gene overexpression in cells: virus transduction or transfection following selec - (Aug/09/2012 )
to selectively overexpress one interested gene in a cell line, we usually have two strategies in our lab. One way is to transfect the cells with a vector carrying this gene and do antibiotics selection. after several weeks, a stably transfectant will be available. the vector is usually drived by CMV-promoter.
another way is construct the cDNA to a lentiviral vector and make lentivirus carrying this gene and then infect cells with the virus. since the gene will be integrated to the host genome as one component of viral genome, expression will be constant without antibiotics selection, at least for months.
sometimes I get confused at which to choose. 1st question is which can provide higher expression level? according to my experience, when using high titer of virus, I can get very high expression of the gene. the first method usually gives a physiological level of expression, although it is also called overexpression. how many copies are there in these two cases? When using lentiviral transduction, I am worried that the result may not truly reflect the real thing in the cell since the gene is much overexpressed. when using the first method, sometimes I cant even see any change.
Now I have one interested gene that expresses at very low level in one kind of cell lines but not another kind. I want to figure out its function first by overexpression experiments and then maybe knockdown in high expressing cells. Is this idea ok? any comments will be appreciated.
Transfection would give you a higher expression level. It is assumed that when you infect cells, each cell gets infected with a single lenti virus and hence each cell would have one copy of whatever it is that you are trying to express. But when you transfect cells, each cell can take up several of the plasmid constructs and hence end up having more copies of the the cDNA and consequently a higher level of expression.
I am not sure I follow the antibiotic part. You usually use the antibiotics to select for the cells that were transfected or infected. When you make the lenti virus, you usually clone your cDNA first into a plasmid and then use that along with the various lenti viral component plasmids to generate lenti virus. You then infect the cells with high titer lenti virus and you still have to select the cells on antibiotic to eliminate the non infected cells. I agree with you that in this case, the expression would be constant in the infected cells.
(I have never worked with over expression vectors. But have done a lot of transfections/infections with shRNA constructs and generated KD mice)
Overexpression of genes is well known to potentially cause changes in how the protein behaves - especially in the form of non-specific interactions. However, currently there isn't a lot of choice about such systems, as they allow us to determine in "isolation" some of the functions of the gene in a living system. It is possible to do single integrant transfections using systems such as the Life Technologies FLP-IN T-Rex system, which produce more "normal" levels of expression, but are typically still driven by a viral (high expression) promoter. You can also get plasmids that have mammalian promoters that allow approximately physiological level expression of the gene. In many instances you don't really need to do the stable integrations that you are performing, you should be able to get away with transient transfections.
Both systems will produce overexpression due to the promoters that are running the gene. Both systems are widely used for protein studies, so I wouldn't worry too much about the validity of using the systems. However, selecting cells with *anything* could potentially alter how those cells behave, and, if you are comparing with the parent cell line, it is important that you compare closely behaviour (attachment, growth rates, etc.) and perhaps gene expression of certain genes, to ensure that you havn't changed the selected cell line so much that you can no longer easily compare it to the parent line.
Your knockdown/expression suggestion is pretty much a standard procedure for these style experiments. Note that knocking down a gene that is being overexpressed can be difficult due to the levels of expression of viral promoters.
zodiac1505 on Tue Sep 11 20:33:02 2012 said:
Transfection would give you a higher expression level. It is assumed that when you infect cells, each cell gets infected with a single lenti virus and hence each cell would have one copy of whatever it is that you are trying to express. But when you transfect cells, each cell can take up several of the plasmid constructs and hence end up having more copies of the the cDNA and consequently a higher level of expression.
I am not sure I follow the antibiotic part. You usually use the antibiotics to select for the cells that were transfected or infected. When you make the lenti virus, you usually clone your cDNA first into a plasmid and then use that along with the various lenti viral component plasmids to generate lenti virus. You then infect the cells with high titer lenti virus and you still have to select the cells on antibiotic to eliminate the non infected cells. I agree with you that in this case, the expression would be constant in the infected cells.
(I have never worked with over expression vectors. But have done a lot of transfections/infections with shRNA constructs and generated KD mice)
in the case of transfection, the problem is, when using cells that have low transfection efficiency, after the stable clone is selected, the expression of transfected gene is usually not high. It will be much higher if you use 293FT cells I think, though I never tried.
bob1 on Tue Sep 11 22:59:40 2012 said:
Overexpression of genes is well known to potentially cause changes in how the protein behaves - especially in the form of non-specific interactions. However, currently there isn't a lot of choice about such systems, as they allow us to determine in "isolation" some of the functions of the gene in a living system. It is possible to do single integrant transfections using systems such as the Life Technologies FLP-IN T-Rex system, which produce more "normal" levels of expression, but are typically still driven by a viral (high expression) promoter. You can also get plasmids that have mammalian promoters that allow approximately physiological level expression of the gene. In many instances you don't really need to do the stable integrations that you are performing, you should be able to get away with transient transfections.
Both systems will produce overexpression due to the promoters that are running the gene. Both systems are widely used for protein studies, so I wouldn't worry too much about the validity of using the systems. However, selecting cells with *anything* could potentially alter how those cells behave, and, if you are comparing with the parent cell line, it is important that you compare closely behaviour (attachment, growth rates, etc.) and perhaps gene expression of certain genes, to ensure that you havn't changed the selected cell line so much that you can no longer easily compare it to the parent line.
Your knockdown/expression suggestion is pretty much a standard procedure for these style experiments. Note that knocking down a gene that is being overexpressed can be difficult due to the levels of expression of viral promoters.