Analysis of amelogenin mRNA during bovine tooth development

~ Pergamon Archs oral Biol. Vol. 41, No. 2, pp. 205-213, 1996 Copyright © 1996. Elsevier Science Ltd. All rights reserved

Printed in Great Britain 0003-9969(95)00119-0 0003-9969/96 $15.00 + 0.00

ANALYSIS OF AMELOGENIN mRNA DURING BOVINE TOOTH DEVELOPMENT

Z. A. YUAN, P. M. COLLIER, J. ROSENBLOOM and C. W. GIBSON* Department of Anatomy, University of Pennsylvania School of Dental Medicine, 4001 Spruce Street,

Philadelphia, PA 19104, U.S.A.

(Accepted 10 September 1995)

Summary--Th,~ amelogenins are highly conserved enamel-matrix proteins that are essential for proper mineral forrnalion. Transcriptionally active genes encoding the bovine amelogenin proteins reside on both the X and Y chromosomes. Comparison of relative levels of amelogenin mRNAs at various stages of development indicated that the X-chromosomal amelogenin message is at least six fold more abundant than the Y. Alternative splicing generates at least seven messages, five from the X primary transcript, and two from the Y. The two most abundant X-chromosomal amelogenin messages are approx. 850 and 450 nucleotides long, and nearly 10-fold more 850-nucleotide mRNA can be measured than 450 nucleotide, which has lost :most ofexon 6 by splicing. The predominant small message encodes leucine-rich amelogenin protein (LRAP), and amounts of LRAP message are relatively constant during development. However, the amelogeni:a message from which exon 3 has been spliced declines approximately 2.3-fold, when compared to total X chromosomal amelogenin transcripts, suggesting differential regulation of alternative splicing. In addition, a new exon was identified within genomic DNA, which was shown to be expressed by the use of reverse transcriptase-polymerase chain reaction, and the exons were renamed accordingly. This new exon-4 sequence is unusual in that it is not highly conserved between species.

Key words: arnelogenin, gene expression, tooth development, bovine.

INTRODUCTION

Dental enamel, the highly mineralized tissue that covers the crown of the tooth, is secreted by cells called ameloblasts. The amelogenin proteins comprise about 90% of the extracellular matrix secreted by these cells, and the enamel organic matrix begins to mineralize shortly after secretion begins. The primary structure of the antelogenin proteins is conserved across species and these proteins have an essential role in normal ename,1 mineral formation. The import- ance of the amelogenin proteins for enamel integrity was verified in several families with different mutations in the X-chromosomal amelogenin gene. The diagnosis in each family member with the mutation was the inherited enamel defect amelogenesis imperfecta, in which the enamel appeared either hypoplastic or hypomineralized (Lagerstrom et al., 1991; Aldred et al., 1992; Lench, Brook and Winter, 1994; Lagerstrom-Fermer et al., 1995).

It is thought that amelogenin proteins may control the direction or rate of mineral crystal growth, through binding to developing hydroxyapatite crystals, although the exact role they perform is not well understood (reviewed by Deutsch, 1989; Brookes

*To whom all correspondence should be addressed. Abbreviations: LRAP, leucine-rich amelogenin protein; RT-

PCR, reverse transcriptase-polymerase chain reaction.

et al., 1995; Simmer and Fincham, 1995). Ameiogenins aggregate, and are relatively insoluble, at physiological pH, and they are subject to proteolytic breakdown as mineral crystals increase in size and number (Fincham et al., 1991a; Limeback and Simic, 1990; Robinson et al., 1990). Both protein aggregation and degradation are thought to be involved in the controlled orchestration of enamel crystal growth and maturation (Brookes et al., 1995; Limeback, 1991; Fincham et al,, 1994; Diekwisch et al., 1993).

Proteolytic degradation results in size heterogeneity of enamel matrix proteins, and the average size decreases during development as enamel matures to become the hardest tissue in the body. It is thought that the ameloblasts remove amelogenin peptides from the maturing matrix by secreting enzymes responsible for proteolytic breakdown (DenBesten and Heffernan, 1989; Overall and Limeback, 1988; Shimizu, Tanabe and Fukae, 1979). While numerous amelogenin peptides result from protein processing, extensive alternative splicing of the primary amelogenin transcript, which is a characteristic of all species examined thus far, may also contribute as many as seven discrete primary translation products (Yamakoshi et al., 1994; Lau et al., 1992; Simmer et al., 1994; Gibson et al., 1992). In some species, such as human and bovine, an additional active amelogenin gene resides on the Y chromosome, and Y-specific transcripts are also alternatively spliced (Salido et al., 1992;

205

206 Z.A. Yuan et al.

Gibson et al., 1995). When this complex situation is analysed at the level of the mRNA, that heterogeneity attributable to proteolysis can be disregarded. The bovine tooth was chosen as the model system because both X- and Y-chromosomal amelogenin genes are active in human and bovine species, while in the mouse there is an amelogenin gene only on the X chromosome (Lau et al., 1989). The bovine X- and Y-chromosomal genes have been cloned and sequences determined, and comparisons have revealed approx.

1.2% formaidehyde-agarose gel and transferred to a nitrocellulose membrane, which was cross-linked by ultraviolet light (Stratagene Cloning Systems, La Jolla, CA). Oligomer probes were radioactively end-labelled using T4 polynucleotide kinase, and hybridized in 0.9 M NaC1, 90mM sodium citrate, 20mM NaPO4 pH 6.5, 0.1% sodium dodecyl sul- phate, 5 × Denhardt's solution, and 0.5 mg/ml denatured salmon sperm DNA, at 43°C. Amelogenin oligonucleotide probe sequences were:

No. 663 < T G G G G G G C A G A G G C G G G T G T C T T A T C for the LRAP splice junction, from exon 5 to the internal site in exon 6;

No. 664 < G A G G G G T G A G C A C A G G C A T A G A G A A G for the exon 2-5 splice junction;

No. 707 < TCTCTTCTCATTTTCTGATCTTTTA for X-chromosomal amelogenin;

No. 708 < GGTTTTGTTCTCTCATTTTCTTTTA for Y-chromosomal amelogenin;

in which < indicates the antisense sequence.

83% homology at the amino acid level (Gibson et al., 1992). Sequence information has been reported for X and Y specific amelogenin cDNAs (Salido et al., 1992; Gibson et al., 1991a, 1992), and preliminary evidence exists for Y-specific amelogenin translation products (Fincham et al., 1991; Gibson et al., 1995). The present study sought to analyse how amounts of various amelogenin transcripts vary in relation to each other as suggested for mouse (Couwenhoven and Snead, 1994), as it is probable that all primary translation products will be shown to have distinct roles in enamel crystal formation.

MATERIALS AND METHODS

R N A isolation

Bovine fetuses were measured from crown to rump for age approximation (Eckstein et al., 1977), and unerupted mandibular molars were removed and placed in liquid nitrogen. In the laboratory, enamel organs were dissected and RNA was extracted separately for each age, using the guanidinium isothiocyanate procedure (Chirgwin et al., 1979). RNA quality was checked by denaturing agarose gel electrophoresis.

Reverse transcriptase-polymerase chain reaction

RT-PCR was carried out according to the procedure in the Gene-Amp RNA PCR kit from Perkin Elmer Corp. (Foster City, CA). cDNA was synthesized in a volume of 20#1 containing 10mM tris-HC1, pH 8.3, 5 mM MgC12, 50 mM KCI, 1 mM each of dATP, dGTP, dCTP and dTTP, 1 unit of RNase inhibitor, 2 p g of total RNA, 2.5 units of MuLV reverse transcriptase and 2.5 #M oligo-d(T)j 6 primer. The cDNA synthesis was carried out for one cycle at 42°C for 15 min, 99°C for 5 min, 5°C for 5 min in a Perkin Elmer DNA thermocycler 480. PCR amplification was done immediately after cDNA synthesis in a total volume of 100#l containing 10 mM tris-HCl pH 8.3, 50 mM KCI, 4 mM MgC12, 2.5 units of Taq DNA polymerase and 35 pmol of each primer. The amplification was done for 35 cycles of 95°C for 1 min and 60°C for 1 min, with an additional extension time of 7 min at 60°C. RT-PCR products were analysed by agarose gel electrophoresis using Metaphor agarose (FMC Bioproducts, Rockland, ME).

The sequence of PCR primers used was (see Fig. 1):

No. 706 > C A G G G T A T C A G T A T T G A C G A G A C T G C for newly described exon 4 of amelogenin;

No. 479 > GGGGAATTCTTTGCCTGCCT CCT G G G A G CA G CCT for exon 2, amino acids - 9 to - 2 ;

No. 480 < GGGGAATTCCTCCCGCTTGGTCTTGTCTGTTGCT for C-terminus;

No. 731 > G C A G G T C G A C A A C T T A C T G A G A A C G T G T G T T C for exon l;

No. 733 > CATGGGGACCTGGATTTTGT T T G for exon 2, amino acids - 1 6 to - 9 ;

No. 738 < G T G A G G A T C T T C A T G A G G T A G T for 3' bovine and human fl-actin;

No. 739 > A C C A A C T G G G A C G A C A T G G A G A for 5' bovine and human fl-actin.

Northern blot and amelogenin probes

Twenty micrograms of total RNA for each sample were fractionated by electrophoresis in a

Probes and primers were synthesized by Wistar Biotechnology Resource Facilities, Philadelphia, PA.

Bovine amelogenin mRNA 207

731 479 480 707 (7O8)

733

706

Fig. 1. Diagram of the bovine X-chromosomal amelogenin cDNA, divided into exons, with exon numbers within the bar. The newly identified exon 4 was rarely found in transcripts and therefore was placed below the bar. Oligo- nucleotides used as primers or probes are indicated by number and direction (-< and > ). No. 708 is comple- mentary to the Y transcript at the same position as No. 707 anneals to the X transcript. The vertical dashed line is the site of alternative splicing within exon 6 to produce LRAP. Oligomer No. 663 is the junction between exon 5 and the 3' end of exon 6, as used in LRAP. Oligomer No. 664 is the junction between exons 2 and 5, containing 13 nucleotides

from each exon.

DNA sequence analysis

DNA sequence was obtained by the dideoxynucleo- tide chain terminatior, procedure (Sanger, Nicklen and Coulson, 1977) using the PCR amplification technique with fluorescent dye-labelled terminators on an ABI instrument, according to the manufac- turer 's recommendations (Applied Biosystems, Inc., Foster City, CA). Sequence was assembled with Sequencher software (Gene Codes Corp., Ann Arbor, MI) and was analysed using the Genetics Computer Group 's Sequence Analysis Software Package.

RESULTS

A series of oligonucleotides corresponding to bovine amelogenin cDNA or genomic sequences was synthesized. The location and orientation of these oligomers are shown in the cartoon of cDNA organization (Fig. 1). Because an additional exon that is included in some of the alternative splice products (exon 4 in Fig. 1) will be described, exons 4 - 6 (Gibson et al., 1991a, 1992) have been renumbered to become exons 5-7, which is consistent with the human and mouse terminology (Salido et al., 1992; Simmer et al., 1994).

X and Y chromosomal amelogenin mRN As

R NA was extracted from the enamel organs and duplicate Northern blots were made containing enamel-organ RNA from male and female animals of various ages, which were hybridized under identical conditions using the X-chromosome (No. 707) or Y-chromosome (No. 708) amelogenin-specific probe, radiolabelled to the same specific activity (Fig. 2). Relatively wide bands at approx. 850 and 450 bases are evident in the Northern blot probed with the X-specific oligomer. When the Y-specific probe was used, there was no hybridization to samples from female animals, and the 450-base band was not apparent, as expected (Gibson et al., 1992).

The Northern blot hybridized with the Y-specific probe was then rehybridized with the X-specific probe to indicate the presence of amelogenin R NA in all lanes (Fig. 2, right). Densitometric measurement showed at least six fold more X- than Y-chromosomal transcripts at these stages of development. These

7 0 7 7 0 8 7 0 8 / 7 0 7 X-l inked Y-l inked Combined

1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6

Probe

lane #

kb

__ g.5 - - 7.5

m 4.4

r R N A

- - 28S

m 2 . 4 185

1 . 4

Fig. 2. Northern blot showing bovine X- and Y-chromosomal amelogenin mRNAs. Two Northern blots were prepared and hybridized under identical conditions, using oligomer No. 707 (X amelogenin) or No. 708 (Y amelogenin) as probes, which were radiolabelled to the same specific activity. Subsequently, the Northern blot hybridized with No. 708 was rehybridized with No. 707, without removing the first probe, to show the presence of amelogenin mRNA in all lanes. Lane 1,142 days--F; lane 2, 142 days--M; lane 3, 151 days--F; lane 4, 158 days--M; lane 5, 182 days--F; lane 6, 184 days--M: F, female; M, male.


experiments were repeated twice more with similar results.

L R A P m R N A during bovine tooth development

Densitometric comparison of large (850 nucleotide) to small (450 nucleotide) alternatively spliced mRNAs, detected with the X-chromosomal amelogenin probe, revealed more than nine times more large mRNAs than small amelogenin mRNAs. The band width in the Northern analysis (Fig. 2, left) is due to the fact that at least two large and two small alternatively spliced amelogenin messages have been identified. The LRAP is thought to be the predominant small mRNA seen in Northern blots, as LRAP protein is abundant in protein extracts (Fincham et al., 1983). LRAP mRNA will contain exons 1, 2, 3, 5, and a small segment of exons 6 and 7 (see Fig. 1). In order to determine whether LRAP mRNA levels vary during development, oligomer primers Nos 479 and 480 were used to amplify this alternatively spliced

A 1 2 3 4 5 6 C

B M 1 2 3 4 5 6 C

kb

310 287 /271

234 194

C

I

w

M I 2 3 4 5 6

3 1 o 287 /271 . . . . . . . . . . . . . . . ,~ . . . . . . . . . . . . . . . .

2 3 4

194 Fig~ 3. Analysis of bovine enamel organ mRNA by RT-PCR. (A) LRAP RT-PCR product of 200 bp using primers No. 479 and 480; lane C contains control RT-PCR product from the kit. (B) RT-PCR with various combinations of PCR primers amplified the LRAP message; lane l, primers No. 479 and 480; lane 2, primers No. 479 and 707; lane 3, primers No. 731 and 480; lane 4, primers No. 733 and 480; lane 5, primers No. 731 and 707; lane 6, primers No. 733 and 707; lane C contains control PCR product from the kit. (C) fl-Globin primers correctly amplified a 353-bp PCR product in all samples. (A) and (C) Lanes 1-6 same as for Fig. 2; (B) lanes 1, 3 and 5, 182 days--F; lanes 2, 4 and

6, 184 days--M.

mRNA from animals at different stages of development, using RT-PCR. RNA samples from male and female fetuses between 142 days and 184 days of gestation were subjected to RT-PCR. For each sample the expected 200-bp product was seen, with no significant differences in amount between ages (Fig. 3A).

A control for this experiment was bovine fl-actin. Primers No. 738 and No. 739 were made from the region of the fl-actin coding sequence that was identical in bovine and human (Degen et al., 1983; Nakajima-Iijima et al., 1985), and plasmid pHFflA-1 containing human fl-actin cDNA was used as the positive control for PCR (Gunning et aL, 1983). The fl-actin control RT-PCR indicates that each sample contained a similar amount of RNA (Fig. 3C).

We had predicted that the amelogenin primers would also amplify a full-length 600-bp product but this was not obtained under these conditions, pre- sumably due to the high GC content of the internal segment. A similar result was also obtained when an enamel-organ cDNA library (Z. A. Yuan and C. W. Gibson, unpublished) was the target for PCR; again

kb 9.50 7.50

4.40

1 2 3 4 5

2.40

1.40

0.24 Fig. 4. Northern blot of bovine enamel organ mRNA. Probe No. 663 was designed specifically to detect the splice junction within exon 6, as found in LRAP, and did not detect full-length amelogenin. Lane 1, 140 days--M; lane 2, 156 days--M; lane 3, 158 days--F; lane 4, 179 days--F;

lane 5, 200 days--M.


no full-length product was obtained using primer No. 479 or No. 480 and a vector-specific primer (not shown). Two of the RNA samples were analysed by

0 . 8 -

O

0.6-

0.4-

0.2-

-o- A

I I I I

0.8

0 . 6 -

O

0.4-

0.2-

135 150 165 180

Fetal Age (days)

195

[ - * - B

0 I I I I

135 150 165 180 195

Fetal Age (days)

Fig. 5. Densitometry of Northern blots. (A) Band intensity for LRAP or (B) mRNA missing exon 3 as a ratio with total X-chromosomal amelogenin message, obtained by

rehybridizing th,: blots with probe No. 707.

RT-PCR with a variety of sets of primers (Fig. 3B). Each should be able to amplify any cDNA containing exons 2 and 6. It was possible to produce a full-length PCR product only when the samples were denatured for 10 min at 85°C before RT-PCR.

As quantification may prove difficult when a large number of alternative splice products are expected, especially when some of them are very GC rich (Templeton, Urcelay and Safer, 1993), a Northern blot was used to confirm the LRAP abundance during development. An antisense oligomer probe was synthesized to span the splice junction with 13 nucleotides on each side (No. 663; Fig. 4). This blot was subsequently stripped of probe and rehybridized with No. 707, which recognizes all X-chromosomal amelogenin transcripts. By densitometry it was determined that the ratio of LRAP to X-chromosomal amelogenin mRNA did not change appreciably during these stages of development (Fig. 5A).

The amelogenin message from which exon 3 had been spliced

The coding sequence for exon 3 may be spliced from bovine, human, and porcine mRNAs (Salido

1 2 3 4 5 6 7

kb 9 .50

7 .50

4 .40

2 .40 m

1.40

0 .24

Fig. 6. Northern blot showing amelogenin mRNAs from which exon 3 has been removed by alternative splicing. Probe No. 664 detects the junction between exons 2 and 5. Lanes 1-5, same as for Fig. 4. Lane 6, positive control plasmid with only exons 2 and 5; lane 7, negative control

plasmid with exons 2, 3 and 5.


e t al., 1992; Yamakoshi et al., 1994; Gibson et al., 1995). In order to determine whether this mRNA is prominent among the alternatively spliced amelogenin messages, primer No. 664 was synthesized specifically to detect messages missing exon 3, again by spanning the splice junction. Figure 6 shows that alternative splicing, of exon 3 by Northern blot was detectable at all ages examined. The controls are plasmids without exon 3 (+ control; lane 6) or with exon 3 ( - control; lane 7) within the cloned cDNA (Gibson et al., 1995). The blot was subsequently stripped of probe and rehybridized with No. 707 as before. In this case there was a decrease in message missing exon 3 relative to total amelogenin X-chromosomal mRNA (Fig. 5B). This experiment was repeated three times, with decreases of 2.3-, 2.6- and 2.1-fold, as measured by densitometry, during these stages of development.

A new bovine amelogen in exon

An additional exon had been found in the sequence of one of the human X-chromosomal amelogenin PCR products, appearing at low frequency using RT-PCR (Salido et al., 1992). The bovine X-chromo-

somal amelogenin gene was examined for the presence of this exon; Fig. 7(A) shows a potential exon coding for 14 amino acids and the adjacent sequences. To determine whether this exon is included in bovine alternatively spliced mRNAs, oligomer No. 706 was synthesized, corresponding to the predicted genomic exon-4 sequence (outlined letters in Fig. 7A). This oligomer was used with the 3'-oligo No. 480 for RT-PCR. The product of this reaction was cloned and sequenced. Four clones contained the new exon sequence plus the downstream intron (63 bases) followed by exon 5 and the 3' end of exon 6 at the internal splice site (vertical broken line in Fig. 1) found in the LRAP cDNAs. As intron 5 and part of exon 6 are spliced out, this product could not be a PCR product resulting from amplification of genomic DNA. Figure 7(B) shows the predicted amino acid sequence for the exon and the extension into the intron, along with exon-4 sequences from human, mouse and pig for comparison. Translation would end at the termination codon found in the intron, and would yield a peptide containing 41 amino acids, assuming upstream exons did not vary. If this mRNA

(A) irrtron 3 exon 4

TTCTTTCTTCCTTAG I AAC TCC TAT TTT / ash ser tyr phe

gin gly lie set ile asp lys thr

exon 4 Intron 4

~CA TTA I GTGAGTCTATATTTCATGCAA / ala leu

CTAAATTAAAACAAATGTATTCTAATCTTTC

Intron 4 exon 5

TTrCTCITAAG I GTG CTC / val leu

(e)

bovine N S Y F Q G I S I D K T A L V S L Y F M Q L N T m

human N S H S Q A I N V D R T A L

murine K S H S Q A I N T D R T A L

porcine K S G S W G A x L T A F V S ~ V Q

Fig. 7. Bovine amelogenin exon 4 DNA sequence, and predicted amino acid sequence. (A) DNA sequence of predicted bovine X-chromosomal amelogenin exon 4 and adjacent intron sequences. The PCR primer No. 706 sequence is indicated in outlined letters. The predicted amino acid sequence is listed beneath the respective codons, and the translation termination codon in the intron is marked ***. (B) Exon 4 sequences for four species. The bovine predicted sequence ends within the intron: human (Salido et al., 1992), murine

(Simmer et al., 1994) and porcine (Yamakoshi et al., 1994) are also given.


were translated, it would yield a peptide with a C-terminus unique fi:om any previously described amelogenin protein.

DISCUSSION

The expression of the amelogenin genes, in both human and bovine species, is unusual in having genes on both the X and Y chromosomes that are transcriptionally act~ve. In addition, alternative splicing of the primary transcript can be extensive, as all or part of most coding-region exons can be deleted. The observation that some of the alternative splicing is developmentally regulated, while other spliced messages remain relatively constant in amount, provides some insight into this highly ordered process.

There are approx, six times more X- than Y- specific transcripts in bovine, similar to the situation found in human where the frequency of appearance of chromosome-specilic RT-PCR products has been examined (Salido et al., 1992). This difference in abundance is probably due to the regulatory regions of the genes upstream from the transcription start sites, an area currently under investigation in this laboratory. One significance of the bovine X-chromosomal gene product i,; that this amelogenin contains a domain that is predicted to form a fl-spiral thought to participate in calcium movement (Renugopala- krishnan et al., 1986; Renugopalakrishnan et al., 1988). The bovine Y-specific amelogenin is without the E-spiral domain, and there are 13 other amino acid differences from the X-specific protein (Gibson et al., 1992). In addition, the Y-specific primary transcript cannot be spliced to form an LRAP peptide, as the required consensus sequence AG for splicing has been replaced by AA. Therefore, the Y-specific transcript contributes two mRNAs with exons 1, 2, 3, 5, 6 and 7, and a similar mRNA missing exon 3 (Gibson et al., 1995). Alternative splicing of the X-specific primary transcript may delete exons 3 and 4, or most of exon 6, plus various combinations of these splicings.

The structure of the amelogenin genes is unusual in that several exons are much shorter than the average length of 137 nucleotides (Hawkins, 1988). Exons 3, 4, and 5 are 48, 42 and 45 nucleotides in length, and exons containing fewer than 50 nucleotides are frequently deleted by alternative splicing. One explan- ation relies on the 'exon definition' model of splice-site determination in which exon recognition is determined by the recognition of exon boundaries rather than intron ends. This type of regulation is thought to occur in higher euka;yotes containing long introns (Berget, 1995). Short exons are removed, as their length is not sufficient for proper splicing most of the time. The size of the amelogenin exons correlates with their appearance in that the shortest (42 base) exon 4 is most frequently spliced out. In agreement is our observation that when exon 4 is present, intron 4 is

frequently not removed by splicing. This incompletely spliced message could therefore represent a splicing intermediate, remaining partly unspliced due to small exon and intron size. Exon 6 is also unusual, as at 489 bases it is longer than more than 99% of primate internal exons (Berget, 1995). The exon definition model also predicts that long exons are frequently spliced within the exon; for amelogenin, internal splicing produces LRAP.

The question remains as to why these genes have evolved to contain exons too small or too large to be spliced efficiently. The predicted result would be a population of proteins existing in a ratio determined by splicing efficiency, each, perhaps, with a function according to the coding regions present. Rather than the development of a family of genes, this may be the favoured mechanism where a gene is expressed in only one tissue and alternative tissue-specific regulation is not required.

RT-PCR is occasionally used for quantitative purposes to compare relative amounts of transcripts. Investigators can be troubled by extra bands, which prove to be either single-stranded products or hybrids rather than authentic products (Zacharias, Garamszegi and Strehler, 1994). In the case of bovine amelogenin, some of the predicted products did not appear after RT-PCR, even though bands of the appropriate size could be detected by Northern blot. However, using internal primers with RT-PCR or with PCR of a cDNA library, the appropriately spliced message of interest could be detected (not shown). It was con- cluded that there is difficulty in comparing amounts of mRNAs when there is a high level of alternative splicing, especially when some of the products are GC-rich, and in these cases other techniques such as Northern blots combined with densitometry are preferable.

PCR amplification using primers that specifically detect exon 4 gave a low-abundance product that contained exon 4 plus intron 4, exon 5 and the 3' end of exon 6. If this message were translated, it would end at the translation termination codon within the intron, and yield a 41-amino acid amelogenin with a C-terminus different from any thus far reported for any species. When an antibody to mouse exon 4 was tested with a Western blot containing murine, bovine and porcine enamel proteins, strong stain was observed with the murine sample, weak staining resulted for the bovine sample, and the porcine sample was not detected (Simmer et al., 1994). This is not unexpected since only 7 of the predicted 14 amino acids are identical in mouse and bovine, and only 3 of 14 are identical in the pig. Although amelogenins are highly conserved between species, there is minor similarity among the four species for the exon-4 sequence.

Other differences between species are equally strik- ing. For example, (a) genes are found on the X and Y chromosomes in man and bovines, but only on the X chromosome in mouse and rat; (b) exon 3 can be


spliced out of the bovine X- and Y-specific amelogenins in bovine, but is present in all mouse c D N A s described thus far (Simmer et aL, 1994); (c) exon 6 has several internal sites used for splicing in the mouse, but only one has been detected in bovine (Simmer et al., 1994; Gibson et al., 1995); and (d) the fl-spiral domain is found only in the bovine X-chromosomal amelogenin.

An intriguing question for future studies will be how the presence of various exons and domains can alter the structure-function relations of amelogenin proteins and interactions between organic matrix molecules and the developing enamel mineral. In order to understand the structure-function relations, in vitro expression systems will be useful, but it will also be necessary to determine whether relative levels of individual proteins match their respective m R N A levels.

Acknowledgements--We thank H. Koo and M. Bashir for assistance with dissections; T. Tucker of the Biopolymer Analysis Laboratory for DNA sequence determination; P. Leboy for critical reading of the manuscript; W. Abrams and J. Taplin for help in assembling the figures; R. Ma for assistance with oligomers; and Y. Shen for the gift of pHFflA-I. Support for this work was provided from grant DE 10149 from the National Institute of Dental Research (C.W.G.).

REFERENCES

Aldred M. J., Crawford P. J. M., Roberts E. and Thomas N. S. T. (1992) Identification of a nonsense mutation in the amelogenin gene (AMELX) in a family with X-linked amelogenesis imperfecta (AIHI). Hum. Genet. 90, 413-416.

Berget S. M. (1995) Exon recognition in vertebrate splicing. J. bioL Chem. 270, 2411-2414.

Brookes S. J., Robinson C., Kirkham J. and Bonass W. A. (1995) Biochemistry and molecular biology of amelogenin proteins of developing dental enamel. Archs oral Biol. 40, 1-14.

Chirgwin J. M., Przybyla A. E., MacDonald R. J. and Rutter W. J. (1979) Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18, 5294-5299.

Couwenhoven R. I. and Snead M. L. (1994) Early determination and permissive expression of amelogenin transcription during mouse mandibular first molar development. Dev. Biol. 164, 290-299.

Degen J. L., Neubauer M. G., Friezner-Degen S. J., Seyfried C. E. and Morris D. R. (1983) Regulation of protein synthesis in mitogen-activated bovine lymphocytes: analysis of actin-specific and total mRNA accumulation and utilization. J. biol. Chem. 258, 12153-12162.

DenBesten P. K. and Heffernan L. M. (1989) Separation by polyacrylamide gel electrophoresis of multiple proteases in rat and bovine enamel. Archs oral BioL 34, 399-404.

Deutsch D. (1989) Structure and function of enamel gene products. Anat. Rec. 224, 189-210.

Diekwisch T., David S., Bringas P. Jr, Santos V. and Slavkin H. C. (1993) Antisense inhibition of AMEL translation demonstrates supramolecular controls for enamel HAP crystal growth during embryonic mouse molar development. Development 117, 471-482.

Eckstein P. and Kelly W. A. (1977) Implantation and Development of the Conceptus. Reproduction of Domestic Animals (Eds Cole H. H. and Cupps P. T.), 3rd edn., p. 335. Academic Press, San Diego, CA.

Fincham A. G., Hu Y., Lau E. C., Slavkin H. C. and Snead M. L. (1991a) Amelogenin post-secretory processing during biomineralization in the postnatal mouse molar tooth. Archs oral Biol. 36, 305-317.

Fincham A. G., Bessem C. C., Lau E. C., Pavlova Z., Shuler C., Slavkin H. C. and Snead M. L. (1991b) Human developing enamel proteins exhibit a sex-linked dimorphism. Calcif. Tissue Int. 48, 288-290.

Fincham A. G., Belcourt A. B., Termine J. D., Butler W. T. and Cothran W. C. (1983) Amelogenins, sequence hom- ologies in enamel-matrix proteins from three mammalian species. Biochem. J. 211, 149-154.

Fincham A. G., Moradian-Oldak J., Simmer J. P., Sarte P., Lau E. C., Diekwisch T. and Slavkin H. C. (1994) Self-assembly of a recombinant amelogenin protein generates supramolecular structures. J. struct. Biol. 112, 103-109.

Gibson C., Golub E., Herold R., Risser M., Ding W., Shimokawa H., Young M., Termine J. and Rosenbloom J. (1991a) Structure and expression of the bovine amelogenin gene. Biochemistry 30, 1075-1079.

Gibson C. W., Golub E., Ding W., Shimokawa H., Young M., Termine J. T. and Rosenbloom J. (1991b) Identi- fication of the leucine-rich amelogenin peptide (LRAP) as the translation product of an alternatively-spliced transcript. Biochem. biophys, res. Commun. 174, 1306-1312.

Gibson C. W., Golub E. E., Abrams W. R., Shen G., Ding W. and Rosenbloom J. (1992) Bovine amelogenin message heterogeneity: alternative splicing and Y-chromosomal gene transcription. Biochemistry 31, 8384-8388.

Gibson C. W., Kucich U., Collier P., Shen G., Decker S., Bashir M. and Rosenbloom J. (1995) Analysis of amelogenin proteins using monospecific antibodies to defined sequences. Connective Tissue Res. 32, 109-114.

Gunning P., Ponte P., Okayama H., Engel J., Blau H. and Kedes L. (1983) Isolation and characterization of full-length cDNA clones for human ~-, fl- and F-actin mRNAs: skeletal but not cytoplasmic actins have an amino-terminal cysteine that is subsequently removed. Mol. Cell. Biol. 3, 787-795.

Hawkins J. D. (1988) A survey on intron and exon lengths. Nucleic acids Res. 16, 9893-9908.

Lagerstrom M., Dahl N., Nakahori Y., Nakagome Y., Backman B., Landegren U. and Pettersson U. (1991) A deletion in the amelogenin gene (AMG) causes X- linked amelogenesis imperfecta (AIHI). Genomics 10, 971 975.

Lagerstrom-Fermer M., Nilsson M., Backman B., Salido E., Shapiro L., Pettersson U. and Landegren U. (1995) Amelogenin signal peptide mutation: correlation between mutations in the amelogenin gene (AMGX) and manifes- tations of X-linked amelogenesis imperfecta. Genomics 26, 159-162.

Lau E. C., Simmer J. P., Bringas P. Jr, Hsu D. D.-J., Hu C.-C., Zeichner-David M., Thiemann F., Snead M. L., Slavkin H. C. and Fincham A. G. (1992) Alternative splicing of the mouse amelogenin primary RNA transcript contributes to amelogenin heterogeneity. Biochem. biophys. Res. Commun. 188, 1253-1260.

Lau E. C., Mohandas T. K., Shapiro L. J., Slavkin H. C. and Snead M. L. (1989) Human and mouse amelogenin gene loci are on the sex chromosomes. Genomics 4, 162-168.

Lench N. J., Brook A. H. and Winter G. B. (1994) SSCP detection of a nonsense mutation in exon 5 of the amelogenin gene (AMGX) causing X-linked amelogenesis imperfecta (AIHI). Human mol. Genet. 3, 827-828.

Limeback H. (1991) Molecular mechanisms in dental hard tissue mineralization. Curt. Op. Dentistry 1, 826-835.

Limeback H. and Simic A. (1990) Biochemical characterization of stable high molecular-weight aggregates of amelogenins formed during porcine enamel development. Archs oral Biol. 35, 459-468.


Nakajima-Iijima S., Hamada H., Reddy P. and Kakunaga T. (1985) Molecular structure of the human cytoplasmic beta-actin gene: inter,;pecies homology of sequences in the introns. Proc. Natl. Acad. Sci. USA 82, 6133-6137.

Overall C. M. and Limeback H. (1988) Identification and characterization of enamel proteinases isolated from developing enamel. Biochem. J. 256, 965-972.

Renugopalakrishnan V., Strawich E. S., Horowitz P. M. and Glimcher M. J. (1986) Studies of the secondary structures of amelogenin from bovine tooth enamel. Biochemistry 25, 4879-4887.

Renugopalakrishnan V., Pattabiraman N., Langridge R., Strawich E., Glimcher M. J., Horowitz P. M., Zheng S., Tu A. T., Huang S. G., Prabhakaran M., Duzgunes N. and Rapaka R. S. (1988) Unique secondary structure of amelogenin and its possible implications for the mineralization of enamel. In Recent Progress in Chemistry and Biology of Centrally Acting Peptides (Eds Dhawan B. N. and Rapaka R. S.), p. 87-107. Central Drug Research Institute. Lucknow, India.

Robinson C., Shore R. C., Kirkham J. and Stonehouse N. J. (1990) Extracellular processing of enamel matrix proteins and the control of crystal growth. J. Biol. Buccale lg, 355-361.

Salido E. C., Yen P. H., Koprivnikar K., Yu L.-C. and Shapiro L. J. (1992) The human enamel protein gene amelogenin is expressed from both the X and the Y chromosomes. Am. J. hum. Genet. 50, 303-316.

Sanger F., Nicklen S. and Coulson A. (1977) DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467.

Shimizu M., Tanabe T. and Fukae M. (1979) Proteolytic enzyme in porcine immature enamel. J. dent. Res. 58(B), 782-789.

Simmer J. P. and Fincham A. G. (1995) Molecular mechanisms of dental enamel formation. Crit. Rev. oral Biol. Med. 6, 84-108.

Simmer J. P., Hu C.-C., Lau E. C., Sarte P., Slavkin H. C. and Fincham A. G. (1994) Alternative splicing of the mouse amelogenin primary RNA transcript. Calcif. Tissue Int. 55, 302-310.

Snead M. L., Lau E. C., Zeichner-David M., Fincham A. G., Woo S. L. C. and Slavkin H. C. (1985) DNA sequence for cloned cDNA for murine amelogenin reveal the amino acid sequence for enamel- specific protein. Biochem. biophys. Res. Commun. 129, 812-818.

Templeton N. S., Urcelay E. and Safer B. (1993) Reduc- ing artifact and increasing the yield of specific DNA target fragments during PCR-RACE or anchor PCR. Biotechniques 15, 48-51.

Yamakoshi Y., Tanabe T., Fukae M. and Shimizu M. (1994) Porcine amelogenins. Calcif. Tissue Int. 54, 69-75.

Zacharias D. A., Garamszegi N. and Strehler E. E. (1994) Characterization of persistent artifacts resulting from RT-PCR of alternatively spliced mRNAs. Biotechniques 17, 652-655.

Documents

Analysis of amelogenin mRNA during bovine tooth development