Vol. 189, No.‘~, December
BIOCHEMICAL
1992
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Pages 1561-1568
30, 1992
A XENOPUS PROTEIN:
HOMOLOGUE
DEVELOPMENTAL
OF THE HUMAN REGULATION
P-AMYLOID
PRECURSOR
OF ITS GENE EXPRESSION
Haruo Okado and Harumasa Okamoto Department of Neurobiology, Institute for Brain Research, Faculty of Medicine, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan Received
November
25,
1992
SUMMARY: Complimentary DNA clones have been isolated from Xenopus larva to delineate a protein highly homologous to the human b-amyloid precursor protein (APP). Developmental change of Xenopus APP gene expression has been analyzed with molecular probes. From early oogenesis, there is a high accumulation of maternal APP. After fertilization, the mRNA is degraded, reaching a minimum level around the gastrula stage. Then zygotic transcription appears to be initiated, and this continues during the subsequent embryonic and larval stages. Splicing patterns differ between the maternal and zygotic transcripts. The ratio of mRNA including the protease inhibitor domain (PID) sequence is extremely low for the transcript of maternal origin as compared to that for the transcript of zygotic origin. These results suggest some roles for the APP molecule in Xenopus early 0 1992 Academic Press, Inc. development.
The p-amyloid precursor protein (APP) gives rise to a 4.2~kDa peptide that is deposited in extracellular dense-core amyloid plaque of Alzheimer’s disease (1,2). Molecular cloning studies have revealed that there are at least three types of APP mRNA produced by alternative splicing, each encoding 695,75 1 and 770 amino acid residues (APP695, APP751 and APP770) in human (2, 3-5), mouse (6,7) and rat (8,9). The sequence of the protease inhibitor domain (PID) is included in the latter two types (3-5). Since human APP transcripts are abundant in fetal tissues as well as in adults (4, 5) and mouse APP transcripts can be detected in oocytes and early embryos by the PCR technique (lo), a developmental role for APP is highly likely, Recently an APP homologue of but the data so far obtained seem rather fragmented. Drosophila
was identified and suggested to be required for embryonic neural development (11,
12). In the present experiments, we have isolated cDNA clones for the APP homologue from Xenopus as an initial step to provide insight into the normal function of APP in vertebrate development. Due to the ease of experimental manipulation, this species has already been used to obtain a considerable amount of data on developmental processes, including the neural one (13), at the cellular and molecular levels We report here that the APP homologue of Xenopus is highly conservative to the mammalian APP and that the Xenopus APP transcript exists from early oogenesis, remaining throughout the embryonic and larval stages, although its relative amount and splicing pattern changes during development. 0006-291X/92
1561
$4.00
Copyright 0 1992 bJ Academk Press. Inc. All rights of reproduction in any jimn reserved.
Vol. 189, No. 3, 1992
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
MATERIALS
AND METHODS
Total RNA was extracted from Xenopus Construction of two sets of cDNA libraries: tadpoles with guanidinium thiocyanate, followed by centrifugation in CsCl solution (14). Poly(A)+RNA was purified by an oligo(dT)-cellulose spun column (mRNA Purification kit, Pharmacia) and employed for construction of a unidirectional cDNA library using a ZAPcDNA synthesis kit (Stratagene) with oligo(dT)-linker primers. A random primed cDNA library was also made from poly(A)+RNA using a You-prime cDNA synthesis kit (Pharmacia). The amplified unidirectional cDNA library (7X 1O5 Screening of the cDNA libraries: plaques) was screened with human APP cDNA fragment probes, KX and XB, which were kindly provided by Drs. T. Ishida and Y. Takahashi (Asahi Chemical Industry Co.). The plaque-transferred filters (Hybond-N+; Amersham) were hybridized with the two radiolabelled probes for 2 days at 37’C in 5x SSPE, 0.5% SDS, 5x Denhardt’s solution, 35% formamide and 20 pg/ml sperm DNA. After this, the filters were washed twice for 10 min. at room temperature in 2x SSPE and 0.1% SDS, followed by washing with the same solution for 15 min. at 50°C and then exposed to Kodak XAR film for about one day at -7O’C. The original random primed library (5~10~ plaques) was screened with a 191-bp fragment of the Xenopus APP homologue to obtain the 5’portion of the coding sequence. This fragment was constructed by PCR with the following degenerate primers, the sequences of which are highly conserved between the human APP and its Drosophila homologue at their 5’regions: S>primers: S>ATATGGATCC(AT)(CG)IGA(TC)GCI(TC)TI(TC)TIGT(ATCG)CCDrimers: 5>CGTAGAATTCGG(AG)CA(AG)CA(ATCG)AC(AG)AA(TC)TC(ATCG)AC< 3, in which GAATTC is the EcoRI site and CGTA is a nonsense sequence. These primers were applied to single strand cDNA made from the poly(A)+RNA fraction with a random hexamer. The template was then amplified under the following conditions: 94°C (1 min.), 37’C (1 min.), transition time (2min.), 72’C (2 min.) in the first 5 cycles and then 94’C (1 min.), 50°C (1 min.), 72’C (2 min.) in 30 cycles. The plaque-transferred nylon filters were hybridized with the radiolabelled 191-bp probe for one day at 42’C in 5x SSPE, 0.5% SDS, 5x Denhardt’s solution, 50% formamide and 20 &ml sperm DNA. They were washed twice at room temperature for 10 min. and once at 42’C for 15 min. in 2x SSPE and 0.1% SDS. Nucleotide sequences: DNA sequencing was performed by the dideoxynucleotide chain termination method with Sequenase Version 2 (United States Biochemical Corp.). Northern blotting analysis: Total RNA was extracted from oocytes, embryos and larvae at various developmental stages by the proteinase K method (15). They were electrophoresed through a 1% agarose-formaldehyde gel (30 @ lane), blotted to Hybond-N (Amersham), fixed by exposure to UV light, stained with methylene blue (16) and then hybridized at 43’C in 50% formamide, 5x SSPE, 5x Denhardt’s solution and 0.5% SDS. The probe applied was a cDNA clone of the Xenopus APP homologue (KB2, see text), which was labelled by a Multiprime Labeling System (Amersham). The filter membrane was washed in 0.1% SDS, 0. lx SSPE at 65*C. PCR for determining the splicing pattern: Two sense primers were used: ATCTGGATCC TGCAGTGAGAAGAGCATGAG (designated as 1 l), in which GGATCC is the BarnHI site and ATCT is a nonsense sequence, and CTTTGGATCCCGAGGAACCCTATGAAGAAG (designated as 51), in which GGATCC is the BamHI site and CITT is a nonsense sequence. One antisense primer was used: CCTCTCCTTTGCTTTCA (designated as 106). The two pairs of sense and antisense primers were applied to single strand cDNA made from the total RNA using random hexamers. The template was amplified at 94’C (1 min.), 55’C (1 min.) and 72°C (1 min.) for 30 cycles (see the Fig. 5 legend for further details).
RESULTS AND DISCUSSION Molecular cloning of Xenopus APP cDNA: The unidirectional cDNA library generated from Xenopus tadpole was screened with two radiolabelled human APP cDNA probes, KX and XB. The longest clone (KB2, 2kb) out of three positives was still short of the 5’portion of the coding region when compared with the published human APP cDNA sequence. To isolate the 1562
Vol.
189,
No.
BIOCHEMICAL
3, 1992
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
5’ portion, the random-primed cDNA library was screened with a 191-bp fragment of the Xenopus
APP homologue in this region, which was made by the PCR technique (see
By sequencing some of the nine positive clones obtained the total coding region was determined, Figure 1 shows the delineated nucleotide sequence of the Xenopus APP homologue that includes a coding region of 2241 bp, corresponding to 747 amino acid residues; a 5’ untranslated region of 33 bp; and a 3’untranslated region of 720 bp. When the deduced amino acid sequence is compared with that of human APP7.51 (Fig. 2). MATERIALS
AND METHODS).
they share a high degree of homology (87.5%). human APP is 78.3%.
Fig. 1. Xenopus code).
At the nucleic acid level, the homology to
The Xenopus sequence contains PID (84.2% homologous to human
APP nucleotide
sequence and deduced amino acid sequence (single-letter
1563
Vol. 189, No. 3, 1992
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
42
YLP”KLARADCkRLALEYLVDG~GQLLAEPOI*UFSVARLN~ I :; :::: .:::::::::::::::: ::::: : : : .ALEVPTDONAGLLAEPOIAMFc.o~~N~ I YLPGLALLLLAAYTAR 43 44
43 ss
H.HVONGKIiTDVSG. .ClGTKEGlLOYiOEV”PELOliNVVEANOPV~ :: I:::::::::::::I:::::::::::::: :::::::::: :.:: HUNVONOKYDSDPSOTKTClOTKEOlLOYCOEVYPELPITNVVEANOPVT
9s
SO lONWCKKORKOCKSRTHlVVP”flCLVGEFVSDALLVPDK~KF~HOE~ND~ :::::: ::I::: ;.; :;;::I::I:;:;;III;:::;I:::;;:.
130
94
IONWCKRQRKOCKTHPHFVIPYRCLVGEFVSDALLVPDKCKFLHOERYDV
143
CETHLHYHTVAKESCSEKSUSLHEYGMLLPCGlDKFROVEFVCCPSAEES I., , I,, I,, ::::: :: ::::::::::::::::::::: :::: ,.,,.,,,,.,,, CETHLHYHTVAKETCSEKSTNLHDYGMLLPCOlDKFRGVEFYCCPLAEES
1.3s
ESFDSADAAEDDCDVYYGGADAD”VDRSDDKAVEAOPDEEEEVVEVEEEE ,,,,, .,. ,,,,,,.. ,,... .,,,.. ,,.., ,,. ,,,..,., :: : : ;: :: ,,.,. III,.. DNVDSADAEEDDSOVYWGGADTOYPlDGSEDKVYEV AEEEEVAEVEEEE
239
TODDED DGDEAEEEPEEPYEEATERTTSIATTTTTTTESVEEVVREVC ,,,,, :,:: ::: ,::::::::::::‘::::::::::::::::::: ADOOEDDEDGDE”EEEAEEP”E~A,~~~~S,A~~~~~~,ESVEEVV~E~
287
SEOAETGPCRAMISRWYYDVTESKCAOFIYGGCGGNRNNFESDDYCMAVC .,,,,,,,,.,,,,.,, .,,, I,, ,I.,,,,,,., ~IIIIII.,III,III, II,, .,, : ,I,,,~,,... :::::: SEOAETGPCRAUISRW”FD”TEGKCCPFF”GGCGGNRNNFDTEEYCW*VC
337
KEWEEAEROAKNLPKADKKA”IOHFOEK”ESLEOEAAKOIROOL”E~“~A~ ,,,,.,,,,,,,,, ,,,,,., ,..I ,,,I,,,, ,. I, ,,,,,/,,,,, ,,,,I.,,~,I,.I,I~~IIII ,,,, ,I,,,,, RE~EEAEROAKNLPKACKKA”~OHFOEK”ESLEOEAANEROOL”E,”~A~
437
140 144 IS0 IS4 240 242
193
241
+
origin
201 341
I,
441
HFEHVRM”DPKKAAOIRSO”MTHLRVlNERMNOSFSLLYKYPA”AEEIOD ~,,,,.,I,~,,II~,,I~,,I~~,,~ ,,.,,: ,I,, ,.,,,,,.,, ,.,,,,,,,,, ,I,,,,, ,I,,,,, ,,I, ,,,I ,,,,,,,,, HFEH”RMYDPKKAAOIRSO”MT”LR”IYERMNOSLSLLYN”PAYAEE,OD
537
GEFNlEDLOPVHSFG”DS”PA~~~NS”EPYO*RP*IIDRGL~,~PGSGL~~ I..,,.,,, .~~III/II~~III~II,II..,,,,,,,,,.,, : : : ,,,11,,11 ~~I~~l~I~~III,II,,,,.,/,,,,,,,,,,, GEFSLDDLOPWHSFGADSYPP~N~ENEYEPVDIRPIP~DRL,~
637
IKTEEISEVKHDSEYRHDTAYEVHHOKLVFFAEEVGSNKGAIIGLMVGGV ,.IIIIII.II. : :;; .,II~,,I,,,,, ..,,,,,,,,,,,,,, ,III,~I,..II .,,,,,,,,,I,, ~.,,,,,,,.,,,,,, ,KIEEISE”KYOAEFRHDSG”EVHHOKLVFFAE~”GSNKGA,,GL~VGGV
587
VI*TVIVITLVMLKKKOYTTIHHGVVEVOAAVTPEERHLTKN~~NG~ENP ::::::::::::I::::;: :::::::::::::::::I: I;::;::;:: ~TVIVITLVMLKKKOYTSIHHGVVEVDAAVTPEERHLSKMOONGYENP
737
TYKFFEOMON ,I,IIII,.. ,,,,,,,,.. TYKFFEOMON
1 M 0 G N 22283742463W
54,
.-
+
2SS
f-
185
^-----^
+
84,
SO,
741
747
03
751
m Comparison of Xenopus APP protein sequence (top line) and human APP 75 1 (bottom line). A vertical line indicates amino acid identity, whereas a dot indicates a conservative change. The PID region is underlined. The location of the p-protein region is shown by a double underline. Fig. Northern blotting analysis of the total RNA (30 &lane) from Xenopus embryos of various developmental stages. 1, one-cell stage; M, morula; B, blastula; G, gastrula; N, neurula; 22, stage 22; 28, stage 28; 37, stage 37; 42, stage 42; 46, stage 46; 3W, the tadpole three weeks after fertilization. The size of the transcripts was estimated as 4.2 kb by referring to the positions of 28s and 18s RNA. In the upper blot, the film was exposed for one day, whereas exposure was performed for two days in the lower blot. The transferred RNA was stained with methylene blue to verify that the efficiency of transferring RNA was the same in each lane of nylon membrane.
PID at the amino acid level and 83.3% at the nucleic acid level), but lacks a small exon that is
only included in human APP770.
The predicted protein product is thus thought to be a
Xenopus homologue of the human APP751 (Xenopus APW47). Phylogenic tree analysis by the PILEUP program UWGCG indicates that the Xenopus APP is closer to mammalian APPs than to Drosophila APP. Expression of APP transcripts during Xenoprrs development: The transcription pattern of the APP gene during embryonic and larval development of Xenopus was examined by Northern blot analysis (Fig. 3). The 4.2-kb-long transcript was most abundant at about stage 1564
185
Vol. 189, No. 3, 1992
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
42 and decreased slightly during later larval development (Fig. 3, upper blot).
When the
exposure time was extended to detect the transcript in earlier stages, the APP mRNA was clearly seen at the one-cell stage (Fig. 3, lower blot).
The transcript decreased rapidly during
subsequent embryonic periods up to the gastrula stage, but then started to increase toward the late larval stage. Since the presence of the APP transcript at the one-cell stage indicates its maternal origin. we also examined the transcription
pattern in oocytes by Northern blotting
(Fig. 4).
Accumulation of APP mRNA had already started in young oocytes (lane Oy). They appeared to have much more APP mRNA than fully-grown oocytes (lane 0) when compared on the basis of total RNA, but the relation might be reversed when compared on the basis of each oocyte. Fully grown oocytes (lane 0) had more APP mRNA in their total RNA than one-cell stage embryos (lane 1). Defolliculation of fully-grown oocytes by collagenase did not change the quantity of APP mRNA (lanes Odl and Od2). However, when they are defolliculated further intensively by both collagenase and pronase (lane Od3), APP mRNA decreased considerably. These results suggest that follicular cells have a significant level of APP mRNA.
This was confirmed by staining of the follicular cells with anti-APP antibody (nor
shown). When fully grown oocytes were induced to mature with progesterone, there seemed to be a slight decrease of the APP mRNA level (Om versus Od2), which may also contribute to the difference of APP mRNA content between fully-grown embryos.
oocytes and one-cell stage
In the mouse, the presence of APP transcripts in oocytes and early embryos has
been demonstrated by the PCR technique (10). In conclusion, there are at least two sources for the APP transcript in Xenopus early development: one is of maternal origin and the other is of zygotic origin. From early oogenesis, there is a high accumulation of the maternal transcript. After fertilization, the maternal transcript is degraded to a minimum level at about the gastrula stage, possibly by a post-transcriptional regulation.
A similar change of maternal transcript was reported for the
Xenopus c-myc gene (17). The zygotic APP gene expression appears to start after the gastrula stage.
Developmental change of the splicing pattern of APP mRNA:
We obtained an APP-
homologous cDNA clone that lacked the PID sequence, although the surrounding sequence was identical to that of the Xenopus APP747 cDNA (clone 22). To verify the presence of alternatively spliced APP mRNAs as shown in mammals and possible developmental change of the splicing pattern, PCR was used to amplify the corresponding sequences from the tot.al RNA extracted from oocytes, gastrula embryos and stage 42 tadpoles. The strategy is shown in Fig. 5A(see also MATERIALSANDMETHODS). The tadpole RNA yielded two major bands (lanes 3 and 7 in Fig. 5B and C). The size of the larger band (a and c in Fig. 5B and C) in each lane is just as expected from the transcript containing PID (Xenopur APP747 mRNA). The size of the smaller band (b and d in Fig. SB and C) is the one expected from the transcript lacking PID. This transcript may correspond to a Xenopus homologue of human APP695 mRNA. A possible homologue of human APP770 mRNA, which would give a longer band than those seen in Fig. 5B and C, could not be
1565
Vol. 189, No. 3, 1992
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
B
+
f-
12345MC6789M
c-origin
origin
C
+--origin
a+
--c+d +I
28s
+ ias 04Fig. 4. Northern blotting analysis of the total RNA05 (30b-,&lane)
from Xenopu,s one-cell stage embrvo and oocvtes. 1. one-cell stage: 0. untreated full-grown oocvte: Odl. full-mown oocy& defollicuiated by ‘treating with”cbllagenase for one “hour; Od2: full-grown Gocyte defolliculated with collagenase overnight; Od3, full-grown oocyte defolliculated with collagenase overnight and then pronase for 3 hours; Om, matured oocyte obtained by treating defolliculated full-grown oocyte (Od2) with progesterone; Oy, young oocytes (oocyte stage I4) defolliculated with collagenase overnight. m PCR analysis of the splicing pattern of APP mRNA. In A, the strategy is shown schematicallv. When the mRNA contains PID, products a (383 bp) and c (673 bp) will be produced in ihe presence of a pair of sense and at&sense primers (51 and 106) and of a pair of sense and antisense primers (11 and 106), respectively. When the mRNA does not contain PID, products b (215 bp) and d (505 bp) will be produced in the presence of primers 51 and 106 and of primers 11 and 106, respectively (see MATERIALS AND METHODS). In B and C, PCR products with primers 51 and 106 were electrophoresed from lanes l-5. PCR products with primers 11 and 106 were electrophoresed from lanes 6-9. Lane M: DNA size marker, pGEM DNA Markers of Promega (2645, 1605, 1198,676,517,460, 396, 350, 222, and 179). Lane 1: PCR product from the plasmid containing clone KB2, which has PID; this is a positive control for the mRNA containing PID. Lanes 2 and 6: the PCR product from the plasmid containing clone 22, which does-not contain PID (see text); this is-a positive controi for the mRNA without PID. Lane C: PCR product from water with primers 11 and 106; this is the negative control without the template DNA. Lanes 3 and 7: total RNA of stage-42 tadpoles. Lanes 4 and 8: total RNA of gastrula embryos. Lanes 5 and 9; total RNA of full-grown oocytes. In C, 3-fold higher amounts of the PCR product were loaded in lanes 4, 5, 8 and 9.
detected. The origin of the faint band between the two major bands (x in lane 7) is not clear at the moment. A similar analysis was done on the total RNA from gastrula embryos (lanes 4 and 8 in Fig. 5B) and oocytes (lanes 5 and 9). PID was clearly
In both cases, the PCR product from the transcript lacking
seen, but the PCR product
from the transcript
with PID could hardly
be
detected. The latter was observed, however, when a 3-times larger amount of the PCR product from gastrula or oocytes was electrophoresed (Fig. 5C). However, the PCR product with the PID from gastrula or oocytes was still much less than that from tadpoles under this condition. It should be noted that the total amount of PCR product for tadpoles loaded on the gel was the same in Fig. 5B and 5C. We may conclude that the ratio of the transcript with PID 1566
Vol:189,
No.
3, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
against that without PID is extremely low for the transcript of maternal origin as compared to the ratio for the transcript of zygotic origin.
An essentially similar change in the splicing pattern of APP mRNA was observed in mouse early development (10). In most of the
mammalian tissues, the transcript with PID is dominant with the exception of the central nervous system (7, 18), in which the transcript without PID is dominant, and this type of transcript is found principally in neurons (18). It is interesting that the splicing pattern of Xenopus or mouse maternal APP mRNA is similar to the rather exceptional splicing pattern in neurons. Our results show that the APP gene is expressed in Xenopus early development. In Drosophila, the APP homologue appears to be required for embryonic neural development (11, 12). A similar role of the APP molecule may be shared with Xenopus and mammals, as inferred from the high degree conservation of the amino acid sequence of the APP molecule among these species.
The Xenopus system described here would be helpful for further
understanding of the normal function of APP during vertebrate development.
ACKNOWLEDGMENTS
We thank Drs. T. Ishida and
Y. Takahashi (Asahi Chemical Industry Co.) for their gifts of cDNA clones of the human APP. We appreciate the help of Mr. David Kramer for checking our manuscript and giving useful suggestions. Special thanks to Professor Kunitaro Takahashi for his constant encouragement and valuable suggestions during the present experiments. We also thank Dr. Yasushi Okamura for the protocol for PCR. This work was supported by the Japan Society for the Promotion of Science, a grant to H. Okado, and a Grant-in-Aid from the Japanese Ministry of Education, Science and Culture, Japan to Professor K. Takahashi.
REFERENCES
1. Glenner, G.G., and Wong, C.W. (1984) Biochem. Biophys. Res. Commun. 120,885890. 2. Kang, J., Lemire, H-G, Unterbeck, A., Salbaum, J.M., Masters, C.L., Grzeschik, K-H, Multhaup, G., Beyreuther, K. and Muller-Hill, B. (1987) Nature 325,733-736. 3. Ponte, P., Gonzalez-DeWhitt, P., Schilling, J., Miller, J., Hsu, D., Greenberg, B., Davis, K., Wallace, W, Lieberburg, I., Fuller, F. and Cordel, B. (1988) Nature 331,525-527. 4. Tanzi, R.E., McClatchey, A.I., Lamperti, E.D., Villa-Komaroff, L., Gusella, J.F. and Neve, R.L. (1988) Nature 331,528-530. 5. Kitaguchi, N. Takahashi, Y., Tokushima, Y, Shiojiri, S. and Ito, H. (1988) Nature 331, 530-532. 6. Yamada, T., Sasaki, H, Furuya, H, Miyata, T, Goto, I and Sakaki, Y. (1987) Biochem. Biophys. Res. Commun. 149,665-671. 7. Yamada, T., Sasaki, H., Dohura, K., Goto, I and Sakaki. Y. (1989) B&hem. Biophys. Res. Commun. 158,906-912. 8. Kang, J. and Mtiller-Hill, B. (1989) Nucleic Acids Res. 17,213O. 9. Shivers, B.D., Hilbich, C., Multhaup, G., Salbaum, M., Beyreuther K. and Seeburg, P.H. (1988) EMBO J. 7, 1365-1370. 10. Fisher, S., Gearhart, J.D. and Oster-Granite, M.L. (1991) Proc. Natl. Acad. Sci. USA 88, 1779-1782. 11. Rosen, D.R. Martin-Morris, L. Luo, L. and White K. (1989) Proc. Natl. Acad. USA 86, 2478-2482. 12. Martin-Morris,
L.E. and White, K. (1990) Development 110, 185-195. 1567
Vol. 189, No. 3, 1992
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
13. Slack, J.M.W. (1991) In Early Development. Second ed. Cambridge University Press. 14. Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Rutter, W.J. (1979) Biochemistry 18, 5294-5299. 15. Sambrook,J., Fritsch, E.F. and Maniatis, T. (1989) In Molecular cloning, vol. 1 (second edition), New York: Cold Spring Harbour Laboratory Press. 16. Herrin, D.L. and Schmidt, G.W. (1988) Bio-Techiques 6, 196-200. 17. Taylor, M.V., Gusse, M., Evan, G.I., Dathan, N. and Mechali, M. (1986) EMBO J. 5, 3563-3570. 18. Neve, R.L., Finch, E.A. and Dawes, L.R. (1988) Neuron 1,669-677.
1568
BIOCHEMICAL
1992
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Pages 1561-1568
30, 1992
A XENOPUS PROTEIN:
HOMOLOGUE
DEVELOPMENTAL
OF THE HUMAN REGULATION
P-AMYLOID
PRECURSOR
OF ITS GENE EXPRESSION
Haruo Okado and Harumasa Okamoto Department of Neurobiology, Institute for Brain Research, Faculty of Medicine, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan Received
November
25,
1992
SUMMARY: Complimentary DNA clones have been isolated from Xenopus larva to delineate a protein highly homologous to the human b-amyloid precursor protein (APP). Developmental change of Xenopus APP gene expression has been analyzed with molecular probes. From early oogenesis, there is a high accumulation of maternal APP. After fertilization, the mRNA is degraded, reaching a minimum level around the gastrula stage. Then zygotic transcription appears to be initiated, and this continues during the subsequent embryonic and larval stages. Splicing patterns differ between the maternal and zygotic transcripts. The ratio of mRNA including the protease inhibitor domain (PID) sequence is extremely low for the transcript of maternal origin as compared to that for the transcript of zygotic origin. These results suggest some roles for the APP molecule in Xenopus early 0 1992 Academic Press, Inc. development.
The p-amyloid precursor protein (APP) gives rise to a 4.2~kDa peptide that is deposited in extracellular dense-core amyloid plaque of Alzheimer’s disease (1,2). Molecular cloning studies have revealed that there are at least three types of APP mRNA produced by alternative splicing, each encoding 695,75 1 and 770 amino acid residues (APP695, APP751 and APP770) in human (2, 3-5), mouse (6,7) and rat (8,9). The sequence of the protease inhibitor domain (PID) is included in the latter two types (3-5). Since human APP transcripts are abundant in fetal tissues as well as in adults (4, 5) and mouse APP transcripts can be detected in oocytes and early embryos by the PCR technique (lo), a developmental role for APP is highly likely, Recently an APP homologue of but the data so far obtained seem rather fragmented. Drosophila
was identified and suggested to be required for embryonic neural development (11,
12). In the present experiments, we have isolated cDNA clones for the APP homologue from Xenopus as an initial step to provide insight into the normal function of APP in vertebrate development. Due to the ease of experimental manipulation, this species has already been used to obtain a considerable amount of data on developmental processes, including the neural one (13), at the cellular and molecular levels We report here that the APP homologue of Xenopus is highly conservative to the mammalian APP and that the Xenopus APP transcript exists from early oogenesis, remaining throughout the embryonic and larval stages, although its relative amount and splicing pattern changes during development. 0006-291X/92
1561
$4.00
Copyright 0 1992 bJ Academk Press. Inc. All rights of reproduction in any jimn reserved.
Vol. 189, No. 3, 1992
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
MATERIALS
AND METHODS
Total RNA was extracted from Xenopus Construction of two sets of cDNA libraries: tadpoles with guanidinium thiocyanate, followed by centrifugation in CsCl solution (14). Poly(A)+RNA was purified by an oligo(dT)-cellulose spun column (mRNA Purification kit, Pharmacia) and employed for construction of a unidirectional cDNA library using a ZAPcDNA synthesis kit (Stratagene) with oligo(dT)-linker primers. A random primed cDNA library was also made from poly(A)+RNA using a You-prime cDNA synthesis kit (Pharmacia). The amplified unidirectional cDNA library (7X 1O5 Screening of the cDNA libraries: plaques) was screened with human APP cDNA fragment probes, KX and XB, which were kindly provided by Drs. T. Ishida and Y. Takahashi (Asahi Chemical Industry Co.). The plaque-transferred filters (Hybond-N+; Amersham) were hybridized with the two radiolabelled probes for 2 days at 37’C in 5x SSPE, 0.5% SDS, 5x Denhardt’s solution, 35% formamide and 20 pg/ml sperm DNA. After this, the filters were washed twice for 10 min. at room temperature in 2x SSPE and 0.1% SDS, followed by washing with the same solution for 15 min. at 50°C and then exposed to Kodak XAR film for about one day at -7O’C. The original random primed library (5~10~ plaques) was screened with a 191-bp fragment of the Xenopus APP homologue to obtain the 5’portion of the coding sequence. This fragment was constructed by PCR with the following degenerate primers, the sequences of which are highly conserved between the human APP and its Drosophila homologue at their 5’regions: S>primers: S>ATATGGATCC(AT)(CG)IGA(TC)GCI(TC)TI(TC)TIGT(ATCG)CCDrimers: 5>CGTAGAATTCGG(AG)CA(AG)CA(ATCG)AC(AG)AA(TC)TC(ATCG)AC< 3, in which GAATTC is the EcoRI site and CGTA is a nonsense sequence. These primers were applied to single strand cDNA made from the poly(A)+RNA fraction with a random hexamer. The template was then amplified under the following conditions: 94°C (1 min.), 37’C (1 min.), transition time (2min.), 72’C (2 min.) in the first 5 cycles and then 94’C (1 min.), 50°C (1 min.), 72’C (2 min.) in 30 cycles. The plaque-transferred nylon filters were hybridized with the radiolabelled 191-bp probe for one day at 42’C in 5x SSPE, 0.5% SDS, 5x Denhardt’s solution, 50% formamide and 20 &ml sperm DNA. They were washed twice at room temperature for 10 min. and once at 42’C for 15 min. in 2x SSPE and 0.1% SDS. Nucleotide sequences: DNA sequencing was performed by the dideoxynucleotide chain termination method with Sequenase Version 2 (United States Biochemical Corp.). Northern blotting analysis: Total RNA was extracted from oocytes, embryos and larvae at various developmental stages by the proteinase K method (15). They were electrophoresed through a 1% agarose-formaldehyde gel (30 @ lane), blotted to Hybond-N (Amersham), fixed by exposure to UV light, stained with methylene blue (16) and then hybridized at 43’C in 50% formamide, 5x SSPE, 5x Denhardt’s solution and 0.5% SDS. The probe applied was a cDNA clone of the Xenopus APP homologue (KB2, see text), which was labelled by a Multiprime Labeling System (Amersham). The filter membrane was washed in 0.1% SDS, 0. lx SSPE at 65*C. PCR for determining the splicing pattern: Two sense primers were used: ATCTGGATCC TGCAGTGAGAAGAGCATGAG (designated as 1 l), in which GGATCC is the BarnHI site and ATCT is a nonsense sequence, and CTTTGGATCCCGAGGAACCCTATGAAGAAG (designated as 51), in which GGATCC is the BamHI site and CITT is a nonsense sequence. One antisense primer was used: CCTCTCCTTTGCTTTCA (designated as 106). The two pairs of sense and antisense primers were applied to single strand cDNA made from the total RNA using random hexamers. The template was amplified at 94’C (1 min.), 55’C (1 min.) and 72°C (1 min.) for 30 cycles (see the Fig. 5 legend for further details).
RESULTS AND DISCUSSION Molecular cloning of Xenopus APP cDNA: The unidirectional cDNA library generated from Xenopus tadpole was screened with two radiolabelled human APP cDNA probes, KX and XB. The longest clone (KB2, 2kb) out of three positives was still short of the 5’portion of the coding region when compared with the published human APP cDNA sequence. To isolate the 1562
Vol.
189,
No.
BIOCHEMICAL
3, 1992
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
5’ portion, the random-primed cDNA library was screened with a 191-bp fragment of the Xenopus
APP homologue in this region, which was made by the PCR technique (see
By sequencing some of the nine positive clones obtained the total coding region was determined, Figure 1 shows the delineated nucleotide sequence of the Xenopus APP homologue that includes a coding region of 2241 bp, corresponding to 747 amino acid residues; a 5’ untranslated region of 33 bp; and a 3’untranslated region of 720 bp. When the deduced amino acid sequence is compared with that of human APP7.51 (Fig. 2). MATERIALS
AND METHODS).
they share a high degree of homology (87.5%). human APP is 78.3%.
Fig. 1. Xenopus code).
At the nucleic acid level, the homology to
The Xenopus sequence contains PID (84.2% homologous to human
APP nucleotide
sequence and deduced amino acid sequence (single-letter
1563
Vol. 189, No. 3, 1992
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
42
YLP”KLARADCkRLALEYLVDG~GQLLAEPOI*UFSVARLN~ I :; :::: .:::::::::::::::: ::::: : : : .ALEVPTDONAGLLAEPOIAMFc.o~~N~ I YLPGLALLLLAAYTAR 43 44
43 ss
H.HVONGKIiTDVSG. .ClGTKEGlLOYiOEV”PELOliNVVEANOPV~ :: I:::::::::::::I:::::::::::::: :::::::::: :.:: HUNVONOKYDSDPSOTKTClOTKEOlLOYCOEVYPELPITNVVEANOPVT
9s
SO lONWCKKORKOCKSRTHlVVP”flCLVGEFVSDALLVPDK~KF~HOE~ND~ :::::: ::I::: ;.; :;;::I::I:;:;;III;:::;I:::;;:.
130
94
IONWCKRQRKOCKTHPHFVIPYRCLVGEFVSDALLVPDKCKFLHOERYDV
143
CETHLHYHTVAKESCSEKSUSLHEYGMLLPCGlDKFROVEFVCCPSAEES I., , I,, I,, ::::: :: ::::::::::::::::::::: :::: ,.,,.,,,,.,,, CETHLHYHTVAKETCSEKSTNLHDYGMLLPCOlDKFRGVEFYCCPLAEES
1.3s
ESFDSADAAEDDCDVYYGGADAD”VDRSDDKAVEAOPDEEEEVVEVEEEE ,,,,, .,. ,,,,,,.. ,,... .,,,.. ,,.., ,,. ,,,..,., :: : : ;: :: ,,.,. III,.. DNVDSADAEEDDSOVYWGGADTOYPlDGSEDKVYEV AEEEEVAEVEEEE
239
TODDED DGDEAEEEPEEPYEEATERTTSIATTTTTTTESVEEVVREVC ,,,,, :,:: ::: ,::::::::::::‘::::::::::::::::::: ADOOEDDEDGDE”EEEAEEP”E~A,~~~~S,A~~~~~~,ESVEEVV~E~
287
SEOAETGPCRAMISRWYYDVTESKCAOFIYGGCGGNRNNFESDDYCMAVC .,,,,,,,,.,,,,.,, .,,, I,, ,I.,,,,,,., ~IIIIII.,III,III, II,, .,, : ,I,,,~,,... :::::: SEOAETGPCRAUISRW”FD”TEGKCCPFF”GGCGGNRNNFDTEEYCW*VC
337
KEWEEAEROAKNLPKADKKA”IOHFOEK”ESLEOEAAKOIROOL”E~“~A~ ,,,,.,,,,,,,,, ,,,,,., ,..I ,,,I,,,, ,. I, ,,,,,/,,,,, ,,,,I.,,~,I,.I,I~~IIII ,,,, ,I,,,,, RE~EEAEROAKNLPKACKKA”~OHFOEK”ESLEOEAANEROOL”E,”~A~
437
140 144 IS0 IS4 240 242
193
241
+
origin
201 341
I,
441
HFEHVRM”DPKKAAOIRSO”MTHLRVlNERMNOSFSLLYKYPA”AEEIOD ~,,,,.,I,~,,II~,,I~,,I~~,,~ ,,.,,: ,I,, ,.,,,,,.,, ,.,,,,,,,,, ,I,,,,, ,I,,,,, ,,I, ,,,I ,,,,,,,,, HFEH”RMYDPKKAAOIRSO”MT”LR”IYERMNOSLSLLYN”PAYAEE,OD
537
GEFNlEDLOPVHSFG”DS”PA~~~NS”EPYO*RP*IIDRGL~,~PGSGL~~ I..,,.,,, .~~III/II~~III~II,II..,,,,,,,,,.,, : : : ,,,11,,11 ~~I~~l~I~~III,II,,,,.,/,,,,,,,,,,, GEFSLDDLOPWHSFGADSYPP~N~ENEYEPVDIRPIP~DRL,~
637
IKTEEISEVKHDSEYRHDTAYEVHHOKLVFFAEEVGSNKGAIIGLMVGGV ,.IIIIII.II. : :;; .,II~,,I,,,,, ..,,,,,,,,,,,,,, ,III,~I,..II .,,,,,,,,,I,, ~.,,,,,,,.,,,,,, ,KIEEISE”KYOAEFRHDSG”EVHHOKLVFFAE~”GSNKGA,,GL~VGGV
587
VI*TVIVITLVMLKKKOYTTIHHGVVEVOAAVTPEERHLTKN~~NG~ENP ::::::::::::I::::;: :::::::::::::::::I: I;::;::;:: ~TVIVITLVMLKKKOYTSIHHGVVEVDAAVTPEERHLSKMOONGYENP
737
TYKFFEOMON ,I,IIII,.. ,,,,,,,,.. TYKFFEOMON
1 M 0 G N 22283742463W
54,
.-
+
2SS
f-
185
^-----^
+
84,
SO,
741
747
03
751
m Comparison of Xenopus APP protein sequence (top line) and human APP 75 1 (bottom line). A vertical line indicates amino acid identity, whereas a dot indicates a conservative change. The PID region is underlined. The location of the p-protein region is shown by a double underline. Fig. Northern blotting analysis of the total RNA (30 &lane) from Xenopus embryos of various developmental stages. 1, one-cell stage; M, morula; B, blastula; G, gastrula; N, neurula; 22, stage 22; 28, stage 28; 37, stage 37; 42, stage 42; 46, stage 46; 3W, the tadpole three weeks after fertilization. The size of the transcripts was estimated as 4.2 kb by referring to the positions of 28s and 18s RNA. In the upper blot, the film was exposed for one day, whereas exposure was performed for two days in the lower blot. The transferred RNA was stained with methylene blue to verify that the efficiency of transferring RNA was the same in each lane of nylon membrane.
PID at the amino acid level and 83.3% at the nucleic acid level), but lacks a small exon that is
only included in human APP770.
The predicted protein product is thus thought to be a
Xenopus homologue of the human APP751 (Xenopus APW47). Phylogenic tree analysis by the PILEUP program UWGCG indicates that the Xenopus APP is closer to mammalian APPs than to Drosophila APP. Expression of APP transcripts during Xenoprrs development: The transcription pattern of the APP gene during embryonic and larval development of Xenopus was examined by Northern blot analysis (Fig. 3). The 4.2-kb-long transcript was most abundant at about stage 1564
185
Vol. 189, No. 3, 1992
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
42 and decreased slightly during later larval development (Fig. 3, upper blot).
When the
exposure time was extended to detect the transcript in earlier stages, the APP mRNA was clearly seen at the one-cell stage (Fig. 3, lower blot).
The transcript decreased rapidly during
subsequent embryonic periods up to the gastrula stage, but then started to increase toward the late larval stage. Since the presence of the APP transcript at the one-cell stage indicates its maternal origin. we also examined the transcription
pattern in oocytes by Northern blotting
(Fig. 4).
Accumulation of APP mRNA had already started in young oocytes (lane Oy). They appeared to have much more APP mRNA than fully-grown oocytes (lane 0) when compared on the basis of total RNA, but the relation might be reversed when compared on the basis of each oocyte. Fully grown oocytes (lane 0) had more APP mRNA in their total RNA than one-cell stage embryos (lane 1). Defolliculation of fully-grown oocytes by collagenase did not change the quantity of APP mRNA (lanes Odl and Od2). However, when they are defolliculated further intensively by both collagenase and pronase (lane Od3), APP mRNA decreased considerably. These results suggest that follicular cells have a significant level of APP mRNA.
This was confirmed by staining of the follicular cells with anti-APP antibody (nor
shown). When fully grown oocytes were induced to mature with progesterone, there seemed to be a slight decrease of the APP mRNA level (Om versus Od2), which may also contribute to the difference of APP mRNA content between fully-grown embryos.
oocytes and one-cell stage
In the mouse, the presence of APP transcripts in oocytes and early embryos has
been demonstrated by the PCR technique (10). In conclusion, there are at least two sources for the APP transcript in Xenopus early development: one is of maternal origin and the other is of zygotic origin. From early oogenesis, there is a high accumulation of the maternal transcript. After fertilization, the maternal transcript is degraded to a minimum level at about the gastrula stage, possibly by a post-transcriptional regulation.
A similar change of maternal transcript was reported for the
Xenopus c-myc gene (17). The zygotic APP gene expression appears to start after the gastrula stage.
Developmental change of the splicing pattern of APP mRNA:
We obtained an APP-
homologous cDNA clone that lacked the PID sequence, although the surrounding sequence was identical to that of the Xenopus APP747 cDNA (clone 22). To verify the presence of alternatively spliced APP mRNAs as shown in mammals and possible developmental change of the splicing pattern, PCR was used to amplify the corresponding sequences from the tot.al RNA extracted from oocytes, gastrula embryos and stage 42 tadpoles. The strategy is shown in Fig. 5A(see also MATERIALSANDMETHODS). The tadpole RNA yielded two major bands (lanes 3 and 7 in Fig. 5B and C). The size of the larger band (a and c in Fig. 5B and C) in each lane is just as expected from the transcript containing PID (Xenopur APP747 mRNA). The size of the smaller band (b and d in Fig. SB and C) is the one expected from the transcript lacking PID. This transcript may correspond to a Xenopus homologue of human APP695 mRNA. A possible homologue of human APP770 mRNA, which would give a longer band than those seen in Fig. 5B and C, could not be
1565
Vol. 189, No. 3, 1992
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
B
+
f-
12345MC6789M
c-origin
origin
C
+--origin
a+
--c+d +I
28s
+ ias 04Fig. 4. Northern blotting analysis of the total RNA05 (30b-,&lane)
from Xenopu,s one-cell stage embrvo and oocvtes. 1. one-cell stage: 0. untreated full-grown oocvte: Odl. full-mown oocy& defollicuiated by ‘treating with”cbllagenase for one “hour; Od2: full-grown Gocyte defolliculated with collagenase overnight; Od3, full-grown oocyte defolliculated with collagenase overnight and then pronase for 3 hours; Om, matured oocyte obtained by treating defolliculated full-grown oocyte (Od2) with progesterone; Oy, young oocytes (oocyte stage I4) defolliculated with collagenase overnight. m PCR analysis of the splicing pattern of APP mRNA. In A, the strategy is shown schematicallv. When the mRNA contains PID, products a (383 bp) and c (673 bp) will be produced in ihe presence of a pair of sense and at&sense primers (51 and 106) and of a pair of sense and antisense primers (11 and 106), respectively. When the mRNA does not contain PID, products b (215 bp) and d (505 bp) will be produced in the presence of primers 51 and 106 and of primers 11 and 106, respectively (see MATERIALS AND METHODS). In B and C, PCR products with primers 51 and 106 were electrophoresed from lanes l-5. PCR products with primers 11 and 106 were electrophoresed from lanes 6-9. Lane M: DNA size marker, pGEM DNA Markers of Promega (2645, 1605, 1198,676,517,460, 396, 350, 222, and 179). Lane 1: PCR product from the plasmid containing clone KB2, which has PID; this is a positive control for the mRNA containing PID. Lanes 2 and 6: the PCR product from the plasmid containing clone 22, which does-not contain PID (see text); this is-a positive controi for the mRNA without PID. Lane C: PCR product from water with primers 11 and 106; this is the negative control without the template DNA. Lanes 3 and 7: total RNA of stage-42 tadpoles. Lanes 4 and 8: total RNA of gastrula embryos. Lanes 5 and 9; total RNA of full-grown oocytes. In C, 3-fold higher amounts of the PCR product were loaded in lanes 4, 5, 8 and 9.
detected. The origin of the faint band between the two major bands (x in lane 7) is not clear at the moment. A similar analysis was done on the total RNA from gastrula embryos (lanes 4 and 8 in Fig. 5B) and oocytes (lanes 5 and 9). PID was clearly
In both cases, the PCR product from the transcript lacking
seen, but the PCR product
from the transcript
with PID could hardly
be
detected. The latter was observed, however, when a 3-times larger amount of the PCR product from gastrula or oocytes was electrophoresed (Fig. 5C). However, the PCR product with the PID from gastrula or oocytes was still much less than that from tadpoles under this condition. It should be noted that the total amount of PCR product for tadpoles loaded on the gel was the same in Fig. 5B and 5C. We may conclude that the ratio of the transcript with PID 1566
Vol:189,
No.
3, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
against that without PID is extremely low for the transcript of maternal origin as compared to the ratio for the transcript of zygotic origin.
An essentially similar change in the splicing pattern of APP mRNA was observed in mouse early development (10). In most of the
mammalian tissues, the transcript with PID is dominant with the exception of the central nervous system (7, 18), in which the transcript without PID is dominant, and this type of transcript is found principally in neurons (18). It is interesting that the splicing pattern of Xenopus or mouse maternal APP mRNA is similar to the rather exceptional splicing pattern in neurons. Our results show that the APP gene is expressed in Xenopus early development. In Drosophila, the APP homologue appears to be required for embryonic neural development (11, 12). A similar role of the APP molecule may be shared with Xenopus and mammals, as inferred from the high degree conservation of the amino acid sequence of the APP molecule among these species.
The Xenopus system described here would be helpful for further
understanding of the normal function of APP during vertebrate development.
ACKNOWLEDGMENTS
We thank Drs. T. Ishida and
Y. Takahashi (Asahi Chemical Industry Co.) for their gifts of cDNA clones of the human APP. We appreciate the help of Mr. David Kramer for checking our manuscript and giving useful suggestions. Special thanks to Professor Kunitaro Takahashi for his constant encouragement and valuable suggestions during the present experiments. We also thank Dr. Yasushi Okamura for the protocol for PCR. This work was supported by the Japan Society for the Promotion of Science, a grant to H. Okado, and a Grant-in-Aid from the Japanese Ministry of Education, Science and Culture, Japan to Professor K. Takahashi.
REFERENCES
1. Glenner, G.G., and Wong, C.W. (1984) Biochem. Biophys. Res. Commun. 120,885890. 2. Kang, J., Lemire, H-G, Unterbeck, A., Salbaum, J.M., Masters, C.L., Grzeschik, K-H, Multhaup, G., Beyreuther, K. and Muller-Hill, B. (1987) Nature 325,733-736. 3. Ponte, P., Gonzalez-DeWhitt, P., Schilling, J., Miller, J., Hsu, D., Greenberg, B., Davis, K., Wallace, W, Lieberburg, I., Fuller, F. and Cordel, B. (1988) Nature 331,525-527. 4. Tanzi, R.E., McClatchey, A.I., Lamperti, E.D., Villa-Komaroff, L., Gusella, J.F. and Neve, R.L. (1988) Nature 331,528-530. 5. Kitaguchi, N. Takahashi, Y., Tokushima, Y, Shiojiri, S. and Ito, H. (1988) Nature 331, 530-532. 6. Yamada, T., Sasaki, H, Furuya, H, Miyata, T, Goto, I and Sakaki, Y. (1987) Biochem. Biophys. Res. Commun. 149,665-671. 7. Yamada, T., Sasaki, H., Dohura, K., Goto, I and Sakaki. Y. (1989) B&hem. Biophys. Res. Commun. 158,906-912. 8. Kang, J. and Mtiller-Hill, B. (1989) Nucleic Acids Res. 17,213O. 9. Shivers, B.D., Hilbich, C., Multhaup, G., Salbaum, M., Beyreuther K. and Seeburg, P.H. (1988) EMBO J. 7, 1365-1370. 10. Fisher, S., Gearhart, J.D. and Oster-Granite, M.L. (1991) Proc. Natl. Acad. Sci. USA 88, 1779-1782. 11. Rosen, D.R. Martin-Morris, L. Luo, L. and White K. (1989) Proc. Natl. Acad. USA 86, 2478-2482. 12. Martin-Morris,
L.E. and White, K. (1990) Development 110, 185-195. 1567
Vol. 189, No. 3, 1992
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
13. Slack, J.M.W. (1991) In Early Development. Second ed. Cambridge University Press. 14. Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Rutter, W.J. (1979) Biochemistry 18, 5294-5299. 15. Sambrook,J., Fritsch, E.F. and Maniatis, T. (1989) In Molecular cloning, vol. 1 (second edition), New York: Cold Spring Harbour Laboratory Press. 16. Herrin, D.L. and Schmidt, G.W. (1988) Bio-Techiques 6, 196-200. 17. Taylor, M.V., Gusse, M., Evan, G.I., Dathan, N. and Mechali, M. (1986) EMBO J. 5, 3563-3570. 18. Neve, R.L., Finch, E.A. and Dawes, L.R. (1988) Neuron 1,669-677.
1568
