All fall into black again

I discussed with my mentors about my recent results. They should be right. All my recent results tell me only a negative result -- there is not any relation between the transformation and ycaI and dprA mutation.



Firstly, the effect of ampicillin should be just a result of lottery. If I continue increase the repeats of my experiments, the SD will become larger and larger and the difference between the mutants and their parents will become smaller and smaller. That is, actually neither increase nor decrease when I mutate the two genes.



Secondly, the reason for why cells pretreated with antibiotics produce transformants quicker should be a reason of metabolic reason. Because antibiotics kill non-transformants and make transformants grow better. All the results has little relation to transformation mechanism. And I was cheated by the superficial phenomenon.



Nonetheless, these results absolutely demonstrated that ycaI and dprA are not necessary to this type of transformation. We still know nothing about the mechanism of natural transformation in E. coli.



This is a bad result. But I have to face to the fact.

The conclusion drawn from my older posts

I will discuss with my mentors about my recent results. After I prepare the PPT to be shown tomorrow, I am aware of the following points.

1. I have no need to add DNase in the replica plate to confirm that transformation occur before replica. Because the the replica plate with high concentration of ampicillin and the replica with high concentration of spectinomycin has the same layout when cells pretreated with antibiotics. Transformation must have happened before replica. Otherwise, one transformant can not appear in two different places. This result confirmed that transformation process should finished 10 h after plating.

2. The discrepancy of the layout in the replica containing spectinomycin and ampicillin in the wild type cells might imply that transformation in the wild type can still take place after replcia. However, in the mutant this ability was lost. But the layout discrepancy between replica-Amp and replica-Spt does not appear stably. It is repeatable but not always like this. Nevertheless, if the latter one is really chromosomal transformation. The chromosomal marker should be able to transform in this way. Nevertheless, I will try this simple experiment recently to see whether it is the real transformation.

3. similar effect of low concentration of ampicillin and spectinomycin on transformation has been confirmed recently. The data is better than that I get previously from the wild type. However, several conflicts come out. I can not explain which one is true which one is false. I will discuss them tomorrow.

New finding with some bugs

I have done 10 independent experiments to show that transformation frequency in ycaI mutant is about 5 folds higher than that of its wild type when Ampicillin concentration is 5 microgram per mL. Why my recent data shows that such effect disappeared, and transformation frequency of ycaI mutant is extremely low? Maybe the plate with which I picked up single colony to culutre is too old, I might have to restreak my plates with cells from the -70 refrigerator.

Nevertheless, after several repeats with large amount work (I am very tired this week because I have to go back home after 23:00 in these days), several points are clear. And I know some of the reasons why transformants can not be observed when protein inhibitor was used as the selection marker. Transformants appear much slower when protein inhibitor was used in transformation. The replica was done 10 h later than spreading and transformants can be observed when ampicillin was used as the selection marker 12 h later than replica. Few transformants can be observed when protein synthesis inhibitor was used as selection marker at this time. The latter type of transformants appear 12 h later than replica. That is, in all about 36 h is required for the formation of visible transformants. Transformants can be obviously observed until 48 h after plating or even later. This is quite a long time. In my previous experiments, I should have thrown plates away before the transformants visible. For transformants appeared on ampicillin plates, no such delay can be observed. All transformants appeared after 12 h later than replica. I also noticed that the 48 later appeared transformants has completely different layout comparing with other samples. Only the wild type behave like this. YcaI mutant has no such phenomenon. Both transformants appear 48 h later and the layout of transformants did not appear in YcaI mutant and samples pretreated with antibiotic (neither ampicillin pretreatment and spectinomycin pretreatment). With these information accumulated, it is quite possible that transformants appeared 48 h later should be the chromosomal transformants.

Strangely, MG1655 wild type did not produce transformants after 48 h. I need to redo this experiment and compare it with MG1655::dprA. Or I can do experiment in RR1, because recently I am very lucky in mutant construction. For more than one month, I failed to get dprA mutant in RR1. But luckily, I get it this week, and hofQ as well. Next week, I will have a lot of work to do. Firstly I need to isolate a large amount of DNA for my following experiments. But first of all, I need some rests......

Similar phenomenon in dprA mutant

For days, I tried to transform MG1655 with my previous natural transformation protocol. It did not work very well at that time. I constructed MG1655::dprA mutant and tested it with my modified transformation protocol. It seems work. Similar phenomenon in ycaI and its wild type counterpart was observed. No transformants can be observed in the wild type MG1655 after 12 h of replicate the lawn from LA to LA containing spectinomycin. However, transformants can be observed in the MG1655::dprA mutant sample. This is in consistant with the phenomenon which I observed in YcaI mutant. I have to wait until tomorrow to see the layout of the wildtype and compare it with the dprA mutant. Also, I observed the same phenomenon that cells pretreated by antibiotic appear earlier and grow quicker. The transformation induction effect of antibiotics also exist in MG1655 and MG1655::dprA.

Although the data seem good, I was troubled by some personal affairs. The fund for my phD study is available for only one year. When I tried to apply the fund for another year, so few programs about this field are available for my current identity (neither the beginner nor the postponer) while many programs supporting cancer research can be found. Without financial support, my research for the next year will get slow again. Currently, I need to spend some time to try to find an appropriate program to continue my study.

Two Evidences Proving the Activity of YcaI (com2 homologue)

Firstly I provide the two observations which strongly suggest that ycaI dependent transformation actually also participate in our transformation system.

1. 24 h after replica, only the wild type without antibiotic stimulation can still produce transformants. The ycaI mutant, the wild type and the ycaI mutant pretreated with low concentration of antibiotics gives few (most are 0) transformants at this period. It seems that antibiotics induce transformation occurring earlier. This might be true for the ycaI-independent pathway, but another evidence reveal that the two mechanism seems not be the same.

2. When I do the replica experiment, I replicate the lawn from LA plates to two plates, one contains high concentration of Ampicillin and the other contains high concentration of spectinomycin. In ycaI mutant, comparing the LA-Amp plate and LA-Spc plate, the layouts of transformants are almost the same. However, for the wild type, the layouts of transformants on LA-Amp and LA-Spc are completely different.

How to explain this phenomenon. Remember my last post, since transformation occurrs only on plates we can observe transformants in situ. Two possibilities might be able to explain this phenomenon: 1. transformants moved. 2. the two transformantion mechanism are different, so the transformants are derived from different types of competent cells. So they are intrincically different. If first possibility is true, how cells move? Does this means that type IV pili is active at this moment and largely enhanced the mobility of transformants cells? One thing is clear that ycaI-dependent transformation behavior is slower than that of ycaI-independent transformation behavior (the first evidence I posted here). If the second possibility is true, that is the transformants observed on ampicillin plate and spectinomycin plate derived from different cells. Why one type of cells can grow on ampicillin plate and the other grow only on spectinomycin? Since they have finished transformation process and both of the amp resistance gene and spc resistance genes have entered into cells, they should be able to resistant to both of the two type of transformants. The situation is similar to the paper published more than 20 years ago. The author said that they can not explain the phenomenon, that transformation of ampicillin marker in ycaI mutant with a plasmid containing USS is much lower than transformation of ampicillin marker in the wild type cell with the same plasmid without USS. But for nov resistance marker, the former for of transformation was much stronger than the latter one. Both of my results and the author's support that once cells decide to adpot one type of transformation, the other type is closed.

Dialectics—The Potential of Application Transformation on Solid Surface Technique

Some of my words might be illogical because of my poor writting. Of course, researchers concern about the factor - time. What I want to express is that few people concern on the time when transformants begin to accumulate on plates because they think that transformation occur only in liquid culture in transformation study. It is possible that this is true for other bacteria. But in E. coli, I have to keep this in mind that transformation occurs only on plates. To evaluate transformability, the number of transformants (transformation frequency and transformation efficiency) should not be the only index. There should be some other aspects which are ignored by others but we should take care . Goodgal published his paper in Nature about 35 years ago and mentioned that different types of H. influenzae mutants with transformation deficiency has different irregular shapes. However, this is just the property of the donor strain not be transfomed. Dubnau's group finds that many genes regulated by comK do not function in competence and DNA uptake and proposed K-state. But no little functions were found in these genes when the competence occur. Claverys' group claims that fratricide accompanies with competence and named X-state. This shows the effect of competence on other cells. No referrences can be found about the properties of transformants in themselves other than the transformation behavior. Is it possible that the transformants which have absorbed exogenous DNA has other abnormal phynotypes (e.g. shape, size) just during transformation? Can they be the indexs for transformation ability evaluation and classification. The transformation process might be dynamic and has an integral impact on the recipients.

We should not just think the black side of transformation on plates (not easy to monitor, not easy to be observed under microscopy and so it is difficult to assay with fluorescence or beta-gal). Nevertheless, we can see the in situ landscape: diversity of transformants just at the position where they were transformed. In liquid cultures, we have no way to discriminate the diversity, however, on plates we can open a window to see its original appearance. The techniques in developing transformation on solid surface could provide much transformation information unconscious before and in turn it calls for other novel techniques to overcome its shortcomings in that not be able to be assayed in liquid culture.

The picture getting clearer

The effect of rec-2 homologue on transformation can not be observed directly because some other factors influenced the results. Often, we try to get the final result, we test all kinds of conditions and factors, but we forget another parameter--the time. In my opinion, gene regulation should not be statical during cell life. On the contrary, it should be variable with the growth. When I counting my transformants, I was always bothered by transformants which occur after I had already counted. I have never thought of the effect of time on transformation until recently. I compared transformants at different time incubation on plates and found that the response to incubation time is so different between rec-2 homologue mutant and the wild type. Especially between 24 h to 36 h incubation on selective plates. Wild type cells that without stimulation by the first round of low concentration of antibiotics behaves weak in the first 12 h but very strong at the last 12 h. 24 h after incubation on plates, ycaI mutant stoped producing transformants, but wild type cell produced a lot of transformants. This strongly suggested that ycaI gene faciliated transformation after 24 h of incubation on plates. Although there are many other important conclusions can be drawn from my recent results, this should be the most important one. Before making any conclusions, I have to repeat all my experiments with more controls to eliminate other possibilities.

Modification on Transformation protocol opens new windows

I modified my transformation protocol by chance. Just as I found the transformation protocol by a tube to be thrown away. This time it is the plate which seems of no use tells me something really interesting. Firstly, when the stimulating ampicillin is about 5 microgram per mililiter (actually it should be 2 microgram per mililiter with particularly new). Ampicillin degrades very quick. Even after several days, you will find that one has to add two volume of that used previously for suppressing cells at the threshold concentration. It is quite complicate to explain all my data that I recently get and I do not have much time to write at this moment because I have experiments to do and I am hurry to go back for my meal. I present some of my conclusions here. Ampicillin do affect transformation. But it is difficult to detect when you just use the wild type cell and the rec-2 homologue (I use rec-2 here because many people might not be familiar with ycaI in E. coli, I hope this can be changed after my results are published). When the concentration of ampicillin is fixed at 5, the difference between wild type and the rec-2 homologue mutant is quite obvious (more than 6-fold difference without overlapping in the SD). The result can be covered if I try to test it is ampicillin or the ycaI lead to this difference. Importantly, it is ycaI mutant which has higher transformation frequency. I tested transformation with a plasmid conferring both amp and spc resistant, the second selection marker of an antibiotic which inhibits protein synthesis. Similar results were obteined since no difference if the second selection plates containing ampicillin or spectinomycine. This further confirm that ampicillin should has an effect to enlarge the difference between wild type and rec-2 mutant. This implied rec-2 homologue in E. coli is active although with an adversal effect. Often the friend and the enemy come together. I have many words to say, but I have no time now. Talk later.