Results

pProtect is our first plasmid to be transferred, it was constructed based on pHT01 and composed of three functional modules listed below. The process by which we constucted the plasmid is described carefully in the following pages.

cotB and cotE proteins are both innate proteins of B.subtilis which will express when sporulation is induced. They help organize the outer spore surface and therefore can be chosen as the anchoring proteins used in the spore display system, fused with exogeneous proteins to display them on the spore surface.

Firstly, since cotB and cotE proteins are spore surface proteins of Bacillus subtilis and their coding genes are available in their genome, we extracted the genome of Bacillus subtilis strain WB800N, and then used this genome as a template to clone the gene sequences corresponding to cotE and cotB proteins from the genome. With the design of PCR primers, we added the cotE to the homologous arm of the pHT01 plasmid sequence.A RBS sequence was added to the 5' end of the cotE gene sequence, and the stop codon was removed at its 3' end while the linker sequence was added; the homologous arm of the tyrosinase coding sequence and the RBS sequence were added to the 5' end of the cotB gene sequence, and the stop codon was removed at its 3' end and the FLAG tag was added.

Fig1. Electrophoresis result after PCR of cotE and the first PCR electrophoresis result of cotB. The picture indicates a positive result.

Next, running a PCR on the cotB gene sequence again. 4B4 melanin-binding peptide sequence and the stop codon can be added to the 3' end of the original sequence by designing specific primers for the operation.

Fig2.Electrophoresis of cotB after the second PCR and Dsup-encoded gene after PCR. The picture indicates a positive result.

The Dsup protein existed in the Tardigrade genome, but since it was difficult to extract the the Tardigrade genome as it was inaccessible, we chose to synthesize the coding sequence of Dsup directly. We added a 6*His tag after the start codon at the 5' end of the synthesized sequence to facilitate subsequent detections. After receiving the sequence, we modified the synthesized Dsup sequence by adding part of the homology arm of cotB gene sequence and RBS sequence at the 5' end and part of the homology arm of pHT01 plasmid sequence at the 3' end.

Fig3.Image of tyrosinase sequence by PCR followed by electrophoresis. The picture indicates a positive result.

Finally, the coding sequence of tyrosinase was present in the genome of Bacillus polymorpha, but we can't get the Bacillus polymorpha strain, so we still chose to synthesize tyrosinase commercially. The sequence was synthesized by removing the start codon and adding the linker sequence at the 5' end and adding a 6*His tag before the stop codon at the 3' end for subsequent detections.

Fig4.Electrophoresis results after direct PCR of the plasmid. The blurry line suggested the impurity of target sequence.

First, we tried to linearize the plasmids directly using PCR, but the efficiency of our gel purification after PCR was too low resulting in failure to extract sufficient amount of linearized plasmids.

Fig5.Electrophoresis products of the plasmid after enzymatic digestion

After this, we re-cultured the glycerol containing plasmid pHT01, and after extracting the plasmid we changed the linearization method, we chose to use single digestion followed by PCR to achieve linearization. The pHT01 plasmid was digested by nucleic acid endonuclease BamHⅠ before PCR, but it turned out that not many vectors were successfully linearized after gum purification, so we used the Plus Multi One Step Cloning Kit to ligate it with the above four insertion sequences, finding that the four insertion sequences could be successfully ligated.However, the insertion sequences failed to ligate with the The four inserts were successfully ligated, but the inserts were not successfully ligated to the plasmid.

Fig6.Electrophoresis products of PCR after enzymatic digestion of plasmids.

Finally, we linearized the plasmid vector by double digestion after extracting a large amount of plasmids,digested the pHT01 plasmid with nucleic acid endonucleases BamHⅠ and XbaⅠ, and then ligated the plasmid vector with the four insert sequences completed by Plus Multi One Step Cloning Kit. The ligation was successful.

Fig7.Electrophoresis of the plasmid after double digestion.

When choosing to use single digestion followed by PCR to achieve linearized plasmid vectors, we redesigned the PCR primers purchased for the cotE gene sequence and the Dsup coding sequence to change the homologous arms of the partial pHT01 plasmid sequence on either side of the insert sequence. After that, the pHT01 plasmid was digested by nucleic acid endonuclease BamHⅠ before PCR. Then it was ligated with the above four insertion sequences using Plus Multi One Step Cloning Kit. We found that the four insertion sequences could be successfully ligated, but the insertion sequences were not successfully ligated with the plasmid.

Fig8.Graph of the electrophoresis results after PCR of Dsup for the second time. The blurry line indicated the impurity.

Fig9.Electrophoresis of the cotE after PCR for the second time.

Fig10.Picture above showed the electrophoresis results of the ligated components using the ligation kit on the left side and the plasmid digestion product on the right side, which shows that the largest fragment in the ligated product is the same size as the plasmid digestion product, which means that the four inserting fragments are not linked to the plasmid.

When choosing to use double digestion to achieve linearization of the plasmid vector, we redesigned the PCR primers purchased for the cotE coding sequence to alter the homologous arm of part of the pHT01 plasmid sequence at the 3' end of the insert sequence. The four fragments were then ligated together using Overlap PCR, after which the pHT01 plasmid was digested by the nucleic acid endonucleases BamHⅠ and and XbaⅠ.

Fig11.Electrophoresis of the cotE after PCR the second time

Fig12.Electrophoresis results of the four inserting fragments after Overlap PCR.

Next, the plasmid vector was ligated with the four insert sequences completed by the Plus Multi One Step. The plasmid vector was then ligated with the four insert sequences by Plus Multi One Step Cloning Kit and the ligation was found to be successful.

Fig13.Sequencing the synthesized plasmids and comparing the sequencing results with the theoretical sequences. These results altogether demonstrated the engineering success.

Firstly, the successfully ligated plasmid expression vector was transformed into the receptor E. coli strain DH5α by heat stimulation, and the transformed E. coli was incubated for 4h and then coated, and after overnight incubation, a single colony was selected for PCR using identification primers, and the plasmid expression vector was successfully transformed into E. coli after identification .

Fig14.Electrophoresis images of single colonies of transformed E. coli identified by colony PCR using cotE side identification primers

Fig15.Electrophoresis images of single colonies of transformed E. coli identified by colony PCR using Dsup side identification primers

The plasmid was extracted and transformed into Bacillus subtilis strain WB800N by electrotransformation. After incubation overnight , single colonies were selected for amplifying by PCR with identification primers. Successful transformants with the correct plasmids can then be identified.

Fig16.Electrophoresis images of single colonies of transformed Bacillus subtilis identified by colony PCR using cotE side identification primers

Fig17.Electrophoresis images of single colonies of transformed Bacillus subtilis identified by colony PCR using Dsup side identification primers

We have confirmed that our plasmid has successfully transformed into the WB800N, then the next step ahead is to induce the plasmid for expressing targeted proteins. As it is mentioned above, there's a lac operon on the pHT01, so additional IPTG had to be added.

According to the protocols we found on some published papers and previous iGEM teams, it is better to add IPTG when the B.subtilis is at OD=0.6-0.8. In theory, too early stage of the growing curve will harm the normal metabolism and threaten the survival of bacteria while too late the stage will not induce enough protein production.

23.8μl 1mM IPTG was added to 25ml LB medium where the bacteria was at proper OD index and then put the system on the shaker for 9 hours at 37℃, 200 rmp. If the induction time is too short, IPTG could not induce enough proteins while the time is too long, IPTG itself can be harmful for bacteria.

In our design, we add His-tag to Dsup protein and this protein is theoretically bind to the bacterial DNA inside the cell. So, it is a practical idea to detect Dsup protein use Western Blot. Anti-His tag antibody from mouse was used as the primary antibody and the secondary antibody binds with HRP. The result of our WB test is shown below:

Fig18. The result of Western blotting, sample names are on the top. It can be clearly recognize that two lines are visible while the WT (wildtype) with no line, despite the improper position.

From the photo above, we could recgonize a relatively clear line in the middle of the marker ranging from 55kD to 70kD in the IPTG induced sample and only-plasmid sample, while no line could be observed in the control sample, which was the wildtype WB800N with no plasmid transferred. It was not very obvious and this indicated a relatively low propotion of targeted protein.

Before we conducted the WB assay, we used Bradford protein assay to determine the concentration of total protein in solution, which was extracted directly after treatment of lysozyme. It provided a standard curve of our samples compared to the the standard concentration. Therefore, we indicate that this is because of the following reasons：firstly, as a bacteria chassis known for its production of proteins, B.subtilis had lots of proteins intrinsically; second, we added 30μl protein marker because of the total protein quntity is beyond our expectation so we decided to add more, but this in turn made the marker so bright that when we took the photo, it was indeed a trouble to modulate the time of light exposure; thirdly, there may need some adjustments to our IPTG induction time or to our plasmid.

However, Dsup protein, which is 1338bp in sequence, is 42.8kD if calculated by the rule of thumb. Therefore, we had to find out what is wrong in our experiments or the reasons behind this relatively larger results. After disccustion with our advisors and instructors, we thought the result was produced because of the following reasons: firstly, as we could see from the picture, the standard lines of the markers were also a bit close, which means the time of running the electrophoresis is a little bit short and thus there existed the possibility that the different quantity and quanlity of proteins hadn't seperated enough before the reaction stopped and membrane transfer begun. Therefore, our instructor suggested that we can treat this result as a successful positive one which is not so solid and perfect. Besides, he advised us to do another test to fully validate the expression of Dsup, thus we conducted a straight-forward SDS-PAGE to make further improvement.

Besides, as the WB indicated the low production rate of our protein, we decided to entend our induction time into 9 hours and do not refer to the total protein quantity as a parameter. By the same time, we tried to look for a stronger promoter because we suspected the lac operon used by now may not be potent and productive enough.

We conducted SDS-PAGE as a complement and verification to the result of Western blot. The procedures were much easier and we used Coomassie Brilliant Blue G250 to stain. The result is shown below:

Fig19. Result of SDS-PAGE to detect Dsup protein

It was easy to figure out that an obvious line was in between of 40kD-50kD of our positive group, while in the negative control, the line was much blurrier and nearly no line can be observed in the wildtype control. Since our target protein was at approximately 42.3kD, it provided another proof for the expression of Dsup protein.

This result was in line with our previous Western blot assay in the following aspects: firstly, there was leakage of our plasmid but the concentration is relatively low, this could be indicated from lines in negative control group in both the gel. Secondly, the line was at approximately the correct position. Since in the WB experiment, our marker line was so close to each other, we extended the time of electrophoresis and this led to better separating our markers. Our instructors agreed with us and suggested both the experiments could demonstrate the successful expression of our target proteins.

Fig20.Diagram of sporulation medium. On the left is the one sporulating and on the right hand is one haven't used yet.

In our flowing experiments, 3 groups of samples were used:
+: The bacteria with plasmid transferred and IPTG induced;
-: The bacteria with plasmid transferred and no IPTG induced;
WB: The bacteria without plasmid and no IPTG induced.

As mentioned above, IPTG was added in the time when the OD reached 0.6-0.8, and induced for 9 hours before inoculating into the sporulation medium. No extra IPTG was added after transferring to DSM because the metabolic impair IPTG may cause negative effects on the growing and sporulating bacteria.

We had successfully induced the production of spores with sporulation medium as the solution of the liquid was turning from clear to turbid.After culturing for 48 hours, we sequentially extract spores from vegetative cells with lysozyme and purify the spores several times. Pictures shown below stands for conditions before and after extraction and pufication of spores. Malachite Green and Safranin were used for endospore staining.

Fig21.Photographs of the simple staining method for spores

Endospore staning before and after the process of sporulation and purification. Spores is in green by Malachite Green dye, while B.sutilis cells in in red because of the Safranin.

Tyrosinase and the melanin-binding-peptide were the next to be demonstrated because of their display on the spore surface, and the process of sporulation has been mentioned above. Because there was his-tag linking with our tyrosinase protein, we decided to use immunofluorescence to detect it. The following is the result of our immunofluorescence test:

Fig22. Immunofluorescence assay on tyrosinase anchoring on the spore surface. Groups: Samples containing plasmid with IPTG induced; Samples containing plasmid without IPTG induced; WB800N wildtype.

We used Alexa Fluor® 488 anti-His Tag Antibody provided by BioLegend company and took the pictures using 100× fluorescence microscope. The excitation light is 488nm which is luckily to be the same as the excitation wave length of GFP, so we use GFP module in our fluorescence microscope.

The results showed a dark background with sparkles of fluorescent light, indicating for the thorough removal of the antibody which was coupled with fluorescence molecule, guaranteeing the reliability of our experiments. Therefore, existence of fluorescence in the vision of spores demonstrated the existence of his-tag. It could be clearly observed that in the sample with IPTG-induced plasmid, strong bright green fluorescence can be detected and in a comparatively large density. To the contrast, only very dim and sparse green dot can be witnessed in the sample with plasmid but no IPTG had been added. In the sample of wildtype WB880N with no plasmid transferred, no fluorescence had been detected.

The results showed above indicated the successful expression of tyrosinase on the outer surface of spores. As we integrated the Dsup proteins, tyrosinase and melanin-binding-peptide on the same plasmid induced by the same lac operon, it was no surprise for us to discover the leakage of tyrosinase expression observed with the IF assay, just like in the previous experiments of Western blot and SDS-PAGE.

We had repeated the IF experiment for two times before we finally got the results above. And in the first two attempts, we use both the primary and secondary antibodies. However, at the first time, we were unable to discover any fluorescent dot. After discussion with our advisors, we thought it might be the problem of the failing to attach the primary antibody to his-tag, since there was sediment at the bottom of the tube and may hinder the further attachment. Therefore, at second time, we shake the tube every half an hour to assist the sufficient exposure between antigen and antibody, we were able to detect fluorescence under eyepieces by confocal microscopy but failed to capture the right pictures on the computer.

Fig23. Failed immunofluorescence assay on tyrosinase proteins anchoring at the spore surface

We thought this is because: firstly, the maximum magnification of our confocal microscope is 63 times and under these conditions, the spores were absolutely too small to be clearly observed. Secondly, since the spores were so small, it was almost impossible for us to distinguish it from the spots probably on the slides. Our instructors suggested us to first distinguish the spores and then we can observe, just like researchers staining DAPI when positioning cells. It is difficult to satin the spores, however, so we had to find other ways out.

In our third experiments , we use a microscope which can both observe in phase contract mode and fluorescence mode, which brought great convenient to us because it is easy to distinguish spores under phase contract microscope. The B.subtilis is supposed to be transparent while the spores is somewhat dark and solid.

With the two pictures under the same vision, we can therefore find the spores and the spores successfully stained by the antibody. Thus, the result revealed the expression of tyrosinase on the spore surface. However, as it is indicated, the proportion of spores been stained to the total number is relatively low.

FCM was also conducted by similar procedures and the results are as below:

Fig24. Flow cytometry for the detection of tyrosinase. Experimental group (Tyrosinase+) showed significant positive signals compare to the control groups (Tyrosinase-, Tyrosinase WB800N, Tyrosinase Control)

From the results, compared with the control group, samples with IPTG-induced plasmids showed more positive signals, whilst nearly no signals had been detected in the negative controls, suggesting a similar results as the immunofluoresecnce results.

However, the peak of positive signals is just by the side of the negative peak, indicating a weak fluorescence intensity. From the result of FCM, it is more obvious that the total expression proportion of the target protein is relatively low.

After discussion, we thought this was caused by several reasons.

Firstly, during the growth and propagation of B.subtilis, the plasmid may be lost and thus no proteins can be generated.

Secondly, as we could observe a large amount of spores under the phase-contract microscopy, it reminded us that we added IPTG for induction at the time when the OD of becteria was ranging from 0.6-0.8. This concentration was before the log-growth phase so that the bacteria would still propogate after IPTG induction. With the mother cells absorbing IPTG, less IPTG is left for daughter cells so that the bacteria produced by binary fissions later could not exposed to suffient amount of IPTG. Of course, as it can be indicated by fig20, B.subtilis would go through a quick propagation after transferred into sporulation medium, the concentration of ITPG at that stage of experiment could went even lower since no extra IPTG was added given its negative effects on the bacteria. Therefore, the concentration might be too low to induce the later produced bacteria generating proteins.

Thirdly, we thought this might be caused by competitions between endogeneous cot proteins and the ones we introduced in by exogeneous plasmids. As it is mentioned in our design, we cloned cotB and cotE proteins directly form the bacterial genome. So, there’re intrinsic prioteins to express under the regulation of sporulation signals naturally and thus compete with our exogeneous proteins to occupy the space on the spores’ outer surface. According to our results from Western blot, the expression level of our target priteins was not so much. In this regard, an unclear competitive advantage of exogeneous introduced proteins over the endogeneous ones might lead to the low anchoring rate of target proteins displayed in the outer surface.

Generally, these results suggested we had successfully induced the expression and anchoring of tyrosinase on the spore surface. And there was notable expression disparity between the bacteria with IPTG-induced plasmid and the bacteria with no IPTG-induced plasmid and the wildtype.

The melanin-binding-peptide 4B4 is an oligopeptide with only 10 amino acids, and according to the procedures mentioned above, we had added the flag-tag to it for verification. As another peptide on the spore surface, we chose the immunofluorescence assay similar to the ones done on tyrosinase. Anti-flag antibody coupled with PE-CY7 was used and we observe the results both from phase contract/ fluorescecnce microscopy and the flow cytometer. The excitation light is at 562nm and the emitted light is 770nm, so theoritically, red dots could be recognized if there were 4B4 expression on the surface. The microscope we used only got the excitation wave length for RFP, which is 532nm. Given that there must be a error band for the real wave length, we determined to have a try using 532nm light as the excitation light of CY7. The results of immunofluorescence is as below.

Fig25.4B4 detection under immunofluorence microscopy

From the fig nn, it can be clearly identified the red dots in the middle of the vision in the same place as where the spores were observed by the phase contrast microscopy. Fig nn is the sample with plasmid but no IPTG induced plasmid and figmm is the control group with no plasmid at all.

The results revealed that our 4B4 peptide was successfully expressed on the spore surface compared to the nagative controls. In line with the results from tyrosinase, the proportion of spores carrying fluorescence is not so much. The possible reasons and have been illustrated above. Combined with the results we got last time with Alexa Floura 488, it was deduced that the microscopy had obervable attennuation of light in the whole vision, so even if the part around is dark without fluorescence, it could not indicate no fluorenscece existed in those places. This attenuation added difficulties for us to calculate the accurate expression rate so that a flow cytometry assay was conducted later with similar procedures.

We have conducted FCM on the flag-tag as well! The pictures indicated a similar weak positive signals in the experimental group as mentioned above.

Fig26.4B4 detection under immunofluorence microscopy

After verification of protein expression, we next turn to characterize the proteins and test their functions under the target conditions of UV radiations.

The first experiments was test on Dsup protein. This protein is found to bind the genome by its relatively flexible sequence. We firstly induced target protein expression in the three B.subilis groups (as mentioned under the subtitles of sporulation) and spreaded them on the plate. Next, the plates were exposed to a grandient of UV intensities and then cultured under 37℃ overnight. Bacteria survival rate was then analysed by counting colony numbers and make comparison between our results. The photoes taken for our 45 plates and the statistical analyses are showing below.

Fig27.Pictures of 45 plates(9*5) and statistical results

Preliminary experiments have taken place to test the best concentration of bacterial solution to be spreaded on the plates for colony counting. Concentration gradients were employed from 1 to 10^-7, and the dilustion radio shows the best between 10^3 and 10^4 when the OD=0.8

We then diluted 10^ 4 at our formal experiments, but it seems a little bit crowded for the conoly especially under no UV exposure. However, despite the errors caused by counting artificially (we try to count with the aid from some mobilphone software, but gave up for the terrible accuracy), the result can be analysed from the following aspects:

To guarantee the fidelity of our results, we had used three paralle plates for each sample under the same radiation density. However, it is out of our expectation that some of the paralle samples had large diaparity from each other on the colony numbers. This maybe partly due to the disparity in the plates which we use, because we had to prepared 600mL to make the 45 plates, but the maximum volume of conical flasks in our lab is 1L, which is not allowed to contain 600ml liquid for autoclave. Therefore, 3 flasks with the total volume of 500ml were used and thus disparity may derived from the dofference of LB medium added in the flasks. Other reasons for this difference on colony number is unknown.

However, if we trying to ignore the disparity by removing those ‘strange number’, the data seems much more explicable. Firstly, with the increase of the UV itensity, the survival rate of our bacteria decresed as expected, 0.05 kJ/cm2 were tolerable for most bacteria from all the three groups and 0.1 kJ/cm2 kick the wildtype off the game. However, when the intensity rised to 0.3 kJ/cm2 or beyond, 0.5 kJ/cm2, no bacteria cell had grwon on the plate, suggesting that DNA damage caused by UV light at these levels were fatal for the B.subtilis.

We learned that ordinary human cells can normally tolerate UV intensity at 0.03 kJ/cm2, and our bacteria grown really well under 0.05 kJ/cm2, demonstrating the rule of thumb: ‘bacteria can withstand up to 1,000 times more radiation than eukaryotic cells’. Besides, it is obvious that the samples with plasmids transferred withstand much higher intensity of radiations than the wildtype. Given the possibility of leakage which was also detected in previous verification test, it could lead to a positive result, that bacteria with the Dsup proteins expressed had a stonger resistance to the UV radiations. It must be note that here we did not detect and consider the effects cuased by the tyrosinase and 4B4, even if they were induced by the same lac operon and expressed as well. However, their exsitances in the B,sutilis cells theoritically had no influence on the cells’ resistance to radiations since it was the malenin that counts instead of the enzymes or binding peptide, thus no effect can be triggered without the right conditions to produce melanin.

However, from the results we gained under the intensity of 0.05 kJ/cm2, the survival rate of the wild-type is notably higher than the bacteria with plasmid transferred. We disccused this phenomenon with our instructors and they suggested that this could be due to the expression of exogeneous proteins in the bacteria which hinder their own metabolic pathways or consume extra nutrience of the cell. In this perspective, it is accountable for the results that with the IPTG induced, bacteria spent the most to support expression of foreign proteins and thus with least survival rate due to the burgen on metabolisms.

Under the 0.05 kJ/cm2 intensity, the experimental results were in agreement with expectations. That is, IPTG-induced cells produced Dsup proteins so that it survive the most, and a decreasing of survival rate was observed in the negative controls. However, there was little difference between the bateria with plasmids, whether adding IPTG or not. This could mostly be caused by the metabolic hinderance in samples with IPTG and plasmid leakage in the samples without.

Generally speaking, we have proved the function of Dsup basically by analysed the results of our viability test. It could increase the viability of cells under UV light and increase bacteria’s resistance and tolerance to radiations. The results was theoriotically accountable even if it departed from our original expectations. Besides, we found the leakage of plasmid is most severe in this experiments than any of the previous ones. It is suspected that the extended culturing time explained this results.

In the viability test, we simply detected the survival rate of bacteria under the protection of Dsup protein when exposed to UV light. However, whether and how the Dsup play its role in this process is unclear. Previous report showed that Dsup could bind the DNA and protect them from physical damage such as double strands breaks. Therefore, our next step was to detect whether Dsup could prevent damages in DNA.

We choose comet assay as our methods for detection. Comet assay is principally similar to electrophoresis gel and the length of comet tail is a good parameter to the degree of DNA damage. Single cell trapping was used in our experiment by the help of an advisor. Single cell trap uses the gel specifically embedded with many little cavities to just contain one cell, by this way, it could clearly show the damage of one single cell in an orderly arranged way. The pictures below showed the little hole and the ‘comet’ from 10× and 40× magnification under fluorescence microscopy.

Fig28. Comet assay. Comet head length and tail length have been labelled in the figure.

When the picture is analyzed, how long the comet tail is playing an important role of indicating the degree of DNA’s fragmentation. We measured the length of comet heads and comet tails and calculated the tail DNA % by dividing the tail length with entire length. The higher the percentage was, the debris of the DNA sequences were under exposure to 0.2kJ/cm2 UV light energy.

The number is showing in fig28 and the percentage is calculated according to the formula mentioned above. From the results, it can be clearly recognized that the percentage of the IPTG-induced bacteria is with the least tail length, indicating for a relatively intact genomic sequence. However, the bacteria without IPTG and the one without plasmid showed a significantly longer tail, which suggested the serious damages and fragmentations existed in the test.

Of course, apart from the tail length, which can be the indicator of fragmentation, the color or luminance of the tail is also an important parameter to determine the proportion of the fragmentation. Since we staining the DNA with fluorescent dye, the brighter the fluorescence, the severer the damage. It can be apparently observed from the photos that the bacterial sample with even no plasmid obtaining the brightest fluorescence. In another hand, the sample with plasmid and no IPTG, even if it nearly obtained the same degree of fragmentation, the luminance is much blurry thus indicating a potential leakage of Dsup expression. Of course, this might also because of the quenching for long time had been spent on adjusting the focus under excitation light, which may lead to fluorescence quenching.

Generally speaking, we have confirmed the protective function of the Dsup proteins. It can increase the resistance of B.subtilis to radiations, and this effect is at least partly caused by the protection on DNA to keep its integrity. In addition, we have tried to improve this protein for a better binding affinity and a better protection performance by dry-lab prediction and wet-lab experiments. For more information, please visit our Improvement of Part!

Activity of the tyrosinase was going at the same time. We found the original information of this enzyme on some papers, and figured out the best condition of its activity is under pH=7.0 while the temperature at 50℃. We realized that this could be challenged by our spore display system, since the microenvironment on the spore surface could be really different from the environment when the enzyme is dissociative and pure in the liquid.

pH and temperature, as two of the most sensitive factors affecting the activity of enzymes, were under our considerations. Therefore, we set a series of pH and temperature gradients in order to identify the best conditions that we can use to produce melanin in our following experiments.

We firstly tested the pH gradients and the results was shown in a line chart below.

Fig29. Testing the activity of tyrosinase with pH gradients.

Experimental group was the one with both plasmid and IPTG, while the controls were bacteria with plasmid but no IPTG and L-DOPA solution with the same concentration and solute.

From the chart above, it can be easily recognized that, for bacteria with IPTG-induced plasmids, pH=7.5 showed the best activity to promote catalytic reaction with the time continuing. This pH is different from the optimal pH found in original purified enzyme. On the other hand, pH=8.0 had the secondary highest increase on the absorbance, but we had found some published materials pointing out that L-DOPA can be self-polymerization at approximately pH=8. However, figure 29 seemed not to support this point.

The analysis chart below shows the difference between the case and control groups. Generally speaking, the spores with IPTG-induced plasmids obtaining the highest enzymetic activity, which is in agreement with our expectations.

Results for three groups of spores put in 5 temperature gradients were showing below.

Fig30.Testing the activity of tyrosinase with temperature gradients.

It must admitted that the results are extremely confused by the first sight. Absorbance increased quickly in both of the groups before 30 min and perform a sudden drop at 45 min, seems like a system error occurring in the middle of our experiments!

We demonstrated the function melanin-binding peptide by observation.Melanin solution was prepared preliminarily from 1mM L-DOPA, as it is mentioned above that L-DOPA can go polymerization naturally when the pH>8. Then the black liquid containing melanin molecule shall be added to the 3 sample groups to mix and incubate for just 1 min in order to prevent further reaction of L-DOPA and tyrosinase. And then several rounds of centrifuge and wash shall take place to detect the 4B4 peptide’s function to bind with the melanin molecule.

The experimental results were shown below:

Fig31.Detection of Melanin-binding-peptide. b is for the first round of washing; c is the second round of washing; d is the third round of washing. It can be clearly observed that the experimental group is the darkest at last and therefore demonstrate the function of 4B4 as the melanin-binding-peptide.

Fig31 were can easily distinguish the samples without plasmids and samples with plasmids. Despite the little number of sediments, it can still be observed that spores without plasmid is less dark than the two samples placed on its right. However, after a few round of experiments, the final color of IPTG-induced bacteria is the darkest, demonstrated the expression and functions of this little peptide.

Analysis of fig31
1. Plasmid leakage resulted in the expression of 4B4-binding peptide on the uninduced spores, which resulted in the similar degree of black color in the sample at the early stage of experiments.
2. Different concentrations of spores were observed. The higher the concentration of spores, the more precipitation and accumulation of melanin will be, thus make the color appear darker than it really is.
3. Adsorption on the spore surface is definitely happened. We suspected that the large number of proteins presented on the surface of the spore can naturally adsorb melanin to some extent, which may explain why the spore is almost black regardless of whether it is induced with IPTG or not by the time melanin is added. It may be a nonspecific and loose bind that is relatively easy to separate during subsequent centrifugation process. This also explains why IPTG-induced spores showed the deepest color after three times of washes.

Plasmid pDG1730 is an integrated plasmid of Bacillus subtilis, in which we insert the information sequence into the specific amyE site, which can then integrate into chromosomal DNA by homologous recombination. To do this, we built accordingly and came up with the following results.

Fig32.Construction of pInfo composed of pDG1730 and the information sequence.

According to the design principles of the Yin-Yang codec, we have designed an information sequence DNA containing the lyrics of Fly Me To The Moon, the length of this sequence is 792bp. The sequence was synthesized commercially and therefore, we obtained it from the plasmid. The electrophoresis plot results as follows:

Fig34.Electrophoresis results of the information sequence.

We wanted to treat the plasmid with two enzymes, BamI and HimIII, to get alinearized plasmid junction vector, in which the information sequence shall be inserted, but some problems arose in our process

In the latter few times of the experiment, our digestion results were very unsatisfactory, with the concentration determination less than 10ng/μl per time. The results of the analysis are as follows:
（1）The plasmid is constantly degraded under natural conditions and its own concentration decreases .
（2）The reuse of the enzyme leads to adecrease in the activity of the enzyme
（3）There is aproblem with the purification operation .

In our successful construction process, the plasmid linear vector used is the most derived from atube of pure plasmid, and its concentration and purity are unique, so we can obtain the corresponding vector.

The information sequence and plasmid pDG1730 have the same BamI and HimIII digestion sites at both ends, which then be linked by T4 ligase and we constuct our second plasmid to be electrotransferred, pInfo. After successfully transferring the plasmid into DH5α, we sent some samples for preliminary identifiation. We sequence the information-containing site and the result is shown in the graph below. This step was very necessary, for we not only have to identify whether we have successfully transfer the gene with target sequence, but also we have to ensure the accuracy of the sequence, for its bases are exactly where our information in and any mutation or mistake could lead to information damage. After analysing the results, it was recognized that no mutation have taken place yet.

Fig35.Preliminary sequencing results for the information sequence inserted into pInfo.

We first transferred the plasmid into DH5α for amplifying. And then extract it from the E.coli for the following electrotransformation. As what we have done in the first stage of transferring pProtect, WB800N with pProtect was cultured to OD600=8.5~9.5, then competent cells were made.

Single colonies were obtained from the plate with Spectinomycin，which potentially obtain the right plasmid with information sequence inserted in. It must be noted that the antibiotics resistance gene is situated within the recombinant sequence of amyE sites, so the colonies growing on the plate might also be the one with no information sequence in.

Step by step, we first identified the existence of pInfo inside the bacteria,to verify the success of our electrotransformation.

Fig33. Gel electrophoresis to identify the existence of pInfo in the B.subtilis after electrotransformation.

As mentioned above, our next step was to sequence the information-insertion site of the plasmid to identify the general existence of our information sequence. The result of sequencing was showing below. The result told that mutations have already been identified as the analysis pointed out few bases have changed. To our surprise, these mutations were not distributed in the sequence uniformly as the probability theory proposed. In contrast, they accumulated in one end of the sequence.

Fig36.Sequencing results

In the coming period our plans are as follows：

1. Continue the electric rotation Observe the mutation rate of the sequence. Due to the impact of the epidemic, the progress of our experiment has been delayed, and the second plasmid originally planned has not yet been perfectly introduced，we will complete the corresponding experiment according to the following conditions
a.We already have a well-constructedplasmid
b.We have successfully introduced a plasmid into WB800N
c.Our electroconversion pre-experiments have been done many times and improved

2. build a model tointerpret and minimize its mutation rate. we found that the probability of mutation in the information sequence is very high, about 10 mutations in 700 bp, which we speculate is related to the secondary structure. So we also use modeling to verify and try to solve it.

Since our information is stored in the spores, to read the information we must lyse the spores and extract the complete DNA information from it.

1.In our project, we first tried the mechanical cracking method. Since we don't have a horizontal vortex adapter, we made a simple cracking instrument using a vortex oscillator and laboratory materials. Unfortunately, after 3 attempts, we were unable to crack the spores.
2.We then tried a chemical lysis method in which we first used decoating Buffer to remove the spores' hardest coats, followed by lysozyme to lyse the remaining components of the spores and extract the DNA. However, to our surprise, after adding the decoating buffer, the original turbid bacteria liquid became very clear, and it was difficult to find the presence of the bacteria. We continued the extraction following the experimental steps and obtained a DNA content of only 5 ng/μl. After adjusting the ratio of spores to Decoating buffer, 22.4 ng/μl of 50 μl of DNA was obtained in the second extraction. DNA was successfully extracted.

Our analysis believes that there are two reasons for the failure of the mechanical method: First, the instrument is too simple, and the self-made instrument may not be able to ensure that the frequency of oscillation is the same as the set speed; The second is the size of the cracking beads, we use 0.5mm glass beads. Maybe using 0.2mm & 0.5mm beads will have a better effect.For the chemical cracking method, we do not have a definite conclusion. The main ingredients of Decoating Buffer are DTT, SDS and urea. All three substances have greater damage to proteins, and SDS can also form complexes with polysaccharides. Therefore, we speculate that it is possible that the decoating buffer destroyed the protein structure and polysaccharide structure of the spore wall, so that the spores directly lysed. Correspondingly, we detected high levels of DNA in the supernatant after adding the decoating buffer and centrifugation at 8000r/min.

Fig 37. Electrophoresis for our gemome extracted and amplified. The picture indicated that we have successfully extracted genome from spores.

As mentioned above, the process of lysing the budding spores was not easy, but we still managed to extracted a tube of 400ng of spore DNA after trying multiple methods and using a combination of methods learned from multiple protocols. The DNA extracted from the UV-irradiated budding spores was sent to the sequencing commercially and found that the mutation rate of out target sequence was extremely low and could be approximated to no mutation. However, this result could not be considered valid since there was control groups or any parallel groups. To some extent, this result confirms that Dsup, melanin and the bacterium itself are very protective of DNA and that our various modifications of Bacillus subtilis can indeed make it a good and stable information carrier. But on the other hand, since the sequencing scale was not large enough and the amount of bacterial genome was not so much, the result is till under questions.

Fig38. Sequencing results for our sample, no mutation was observed.

We will continue to improve chemical lysis methods, such as adjusting the content of the components of the Decoating buffer so that the spores can remain intact, while extracting and purifying DNA directly in the supernatant in order to obtain higher levels of DNA.
the same time, we are also collaborating with professors from other universities to use professional spore lysis instruments to read DNA information within spores.

In order to verify the biosafety of our engineered spores, we conducted the following experiments. We have demonstrated our engineering success on the increase of spores’ resistance to radiations from different aspects as mentioned above. However, this may led to potential anti-microbial issues. In out safety-proof experiments, we tested the autoclave, one of the most commonly-used strategy in labs, can still efficiently kill our spores even if their tolerance to radiations had improved. Spores once been induced with IPTG, spores with plasmid but no IPTG, and wild spores were autoclaved. The maximum temperature is 120℃, lasting for 30 mins. After the autoclaving, these spores, as well as the control groups without any sterilization procedures, were spread on the plates for further testing.

Fig39

As it was indicated on the photo, that spores without been autoclaved were rapidly proliferating and by the time of 12 hours culturing, the plates’ surfaces were almost occupied with colonies. However, in contract, there was no colonies at all on the 6 plates of different groups. These results indicated that, autoclaving not only can largely kill the ordinary spores, but also can efficiently kill our spores with an increased tolerance to radiations. Therefore, it can be a little bit relieved when handling these spores!

Bio-safety is always important so we conducted time to test for out bio-safety issues. However, there’re still many things to do in order to prove and improve our bio-security. And the things in out future to-do list is listed below:

1.We can try other sterilization test on the anti-microbial issues. Commonly used anti-biotics are worth trying. However, it must be noted that in most occasions, radiations are mutagenesis, while mutation serves as the sources for various bacteria strains to evolve and rise anti-microbial concerns among people. Whether the anti-radiation functions of bacteria could decrease the mutation rate or in turn improve the damage repair system is still unknown. In this regard, the problems related shall be very interesting for us to investigate in the future researches.

2.The ultimate solution to be against contamination caused by spore-forming organisms is to prevent spores to germinate. As in the B.subtilis cells, key genes to enable germination have already been identified and strains that can’t germinate while performing well in other aspects are also available. Therefore, we can employ these strains in our design to further avoid the accidental release or contamination. However, the strain we use now is the WB800N,it is outstanding because of its lack on various proteinase which help our proteins and peptide anchoring to the outer surface of spores. The strain whose spores can’t germinate, though, express proteinase themselves. Thus, a problem of protein induction and expression may become the dominant and need further adjustment.

After directed evolution of the Dsup protein, we found that mutations in amino acids at positions 195 and 198 could alter the binding and protective effects of Dsup on DNA. Through calculation and evaluation, we screened out 4 Dsup mutations based on our dry-lab which were most potentially to be the one with both high-affinity and stability.

Because of the time is urgent, we directly made the piont mutation on our pHT01 palsmid with four sequences already inserted. introduced mutations through point mutation strategies to obtain a plasmid containing the mutated nucleotides. We then digested the template chain using the DnpI enzyme because the DnpI enzyme specifically cleaves the methylated DNA strands, and only the template strands are methylated.

In experimental designing, we shall then transform the plasmids into Escherichia coli for amplification and expression induction. After UV irradiation as what's been mentioned above, we will viability tests to count the number of colony and calculate the survival rate for comparing the effects of different mutations on Dsup protection. If our prediction by the dry-lab model is alright, then we can expect for a promotion in the viability in the bacteria with evolved proteins compared to the old protein.

(Due to time issues, we did not complete all the experiments. Here's what we plan to do for the piont mutation on DNA sequences):
1. Amplify the plasmid using Green Taq Mix
2.Use gel electrophoresis ito dentify the plasmid
3.Use DnpI enzyme to digest template
4. Transformation of this plasmid to E. coli and enlarged culture (unfinished)

Fig40.Directed evolution of the Dsup protein

So far we have successfully clone the linearized plasmid and identified the plasmid by electrophoresis. The pictures above showed a positive result but further improvements have to be made. Transferring the plasmid into DH5alpha was sequentially done and no colony grew on the plates, indicating a failure. Our instructors suggested that this is because of the lack of target sequence. From the picture of electrophoresis, the line is dim and thin, suggesting that only little quanlity of product have been made. Since it seems pure of our products, we thought it may be because of the conditons for the pcr cycles. It is quite a pity that no time is left for us to validate our predictions before the wiki freezes, but we will definitely continue exploring the pcr contitions and try again in the nuear futures. Hope we can share the results with other teams later!

In our wet-lab experiments, we have achieved the following points:

1. We have successfully constructed pProtect, the plasmid with three functional proteins on;
2. We have verified the expression of three proteins using different methods;
3. We have characterized the functions of different proteins from different perspectives;
4. We have ligated the plasmid pInfo and transferred it in to B.subtilis.
5. We have induced sporulation and operated the lysis of spores for genome extraction.

In the future, we hope to advance our design from different of perspective. Firstly, as we all know that GCRs are one of the most fatal factors in the universe, it is a pity that we can only check the influence of the UV light in our experiments. This is mostly because that we were trapped inside the school and we couldn’t go to one of our university’s affiliated hospitals where the researchers have agreed us to borrow their instruments for X-ray imaging to irradiate our spore. Besides, we may also consider to involve different radiations together and, to the best, we can carry out our experiments under the conditions simulating the space.

Secondly, in our experiments, we only use a sequence at approximately 1000bp for the insertion, and this is definitely not enough for transporting a large amount of information. As we’ve learned from some published papers, that scientists can insert a large piece of information into the spores and we may employ their methods and adjust their method into our chassis for wider application.

Thirdly, as it can be deduced from our experiments mentioned above, some works are still undergoing, especially with regard to the genome extraction and sequencing, Dsup evolution and verification.