• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Prospects for tissue-specific analysis of gene expression in Xenopus embryos

Posted on: Thursday, 2 October 2003, 06:00 CDT

Summary

Analysis of spatiotemporal patterns of gene expression is an important prerequisite for understanding the molecular basis of embryogenesis. Tissue-specific resolution is desirable, but often not achieved owing to methodical limitations. We used a common model system for embryonic development - the South African clawed frog Xenopus laevis - to demonstrate that laser microdissection and laser- mediated catapulting of tissue samples from histologic sections are feasible even for yolk-rich, fragile embryonic tissue. A combination with RT-PCR provides the possibility of detecting tissue-specific gene expression with high resolution and fidelity. We show that specimens of various sizes and shapes can easily be procured by laser microdissection and pressure catapulting (LMPC). Subsequent RNA-isolation and nested RT-PCR for marker genes revealed that the combination of these methods allows for analysis of specific gene expression in micro-areas. We report on the efficiency and reliability of detection of marker genes in dissected tissue. We further discuss the question of whether such a combination can be applied to certain issues raised in developmental biology with regard to other techniques.

Key words: Xenopus laevis - Laser microbeam microdissection (LMM) - Laser mediated catapulting (LMC) RT-PCR - Developmental biology

Introduction

One of the key questions raised in developmental biology refers to the regulation of spatiotemporal gene expression during embryogenesis. For this purpose, whole-mount in situ hybridization or RT-PCR is carried out. These two methods, however, are confined methodologically, because the former lacks sensitivity to low expressed transcripts and has difficulties analyzing more than two transcripts at the same time, and the latter has limitations regarding tissue-specific resolution. Tissue-specific resolution is desirable in many cases, but is difficult to achieve with manual dissection. This problem can be solved by combining laser microbeam microdissection (LMM) of histologic sections with RT-PCR. Owing to amplification steps, this technique is more sensitive than in situ hybridization. In addition, several different transcripts can be simultaneously assayed from only one section. A lot of different methods for RNA isolation and RT-PCR from cryopreserved or archival tissue have been reported to date [1, 6, 8, 12]. The ability to study preserved tissues at the molecular level is desirable, because it allows us to conduct retrospective studies of large numbers of embryonic tissue of any source, and may permit analysis of distinct expression patterns during development. For highly sensitive techniques, such as nested PCR, special care must be taken regarding sample preparation. In this study, we report on our experience in dealing with a laser microdis-section technique with non-contact sample procurement and optimized RNA isolation and RT-PCR, allowing for marker gene detection in small tissue areas from sections of formalin-fixed Xenopus embryos. Although formalin fixation causes strand breaks in nucleic acids, the extracted RNA is suitable for RT, followed by nested PCR amplification. Laser-mediated microdissection techniques allow us to extract morphologically distinct cells for molecular analysis [5]. A technical progress further allows for precise removal of pure cell populations from morphologically preserved cells and tissue sections without mechanical contact [12, 17]. The method is suitable for capturing clean samples of any shape and any size between one and several hundred micrometers in diameter. This is achieved by employing the laser not only for microdissection, but also for "beaming" selected samples directly into the cap of a common microfuge tube mounted on a special collector device - a process named "laser pressure catapulting" (LPC) [11, 18]. As LPC is usually combined with LMM, the method has recently been referred to as "laser microdissection and non-contact pressure catapulting" (LMPC). For ease of observation, the yolk-rich Xenopus embryo, which has large sized fragile cells, was used to improve this technique and to allow access for embryonic tissues in general. Standard techniques, such as microinjection and animal cap assays, can be applied to a broader range of problems when combined with LMPC. If microinjection is applied to studying the effect of ectopic gene expression on induction of other genes, detection of marker gene expression in cell subpopulations is a prerequisite for the study of cell differentiation controlled by the gene of interest. For example, it is possible that some induced genes are abundantly expressed in other embryonic tissue, thus limiting the detection of induction in whole-embryo RNA even in quantitative PCR analysis. Animal cap assays allow for the examination of the influence of various growth factors on differentiation of uncommitted ectodermic cells. However, these cells usually do not differentiate uniformly, but into a variety of cell types. In this case, LMPC would be a useful tool for selecting and pooling morphologically similar cells, where manual segregation proves virtually impossible. This opens the field for differential gene expression screens between micro-areas.

Materials and Methods

Specimen preparation

Xenopus embryos were staged according to Niewkoop and Faber [13], the vitelline membrane was removed manually with two forceps, and embryos were fixed in 3% paraformaldehyde in APBS, (amphibian phosphate buffered saline: 2 mM KCl, 0.15 mM KH^sub 2^PO^sub 4^, 103 mM NaCl, 0.7 mM Na^sub 2^HPO^sub 4^, pH 7.5) for 1 h at room temperature. Fixed embryos were washed twice in APBS and dehydrated through 25%, 50%, 75%, 90%, and 100% ethanol for 10 minutes each. Ethanol was then replaced by isopropanol in three washes for 10 minutes. Embryos were then incubated in xylol/isopropanol 1:1, and pure xylol for half an hour at room temperature. After transfer to an oven at 65 [degrees]C, they were passed through paraffin/xylol 1:3, 1:2, 1:1, and pure paraffin for one hour each. The embryos were then embedded in fresh paraffin and left to cool down overnight at 4 [degrees]C. Serial sections of 6 to 8 [mu]m were mounted on two types of glass slides. Glass slides covered with a 1.35-[mu]m thin UV-radiated polyethylene-naphthalene membrane (P.A.L.M. Bernried, Germany) were coated with 1% poly-L-lysine (Sigma, Deisenhofen, Germany) to enhance adhesion of the sections to the surface. Plus charged slides (Super-Frost plus; Menzel-Glaser, Braunschweig, Germany), suitable for direct LPC, were used without further preparation. Mounted sections were dewaxed by incubation in two changes of xylol for five minutes and a brief wash in acetone. For frozen sections, embryos were fixed in zincfix (5 g ZnCl^sub 2^, 6 g ZnAc^sub 2^ x 2H^sub 2^O, 0.1 g Ca Ac^sub 2^ in 0.1 M Tris, pH 7.4) for 1.5 h at 4 [degrees]C, and incubated overnight in zinc-fix with 30% sucrose. Embryos were mounted in Tissue-Tek OCT compound (Sakura Finetek; Torrence, Canada), 30-[mu]m frozen sections were mounted on Super-Frost slides and immediately stored at -80 [degrees]C. Prior to laser microdissection, slides were incubated on ice in zinc-fix for 5 min to dissolve the OCT, and dehydrated by incubation in 70%, 95%, and pure ice-cold ethanol for l min each and covered with 1:2 liquid coverglass (P.A.L.M., Bernried, Germany) in isopropanol to improve visualization.

Laser microdissection and non-contact pressure catapulting (LMPC)

Microdissection and sample collection were performed using the commercially available PALM(R)MicroBeam (P.A.L.M., Bernried, Germany). The device consists of a pulsed, low energy 337-nm nitrogen laser coupled into an inverted microscope (Axiovert 135; Carl Zeiss, Gottingen, Germany) via the epifluorescence path. The laser microbeam was focussed through a 10x, 20x, or 40x dry objective lens. For microdissection, the laser energy was adjusted to cut the entire membrane-tissue stack without scattering the cells. Settings were dependent on the absorption behavior of the specimen and on the transmission of the respective objective lens. Samples for microdissection were fixed on a motorized, computer- controlled microscope stage (RoboStage, P.A.L.M., Bernried, Germany) that can be directed by a mouse cursor. The track of the laser microbeam can be followed on a video screen. For the catapulting of samples, energy was sufficiently increased to deliver the specimen into a PCR-oil-coated microfuge tube cap, which was held and centered closely above the line of laser fire by a special micromanipulator (CapMover, P.A.L.M., Bernried, Germany). Catapulting rather than photolysis of the sample was achieved by delivering a single high-energy laser-shot after focussing slightly below the section.

Samples for RNA-preparation were procured in two ways. First, the desired tissue was cut out of membrane-mounted sections along with the supporting membrane and catapulted into a microfuge tube cap. In that case, the morphology of the catapulted specimen was preserved, and it was possible to check the size and number of dissectates in the cap. Second, desired areas of sections mounted on Super-Frost slides were directly catapulted employing the auto-LPC function of the device. Using this function, all the material in a marked area is directly catapulted in small portions by a series of high-energy laser-shots slightly below the focus of the section. In that case, the morphology of the catapulted specimen was not preserved, however, it allowed for quicker procurement of the samples.

RNA isolation and RT-PCR

Total RNA was isolated using the PUREscript RNA-isolation kit (Biozym, Hessisch Oldendorf, Germany) according to a modified manufacturer's protocol. Volumes were scaled down to one tenth of the indicated minimal volume, and centrifugation steps were carried out at 4 [degrees]C for 15 min. Thirty microliters of cell lysis solution was pipetted into the caps containing dissected specimens, and the corresponding tubes were closed onto them. Cell lysis was achieved through incubation of the inverted tubes on a vortex for 15 min at room temperature. The tubes were then briefly centrifuged to allow the lysate to flow to the tube bottom, and 10 [mu]l of DNA/ protein precipitation buffer was added and mixed by gently flicking the tubes. After incubation on ice for 5 min, tubes were centrifuged at maximum speed in a common microfuge. The supernatant was saved, and total RNA precipitated with an equal volume of isopropanol containing 1 [mu]l of 20 mg/ml glycogen. The RNA-pellet was allowed to dry at room temperature for 15 min and then hydrated in 10 [mu]l RNase-free water containing 10 U RNase-inhibitor in an ice bath for 30 min.

For RT-PCR analysis, the ExpeRT-PCR Kit from Hybaid-AGS (Hybaid- AGS, Heidelberg, Germany) was used, since the RT reaction and PCR could be carried out in the same buffer optimized for both reactions. Nevertheless, the ExpeRT-PCR kit allowed for a two-step reaction, which was necessary for the use of random hexamer primers rather than gene-specific primers for the RT reaction. Random hexamer primers gave better results in RT reactions than did gene- specific primers, and allowed for the analysis of several genes from one RNA sample. The RT reaction was set up in a 25-[mu]l reaction volume, basically following the two-step protocol of the ExpeRT-PCR kit. In addition, 50 U of Superscript RT (GIBCO BRL, Life Technologies, Karlsruhe, Germany) was added to the reaction, as this increased the yield of cDNA. Isolated RNA from one dissectate was subjected in total to one reverse transcriptase (RT) reaction. A negative control reaction lacking RT could be avoided, as subsequent PCRs were performed by employing primers annealing to different exons of a given marker gene. Thus, amplification products deriving from mRNA and genomic DNA could be distinguished by size. PCRs were set up under a PCR/UV workstation. We routinely used 5 [mu]l of the RT reaction as an input for the first PCR employing gene specific primers. Thus, up to 5 different marker genes were assayed from one microdissected sample. For the first PCR, we performed 35 cycles in a 25-[mu]l reaction. Nested PCR amplifications were carried out in a 25-[mu]l reaction volume with a 5-[mu]l aliquot of the first PCR using nested primers. The nested PCRs were carried out with Taq polymerase from Eurogentech (Eurogentech Darmstadt, Germany), as this enzyme showed a significantly higher activity than did the Taq/ PWO mix from the ExpeRT-Kit. We routinely performed further 35 cycles for the second PCR. Positive controls with whole embryo RNA and negative controls with water or RNA from tissue supposedly negative for the gene were run in parallel. PCR products were sequenced to verify their identity in addition to comparing the band size to a hundred base pair ladder after electrophoretic separation on a 2% agarose gel.

The following thermocycler conditions were used: 2 min at 94 [degrees]C (30 s at 94 [degrees]C; 30 s at 55 [degrees]C; 1 min at 72 [degrees]C) 5 min at 72 [degrees]C. An annealing temperature of 52 [degrees]C was used for the primer pair myoD F2/R2.

Primers for nested PCR:

Endodermin:

F1: TATTCTGACTCCTGAAGGTG,

R1: TAAAGCGCACAGAGACGCACG;

F2: GTGCAGTGTTTCACTACTTGG,

R2: GAGAACTGCCCATGTGCCTC;

FoxA4b:

F1: CCTCCGCTGTGGACTCCAGA,

R1: GCCTATGTTTCCCTGAATACTGG;

F2: GAGTCAAATTGGAAATTAAGGATCC,

R2: CATTCCTTGAGGCATGGAAC;

myoD:

F1: AACTGCTCCGATGGCATGATGGATTA,

R1: GATTACTCTCTCGCTCAATGTCT;

F2: CCTGCGGCTCCAGGAGAAGGAAC,

R2: ATTGCTGGGAGAAGGGATGGTGATTA;

NCAM:

F1: AAATGCTGTGGATTGGAAACCAG,

R2: GGCTTGGCTGTGGTTCTGAAGG;

F2: GCGGGTACCTTCTAATAGTCAC,

R2: GCTGGCACTTGTGGCTGTAGTAG;

Results

For microdissection, the laser microbeam was focussed through a 10x, 20x, or 40x dry objective lens. The laser energy was adjusted to cut the entire membrane-embryo stack. Figure 1 shows how a large tissue area from a transversal membrane-mounted paraffin section of a stage-19 embryo can be removed in one single step (Figs. 1A-C). Laser microdissection and pressure catapulting were performed using the 10x objective lens, while photographs were taken with the 5x objective lens for achieving better overview. Precise microdissection by laser microbeam, without scattering the cells, is possible even from large yolk-rich areas represented by the endodermal yolk-mass (Figs. 1A and C). One laser shot directly catapulted the dissected endoderm into the cap of a common microfuge tube, which was centered closely above the section, as seen in Figure 1C. It is noteworthy that a dissectate as large as 1 mm in diameter can be catapulted in one step. The cells were simply covered with the buffer for total RNA isolation. The attached tubes were closed onto the caps and processed as described. As shown in Figure 1D, expression of the panendodermal gene endodermin, the first molecular marker of endoderm specification [15], could be detected in the dissectate after nested RT-PCR.

Figure 2 shows the LMPC of a much smaller structure - the neural tube. In this case, the laser was focussed through a 20x objective lens, while the 10x objective lens was used for documentation. As can be seen in the image of the catapulted neural tube in the cap (Fig. 2C), the morphology of the catapulted specimen is well preserved. The piece of tissue shown yielded enough RNA to obtain a clear band for neural cell adhesion molecule (NCAM) [11] after nested RT-PCR with 70 cycles. NCAM is involved in cell-cell interactions during the development of the vertebrate nervous system, and is restricted to the neuroectoderm during formation and closure of the neural tube. Figure 3 shows a LPC sequence for the notochord of a stage-34 embryo. A 30 [mu]m frozen section was mounted on a plus-charged slide without supporting membrane. Figure 3A shows the intact section. In spite of improved visualization through covering with liquid coverglass, morphologic tissue perseverance is inferior to formalin-fixed, paraffin-embedded tissue. After encircling the requested region via mouse cursor (Fig. 3B), LPC is performed automatically. As shown in Figure 3C, the marked region is well cleared of cells. This means that samples can be procured without effort, even from thick sections. Nevertheless, we used membrane-mounted, paraffin-embedded tissue for systematic analysis on PCR-efficiency, as the boundaries between different tissues were more clearly visible. In addition, the supporting membrane provides a backbone for catapulted tissue, helping to preserve the morphology of catapulted specimen, which allows us to check the quality of the isolated material prior to RNA-isolation and RT-PCR-analysis.

Fig. 1. One specific experiment showing LMPC images of the endoderm and the corresponding RT-PCR. A: Overview of a membrane- mounted transversal section of a stage-19 embryo. B: The same section after microdissection and catapulting of the endoderm. C: Catapulted endoderm in the tube cap. D: RT-PCR for endodermin C. Lane 1: Endodermin PCR-product derived from RNA isolated out of the dissectate shown in C. Lane 2: Negative control (water). Lane 3: Positive control (RNA from whole embryo). Scale bars, 200 [mu]m.

Table 1. RT-PCR amplification efficiency of marker genes

Figure 4 shows a LMPC sequence for the notochord of a formalin- fixed, paraffin-embedded membrane-mounted embryo and the set-up for the experiments leading to the figures in Table 1. Figure 4A shows the dorsal part of an intact transversal section from the middle part of a stage-20 embryo. Figure 4B shows the section after microdissection of the notochord prior to catapulting. In Figure 4C, the high-energy shot for catapulting was delivered, and 4D shows the catapulted notochord in the cap. The sample was then submitted to RNA-isolation, and RT-PCR was carried out for the notochord- specific transcription factor FoxA4b, which is implicated in embryonic pattern formation and expressed in the Spemann-Mangold organizer region and later in the embryonic notochord [10], and for the somite-specific transcription factor myoD, which is a key regulator of muscle determination restricted to muscle precursor cells after mesoderm induction [14]. As expected, only the notochord- specific marker gene could be detected in the captured notochord. We chose to assay myoD expression as a control for pureness of the captured notochord cells, as the somites are tissues adjoining the notochord (Fig. 4E, lanes 1 and 2). In lanes 3 and 4, positive and negative controls for PCR are shown. We conducted similar experiments to achieve data on the reliability and efficiency of marker gene detection after LMPC. These data are summarized in Table 1. The amplification efficiency of a marker gene is expressed in the ratio of cases, where expected gene expression could be detected in dissected tissue. At least 5 independent experiments were performed for each gene. False positives stand for the detection of gene expression in tissue supposedly negative for this gene. Apart from genes already mentioned, we assayed epidermal cytokeratin and E- cadherin in the epidermis [3, 9], and sonic hedgehog in the floor plate - the ventral part of the neural tube - [4]. In all the tissues, Histone H4 was assayed as a positive control, as this is a ubiquitously expressed gene. We obtained amplification efficiencies of 20 to 50% depending on the marker gene. False positives were only detected in the case of FoxA4b in the endoderm, which is less than 2% of the total cases tested.

Fig. 2. LMPC of the neural tube. A: Intact section seen with the 10x objective. B: Section after laser microdissection and pressure catapulting. C: Neural-tube-fragment in the cap. D: RT-PCR for NCAM from C. Scale Bars, 100 [mu]m.

We also compared two methods of sample procurement - direct LPC versus membrane-supported dissection and catapulting. We always observed higher amplification efficiency in membrane-mounted tissue. Figure 5 shows the outcome of a representative experiment. Lanes 1 and 2 display the RT-PCR results for four notochords dissected and catapulted with the sustaining membrane, and 8 notochords procured by direct LPC. A positive signal for FoxA4b (arrow) was obtained only for the membrane-mounted tissue, although starting material was doubled for direct LPC. Lanes 3 and 4 show negative and positive controls, respectively. The arrow-head marks primer dimers, which can form in the absence of amplifiable material.

Fig. 3. LMPC of the notochord from a 30-[mu]m frozen section of a stage-34 embryo. A: Intact section covered with liquid cover-glass. B: Screen-view after marking the notochord for LPC. C: Section after LPC. Scale Bars, 100 [mu]m.

Discussion

Tissue microdissection and single-cell procurement is one of the most advanced techniques in modern gene analysis. LMPC is a non- contact technique for specific, reproducible, and efficient sample isolation. We have already demonstrated that it is applicable to the Xenopus system, and that commonly used dyes, such as HE-staining and DAB-immunostaining, do not negatively impact mRNA integrity and RT- PCR [7]. Compared to mouse or chicken embryos, one problem arising in the processing of embryonic Xenopus tissue lies in the high proportion of yolk, which absorbs much of the laser energy and might interfere with mRNA accessibility. Nevertheless, our results prove that any desired tissue area can be cut out by selecting the appropriate objective lens and adjusting laser energy levels accordingly. Automation of microdissection and catapulting is possible to a certain degree, as shown for LMPC of frozen sections in Fig. 3. However, manual control should be preferred, as sections of Xenopus embryos are very inhomogeneous with regard to absorption of laser energy, depending on the density and the yolk content of different tissues.

For specific PCRs, we designed primers that spanned intron-exon- sequences to allow for the distinction of PCR products derived from mRNA or DNA. Thus, minus RT-reactions, as a negative control, can be avoided, and RNA for specific marker amplification can be saved. This is favorable, as the selection of small areas or even individual cells results in the recovery of very little nucleic acids. Regulatory genes, such as signalling molecules or tissue- specific transcription factors, are often expressed at relatively low levels, compared to abundant house keeping genes. In addition, these molecules are often expressed by a small cluster of cells. An advantage of the LMPC-method lies in the fact that these cells or cell clusters can be precisely separated from the surrounding tissue. In this study, false positive signals were limited to endoderm samples, where expression of FoxA4b could unexpectedly be detected in 11% of the cases. It is therefore conceivable that low FoxA4b expression is present at the notochord/endoderm boundary, and has not been reported previously because of the low sensitivity of other techniques. The minimal number of false positive signals (in total less than 2%) in all the tissues examined demonstrates that, owing to the precision of laser microdissection and laser-mediated sample procurement, studies can be conducted in which the proof of gene induction in small tissue areas is required. If a knock down of genes is to be demonstrated, however, amplification efficiency has to be improved. This should be readily achieved by pooling several samples of the same kind before analysis of mRNA expression. Alternatively, thicker sections could be used to collect desired tissue samples. As shown in Figure 3, 30 [mu]m thick sections can be processed by LPC without problems. Considering the low amount of starting material used in this study, the amplification efficiencies of marker genes are promising. They ranged from 20 to 50 percent depending on the size of the dissectate and the number of captured cells. This is comparable to results obtained from mammalian tissues [6, 17]. Several factors might account for the result: sections are only 6 [mu]m thick so that part of the cytoplasm is cut off. Furthermore, the yield of RT-reaction itself can be highly variable even in samples with abundant mRNA. As formalin induces strand breaks in nucleic acids, including mRNA molecules, we amplified only small RT-PCR fragments [2, 12, 17]. However, there was no apparent correlation between product length and amplification efficiency of a marker gene, nor did the estimated abundance of mRNA molecules account for the different results. It is known that efficiency of PCR can vary for different primers. Thus, in cases such as this, with the amount of starting material being critically low, special care should be given to primer design.

Fig. 4. LMPC images of the notochord. A: Intact section seen with the 20x objective. B: Section after laser microdissection of the notochord. C: Section after the catapulting of the notochord. D: catapulted notochord in the tube cap. E: RT-PCR of the catapulted specimen and controls. Lanes 1 and 2: FoxA4b, but not myoD, could be detected in RNA from the dissected notochord shown in C. Lanes 3 and 4: Negative and positive controls for FoxA4b (water and RNA from whole embryo). Scale Bars, 100 [mu]m.

Fig. 5. One representative experiment comparing RT-PCR efficiency in samples procured by catapulting morphologically intact samples with sustaining membrane, and samples of the same tissue procured by direct LPC.

Lane 1: four notochords dissected and catapulted as shown in Fig. 4; lane 2: equivalent of eight notochords procured by direct LPC; lane 3: negative control (water); lane 4: positive control (whole embryo). Arrowhead: primer dimers.

It is important to use the supporting membrane with regard to quick isolation and collection of entire tissue areas, e.g. notochord, neural tube, floor plate. As the captured cells are not further damaged, this leads to a higher recovery of intact mRNA molecules. We observed a striking difference in the efficiency of RT- PCR amplification from material collected by cutting and catapulting entire tissue areas with the supporting membrane or by direct LPC. Apart from the fact that only cell fragments can be gained by direct LPC, this is possibly due to electrostatic forces evoked by repeated high-energy laser shots, sending small particles off track so that some do not reach the collector device (Dr. K. Schutze, personal communication). This could be avoided by diminishing the travelling distance.

LMPC is even applicable to thick frozen sections, as shown in Figure 3. Therefore, it can be used for analysis of gene expression in areas marked by in-situ hybridization or immunostaining, even in morphologically similar tissue areas. This is only partly feasible using paraffin sections, as only very intense staining is visible on thin sections, and paraffin-embedding may interfere with immunostaining. LMPC is also suitable for mRNA profiling, as it rapidly allows for dissecting, collecting, and pooling even large numbers of cells [16]. As microarrays are increasingly used, this is certainly an important aspect to keep in mind.

The application of LMPC in combination with other techniques is an important step forward in gaining more information about the molecular events responsible for embryonic development.

Acknowledgements. We thank Petra Dietmann for her excellent technical assistance, and Dr. Georgia Lahr for her helpful discussion. This work was supported by a research fund of the DFG (1208/9-1) given to D.W.

PATHOLOGY

RESEARCH AND PRACTICE

(C) Urban & Fischer Verlag

http://www.urbanfischer.de/journals/prp

Pathol. Res. Pract. 199: 381-389 (2003) 0344-0338/03/199/06-381 $15.00/0

References

1. Bernsen MR, Dijkman HB, de Vries E, Figdor CG, Ruiter DJ, Adema GJ, van Muijen GN (1998) Identification of multiple mRNA and DNA sequences from small tissue samples isolated by laser-assisted microdissection. Lab Invest 78: 1267-1273

2. Bonner RF, Emmert-Buck M, Cole K, Pohida T, Chuaqui R, Goldstein S, Liotta LA (1997) Laser capture microdissection: molecular analysis of tissue. Science 278: 1481-1483

3. Choi YS, Gumbiner B (1989) Expression of cell adhesion molecule E-cadherin begins at gastrulation and predominates in the ectoderm. J Cell Biol 108: 2449-2458.

4. Ekker SC, McGrew LL, Lai CJ, Lee JJ, von Kessler DP, Moon RT, Beachy PA (1995) Distinct expression and shared activities of members of the hedgehog gene family of Xenopus laevis. Development 727: 2337-2347

5. Emmert-Buck MR, Bonner RF, Smith PD, Chuaqui RF, Zhuang Z, Goldstein SR, Weiss RA, Liotta LA (1996) Laser capture microdissection. Science 274: 998-1001

6. Fink L, Kinfe T, Stein MM, Ermert L, Hanze J, Kummer W, Seeger W, Bohle RM (2000) Immunostaining and laser-assisted cell picking for mRNA analysis. Lab Invest 80: 327-333

7. Imamichi Y, Lahr G, Wedlich D (2001) Laser-mediated microdissection of paraffin sections from Xenopus embryos allows detection of tissue-specific expressed mRNAs. Dev Genes Evol 211: 361-366

8. Jin L, Thompson CA, Qian X, Kuecker SJ, Kulig E, Lloyd RV (1999) Analysis of anterior pituitary hormone mRNA expression in immunophenotypically characterized single cells after laser capture microdissection. Lab Invest 79: 511-512

9. Jonas E, Sargent TD, Dawid IB (1985) Epidermal keratin gene expressed in embryos of Xenopus laevis. Proc Natl Acad Sci USA 82: 5413-5417

10. Knochel S, Lef J, Clement J, Klocke B, Mille S, Koster M, Knochel W (1992) Activin A induced expression of a fork head related gene in posterior chordamesoderm (notochord) of Xenopus laevis embryos. Mech Dev 38: 157-165

11. Krieg PA, Sakaguchi DS, Kintner CR (1989) Primary structure and developmental expression of a large cytoplasmic domain form of Xenopus laevis neural cell adhesion molecule (NCAM). Nucleic Acids Res 17: 10321-10335

12. Lahr G (2000) RT-PCR from archival single cells is a suitable method to analyse specific gene expression. Lab Invest 80: 1477- 1479

13. Niewkoop PD, Faber J (1967) Normal table of Xenopus laevis (Daudin). 2nd ed. Hubrecht Lab., Utrecht, North-Holland Publishing Co., Amsterdam

14. Rupp RA, Weintraub H (1991) Ubiquitous MyoD transcription at the midblastula transition precedes induction-dependent MyoD expression in presumptive mesoderm of X. laevis. Cell 65: 927-937

15. Sasai Y, Lu B, Piccolo S, De Robertis EM (1996) Endoderm induction by the organiser-secreted factors chordin and noggin in Xenopus animal caps. EMBO J 15: 4547-4555

16. Scheidl S J, Nilsson S, Kalen M, Hellstrom M, Takemoto M, Hakanson J and Lindahl P (2002) mRNA expression profiling of laser microbeam microdissected cells from slender embryonic structures. Am J Pathol 160: 800-813

17. Schutze K, Lahr G (1998) Identification of expressed genes by laser mediated manipulation of single cells. Nat Biotechnol 16: 737- 742

18. Schutze K, Posl H, Lahr G (1998) Laser micromanipulation systems as universal tools in cellular and molecular biology and in medicine. Cell Mol Biol 44:735-146

Received: March 19, 2003

Accepted: May 8, 2003

Yukiko Imamichi1, Katja Koebernick2, Doris Wedlich2

1Department of Internal Medicine I, University of Ulm, Germany

2Institute of Zoology II, Molecular Embryology and Cell Physiology, University of Karlsruhe (TH), Germany

Address for correspondence: Yukiko Imamichi, University of Ulm, Department of Internal Medicine I, Robert-Koch-Str. 8, D-89081 Ulm, Germany.

E-mail : yukiko.imamichi@medizin.uni-ulm.de

Copyright Urban & Fischer Verlag 2003

More News in this Category