SEARCH

SEARCH BY CITATION

Keywords:

  • Xenopus;
  • hsp70;
  • injection;
  • heat shock;
  • ectopic expression;
  • β-catenin;
  • GFP;
  • gene manipulation

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

The temporal and spatial manipulation of gene expression is useful in analyzing the mechanisms of early embryogenesis. This report describes a modified strategy to achieve controlled gene expression by directed plasmid injection using the hsp70 promoter and heat treatment. Two control genes, enhanced green fluorescent protein (EGFP) and β-catenin, were also expressed by this method. When embryos were injected with HsS1/EGFP and subsequently heat-treated, ectopic EGFP was expressed only in the injected area. No severe defects were attributable to the heat treatment alone. Western blotting confirmed that no EGFP induction occurred in the absence of heat treatment and that, in the presence of heat induction, EGFP expression was detected within 1 hr after treatment. These results suggest that heat-mediated gene expression in the restricted area was regulated temporally. In addition, HsS1/β-catenin injection into the animal pole of 8-cell embryos, followed by heat treatment, caused loss of head formation that was similar to that seen with CS2/β-catenin injection. Although a hormone-inducible gene induction system already exists in Xenopus, our modified technique provides an alternative method for controlling temporal and spatial gene expression. Developmental Dynamics 232:369–376, 2005. © 2004 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

The manipulation of gene expression is an essential tool in the study of development. In Drosophila, a DNA fragment can be integrated into the genome by injection with a plasmid DNA-containing transposable element, such as the P-element into germ cells of Drosophila embryos (Rubin and Spradling,1982). This technique enables a desired gene to be induced under the control of a tissue-specific enhancer. In addition, the promoter sequence of the Drosophila heat shock protein 70 (hsp70) gene has been used to induce temporally controlled ectopic gene expression (Karch et al.,1981; Pelham,1982; Kojima et al.,1993). Xenopus xhsp70 is expressed endogenously during oogenesis (Bienz,1984) and ectopic expression of microinjected xhsp70 gene can be induced by heat treatment (Krone and Heikkila,1988). The xhsp70 promoter sequence also has been used to induce some genes such as chloramphenicol acetyltransferase (CAT; Krone and Heikkila,1989; Brakenhoff et al.,1991).

It is well known that ectopic gene expression can be spatially controlled by injection with mRNA or plasmid DNA into a specific site on the embryo. For example, ventral but not dorsal injection of xWnt8 mRNA induces secondary axis formation in embryos (Sokol et al.,1991; Smith and Harland,1991). Furthermore, transgenic techniques have been established to generate knock-in frogs (Kroll and Amaya,1996) and to regulate gene expression precisely by using a tissue-specific enhancer. The promoter region of Xenopus hsp70 can be used also to induce gene expression manually (Wheeler et al.,2000), as shown in zebrafish where ectopic gene expression was induced in a spatially restricted set of cells in transgenic embryos using a laser microbeam (Halloran et al.,2000).

While widely used and important in developmental studies, the above-discussed methods have several disadvantages. Microinjection methods do not allow for the temporal regulation of ectopic gene expression, because mRNA is translated and begins to function immediately after injection. With transgenic procedures, integration of the gene into the genome is not easy compared with injection, and a tissue-specific enhancer is needed to ensure that the appropriate expression is induced. Furthermore, a given enhancer element is fixed with respect to the timing of expression and cannot be used for a different condition.

We therefore aimed to combine two strategies to produce a simple way to induce ectopic gene expression in a restricted time and place. Plasmid DNA, including the hsp70 promoter followed by the desired gene, was injected into a restricted region of Xenopus embryos, and ectopic gene expression was induced by subsequent heat shock treatment. In short, spatial gene expression was regulated by injection, and temporal expression was controlled by the heat shock induction. Initially, we constructed a modified injection vector (pHsS1) containing the hsp70 promoter. We then tested two genes, enhanced green fluorescent protein (EGFP) and β-catenin, in control experiments to assess whether our method was successful and practical. With the HsS1/EGFP construct, we examined whether ectopic gene expression was induced only in the injected area, while the HsS1/β-catenin was used to check that the intended defect was caused by the gene expression. Our results demonstrate the value of this alternative method for the temporal and spatial control of gene expression.

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

Initially, an expression vector for the heat shock induction system was constructed (Fig. 1A). To do this construction, we obtained a fragment containing the xhsp70 promoter sequence (Bienz,1984) by PCR amplification from Xenopus genomic DNA. Next, we modified the pCS2 vector by substituting the hsp70 promoter for the CMV promoter (pHsS1; Fig. 1A). To induce ectopic gene expression, a fragment containing the desired gene was introduced into the new vector between hsp70 promoter and the SV40 polyA signal. This construct was injected into Xenopus embryos, which were then heat treated at the desired stage. Heat shock (HS) treatments were carried out by pulsing; plasmid-injected embryos were treated at 33°C or 37°C for 15 min followed by an interval at 16°C for 15 min, and these steps were repeated three or six times (Fig. 1B).

thumbnail image

Figure 1. A: Strategy for pHsS1 vector construction. Sequence data for the Xenopus hsp70 promoter was obtained from GenBank (accession no. U77532). Polymerase chain reaction amplification was performed by using the following primers: 5′-CCCTGGTAGCAGGAAATAGCCTTG-3′ (corresponding to nucleotides 1–24) and 5′-GCTGTGCAGTTGCTCTTCTCTGTC-3′ (corresponding to nucleotides 420–397). A 0.4-kb fragment of the hsp70 promoter region (hatched box) was isolated and integrated into pBluescriptII SK(−). DNA sequencing confirmed the correct nucleotide sequence. The plasmid then was digested with EcoRI, creating the blunt end, and was digested with HindIII, and then the hsp70 promoter fragment was inserted into a SalI (creating the blunt end after digestion) -HindIII site of pCS2. B, BamHI; E, EcoRI; H, HindIII; N, NotI; S, SalI. Open horizontal arrows with SP6, T7, and T3 indicate the RNA polymerase promoter region. B: Method for heat shock (HS) treatment. This method is similar to that described in previous studies [5]. Before the experiment, culture dishes containing 0.1× Steinberg's solution were equilibrated at 16°C, 33°C, and 37°C. Injected embryos were cultured at 16°C until HS treatment and transferred into 0.1× Steinberg's solution at 33°C or 37°C, incubated for 15 min, and then transferred and incubated at 16°C for 15 min. These steps were repeated three or six times. These embryos were cultured at 20°C and observed at the appropriate stages.

Download figure to PowerPoint

As a control experiment, the EGFP gene was introduced into pHsS1 (named pHsS/EGFP) to allow easy detection of the expressed construct. pHsS/EGFP was then injected into four-cell stage embryos and cultured until the gastrula stage. Injected embryos were treated at 37°C, and EGFP expression was observed at several stages. In normal embryos, no fluorescence was detected even in the presence of HS treatment (Fig. 2A,E,I). Similarly, without HS treatment, pHsS1/EGFP-injected embryos showed no green fluorescence above background levels (Fig. 2B,F,J). When pHsS1/EGFP-injected embryos were subjected to a 3 × HS treatment, EGFP expression was detected immediately after the treatments (Fig. 2C). The signal increased by 1.5 hr after the treatment (Fig. 2G) and was maintained until at least 16 hr after treatment (Fig. 2K). These results showed that the xhsp70 promoter-contained plasmid could induce gene expression after heat induction. The signal detected after a 6× HS treatment was slightly stronger than that observed with the 3× condition, although the difference was not significant (Fig. 2D,H,L). The increase in treatment time allowed embryos to progress to the developmental stage, and the ability to maintain rigorous gene expression was compromised. Therefore, a 3× HS condition was adopted for the remainder of the experiments. To assess embryo damage due to the heat treatment alone, the superficial phenotype of the embryos was observed in 2-day-old tadpoles. No severe defect was seen under any of the control conditions (Fig. 2M–P).

thumbnail image

Figure 2. A–L: Enhanced green fluorescent protein (EGFP) fluorescent pattern induced by heat-shock (HS) treatment of pHsS1/EGFP-injected embryos. pHsS1-EGFP (100 pg) was injected into the dorsal blastomere of four-cell stage embryos (B–D,F–H,J–L,N–P). Injected embryos were cultured at 16°C in 5% Ficoll until stage 11.5 and then were subjected to the heat treatments (3×: C,G,K; 6×: D,H,L). EGFP expression was observed at 0 hr (A–D), 1.5 hr (E–H), or 16 hr (I–L) after heat treatment. A,E,I,M: Normal embryos showed no fluorescence with or without HS treatment. B,F,J: In the pHsS1/EGFP-injected embryos, almost no fluorescent pattern was seen in the absence of heat shock induction. C,D,G,H,K,L: Conversely, the 3× HS treatment induced EGFP gene expression (C,G,K), and the 6× HS treatment showed more intense EGFP expression (D,H,L). K,L: At 16 hr after the heat treatment, a green fluorescent pattern (FL) continued to be seen. M–P: Superficial phenotypes of the embryos shown in I–L: no injection, no HS (M), pHsS1/enhanced green fluorescent protein (EGFP) injection (no HS, N; 3× HS, O; 6× HS, P). Any embryos, whether heat treated or not, showed almost no defects. Q–V: There were differences in the EGFP expression between pHsS/EGFP-injected embryos treated at 33°C and those treated at 37°C. Expression patterns at 5 hr (Q–S) or 12 hr (T–V) after heat treatment. When embryos were treated at 37°C (S,V), EGFP expression was stronger than in those treated at 33°C (R,U).

Download figure to PowerPoint

In previous studies, heat treatment often was done at 33°C (Krone and Heikkila,1989). We therefore checked differences in gene expression between 33°C-treated and 37°C-treated embryos. When heat treatment was carried out at 33°C, EGFP expression was detected, albeit at lower levels than in the 37°C treatment (Fig. 2S,V compared with R,U). Next, we checked the defect caused by these treatments. Under our experimental conditions, no defect was present in the uninjected embryos (data not shown). We next injected embryos with 100 pg of pHsS/EGFP into the dorsal region of four-cell stage embryos and examined the frequency of defect from the heat treatment. Constitutive expression by injection with pCS2/EGFP did not result in a given phenotype (data not shown); thus, it was expected that the defects of the pHsS1/EGFP-injected and heat-treated embryos reflected abnormalities derived from the heat treatment. Despite no heat treatment, approximately 5% of embryos showed some defect (Table 1). It was thought that these defects were mainly caused by toxicity of the DNA rather than the heat treatment because the dose of plasmid DNA used was high (Sive et al.,2000). Thus, the level of abnormalities seen with heat treatment alone could be regarded as background. In our experiment, at least, the frequency of abnormality at 37°C treatment was less than that induced by 33°C treatment and even by no treatment (Table 1). From these results, we selected a 37°C treatment for our subsequent experiments. Having said this, 37°C treatment in the early gastrula stage causes a moderately severe defect (Table 1). Therefore, alternative conditions such as treatment temperature and injection dose may need to be considered when using this method, according to the kind of gene expressed.

Table 1. Frequency of the Superficial Defects of Injected Embryos by Heat Treatmenta
Stage of heat treatmentHeat treatment (°C)nNo phenotypeDefectb
WeakStrong
  • a

    Four-cell stage embryos were dorsally injected with 100 pg of HsS1/EGFP, cultured until stage 12, and then heat-treated. Superficial phenotypes were observed at stage 38.

  • b

    Weak defect indicates slight head defect (such as small eyes, dorsoanterior index (DAI) >4) or weak curved axis, whereas strong defect indicates severe head defect (DAI <3) or short and severely curved axis).

No treatment3936 (92.3%)2 (5.1%)1 (2.6%)
10.5373430 (88.2%)2 (5.9%)2 (5.9%)
 333735 (94.6%)2 (5.4%)0 (0.0%)
12373534 (97.1%)1 (2.9%)0 (0.0%)
 333433 (97.1%)1 (2.9%)0 (0.0%)
14373332 (97.0%)1 (3.0%)0 (0.0%)
 333532 (91.4%)3 (8.6%)0 (0.0%)

To assess whether ectopic gene expression was coincident with the injected area, we coinjected pHsS1/EGFP with Alexa568 as a tracer. In injected embryos without HS treatment, only the red fluorescent signal of Alexa568 was seen (data not shown). When pHsS1/EGFP was injected into an equatorial region of one dorsal blastomere of four-cell stage embryos, green fluorescent signal was detected in half of the dorsal midlines of injected embryos (Fig. 3B). This expression pattern was overlapping with the red signal (Fig. 3A,C). Similarly, when pHsS1/EGFP was injected ventrally, red and green overlapping signals were seen in the posterior ventral region (Fig. 2D–F). These results suggested that EGFP expression was regulated in an injected area-dependent manner. In both cases, the area of red fluorescence was broader than that of the green signal.

thumbnail image

Figure 3. Coincidence of enhanced green fluorescent protein (EGFP) expression and the injected area. A–F: Embryos were coinjected with HsS1/EGFP and Alexa 568 into the dorsal (A–C) or ventral (D–F) region at the four-cell stage, cultured, heat-treated three times, and then observed at 1 hr after the treatment. A,B,D,E: Red signals indicate Alexa 568 distribution (A,D) and green signals show EGFP expression (B,E). C,F: Merged patterns of both signals are also shown. G–L: Detailed comparison of EGFP expression with Alexa 568 distribution. To do this, HsS1/EGFP and Alexa568 were injected into one blastomere at the 64-cell stage. G,J: No heat treatment (nohs). H,K: Immediately after the treatment (hs+0hr). I,L: Six hours after the treatment (hs+6hr). All figures show the merged patterns of Alexa 568 distribution and EGFP expression. The upper column (G–I) and the lower column (J–L) are individual samples, but show similar results in principle. H–L: In both cases, EGFP expression was correctly observed in the Alexa-positive area. M–O: Magnified figures of I: Alexa 568 distribution (M), EGFP expression (N), merged pattern (O). EGFP expression can be seen only in the Alexa 568–positive region of the cells. O: Arrows indicate an EGFP-positive cell, whereas arrowheads shows an EGFP-negative, Alexa-positive cell. Scale bars = 0.5 mm in C (applies to A–C), F (applies to D–F), I (applies to G–I), L (applies to J–L), 0.2 mm in O (applies to M–O).

Download figure to PowerPoint

To examine whether EGFP expression could be induced in a more restricted region, we coinjected Alexa and pHsS/EGFP into one blastomere of 64-cell stage embryos. EGFP expression was detected to a tightly restricted area within the Alexa-positive field (Fig. 3H,I,K,L). It should be pointed out that the mosaic expression of EGFP could be seen (Fig. 3I,L). To clarify whether EGFP expression was seen only in Alexa-positive cells, we examined EGFP expression at the single-cell level and showed that EGFP-expressing cells were restricted to only a part of the Alexa-positive cells (Fig. 3M–O, indicated by arrows). These results show that, even though the essential problem of unequal distribution with the injection of plasmid DNA is not completely unavoidable, our method has greatly improved the control of spatially restricted gene expression control of gene expression.

To determine whether ectopic gene expression could be controlled in a heat treatment-dependent manner, we performed Western blotting of the injected embryos. As a positive control, we injected CS2/EGFP, which is under the control of the CMV promoter, and examined the levels of EGFP expression. In CS2/EGFP–injected embryos, EGFP expression gradually increased after the mid-blastula transition (MBT) stage (Fig. 4A, row 1). We next injected pHsS/EGFP into the animal pole of two-cell stage embryos and did HS treatment at stages 10, 12, and 14. In all cases, EGFP protein was induced within the next stage after the HS treatment (Fig. 4A, rows 2–4). This result showed that EGFP expression was induced in a HS treatment-dependent manner. In addition, the amount of EGFP protein produced gradually increased with time of culturing (Fig. 4A, row 2), probably because of the high stability of the EGFP protein. Without HS treatment, EGFP protein expression was barely detectable, with a very weak band only detected at stages 15 and 18 (Fig. 4A, row 5). We further examined precisely when ectopic gene expression was initiated after heat induction. In the pHsS1/EGFP–injected embryos, no EGFP expression was detected without HS treatment (Fig. 4B, lane 1), and expression remained undetectable after a single HS treatment (15 min after the beginning of treatment; Fig. 4B, lane 2). After 2× HS treatment (45 min), weak expression was detected (Fig. 4B, lane 3), and after 3× treatment (>75 min), strong EGFP expression was observed, which increased gradually with time (Fig. 4B, lanes 3–9). This result showed that ectopic gene expression occurred within 1 hr, which was consistent with the study using Drosophila hsp70 promoter (Kojima et al.,1993).

thumbnail image

Figure 4. Temporal expression pattern of heat-induced enhanced green fluorescent protein (EGFP) protein in pHsS1/EGFP-injected embryos. A: CS2/EGFP (100 pg; row 1) or HsS1/EGFP (100 pg; rows 2–5) was injected into the dorsal region of four-cell stage embryos. CS2/EGFP-injected embryos were harvested at various stages (row 1). HsS1/EGFP-injected embryos were cultured until stage 10 (row 2), 12 (row 3), or 14 (row 4), followed by heat shock (HS) treatment. Row 5 shows the negative control in which injected embryos had no HS treatment. This figure suggests that the initiation of EGFP expression could be temporally controlled. B: Detailed analysis of the precise initiation of heat-induced gene expression. Injected embryos (lanes 1–9) or normal embryos (lanes 10–12) were heat-treated three times at stage 12 and cultured at 20°C. Embryos were harvested at various stages until stage 18, and then Western blot analysis was performed. EGFP protein was detected by anti-GFP antibody. EGFP expression was seen 1 hr after the start of the HS induction and maintained throughout the stage.

Download figure to PowerPoint

To check whether the vector alone could induce the observed phenotype, we constructed the HsS1/β-catenin expression plasmid. Previous studies have shown that anterior patterning requires inhibition of canonical Wnt signaling in the anterior neuroectoderm at the gastrula–neurula stage (Glinka et al.,1997). Down-regulation of a positive regulator of the canonical Wnt pathway, such as β-catenin, into animal pole (AP) cells causes enlargement of anterior neural tissues (Heasman et al.,2000). Moreover, overexpression of β-catenin at the presumptive anterior neuroectoderm caused loss of head structure (Kiecker and Niehrs,2001; Michiue et al.,2004). Thus, we expected that HsS1/β-catenin–injected embryos would show a similar phenotype after heat induction. To trace the injected area, we coinjected with HsS1/EGFP. In the absence of heat treatment, the injected embryos showed no head defects (Fig. 5E,I), and EGFP induction was not seen in these embryos (Fig. 4F,J). When HsS1/β-catenin was injected into the dorsal region and the embryos were HS treated at stage 11 or 12, a head defect phenotype was observed (Fig. 5G). The combination of AP injection and HS treatment at stage 11/12 also caused severe head defects (Fig. 5K). Intense EGFP expression was seen in the most anterior regions of the coinjected embryos (Fig. 5H,L). When HsS1/β-catenin was injected into the ventral region, no obvious defects were observed with or without HS treatment (Fig. 5M,N). PHsS/EGFP injection alone also did not cause loss of head formation after HS treatment (Fig. 5O,P).

thumbnail image

Figure 5. Heat inducible expression of β-catenin caused loss of head formation owing to the ectopic activation of late canonical Wnt signaling in presumptive neural ectoderm. A–D: As a positive control, CS2/β-catenin (100 pg) was injected into the ventral region (B), the dorsal region (C), or animal pole (AP, D). E–N: HsS1/enhanced green fluorescent protein (EGFP; 100 pg) and HsS1/β-catenin (100 pg) were coinjected into the ventral region (E–H) or dorsal region (M,N) of four-cell embryos, or the AP of eight-cell embryos (I–L), followed by heat shock (HS) induction at stage 11/12 (G,H,K–N) or no treatment (E,F,I,J). When HsS1/β-catenin and HsS1/EGFP were injected into the dorsal region of four-cell embryos or the AP of eight-cell embryos, a green fluorescent signal was seen only at the anterior region where obvious head structure defects were apparent (G,H,K,L). O,P: Only the HsS1/EGFP injection into the dorsal region (O) or AP (P) showed no phenotype despite heat treatment. Q: Head defects are summarized by a graph. The head defect index (described in the Experimental Procedure section) was used to measure the extent of the loss of head structure.

Download figure to PowerPoint

Taken together, we have shown that injection with an hsp70 promoter- containing plasmid can induce heat shock-mediated gene expression in a restricted area of the embryo and still induce the expected phenotype. Previous studies already have shown that microinjection with an hsp70 promoter-containing vector into Xenopus oocytes could induce exogenous gene expression (Krone and Heikkila,1989; Brakenhoff et al.,1991). Nevertheless, the novel point of our studies remains: we have been able to show that injection with pHsS vector into a restricted region of four- or eight-cell stage embryos can spatially target heat-dependent gene expression. This method allows us to assay the effects of temporally and spatially restricted gene expression in Xenopus embryos by a simple technique. Alternative methods to examine restricted gene expression using the hormone-binding domain of the steroid hormone have been reported previously. For example, expression of a fusion protein of glucocorticoid receptor can be induced on exposure to hormones such as dexamethasone (DEX; Kolm and Sive,1995). This method is also very useful in controlling the spatial and temporal expression of given genes (Tada et al.,1997; Gammill and Sive,1997), but our HS induction system holds several advantages. One of the important merits of our method is the versatility in being able to express all genes. The hormone inducible systems are mainly used for nuclear proteins such as the DNA/RNA-binding proteins (Scherrer et al.,1993) and, therefore, not easily applicable for the expression of cytoplasmic proteins. Another advantage is that gene expression in the heat induction system does not require another gene fragment such as the hormone-binding domain, which allows us to analyze the function of the native protein. Furthermore, treatment of the embryos with a special reagent such as DEX is not necessary in our system.

The heat-inducible system does have some minor shortcomings. One is the limitation of the embryonic stage at which heat treatment can be used successfully. When heat treatment is done at the blastula stage, severe defects in cell division result and prove lethal for the embryos before gastrulation (data not shown). Another problem is that the HsS construct-injected embryos showed faint leakage even in the absence of HS treatment (Figs. 1E,H, 4A, row 5). To minimize this leakage, injected embryos have to be set in temperatures less than 20°C. Nevertheless, our system is now available as an alternative method for the spatial and temporal regulation of ectopic gene expression in Xenopus embryos.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

Plasmid Construction

A summarized method for construction of the pHsS1 plasmid is described in the text. To obtain the pHsS1/EGFP plasmid, an EGFP fragment was excised by using an EcoRI–XhoI double digestion and inserted into the EcoRI–XhoI site of pHsS1. To make pHsS1/β-catenin, pBC40 (CS2/β-catenin) was digested with BamHI and NotI, and the fragment containing the complete β-catenin CDS was cloned into the BamHI and NotI sites of pHsS1.

Microinjection Into Xenopus Embryos

Microinjection was basically done as previously described (Sive et al.,2000). Plasmids for injection were purified by using QiaPrep Spin mini-columns (QIAGEN) and the plasmid concentration was checked by ultraviolet spectrophotometry. Xenopus embryos for injection were prepared by artificial fertilization and de-jellied by using 4.6% L-cystine hydrochloride. Plasmid DNA was injected into four- or eight-cell stage embryos. Injected embryos were incubated in 5% Ficoll /1× Steinberg's solution until they reached stage 8. The embryos were then transferred into 0.1× Steinberg's solution and cultured at 16°C until they reached the desired developmental stage. In Figure 5, we used the “head defect index” to measure the level of head defect (Michiue et al.,2004). In this index, the head defect is categorized into four classes: index 0, normal head; index 1, weak head defect with small eye embryo; index 2, strong defect with an extremely small, pin-shaped eye; index 3, embryo with no eye. Both sides of each embryo were classified and the results normalized. Thus, the value of the normal embryo equals 0 on the index, whereas maximum head-defect embryos were scored as 6.

Western Blotting

For Western blot analysis, three embryos were harvested at the appropriate stage then macerated by using a micropipettor in 60 μl of phosphate buffered saline containing a protease inhibitor cocktail (Complete, Roche). Sixty microliters of 2 × sodium dodecyl sulfate-polyacrylamide gel electrophoresis sample buffer was added to the sample, which was boiled for 3 min and centrifuged at 15,000 rpm. The samples were electrophoresed then transferred to polyvinylidene difluoride membrane. Immunoblotting was carried out using an anti–green fluorescent protein antibody (Santa Cruz).

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES

M.A. was funded by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, and this work was supported in part by the CREST and SORST programs of the JST.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES