Amelogenins as potential buffers during secretory-stage amelogenesis

Amelogenins are the most abundant protein species in forming dental enamel, taken to regulate crystal shape and crystal growth. Unprotonated amelogenins can bind protons, suggesting that amelogenins could regulate the pH in enamel in situ. We hypothesized that without amelogenins the enamel would acidify unless ameloblasts were buffered by alternative ways. To investigate this, we measured the mineral and chloride content in incisor enamel of amelogenin-knockout (AmelX(-/-)) mice and determined the pH of enamel by staining with methyl-red. Ameloblasts were immunostained for anion exchanger-2 (Ae2), a transmembrane pH regulator sensitive for acid that secretes bicarbonate in exchange for chloride. The enamel of AmelX(-/-) mice was 10-fold thinner, mineralized in the secretory stage 1.8-fold more than wild-type enamel and containing less chloride (suggesting more bicarbonate secretion). Enamel of AmelX(-/-) mice stained with methyl-red contained no acidic bands in the maturation stage as seen in wild-type enamel. Secretory ameloblasts of AmelX(-/-) mice, but not wild-type mice, were immunopositive for Ae2, and stained more intensely in the maturation stage compared with wild-type mice. Exposure of AmelX(-/-) mice to fluoride enhanced the mineral content in the secretory stage, lowered chloride, and intensified Ae2 immunostaining in the enamel organ in comparison with non-fluorotic mutant teeth. The results suggest that unprotonated amelogenins may regulate the pH of forming enamel in situ. Without amelogenins, Ae2 could compensate for the pH drop associated with crystal formation.

M180 amelogenin processed by MMP20 is sufficient for decussating murine enamel

Amelogenin (AMELX) and matrix metalloproteinase-20 (MMP20) are essential for proper enamel development. Amelx and Mmp20 mutations cause amelogenesis imperfecta. MMP20, a protease secreted by ameloblasts, is responsible for processing enamel proteins, including AMELX, during the secretory stage of enamel formation. Of at least 16 different amelogenin splice products, the most abundant isoform found in murine ameloblasts and developing enamel is the M180 protein. To understand the role of MMP20 processing of M180 AMELX, we generated AmelxKO/Mmp20KO (DKO) mice with an amelogenin (M180Tg) transgene. We analyzed the enamel phenotype by SEM to determine enamel structure and thickness, µCT, and by nanoindentation to quantify enamel mechanical properties. M180Tg/DKO mouse enamel had 37% of the hardness of M180Tg/AmelxKO teeth and demonstrated a complete lack of normal prismatic architecture. Although molar enamel of M180Tg/AmelxKO mice was thinner than WT, it had similar mechanical properties and decussating enamel prisms, which were abolished by the loss of MMP20 in the M180Tg/DKO mice. Retention of the C-terminus or complete lack of this domain is unable to rescue amelogenin null enamel. We conclude that among amelogenins, M180 alone is sufficient for normal enamel mechanical properties and prism patterns, but that additional amelogenin splice products are required to restore enamel thickness.

Expression and function of enamel-related gene products in calvarial development

Enamel-related gene products (ERPs) are detected in non-enamel tissues such as bone. We hypothesized that, if functional, ERP expression corresponds with distinct events during osteoblast differentiation and affects bone development and mineralization. In mouse calvariae and MC3T3 cells, expression profiles of enamel-related gene products (ERPs) correlated with key events in post-natal calvarial development and MC3T3 cell mineralization. Developing skulls from both Amel- and Ambn-deficient animals were approximately 15% shorter when compared with those of wild-type controls, and their sutures remained patent for a longer period of time. Analysis of Amel- and Ambn-deficient calvariae and calvarial osteoblast cultures revealed a dramatic reduction in mineralized nodules, a significant reduction in Runx2, Sp7, Ibsp, and Msx2 expression, and a reduction in Alx4 in Amel-deficient calvariae vs. an increase in Alx4 in Ambn-deficient calvariae. Analysis of these data indicates that ERP expression follows defined developmental profiles and affects osteoblast differentiation, mineralization, and calvarial bone development. We propose that, in parallel to their role in the developing enamel matrix, ERPs have retained an evolutionary conserved function related to the biomineralization of bones.

The use of mouse models to investigate shear bond strength in amelogenesis imperfecta

Patients with amelogenesis imperfecta (AI) have defective enamel; therefore, bonded restorations of patients with AI have variable success rates. To distinguish which cases of AI may have good clinical outcomes with bonded materials, we evaluated etching characteristics and bond strength of enamel in mouse models, comparing wild-type (WT) with those having mutations in amelogenin (Amelx) and matrix metalloproteinase-20 (Mmp20), which mimic 2 forms of human AI. Etched enamel surfaces were compared for roughness by scanning electron microscopy (SEM) images. Bonding was compared through shear bond strength (SBS) studies with 2 different systems (etch-and-rinse and self-etch). Etched enamel surfaces of incisors from Amelx knock-out (AmelxKO) mice appeared randomly organized and non-uniform compared with WT. Etching of Mmp20KO surfaces left little enamel, and the etching pattern was indistinguishable from unetched surfaces. SBS results were significantly different when AmelxKO and Mmp20KO enamel surfaces were compared. A significant increase in SBS was measured for all samples when the self-etch system was compared with the etch-and-rinse system. We have developed a novel system for testing shear bond strength of mouse incisors with AI variants, and analysis of these data may have important clinical implications for the treatment of patients with AI.

The amelogenin C-terminus is required for enamel development

The abundant amelogenin proteins are responsible for generating proper enamel thickness and structure, and most amelogenins include a conserved hydrophilic C-terminus. To evaluate the importance of the C-terminus, we generated transgenic mice that express an amelogenin lacking the C-terminal 13 amino acids (CTRNC). MicroCT analysis of TgCTRNC29 teeth (low transgene number) indicated that molar enamel density was similar to that of wild-type mice, but TgCTRNC18 molar enamel (high transgene number) was deficient, indicating that extra transgene copies were associated with a more severe phenotype. When amelogenin-null (KO) and TgCTRNC transgenic mice were mated, density and volume of molar enamel from TgCTRNCKO offspring were not different from those of KO mice, indicating that neither TgCTRNC18 nor TgCTRNC29 rescued enamel's physical characteristics. Because transgenic full-length amelogenin partially rescues both density and volume of KO molar enamel, it was concluded that the amelogenin C-terminus is essential for proper enamel density, volume, and organization.

Synergistic roles of amelogenin and ameloblastin

Amelogenin and ameloblastin, the major enamel matrix proteins, are important for enamel mineralization. To identify their synergistic roles in enamel development, we generated Amel X(-/-)/Ambn(-/-) mice. These mice showed additional enamel defects in comparison with Amel X(-/-) or Ambn(-/-) mice. In 7-day-old Amel X(-/-)/Ambn(-/-) mice, not only was the ameloblast layer irregular and detached from the enamel surface, as in Ambn(-/-), but also, the enamel width was significantly reduced in the double-null mice as compared with Amel X(-/-) or Ambn(-/-) mice. Proteomic analysis of the double-null teeth revealed increased levels of RhoGDI (Arhgdia), a Rho-family-specific guanine nucleotide dissociation inhibitor, which is involved in important cellular processes, such as cell attachment. Both Amel X(-/-)/Ambn(-/-) mice and Ambn(-/-) mice displayed positive staining with RhoGDI antibody in the irregularly shaped ameloblasts detached from the matrix. Ameloblastin-regulated expression of RhoGDI suggests that Rho-mediated signaling pathway might play a role in enamel formation.

Transgenic mice that express normal and mutated amelogenins

Amelogenin proteins are secreted by ameloblasts within the enamel organ during tooth development. To better understand the function of the 180-amino-acid amelogenin (M180), and to test the hypothesis that a single proline-to-threonine (P70T) change would lead to an enamel defect similar to amelogenesis imperfecta (AI) in humans, we generated transgenic mice with expression of M180, or M180 with the proline-to-threonine (P70T) mutation, under control of the Amelx gene regulatory regions. M180 teeth had a relatively normal phenotype; however, P70T mineral was abnormally porous, with aprismatic regions similar to those in enamel of male amelogenesis imperfecta patients with an identical mutation. When Amelx null females were mated with P70T transgenic males, offspring developed structures similar to calcifying epithelial odontogenic tumors in humans. The phenotype argues for dominant-negative activity for the P70T amelogenin, and for the robust nature of the process of amelogenesis.

Amelogenin-mediated regulation of osteoclastogenesis, and periodontal cell proliferation and migration

We previously reported that amelogenin isoforms M180 and leucine-rich amelogenin peptide (LRAP) are expressed in the periodontal region, and that their absence is associated with increased cementum defects in amelogenin-knockout (KO) mice. The aim of the present study was to characterize the functions of these isoforms in osteoclastogenesis and in the proliferation and migration of cementoblast/periodontal ligament cells. The co-cultures of wild-type (WT) osteoclast progenitor and KO cementoblast/periodontal ligament cells displayed more tartrate-resistant acid phosphatase (TRAP)-positive cells than the co-cultures of WT cells. The addition of LRAP to both co-cultures significantly reduced RANKL expression and the TRAP-positive cells. Proliferation and migration rates of the KO cementoblast/periodontal ligament cells were lower than those of WT cells and increased with the addition of either LRAP or P172 (a porcine homolog of mouse M180). Thus, we demonstrate the regulation of osteoclastogenesis by LRAP, and the proliferation and migration of cementoblast/periodontal ligament cells by LRAP and P172.

Tooth enamel defects in mice with a deletion at the Arhgap 6/Amel X locus

The amelogenin proteins regulate enamel mineral formation in the developing tooth. The human AMELX gene, which encodes the amelogenin proteins, is located within an intron of the Arhgap 6 gene. ARHGAP 6 encodes a Rho GAP, which regulates activity of Rho A, a small G protein involved in intracellular signal transduction. Mice were generated in which the entire ARHGAP 6 gene was deleted by Cre-mediated recombination, which also removed the nested Amel X gene. Enamel from these mice appeared chalky white, and the molars showed excessive wear. The enamel layer was hypoplastic and non-prismatic, whereas other dental tissues had normal morphology. This phenotype is similar to that reported for Amel X null mice, which have a short deletion that removed the region surrounding the translation initiation site, and resembles some forms of X-linked amelogenesis imperfecta in humans. Analysis of the enamel from the Arhgap 6/Amel X-deleted mice verifies that the Amel X gene is nested within the murine Arhgap 6 gene and shows that removal of the entire Amel X gene leads to a phenotype similar to the earlier Amel X null mouse results, in which no amelogenin protein was detected. However, an unusual layer of aprismatic enamel covers the enamel surface, which may be related to the 1.1-Mb deletion, which included Arhgap 6 in these mice.

Relationship of phenotype and genotype in X-linked amelogenesis imperfecta

X-linked amelogenesis imperfectas (AI) resulting from mutations in the amelogenin gene (AMELX) are phenotypically and genetically diverse. Amelogenin is the predominant matrix protein in developing enamel and is essential for normal enamel formation. To date, 12 allelic AMELX mutations have been described that purportedly result in markedly different expressed amelogenin protein products. We hypothesize that these AMELX gene mutations result in unique and functionally altered amelogenin proteins that are associated with distinct amelogenesis imperfecta phenotypes. The AMELX mutations and associated phenotypes fall generally into three categories. (1) Mutations (e.g., signal peptide mutations) causing a total of loss of amelogenin protein are associated with a primarily hypoplastic phenotype (though mineralization defects also can occur). (2) Missense mutations affecting the N-terminal region, especially those causing changes in the putative lectin-binding domain and TRAP (tyrosine rich amelogenin protein) region of the amelogenin molecule, result in a predominantly hypomineralization/hypomaturation AI phenotype with enamel that is discolored and has retained amelogenin. (3) Mutations causing loss of the amelogenin C terminus result in a phenotype characterized by hypoplasia. The consistent association of similar hypoplastic or hypomineralization/hypomaturation AI phenotypes with specific AMELX mutations may help identify distinct functional domains of the amelogenin molecule. The phenotype-genotype correlations in this study suggest there are important functional domains of the amelogenin molecule that are critical for the development of normal enamel structure, composition, and thickness.

Over- and ectopic expression of Wnt3 causes progressive loss of ameloblasts in postnatal mouse incisor teeth

Intercellular signaling is essential for the development of teeth during embryogenesis and in maintenance of the continuously growing incisor teeth in postnatal rodents. WNT intercellular signaling molecules have been implicated in the regulation of tooth development, and the Wnt3 gene shows specific expression in the enamel knot at the cap stage. We demonstrate here that Wnt3 also is expressed in specific epithelial cell layers in postnatal incisor teeth. To begin to delineate the functions of Wnt3 in developing and postnatal teeth, we determined the effects of over- and ectopic expression of Wnt3 in the tooth epithelium of mice carrying a keratin 14-Wnt3 transgene. Expression of the transgene caused a progressive loss of ameloblasts from postnatal lower incisor teeth. Loss of ameloblasts may be due to defective proliferation or differentiation of ameloblast precursors, progressive apoptosis of ameloblasts, or loss of ameloblast stem cells.

The small bovine amelogenin LRAP fails to rescue the amelogenin null phenotype

Amelogenins are the most abundant secreted proteins in developing dental enamel. These evolutionarily-conserved proteins have important roles in enamel mineral formation, as mutations within the amelogenin gene coding region lead to defects in enamel thickness or mineral structure. Because of extensive alternative splicing of the primary RNA transcript and proteolytic processing of the secreted proteins, it has been difficult to assign functions to individual amelogenins. To address the function of one of the amelogenins, we have created a transgenic mouse that expresses bovine leucine-rich amelogenin peptide (LRAP) in the enamel-secreting ameloblast cells of the dental organ. Our strategy was to breed this transgenic mouse with the recently generated amelogenin knockout mouse, which makes none of the amelogenin proteins and has a severe hypoplastic and disorganized enamel phenotype. It was found that LRAP does not rescue the enamel defect in amelogenin null mice, and enamel remains hypoplastic and disorganized in the presence of this small amelogenin. In addition, LRAP overexpression in the transgenic mouse (wildtype background) leads to pitting in the enamel surface, which may result from excess protein production or altered protein processing due to minor differences between the amino acid compositions of murine and bovine LRAP. Since introduction of bovine LRAP into the amelogenin null mouse does not restore normal enamel structure, it is concluded that other amelogenin proteins are essential for normal appearance and function.

A new frameshift mutation encoding a truncated amelogenin leads to X-linked amelogenesis imperfecta

The amelogenin proteins are the most abundant organic components of developing dental enamel. Their importance for the proper mineralization of enamel is evident from the association between previously identified mutations in the X-chromosomal gene that encodes them and the enamel defect amelogenesis imperfecta. In this investigation, an adult male presenting with a severe hypoplastic enamel phenotype was found to have a single base deletion at the codon for amino acid 110 of the X-chromosomal 175-amino acid amelogenin protein. The proband's mother, who also has affected enamel, carries the identical deletion on one of her X-chromosomes, while the father has both normal enamel and DNA sequence. This frameshift mutation deletes part of the coding region for the repetitive portion of amelogenin as well as the hydrophilic tail, replacing them with a 47-amino acid segment containing nine cysteine residues. While greater than 60% of the protein is predicted to be intact, the severity of this phenotype illustrates the importance of the C-terminal region of the amelogenin protein for the formation of enamel with normal thickness.

Model system for evaluation of alternative splicing: exon skipping

Alternative splicing of the primary RNA transcript is a common mechanism for generating protein diversity. A model system was developed to study this process in vitro that is useful for evaluation of splicing of transcripts expressed in cells that do not grow well in culture. The system was used to analyze skipping of exon 4 of the amelogenin message, normally expressed in ameloblast cells for a short interval during tooth enamel development. Amelogenins are highly conserved proteins resulting from extensive alternative splicing, with domains involved in a range of functions, including mineral formation and intercellular signaling. In the bovine gene, the very short intron 4 was predicted to inhibit inclusion of exon 4, because in murine ameloblasts, exon 4 is detectably included in mRNA, and intron 4 is longer than the bovine counterpart. Bovine intron 4 was lengthened, and this size increase enhanced exon 4 inclusion sixfold to eightfold, although splice site selection was inaccurate. Intron length, therefore, is not the sole determinant controlling amelogenin exon 4 inclusion, and cis-acting inhibitory elements may also be involved in exon skipping. This vector system allows evaluation of splicing of a tissue-specific RNA by focusing on exons of interest through transfection of heterologous cultured cells without complications attributable to background transcription of the gene being evaluated.

Amelogenin-deficient mice display an amelogenesis imperfecta phenotype

Dental enamel is the hardest tissue in the body and cannot be replaced or repaired, because the enamel secreting cells are lost at tooth eruption. X-linked amelogenesis imperfecta (MIM 301200), a phenotypically diverse hereditary disorder affecting enamel development, is caused by deletions or point mutations in the human X-chromosomal amelogenin gene. Although the precise functions of the amelogenin proteins in enamel formation are not well defined, these proteins constitute 90% of the enamel organic matrix. We have disrupted the amelogenin locus to generate amelogenin null mice, which display distinctly abnormal teeth as early as 2 weeks of age with chalky-white discoloration. Microradiography revealed broken tips of incisors and molars and scanning electron microscopy analysis indicated disorganized hypoplastic enamel. The amelogenin null phenotype reveals that the amelogenins are apparently not required for initiation of mineral crystal formation but rather for the organization of crystal pattern and regulation of enamel thickness. These null mice will be useful for understanding the functions of amelogenin proteins during enamel formation and for developing therapeutic approaches for treating this developmental defect that affects the enamel.

Reduced hydrolysis of amelogenin may result in X-linked amelogenesis imperfecta

Amelogenesis imperfecta (AI) is a group of inherited disorders with defective tooth enamel formation caused by various gene mutations. One of the mutations substitutes a cytidine to adenine in exon 6 of the X-chromosomal amelogenin gene, which results in a proline to threonine change in the expressed amelogenin. This transformation is four amino acids N terminal to the proteinase cleavage site in amelogenin for enamel matrix metalloproteinase-20 (MMP-20), also known as enamelysin. MMP-20 effects the release of tyrosine rich amelogenin peptide (TRAP) from amelogenin. This study evaluated the rate MMP-20 hydrolyzes the putative mutated amelogenin cleavage site. The proteolytic site was modeled as a substrate by two synthetic peptides, P1 (SYGYEPMGGWLHHQ) and M1 (SYGYETMGGWLHHQ), selected from residue 36-49 of the amino acid sequence for amelogenin and the respective X-linked amelogenin mutant. Recombinant metalloproteinase-20 (rMMP-20) was used to digest the oligopeptides and the truncated peptides were separated by reversed phase HPLC and identified by mass spectrometry. The results demonstrate that both peptides are cleaved between tryptophan and leucine, matching the TRAP cutting site found in tooth enamel. However, the apparent first order rate of digestion of the mutation containing peptide by rMMP-20 was approximately 25 times slower than that of the non-mutated peptide. This study suggests that the reduced rate of TRAP formation due to a single amino acid substitution may alter enamel formation and consequently result in amelogenesis imperfecta.

Regulation of amelogenin gene expression

The X-chromosomal amelogenin gene is expressed at a high level by ameloblast cells within the enamel organ for a short time during tooth development. Therefore, expression is both tooth specific and developmentally regulated. A Y-chromosomal amelogenin gene is also active in human and cow, but has not been detected in mouse. Genes and/or cDNAs have been cloned for mouse, human, cow, rat, pig, opossum, and hamster, and analyses have indicated that coding and upstream regions are conserved across species. Alternative splicing is extensive and produces as many as 9 mRNAs from the 7 exon murine gene, resulting from single and multiple exon skipping and alternate 3' site selection within exon 6. The pattern of alternative splicing varies both between species and during development, which is expected to result in some diversity through varying complements of amelogenin proteins associated with this highly conserved gene.

Comparison of upstream regions of X- and Y-chromosomal amelogenin genes

The amelogenin genes encode abundant enamel proteins that are required for the development of normal tooth enamel. These genes are active only in enamel-forming ameloblasts within the dental organ of the developing tooth, and are part of a small group of genes that are active on both sex chromosomes. The upstream regions of the bovine X- and Y-chromosomal and the sole murine X-chromosomal amelogenin genes have been cloned and sequenced, and conservation at nearly 60% is found in the 300 bp upstream of exon 1 for the 3 genes. A region of the bovine X-chromosomal gene that has inhibitory activity when assayed by gene transfer into heterologous cells includes motifs that have a silencing activity in other genes, and may be important to the mechanism that represses amelogenin expression in non-ameloblast cells in vivo. A comparison of sequences from three genes has led to the identification of several regions with conserved motifs that are strong candidates for having positive or negative regulatory functions, and these regions can now be tested further for interaction with nuclear proteins, and for their ability to regulate expression in vivo.

DNA sequences of amelogenin genes provide clues to regulation of expression

The amelogenins are a heterogeneous group of enamel proteins, which have an important function in enamel formation, as mutations in the amelogenin gene result in the enamel defect amelogenesis imperfecta. The cDNAs that encode murine, bovine, human, porcine, rat and opossum amelogenins have been cloned, and as many as nine alternatively spliced messages can be produced from a single primary transcript, explaining some of the protein heterogeneity. Bovine and human amelogenin genes are found on both X and Y chromosomes, and the sexually dimorphic proteins would have 87-93% identity. A comparison of genes from human, bovine and mouse indicates that they are organized into seven exons, and sequences are highly homologous among species. Bovine, murine and human upstream regions also have similarities, with consensus sequences for potential binding of transcription factors, such as AP1 and CTF/NF1. Transgenic mouse studies have shown that 2300-3500 bp of upstream region are sufficient for expression, while 900 bp are insufficient. Analysis of DNA sequence has identified (a) major homology between species for coding exons with the exception of exon 4, (b) similarities in upstream regions likely involved in tissue specific regulation of expression, and (c) sequences at the RNA splice sites which may determine exon inclusion or skipping.

p53 influences mice skeletal development

The p53 tumor suppressor gene encodes a transcriptional activator whose targets include genes that regulate cell cycle progression and apoptosis. Since we have shown that a critical event in the life history of the chondrocyte is programmed cell death, we asked the question: does loss of the p53 gene influence skeletogenesis? Female p53(+/-) mice were mated with p53(+/-) male mice and 17-day-old fetal mice were studied. Exencephaly was the most profound skeletal defect of the p53 null mutation. This defect was due to failure of formation of the bones that comprise the mouse calvarium. There was also loss of the hyoid bone, and defective mineralization of the manubrium sternum and the terminal phalanges. In the homozygous state (-/-), in the absence of exencephaly, the number of skeletal deformities was markedly reduced. Aside from the gross changes associated with null status, the mutants exhibited alterations in bone length and width. Small differences in the size and orientation of the mineral crystals in embryonic bone, as evaluated by small-angle X-ray scattering, were found to disappear after birth. To explain these observations, we evaluated the extent of apoptosis in the tibial growth plates using the TUNEL stain. In the growth plate of the p53(-/-) homozygote, there was minimal labeling of the hypertrophic layer. Since the p53(-/-) TUNEL stain pattern at 17 days was very similar to the pattern of labeling of the p53(+/+) at 15 days, we concluded that the growth defect reflected a delay in cartilage maturation rather than a change in chondrocyte phenotype. On this basis, we predict that after birth, in mice that survive, differences in bone length would become minimal, and at maturity, the length of the long bones of (+/+) and (-/-) mice would be similar.

An amelogenin gene defect associated with human X-linked amelogenesis imperfecta

Dental enamel is a product of ameloblast cells, which secrete a mineralizing organic matrix, composed primarily of amelogenin proteins. The amelogenins are thought to be crucial for development of normal, highly mineralized enamel. The X-chromosomal amelogenin gene is a candidate gene for those cases of amelogenesis imperfecta, resulting in defective enamel, in which inheritance is X-linked. In this report, a kindred is described that has a C to A mutation resulting in a pro to thr change in exon 6 of the X-chromosomal amelogenin gene in three affected individuals, a change not found in unaffected members of the kindred. The proline that is changed by the mutation is conserved in amelogenin genes from all species examined to date.

Regulation of amelogenin gene expression

The amelogenins are found uniquely in enamel, where they constitute the predominant class of secreted matrix proteins and where they play a fundamental role in normal enamel formation. To better understand the high level of tissue-specific expression, we cloned the bovine X and Y chromosomal amelogenin genes and the murine amelogenin gene and determined the DNA sequences for the regions upstream of the transcription start sites. We observed segments of strong homology among species, and identified consensus sequences for the binding of various transcription factors, including the glucocorticoid receptor, AP1, RXR and p53. Although specific sis-elements conferring enhanced transcription have not yet been identified, elements have been localized that have silencing effect in non-ameloblast cells. Conserved sequences are likely to be involved in tissue-specific expression. Transgenic mouse studies have shown that 3.5 kb of upstream region is sufficient but 900 bp is insufficient for specific expression in vivo. Alternative splicing of the primary transcript is an effective mechanism for generating molecular heterogeneity. Amelogenin genes contain seven exons, and exons 3, 4, 5 and most of 6 can be deleted by alternative splicing. However, the pattern of exon splicing varies according to the species, and skipping of bovine exon 3 appears to be developmentally regulated. It will be important to determine whether the relative amounts of translation products differ among species as do the mRNAs, and to correlate the various protein structures with function. These findings also suggest that the regulation of amelogenin gene expression is complex and takes place at several levels.

Albumin gene expression during mouse odontogenesis

Albumin protein is present in developing teeth of several species. Oligomer primers and cRNA probes specific for albumin were designed to perform RT-PCR, and for in situ hybridization, respectively. In situ hybridization failed to reveal albumin expression in any tooth cells, however, albumin PCR products were amplified from tissues adhering to the roots of developing teeth from four-week-old mice. It is concluded that this source is not the primary source of albumin protein found in developing enamel, because of the location and level of expression of albumin mRNA in periodontal tissue.

Analysis of amelogenin mRNA during bovine tooth development

The amelogenins are highly conserved enamel-matrix proteins that are essential for proper mineral formation. 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 of exon 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 amelogenin 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.

Bovine enamel organ cells express tissue non-specific alkaline phosphatase mRNA

Alkaline phosphatase (AP) is expressed at high levels in all mineralizing tissues, and the isoform identified in developing enamel has biochemical properties similar to that found in bone. While the bone AP is referred to as the liver/bone/kidney or tissue non-specific (TNS) form, other APs are highly specific for tissue of expression. To determine unequivocally the AP isoform made by enamel organ cells, we constructed a fetal bovine enamel organ cDNA library, which yielded eight AP cDNA clones. In each case, the DNA sequence was homologous to the partial cDNA reported for bovine kidney AP (Garattini et al., 1987). It is concluded that enamel organ cells express the TNS-AP isoform. The extended 3' untranslated region of the cDNA has considerable homology to human TNS-AP, and the conservation of sequence suggests that the 3' end may have a role in post-transcriptional regulation of expression.

Analysis of amelogenin proteins using monospecific antibodies to defined sequences

Amelogenins are the predominant proteins found in the developing enamel matrix and are believed to play a crucial role in normal mineralization. Although the amelogenin gene is found as a single copy in all species in which it has been examined, multiple amelogenin polypeptides ranging in size from 5 to 25 kDa are obtained upon extraction of developing enamel matrix, making identification and characterization of individual components difficult. This heterogeneity may be ascribed to transcription of divergent genes located on the X and Y chromosomes, alternative splicing of the primary transcripts, physiologic degradative processing, and artefactual degradation. In order to characterize individual components, antibodies were produced to the following peptides: (1) QPLQPMQPMQPLQPLQPL (corresponding to the repeat sequence encoded only in the bovine X chromosome gene), (2) IRHPPLPP (corresponding to a unique sequence generated by alternative splicing found in leucine-rich amelogenin peptide (LRAP), (3) LPDLPLEAWPATDKTKREEVD corresponding to the amelogenin carboxy-terminus. Amelogenin proteins obtained from fetal bovine molars were subjected to SDS PAGE and Western electrotransfer, and immuno-ultrastructural analysis. These analyses demonstrated that: (1) the distribution of amelogenin polypeptides isolated from male fetuses differed appreciably from that of females, (2) the LRAP junctional peptide sequence can be specifically identified, and (3) the LRAP peptide can be immunolocalized in the enamel matrix of both males and females.

Analysis of the regulatory region of the bovine X-chromosomal amelogenin gene

The amelogenin proteins, which are crucial for normal enamel mineral formation, are secreted by ameloblasts during development of tooth enamel. In order to better understand the mechanisms involved in regulation of expression of the amelogenin genes, the bovine X-chromosomal amelogenin gene was cloned and a 3.5 KB fragment upstream of exon 1 was inserted into a beta galactosidase (beta gal) expression vector for production of transgenic mice. When tissues from these mice were treated with Xgal, a substrate for beta gal, only ameloblasts and some of the adjacent stratum intermedium cells contained blue stain. To obtain further information concerning regulation of expression, the 3.5 KB amelogenin gene fragment was evaluated in transfection experiments. Nonoverlapping 1.9 and 1.5 KB fragments of the upstream region were subcloned separately into a vector that contains the SV40 promoter and the CAT reporter gene. Each amelogenin gene fragment was able to suppress CAT activity driven by the heterologous SV40 promoter in transfected HeLa cells. We theorize that each of these gene fragments contains regulatory elements important for the tissue-specific and developmentally-regulated pattern of expression of the X-chromosomal amelogenin gene.

A beta-galactosidase expression vector for promoter analysis

An expression vector plasmid, designated pMCS beta gal, was constructed to contain a multiple cloning site for insertion of gene fragments with potential regulatory function. This plasmid was designed to test promoter activity in transgenic mice, since digestion with Pst I will release all vector sequences, which could inhibit transgene expression in vivo (Jaenisch, 1988). When this vector, containing the amelogenin promoter, was used to make transgenic mice, expression was tissue specific and developmentally regulated similar to the endogenous gene (Chen et al., 1994). In addition, the herpes simplex virus (HSV) thymidine kinase (TK) minimal promoter was inserted into pMCS beta gal to produce pTK beta gal, leaving an Sph I upstream site available for insertion of gene fragments with potential enhancer or silencer function, which can be assayed following transfection into cultured cells.

Regulation of amelogenin gene expression during tooth development

The amelogenins are the predominant matrix proteins in developing enamel and are crucial for proper enamel mineralization. Transgenic mice were constructed in order to identify the segment of the amelogenin gene required for specific expression in enamel organ cells. A 3.5 kb fragment of the bovine X-chromosomal amelogenin gene that includes a TATA box, the transcription initiation site, and 32 bp of exon 1 was linked to the beta galactosidase gene and injected into fertilized mouse eggs. Newborn transgene positive mice expressed beta galactosidase activity in developing teeth treated with the chromogenic substrate Xgal. Foci of ameloblasts were positive in newborn mice; stain intensity and number of positive ameloblasts increased in 1-day and 2-day postnatal mice. Some of the adjacent stratum intermedium cells also were positive in the later stages. Targeting of the transgene to the enamel organ was specific; the only other cells observed to be positive were macrophages, which have endogenous beta galactosidase activity.

Bovine amelogenin message heterogeneity: alternative splicing and Y-chromosomal gene transcription

The amelogenins are the most abundant proteins in developing tooth enamel. Previous analyses have demonstrated that transcriptionally active genes encoding the proteins are located on both the bovine X and the bovine Y chromosomes. We report here the cloning and sequence analysis of the Y-chromosomal gene and corresponding cDNA. The Y-specific mRNA encodes a translation product in which a 21 amino acid domain has been deleted, relative to the X-specific amelogenin, resulting in loss of a structure tentatively described as a beta-spiral. There are also 13 single amino acid differences compared to the X-specific amelogenin. In addition, we have cloned and sequenced an X-chromosomal alternatively spliced amelogenin cDNA that encodes a 43 amino acid amelogenin primary translation product. Hydrophobicity analysis indicates that all analyzed amelogenin proteins have a mean hydrophilic character and the two peptides translated from alternatively spliced messages have significant increases in percentage of hydrophobic amino acids.

Odontogenic tumors in mice carrying albumin-myc and albumin-rats transgenes

Odontogenic tumors that produce abnormal tooth-like structures are repeatedly observed in mandibles of mice that carry both albumin-myc and albumin-ras transgenes. The earliest lesions appear among the periodontal ligament mesenchymal cells, but later lesions include an epithelial component. Subsequent tumor development recapitulates the process of normal tooth formation, which requires multiple sequential cell signals, and results in cell differentiation, matrix secretion, and mineralization. Tumor cells with epithelial morphology produce ras oncoprotein, consistent with an epithelial origin of these tumors. As albumin regulatory sequences direct oncogene expression in these mice, our findings also suggest that some of the albumin present in normal teeth may be locally produced and have a role in tooth mineral formation. The reproducibility of this phenotype makes these mice an excellent model for studies of both normal and neoplastic odontogenesis.

Structure and function of the B and D genes of the Actinobacillus actinomycetemcomitans leukotoxin complex

The Actinobacillus actinomycetemcomitans leukotoxin gene complex, consisting of four genes, has been cloned and the sequence of the AaLtC and AaLtA genes reported. The present paper details the sequences of the AaLtB and AaLtD genes which, like AaLtC and AaLTA, are also homologues of genes found in other cytolytic toxin complexes of several other Gram-negative bacterial pathogens. When tested in a recombinant expression system, the AaLtB and/or AaLtD genes are required for the translocation and insertion of the A. actinomycetemcomitans leukotoxin (AaLtA) into the cell membrane of Escherichia coli.

Production of recombinant human tropoelastin: characterization and demonstration of immunologic and chemotactic activity

Tropoelastin cannot readily be prepared in quantity from natural sources and this has limited research in several important areas including structure/function relationships and fiber assembly. In order to eliminate this limitation, human tropoelastin has been expressed in a recombinant bacterial system and the protein has been highly purified. The size, amino acid composition, and sequence of the amino terminus of the recombinant tropoelastin (rTE) all agree with values predicted by the nucleotide sequence of the cDNA used in the expression vector. The rTE exhibits cross-reactivity with antibodies directed against a mixture of peptides derived from human elastin as well as antibody against a defined peptide located at the carboxy terminus of the protein. In addition, the rTE is chemotactic for fetal calf ligament fibroblasts. These results suggest that rTE could be a useful reagent for many types of studies.

Analysis of the Actinobacillus actinomycetemcomitans leukotoxin gene. Delineation of unique features and comparison to homologous toxins

Actinobacillus actinomycetemcomitans leukotoxin has been implicated as a virulence factor in human infections. To initiate delineation of leukotoxin structure/function relationships, molecular cloning of the leukotoxin gene was carried out. When an A. actinomycetemcomitans genomic DNA library in lambda EMBL3 was screened using a 1.3-kilobase pair restriction fragment containing a portion of the leukotoxin gene, 13 positive recombinants were identified. One recombinant, designated lambda OP8, containing a 16-kilobase pair insert was selected for detailed study. Lysates from lambda OP8, but not control lysates, exhibited leukotoxic activity with target cell specificity identical to the native toxin. Western blots identified the recombinant-produced toxin as a 125-kDa protein doublet identical in mobility to the native toxin. Restriction enzyme and extensive DNA analyses demonstrated that the leukotoxin gene showed strong homology to two other toxins produced by Escherichia coli and Pasteurella haemolytica. As in the other two species, the A. actinomycetemcomitans toxin is contained in a cluster of four genes in which the A gene encodes the toxin and the products of the B, C, and D genes are involved in posttranslational modification of the toxin and its membrane insertion and secretion. The target cell specificity of the A. actinomycetemcomitans toxin differs from the other two toxins and is restricted to human and some non-human primate cells of the monomyelocytic lineage. The A. actinomycetemcomitans leukotoxin is not secreted but remains associated with the bacterial membrane, possibly through a hydrophobic domain at the carboxyl terminus which distinguishes it from the E. coli and P. haemolytica toxins.

Identification and expression of the Actinobacillus actinomycetemcomitans leukotoxin gene

The leukotoxin produced by the oral bacterium Actinobacillus actinomycetemcomitans has been implicated in the pathogenesis of juvenile periodontitis. In order to elucidate the structure of the leukotoxin, molecular cloning of the leukotoxin gene was carried out. A DNA library of A. actinomycetemcomitans, strain JP2, was constructed by partial digestion of genomic DNA with Sau3AI and ligation of 0.5 to 5.0 kilobase pair fragments into the Bam HI site of the plasmid vector pENN-vrf. After transformation into E. coli RR1 (lambda cI857), the clones were screened for the production of A. actinomycetemcomitans leukotoxin with polyclonal antibody. Six immunoreactive clones were identified. The clones expressed proteins which ranged from 21-80 kilodaltons, and the clone designated pII-2, producing the largest protein was selected for further study. Antibodies eluted from immobilized pII-2 protein also recognized the native A. actinomycetemcomitans leukotoxin molecule indicating that both molecules shared at least one epitope. DNA sequence analysis demonstrated that there are regions of significant amino acid sequence homology between the cloned A. actinomycetemcomitans leukotoxin and two other cytolysins, Escherichia coli alpha-hemolysin and Pasteurella haemolytica leukotoxin, suggesting that a family of cytolysins may exist which share a common mechanism of killing but vary in their target cell specificity.

Cell cycle dependent genes inducible by different mitogens in cells from different species

A number of genes and cDNA sequences (including at least four oncogenes) are known to be expressed in a cell cycle-dependent manner, i.e. the levels of specific mRNAs vary with the phases of the cell cycle. In order to explore the significance of some of these sequences in the mitogenic response, we have investigated the expression of 8 cell cycle-dependent sequences (plus two control sequences, not expressed in a cell cycle-dependent manner) under a variety of conditions. These conditions included cells of different types, from different species, stimulated to proliferate by different mitogens. The genes (or sequences) studied included five cDNA clones whose sequences are preferentially expressed in early G1, i.e. two cDNA clones inducible by platelet-derived growth factor (JE-3 and KC-1), and three cDNA clones inducible by serum (2A9, 2F1, 4F1); and three oncogenes (c-myc, c-rasHa and p53) whose expression is known to be cycle-dependent. All of the tested genes, except 2A9, c-rasHa and the control genes, are expressed in a cell cycle-dependent manner in human peripheral blood mononuclear cells stimulated by phytohemagglutinin and in serum-stimulated mouse and Syrian hamster fibroblasts. The inducibility of these genes by different mitogens in cells of different types and from different species strongly suggests that these genes play a role in cell cycle progression. This conclusion is further supported by the known structural and functional similarities between cell-cycle dependent genes, oncogenes and genes coding for cell-cycle related molecules.

The effect of cycloheximide on the expression of cell cycle dependent genes

We have investigated the inducibility of several cell cycle-dependent genes (plus control sequences, not expressed in a cell cycle-dependent manner) in the presence of cycloheximide, an inhibitor of protein synthesis. The genes studied include: 1) five cDNA clones that are preferentially expressed in the G1 phase of the cell cycle: KC-1, JE-3, 2F1, 4F1 and 2A9; 2) one gene preferentially expressed in late G1/S phase: histone H3; and 3) the cell cycle-dependent oncogene p53. All the genes studied are induced by serum even in the presence of cycloheximide. Previous results in the literature have shown that 2 other oncogenes, c-myc and c-fos, can be induced by growth factors in the presence of cycloheximide. Together with our results, these findings indicate that protein synthesis is not required for the induction of at least nine cell cycle genes by growth factors.

Cell-cycle-specific genes differentially expressed in human leukemias

Three cDNA clones isolated from Syrian hamster cells (p4F1, p2F1, and p2A9) contain sequences that are preferentially expressed in the G1 phase of the cell cycle. The expression of these sequences was investigated in human peripheral blood cells from normal individuals and from patients with leukemia. The expression of p4F1 and p2F1 is clearly dependent on the cell cycle in peripheral blood mononuclear cells stimulated to proliferate with phytohemagglutinin; the p2A9 sequences cannot be clearly detected in human lymphocytes but are expressed in a cell-cycle-dependent manner in human diploid fibroblasts (WI-38). These genes also show different levels of expression in lymphoid and myeloid leukemias. The highest level of expression for p2A9 is found in patients with chronic myelogenous leukemia, and the lowest in patients with chronic lymphocytic leukemia. For p2F1 and p4F1, the highest levels of expression are found in chronic and acute myelogenous leukemia. At least two other cell-cycle genes are not expressed at detectable levels in human leukemias. These findings suggest that the activation of cell-division-cycle genes might contribute, like cellular oncogenes, to the phenotype of human malignancies and that, perhaps, new oncogenes could be found by identifying and isolating genes whose expression is dependent on the cell cycle.

Expression of thymidine kinase and dihydrofolate reductase genes in mammalian ts mutants of the cell cycle

Thymidine kinase and dihydrofolate reductase mRNA levels and enzyme activities were determined in two temperature-sensitive cell lines, tsAF8 and ts13, that growth arrest in the G1 phase of the cell cycle at the restrictive temperature. The levels of thymidine kinase mRNA and enzyme activity increased markedly in both cell lines serum stimulated from quiescence at the permissive temperature. At the nonpermissive temperature, the levels of thymidine kinase mRNA and enzyme activity remain at the low levels of quiescent G0 cells. The levels of dihydrofolate reductase mRNA as well as the enzyme activity also increase when both cell lines are serum stimulated at the permissive temperature. When ts13 cells are serum stimulated at the nonpermissive temperature dihydrofolate reductase enzyme activity declines rapidly and dihydrofolate reductase mRNA is below detectable levels. On the contrary, when tsAF8 cells are serum stimulated at the nonpermissive temperature dihydrofolate reductase enzyme activity increases and mRNA levels are detectable slightly above G0 levels, even though the cells are blocked in the G1 phase. Studies with 2 other cDNA clones (one with an insert whose expression is cell cycle dependent and the other with an insert whose expression is not cell cycle dependent) indicate that the results are not due to aspecific toxicity or the effect of temperature. We conclude that the expression of different genes is affected differently by the ts block in G1, even when these genes are all growth-related.

Macrophage-mediated mitogenic suppression induced in mice of the C3H lineage by a vaccine strain of Salmonella typhimurium

Salmonella typhimurium, strain SL3235, an avirulent organism, has been used as a live vaccine in mice of the C3H lineage and has been found to confer high levels of protection. In the present study, it was found that intraperitoneal injection of approximately 5 X 10(5) live SL3235 induced potent suppression of spleen cell mitogenic responses to a panel of B- and T-cell mitogens in the Salmonella-hypersusceptible C3H/HeJ and C3HeB/FeJ, and the inherently resistant C3H/HeNCrlBR mice. Maximal suppression (greater than 99%) was seen at 1 week, and was still significant but waning (50%) at 3 weeks postimmunization. In contrast, cells of mice receiving acetone-killed cells were not suppressed. Removal of macrophages, but not T or B cells, restored responsiveness, indicating that suppression was macrophage mediated. Prostaglandins were not the major mediator of suppression, as in vitro administration of indomethacin failed to abrogate suppression. As mitogenic suppression occurred in mice with high levels of Salmonella immunity, the suppression is interpreted as a marker of a powerful immunomodulatory process induced by live cells, rather than as an indication of poor immune status of the host.

Analysis of nutritional shift-up of Streptococcus faecium

Three-dimensional reconstruction methods were applied to electron micrographs of Streptococcus faecium to study the initiation of cell wall growth sites during a nutritional shift-up experiment. Upon lowering the mass doubling time from 76 to 33 min by the addition of excess glutamate, the formation of new cell wall growth sites was studied in relation to other growth parameters (autolytic capacity, cell number, mass, RNA, DNA and peptidoglycan). The findings from these studies, to be described below, support a model in which new sites are introduced when cells grow to a relatively constant, growth-rate-independent size, while the rate at which sites form and grow increases with growth rate. In this model, chromosome synthesis does not regulate the formation of new sites of cell wall growth, but existing sites cannot be completed until rounds of chromosome synthesis are completed.

Analysis of initiation of sites of cell wall growth in Streptococcus faecium during a nutritional shift

Three-dimensional reconstruction methods were applied to electron micrographs of Streptococcus faecium to study the initiation of cell wall growth sites during a nutritional shift experiment. Upon lowering the mass doubling time from 76 to 33 min by the addition of excess glutamate, the formation of new cell wall growth sites accelerated above the old steady-state rate at about the same time (10 to 15 min) as did mass, RNA, protein, cell numbers, and autolytic capacity but considerably before DNA (30 min) and peptidoglycan (20 min) synthesis did. During the shift, the average range of cell volumes over which new wall growth sites were introduced did not change significantly. However, upon the shift there was an increase in the frequency of cells having new sites, which was due to the faster-growing cells initiating more new sites in peripheral locations before division. After a transition period, the number of new sites per milliliter of culture increased at a rate that paralleled that of the culture mass. These findings support a model in which new sites are introduced when cells grow to a relatively constant, growth rate-independent size, while the rate at which sites form and grow increases with the growth rate. In this model, chromosome synthesis does not regulate the formation of new sites of cell wall growth, but existing sites cannot be completed until rounds of chromosome synthesis are completed.

Cell-cycle-specific cDNAs from mammalian cells temperature sensitive for growth

A library of double-stranded cDNA was constructed from ts13 cells, a G1-specific temperature-sensitive hamster cell line. The cDNAs, cloned into pBR322, were prepared from poly(A)+ mRNA isolated from ts13 cells 6 hr after serum stimulation at the permissive temperature of 34 degrees C. Differential screening of the library with G1-specific and G0-specific single-stranded cDNA probes prepared from the same cells identified five cDNA clones whose sequences were preferentially expressed in G1. Levels of RNA complementary to these clones were 3- to 6-fold higher in G1 than in other phases of the cell cycle. When ts13 cells were arrested in G1 at the restrictive temperature of 39.6 degrees C, the levels of RNA complementary to p13-2A9 and p13-4F1 were as high as 10 times that found in a resting population, while the expression of sequences complementary to p13-2A8 did not significantly change from levels found in G0. RNA and Southern gel blot analysis suggest that these cell-cycle-specific clones represent either low copy or moderately repetitive gene sequences. Results with another ts mutant of the cell cycle, tsAF8, which is a ts mutant of RNA polymerase II, showed that these cell-cycle-specific sequences have a rapid turnover. The use of G1-specific ts mutants of the cell cycle provides an approach to determine which cell-cycle-dependent genes are most relevant to cell cycle progression.

Cell wall assembly during inhibition of DNA synthesis in Streptococcus faecium

Growth sites which are bounded by raised wall bands can be observed in electron micrographs of replicas of Streptococcus faecium. When mitomycin C was added to an exponential-phase culture doubling in mass every 64 min, DNA synthesis was inhibited, and eventually cell division stopped. The growth sites formed before and after inhibition of DNA synthesis enlarged until they contained about 0.25 micron3 of cell volume, at which point they ceased to increase in size. When these sites approached this 0.25-micron3 limit, new sites were initiated; this result had also been observed in untreated cells undergoing a large range of exponential-phase mass doubling times. Thus, regardless of whether chromosome replication is inhibited or uninhibited, sites have the same finite capacity to enlarge to about 0.25 micron3, and when this capacity is reached, new sites are initiated. Although initiation of new growth sites seems to be independent of normal chromosome replication, these results confirm previous studies showing that chromosome replication is necessary for the terminal events of growth site development which result in the division of a site into two separate poles. Two classes of models for the regulation of growth site initiation are discussed.

Initiation of wall assembly sites in Streptococcus faecium

In electron micrographs of replicas of Streptococcus faecium, sites of wall growth are located between pairs of raised equatorial bands. Analysis of cells taken from cultures with mass doubling times between 30 and 125 min indicates that rounds of wall synthesis are initiated at a time close to division, which is temporally unrelated to the initiation or termination of chromosome replication. Growth sites are initiated at a relatively constant volume independent of growth rate when the volume contained within the two segments of wall adjoining an equatorial band marker approaches ca. 0.26 micrometer 3.