|Home | About | Journals | Submit | Contact Us | Français|
This unit describes a robust protocol for producing multipotent Kdr-expressing mesoderm progenitor cells in serum-free conditions and functional genomics screening using these cells. Kdr-positive cells are known to be able to differentiate into a wide array of mesoderm derivatives including, vascular endothelial cells, cardiomyocytes, hematopietic progenitors and smooth muscle cells. The efficient generation of such progenitor cells is of particular interest because it permits subsequent steps in cardiovascular development to be analyzed in detail, including deciphering the mechanisms that direct differentiation. The oligonucleotide transfection protocol used to functionally screen siRNA and microRNA libraries is a powerful tool to reveal networks of genes, signaling proteins and microRNAs that control the diversification of cardiovascular lineages from multipotent progenitors. The discussion addresses technical limitations, troubleshooting and potential applications.
Mesoderm is one of the three germ layers that constitute the building blocks of the body plan in most animals. Organs such as the heart, kidney, skeletal muscle, blood as well as the lymphatics and blood vasculature mainly derive from mesoderm cells. Understanding genetic and ageing-associated pathologies in these organ systems and the potential for regeneration motivated our study of their respective developmental programs. To date our understanding of how these different organs are formed and differentiate remains incomplete, primarily because studing these processes directly in embryos of higher vertebrates is inherently difficult. In the past few years, the development of in vitro differentiation assays using embryonic stem cells or induced pluripotent cells has greatly improved our ability to study these genetic programs. Here we describe a chemically-defined protocol that robustly produces multipotent mesoderm (Kdr-positive) cells that can subsequently differentiate into cardiomyocytes, vascular endothelial cells and vascular smooth muscle cells from mouse embryonic stem cells. We describe the use of these cells in functional genomics screens of siRNA and microRNA libraries that are potent tools for deciphering the genetic and signaling hierarchies that control development.
The purpose of Basic Protocol 1 is to prepare the mESCs for cryopreservation and scale up cultures for the differentiation and screening protocol (Basic Protocol 2 and Alternate Protocol 2).
NOTE: All steps of the protocol should be performed using proper aseptic technique.
NOTE: Before using the mESC Growth Media (mESC GM), chemically defined medium (CDM), Freezing Medium (FM), trypsin, or PBS warm to 37°C.
The purpose of this protocol is to plate mESC-derived cardiovascular progenitors into cell culture dishes and probe differentiation by functional genomics screening. mESCs cultured according to the Basic Protocol 1 above are differentiated in bulk EB culture until day 4 of differentiation, at which point they are dispersed and plated onto 384-well plates in the presence of oligonucleotide and transfection reagents to induce and probe differentiation.
NOTE: All CDM must be prepared on the day that it is used. CDM cannot be made in bulk and used throughout the differentiation assay.
Step 3 – In order to achieve the proper cell density for counting, dilute the cells such that the cell density is approximately 2.0×106 cells/mL. From this stock, a 1:5 dilution of cells to media can be made in order to accurately calculate the cell density. Assuming that the cells divide 5 times between day 0 and day 3, then each 10cm low attachment plate should have about 1×106 cells. Add 5mL of CDM for every 10cm low attachment plate combined during the day 3 protocol.
Step 13 - Transfection of siRNA against Acvr1b is used as a robust positive control for generation of a large population of Kdr-eGFP+ cells. However, for the purpose of screening siRNA or miRNA libraries, this experiment can be scaled up as described in Alterative Protocol 2.
This protocol describes the modifications to the Basic Protocol 2 that permit the use of Kdr+ progenitor cells in functional genomics screens. The protocol was developed for screening of libraries of microRNAs and mouse siRNAs that are now available from several commercial vendors.
The purpose of this protocol is to prepare the 384 plates for automated microscopy.
The purpose of this protocol is to quantify the Kdr and Foxa2 fluorescent signals to determine the proportion of cardiovascular and endodermal progenitors in the cultures, respectively. The protocol is for the InCell1000 (GE Healthcare), but is modifiable for other platforms.
Step 5 - Automated Imaging Algorithm in CyteSeer:
The image analysis routine was designed to quantify the phenotypic output - fluorescent protein expression associated with gene activity of lineage markers and immunofluorescent staining of proteins of interest. Images are routinely collected in 3 channels (e.g. blue, green and red fluorescent spectra). The algorithm can be used on any of three spectra by simply adjusting the image capture sequence on the imaging instrument. If a channel is unused, we have found it advantageous to use it as a non-specific measure of broadband fluorescence and to subtract this from data quantified from a channel with relevant fluorescence data. The image processing and analysis routine described here is specific to the Cyteseer software package, but we have implemented the basic principle of image thresholding on other commercially available image processing and quantification packages. A pipeline of Cyteseer image processing plugins used in the algorithm is outlined below in the order executed in the routine:
NB: The spelling is as displayed in Cyteseer. As an example “GrayImage” (sic).
To ensure consistent results from one differentiation to another, it is essential that an ample frozen stock is generated, and that the cells are passaged the same number of times before beginning the differentiation protocol. Moreover, one vial of stock should be used to generate a single differentiation and then be discarded. mESCs should not be kept in culture when not planning an experiment. To increase the recovery rate when thawing cells, the vial should be placed immediately into a 37°C water bath and transferred to serum containing media as quickly as possible. When passaging the cells, the critical step is the disassociation. The cells should be suspended in trypsin for the least amount of time necessary to obtain a single cell suspension, and gently resuspended in serum containing media. The quality of the culture is dependent the disassociation step.
Obtaining consistent results from one differentiation to the next is the most difficult aspect of the protocol. However, consistent results can be achieved by meticulously performing a few critical steps of the differentiation protocol (Basic Protocol 2). When growing the mESCs for differentiation, ensure that the cells never become >80% confluent as passaging then only a single time two days prior to day 0 of the protocol is crucial. Over-trypsining the cells can greatly decrease viability, and is the most critical step in executing this protocol efficiently. Additionally, the potency of each lot of Activin A should be assayed to ensure the proper activation of the pathway by transfecting an activin response element expressing construct with varying doses of Acitivin A. Without proper activation of the Activin/Nodal signaling pathway, the assay will not work properly. We have noted as much as a 10-fold variation in potency between suppliers, and individual suppliers can have several-fold lot to lot variation in potency. When both aspirating and replacing media on the cells, the speed of aspiration and dispensation should be slow and gentle to avoid lifting the cells from the plate.
One should evaluate the intensity of each signal being quantified so to avoid under/over-exposure of the signal. Saturation can compromise assay dynamic range and can be avoided by looking for saturated pixels in the specific channel or positive values too close to background signal (e.g. within 2-fold) and adjusting the exposure time appropriately. A second issue to consider is that the threshold for the channel masks in the routine we developed for Cyteseer is determined relative to the global image average pixel intensity. This means that the algorithm functions with greatest accuracy when the number of cells expressing the signal of interest comprises a minor proportion of the total image field. This approach is appropriate for cardiovascular differentiation from Kdr+ progenitors under our screen conditions. However, If this condition is dramatically altered (e.g. 90% of cells are producing positive signal), then the algorithm will score this condition very conservatively since a high proportion of responding cells in a well will skew the global average higher and, consequently, only the highest pixel values in the population will be scored. Thus, method used for setting the threshold for the channel mask is critical and might need to be altered if a high proportion of cells will produce the signal of interest.
During days 0–2 of the differentiation protocol (Basic Protocol 2), the mESCs should appear as compact spherical structures, and proliferate rapidly during the time between day 2 and day 3 after Activin A has been added to the media. After day 3 of the differentiation protocol has been completed and the cells have settled to the bottom of the plate, the cells should be evenly spaced and aggregate into EB structures that are attached to the plate about 12 hours later. On day 5 of the differentiation assay (Basic Protocol 2), eGFP fluorescence should begin to appear in cells that surround the EBs and are migrating away from the EB. eGFP fluorescence will peak on day 6 of the differentiation protocol (Basic Protocol 2) when the cells are to be fixed. After the immunofluorescence has been performed, the Kdr positive cells should be flat, broad, and non-overlapping with the Foxa2 positive cells, which appear as tightly clustered groups of cells. The Foxa2 staining is nuclear while the eGFP is cytoplasmic. The ratio of Kdr positive cells to Foxa2 positive cells is much greater in the siAcvr1b condition versus the control condition.
The time required to generate a proper stock of frozen cells is dependent on the initial confluency of the 10cm tissue culture dish. As long and the stock vial used for the expansion recovers properly, then the following time considerations will be valid. Once the cells are plated, the media should be change two days post-plating, and then allowed to expand for another two days before being trypsinized and frozen in cryotubs. After one week, the cells should be 70%–80% confluent and ready for cryopreservation although only three days are necessary to complete the protocol steps.
The time considerations for Basic Protocol 2 should take into account the time required for the cells to recover after being thawed, which usually takes two passages (the same time requirement for Basic Protocol 1). Once the cells are ready for differentiation, the assay takes an additional six days to reach the peak of Kdr-eGFP expression. Throughout the protocol, there are certain days (days 1 and 4) that the cells incubate for the full day. Therefore, eight days are required to complete the protocol steps necessary for successful generation of Kdr-eGFP positive cells.
The purpose of Alternate Protocol 2 is to generate fully differentiated mesoderm derivatives. In order to allow the cells to fully differentiate, seventeen days total will be necessary to complete the protocol. However, since there are days when the cells are left to incubate (days 1,4,6,8,10) and CDM is changed every other day after day 5 of Alternate Protocol 2, a total of nine days is necessary to complete all steps of the protocol.
The fixation and immunostaining requires four hours of incubation in addition to the time necessary to prepare the reagents for the protocol. Support Protocol 2 can be completed in a single day.
The time required to image and quantify a single assay depends on the size of the assay. Using the exposure times described in Basic Protocol 3, a single plate can be imaged in two hours with another hour necessary to adjust the Cyteseer algorithm to quantify true signal and eliminate background fluorescence.
The authors acknowledge the help of Jeff Price (SBMRI and Vala Sciences, San Diego, CA) and Casey Laris (Vala Sciences) in developing algorithms for CyteSeer. The research described was supported by grants from the NIH (HL059502 and HL113601) and the California Institute for Regenerative Medicine (CIRM, RC1-000132) to MM. AC was a postdoctoral fellow of the CIRM Training Grant at SBMRI.