Silencing kinases in mammalian cells large screen

9 posts / 0 new
Last post
varsha
varsha's picture
Silencing kinases in mammalian cells large screen

Plasmodium spp responsible for causing malaria are hard to manipulate genetically although there have been advances in knockdown by RNAi. On the other hand, RNAi in the cell culture has become relatively very easy and economical to do. Malarial parasites, sporozoites, are delivered into the bloodstream thru a mosquito bite... sporozoites need to go thru a heaptic stage before they can differentiate into merozites which causes infection of RBCs in the blood.
Understanding the hepatic phase of infection is very important and may provide targets for boosting anti- malarial immunity. Using RNAi of kinases in the hepatic cells, a number of kinases were identified to be required for infection by Plasmodium, PKCgamma being one of the five. PKCgamma appears to be regulating the invasion of hepatocytes by sporozoites.

Plos Pathogens

R Bishop
R Bishop's picture
Interesting paper Varsha. I

Interesting paper Varsha. I published some work on malaria kinases that are involved in hepatocyte invasion in collaboration with Photini Sinnis at NYU Medical. Got the cover of Cell Host and Microbe last year. Cell Host and Microbe Paper

So Im going through the methods on this paper and some things caught my interest. One is this method for screening. Pretty effective. This method would be useful for all kinds of screens.

High-throughput siRNA screening of Plasmodium infection

"Huh7 cells (4500 per well) were seeded in 100 µl complete RPMI medium in optical 96-well plates (Costar) and incubated at 37°C in 5% CO2. Twenty-four h after seeding, cells were transfected with individual siRNAs in a final concentration of 100 nM per lipofection. Each siRNA was transfected in triplicate. Briefly, for each well, cell supernatant was replaced by 80 µl of serum-free culture medium without antibiotics. One µl of 10 µM siRNA diluted in 16 µl of Opti-MEM (Invitrogen) was complexed with 0.4 µl Oligofectamine (Invitrogen) diluted with 2.6 µl Opti-MEM and added onto the cells following the manufacturer's protocol. Four h after addition of the complex, 50 µl of fresh RPMI medium, supplemented with 30% FCS, 3% pen/strep, 3% non-essential amino acid, 3% glutamine and 3% HEPES were added to the cells. Two d after siRNA transfection, cells were infected with 104P. berghei sporozoites/well. Twenty-four h after infection, cells were fixed with 4% paraformaldehyde (PFA) in PBS and permeabilized with 0.2% saponin in PBS. Cell nuclei were stained with Hoechst-33342 (Molecular Probes/Invitrogen), filamentous actin was stained with Phalloidin AlexaFluor488 (Molecular Probes/Invitrogen), EEFs were detected using the mouse monoclonal antibody 2E6 and an AlexaFluor555 labeled goat anti-mouse secondary antibody (Molecular Probes/Invitrogen).

Automated image acquisition and analysis

Plates were acquired with a Discovery1 automated fluorescence microscope (Molecular Devices Corporation, CA, USA) using a 10× lens. In each well, cell nuclei, actin and EEFs were imaged in 9 fields covering a total area of 2.7×2.0 mm. Image data was analyzed using a custom MetaMorph (Molecular Devices Corporation, CA, USA) based algorithm extracting the following values for each imaged field: cell proliferation as measured by the number of nuclei per imaged field (Hoechst staining), cell confluency as measured by the percentage of the imaged field covered by actin staining and number of EEFs as number of compact, high contrast objects in a size range from 16 to 150 µm2. Within each field, the number of EEFs was normalized to the cell confluency. Normalized EEF numbers and number of nuclei were averaged between the 9 imaged fields within each well. Mean and standard deviations were calculated for each experimental triplicate.

R Bishop
R Bishop's picture
Also, they knocked down 727

Also, they knocked down 727 genes that are in the kinome. They bought these siRNAs from Ambion. Check out the cost!
Kinase siRNA library

This is an ambitious experiment to say the least. So Im wondering how each siRNA was validated?

Rb

R Bishop
R Bishop's picture
The answer is they didnt

The answer is they didnt validate them all.

"The RNAi strategy employed was validated by targeting 53 randomly chosen genes with 3 siRNAs each and performing quantitative real-time PCR (qRT-PCR) analysis to determine the level of knock-down achieved in each case. For 13 of these genes either expression was too low to be correctly assessed or primer specificity was insufficient. Most importantly, for 85% of the genes whose expression could be determined, at least 1 of the siRNAs led to an expression knock-down greater than 70%."

So that's 53/727 validated or 7.2%, but 13 weren't expressed, so 40/727 5.5% were validated by qPCR. What a crazy experiment.

R Bishop
R Bishop's picture
this is a very cool technique

this is a very cool technique. i didnt know these particles were commercially available?

In vivo RNAi

C57Bl/6 mice (male, 6–8 weeks) were treated with a single intravenous (i.v.) administration of 5 mg/kg of siRNA formulated in liposomal nanoparticles (Alnylam). Three different modified siRNAs targeting PKC? were used: siRNA#1 – 5'-GGGAcAGcAAcAAcuGcuudTsdT-3'; siRNA#2 – 5'-GGccucAcAcGucuuAAAAdTsdT-3'; siRNA#3 – 5'-cccuuAAcuAcAGcAuAuGdTsdT-3. A modified siRNA targeting luciferase was used as control (5'- cuuAcGcuGAGuAcuucGATsT-3'). Lower case letters represent 2'OMe nucleotides and “s” represents phosphorothioate linkage. Thirty-six h after siRNA administration mice were infected by i.v. injection of 2×104P. berghei sporozoites. Remaining PKC? mRNA levels, parasite load in the livers of infected mice were determined by qRT-PCR 40 h after sporozoite injection, 76 h after siRNA administration. Infection of mice treated with one PKC? siRNA was allowed to proceed onto the blood stage and parasitemia (% of infected red blood cells) was measured daily. The PKC? protein level in the liver of siRNA-treated mice was determined by Western blot.

varsha
varsha's picture
R Bishop wrote:Interesting

R Bishop wrote:

Interesting paper Varsha. I published some work on malaria kinases that are involved in hepatocyte invasion in collaboration with Photini Sinnis at NYU Medical. Got the cover of Cell Host and Microbe last year. Cell Host and Microbe Paper

So Im going through the methods on this paper and some things caught my interest. One is this method for screening. Pretty effective. This method would be useful for all kinds of screens.

High-throughput siRNA screening of Plasmodium infection

"Huh7 cells (4500 per well) were seeded in 100 µl complete RPMI medium in optical 96-well plates (Costar) and incubated at 37°C in 5% CO2. Twenty-four h after seeding, cells were transfected with individual siRNAs in a final concentration of 100 nM per lipofection. Each siRNA was transfected in triplicate. Briefly, for each well, cell supernatant was replaced by 80 µl of serum-free culture medium without antibiotics. One µl of 10 µM siRNA diluted in 16 µl of Opti-MEM (Invitrogen) was complexed with 0.4 µl Oligofectamine (Invitrogen) diluted with 2.6 µl Opti-MEM and added onto the cells following the manufacturer's protocol. Four h after addition of the complex, 50 µl of fresh RPMI medium, supplemented with 30% FCS, 3% pen/strep, 3% non-essential amino acid, 3% glutamine and 3% HEPES were added to the cells. Two d after siRNA transfection, cells were infected with 104P. berghei sporozoites/well. Twenty-four h after infection, cells were fixed with 4% paraformaldehyde (PFA) in PBS and permeabilized with 0.2% saponin in PBS. Cell nuclei were stained with Hoechst-33342 (Molecular Probes/Invitrogen), filamentous actin was stained with Phalloidin AlexaFluor488 (Molecular Probes/Invitrogen), EEFs were detected using the mouse monoclonal antibody 2E6 and an AlexaFluor555 labeled goat anti-mouse secondary antibody (Molecular Probes/Invitrogen).

Automated image acquisition and analysis

Plates were acquired with a Discovery1 automated fluorescence microscope (Molecular Devices Corporation, CA, USA) using a 10× lens. In each well, cell nuclei, actin and EEFs were imaged in 9 fields covering a total area of 2.7×2.0 mm. Image data was analyzed using a custom MetaMorph (Molecular Devices Corporation, CA, USA) based algorithm extracting the following values for each imaged field: cell proliferation as measured by the number of nuclei per imaged field (Hoechst staining), cell confluency as measured by the percentage of the imaged field covered by actin staining and number of EEFs as number of compact, high contrast objects in a size range from 16 to 150 µm2. Within each field, the number of EEFs was normalized to the cell confluency. Normalized EEF numbers and number of nuclei were averaged between the 9 imaged fields within each well. Mean and standard deviations were calculated for each experimental triplicate.

great work, Rusty. liked your paper esp the cleavage of CSP. cool!
looks like CSP vaccine may be very useful http://content.nejm.org/cgi/content/full/NEJMoa0807381?query=TOC

varsha
varsha's picture
R Bishop wrote:The answer is

R Bishop wrote:

The answer is they didnt validate them all.

"The RNAi strategy employed was validated by targeting 53 randomly chosen genes with 3 siRNAs each and performing quantitative real-time PCR (qRT-PCR) analysis to determine the level of knock-down achieved in each case. For 13 of these genes either expression was too low to be correctly assessed or primer specificity was insufficient. Most importantly, for 85% of the genes whose expression could be determined, at least 1 of the siRNAs led to an expression knock-down greater than 70%."

So that's 53/727 validated or 7.2%, but 13 weren't expressed, so 40/727 5.5% were validated by qPCR. What a crazy experiment.

what fraction of a sample should be validated e.g. rnai knockdown here. i realize that the whole process, rnai in the mouse and then validation by RT-PCR is pretty tedious (relative to what we do).
How do scientists validate microarray results i.e. what fraction of genes do they need to show bt real time RTPCR?

Jon Moulton
Jon Moulton's picture
R Bishop wrote:The answer is

R Bishop wrote:

The answer is they didnt validate them all.

So that's 53/727 validated or 7.2%, but 13 weren't expressed, so 40/727 5.5% were validated by qPCR. What a crazy experiment.

The validation doesn't seem to investigate specificity. I'd be concerned about the fraction of the phenotypic outcomes that are mediated by off-target interactions of the siRNAs, either as off-target mRNA cleavage or inhibition of the expression of off-target genes. The RISC complex only needs about 8 bases of seed complementarity to inhibit translation; there are many of those matches in a transcriptome, so rigorous validation should include both an efficacy test and a specificity test.

However, this is an exploration of high-throughput techniques so in that context I suppose we should cut some slack. The question that remains for the next phase is, how many of those siRNA created their phenotypic outcomes through effective and specific knockdown of their putative targets?

R Bishop
R Bishop's picture
Now thats something I hadn't

Now thats something I hadn't thought of, "off target effects".

However I agree definitely cut them some slack, its a monumental experiment and they got some interesting results. I'll be curious to see any follow-up papers from this group.

Rb