Has anyone compared the effects of different plasmid purification methods on the mammalian cell transfection efficiency? I am trying to make MDA-MB-231 cell lines that stably express my proteins. After G418 selection and expansion, the western blot showed me big variations in the expression level in different constructs (these constructs are only different by one amino acid). It turned out the good expression came from the CsCl purified plasmid and the bad expression came from the Qiagen Mega Kit. Both DNAs looks very pure from the OD reading and gel running. Can any one tell me, from experience or theoretically, whether these two purification methods would make the difference? Your answer will help me decide whether I need to do CsCl plasmid purification all the time in the future.
My recommendation would be to use endotoxin free purification kit. Any suppliers will be OK but it is much better for transfection with endotoxin free kit. The importnat criterai also is the amount of supercoiled form in the preparation. I hope this can help
My recommendation would be to use endotoxin free purification kit. Any suppliers will be OK but it is much better for transfection with endotoxin free kit. The importnat criterai also is the amount of supercoiled form in the preparation. I hope this can help
Years ago I tested CsCl gradient purified plasmids vs corresponding plasmids purified with Qiagen columns. CsCl purified DNA was always superior. I did not find out why CsCl purification was better in transfections. CsCl purified DNA was also easier to sub clone fragments from. However, if the transfections are good enough for you with Qiagen purified plasmids, well then use that as much as possible because CsCl gradients are tedious.
Variation in the expression levels after selection for stable transfectants would not be due to the transfection method. Affects due to impurities would show up in percentage of intial cells transfected and cytotoxic effects after transfection. Insertion sites and random breakpoints to linearize the supercoiled plasmid for insertion into a chromosome are the likely culprits, as well as number of copies inserted in a given clone. Linearizing plasmids may decrese transfection efficiency, but alleviates the random breaks, and gives a higher percentage of expressing clones in stable transfections in my experience. I always clean up the linear DNA after digestion. Phenol:chloroform, ethanol precipitation and 75% ethanol washes will work for that. Be sure to linearize well away form the region with the expression cassette, somewhere in the plasmid backbone if possible.
CsCl purified plasmids work better, but we usually only used column purified plasmids. This was "good enough" for our experiments, but this can be different for you. Doing the CsCl purification takes time and you need a lot of starting material. Not to speak of the loads of EtBr you have to use. If you want to test some plasmids which need to be more pure, you could start with an additional EtOH precipitation.
CsCl DNA prep is always better in quality AND quantity. It may seem to cost more or take more time--but in long run, it never fails to work with CsCl DNA. It really does not cost more than kits. It also only takes at best, three extra hours of man power involvement than kit . It is all worthy it.
You didn't mention what transfection method you are using? Some methods require cleaner dna. Endotoxin free is good enough for most transfection reagents. Linearization is not necessary most of the time (used to do it routinely- don't anymore- been transfecting cells over 30 years)
I think when you have stablished resistant cells with geneticin, neither plasmid extraction method nor transfection method would be an important matter, unless you use a more powerful method like nucleofection, in which most of cells receive more than one plasmid. In my experience, endotoxin does not effect significantly on transfection efficiency. May be differences in sample preparation for western blot have made this variation.
For stable cell line, not like transient transfection, it is not necessary to consider which method is better for purification and transfection. Routine method is OK. Because after G418 selection, only the cells its genome containing your express construction (plasmid) can survive and grow up. It is a random event for cells to stably incorporate the plasmid into their genomic DNA. Thus the expression levels are mainly dependent on the microenvironment of the site where the insert located in. If you do more you would find it should has big various expression levels during the cell lines established at the same time, same construction but from different colonies. This is important and why we need to measure the expression levels to make a choice at last. Here I need to say that different construction also can, sometimes significantly, affect the expression levels but the main affecting factor is involved in the mechanism of expression regulation not others. I hope the answer is helpful.
It is possible that the differences in the expression are not the result of the differences in the DNA quality (if it got into genome it is OK) but in the presence of the fusions head-to-tail of Your construct. Check this publication: http://www.plantmethods.com/content/4/1/22
It shows very nice that many copies of the luciferase gene can be inserted into one place in the genome during transfection, which can, of course, influence the level of expression.
The method of DNA isolation is not the big problem in transfection. I have used for transfection DNA isolated using alkalic lysis method with additional steps with RNAse, Proteinase K before phenol/chloroform and with triple washing with 70% EtOH at the end and had a very nice efficiency of transfection. I have done it twice this way (without DNA isolation kit): 1. with a very big targeting plasmid for homologous recombination in murine embryonic stem cells using electroporation (it was too big plasmid to even use the Qiagen kit for isolation of big plasmids) and the efficiency was great , 2. with lentiviral plasmid (which was also quite big as I had to even additionally insert 10 kb-long enhancer!) using TurboFect from Fermentas and in control cells HEK293T the efficiency was almost 100% and in my target cells 30-40%. There was probably more pollutants in the DNA mix than in DNA isolated using kits but the quality was also very good and the concentration was much better than using kits. And no CsCl was used.
Maybe it would be interesting to check if the presence of pollutants in DNA mix can influence the amount of the head-to-tail fusions of DNA constructs during transfection?
While traditionally CsCl has been viewed as better method, arguably so, the risks compared to the added benefits are not worth it. There are many extremely good commercial kits that perform as well. I second someone else's view that endotoxin free kits are best. The drawback is the lower yield these kits provide.
Having said so, I don't think the variation you see in the expression level is attributed to the DNA purification method. I make a lot of stable clones, and even using same starting DNA constructs, I get wide variability depending on copy number and location of integration in genome.
If you haven't selected single cell clones, I strongly advice you to do so and select ones with similar expression.
Thank you guys so much for the valuable input. I learned a lot. It is such a good discussion. Mean while, I tested my transient transfection levels and compared the transfection efficiency by using CsCl purified DNA vs Qiagen mega kit purified DNA on same construct. It turned out the Qiagen mega kit purified DNA was as good as CsCl purified DNA. The variation I saw earlier was kind of a random thing. I noticed big variation in transfection efficiency from time to time even in my transient transfection. I wonder if it is my pipetting variation. Because my DNA stock concentrations are all quite high, from 1.7ug/ul to 3.7 ug/ul, so each time for one 6 well transfection, I usually end up pipetting 0.5-1ul. That could be an issue. I am planning to dilute the DNA. My conclusion is good quality DNA is important but Qiagen kit is good enough. For those of you who don't think the DNA quality is a big issue in stable cell line generation, here I even have a supporting result. For one construct, I tested Qiagen miniprepped DNA and got beautiful stable clone. So it depends, for some really easy case, miniprep DNA is good enough but for some really difficult case, higher quality DNA is necessary.
Diluting your stocks is a good idea, so you handle bigger volumes. Aliquot them as well to avoid freeze-thaw cycles, which will affect the quality of your DNA as well.
I don´t think that the method of purification may cause variation in the level of gene expression. Culumn purification is sufficient, and I have obtained more then 80% of transfection efficiency using culumn purification. Different levels of expression are logic since the plasmid is integred in different sites in the chromosome, and multicopies are integred also. I have experimented that inceasing the concentration of selection marker raises the number of plasmid copies in the genome, providing higher expression levels.
Last time I saw cesium chloride was Middle Ages, for pure super-coiled form was perfect;) If you expect any contaminants I would recommend you isopropanol (isopropyl alcohol) to precipitate. You just add 5-10 volumes of your sample volume, freeze in liquid nitrogen, spin down (the highest g-force bench centrifuge you can find) during lunch break. Than one ethanol wash (70%). You don't add any salt here can affect your transfection. Never try it with 1-10 nanogram DNA concentration without precipitation enhancer.
- Badges
- Science topic