Draft of 2nd Project report - 2nd version
Abstract
During the development of Drosophila Melanogaster embryo, the product of Decapentaplegic (Dpp) gene plays an essential role in the final organization of dorsal-ventral pattern. The Dpp is a homologue of Bone Morphogenetic Proteins 2 and 4 (BMP 2/4) and acts as an extracellular morphogen to form a step gradient with the effect of another BMP homologue Screw (Scw) and some other regulators including short gastrulation (Sog), twisted gastrulation (Tsg) and Tolloid (Tld). These regulators are all products of zygotic genes, and Tld is metalloprotease. Moreover, two type IV collagens Dcg1 and Viking (Vkg) may also participate in the BMP signalling. Interactions between Tsg, Vkg, Sog and Dpp were investigated by using GST pull down assay, and the exact region where the interaction occurs was tried to find out in this study. The enhancement of Sog to Tsg on affinity to Dpp was established, and it is observed that Sog reduces the binding activity of Vkg C terminus part to Dpp. In situ hybridization was also performed on Vkg and Dcg1 mutated embryos and wild type embryos to compare the expression patterns of Race and U-shaped (Ush) genes which are reflections to the activity pattern of Dpp. Differences between the wild type and mutated embryos were observed, and the gene level patterns in the mutated embryos are found to be different too. And a conserved site in type IV collagens/procollagens among different species was found.
Introduction
As kind of insect, Drosophila melanogaster undergoes gastrulation germ layer differentiation to form endoderm, mesoderm, ectoderm and then their organelles during embryo development. This differentiation process rises from the different spatial positions of the cells in the embryo and is mostly controlled by morphogens. (Ashe and Briscoe, 2006)
During the early stages of embryo Drosophila development, the product of spätzle gene acts as a morphogen to form the Dorsal (Dl) activity gradient and thus establishes the dorsal-ventral pattern (Morisato and Anderson, 1994; Raftery and Sutherland, 2003). Then, the control is taken away by other morphogens. In previous studies, it has been reported that two homologs of Bone Morphogenetic Proteins 2 and 4 (BMP 2/4), Decapentaplegic (Dpp) and Screw (Scw) which are members of TGF-β superfamily, are the control factors for the formation of dorsoal ectoderm in Drosophila. Further researches indicated that Dpp plays the main role since all dorsal cells assume ventral lateral fates and the mutant phenotype of the zygotically required genes appears the most severely with the lack of Dpp (Arora and Nusslein-volhard, 1992; Ferguson and Anderson, 1992a). Based on these discoveries, the studies in this field are mainly emphasized on the Dpp.
The signal transduction of Dpp/Scw has been well described now. The Dpp/Scw functions by binding to a receptor complex which is supposed to be composed of type I and type II transmembrane serine/threonine kinases (Hogan, 1996; Nakayama et al., 2000). It has been elucidated that in Drosophila, the type I receptor Thick veins (Tkv) interact with the Dpp molecule and the type II receptor Punt (Put) to form a heteromeric complex. The active Put within this complex phosphorylate the Tkv consequently and thus activates the associated type I kinases which will phosphorylate the cytoplasmic protein Mothers against Dpp (Mad). The Mad which belongs to the Smad superfamily then function inside of the cell and lead to the differentiation and in this way the signal of Dpp is transduced. Moreover, in the signal transduction of Scw, it seems that another type I receptor Saxophone (Sax) takes the place of Tkv in Dpp signal transduction. However experiment also showed that high level of Tkv is able to compensate the loss of Sax, and indicates that these two morphogens shares a similar intracellular pathway. (Affolter et al., 2001) Moreover, it is thus able to visualize the Dpp/Scw activity level by detecting the level of Smad due to these discoveries on the Dpp signal transduction.
Studies had showed that after the dorsal-ventral (DV) axes is patterned, the effect of the morphogenetic Dpp depends on the local concentration level. In detail, high Dpp level leads to the fate of being amnioserosa whilst lower levels make the cell become dorsal ectoderm tissue (Ashe et al., 2000; Shimmi and O'Connor, 2003). And it has also been reported that the Dpp and Scw function much more effectively by forming a Dpp-Scw heterodimer which is able to induce a 10 to 100 times higher phosphorylated Mad accumulation than Dpp or Scw homodimer do within the cells exposed to these molecules. The receptor Sax and Tkv are suggested to be related to this phenomenon, different kinds of receptor complexes may form. (Shimmi et al., 2005; Wang and Ferguson, 2005)
Based on this knowledge, the regulation of dorsal ectoderm and amnioserosa formation by BMP signalling pathway is suggested to be achieved by the gradient distribution of Dpp/Scw. However, an interesting and indispensable phenomenon of the gradient of Dpp is that it is not smoothly distributed. The visualized Dpp distribution in dorsal ectoderm formation stage Drosophila embryos showed that Dpp activity has a peak level at the 8-10 dorsalmost cells which will later form the amnioserosa, and there is a sharp fall of Dpp level at the edge of the dorsalmost cells where cells on the different sides of the confine receive high and very low BMP signal (Shimmi and O'Connor, 2003). This step gradient specifies the very small region to amnioserosa and ensures the normal development of amnioserosa and dorsal epidermis.
However, the Dpp protein itself is nonautonomous and is only able to diffuse under a narrow limitation because it is easily captured by its receptor without protection. The Dpp protein direct visualization is possible in present time, and the result shows that there is an even level of Dpp throughout the dorsal 40% of Drosophila embryo before the late cellularisation stage when Dpp starts to function. Moreover, though this uniform distribution will lost in later stages of normal embryo development, it can be maintained by Scw mutation, and this also prompts the involvement of Scw in this patterning. The pattern of Smad observed in the embryo also agrees this distribution. (Raftery and Sutherland, 2003; Shimmi et al., 2005; Wang and Ferguson, 2005) It is still not possible to detect and visualize the intracellular concentration of Dpp or Scw directly so far, but the phenomenon had been observed already supports the hypothesis that the Dpp may be evenly secreted in the dorsal part of the embryo after the patterning of DV axes just as the same as before. In this way, there must be some other regulators participated in the formation of the Dpp gradient. Furthermore, according to the classical model of morphogen distribution which concerns the diffusion alone, specified cell or region produces the morphogen and the morphogen diffuses to the tissue nearby via extracellular or intracellular pathway. Thus a morphogen concentration gradient which peaks at the secreting cell and smoothly deduced along with the increasing distance to the secreting cell is formed(Ashe and Briscoe, 2006). The step gradient of Dpp activity is incompatible with this model, thus implies that the formation of this gradient should be created based on another mechanism which is not simply considering the diffusion only.
There is a good hypothesis which gives the explanation for this step gradient that had been widely accepted now. In this hypothesis, three regulators are accounted to be responsible for the formation of the BMP step gradient (Marqués et al., 1997; Shimmi and O'Connor, 2003). These three regulators are the products of three zygotic genes: short gastrulation (Sog), twisted gastrulation (Tsg) and Tolloid (Tld) (Ashe, 2005). And the product of Tld gene is a metalloprotease. They are able to regulate the BMP signalling activity and are essential in the determination of the peak level and the step gradient distribution of BMPs since the loss of any one of these three regulators will lead to the loss of the step gradient. It is long known that Sog gene is a homologue of Xenopus Chordin (Chd) gene which produces antagonist of BMPs both structural and functional. (Ferguson and Anderson, 1992b; Piccolo et al., 1996). And later studies discovered that Tsg is also a secreted protein which acts as extracellular antagonists of BMP (Chang et al., 2001; Ross et al., 2001). These works gave the evidence that Sog and Tsg bind to BMP directly and inhibit the interaction between BMP and its receptor (Tkv/Sax and Put) to form a complex in order to block the signal transduction (Ashe, 2005). An important discovery in the binding mechanism is that the common binding domain among them is reported to be a cysteine-rich repeats (CRs) (Larraín et al., 2000). However, these interactions will not destroy the BMP molecule itself, and the formed complex can be divided into the original single molecular to regain its function too. In Drosophila, this cleavage process is operated by the protease Tld which was first recognized as an activator of Dpp (Ferguson and Anderson, 1992a).
Function of these regulators has been investigated in great detail. Start from Sog, it is produced at the presumptive neuroectoderm and the distribution of Sog in Drosophila embryos had been directly shown to be in a graded fashion and limited by Tld in dorsal region (Srinivasan et al., 2002). The distribution of Sog protein agrees with the classical passive morphogen distribution model since the level of Sog reach the highest at the producing region and falls whilst the distance from the presumptive neuroectoderm increases. However, the Sog detected is thought to be in the complex including Sog and Dpp and Sog single molecule itself will be degraded via endocytosis. Thus it is also suggested that the shaping of the lowest level of Sog at the dorsalmost region is associated by the protease Tld which is able to cleave the complex and release Sog to be degraded. The interaction between Sog and Dpp here not only contributes to the formation of Sog gradient, but also help the diffusion of Dpp. As mentioned, either the Dpp homodimer or Dpp-Scw heterodimer has the activity to bind the receptor on cell membrane and then are caught by the receptors and lost the diffusion ability. Thus, the binding between Sog and Dpp on the one hand is an inhibition on Dpp, on the another hand is a protection of Dpp signalling activity and helps the diffusion of Dpp. Since the Tld restores the activity of Dpp by cleaving the complex when the complex reaches that region, a long-range enhancement and local inhibition of Dpp activity by Sog is then created. (Ashe, 2002; Ashe and Levine, 1999)
A notable fact is that Tld is distributed evenly in the dorsal ectoderm formation embryo (Shimell et al., 1991), thus the cleavage rate of the Sog-Dpp complex is the same throughout the embryo. However, only in the dorsalmost region where is lack of Sog, the Dpp released by Tld cleavage is able to bind to the receptor before forming the complex with the Sog protein again.
Later studies demonstrated that Tsg is another inhibitor of Dpp activity, its role is generally recognized to be similar to Sog. Tsg and Sog are enhancers to each other since the loss of either Tsg or Sog reduces the efficiency of Dpp inhibition in the embryo and they show strong Dpp affinity when existing together. Thus the mechanism of the enhancement is suggested to be the formation of a Tsg-Sog-Dpp complex (Chang et al., 2001; Shimmi and O'Connor, 2003). The knowledge on the complex is further advanced by the recent discoveries mentioned above that gave evidence for the existence of Tsg-Sog-Dpp-Scw complex. It is also discovered that the affinity of Sog/Tsg for Dpp and Scw homodimer is weaker than Dpp-Scw heterodimer.
And some new discoveries recently gave addition to this hypothesis: Wang and Ferguson group (Wang and Ferguson, 2005) found that with the lack of Scw the interpreted results of their experiments support the mechanism that Dpp and Scw homodimers function cooperatively in the BMP signalling, and the report from Shimmi et al. (Shimmi et al., 2005) group gave evidence for the actual existence of Dpp-Scw heterodimer. According to these reports, it is suggested there are three kinds of receptors on the cell membrane for the BMPs which are able to specially bind to Dpp homodimer, Scw homodimer or Dpp-Scw heterodimer. There is a positive feedback with unclear mechanism between the interaction of Dpp-Scw heterodimer and the Dpp-Scw-receptor. Combining the potent signalling of Dpp-Scw heterodimer and the effect of Sog, Tsg and Tld, this feedback helps the formation of peak BMPs signalling level in the dorsalmost cells. And in contrary, in other regions where the Sog/Tsg level is higher the positive feedback is not activated due to the lower free BMPs level, and the main functioning BMPs are Dpp and Scw homodimers. Thus the signalling strength is much weaker in these regions. (Ashe, 2005; Wang and Ferguson, 2005)
This hypothesis is very reasonable. However, two proteins, Dcg1 (Blumberg et al., 1988) and Viking (Vkg) (Yasothornsrikula et al., 1997), may be put into the patterning of BMPs in Drosophila embryos. Vkg and Dcg1 are type IV collagens which are the components of extracellular matrix (Timpl and Brown, 1996) and may limit the diffusion of Dpp/Scw via binding. This idea rose from the observation that in vertebrates BMP 2B/4 is able to bind to type IV collagens (Paralkar et al., 1992). This result was acquired in human cells, but because the BMP signalling pathway is a common and conserved pathway among the animals undergoing gastrulation, it is possible that type IV collagens in Drosophila are also able to bind to BMPs. However, the type IV collagens have not been regarded to be important in the BMP signalling pathway since they lack the widely accepted BMP binding region, CRs region, in the past years whilst some other procollagen types with the CRs region are considered to be necessary for the normal BMP route. And though previous studies had given some information about the distribution of type IV procollagens in Drosophila, the investigation in type IV collagens expression in Drosophila early embryos was failed (Lunstrum et al., 1988).
Here, to further understand the BMP signalling pathway, function of Vkg and Dcg1 was investigated by deletion experiments and interactions among the BMPs and regulators (including type IV collagen) were tested in more detail. Vkg and Dcg1 mutant Drosophila with lower Vkg or Dcg1 activity were first prepared (Wang and Ashe, submitted). Marker genes Race (Tatei et al., 1994) and U-shaped (Ush) (Cubadda et al., 1997) were detected using in situ hybridization method in these Vkg and Dcg1 mutated Drosophila embryos to visualize the Dpp patterning and threshold (Ashe et al., 2000). The interactions among Tsg, Sog, Vkg and Dpp were tested by GST pull down assay. To study the interactions more clearly, the Dpp and Sog protein were tried to be spitted into smaller fragments and GST pull down assay was performed on them consequently to find out and confirm where the interaction region is on these proteins. The Vkg used in the interaction analyse was the C terminus Vkg fragment since this part is reported to contain the interaction region and the full length Vkg protein is insoluble (Wang and Ashe, submitted). Moreover, the type IV collagens in different species were compared to find out whether there is a conserved site which may related to the BMP signalling pathway.
Materials & Methods
Mutant Drosophila melanogaster
The yw67c23 Drosophila from lab stock is used as the wild type control. Vkgk00236 and Dcg1k00405 flies (gift from Xiaomeng Wang, University of Manchester) are used as Vkg and Dcg1 mutants. These two mutants have weaker Vkg or Dcg1 expression, because the completely disappearance of functional Vkg or Dcg1 leads to flies unable to lay eggs. (Wang and Ashe, submitted)
Embryo collection
The embryos from the wild type, Vkg mutant and Dcg1 mutant flies were collected every 2 hours, and the flies were kept in 25 ºC incubator. The collected embryos were left to develop for another period equal to 2 hours at 25 ºC before fixation. The embryos were fixed using fixing buffer (check recipe), 37% formaldehyde and heptane mixture with the volume ratio 4:1:3 by fiercely shaking for 25 minutes and stored in ethanol at -20 ºC. Thus the embryos collected were at the development stages between 2-4 hours. A small number of overnight embryos were also fixed and stored with the 2-4 hours embryos as the control.
In situ hybridization
In situ hybridization was performed with the embryos collected described above. About 50 µL of embryos in volume was used for each in situ test and the hybridization buffer was prepared prior to the in situ hybridization. The recipe of hybridization buffer is (in volume percentage): formamide 50%, 20x SSC buffer 25%, ssDNA 1%, Heparin 0.5%, Tween 0.1%, and H2O 37.3%. Race and Ush digoxigenin-labelled RNA probes were used to visualize the distribution of Race and Ush gene.
The embryos were put into 1.5 mL eppendorf tube, washed with ethanol for 4 times and then methanol for 3 times, each wash lasts for 3 minutes. Then the embryos were shake in 1 mL PBT (0.5% tween in 1x PBS) for 1 hour, continued by shaking in 0.5 mL PBT and 0.5 mL hybridization buffer for 10 minutes, and at last the embryos were washed in 1 mL hybridization buffer for at least 2 minutes.
After the washing, embryos were prehybridized in 1 mL hybridization buffer in 55 ºC oven for at least 1 hour, the tubes with embryos inside was inverted every 15 minutes. And at the same time the RNA probes were denatured at 80 ºC for 4 minutes.
When the prehybridization is done, the hybridization buffer was removed and 1 µL of probe with 50 µL of hybridization buffer was added to the embryos. The embryos were then hybridized in 55 ºC oven for 18 hours.
When the hybridization is finished, 1 mL of 55 ºC hybridization buffer was added to the embryos. The embryos were kept in the 55 ºC oven for another 1 hour, and the tube was inverted every 15 minutes. Then the embryos were washed with 1 mL hybridization buffer for 4 times, for each time the embryos were incubated in the 55 ºC oven for 30 minutes. After this, the embryos were washed with 0.5 mL PBT + 0.5 mL hybridization buffer for 2 times, each time the embryos were shaked for 15 minutes. Finally the embryos were washed with 1 mL PBT for 5 times, each time the embryos were shaked for 10 minutes.
After the washing, 2.5 µL of anti-digoxigenin antibody was added with 0.5 mL PBT, and the embryos were shaked at 4 ºC for at least 12 hours. And then the embryos were ready for staining.
The staining buffer was made freshly before staining, the recipe is: 100 mM NaCl, 50mM MgCl2, 100mM pH 9.5 Tris, 0.1% (in volume) tween and 99.9% (in volume) H2O.
For staining, the embryos were washed with 1 mL PBT for 4 times, 15 minutes each time. Then the staining buffer was used to wash the embryos for 2 times, 5 minutes each time.
After the washing, 400 µL staining buffer with 3.6 µL NBT and 2.8 µL BCIP were added into the tube, and was mixed to begin staining. To stop staining, the embryos were transferred into 1 mL PBT.
After staining, the embryos were washed with 1 mL PBT for 2 times, continued with washing with 0.5 mL PBT + 0.75 mL ethanol (the ethanol was added in 3 times, 0.25 mL each time) once, and 1 mL ethanol for 10 times (at least 5 minutes each time).
When washing is finished, ethanol was removed, and the embryos were washed with xylene once. Then approximate 400 µL of premount buffer (recipe) was added and mixed with the embryos. The embryos were subsequently mounted on slides and left to be dried.
The slides were observed under microscope and images were taken at the same time.
Plasmids
GST protein expression plasmid pGEX4T-3 (lab stock) and pGEX4T-1 (lab stock) was used for GST fusion protein construction cloning. The production of Dpp and Sog proteins fragments were based on two Drosophila cell expression constructs which express full length Dpp with Hemagglutinin (HA) tag (Dpp-HA) and full length Sog with Myc tag (Sog-Myc) (Yu et al., 2000).
The premature Dpp protein will be cleaved in cytosol and the CDS region of the Dpp gene which encodes the mature functional protein is localized at the very end of the full Dpp DNA sequence (The full length of Dpp is 588 amino acids, however only the region between 487-588 amino acids is active). This CDS region is 402 base pairs long. In the Dpp-HA expressing construct, a triple HA tag encoding sequence was inserted after the 90th base pair in this CDS region and is 195 base pairs long. And in the Sog-Myc expressing construct, the Myc tag is inserted in front of the Sog DNA sequence.
Cloning Materials
Phusion (purchased from Finnzyme) polymerase was used for PCR. And all primers for PCR were ordered from Sigma-Aldrich. All restriction digestion enzymes used in cloning were products of New England Biolabs except EcoRV which is purchased from Rosche. The T4 Ligase used in ligations was produced by Rosche.
T-easy ligation kit (purchased from Promega) was used in the GST fusion protein construction cloning work.
DH5α E.Coli cell strain was used for transformation for plasmid amplification and selection.
GST fusion protein construction
As described, plasmids pGEX4T-3 and pGEX4T-1 was used. A pGEX4T-1 plasmid with Vkg protein DNA sequence of the C terminus part inserted within Multi Cloning Site (MCS) was a gift from Xiaomeng Wang (University of Manchester) and the expressed protein is named GST-VkgC (Wang and Ashe, submitted).
Tsg DNA sequence taken from SKAsc2-Tsg (gift from ___,) which is originated from cDNA bank was inserted between the EcoRI and XhoI restriction digestion sites within the MCS of pGEX4T-3. The restriction sites used were chosen according to the plasmid and Tsg DNA sequence. The primers used were: Tsg prime primer 5’-___-3’ and Tsg reverse primer 5’-___-3’. Note that a 5’-tag-3’ stop codon was added in the prime primer. The acquired PCR product was processed using T-easy kit first and was amplified subsequently, the purified plasmid DNA was then double digested with EcoRI and XhoI, and the small fragment was purified. This small fragment was then ligated to the prepared pGEX4T-3 vector which was digested with EcoRI and XhoI and dephosphorylated. The ligated product was consequently transformed and selected on plates. The final product was check by sequencing and is proved to be correct. The protein expressed by this plasmid was named GST-Tsg.
Construction of plasmid for Drosophila cell expression
Since the full sequence of the Dpp-HA and Sog-Myc expressing construct is unknown to us, restriction digestion test was first performed on them to find out the usable restriction enzymes which do not cut the constructs. From the results it can be found that for Dpp-HA expressing construct, BglII restriction digestion enzyme can be used and for Sog-Myc expressing construct EcoRV is the only available enzyme to be used.
As mentioned, the triple HA within the Dpp-HA is localized at the front of the Dpp functional region. Thus to split it into two fragments without losing the HA tag, the first 285 base pairs in the Dpp-HA CDS region was kept in both fragments, and the latter 312 base pairs was evenly spitted into two parts in the cloning. Primers were designed to amplify the whole plasmid except the small part which should be deleted and restriction digestion sites were added at the end of the PCR product. The cloning strategy for this work is outlined on Figure 3. The primers used were: Dpp N terminus prime primer 5’-GTA GAT CTA TTC GCA CCA CCA TCG CAC C-3’ and reverse primer 5’-GTA GAT CTC TAC ACC ACG GCG TGA TTG GTC G-3’; Dpp C terminus prime primer 5’-GTA GAT CTC AGA CCC TGG TCA ACA ATA T-3’ and reverse primer 5’-GTA GAT CTA TTC AAG TCC TCT TCA GAA ATA-3’. Note that 5’-TAG-3’ stop codon was added in the prime primer for Dpp N and not the one for Dpp C. The PCR products were digested by the BglII restriction digestion enzyme, linked using T4 ligase, and subsequently transformed to be selected on plates. The expressed proteins were named Dpp N and Dpp C.
For the Sog-Myc expressing construct, the Sog DNA sequence was divided into three fragments in order to express C terminus, N terminus and the central part of the Sog protein since mature Sog protein is much larger than Dpp – the Sog DNA sequence is 4818 base pairs in length. There is a transmembrane signal region localized at 1327-1383 base pairs and the AUG codon of this region is localized at the 1162nd base pair, thus this region (including the whole sequence from the AUG codon) should be kept in all fragments to maintain the transmembrane ability. Thus the region before the 1384th base pair was shared in all three fragments. Moreover, the very end region of Sog DNA sequence is too AT rich, and because that the sequence outside of the Sog DNA region is unknown; the very end part of the Sog DNA was kept as a common part in the fragments as well. (See Figure 4 for the model graph of this cloning work.) Primers were designed according to these conditions: Sog N terminus prime primer 5’-GTG ATA TCC CTC CCC AAC CAA CAA ACA CC-3’ and reverse primer 5’-GTG ATA TCC TAC TGG ATG CGC AGA TGT GGG TA-3’; Sog Central part prime primer 5’-GTG ATA TCG GAC ACA TCG TGA CCC GAG CC-3’ and reverse primer 5’-GTG ATA TCC TAA TTG GGC GGC AGG AAT GGA TG-3’; Sog C terminus prime primer 5’-GTG ATA TCG GCT TCG ATA CCT GCA CCA CC-3’ and reverse primer 5’-GTG ATA TCC ACG CCC GCC AAG CAG ACG AT-3’. The PCR products were processed following standard cloning protocol consequently. However, the self ligation of the Sog central part cloning is very serve and the expected product have not been got so far. The expressed Sog C terminus protein fragment was named Sog C and the N terminus part was named Sog N.
The final products of the cloning works were sequenced and proved to be correct.
GST fusion protein expression & purification
For GST fusion protein expression, BL21 E.Coli cell strain was used for transformation. The plasmid DNA was first transformed into the BL21 cells. Then single colony on transformed plate was picked to set up start culture and the start culture was later seeded into more medium to make expression culture (7 mL culture from 10 mL of start culture was seeded into every 100 mL expression culture). These cultures were all incubated 200 rpm shaking at 37 ºC. 50 µL of 0.1 mM IPTG was added to the expression culture to induce protein expression when the OD600 reached 0.6, and the culture was moved to 25 ºC 200rpm shaker. After 3 hours of incubation, the cells were harvested by 5,000 g 15 minutes centrifugation. Then the cells were resuspended in 8.35 mL NTN buffer (H2O solution with 20mM ph 7.5 Tris, 100mM NaCl and 0.5% NP-40 by volume inside) with 0.0085 g lysozyme, 7.5 µM phenylmethylsulphonyl fluoride (PMSF) and 75 mg n-lauroylsarcosine sodium salt (sarkosyl), and the mixture was kept on ice for 30 minutes. After this step, the mixture was sonicated for 3 times, 30 seconds each time. The debris of the cells and the supernatant were then sperated by 20,000 g 10 minutes centrifugation. The debris was discarded and GSH-sepharose beads were added into the supernatant to collect the GST tagged protein by shaking in 4 ºC cold room for at least 1 hour. The beads were subsequently washed with NTN buffer for 4 times and the purified protein bound on the beads was checked on SDS gel and was stored at 4 ºC.
Cell transfection, expression & harvesting
After testing of expression efficiency using Sus529, S2 R+ and Flag-Mad cell lines, S2 R+ was chosen for its high expression efficiency. The S2 R+ cells to be used for protein expression were incubated at 25 ºC in FCS medium (medium detail here) until it reaches the suitable cell density suitable for transfection (>1.5×106 cells per mL). For the transfection, Effectgene transfection kit (purchased from QIAGEN) was used following the manual, and 0.8 µg DNA of Dpp-HA, Dpp N, Dpp C, Sog N and Sog C expressing constructs were transfected. described above prepared using Midi Prep kit (purchased from QIAGEN) was used. CuSO4 (concentration) was added 24 hours after transfection to induce the expression of transfected plasmid DNA, and then the transfected cells were incubated for another 72 hours. The cell cultures were harvested afterwards. Cells and medium were first separated by 5,000 g 2 minutes centrifugation, and the cell pellets were lysed with 50 µL lysis buffer (H2O solution with 50 mM pH7.5 Tris-HCl, 150 mM NaCl, 1% NP-40 by volume and ~10% g/mL protease inhibitor) by 10 minutes incubating at 37 ºC subsequently. Both the supernatant and pellet lysate were loaded on 10% SDS gel (20 µL was loaded for supernatant and 7 µL was loaded for pellet lysate) and checked by western blotting and were stored in -80 ºC freezer. The recipe for the 10% SDS resolving gel is: 30% Acrylamide/Bis (Ac/Bis) 1.25 mL, 1.5 M pH 8.8 Tris 1.25 mL, 10% SDS 50 µL, H2O 2.75 mL, N,N,N,N -Tetramethyl-Ethylenediamine (TEMED) 11 µL and fresh 30% Ammonium Persulfate (Aps) 30 µL. And the recipe for the stacking gel is: 30% Ac/Bis 0.5 mL, 1.5 M pH 8.8 Tris 1.25 mL, 10% SDS 50 µL, H2O 3.2 mL, TEMED 15 µL and fresh 30% Aps 50 µL.
GST pull down assay
Interactions between GST fusion proteins and proteins expressed in Drosophila cells were tested by GST pull down assay.
At the beginning the quantity of the GST fusion proteins were estimated on 10% SDS gel and the amount of GST fusion proteins to be used in GST pull down assay was neutralized. The pull down buffer (PD buffer) was freshly prepared. The recipe of PD buffer is: 100 mM NaCl, 4% (by volume) 1 M ph 7.9 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 5 mM MgCl2, 0.5 mM EDTA, 1 mM Dithiothreitol (dTT), 0.5 mM PMSF and 0.05% (by volume) NP-40 dissolved in distilled H2O.
Then GST fusion proteins bound to the GSH-sepharose beads and the product of S2 R+ cell harvesting mentioned above were mixed together in PD buffer. The quantities of GST fusion protein used were calculated by the result of the GST protein normalization. And the quantities of proteins from Drosophila cells used were: all supernatant 50 µL, all pellet lysate 7 µL. 2x Laemmli buffer (2 mL 0.5 M ph 6.8 Tris, 3.2 mL glycerol, 1.6 mL 20% SDS, 0.8 mL β-Mercaptoethanol and 1.6 mL 1% bromphenol blue dissolved in 6.8 mL H2O) equal to the amount of the GST fusion protein was added at the same time.
The mixture was kept at 4 ºC on rotator for 2 hours and was washed by PD buffer for 4 times (5 minutes on ice each time) consequently, thus the protein which is not able to bind to the GST fusion protein on the beads was washed away. Finally, all the beads of each sample was loaded on 10% SDS gel and analyzed by western blotting.
Western blotting
The 10% SDS gel used to separate the proteins were made as described. After the proteins were separated, the proteins on the gel were transferred to western blotting membranes by wet transfer method. The membrane was washed with 5% milk in PBT and then blotted with first antibody in 4 ºC cold room on a rocker for at least 12 hours. First antibodies used in western blotting including monoclonal anti-HA antibody produced in mouse and monoclonal anti-Myc antibody produced in mouse, all worked by using 1:2,000 dilution in 5% milk in PBT.
After blotting with the first antibody, the membrane was washed with PBT for 2 times, 15 minutes each. Then the membrane was blotted with secondary antibody at room temperature for 1 hour. Secondary antibody used was anti-mouse antibody, worked by using 1:10,000 dilution in 5% milk in PBT.
After blotting, the membrane was washed with PBT for 4 times, 15 minutes each. The membrane was subsequently stained with either ECL kit (purchased from GE health care) for weaker signal strength or Supersignal kit (purchased from Pierce) for 5 minutes. Finally, Medical X-ray films (purchased from Kodak) were used to visualize the signal on the stained membrane.
Results
In situ hybridization
The in situ hybridization using Race and Ush RNA probe will visualize the expression level and distribution of Race and Ush, which are reflections to the Dpp threshold (Ashe et al., 2000). Thus the visualized Race and Ush image will help show the changes of Dpp activity pattern.
Here, photos of the embryos with targeted mutated gene at the stage of dorsal ectoderm formation were taken from the top view. Photos of wild type embryos were also taken for comparison at the same time. The photos are shown in Figure 1.
From the photos, it can be seen that the Race and Ush staining is much weaker in Vkgk00236 and Dcg1k00405 embryos than the wild type ones which indicates the fail of normal Dpp threshold formation in the embryos. Moreover, the expressing pattern of these two genes is changed, which implies the different in function of Vkg and Dcg1. This result proved that Vkg and Dcg1 protein are involved in the Dpp signalling pathway and is important for the formation of Dpp step gradient.
GST fusion protein purification
The purification of GST, GST-VkgC and GST-Tsg proteins were performed as described. The products were check and normalized simultaneously on 10% SDS resolving gel. The sample quantity loaded was: GST 7 µL, GST-VkgC 50 µL and GST-Tsg 50 µL. The gel photo is shown in Figure 2. The normalized ratio of these three products is 7:25:50 due to this result.
Drosophila S2 R+ cell protein expression
The western blotting results for the expression showed that Dpp-HA and Dpp N was expressed perfectly. However there was no expression of Sog N and Sog detected neither in supernatant nor pellet lysate, and Sog C which supposed to be detected in supernatant was in fact be detected in the pellet lysate. The western blotting results are shown in Figure 5 (Sog-Myc data not included).
GST pull down assay result
In GST pull down assay, the quantities of the GST fusion proteins used were: GST 7 µL, GST-VkgC 25 µL and GST-Tsg 50 µL based on the normalization.
To test the interactions centralized on Dpp, Dpp-HA (only supernatant), Dpp N (both supernatant and pellet lysate) and Dpp C (both supernatant and pellet lysate) were added to GST-VkgC and GST-Tsg. And to observe the effect of Sog to these interactions, Sog-Myc (only supernatant) or Sog C (only pellet lysate) was added with Dpp-HA in addition to the samples which Dpp-HA was added with condition mock. These tests were detected using anti-HA antibody.
To investigate whether there is direct binding between Sog and Tsg/Vkg, Sog-Myc (only supernatant) and Sog C (only pellet lysate) were added to GST-VkgC and GST-Tsg. These tests were detected using anti-Myc antibody.
GST protein was also tested with the proteins prepared from S2 R+ cells as control for these tests and the results were all blank which proved that there was no interaction between GST protein and the other proteins tested (data not shown).
The western blotting photo of Group A based on GST-VkgC is shown in Figure 6 and the photos of Group A based on GST-Tsg are shown in Figure 7. No signal is detected in the Group B tests (data not shown). The results showed that the existence of Sog enhanced the affinity of Tsg to Dpp but reduced the signal strength of VkgC + Dpp-HA test, VkgC is able to bind Dpp, and Dpp N showed to be the binding region in Dpp.
Alignment of different type IV collagens
The amino acid sequence of type IV collagens from NCBI database were aligned using bioedit (Hall, 1999). Different type IV collagen and procollagen isoforms from Bos Taurus, Rattus norvegicus, Mus musculus, Homo sapiens, Caenorhabditis elegans, Drosophila melanogaster and Danio rerio plus Vkg protein were included. The result is shown in Figure 8, from which it is showed that there is a short highly conserved site in the type IV collagens /procollagens from the different species checked with the amino acid sequence of “SRCXVCE”.
Discussion
The in situ hybridization experiments gave strong evidence for the involvement of Vkg and Dcg1 in BMP signalling. The C terminus part of Vkg protein was used in the latter experiments because Xiaomeng has reported that the Dpp binding region on Vkg is localized within this part (Wang and Ashe, submitted). The idea of the involvement of Vkg is further advanced by the GST pull down assay results which directly showed that GST-VkgC binds to Dpp. These interactions must occur extracellular since Vkg and Dcg1 are components of extracellular matrix. And this result indicates that at least the Vkg should be put into consideration of the BMP signalling model in the future. The role of Vkg and Dcg1 is suggested to be catching the Dpp and limiting its diffusion ability without damage its signalling potential. This is like the Vkg and Dcg1 are keeping an extracellular Dpp “bank” which may stabilize the supply of Dpp. An interesting observation of the GST pull down assay is that the existence of Sog reduces the efficiency of Vkg C terminus part binding with Dpp. This may imply a competition of Dpp binding between Sog and Vkg. This finding is an addition to the hypothesis mentioned by Xiaomeng (Wang and Ashe, submitted) which suggests that Sog and Tsg together induce the release of the Vkg bounded Dpp-Scw heterodimer from Vkg. Nevertheless, the interaction network of Vkg/Dcg1, Dpp/Scw and Tsg/Sog is mostly unknown.
Because the involvement of type IV collagens in the BMP signalling pathway discovered in Drosophila, and both the type IV collagens as components of extracellular matrix and BMPs as morphogens directing the differentiation are essential in the animals, we suggest that this involvement may be common among the species which undergo germ layer differentiation. The BMPs are highly conserved in different species which implies if the type IV collagens do function in the BMP signalling pathway, the interaction site of these type IV collagens with BMPs should be highly conserved as well. Depending on this conception, we compared the amino acid sequence of type IV collagen and procollagen proteins (some of these sequences are predicted translation) from the NCBI database. The result is positive: a within a very small region of 7 amino acids, 6 amino acids are discovered to be highly conserved in many different species as described in the result. It is not established whether this conserved region is the site that binds to BMPs, but considering that the established BMPs binding protein Vkg and Dcg1 also have this region, it is possible. Some structure analyzes may be helpful to this research and it will be interesting to do some point mutants on this conserved site to see the consequences. Via these work, the function of this region can be investigated. And if this site is confirmed to be the interaction site, the involvement of type IV collagens in the BMP signalling pathway is consequently established.
From the interaction test among GST-Tsg, Dpp and Sog, it can be seen that with the existence of Sog, the Dpp bound by Tsg was increased. This enhancement effect by Sog to Tsg affinity to Dpp was first reported to be established without the direct disturbing of other proteins (though there were other secreted proteins in the supernatant) in vitro. Moreover, the Dpp and Sog protein were produced in Drosophila cell line which implies the normal translation and post translational process and thus improved the reliability of the result. This result also confirmed the basic interaction theory of the step gradient formation hypothesis mentioned in the introduction.
From the GST pull down assays, we also possibly found the interaction region of Dpp with Tsg. As mentioned, Dpp protein was separated into the N terminus part and the C terminus part. Though the C terminus part may be expressed but is not successfully secreted and no binding between Dpp C terminus part and Tsg was found, we observed that the Dpp N terminus part which was normally secreted bound to the GST-Tsg fusion protein. This may be the evidence that clues on the binding region of Dpp to Tsg is on the Dpp N terminus part. The Dpp N terminus part is small enough (the DNA sequence is 441 base pairs in length) to perform point mutants tests and the exact interaction site may be found out in this way. However, this conclusion is not strong. The Dpp C terminus part is supposed to be existed in the pellet lysate since a band of the right size can be found from the western blotting film, but the reason for it not being secreted is unknown. And there is no interaction detected between the Dpp C pellet lysate and Tsg. We suggest that the reason for the dissecreting of Dpp C is due to the deletion of the N terminus unique part which may be essential for secreting. There are two possibilities for the GST pull down assay result of Dpp C: first, there may truly be no interaction between Dpp C and Tsg; and the second, the proteins along with the Dpp C terminus protein within the pellet lysate may interfere with the interaction between Tsg and Dpp C terminus protein. What to mention here is that there is also a band detected in the pellet lysate of Dpp N terminus part expression cells which is of the right size of Dpp N terminus part though the band is thinner than in the supernatant. However there is no bound observed between this protein in the pellet lysate and Tsg in GST pull down assay. The two possibilities remarked may apply on this Dpp N pellet lysate result as well. In sum, the interaction between Dpp C terminus part and Tsg is unsure according to the results, there is no decisively evidence to show that there is no bound between them, thus though the interaction between Dpp N terminus part and Tsg is established, the possible Dpp interaction region range with Tsg is still in doubt.
Similar argument may apply on the GST pull down assay results for Sog fragments. A difficulty we encountered in the experiments related to Sog protein is that normally we can not detect full length Sog by western blotting. It seems that the full length Sog-Myc protein expressed in Drosophila cell may exist in a very low concentration that is not able to be detected by western blot. There is evidence for the existence of Sog protein: the enhancement of Tsg affinity to Dpp by the additional Sog. This may indicate that the biological function of Sog-Myc is kept though it is in a very low concentration (this in another hand means the effect of Sog is very strong). Concerning the Sog fragments, no signal that can be the candidate for Sog N terminus part was detected by western blotting, but considering the phenomenon on the full length Sog, this may happen to the Sog N terminus part as well. However this is lack of evidence and is doubtful. About the Sog C terminus part, signal of the expected size was only detected in the pellet lysate, and no interaction between this fragment and Tsg or VkgC was observed. Here because we reserved the transmembrane region in all Sog fragments by cloning, it is uncertain whether this Sog C terminus part protein is affected by the deletion just like the doubt on Dpp C terminus fragment. Also whether there is no interaction between Sog C terminus part and Tsg remains doubtful in the same way as the Dpp C terminus. However, the fact that the existence of Sog C significantly lowered the binding efficiency between VkgC and Dpp proved it to be functional. This point thus needs more exploring. In this research work, the reason for the arresting of Sog by stacking gel is worth investigation. To further advance the understanding on Sog protein, more analyze on the structure of Sog may be necessary, and if the Sog central part cloning can be done in the future, it will give more ideas in this field.
References
Affolter, M., T. Marty, M.A. Vigano, and A.J. ska. 2001. Nuclear interpretation of Dpp signaling in Drosophila. EMBO J. 20:3298-3305.
Arora, K., and C. Nusslein-volhard. 1992. Altered mitotic domains reveal fate map chages in Drosophila embryos mutant for zygotic dorsoventral patterning genes. Development. 114:1003-1024.
Ashe, H.L. 2002. BMP signalling: visualisation of the Sog protein gradient. Current Biology. 12:273-275.
Ashe, H.L. 2005. BMP Signalling: Synergy and Feedback Create a Step Gradient. Current Biology. 15:375-377.
Ashe, H.L., and J. Briscoe. 2006. The interpretation of morphogen gradients. Development. 133:385-394.
Ashe, H.L., and M. Levine. 1999. Local inhibition and long-range enhancement of Dpp signal transduction by Sog. NATURE. 398:427-431.
Ashe, H.L., M. Mannervik, and M. Levine. 2000. Dpp signaling thresholds in the dorsal ectoderm of the Drosophila embryo. Development. 127:3305-3312.
Blumberg, B., A.J. MacKrell, and J.H. Fessler. 1988. Drosophila basement membrane procollagen alpha 1(IV). II. Complete cDNA sequence, genomic structure, and general implications for supramolecular assemblies. J. Biol. Chem. 263:18328-18337.
Chang, C., D.A. Holtzman, S. Chau, T. Chickering, E.A. Woolf, L.M. Holmgren, J. Bodorova, D.P. Gearing, W.E. Holmes, and A.H. Brivanlou. 2001. Twisted gastrulation can function as a BMP antagonist. Nature. 410:483-487.
Cubadda, Y., P. Heitzler, R.P. Ray, M. Bourouis, P. Ramain, W. Gelbart, P. Simpson, and M. Haenlin. 1997. u-shaped encodes a zinc finger protein that regulates the proneural genes achaete and scute during the formation of bristles in Drosophila. Genes & Development. 11:3083-3095.
Ferguson, E.L., and K.V. Anderson. 1992a. decapentaplegic Acts As a Morphogen to Organize Dorsal-Ventral Pattern in the Drosophila Embryo. Cell. 71:451-461.
Ferguson, E.L., and K.V. Anderson. 1992b. Localized enhancement and repression of the activity of the TGF-β family member, decapentaplegic, is necessary for dorsal-ventral pattern formation in the Drosophila embryo. 113:583-597.
Hall, T.A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids. Symp. Ser. 41:95-98.
Hogan, B.L. 1996. Bone morphogenetic proteins in development. Current Opinion in Genetics & Development. 6:432-438.
Larraín, J., D. Bachiller, B. Lu, E. Agius, S. Piccolo, and E.M.D. Robertis. 2000. BMP-binding modules in chordin: a model for signalling regulation in the extracellular space. Development. 127:821-830.
Lunstrum, G.P., H.-P. Bachinger, L.I. Fessler, K.G. Duncan, R.E. Nelson, and J.H. Fessler. 1988. Drosophila basement membrane procollagen IV. I. Protein characterization and distribution. THE JOURNAL OF BIOLOGICAL CHEMISTRY. 263:18318-18327.
Marqués, G., M. Musacchio, M.J. Shimell, K. Wünnenberg-Stapleton, K.W.Y. Cho, and M.B. O'Connor. 1997. Production of a DPP Activity Gradient in the Early Drosophila Embryo through the Opposing Actions of the SOG and TLD Proteins Cell. 91:417-426.
Morisato, D., and K.V. Anderson. 1994. The spätzle gene encodes a component of the extracellular signaling pathway establishing the dorsal-ventral pattern of the Drosophila embryo. Cell. 76:677-688.
Nakayama, T., Y. Cui, and J.L. Christian. 2000. Regulation of BMP/Dpp signaling during embryonic development. Cell Mol Life Sci. . 57:943-956.
Paralkar, V.M., B.S. Weeks, Y.M. Yu, H.K. Kleinman, and A.H. Reddi. 1992. Recombinant human bone morphogenetic protein 2B stimulates PC12 cell differentiation: potentiation and binding to type IV collagen. The Journal of Cell Biology. 119:1721-1728.
Piccolo, S., Y. Sasai, B. Lu, and E.M.D. Robertis. 1996. Dorsoventral Patterning in Xenopus: Inhibition of Ventral Signals by Direct Binding of Chordin to BMP-4. Cell. 86:589–598.
Raftery, L.A., and D.J. Sutherland. 2003. Gradients and thresholds: BMP response gradients unveiled in Drosophila embryos. Trends In Genetics. 19:701-708.
Ross, J.J., O. Shimmi, P. Vilmos, A. Petryk, H. Kim, K. Gaudenz, S. Hermanson, S.C. Ekker, M.B. O'Connor, and J.L. Marsh. 2001. Twisted gastrulation is a conserved extracellular BMP antagonist. Nature. 410:479-483.
Shimell, M.J., E.L. Ferguson, S.R. Childs, and M.B. O'Connor. 1991. The Drosophila dorsal-ventral patterning gene tolloid is related to human bone morphogenetic protein 1 Cell. 67:469-481.
Shimmi, O., and M.B. O'Connor. 2003. Physical properties of Tld, Sog, Tsg and Dpp protein interactions are predicted to help create a sharp boundary in Bmp signals during dorsoventral patterning of the Drosophila embryo. Development. 130:4673-4682.
Shimmi, O., D. Umulis, H. Othmer, and M.B. O'Connor. 2005. Facilitated transport of a Dpp/Scw heterodimer by Sog/Tsg leads to robust patterning of the Drosophila blastoderm embryo. Cell. 120:873-886.
Srinivasan, S., K.E. Rashka, and E. Bier. 2002. Creation of a Sog Morphogen Gradient in the Drosophila Embryo Cell. 2:91-101.
Tatei, K., H. Cai, Y.T. Ip, and M. Levine. 1994. Race: a drosophila homologue of the angiotensin converting enzyme. Mechanisms of Development 51:157-168.
Timpl, R., and J.C. Brown. 1996. Supramolecular assembly of basement membranes. BioEssays. 18:123 - 132.
Wang, X., and H.L. Ashe. submitted. The extracellular matrix regulates BMP gradient formation in the Drosophila embryo. Nature.
Wang, Y.-C., and E.L. Ferguson. 2005. Spatial bistability of Dpp-receptor interactions during Drosophila dorsal-ventral patterning. Nature. 283:583-583.
Yasothornsrikula, S., W.J. Davis, G. Cramer, D.A. Kimbrell, and C.R. Dearolf. 1997. viking: identification and characterization of a second type IV collagen in Drosophila Gene 198:17-25.
Yu, K., S. Srinivasan, O. Shimmi, B. Biehs, K.E. Rashka, D. Kimelman, M.B. O'Connor, and E. Bier. 2000. Processing of the Drosophila Sog protein creates a novel BMP inhibitory activity. Development. 127:2143-2154.
During the development of Drosophila Melanogaster embryo, the product of Decapentaplegic (Dpp) gene plays an essential role in the final organization of dorsal-ventral pattern. The Dpp is a homologue of Bone Morphogenetic Proteins 2 and 4 (BMP 2/4) and acts as an extracellular morphogen to form a step gradient with the effect of another BMP homologue Screw (Scw) and some other regulators including short gastrulation (Sog), twisted gastrulation (Tsg) and Tolloid (Tld). These regulators are all products of zygotic genes, and Tld is metalloprotease. Moreover, two type IV collagens Dcg1 and Viking (Vkg) may also participate in the BMP signalling. Interactions between Tsg, Vkg, Sog and Dpp were investigated by using GST pull down assay, and the exact region where the interaction occurs was tried to find out in this study. The enhancement of Sog to Tsg on affinity to Dpp was established, and it is observed that Sog reduces the binding activity of Vkg C terminus part to Dpp. In situ hybridization was also performed on Vkg and Dcg1 mutated embryos and wild type embryos to compare the expression patterns of Race and U-shaped (Ush) genes which are reflections to the activity pattern of Dpp. Differences between the wild type and mutated embryos were observed, and the gene level patterns in the mutated embryos are found to be different too. And a conserved site in type IV collagens/procollagens among different species was found.
Introduction
As kind of insect, Drosophila melanogaster undergoes gastrulation germ layer differentiation to form endoderm, mesoderm, ectoderm and then their organelles during embryo development. This differentiation process rises from the different spatial positions of the cells in the embryo and is mostly controlled by morphogens. (Ashe and Briscoe, 2006)
During the early stages of embryo Drosophila development, the product of spätzle gene acts as a morphogen to form the Dorsal (Dl) activity gradient and thus establishes the dorsal-ventral pattern (Morisato and Anderson, 1994; Raftery and Sutherland, 2003). Then, the control is taken away by other morphogens. In previous studies, it has been reported that two homologs of Bone Morphogenetic Proteins 2 and 4 (BMP 2/4), Decapentaplegic (Dpp) and Screw (Scw) which are members of TGF-β superfamily, are the control factors for the formation of dorsoal ectoderm in Drosophila. Further researches indicated that Dpp plays the main role since all dorsal cells assume ventral lateral fates and the mutant phenotype of the zygotically required genes appears the most severely with the lack of Dpp (Arora and Nusslein-volhard, 1992; Ferguson and Anderson, 1992a). Based on these discoveries, the studies in this field are mainly emphasized on the Dpp.
The signal transduction of Dpp/Scw has been well described now. The Dpp/Scw functions by binding to a receptor complex which is supposed to be composed of type I and type II transmembrane serine/threonine kinases (Hogan, 1996; Nakayama et al., 2000). It has been elucidated that in Drosophila, the type I receptor Thick veins (Tkv) interact with the Dpp molecule and the type II receptor Punt (Put) to form a heteromeric complex. The active Put within this complex phosphorylate the Tkv consequently and thus activates the associated type I kinases which will phosphorylate the cytoplasmic protein Mothers against Dpp (Mad). The Mad which belongs to the Smad superfamily then function inside of the cell and lead to the differentiation and in this way the signal of Dpp is transduced. Moreover, in the signal transduction of Scw, it seems that another type I receptor Saxophone (Sax) takes the place of Tkv in Dpp signal transduction. However experiment also showed that high level of Tkv is able to compensate the loss of Sax, and indicates that these two morphogens shares a similar intracellular pathway. (Affolter et al., 2001) Moreover, it is thus able to visualize the Dpp/Scw activity level by detecting the level of Smad due to these discoveries on the Dpp signal transduction.
Studies had showed that after the dorsal-ventral (DV) axes is patterned, the effect of the morphogenetic Dpp depends on the local concentration level. In detail, high Dpp level leads to the fate of being amnioserosa whilst lower levels make the cell become dorsal ectoderm tissue (Ashe et al., 2000; Shimmi and O'Connor, 2003). And it has also been reported that the Dpp and Scw function much more effectively by forming a Dpp-Scw heterodimer which is able to induce a 10 to 100 times higher phosphorylated Mad accumulation than Dpp or Scw homodimer do within the cells exposed to these molecules. The receptor Sax and Tkv are suggested to be related to this phenomenon, different kinds of receptor complexes may form. (Shimmi et al., 2005; Wang and Ferguson, 2005)
Based on this knowledge, the regulation of dorsal ectoderm and amnioserosa formation by BMP signalling pathway is suggested to be achieved by the gradient distribution of Dpp/Scw. However, an interesting and indispensable phenomenon of the gradient of Dpp is that it is not smoothly distributed. The visualized Dpp distribution in dorsal ectoderm formation stage Drosophila embryos showed that Dpp activity has a peak level at the 8-10 dorsalmost cells which will later form the amnioserosa, and there is a sharp fall of Dpp level at the edge of the dorsalmost cells where cells on the different sides of the confine receive high and very low BMP signal (Shimmi and O'Connor, 2003). This step gradient specifies the very small region to amnioserosa and ensures the normal development of amnioserosa and dorsal epidermis.
However, the Dpp protein itself is nonautonomous and is only able to diffuse under a narrow limitation because it is easily captured by its receptor without protection. The Dpp protein direct visualization is possible in present time, and the result shows that there is an even level of Dpp throughout the dorsal 40% of Drosophila embryo before the late cellularisation stage when Dpp starts to function. Moreover, though this uniform distribution will lost in later stages of normal embryo development, it can be maintained by Scw mutation, and this also prompts the involvement of Scw in this patterning. The pattern of Smad observed in the embryo also agrees this distribution. (Raftery and Sutherland, 2003; Shimmi et al., 2005; Wang and Ferguson, 2005) It is still not possible to detect and visualize the intracellular concentration of Dpp or Scw directly so far, but the phenomenon had been observed already supports the hypothesis that the Dpp may be evenly secreted in the dorsal part of the embryo after the patterning of DV axes just as the same as before. In this way, there must be some other regulators participated in the formation of the Dpp gradient. Furthermore, according to the classical model of morphogen distribution which concerns the diffusion alone, specified cell or region produces the morphogen and the morphogen diffuses to the tissue nearby via extracellular or intracellular pathway. Thus a morphogen concentration gradient which peaks at the secreting cell and smoothly deduced along with the increasing distance to the secreting cell is formed(Ashe and Briscoe, 2006). The step gradient of Dpp activity is incompatible with this model, thus implies that the formation of this gradient should be created based on another mechanism which is not simply considering the diffusion only.
There is a good hypothesis which gives the explanation for this step gradient that had been widely accepted now. In this hypothesis, three regulators are accounted to be responsible for the formation of the BMP step gradient (Marqués et al., 1997; Shimmi and O'Connor, 2003). These three regulators are the products of three zygotic genes: short gastrulation (Sog), twisted gastrulation (Tsg) and Tolloid (Tld) (Ashe, 2005). And the product of Tld gene is a metalloprotease. They are able to regulate the BMP signalling activity and are essential in the determination of the peak level and the step gradient distribution of BMPs since the loss of any one of these three regulators will lead to the loss of the step gradient. It is long known that Sog gene is a homologue of Xenopus Chordin (Chd) gene which produces antagonist of BMPs both structural and functional. (Ferguson and Anderson, 1992b; Piccolo et al., 1996). And later studies discovered that Tsg is also a secreted protein which acts as extracellular antagonists of BMP (Chang et al., 2001; Ross et al., 2001). These works gave the evidence that Sog and Tsg bind to BMP directly and inhibit the interaction between BMP and its receptor (Tkv/Sax and Put) to form a complex in order to block the signal transduction (Ashe, 2005). An important discovery in the binding mechanism is that the common binding domain among them is reported to be a cysteine-rich repeats (CRs) (Larraín et al., 2000). However, these interactions will not destroy the BMP molecule itself, and the formed complex can be divided into the original single molecular to regain its function too. In Drosophila, this cleavage process is operated by the protease Tld which was first recognized as an activator of Dpp (Ferguson and Anderson, 1992a).
Function of these regulators has been investigated in great detail. Start from Sog, it is produced at the presumptive neuroectoderm and the distribution of Sog in Drosophila embryos had been directly shown to be in a graded fashion and limited by Tld in dorsal region (Srinivasan et al., 2002). The distribution of Sog protein agrees with the classical passive morphogen distribution model since the level of Sog reach the highest at the producing region and falls whilst the distance from the presumptive neuroectoderm increases. However, the Sog detected is thought to be in the complex including Sog and Dpp and Sog single molecule itself will be degraded via endocytosis. Thus it is also suggested that the shaping of the lowest level of Sog at the dorsalmost region is associated by the protease Tld which is able to cleave the complex and release Sog to be degraded. The interaction between Sog and Dpp here not only contributes to the formation of Sog gradient, but also help the diffusion of Dpp. As mentioned, either the Dpp homodimer or Dpp-Scw heterodimer has the activity to bind the receptor on cell membrane and then are caught by the receptors and lost the diffusion ability. Thus, the binding between Sog and Dpp on the one hand is an inhibition on Dpp, on the another hand is a protection of Dpp signalling activity and helps the diffusion of Dpp. Since the Tld restores the activity of Dpp by cleaving the complex when the complex reaches that region, a long-range enhancement and local inhibition of Dpp activity by Sog is then created. (Ashe, 2002; Ashe and Levine, 1999)
A notable fact is that Tld is distributed evenly in the dorsal ectoderm formation embryo (Shimell et al., 1991), thus the cleavage rate of the Sog-Dpp complex is the same throughout the embryo. However, only in the dorsalmost region where is lack of Sog, the Dpp released by Tld cleavage is able to bind to the receptor before forming the complex with the Sog protein again.
Later studies demonstrated that Tsg is another inhibitor of Dpp activity, its role is generally recognized to be similar to Sog. Tsg and Sog are enhancers to each other since the loss of either Tsg or Sog reduces the efficiency of Dpp inhibition in the embryo and they show strong Dpp affinity when existing together. Thus the mechanism of the enhancement is suggested to be the formation of a Tsg-Sog-Dpp complex (Chang et al., 2001; Shimmi and O'Connor, 2003). The knowledge on the complex is further advanced by the recent discoveries mentioned above that gave evidence for the existence of Tsg-Sog-Dpp-Scw complex. It is also discovered that the affinity of Sog/Tsg for Dpp and Scw homodimer is weaker than Dpp-Scw heterodimer.
And some new discoveries recently gave addition to this hypothesis: Wang and Ferguson group (Wang and Ferguson, 2005) found that with the lack of Scw the interpreted results of their experiments support the mechanism that Dpp and Scw homodimers function cooperatively in the BMP signalling, and the report from Shimmi et al. (Shimmi et al., 2005) group gave evidence for the actual existence of Dpp-Scw heterodimer. According to these reports, it is suggested there are three kinds of receptors on the cell membrane for the BMPs which are able to specially bind to Dpp homodimer, Scw homodimer or Dpp-Scw heterodimer. There is a positive feedback with unclear mechanism between the interaction of Dpp-Scw heterodimer and the Dpp-Scw-receptor. Combining the potent signalling of Dpp-Scw heterodimer and the effect of Sog, Tsg and Tld, this feedback helps the formation of peak BMPs signalling level in the dorsalmost cells. And in contrary, in other regions where the Sog/Tsg level is higher the positive feedback is not activated due to the lower free BMPs level, and the main functioning BMPs are Dpp and Scw homodimers. Thus the signalling strength is much weaker in these regions. (Ashe, 2005; Wang and Ferguson, 2005)
This hypothesis is very reasonable. However, two proteins, Dcg1 (Blumberg et al., 1988) and Viking (Vkg) (Yasothornsrikula et al., 1997), may be put into the patterning of BMPs in Drosophila embryos. Vkg and Dcg1 are type IV collagens which are the components of extracellular matrix (Timpl and Brown, 1996) and may limit the diffusion of Dpp/Scw via binding. This idea rose from the observation that in vertebrates BMP 2B/4 is able to bind to type IV collagens (Paralkar et al., 1992). This result was acquired in human cells, but because the BMP signalling pathway is a common and conserved pathway among the animals undergoing gastrulation, it is possible that type IV collagens in Drosophila are also able to bind to BMPs. However, the type IV collagens have not been regarded to be important in the BMP signalling pathway since they lack the widely accepted BMP binding region, CRs region, in the past years whilst some other procollagen types with the CRs region are considered to be necessary for the normal BMP route. And though previous studies had given some information about the distribution of type IV procollagens in Drosophila, the investigation in type IV collagens expression in Drosophila early embryos was failed (Lunstrum et al., 1988).
Here, to further understand the BMP signalling pathway, function of Vkg and Dcg1 was investigated by deletion experiments and interactions among the BMPs and regulators (including type IV collagen) were tested in more detail. Vkg and Dcg1 mutant Drosophila with lower Vkg or Dcg1 activity were first prepared (Wang and Ashe, submitted). Marker genes Race (Tatei et al., 1994) and U-shaped (Ush) (Cubadda et al., 1997) were detected using in situ hybridization method in these Vkg and Dcg1 mutated Drosophila embryos to visualize the Dpp patterning and threshold (Ashe et al., 2000). The interactions among Tsg, Sog, Vkg and Dpp were tested by GST pull down assay. To study the interactions more clearly, the Dpp and Sog protein were tried to be spitted into smaller fragments and GST pull down assay was performed on them consequently to find out and confirm where the interaction region is on these proteins. The Vkg used in the interaction analyse was the C terminus Vkg fragment since this part is reported to contain the interaction region and the full length Vkg protein is insoluble (Wang and Ashe, submitted). Moreover, the type IV collagens in different species were compared to find out whether there is a conserved site which may related to the BMP signalling pathway.
Materials & Methods
Mutant Drosophila melanogaster
The yw67c23 Drosophila from lab stock is used as the wild type control. Vkgk00236 and Dcg1k00405 flies (gift from Xiaomeng Wang, University of Manchester) are used as Vkg and Dcg1 mutants. These two mutants have weaker Vkg or Dcg1 expression, because the completely disappearance of functional Vkg or Dcg1 leads to flies unable to lay eggs. (Wang and Ashe, submitted)
Embryo collection
The embryos from the wild type, Vkg mutant and Dcg1 mutant flies were collected every 2 hours, and the flies were kept in 25 ºC incubator. The collected embryos were left to develop for another period equal to 2 hours at 25 ºC before fixation. The embryos were fixed using fixing buffer (check recipe), 37% formaldehyde and heptane mixture with the volume ratio 4:1:3 by fiercely shaking for 25 minutes and stored in ethanol at -20 ºC. Thus the embryos collected were at the development stages between 2-4 hours. A small number of overnight embryos were also fixed and stored with the 2-4 hours embryos as the control.
In situ hybridization
In situ hybridization was performed with the embryos collected described above. About 50 µL of embryos in volume was used for each in situ test and the hybridization buffer was prepared prior to the in situ hybridization. The recipe of hybridization buffer is (in volume percentage): formamide 50%, 20x SSC buffer 25%, ssDNA 1%, Heparin 0.5%, Tween 0.1%, and H2O 37.3%. Race and Ush digoxigenin-labelled RNA probes were used to visualize the distribution of Race and Ush gene.
The embryos were put into 1.5 mL eppendorf tube, washed with ethanol for 4 times and then methanol for 3 times, each wash lasts for 3 minutes. Then the embryos were shake in 1 mL PBT (0.5% tween in 1x PBS) for 1 hour, continued by shaking in 0.5 mL PBT and 0.5 mL hybridization buffer for 10 minutes, and at last the embryos were washed in 1 mL hybridization buffer for at least 2 minutes.
After the washing, embryos were prehybridized in 1 mL hybridization buffer in 55 ºC oven for at least 1 hour, the tubes with embryos inside was inverted every 15 minutes. And at the same time the RNA probes were denatured at 80 ºC for 4 minutes.
When the prehybridization is done, the hybridization buffer was removed and 1 µL of probe with 50 µL of hybridization buffer was added to the embryos. The embryos were then hybridized in 55 ºC oven for 18 hours.
When the hybridization is finished, 1 mL of 55 ºC hybridization buffer was added to the embryos. The embryos were kept in the 55 ºC oven for another 1 hour, and the tube was inverted every 15 minutes. Then the embryos were washed with 1 mL hybridization buffer for 4 times, for each time the embryos were incubated in the 55 ºC oven for 30 minutes. After this, the embryos were washed with 0.5 mL PBT + 0.5 mL hybridization buffer for 2 times, each time the embryos were shaked for 15 minutes. Finally the embryos were washed with 1 mL PBT for 5 times, each time the embryos were shaked for 10 minutes.
After the washing, 2.5 µL of anti-digoxigenin antibody was added with 0.5 mL PBT, and the embryos were shaked at 4 ºC for at least 12 hours. And then the embryos were ready for staining.
The staining buffer was made freshly before staining, the recipe is: 100 mM NaCl, 50mM MgCl2, 100mM pH 9.5 Tris, 0.1% (in volume) tween and 99.9% (in volume) H2O.
For staining, the embryos were washed with 1 mL PBT for 4 times, 15 minutes each time. Then the staining buffer was used to wash the embryos for 2 times, 5 minutes each time.
After the washing, 400 µL staining buffer with 3.6 µL NBT and 2.8 µL BCIP were added into the tube, and was mixed to begin staining. To stop staining, the embryos were transferred into 1 mL PBT.
After staining, the embryos were washed with 1 mL PBT for 2 times, continued with washing with 0.5 mL PBT + 0.75 mL ethanol (the ethanol was added in 3 times, 0.25 mL each time) once, and 1 mL ethanol for 10 times (at least 5 minutes each time).
When washing is finished, ethanol was removed, and the embryos were washed with xylene once. Then approximate 400 µL of premount buffer (recipe) was added and mixed with the embryos. The embryos were subsequently mounted on slides and left to be dried.
The slides were observed under microscope and images were taken at the same time.
Plasmids
GST protein expression plasmid pGEX4T-3 (lab stock) and pGEX4T-1 (lab stock) was used for GST fusion protein construction cloning. The production of Dpp and Sog proteins fragments were based on two Drosophila cell expression constructs which express full length Dpp with Hemagglutinin (HA) tag (Dpp-HA) and full length Sog with Myc tag (Sog-Myc) (Yu et al., 2000).
The premature Dpp protein will be cleaved in cytosol and the CDS region of the Dpp gene which encodes the mature functional protein is localized at the very end of the full Dpp DNA sequence (The full length of Dpp is 588 amino acids, however only the region between 487-588 amino acids is active). This CDS region is 402 base pairs long. In the Dpp-HA expressing construct, a triple HA tag encoding sequence was inserted after the 90th base pair in this CDS region and is 195 base pairs long. And in the Sog-Myc expressing construct, the Myc tag is inserted in front of the Sog DNA sequence.
Cloning Materials
Phusion (purchased from Finnzyme) polymerase was used for PCR. And all primers for PCR were ordered from Sigma-Aldrich. All restriction digestion enzymes used in cloning were products of New England Biolabs except EcoRV which is purchased from Rosche. The T4 Ligase used in ligations was produced by Rosche.
T-easy ligation kit (purchased from Promega) was used in the GST fusion protein construction cloning work.
DH5α E.Coli cell strain was used for transformation for plasmid amplification and selection.
GST fusion protein construction
As described, plasmids pGEX4T-3 and pGEX4T-1 was used. A pGEX4T-1 plasmid with Vkg protein DNA sequence of the C terminus part inserted within Multi Cloning Site (MCS) was a gift from Xiaomeng Wang (University of Manchester) and the expressed protein is named GST-VkgC (Wang and Ashe, submitted).
Tsg DNA sequence taken from SKAsc2-Tsg (gift from ___,) which is originated from cDNA bank was inserted between the EcoRI and XhoI restriction digestion sites within the MCS of pGEX4T-3. The restriction sites used were chosen according to the plasmid and Tsg DNA sequence. The primers used were: Tsg prime primer 5’-___-3’ and Tsg reverse primer 5’-___-3’. Note that a 5’-tag-3’ stop codon was added in the prime primer. The acquired PCR product was processed using T-easy kit first and was amplified subsequently, the purified plasmid DNA was then double digested with EcoRI and XhoI, and the small fragment was purified. This small fragment was then ligated to the prepared pGEX4T-3 vector which was digested with EcoRI and XhoI and dephosphorylated. The ligated product was consequently transformed and selected on plates. The final product was check by sequencing and is proved to be correct. The protein expressed by this plasmid was named GST-Tsg.
Construction of plasmid for Drosophila cell expression
Since the full sequence of the Dpp-HA and Sog-Myc expressing construct is unknown to us, restriction digestion test was first performed on them to find out the usable restriction enzymes which do not cut the constructs. From the results it can be found that for Dpp-HA expressing construct, BglII restriction digestion enzyme can be used and for Sog-Myc expressing construct EcoRV is the only available enzyme to be used.
As mentioned, the triple HA within the Dpp-HA is localized at the front of the Dpp functional region. Thus to split it into two fragments without losing the HA tag, the first 285 base pairs in the Dpp-HA CDS region was kept in both fragments, and the latter 312 base pairs was evenly spitted into two parts in the cloning. Primers were designed to amplify the whole plasmid except the small part which should be deleted and restriction digestion sites were added at the end of the PCR product. The cloning strategy for this work is outlined on Figure 3. The primers used were: Dpp N terminus prime primer 5’-GTA GAT CTA TTC GCA CCA CCA TCG CAC C-3’ and reverse primer 5’-GTA GAT CTC TAC ACC ACG GCG TGA TTG GTC G-3’; Dpp C terminus prime primer 5’-GTA GAT CTC AGA CCC TGG TCA ACA ATA T-3’ and reverse primer 5’-GTA GAT CTA TTC AAG TCC TCT TCA GAA ATA-3’. Note that 5’-TAG-3’ stop codon was added in the prime primer for Dpp N and not the one for Dpp C. The PCR products were digested by the BglII restriction digestion enzyme, linked using T4 ligase, and subsequently transformed to be selected on plates. The expressed proteins were named Dpp N and Dpp C.
For the Sog-Myc expressing construct, the Sog DNA sequence was divided into three fragments in order to express C terminus, N terminus and the central part of the Sog protein since mature Sog protein is much larger than Dpp – the Sog DNA sequence is 4818 base pairs in length. There is a transmembrane signal region localized at 1327-1383 base pairs and the AUG codon of this region is localized at the 1162nd base pair, thus this region (including the whole sequence from the AUG codon) should be kept in all fragments to maintain the transmembrane ability. Thus the region before the 1384th base pair was shared in all three fragments. Moreover, the very end region of Sog DNA sequence is too AT rich, and because that the sequence outside of the Sog DNA region is unknown; the very end part of the Sog DNA was kept as a common part in the fragments as well. (See Figure 4 for the model graph of this cloning work.) Primers were designed according to these conditions: Sog N terminus prime primer 5’-GTG ATA TCC CTC CCC AAC CAA CAA ACA CC-3’ and reverse primer 5’-GTG ATA TCC TAC TGG ATG CGC AGA TGT GGG TA-3’; Sog Central part prime primer 5’-GTG ATA TCG GAC ACA TCG TGA CCC GAG CC-3’ and reverse primer 5’-GTG ATA TCC TAA TTG GGC GGC AGG AAT GGA TG-3’; Sog C terminus prime primer 5’-GTG ATA TCG GCT TCG ATA CCT GCA CCA CC-3’ and reverse primer 5’-GTG ATA TCC ACG CCC GCC AAG CAG ACG AT-3’. The PCR products were processed following standard cloning protocol consequently. However, the self ligation of the Sog central part cloning is very serve and the expected product have not been got so far. The expressed Sog C terminus protein fragment was named Sog C and the N terminus part was named Sog N.
The final products of the cloning works were sequenced and proved to be correct.
GST fusion protein expression & purification
For GST fusion protein expression, BL21 E.Coli cell strain was used for transformation. The plasmid DNA was first transformed into the BL21 cells. Then single colony on transformed plate was picked to set up start culture and the start culture was later seeded into more medium to make expression culture (7 mL culture from 10 mL of start culture was seeded into every 100 mL expression culture). These cultures were all incubated 200 rpm shaking at 37 ºC. 50 µL of 0.1 mM IPTG was added to the expression culture to induce protein expression when the OD600 reached 0.6, and the culture was moved to 25 ºC 200rpm shaker. After 3 hours of incubation, the cells were harvested by 5,000 g 15 minutes centrifugation. Then the cells were resuspended in 8.35 mL NTN buffer (H2O solution with 20mM ph 7.5 Tris, 100mM NaCl and 0.5% NP-40 by volume inside) with 0.0085 g lysozyme, 7.5 µM phenylmethylsulphonyl fluoride (PMSF) and 75 mg n-lauroylsarcosine sodium salt (sarkosyl), and the mixture was kept on ice for 30 minutes. After this step, the mixture was sonicated for 3 times, 30 seconds each time. The debris of the cells and the supernatant were then sperated by 20,000 g 10 minutes centrifugation. The debris was discarded and GSH-sepharose beads were added into the supernatant to collect the GST tagged protein by shaking in 4 ºC cold room for at least 1 hour. The beads were subsequently washed with NTN buffer for 4 times and the purified protein bound on the beads was checked on SDS gel and was stored at 4 ºC.
Cell transfection, expression & harvesting
After testing of expression efficiency using Sus529, S2 R+ and Flag-Mad cell lines, S2 R+ was chosen for its high expression efficiency. The S2 R+ cells to be used for protein expression were incubated at 25 ºC in FCS medium (medium detail here) until it reaches the suitable cell density suitable for transfection (>1.5×106 cells per mL). For the transfection, Effectgene transfection kit (purchased from QIAGEN) was used following the manual, and 0.8 µg DNA of Dpp-HA, Dpp N, Dpp C, Sog N and Sog C expressing constructs were transfected. described above prepared using Midi Prep kit (purchased from QIAGEN) was used. CuSO4 (concentration) was added 24 hours after transfection to induce the expression of transfected plasmid DNA, and then the transfected cells were incubated for another 72 hours. The cell cultures were harvested afterwards. Cells and medium were first separated by 5,000 g 2 minutes centrifugation, and the cell pellets were lysed with 50 µL lysis buffer (H2O solution with 50 mM pH7.5 Tris-HCl, 150 mM NaCl, 1% NP-40 by volume and ~10% g/mL protease inhibitor) by 10 minutes incubating at 37 ºC subsequently. Both the supernatant and pellet lysate were loaded on 10% SDS gel (20 µL was loaded for supernatant and 7 µL was loaded for pellet lysate) and checked by western blotting and were stored in -80 ºC freezer. The recipe for the 10% SDS resolving gel is: 30% Acrylamide/Bis (Ac/Bis) 1.25 mL, 1.5 M pH 8.8 Tris 1.25 mL, 10% SDS 50 µL, H2O 2.75 mL, N,N,N,N -Tetramethyl-Ethylenediamine (TEMED) 11 µL and fresh 30% Ammonium Persulfate (Aps) 30 µL. And the recipe for the stacking gel is: 30% Ac/Bis 0.5 mL, 1.5 M pH 8.8 Tris 1.25 mL, 10% SDS 50 µL, H2O 3.2 mL, TEMED 15 µL and fresh 30% Aps 50 µL.
GST pull down assay
Interactions between GST fusion proteins and proteins expressed in Drosophila cells were tested by GST pull down assay.
At the beginning the quantity of the GST fusion proteins were estimated on 10% SDS gel and the amount of GST fusion proteins to be used in GST pull down assay was neutralized. The pull down buffer (PD buffer) was freshly prepared. The recipe of PD buffer is: 100 mM NaCl, 4% (by volume) 1 M ph 7.9 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 5 mM MgCl2, 0.5 mM EDTA, 1 mM Dithiothreitol (dTT), 0.5 mM PMSF and 0.05% (by volume) NP-40 dissolved in distilled H2O.
Then GST fusion proteins bound to the GSH-sepharose beads and the product of S2 R+ cell harvesting mentioned above were mixed together in PD buffer. The quantities of GST fusion protein used were calculated by the result of the GST protein normalization. And the quantities of proteins from Drosophila cells used were: all supernatant 50 µL, all pellet lysate 7 µL. 2x Laemmli buffer (2 mL 0.5 M ph 6.8 Tris, 3.2 mL glycerol, 1.6 mL 20% SDS, 0.8 mL β-Mercaptoethanol and 1.6 mL 1% bromphenol blue dissolved in 6.8 mL H2O) equal to the amount of the GST fusion protein was added at the same time.
The mixture was kept at 4 ºC on rotator for 2 hours and was washed by PD buffer for 4 times (5 minutes on ice each time) consequently, thus the protein which is not able to bind to the GST fusion protein on the beads was washed away. Finally, all the beads of each sample was loaded on 10% SDS gel and analyzed by western blotting.
Western blotting
The 10% SDS gel used to separate the proteins were made as described. After the proteins were separated, the proteins on the gel were transferred to western blotting membranes by wet transfer method. The membrane was washed with 5% milk in PBT and then blotted with first antibody in 4 ºC cold room on a rocker for at least 12 hours. First antibodies used in western blotting including monoclonal anti-HA antibody produced in mouse and monoclonal anti-Myc antibody produced in mouse, all worked by using 1:2,000 dilution in 5% milk in PBT.
After blotting with the first antibody, the membrane was washed with PBT for 2 times, 15 minutes each. Then the membrane was blotted with secondary antibody at room temperature for 1 hour. Secondary antibody used was anti-mouse antibody, worked by using 1:10,000 dilution in 5% milk in PBT.
After blotting, the membrane was washed with PBT for 4 times, 15 minutes each. The membrane was subsequently stained with either ECL kit (purchased from GE health care) for weaker signal strength or Supersignal kit (purchased from Pierce) for 5 minutes. Finally, Medical X-ray films (purchased from Kodak) were used to visualize the signal on the stained membrane.
Results
In situ hybridization
The in situ hybridization using Race and Ush RNA probe will visualize the expression level and distribution of Race and Ush, which are reflections to the Dpp threshold (Ashe et al., 2000). Thus the visualized Race and Ush image will help show the changes of Dpp activity pattern.
Here, photos of the embryos with targeted mutated gene at the stage of dorsal ectoderm formation were taken from the top view. Photos of wild type embryos were also taken for comparison at the same time. The photos are shown in Figure 1.
From the photos, it can be seen that the Race and Ush staining is much weaker in Vkgk00236 and Dcg1k00405 embryos than the wild type ones which indicates the fail of normal Dpp threshold formation in the embryos. Moreover, the expressing pattern of these two genes is changed, which implies the different in function of Vkg and Dcg1. This result proved that Vkg and Dcg1 protein are involved in the Dpp signalling pathway and is important for the formation of Dpp step gradient.
GST fusion protein purification
The purification of GST, GST-VkgC and GST-Tsg proteins were performed as described. The products were check and normalized simultaneously on 10% SDS resolving gel. The sample quantity loaded was: GST 7 µL, GST-VkgC 50 µL and GST-Tsg 50 µL. The gel photo is shown in Figure 2. The normalized ratio of these three products is 7:25:50 due to this result.
Drosophila S2 R+ cell protein expression
The western blotting results for the expression showed that Dpp-HA and Dpp N was expressed perfectly. However there was no expression of Sog N and Sog detected neither in supernatant nor pellet lysate, and Sog C which supposed to be detected in supernatant was in fact be detected in the pellet lysate. The western blotting results are shown in Figure 5 (Sog-Myc data not included).
GST pull down assay result
In GST pull down assay, the quantities of the GST fusion proteins used were: GST 7 µL, GST-VkgC 25 µL and GST-Tsg 50 µL based on the normalization.
To test the interactions centralized on Dpp, Dpp-HA (only supernatant), Dpp N (both supernatant and pellet lysate) and Dpp C (both supernatant and pellet lysate) were added to GST-VkgC and GST-Tsg. And to observe the effect of Sog to these interactions, Sog-Myc (only supernatant) or Sog C (only pellet lysate) was added with Dpp-HA in addition to the samples which Dpp-HA was added with condition mock. These tests were detected using anti-HA antibody.
To investigate whether there is direct binding between Sog and Tsg/Vkg, Sog-Myc (only supernatant) and Sog C (only pellet lysate) were added to GST-VkgC and GST-Tsg. These tests were detected using anti-Myc antibody.
GST protein was also tested with the proteins prepared from S2 R+ cells as control for these tests and the results were all blank which proved that there was no interaction between GST protein and the other proteins tested (data not shown).
The western blotting photo of Group A based on GST-VkgC is shown in Figure 6 and the photos of Group A based on GST-Tsg are shown in Figure 7. No signal is detected in the Group B tests (data not shown). The results showed that the existence of Sog enhanced the affinity of Tsg to Dpp but reduced the signal strength of VkgC + Dpp-HA test, VkgC is able to bind Dpp, and Dpp N showed to be the binding region in Dpp.
Alignment of different type IV collagens
The amino acid sequence of type IV collagens from NCBI database were aligned using bioedit (Hall, 1999). Different type IV collagen and procollagen isoforms from Bos Taurus, Rattus norvegicus, Mus musculus, Homo sapiens, Caenorhabditis elegans, Drosophila melanogaster and Danio rerio plus Vkg protein were included. The result is shown in Figure 8, from which it is showed that there is a short highly conserved site in the type IV collagens /procollagens from the different species checked with the amino acid sequence of “SRCXVCE”.
Discussion
The in situ hybridization experiments gave strong evidence for the involvement of Vkg and Dcg1 in BMP signalling. The C terminus part of Vkg protein was used in the latter experiments because Xiaomeng has reported that the Dpp binding region on Vkg is localized within this part (Wang and Ashe, submitted). The idea of the involvement of Vkg is further advanced by the GST pull down assay results which directly showed that GST-VkgC binds to Dpp. These interactions must occur extracellular since Vkg and Dcg1 are components of extracellular matrix. And this result indicates that at least the Vkg should be put into consideration of the BMP signalling model in the future. The role of Vkg and Dcg1 is suggested to be catching the Dpp and limiting its diffusion ability without damage its signalling potential. This is like the Vkg and Dcg1 are keeping an extracellular Dpp “bank” which may stabilize the supply of Dpp. An interesting observation of the GST pull down assay is that the existence of Sog reduces the efficiency of Vkg C terminus part binding with Dpp. This may imply a competition of Dpp binding between Sog and Vkg. This finding is an addition to the hypothesis mentioned by Xiaomeng (Wang and Ashe, submitted) which suggests that Sog and Tsg together induce the release of the Vkg bounded Dpp-Scw heterodimer from Vkg. Nevertheless, the interaction network of Vkg/Dcg1, Dpp/Scw and Tsg/Sog is mostly unknown.
Because the involvement of type IV collagens in the BMP signalling pathway discovered in Drosophila, and both the type IV collagens as components of extracellular matrix and BMPs as morphogens directing the differentiation are essential in the animals, we suggest that this involvement may be common among the species which undergo germ layer differentiation. The BMPs are highly conserved in different species which implies if the type IV collagens do function in the BMP signalling pathway, the interaction site of these type IV collagens with BMPs should be highly conserved as well. Depending on this conception, we compared the amino acid sequence of type IV collagen and procollagen proteins (some of these sequences are predicted translation) from the NCBI database. The result is positive: a within a very small region of 7 amino acids, 6 amino acids are discovered to be highly conserved in many different species as described in the result. It is not established whether this conserved region is the site that binds to BMPs, but considering that the established BMPs binding protein Vkg and Dcg1 also have this region, it is possible. Some structure analyzes may be helpful to this research and it will be interesting to do some point mutants on this conserved site to see the consequences. Via these work, the function of this region can be investigated. And if this site is confirmed to be the interaction site, the involvement of type IV collagens in the BMP signalling pathway is consequently established.
From the interaction test among GST-Tsg, Dpp and Sog, it can be seen that with the existence of Sog, the Dpp bound by Tsg was increased. This enhancement effect by Sog to Tsg affinity to Dpp was first reported to be established without the direct disturbing of other proteins (though there were other secreted proteins in the supernatant) in vitro. Moreover, the Dpp and Sog protein were produced in Drosophila cell line which implies the normal translation and post translational process and thus improved the reliability of the result. This result also confirmed the basic interaction theory of the step gradient formation hypothesis mentioned in the introduction.
From the GST pull down assays, we also possibly found the interaction region of Dpp with Tsg. As mentioned, Dpp protein was separated into the N terminus part and the C terminus part. Though the C terminus part may be expressed but is not successfully secreted and no binding between Dpp C terminus part and Tsg was found, we observed that the Dpp N terminus part which was normally secreted bound to the GST-Tsg fusion protein. This may be the evidence that clues on the binding region of Dpp to Tsg is on the Dpp N terminus part. The Dpp N terminus part is small enough (the DNA sequence is 441 base pairs in length) to perform point mutants tests and the exact interaction site may be found out in this way. However, this conclusion is not strong. The Dpp C terminus part is supposed to be existed in the pellet lysate since a band of the right size can be found from the western blotting film, but the reason for it not being secreted is unknown. And there is no interaction detected between the Dpp C pellet lysate and Tsg. We suggest that the reason for the dissecreting of Dpp C is due to the deletion of the N terminus unique part which may be essential for secreting. There are two possibilities for the GST pull down assay result of Dpp C: first, there may truly be no interaction between Dpp C and Tsg; and the second, the proteins along with the Dpp C terminus protein within the pellet lysate may interfere with the interaction between Tsg and Dpp C terminus protein. What to mention here is that there is also a band detected in the pellet lysate of Dpp N terminus part expression cells which is of the right size of Dpp N terminus part though the band is thinner than in the supernatant. However there is no bound observed between this protein in the pellet lysate and Tsg in GST pull down assay. The two possibilities remarked may apply on this Dpp N pellet lysate result as well. In sum, the interaction between Dpp C terminus part and Tsg is unsure according to the results, there is no decisively evidence to show that there is no bound between them, thus though the interaction between Dpp N terminus part and Tsg is established, the possible Dpp interaction region range with Tsg is still in doubt.
Similar argument may apply on the GST pull down assay results for Sog fragments. A difficulty we encountered in the experiments related to Sog protein is that normally we can not detect full length Sog by western blotting. It seems that the full length Sog-Myc protein expressed in Drosophila cell may exist in a very low concentration that is not able to be detected by western blot. There is evidence for the existence of Sog protein: the enhancement of Tsg affinity to Dpp by the additional Sog. This may indicate that the biological function of Sog-Myc is kept though it is in a very low concentration (this in another hand means the effect of Sog is very strong). Concerning the Sog fragments, no signal that can be the candidate for Sog N terminus part was detected by western blotting, but considering the phenomenon on the full length Sog, this may happen to the Sog N terminus part as well. However this is lack of evidence and is doubtful. About the Sog C terminus part, signal of the expected size was only detected in the pellet lysate, and no interaction between this fragment and Tsg or VkgC was observed. Here because we reserved the transmembrane region in all Sog fragments by cloning, it is uncertain whether this Sog C terminus part protein is affected by the deletion just like the doubt on Dpp C terminus fragment. Also whether there is no interaction between Sog C terminus part and Tsg remains doubtful in the same way as the Dpp C terminus. However, the fact that the existence of Sog C significantly lowered the binding efficiency between VkgC and Dpp proved it to be functional. This point thus needs more exploring. In this research work, the reason for the arresting of Sog by stacking gel is worth investigation. To further advance the understanding on Sog protein, more analyze on the structure of Sog may be necessary, and if the Sog central part cloning can be done in the future, it will give more ideas in this field.
References
Affolter, M., T. Marty, M.A. Vigano, and A.J. ska. 2001. Nuclear interpretation of Dpp signaling in Drosophila. EMBO J. 20:3298-3305.
Arora, K., and C. Nusslein-volhard. 1992. Altered mitotic domains reveal fate map chages in Drosophila embryos mutant for zygotic dorsoventral patterning genes. Development. 114:1003-1024.
Ashe, H.L. 2002. BMP signalling: visualisation of the Sog protein gradient. Current Biology. 12:273-275.
Ashe, H.L. 2005. BMP Signalling: Synergy and Feedback Create a Step Gradient. Current Biology. 15:375-377.
Ashe, H.L., and J. Briscoe. 2006. The interpretation of morphogen gradients. Development. 133:385-394.
Ashe, H.L., and M. Levine. 1999. Local inhibition and long-range enhancement of Dpp signal transduction by Sog. NATURE. 398:427-431.
Ashe, H.L., M. Mannervik, and M. Levine. 2000. Dpp signaling thresholds in the dorsal ectoderm of the Drosophila embryo. Development. 127:3305-3312.
Blumberg, B., A.J. MacKrell, and J.H. Fessler. 1988. Drosophila basement membrane procollagen alpha 1(IV). II. Complete cDNA sequence, genomic structure, and general implications for supramolecular assemblies. J. Biol. Chem. 263:18328-18337.
Chang, C., D.A. Holtzman, S. Chau, T. Chickering, E.A. Woolf, L.M. Holmgren, J. Bodorova, D.P. Gearing, W.E. Holmes, and A.H. Brivanlou. 2001. Twisted gastrulation can function as a BMP antagonist. Nature. 410:483-487.
Cubadda, Y., P. Heitzler, R.P. Ray, M. Bourouis, P. Ramain, W. Gelbart, P. Simpson, and M. Haenlin. 1997. u-shaped encodes a zinc finger protein that regulates the proneural genes achaete and scute during the formation of bristles in Drosophila. Genes & Development. 11:3083-3095.
Ferguson, E.L., and K.V. Anderson. 1992a. decapentaplegic Acts As a Morphogen to Organize Dorsal-Ventral Pattern in the Drosophila Embryo. Cell. 71:451-461.
Ferguson, E.L., and K.V. Anderson. 1992b. Localized enhancement and repression of the activity of the TGF-β family member, decapentaplegic, is necessary for dorsal-ventral pattern formation in the Drosophila embryo. 113:583-597.
Hall, T.A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids. Symp. Ser. 41:95-98.
Hogan, B.L. 1996. Bone morphogenetic proteins in development. Current Opinion in Genetics & Development. 6:432-438.
Larraín, J., D. Bachiller, B. Lu, E. Agius, S. Piccolo, and E.M.D. Robertis. 2000. BMP-binding modules in chordin: a model for signalling regulation in the extracellular space. Development. 127:821-830.
Lunstrum, G.P., H.-P. Bachinger, L.I. Fessler, K.G. Duncan, R.E. Nelson, and J.H. Fessler. 1988. Drosophila basement membrane procollagen IV. I. Protein characterization and distribution. THE JOURNAL OF BIOLOGICAL CHEMISTRY. 263:18318-18327.
Marqués, G., M. Musacchio, M.J. Shimell, K. Wünnenberg-Stapleton, K.W.Y. Cho, and M.B. O'Connor. 1997. Production of a DPP Activity Gradient in the Early Drosophila Embryo through the Opposing Actions of the SOG and TLD Proteins Cell. 91:417-426.
Morisato, D., and K.V. Anderson. 1994. The spätzle gene encodes a component of the extracellular signaling pathway establishing the dorsal-ventral pattern of the Drosophila embryo. Cell. 76:677-688.
Nakayama, T., Y. Cui, and J.L. Christian. 2000. Regulation of BMP/Dpp signaling during embryonic development. Cell Mol Life Sci. . 57:943-956.
Paralkar, V.M., B.S. Weeks, Y.M. Yu, H.K. Kleinman, and A.H. Reddi. 1992. Recombinant human bone morphogenetic protein 2B stimulates PC12 cell differentiation: potentiation and binding to type IV collagen. The Journal of Cell Biology. 119:1721-1728.
Piccolo, S., Y. Sasai, B. Lu, and E.M.D. Robertis. 1996. Dorsoventral Patterning in Xenopus: Inhibition of Ventral Signals by Direct Binding of Chordin to BMP-4. Cell. 86:589–598.
Raftery, L.A., and D.J. Sutherland. 2003. Gradients and thresholds: BMP response gradients unveiled in Drosophila embryos. Trends In Genetics. 19:701-708.
Ross, J.J., O. Shimmi, P. Vilmos, A. Petryk, H. Kim, K. Gaudenz, S. Hermanson, S.C. Ekker, M.B. O'Connor, and J.L. Marsh. 2001. Twisted gastrulation is a conserved extracellular BMP antagonist. Nature. 410:479-483.
Shimell, M.J., E.L. Ferguson, S.R. Childs, and M.B. O'Connor. 1991. The Drosophila dorsal-ventral patterning gene tolloid is related to human bone morphogenetic protein 1 Cell. 67:469-481.
Shimmi, O., and M.B. O'Connor. 2003. Physical properties of Tld, Sog, Tsg and Dpp protein interactions are predicted to help create a sharp boundary in Bmp signals during dorsoventral patterning of the Drosophila embryo. Development. 130:4673-4682.
Shimmi, O., D. Umulis, H. Othmer, and M.B. O'Connor. 2005. Facilitated transport of a Dpp/Scw heterodimer by Sog/Tsg leads to robust patterning of the Drosophila blastoderm embryo. Cell. 120:873-886.
Srinivasan, S., K.E. Rashka, and E. Bier. 2002. Creation of a Sog Morphogen Gradient in the Drosophila Embryo Cell. 2:91-101.
Tatei, K., H. Cai, Y.T. Ip, and M. Levine. 1994. Race: a drosophila homologue of the angiotensin converting enzyme. Mechanisms of Development 51:157-168.
Timpl, R., and J.C. Brown. 1996. Supramolecular assembly of basement membranes. BioEssays. 18:123 - 132.
Wang, X., and H.L. Ashe. submitted. The extracellular matrix regulates BMP gradient formation in the Drosophila embryo. Nature.
Wang, Y.-C., and E.L. Ferguson. 2005. Spatial bistability of Dpp-receptor interactions during Drosophila dorsal-ventral patterning. Nature. 283:583-583.
Yasothornsrikula, S., W.J. Davis, G. Cramer, D.A. Kimbrell, and C.R. Dearolf. 1997. viking: identification and characterization of a second type IV collagen in Drosophila Gene 198:17-25.
Yu, K., S. Srinivasan, O. Shimmi, B. Biehs, K.E. Rashka, D. Kimelman, M.B. O'Connor, and E. Bier. 2000. Processing of the Drosophila Sog protein creates a novel BMP inhibitory activity. Development. 127:2143-2154.
posted by =DOC=Cai-Fan at
8:25 pm
0 Comments:
Post a Comment
Subscribe to Post Comments [Atom]
<< Home