1
Abstract:
Alx1 is a beta catenin dependent homeodomain protein; its expression is necessary for skeletogenesis and ingression of primary mesenchyme cells (PMCs) in Lytechinus variegates (Lv). Previously, Alx1’s role in these processes was determined by blocking its expression using morpholino antisense oligonucleotides, which prevented both ingression of PMCs and skeletogenesis. However, this approach can only be used to determine Alx1’s function during early development because these oligonucleotides are injected at fertilization resulting in persistent repression during development. Our goal is to obtain a greater understanding of Alx1’s function throughout development by temporally controlling its expression. We generated a DNA construct encoding LvAlx1 fused to green fluorescent protein (GFP), and the glucocorticoid receptor (GR) ligand binding domain. Addition of the GR ligand binding domain allows Alx1 localization to be regulated. In the absence of the hormone ligand dexamethasone, the ligand binding domain traps the fusion protein in the cytoplasm by causing it to bind to a cytoplasmic heat shock protein (HSP). Upon addition of dexamethasone, the fusion protein dissociates from the HSP and entered the nucleus where Alx1 is able to activate the transcription of other skeletogenic genes. The green fluorescent protein allows any subcellular migration of Alx1 to be observed. Dexamethasone was added to eggs that had already been injected with the mRNA for the fusion protein at various embryological stages. It was found that this system is both feasible and efficient in sea urchins for controlling the location of the fusion protein in the cell, and promises to be useful for manipulating Alx1 activity.
Introduction:
Sea Urchins as a Model Organism: There are several reasons for studying skeletogenesis in sea urchins as opposed to other genetic model organisms. First of all, sea urchin eggs and embryos are relatively easy to work with for a complex animal. This is because they are transparent, allowing all stages of development to be visible. During the early stages, the nucleus is also visible. Also, they produce a relatively small number of cells, which makes it easier to keep track of cell lineage and developmental stages. Sea urchin embryos are easily available and they develop rapidly in sea water. Eggs can be easily obtained from adult sea urchins in vast quantities. Once the eggs are fertilized, depending on the species and water temperature, the zygote will reach the larval stage within a day or two. Another reason the sea urchin is a good model for skeletogenesis is that when exogenous DNA is injected into the egg cytoplasm, it is rapidly taken up and incorporated into the embryonic genome. This incorporated DNA allows genes of interest to be expressed during development, further facilitating the use of sea urchins as a genetic system.
Sea urchins are also ideal for genetic studies because despite their simplicity, they are close relatives of humans. In fact, they are a closer relative to humans than any other non-chordate animal, including those commonly used for study such as the fruit fly or nematode, because they are a member of the Echinoderm phylum which makes them deuterosomes along with humans1. Flies and nematodes used to have the advantage over sea urchins because their genomes were sequenced first. However, the sequencing of the genome of the sea urchin species, Strongylocentotus purpuratus was completed recently, making sea urchins even more ideal for genetic studies1.
Overview of Sea Urchin Development: The previously mentioned features allow the Primary Mesenchyme cell (PMC) lineage of the sea urchin embryo to be a model system for the analysis of fate specification, cell signaling, morphogenesis, and biomineralization. A brief overview of how a sea urchin embryo develops is helpful in explaining these processes2. First, the egg is fertilized, as shown in the first frame of Figure 1. Note that the nucleus appears as an indented region near the center of the fertilized egg. Due to an uneven distribution of certain proteins and maternal mRNAs in the egg, the egg begins to divide unevenly, creating an animal and vegetal tier that will give rise to different types of cells. The vegetal tier, which makes up the bottom half of the embryo in the 16-cell stage shown in Figure 1, divides unevenly to produce macromeres and micromeres. The macomeres are the larger cells, while the micromeres are the very small cells at the very bottom of the 16-cell embryo. The micromeres will divide unevenly again, creating large and small micromeres.
Primary Mesenchyme Cells: The large micromeres produce the primary mesenchyme cells (PMCs). The PMCs ingress and eventually become the skeleton of the embryo2. Note that at the mid-gastrula stage, shown in Figure 1, there are some small clumps of cells that have gone into the hollow part of the embryo. These are PMCs. By the late gastrula stage they have formed a ring around the coelem of the embryo. They will secrete calcium carbonate at this point to make spicules, the skeletal elements found in the pluteus larvae. Another function of PMCs is signaling of vegetal invagination (seen as the bottom cells folding inwards at the early gastrula stage, frame 4 of Figure 1). Interestingly, if the PMCs are removed, another cell type will adopt the PMC fate and become the skeleton. That is, non-PMC cells will take over the role of the missing PMCs.
These processes are autonomous in that PMCs will still form skeletal elements on their normal time scale even if removed from the embryo and grown in isolation. It has been found that this behavior requires a functional Raf/MEK/ERK signaling pathway that is activated in a cell-autonomous manner3. This path results in the phosphorylation of transcription factors by ERK, which allows genes required for differentiation to be transcribed without signaling from other cells. Therefore, PMCs do not require signaling from other cell types to perform their function.
Figure 1: Stages of Sea Urchin Development. PMCs are indicated by arrows where relevant. A. Fertilized Egg. B. 16-cell stage. C. Mesenchyme blastula. D. Early gastrula. E. Mid Gastrula. F. Late Gastrula. G. Late gastrula. H. Pluteus larva
Skeleton formation in sea urchins occurs once the PMCs have ingressed and formed a ring. The skeleton of the pluteus larvae is made up of long filaments called spicules. The PMCs deposit this skeleton, which is composed of a branched network of magnesium calcite rods and spicule matrix proteins. A sulfated cell-surface glycoprotein that has been implicated in calcium uptake or deposition, as well as several other PMC-specific gene products that participate in the synthesis of the skeleton have been identified4. The entire process, including the formation of PMCs in the early stages of development, is controlled by a gene regulatory network. β-catenin a maternal transcription factor, referred to as such because it is present in the unfertilized egg, is essentially the “on” switch for the network, as shown in Figure 2. The Beta catenin activates the gene for the protein Pmar1, which is a transcriptional repressor of global repressor HesC. Pmar1 is necessary for initiating the micromere specification program and for the expression of inductive signals produced by micromeres5. In the absence of Pmar1, HesC prevents the transcription of other genes involved in skeletogenesis, but once Pmar1 is active, the rest of the network is free to operate and create a skeleton6. Every cell in the sea urchin has the same genes, but not every cell becomes skeleton-forming. This is because the network allows for the repression or expression of skeleton forming genes depending on the presence of β-catenin. β-catenin is only concentrated in a certain area of the unfertilized egg, and then is segregated into only the cells that this area gives rise to.
Figure 2: The Micromere-PMC gene regulatory network 6,7
Alx 1’s Role in the Gene Regulatory Network: An important member of the PMC gene regulatory network is Alx 1,which is a homeodomain protein and transcription factor. This protein is expressed exclusively by large micromeres, implicating it in skeletogenesis and legitimizing study of the gene as a member of the gene regulatory network7. Previous findings verify that Alx1 is required for skeletogenesis and is a member of the network. Alx1 controls genes required for epithelial-mesenchymal transition (PMC ingression) and biomineralization (skeleton formation). Using in situ hybridization, it was found that levels of Alx1 were high in PMCs following ingression and then gradually declined until late gastrula stage where the mRNA for this protein was no longer detectable7. It was determined by morpholino oligonucleotide studies and over-expression studies that Alx1 is essential at an early stage of specification7. Morpholino oligonucleotides (MOs) bind to complementary mRNA, preventing the mRNA from becoming transcribed into protein. Blocking Alx1 expression, through the use of MOs causes delayed invagination, no ingression of PMCs, and failure to form skeletal elements7. Over-expression is induced by flooding the cell with excess mRNA for the protein of interest. Recent data from Dr. Charles Ettensohn’s lab has shown that over-expression of Alx1 causes an increase in the number of PMCs, but the exact point during developmentat which Alx1 exerts this effect is unknown.
In order to study Alx1 at the point in which it naturally acts, further study is required, as well as a new technique. The problem with injection of mRNAs and morpholinos is that they cause immediate and widespread changes in expression. While these techniques were convenient for determining the overall functions of Alx1, it is important to be able to study the protein when it expressed at the appropriate time. The previous methods cause skeleton to be formed in excess, or not at all, which may have dramatic effects on other aspects of the embryo’s development. They also cause effects so profound that it is difficult to determine which more subtle roles Alx1 might play in development. It would also be beneficial to study expression of Alx1 at a later than natural time, which could be helpful in determining unknown roles Alx1 has in development.
A Hormone-Inducible System for Temporally Regulated Study of Alx1: A Hormone-Inducible System is one technique that confers the ability to induce gene expression in a temporally controlled manner. This technique has been used successfully in a whole organism to study Xenopus laevis (frog) development. The Xenopus system used a fusion protein between the hormone-binding domain of the glucocorticoid receptor (GR) and their protein of study, MyoD, a transcription factor needed to express muscle-specific actin8. Our L. variegatus system used a fusion protein between the GR ligand binding domain, Alx1, and green fluorescent protein, which allows for visualization of the migration of the fusion protein. In the absence of the hormone, dexamethasone, the fusion protein is sequestered by heat shock proteins in the cytoplasm, keeping the transcription factor, Alx1, inactive.
After hormone is added, the fusion protein will theoretically be able to enter the nucleus. At this point, the protein of interest, in this case a transcription factor, is able to exert its function. Because a transcription factor must be able to access the nuclear DNA to function, it is inert while trapped in the cytoplasm. The hormone was found to act quickly enough for the system to be useful in the study of frog transcription factors8.Because our protein of interest is also a transcription factor, and our model organism is also aquatic, which would allow the hormone to be applied in the same way, this technique promised to be feasible for studying Alx1. Our goal was to determine if the system also works in sea urchins so that it can be implemented to study Alx1 in a temporally controlled manner.
Materials and Methods:
Cloning: The first step in implementing the hormone-inducible system in sea urchins was the creation of a DNA construct. This was needed to make mRNA in vitro, which would then be used to produce the fusion protein in vivo. We used PCR to generate Alx 1 and GR ligand binding domain fragments from existing pCS2+ vector constructs. This vector includes a variety of promoters and restriction enzyme sites. The pieces were digested with Cla I and BamH I restriction enzymes and ligated to obtain the order: Alx 1, glucocorticoid receptor, GFP. This construct will be abbreviated as Alx1.GR.GFP. Similar constructs made previously in the Ettensohn lab showed that this particular order was needed for the most efficient fusion protein.
Colony Forming Unit PCR was used to screen for Escherichia coli colonies that were transformed with the Alx1.GR.GFP construct, and that had the insert in the correct orientation. The pCS2 vector conferred resistance to ampicillin. Therefore, E. coli cells were transformed with the ligation product and grown overnight at 25˚C on a plate of media containing ampicillin. Then 25 colonies were chosen at random for screening. Most of each colony was deposited in a small PCR tube containing the necessary buffers and enzymes as well as primers complementary to the pCS2 vector and the ALX1 insert. The colonies were also streaked onto a master plate and were allowed to continue growing. After completing PCR, gel loading dye was added to each sample and they were analyzed using gel electrophoresis. Only colonies transformed with pCS2 vector plus the Alx1 and ligand binding domain fragments would produce DNA fragments during PCR in large enough quantities to be seen on the gel. The bands were also checked for appropriate size: it was determined that the band produced by the vector with the insert in the correct orientation would be 1800 base pairs long. In Figure 3, three such colonies were found and can be seen in lanes 5, 6 and 17.
Figure 3: Image of Colony Forming Unit PCR gel. The bands labeled at 1800bp are PCR products resulting from colonies transformed with the complete Alx1.GR.GFP construct. The blurred bands at the bottom of each lane are primer dimmers.
The construct was produced and collected in mass from each of three colonies shown in Figure 3 by growing a large culture of each colony and then mini prepping according to the instructions provided in the Qiagen mini prep kit. The DNA samples from all three were sent out to be sequenced to ensure that no mutations occurred and to verify that the construct was exactly correct. In the meanwhile, all three samples were used to generate mRNA that could later be injected into embryos should they prove to be correct. Mini prepped samples of the construct were used to create mRNA using Ambion MEGAscript kit under RNAse-free conditions. Each completed mRNA synthesis was tested for correct length and successful production of mRNA using gel electrophoresis. As demonstrated in Figure 4, all three reactions were successful. The concentration and purity of mRNA in each sample was determined using spectrophotometry at 260 and 280nm. Then, these were diluted and prepared for the microinjection needles.
Figure 4: Gel Electrophoresis of mRNA production
Preparation of Microinjection Needles: The microinjection needles were prepared on site using small glass tubes and a needle pulling apparatus. Then a small drop of dilute mRNA with a small amount of glycerin was applied to the top of the needle and allowed to enter the needle through capillary action. Microinjection solutions contained 2 mM mRNA and 20% glycerol. Filled needles were stored at - 20˚C until use to prevent decomposition of the mRNA.
Animals: Lytechinus variegatus sea urchins were obtained from the North Carolina coast (Duke Marine Lab). Spawning was induced by intracoelomic injection with 0.5 M KCl. Sperm were kept in ice until use or were stored at 4˚C for up to several weeks until needed. Eggs were used immediately after collection. The eggs were collected and cultured in artificial seawater at 23˚C7.
Both eggs and sperm were examined under a microscope prior to use to ensure both were in good condition. To prepare eggs for injection, they were arranged in injection dishes via rowing. Eggs were rowed on dishes coated with protamine sulfate to cause the eggs to stick to the dish. Rowing was performed by mouth pipetting eggs and depositing them in approximately straight, single file lines on the dish. The eggs were injected within 5 minutes of fertilization with 45-65 pg Alx1.GR.GFP mRNA solution using a Picopritzer II microinjection apparatus (General Valve Corporation).
Confocal Microscopy: Once the embryos had reached the two-cell stage, about 1 hour into normal development, they were analyzed using confocal microscopy, which allows the embryos to be viewed in different planes of vision. The confocal microscope uses point illumination and a pinhole in an optically conjugate plane to eliminate any image that is out of focus. Because only one point is viewed at a time, the complete image must be reconstructed from a scan over the desired image. The apparatus could also be set to view fluorescence due to green florescent protein present in the embryo.
The10 μM dexamethasone solution was prepared in ethanol and a few drops of solution were added to eggs at the two-cell stage. The embryos were incubated with the hormone for 45 min prior to a second confocal microscopy examination. Two focal planes of 2, 4 and 8-cell stage embryos were analyzed under fluorescent light. In one plane of vision, the view was of cytoplasm and the nucleus was seen as a darker area behind the plane of vision. The other plane allowed for a view of inside the nucleus. Fluorescing embryos were examined to determine if GFP was in the nucleus or the cytoplasm before hormone addition and again afterwards. Images were taken in focal planes best displaying the location of the fusion protein.
Optimization of the System: Alx1 expression does not begin naturally until the hatched blastula stage. This causes several complications, including that the hatched blastula is free swimming, making it difficult for image capture, and the nuclei are less visible because the cells are smaller and more numerous at this stage. Therefore, prior to studying embryos injected with Alx1.GR.GFP mRNA, the system was optimized using mRNA from a construct that had β-Catenin in place of Alx1. For this control construct, the β-Catenin sequence was also ligated to the GR ligand binding domain and green fluorescent protein genes to produce β-Catenin.GR.GFP DNA. The upstream network component, β-Catenin, is expressed at the two-cell stage and is naturally nuclear. These aspects made β-Catenin ideal to work with because the embryos could be examined at early stages with fewer, larger cells. Alsoa construct of β-Catenin.GR.GFP was already available and, therefore, could be used to optimize the system while Alx1.GR.GFP was still being constructed. First, mRNA was produced from this construct as previously described. Likewise, eggs were injected as previously described with theβ-Catenin.GR.GFP construct to test the system in sea urchin embryos. The presence of β-Catenin in the cytoplasm prior to hormone addition would demonstrate that the glucocorticoid receptor can exclude proteins from the nucleus, since β-Catenin is naturally nuclear by the 2-cell stage. Also, if this protein is shown to migrate to the nucleus upon addition of hormone, then the system is feasible in sea urchins. Because the β-Catenin system can be tested within a few hours, it is easier to use in determining the appropriate amount of hormone to add and how long the embryos should be incubated with the hormone.