How do you know these transformants actually contain your recombinant vector of interest when cultured on the plate? Do you have any confirmatory information other than the fact that they are antibiotic-resistant? Have you tried cloning in the reverse orientation or in a low copy number plasmid? What is the source of your insert DNA? Some more details here might help in the troubleshooting.

We have used colony PCR to confirm the gene is present.

The insert is from a restriction digest of a synthetic DNA construct, which was supplied to us in a pUC vector. Why would cloning into a low copy number plasmid help? We are restricted in what vectors we can use for expression because of the Pichia system we are using.

Copy number could matter if your gene were undergoing homologous recombination or were harboring a toxic cDNA (assuming a cryptic promoter that you didn't anticipate to be present). Since JM109 is RecA negative, recombination shouldn't be the root cause of the apparent plasmid instability. Have you tried to see if you can successfully grow the parental pUC+insert plasmid in your strains of interest? This would help assess whether the problem is a function of the type of vector or not.

Thanks Daniel, I hadn't thought about an unexpected promoter. I'll try what you have suggested, growing up pUC and also see if I can reclone the gene in the reverse orientation to see what happens. Cloning usually just works for us, especially now we mostly just buy in synthetic constructs for subcloning.

For JM109,can we directly use from the tube for transforming with purified plasmid containing dna of interest?Is it possible to have white colonies on LB plate which is not treated with X-gal and IPTG?I say like that because the plasmid used is pklac2 and normally,.the transformation can not go well if it's blue-white.

Hi Alice

As your initial confirmation of the clones was 'colony PCR'....sometimes it gives false results...if you are attempting it from the original plates on which the transformed cells were plated.....the reason being the transformed cells / or plates also contain the ligation mix on the surface....which can give you good bands in the PCR....the ideal way is to transfer the clones on fresh plates and then do the colony PCR or do a miniprep......and see the results....

If you don't see the insert after the above mentioned steps..there is a possibility that the insert was not there in the clones on the first plate....and it might also be unstable and in not being manintained...in the cell..

bye

Hi Alice,

Probably you are facing a problem of toxicity. This may happens because some element of the plasmid (antibiotic resistance cassette, replication protein etc) have no terminator of transcription or defective and are transcribing your gene of interest. Also because some eukaryot promoters may work in bacteria and then some toxicity appears. You can try another vector or culture the colonies in minimal media may help, but I'm afraid that sometimes the only way is using a different plasmid or incorporate, if possible, a terminator of transcription upstream of your insert. You can reduce also the antibiotic concentration for regrowth, since toxicity of your gene and antibiotic can have a synergistic toxicity. if it is ampicillin you can use in liquid media 50 ug/ml or even 25, that would be enough to maintain the plasmid. Good luck!!

I disagree with Ajay. While PCR colony screening can yield false positives if your primers misanneal to your target plasmid or the E. coli genome; however these bands are usually weak and/or run at a different size than expected. Colony PCR will never pick up linear DNA from your transformation reaction. Linear DNA is degraded inside the bacteria, so could not be sustained through colony growth. Nor could you realistically recover it directly from the LB plate.

Alexandro's suggestion of a toxicity is a good point, which is why it is incumbent to see if the problem reproduces when you use the parental pUC plasmid. Good luck!

Have a go with HMS174 (DE3) cells, they have worked for me in the past with plasmids that I cannot get expressed in other E. coli cell lines.

Titrate the antibiotic. Sensitive constructs sometimes grow on plates, but not well in culture. Perhaps this has to do with the fact that the plates are somewhat older or the antibiotic was added to still relatively hot medium after autoclaving.

I then reduce the amount of antibiotic in the culture. For example if my plates were Amp100, and the colonies do not grow well in culture using that concentration, I try Amp50 and Amp25. Anyways, there must be something sensitive about what you try to clone. Hope that works.

Alice,

Sometimes low incubation temperature (18-20oC) may help. Growing of your cells may take 2-3 days.

As Peter suggested, low temp low agitation over a longer period of time sometime could reduce toxicity and let the bacteria recovers or overcome a toxic effect from your insert.

Hi Alice,

When bacteria grow on an agar plate containing an antibiotic, the cells do not see antibiotic from all directions (cells on top of a colony will have many cells between them and the plate with antibiotic) so an antibiotic plasmid does not have to be replicated as much in "all" the cells, while in a liquid culture all the cells are immersed in the antibiotic. So cells on a plate can handle a higher concentration of antibiotic, in addition, if the antibiotic was added to the agar/media while it was hot, that could result in the inactivation of some of the antibiotic reducing the effective concentration. You could try lowering the antibiotic concentration in the liquid culture by half, and just be careful to not let the culture go too long without the cells being pelleted and added to fresh media/antibiotic (the antibiotic resistance proteins are secreted into the media and can deplete the antibiotic allowing the cells to get rid of the plasmid). You said you had some colonies that grew slowly, I would try them with less antibiotic in the liquid culture (if your using 100 ug/mL of Amp, try lowering it to 50 ug/mL). It sounds like you are just using the bacteria to make the plasmid, so just let them grow longer as well (just make sure to transfer them to fresh media/antibiotic every ~8 hr or so), and reducing the temperature to 30oC may help too, see last line of the 3rd paragraph.

If cells carry a plasmid that contains a toxic gene product, and are exposed to high levels of antibiotic, they will not be able to keep the plasmid since they will need a lot of the plasmid to deal with the antibiotic and that will result in more of the toxic gene being expressed when it has a bacterial promoter (with or without a repressor). But you said your construct does not have a bacterial promoter (see below).

Your construct could have a cryptic promoter like Daniel mentioned, or it could just be that the cells are randomly transcribing the plasmid (since they are already transcribing it for the antibiotic gene), and since your construct contains a start ATG it then gets translated and that protein is "very" toxic so very little has to be made to affect the bacteria. Do you have any reason to believe that the expressed protein would be toxic to the bacteria? If a toxic protein is being made, the higher the temperature you grow the cells at the more it may be toxic, especially if it is an enzyme.

Hi Alice,

I agree with Alexandro. What we do is to insert an intron between certain exons of the target cDNA sequence. It will stop the expression of the target gene in E.coli due to lack of splicing machinery, but not affect the expression in eukaryotic cells.

Check out the methods part in this paper: J Virol. 2011 Nov;85(21):11220-34. PMID:21835787.

Hopefully it will help.

The intron suggestion is a good one, but I don't think you have enough information yet to conclude that the issue is one of gene toxicity. There could be something unusual going on with your DNA sequence affecting plasmid replication under certain conditions. Since your answer only reference a synthetic DNA, it is impossible to tell whether this sequence includes DNA regulatory sequences or is merely a synthesized cDNA. I would definitely try to figure out if it is indeed toxicity before modifying your sequence by addition of an intron.

I think you may be using the wrong antibiotic concentrations in your liquid broth. The other problem could be with your cloning. Do not rely on the PCR alone. But you could take your clone, extract the plasmid, cut it and run it on an agarose gel to check if you have both the vector and the insert. If you have both then your cloning has worked. The plating can grow false positive colonies. You can also try a lower temperature for expressing your protein. Try from anywhere at 4C to 25C and 35C. Are you using IPTG for expression. IPTG amount should not exceed 1mM or 2mM.

Good luck.

Two unrelated comments:

You could still have phage infection, but not generally, just in this experiment, if the phage comes with the insert you prepared for instance.

Did you check if the insert is there in the colonies by colony PCR, i.e. if the cloning worked?

the colonies you got for the culture in liquid medium were small or big?

If they were small: they may not be colonies bearing your construct: these colonies could grow by using the rest of medium you plated out from the transformation step or the petri dish you used for plating out were not fresh enough. Some antibiotics are not stable even in a low temperature. In this case, the competence stage of the cells or your ligation might be the reasons for the trouble. By checking if your ligation was worked, a few µl of the ligation can be used as sample for PCR. PCR product must be biger than the insert.

In the case they were big colonies but did not further grow in the liquid medium with the same selection condition, you can think about a possilbility that your constructed plasmid was not stable or product of this construct is toxic for this E.coli strain. In this case you can try to get some cells from these colonies for PCR to look for your construct. If the contruct is found again from theses colonies, they are unstable or toxic for the cells.

Daniel's comment is a good one and the final one by Tam is also worth noting. From experience, I have had a similar problem. Small colonies growing and poor / no growth in liquid. What was happening for me was the plasmid/construct was not correctly assembled.

I would definitely try and colony PCR screen your plate coliniues and check your construct or if you can, check the plasmid is actually in there. You may be having problems as fas back as digestion and ligation prior to transformation. Can you not try a positive control?

Thanks to everyone for replying. We only have this problem for this particular construct; our protocols work well 99% of the time for cloning. We are using LB plates with carbenicillin; these are prepared for us in a dedicated media kitchen. Since this problem only arises (and repeatedly) for this construct I am confident it is not related to the plates, or the cells we are using (in the sense that we are not using them correctly).

I checked and the original pUC plasmid containing the gene did grow normally in liquid culture. The gene is merely synthesised DNA and does not contain any regulatory sequences.

For everyone who suggested the ligation may not have worked, we did use colony PCR to check for the inserted gene. I assumed that in the case the ligation didn't work, the picked colonies should still have grown as that is how we generate our vector stocks - an insert is not required.

We are currently working on Daniel's suggestion to insert the gene backwards to see if that affects the behaviour. It's entirely typical for this entire project that the one time we manage to express the protein is not when we want to and it's toxic. I will also try the various temperature and antibiotic strategies outlined by several people.

Thanks again to everyone for replying.

Does it happen with other plasmids (besides pUC) using the same target sequence? I have been successful at growing colonies that did not grow at 37 degrees and agitation. The strategy was simple: I grew them with no agitation (lab bench) and at room temperature. It took longer but it was worth the wait. No need for adjusting the pH, no need for keeping them in the dark. Hope it helps.

Hi, Alice, I faced the same problem some time back while cloning a gene in XL-1 blue. Then made some modifications which worked;

1. I used a different host E.coli stbl2, incubation temperature 30oC after plating. typically stble2 colonies take ~30 h to grow.

2. After picking up the colonies and inoculation in LB medium I incubated them for 30-45 min before adding ampicillin.

good luck,

Hello Alice, Try using glucose in your LB medium, it worked for me...

1. Is it phage? respread your 'colonies' in/on media and incubate for longer time to see lysis/plaques.

2. Are the colonies you see actually the bacteria you are interested in? Very small slow growing and or transparent colonies could indicate a bacterial contamination that do not survive in your liquid culture. Increasing the antibiotic would get rid of this although may not actually improve your likleyhood of transformation. Colony PCR would indicate presence of plasmid.

3. Other than the addition of agar is your media the same? pH, Salt concentration, even sterilization conditions could (but probably shouldnt) have an effect?

4. Could there be detergent residue in your glassware? This sometimes happens if you use a washup service, either lack of rinsing or new detergents that 'cling' to the glass. Ask them if anything has changed.

5. Are you using the correct antibiotic? Do you get the same effect with a control plasmid?

6. Is your plasmid really what you think it is? Does it work for other experiments?

7. Are the products of transcription toxic? Is it possible to regulate this using different strains. I would then go with some of the other suggestions.

Goodluck.

We had (have) a similar problem with one of our inserts. We were successful by

- using stbl2 cells and incubation at 30°C

- using Circlegrow medium

- picking the *small* (slower) colonies (the big colonies had the resistance retained but the insert disrupted)

it would be worth checking if all the other clonings in the lab as well as the equipment *really* do work properly. For example, we discovered that one of our incubators generated higher Temp than indicated. Also, dependent on manufaturer, enzymes can have different efficacy and purity (e.g., they may arrive contaminated by an exonuclease that blunts the fragments).

I agreed that the problem in this case may be resulted from toxicity of the inserted genes or recombinant proteins if you could make sure that your transformants are exactly carrying recombinant plasmids by colony PCR.

you can pick the colony, dip it into 200ul LB, shake well, then spread the 200ul into plate to incubate o/n. the next day you will have a layer of bacteria. Scrip it of the bacteria to isolate DNA from there. I have done cloning many many years and met all kind of problems. email me if you have more questions: [email protected]

Good luck

Had the same problem, one out of 12 parallel constructs one was always dead in the liquid culture. What helps is the late evening transformation, early morning inoculation of colonies to small volume of warm liquid medium (0.1 ml). We explain it, they like society, like HELA cells. When they start growing never dilute them more than factor 5 at the OD of more than 0.2 (measured by eye of course).

The idea of Bailing Li is excelent!!! Could work better than mine, was set up for BL21 and protein expression.

My suggestions:

- checking carefully your ligated plasmid before transformation (correct ligation, circular plasmid (with inserts, without insert but self-ligated one, not linear one). Somtimes when you excised plasmid, UV damage, TE buffer contamination with endonuclease enzyme etc, and if you used glass milk for purification, be sure that there is no any glass milk contamination in your DNA, if there is, glass milk will absorb DNA once you do transformation and may be toxic for cell somehow.

- be sure that you used right antibiotic (e.g., 100 mg/mL amp.) in LB (20 g/L LB broth, 8 mL of glycerol 50%)

- assay competent cells if it is still in good condition (transform 50-100 ng DNA plasmid to DH5alpha or DE3, protocol is available elsewhere, spread out on LB with and without right antibiotic, colonies observation, counting etc)

Hopefully tip some points to you

If you just need the plasmid DNA with the cloned insert then Bailing's suggestion should solve the problem. Bacteria transcribe all plasmid DNA to some level regardless of whether it has a bacterial promoter or not, and they do have to transcribe the plasmid for antibiotic resistance increasing the likelihood they transcribe your cloned gene. So your gene is likely toxic to bacteria. A colony growing on a plate only sees antibiotic from below, and since beta lactamase is secreted cells above the first couple of layers on the plate can be protected and not have to transcribe or even carry the plasmid. But in a liquid culture all the cells see antibiotic equally and have to transcribe the plasmid increasing the chance that they also transcribe the cloned gene, and if that gene is toxic the cells will either die, or get rid of the plasmid and then die due to lack of antibiotic resistance.

I agree with the ideas of the above researchers. The product after ligation of the DNA into a plasmid vector and cloning with the insert could be toxic to the bacterium. Have you checked if your promoter which comes from a known source can transcribe your clone in this bacterium. You can also consider the quality of the medium or culture where you should recover your cloned product.

I think also is important the combination of vector and insert that you are using. If the vector is a high number copy plasmid and if the insert is a DNA codifying somethig with toxicity for the cell at high dosis... Have you tried to use a low copy number plasmid?

I realised that people are still adding answers to this question so wanted to come back with an update. In the end I concluded that there was something horribly toxic about the construct. All the other cloning experiments in the lab (mine and other people's) ran routinely so I am confident it was not a problem with media, incubators, cells ...

I did mean to try ligating the gene in reverse into the vector following a suggestion made early on in the discussion but we finally managed to get one colony to grow before I did that experiment. Since we are a structural biology lab not a molecular biology lab (however it feels sometimes!) I have never quite got round to completing that experiment.

Thanks again to everyone who made suggestions, the discussion was very helpful.