacteria identification was based on two different approaches:
- We are creating a bacterial database. It implies culture of the bacteria on different selective and non-selective media, and characterization of the strains. This identification was managed through 16s rRNA PCR, known as the most common housekeeping genetic marker in bacteria. At the end, we would test the database directly on fabrics stained with wine, or only anthocyanin, looking for colour degradation.
The main advantage here is to select potential useful bacteria from soil, isolate them, and screen all of them, on fabrics, thanks to a high throughput assay. Nevertheless, it implies a selective bias, as some interesting microorganisms may not grow in the media we chose, or cannot compete with other microorganisms present in the soil.
To tackle this problem, we also followed another approach:
- We screened our samples for a microorganism growth directly on anthocyanin-enriched media, and for anthocyanin degradation by absorbance measurement. The idea here is to screen for bacteria that could use anthocyanin as their sole carbon source, and thus degrade it, or that could simply metabolize it. After identification of potential interesting microorganisms, we would isolate them, characterize them, and then test them on stained fabrics.
Therefore, the main point of our team is to put in evidence already existing anthocyanin-degradation metabolism in nature, so that we could isolate the enzymes, and potentially optimize them thanks to the binding domain team results.
Week 27th June - 3rd July
The first step is to define a systematic protocol for the selection of microorganisms from our samples.
It includes :
- Determine samples conditions of collection, storage, and treatment before microbes culture.
- Determine the selective and nonselective media we will use for culture and the number of dilutions per samples.
- After restreak and isolation, determine an efficient way to characterize the strains.
- For the second approach, where we test directly an anthocyanin-degradation natural metabolism, define the controls.
Therefore, during this week, we focused our work reading articles dealing with bioremediation and isolation of bacteria from soil samples.
Nonetheless, we rapidly had to tackle the following issue: Such approaches implies protocols of microorganisms culture on anthocyanin enriched media or even as a single carbon source. Or, anthocyanin isolation and purification is hard to achieve, and consequently, this chemical compound is really expensive. We couldn’t afford to buy it in high quantity.
Considering this problem, we decided to take a deeper look in the anthocyanin chemical structure, so that we could potentially find a cheaper substitute molecule for our assays.
Anthocyanin is a phenolic compound which belongs to the Flavonoid family. Flavonoids have a basic structure of C6–C3–C6. Depending on their structures, flavonoids may be classified into about a dozen groups, such as chalcones, flavones, flavonols and anthocyanins.
If we take a deeper look in the anthocyanin structure, we observe that it’s composed of :
- a chroman ring
- an additional aromatic ring on C2
There are 31 different monomeric anthocyanins known. They differ from number and position of the hydroxyl and/or methyl, ether groups.
But, 90% of the naturally occurring anthocyanins are based on only six common anthocyanidins. among them :
- Cyanidin,the major form in nature
- Malvidin, the most commonly found in wine
As we can see in this figure, they only differ by their cycle B groups, so the chroman ring remains unchanged.
Therefore, we believe that enzymes able of degrade a lot of different types of anthocyanins would preferably attack the chroman ring structure as it is well conserved among the family.
Thus, we thought that the ideal cheap substitute molecule should present a very similar basic structure.
After some researches, quercetin, a flavonol molecule, appeared to be a good competitor.
We can in fact see that the only difference remains in the presence of a carbonyl group in 4. Another advantage is that Quercetin is present in Wine, and is partly responsible for its color. Thus, even though the assays may find some enzymes able to degrade specifically quercetin instead of anthocyanin, we should have a decrease in the wine stain color intensity. In addition, because of the co-pigmentation chemical interaction between quercetin and anthocyanin, degrading quercetin could also have an impact on anthocyanin stability.
For these reasons, quercetin was chosen as our substitute molecules for running the microbiological assays.
However, in parallel, we decided to develop a protocol for purifying anthocyanin from grapes, so that we could also tests our microbes directly on anthocyanin. (See the Anthocyanin section for the evolution of the protocol).
Waiting for vineyard samples, we went to Cochin Port Royal’s garden next to our lab, and took some 15 mL of soil.
Week 11th - 17th July
Week 17th - 25th July
Church paper
We found a paper describing the isolation of antibiotic degrading bacteria from soil samples. It appears that the goal was very similar to ours, the only difference remaining in the nature of the molecule. As they wanted to isolate bacteria that can grow on antibiotics as a unique carbon source, their main concern was carbon contamination from the soil sample.
To avoid this contamination they inoculated the samples in a SCS (single carbon source) liquid medium. They let it grow seven days at 22°C then use the broth to inoculate a new SCS liquid medium. This step is repeated two more times. Then the culture broth is plated on SCS plate and the degrading bacteria isolated. In these conditions the carbon from the soil is consumed during the first cultivation steps and the bacteria that cannot use antibiotics as a carbon source do not survive during the next cultivation step.
This protocol have some issues.
- It take 21 days before plating to isolate the microorganisms, this is very long.
- If an organism can degrade anthocyanins but cannot use it as a carbon source this organism is not isolated.
As we found in several papers, the regular technique to isolate microorganisms from the soil is to dilute the sample and directly plate on agar. A dilution of 10^(-1) correspond to a diltion of 1g of soil in 9 mL of PBS.
Filtration of the soil sample
A member from the protein group told us that he used a different protocol to isolate a toluene degrading bacteria. To remove the carbon contamination from the soil he filtered the sample with a 0.22µL filter. Then he cultivated the filter in a liquid medium before plating. We decided to test this protocol.
Screening assay design
In church lab's article, they cultivate the bacteria at 22°C and pH 5.5 . We found that pH 5.5 inhibit most bacterial growth so we decided to stay at a neutral pH.
We did not have an incubator that could be set at a temperature of 22°C, after a discussion with our advisors and a microbiologist from Cochin hospital it was decided that 30°C should be sufficient for our experiement. Some bacterial species would not survive at 30°C but a sufficient number would, and they may grow faster at 30°C than at 22°C.
We decided to chose 5 dilutions, from 10^(-2) to 10^(-6). Instead of only plating at the end of the 3 liquid cultivation, we decidede to plate 100 μL of medium just after inoculation and and at the end of each cultivation. The goal of this experiment is to know the dilution needed and to test the impact of the several liquid medium cultivation steps.
We decided to chose 4 different medium (for broth and plate).
-M9 : a medium with salts but no carbon source, this is our negative control medium, if something on that medium this means that there is a carbon contamination.
-M9 glucose (M9 G) : M9 salts plus 1g/L glucose, this medium is our positive control, if nothing grow on it it means that there was no bacteria in the medium.
-M9 quercetin (M9 Q): M9 salts plus 1g/L quercetin (similar condition than in Church's article), this is the meidum were we want to isolate the bacteria.
-M9 glucsose quercetine (M9 GQ) : this medium is necessary to test if quercetin is toxic form microorganisms, if we have less growth than in M9 glucose that means that quercetin is toxic.
As it is done in church's protocol the incubation time between each cultivation step is 7 days.
In parallel we decided to test the filtration of the soil samples. In this part of the assay samples would be filtered, one inoculated in a M9 medium and the other in a M9 quercetin medium.
Week 25th - 31th July
Quercetin medium preparation
As soon as the quercetin arrived, we decided to start the assay. However it was found that unlike anthocyans, quercetin was mostly insoluble in water. this causes problems for medium preparation will the organisms be able to degrade an insoluble molecule in a liquid broth ? But it as also a problem for quercitin quantification, we thought that we could measure the degradation of quercetin with the absorbance, however if quercetin is not soluble in water it is impossible to measure this absorbance.
We tried to solubilize querctin at different concentration in water and ethanol : 1g/L, 0.1g/L, 0.01g/L and 0.001g/L . We did not succeed in dissolving totally the quercitin, so an absorbance assay was impossible. However, even at 1g/L the liquid was a little yellow, so quercitin was at least slightly soluble. We thought that the little quantity of soluble quercitin could be enough for growth, as long as the medium is agitated and the quercetin constantly solubilized during the consumption. So we decided to start the assay anyway. Me managed to have relatively homogenous plate.
Results after two days
- The control plates with M9 showed growth until 10^(-3), that means that if we want to avoid carbon contamination in plate to dilute the samples at 10^(-3) at least
- The plate with M9 G always showed growth even in the more diluted samples there was microorganisms.
- The plates with M9 QG showed less growth than the plate with M9G and there was no growth after a dilution of 10^(-3), that means that quercitin is toxic for some organisms.
- The plates with M9 QG showed growth until a dilution of 10^(-2) because of the carbon contamination.
This part of the experiment seemed compromised as there was no growth in the condition M9Q at a dilution were there was no carbon contamination (below 10^(-3)
Growth results in tubes
-M9 growth with a dilution of 10^-2, the carbon contamiation is avoided with a dilution of 10^-3 and below.
-M9 G groth in every tube, there is always microorganisms in the samples
-M9 Q and M9 QG, it is difficult to say, the quercitin is precipitated and it seams that the experiment is a failure, that the quercetin cannot be used in liquide medium, maybe because the agitation is too low.
Week 01th - 7th August
Results after a week in solid medium
In plates : The results were surprising. In the plates of dilution of 10^(-2) and 10^(-3), with or without glucose, we could see transparent circles. The quercetin seemed to have disappeared. In the middle of these circles there was always something that looked like a filamentous fungi.
As we can see it is easy to identify the cicle were the yellow disappear. The organisms in the center always look like that. This is very similar to a fungi, with the hyphae in the middle and spores in the whole circle. It is to be noted that it is written that the dilution is 10^(-1). This is an annotation mistake. At the beginning we did not counted the initial dilution of the soil sample in water. We have considered that our original sample was 1g of soil in 9mL of water.
results after a week in liquid medium
In liquid medium : With dilution of 10(-2) and 10(-3), in quercetin glucose and in M9 quercetin, there was fungal growth and it seemed that the quercitin was consumed (the medium was less yellow).
conservation of the material
In liquid and solid medium there seems to have a degradation of the quercetin and it is easy to identify and to isolate the seems to be responsible of the degradation. However in these organisms look like filamentous fungi. As our laboratory has no experience with fungi we decided to keep the plates at 4°C before working with them.
Week 8th-14th August
Filamentous fungi good lab practice
We looked for protocol to work safely with fungi. We found a quite complete fungi protocol book "Laboratory protocol in fungal biology" (M. Ayyachamy and al). There was a safety part. The more important was to prevent spore contamination, ethanol is not sufficient to kill fungal spores. A way to disinfect the hood needed to be found. It was said in the book to disinfect with bleach between every experiment and , at the end of the day, to disinfect during 20 minutes with a strong disinfectant like those used in hospitals.
We were able to obtain the disinfectant SURFA’SAFE preminim of the laboratory Anios from the microbiologist of Cochin. It was decided with our advisor to disinfect with bleach between each fungi (to avoid contamination of the different fugi) and, before someone else use the hood, to disinfect during 20 minutes with anios.
Quercetin analysis
We tried to solubilize quercetin. Quercitin was almost insoluble, even at a concentration of 1mg/L the quercetin precipitated. We tried to heat it but after 1 hour of eating the quercetin was still not solubilized. We thought that the solubility of the molecule was of 200g/L so we were very surprised. However we looked again on sigma Aldrich information and that was a solubility in DMSO, an organic solvent.
[photo needed]
We tried to solubilized 1mg of quercetin in ethanol, without results.
[photo needed]
The insolubility of quercetin seemed to be not a real problem for the microorganisms, both in liquid and solid medium. However we wanted to quantify it by absorbance. An insoluble molecule can’t be quantified by absorbance. But Maybe we could use DSMO or another solvent were quercetin was soluble.
Another problem was the separation between the cells and the quercetin. We tried to filter a sample of water with quercetin with a 0.22 mum² filter. All the pigment stayed fixed to filter. The pores were much bigger than the quercetin. It seemed that the quercetin has a strong affinity to the membrane.
[photo needed]
As the quercetin is mostly insoluble in water a centrifugation was a way to remove the cells and the pigment from the broth. All we needed, than, was a medium were the quercetin was soluble, preferably not an expensive and toxic organic solvent.
Week 15th-21th August
Fungal strains collection
We had all we needed to work with filamentous fungi. We isolated the identified strains and inoculated them in Sabourdaud Dextrose Agar, a common plate for filamentous fungi culture. After 4 days the plates were covered of mycelium. Then we inoculated them on M9Q agar Plate and in M9 plates.
test in liquid medium
The isolated fungi were tested in 3 liquid medium :
-M9
-M9 + 1g/L quercetin
-M9 + 1g/L quercetin + 1g/L glucose
there was also a blank without fungi
Second assay
In order to reproduce the experiment another assay was conducted in the same conditions than the first one we used another samples of the soil of Cochin. However a mistake was made and distilled water was used to dilute the samples instead of PBS.
Quercitin analysis
We wanted to see if the disappearance of the color in plates could be due to a pH modification. So we took 1g/L of quercetin and changed the pH with hydrochloric acid. There was no color change at a pH of 1.5.
[photo]
Then we added base and there was a change. Around a pH of 12 we could see a solubilization of quercetin. After a pH of 12.5 quercetin was completely solubilized. The color was orange instead of yellow. There was no color change between a pH of 12.5 and 13.5.
[photo]
We had solubilized quercetin. We managed to solubilize up to 50g/L of quercetin in alkaline water. We had found a way to solubilize quercetin.
[Photo]
We did the absorbance spectrum and we could see that the maxiumum of absorbance was at 315nm.
The R² is good, that confirm that the calibration is working. We have a relation between absorbance and concentration : A = 56,767*C, that mean that C=1/(56,767)*A
Finally C = 0,176*A
Week 22th-28th August
Result of the second assay
There was growth on M9 glucose and M9 quercetin glucose for plates for dilutions of 20^(-2) and 10^(-3). However the variability was much less important. No quercetin degradation organisms were found. This might be due to the osmotic choc caused by the dilution of the samples in distilled water.
Third assay
We did a third assay with a soil sample from Croatia. The conditions were the same than for the two first assays (with PBS and not distilled water !)
Fungal isolation
There was still growth on M9Q plates, we could still see the circle of quercetin degradation. It was confirmed that the strains have been isolated.
Week 29th August-4th September
Week 5th-11th September
Week 12th-18th September
Week 19th-25th September
Week 26th September-2th October
Week 9th-16th October
We did an assay with 4 funghies with anthocyanin to see if they are able to degrade it. We prepare an M9 media with 0,8 g/L of glucose and 12mg/L of anthocyanin. This media was flitrated to sterilize it, since anthocyanin is degraded at high temperature, it cant be autoclaved. Each funghi was made in triplicate and cultivated into 5 mL of media. We made a measure with the TECAN at Day 0 and Day 3.
For each measure, we made a centrifuge for each tube at 6000 rpm for 1 min. Then we took three times 150µL and put them into TECAN plate. We add to each wells 50 µL of HCl 37% and let the color stabilise for 1 min. Then we measure the absorbance at 510nm to compare the degradation.
Week 17th-19th October
Separation in two project
At this point we divided the project in two approaches:
-The goal of the first approach (approach 1) is making a huge library of strains that were isolated from soil sample.
We would like, from a given soil sample, to isolate as much as we can of bacteria diversity.
Then we could test each strain on Quercetin liquid media and report Quercetin degradation.
Thanks to our collaboration with other iGEM team, we have sample from some place all around the world.
We would like to compare the diversity of bacteria between each soil sample taken from all around the world from vineyard for instance.
At the end we will have a library with some strains that are able to degrade Quercetin and some other that are not.
We could quantify the strength of the degradation and also make a link between the species that are able to degrade the Quercetin to make hypothesis about a common enzyme responsible for this degradation.
To achieve that, here are the step of the protocol for approach 1:
We plate at 10^-3 dilution (we choose this dilution thanks to previous experiments) to the soil sample on a non-selective agar media (LB, TSA, M9Glucose) but also on selective media (FTO for micrococcus, Agar from Mossel for Bacillus Cereus, M9 Quercetin).
We incubate for one or two days at 30°C.
When we have some colonies, we restreak all of them on LB or TSA plate in order to isolate a maximum of different strains. Once the restreak bacteria grew enough, we give them a name to store them in the library.
FS_M69 stand for Frank&Stain_Microbesgroup69
For a given strain we have the date of the stock, the species once its sequenced, the original agar media from where it was taken, the origin of the strain etc…
Then we make an overnight culture in Tryptic Soy Broth and systematically we characterise our strain with a 16sPCR from colonies restreaked on LB plates.
When the overnight culture is over, we inoculate 100ul of this directly in a 5ml M9 1g/L Quercetin falcon tube in triplicate. We also make two glycerol stocks. One at -80°C the other at -20°C (to be reused more easily).
For the assay on liquid media Quercetin, we are quantifying Quercetin degradation.
We are able to see Quercetin degradation thanks to the absorbance. Quercetin in solution is not soluble at least just a little. So we cannot detect absorbance at a pH=7.
But, when we put our solution at pH=12, we are able to see a nice absorbance spectrum with a peak of absorption around 315nm.
With a dilution a hundred times in NaOH we are able to have a good absorbance value for Quercetin degradation.
To see if there was a degradation, we need to use a negative and a positive control.
The negative control is just M9 + Quercetin. The positive control is M9 + Quercetin + Pseudomonas Putida.
Indeed, some article show that Pseudomonas is a good candidate for Quercetin degradation. The iGEM Evry team gave us this strain to use it for our assay.
We made an experiment to check how long does pseudomonas needs to degrade Quercetin.
We found that 4 days are enough to see a significant decrease in Quercetin absorbance. But to give a chance to every strains we will assay, we decided to make a 6 days experiment.
We designed the experiment with a Quercetin quantification at Time 0 and at Time 6 to see the evolution of Quercetin concentration.
We used Pseudomonas Putida K2440 as a positive control.
We made two sample: at T0 to make sure that each tube had the same amount of Quercetin. And at T6 because it’s enough to see a significant decrease in Quercetin concentration if there is a strain able to use it as a carbon source.
All culture were made in triplicate.
To present data, we made a ratio of the absorbance at T6 over the absorbance at T0 for every culture. Then we average the three values. And we are able to calculate the standard deviation.
The value in y axis are a percentage of remaining Quercetin.
Here you can see the T6 sample. We took 100µl from the 5ml culture after vortexing it and we put it in a eppendorf tube. Then we add 900µl of NaOH 0.5mM solution at pH=12.
You can see the difference on each tube depending of the % of Quercetin remaining.
As you can see, none of the strains are able to degrade Quercetin except or positive control.
The problem is that these strains were collected from M9 Quercetin plates.
So at least we were expecting some Quercetin degradation.
We suggest that it was a problem with the dilution factor. Indeed 10^-3 is maybe not enough as a dilution factor. The problem is that we are not able to see on Quercetin plate colonies when we restreak them to make sure they are pure.
Because of the green color, we are just able to see big colonies.
The other important point is that some strains are not able to use Quercetin as a carbon source to grow. These strains won’t be selected with this technique using a single source carbon. But these strains maybe able to degrade Quercetin as a toxic compound.
Here are the first 16S PCR we did for strains 52 to 104. We have to notice that 16S rRNA is present in every bacteria, which means that this PCR is really sensitive : if there is contamination of the strain, usually e.coli in a lab, it's 16s rRNA can be amplified instead of the desired strain's. Knowing that, we did numerous negative control to ensure that our PCR were pure, and that no contamination interfere with our sequencing procedure, following exactly the same PCR protocol but without adding bacteria to the mix. Thus, all the PCR were maid under the hood, which was anyway essential for our safety, as we were working with unknown bacteria isolated from soil, in order to identify them All the PCR were made at the same time which explain why the negative control don't appear in all the gels
C stands for control : here, you can see that they are completely negative, which means no contamination should have interfered with our samples. We used a 1Kb generuler from Thermofischer: we get the desired DNA sequence weigh of
So for the next assay we chose to increase the amount of cells in the media.
We inoculate 500 µl instead of 100 µl of overnight culture in 5 ml of M9 Quercetin liquid media.
Also, concerning plating, we noticed that we most of the time had more or less 4 species of bacteria on non-selective agar. That is annoying because our aim is to have a maximum of bacteria diversity for the database.
Yellow plots are some contamination with some fungi in it. But we have at least 3 strains that seems to be very efficient in the degradation.
We took these strains and we plate them from the liquid media to some agar plate and then we re sequenced them to make sure they are the same as those previously tested.
The strains B1 and B2 are strains that were taken from Anthocyanin degradation assay.
The strains number 75 seems to be very promising.
The inoculation was the same as last assay (500ul of overnight culture.
The important thing to notice is that the strains 96, 97, 98, 99, 102, 103, 104 were taken from M9 Quercetin Plate.
We were expecting a degradation...
One reason of this failure maybe that the overnight culture were not enough fresh. Indeed, we had some troubles to organize our self to test so many strains. Some overnight culture were 5 days old before we assayed them.
In fact our 6 incubators are all busy because of the assay, so we have to wait until one is free to begin the assay.
We will organize ourself better next time to avoid this kind of problems.
For this assay we had one contamination and also no degradation from the other strains.
In this experiment we tested also the strains from F1 to F8 that were isolated on Quercetin plate. As you can see, only F4 give a good result with the liquid experiment.
Strains 118 to 136 come from non selective plate (LB agar, Tryptic Soy Agar, M9 agar)
For this experiment, we tested also 5 E. coli strains that were designed by the Protein group to express some enzyme reponsible for Anthocyane degradation. We had no results with these strains but there is maybe a problem of enzyme secretion. We should restart the experiment with cell extract.
For this experiments, the strains come from Australia. We isolated them on non selective agar plate.
For this assay, the strains 190 to 200 are mostly pseudomonas, that explains the Quercetin degradation.
We isolated a lot of Pseudomonas from strains 201 to 222 that's why we had that much degradation in Quercetin.
-The goal of the second approach (approach 2) is to select directly from samples microbes that are able to degrade Quercetin.
In this approach we are only selecting strains degrading Quercetin.
The idea is to inoculate in our singe carbon source directly some soil sample diluted at different range.
+33 1 44 41 25 22/25
igem2016parisbettencourt@gmail.com 