Strict co-linearity of genetic and protein folding domains in an intragenically duplicated rat lens gamma-crystallin gene

alphaB-crystallin: a hybrid solid-state/solution-state NMR investigation reveals structural aspects of the heterogeneous oligomer

Atomic-level structural information on alphaB-Crystallin (alphaB), a prominent member of the small heat-shock protein family, has been a challenge to obtain due its polydisperse oligomeric nature. We show that magic-angle spinning solid-state NMR can be used to obtain high-resolution information on an approximately 580-kDa human alphaB assembled from 175-residue 20-kDa subunits. An approximately 100-residue alpha-crystallin domain is common to all small heat-shock proteins, and solution-state NMR was performed on two different alpha-crystallin domain constructs isolated from alphaB. In vitro, the chaperone-like activities of full-length alphaB and the isolated alpha-crystallin domain are identical. Chemical shifts of the backbone and C(beta) resonances have been obtained for residues 64-162 (alpha-crystallin domain plus part of the C-terminus) in alphaB and the isolated alpha-crystallin domain by solid-state and solution-state NMR, respectively. Both sets of data strongly predict six beta-strands in the alpha-crystallin domain. A majority of residues in the alpha-crystallin domain have similar chemical shifts in both solid-state and solution-state, indicating similar structures for the domain in its isolated and oligomeric forms. Sites of intersubunit interaction are identified from chemical shift differences that cluster to specific regions of the alpha-crystallin domain. Multiple signals are observed for the resonances of M68 in the oligomer, identifying the region containing this residue as existing in heterogeneous environments within alphaB. Evidence for a novel dimerization motif in the human alpha-crystallin domain is obtained by a comparison of (i) solid-state and solution-state chemical shift data and (ii) (1)H-(15)N heteronuclear single quantum coherence spectra as a function of pH. The isolated alpha-crystallin domain undergoes a dimer-monomer transition over the pH range 7.5-6.8. This steep pH-dependent switch may be important for alphaB to function optimally (e.g., to preserve the filament integrity of cardiac muscle proteins such as actin and desmin during cardiac ischemia, which is accompanied by acidosis).

Three murine cataract mutants (Cat2) are defective in different gamma-crystallin genes

A number of murine cataract mutations have been localized to chromosome 1 close to the gamma-crystallin gene cluster (Cryg) (Everett et al., 1994, Genomics 20: 429-434; Löster et al., 1994, Genomics 23: 240-242). Based on the size of the mapping or allelism tests they have not been shown to be genetically distinct and have been assigned to locus symbol Cat2. Here we assign three mutations to the respective gamma-crystallin gene. Using a systematic candidate gene approach to analyze the entire Cryg cluster, an A-->G transition was found in exon 2 of Cryga for the ENU-436 mutation and is designated Cryga1Neu. The mutant allele Crygbnop (formerly Cat2(nop)) is caused by a replacement of 11 bp by 4 bp in the third exon of Crygb, while a C-->G transversion in exon 3 of Cryge has been found for the Cryget (formerly Cat2(t)) mutation. For the mutation Cryga1Neu, an Asp-->Gly exchange is deduced, whereas the mutations Crygbnop and Cryget lead to the formation of in-frame stop codons and give rise to truncated proteins of 144 and 143 amino acids, respectively. The effects of the mutations upon gamma-crystallin structure are likely to be quite different. The Cryga1Neu mutation is expected to affect the link between Greek-key motifs 2 and 3, whereas both Crygbnop and Cryget mutations are supposed to truncate the fourth Greek-key motif. All three mutations are predicted to alter protein folding of the gamma-crystallins and result in lens cataract, but the phenotype for each is quite distinctive.

AIM1, a novel non-lens member of the betagamma-crystallin superfamily, is associated with the control of tumorigenicity in human malignant melanoma

AIM1 is a novel gene whose expression is associated with the experimental reversal of tumorigenicity of human malignant melanoma. The predicted protein product of the major 4.1-kb transcript shows striking similarity to the betagamma-crystallin superfamily. All known members of this superfamily contain two or four characteristic motifs arranged as one or two symmetrical domains. AIM1, in contrast, contains 12 betagamma motifs, suggesting a 6-domain structure resembling a trimer of beta- or gamma-crystallin subunits. The structure of the AIM1 gene shows remarkable similarity to beta-crystallin genes, with homologous introns delineating equivalent protein structural units. AIM1 is the first mammalian member of the betagamma superfamily with a primarily non-lens role. Other parts of the predicted AIM1 protein sequence have weak similarity with filament or actin-binding proteins. AIM1 is a good candidate for the putative suppressor of malignant melanoma on chromosome 6, possibly exerting its effects through interactions with the cytoskeleton.

Sequence analysis of the beta B2-crystallin cDNA of hamster containing a domain conserved among vertebrates

The cDNA sequence of the beta B2-cry was determined from hamster (Mesocricetus auratus) and compared to the corresponding genes of bovine, frog, chicken, human, mouse and rat. Multispecies comparison demonstrated high homology between the hamster, rat and mouse gene, but larger distances to man, bovine, chicken and frog. There is striking identity within a strech of 36 deduced amino acids (aa) between the Greek key motif 3 and part of motif 4. This 36-aa domain contains a putative phosphorylation site for protein kinase C and is highly conserved among all known basic beta B-Cry; however, it can neither be detected in the acidic beta A-nor in the gamma-Cry.

Genetic aspects of embryonic eye development in vertebrates

The vertebrate eye comprises tissues from different embryonic origins, e.g., iris and ciliary body are derived from the wall of the diencephalon via optic vesicle and optic cup. Lens and cornea, on the other hand, come from the overlying surface ectoderm. The timely action of transcription factors and inductive signals ensure the correct development of the different eye components. Establishing the genetic basis of eye defects has been an important tool for the detailed analysis of this complex process. One of the main control genes for eye development was discovered by the analysis of the allelic series of the Small eye mouse mutants and characterized as Pax6. It is involved in the interaction between the optic cup and the overlaying ectoderm. The central role for Pax6 in eye development is conserved throughout the animal kingdom as the murine Pax6 gene induces ectopic eyes in transgenic Drosophila despite the obvious diverse organization of the eye in the fruit fly compared to vertebrates. In human, mutations in the PAX6 gene are responsible for aniridia and Peter's anomaly. In addition to Pax6, other mutations affecting the interaction of the optic cup and the lens placode have been documented in the mouse. For the differentiation of the retina from the optic cup several genes are responsible: Mi leads to microphthalmia, if mutated, and encodes for a transcription factor, which is expressed in the melanocytes of the pigmented layer of the retina. In addition, further genes are implicated in the correct development of the retina, e.g., Chx10, Dlx1, GH6, Msx1 and -2, Otx1 and -2, or Wnt7b. Mutations within the retinoblastoma gene (RB1) are responsible for retinal tumors. Knock-out mutants of RB1 exhibit a block of lens differentiation prior to the retinal defect. Besides the influence of Rb1, the lens differentiates under the influence of growth factors (e.g., FGF, IGF, PDGF, TGF), and specific genes become activated encoding cytoskeletal proteins (e.g., filensin, phakinin, vimentin), structural proteins (e.g., crystallins) or membrane proteins (e.g., Mip). The optic nerve originates from the neural retina; ganglion cells grow to the optic stalk, forming the optic nerve. Its retrograde walk to the brain through the rudiment of the optic stalk depends on the correct Pax2 expression.

Computer prediction of the exon-intron structure of mammalian pre-mRNAs

A novel approach to the problem of prediction of protein-coding regions is suggested. This approach combines the site prediction methods to predict splicing sites and the global coding region prediction methods to choose the best variant of spliced mRNA. One of the advantages of the suggested algorithm is that the resulting mRNA or protein sequence may then be immediately analyzed further. The true mRNA either coincides with the predicted one or ranks high in the list of variants. In the latter situation the predicted mRNA usually differs from the true one in only one or two of several exons. The combined approach allows the use of a priori information (e.g. the putative protein length or the number of exons). It is possible to use additional parameters not considered here, such as the preferred lengths of exons and introns, and particularly the preferred position of introns in the reading frame and the preferred codon position of exon termini.

Different evolution rates within the lens-specific beta-crystallin gene family

We have determined the sequence of a rat beta A3/A1-crystallin complementary DNA (cDNA) clone and the (partial) sequence of the human beta B3-crystallin gene. Calculation of the ratio of silent to nonsynonymous substitution between orthologous beta A3/A1-, beta B3-, and other beta- and gamma-crystallin sequences revealed that the region encoding the two globular domains of the beta A3/A1-crystallin sequence is the best conserved during evolution, much better than the corresponding region of the beta B1-, beta B3-, or the gamma-crystallin sequences, and even better (at least in the rodent/frog comparison) than the well-conserved alpha A-crystallin sequence. Remarkably, the rate of change of the beta A3/A1-crystallin coding sequence does not differ in the rodent and primate lineages, in contrast with previous findings concerning the evolution rates of the alpha A- or gamma-crystallin sequences in these two lineages. Comparison of the regions that encode the four motifs of the beta-crystallin between orthologous mammalian sequences showed that the extent of nonsynonymous substitution in each of these four homologous motif regions is the same. However, when the orthologous beta-crystallin genes of more distantly related species (mammals vs chicken or frog) are compared, the extent of non-synonymous substitution is higher in the regions encoding the external motifs I and III than in the regions encoding the internal motifs II and IV. This phenomenon is also observed when paralogous members of the beta/gamma-crystallin supergene family are compared.

The gamma-crystallin gene families: sequence and evolutionary patterns

The gamma-crystallin proteins consist of two topologically equivalent domains, each built up out of two similar motifs. They are encoded by a gene family, which already contained five members before the divergence of rodents and primates. A further gene duplication took place in each lineage. To analyze the pattern of evolution within this gene family, the coding sequences of six human genes, six rat genes, and four mouse genes were compared. Between species, a uniform rate of evolution of all regions of the protein is seen. The ratio of synonymous to nonsynonymous substitution in the human/rat or human/mouse comparison is much lower than the ratio when rat and mouse are compared indicating that the gamma-crystallin proteins are better conserved in the rodent lineage. Within species, the regions encoding the two external motifs I and III of the protein show a greater extent of nonsynonymous substitution than the regions encoding the two internal protein motifs II and IV. The low extent of synonymous substitution between the second exons (encoding motifs I and II) of the rat gamma-crystallin genes suggests the frequent occurrence of gene conversion. In contrast, a high extent of synonymous substitution is found in exon 3 (encoding motifs III and IV) of the rat genes. The same phenomenon is seen within the human gene family. The frequencies of occurrence of the various dinucleotides deviate less from those predicted from the frequencies of occurrence of each individual nucleotide in the second exons than in the third exons. The sequences of the third exons are significantly depleted in CpG, ApA, and GpT and enriched in CpT and GpA.

Evolutionary and functional relationships between the basic and acidic beta-crystallins

beta-Crystallins are complex oligomers composed of many related subunits. In order to understand their interactions we have built molecular models of several bovine beta-crystallins, based on their sequence similarity to the well-defined gamma-II crystallin structure, using interactive computer graphics techniques. Their common origin with gamma-crystallin is displayed in both the retention of four-fold sequence repeats of critical residues involved with stabilizing a folded beta-hairpin and the conservation of core-filling hydrophobic side-chains. The beta-crystallins have been built as bilobal molecules with each domain composed of two 'Greek key' motifs which associate about an approximate two-fold axis to form beta-sheets. The beta-crystallin sequences have previously been shown to comprise two families, the basic and acidic subunits, which have extensions of sequence. The three-dimensional models show how the two families appear to stabilize the folded beta-hairpin in the N- and C-terminal domains in ways which suggest that they have diverged from a common ancestor in different ways. Acidic beta-crystallins, like gamma-crystallins, have a regular array of charges on their N-terminal domain which has been interrupted in basic beta-crystallins by hydrophobic residues which may be related to the presence of a C-terminal extension. beta-Crystallins are more highly charged than gamma-crystallins although their charge density is higher in certain regions of the N-terminal domain, particularly in beta B1-crystallin. beta-crystallins also differ from gamma-crystallins in the virtual absence of core-filling sulphydryl groups whereas they have numerous sulphur-containing side-chains together with tryptophan and histidine rings protruding from the globular domains, particularly in the acidic subunits. The burial of these residues in subunit contacts is consistent with their spectroscopic and electrostatic properties. Protein subunit aggregation commonly occurs through hydrophobic interaction or beta-sheet extension. Analysis of the subunit surfaces has identified an N-terminal hydrophobic region common to beta B1 and beta B2 whereas a C-terminal hydrophobic loop region is common to beta B1 and beta A1 and may be correlated with their association properties. It is suggested that the polar C-terminal domain of beta B2 contributes towards the solubility of higher aggregates by interactions involving beta-sheet structure.

A new 5'-non-coding region for human placental insulin-like growth factor II mRNA expression

A human placenta cDNA library was screened for insulin-like growth factor II (IGF-II). Four clones were selected, which exhibited an IGF-II cDNA coding sequence identical to those previously described for human adult liver IGF-II cDNA. Extensive sequence diversity was observed in the 5'-non-coding region, probably resulting from differential intron splicing.

The mouse eye lens obsolescence (Elo) mutant: studies on crystallin gene expression and linkage analysis between the mutant locus and the gamma-crystallin genes

Previous studies showed that the mouse Elo (eye lens obsolescence) mutation is located on chromosome 1, at a site near the Len-1 locus, which is defined by a set of polymorphic gamma-crystallin proteins. To investigate further the relationship between Elo and the gamma-crystallins, we have examined the steady-state levels of gamma-crystallin transcripts in normal and mutant eyes and analyzed the linkage relationship between the Elo locus and the gamma-crystallin genes. Our data showed that, while gamma-crystallin mRNA levels are preferentially reduced in the mutant eyes, the mutation does not seem to map within the gamma-crystallin gene cluster. The distance between Elo and the gamma 6 gene (the most proximal gamma-crystallin gene member) is estimated to be 1.4 +/- 0.9 cM, whereas that between gamma 6 and the distantly linked gamma 2 gene is 2.7 +/- 1.3 cM. Our data also suggest the possibility of recombination hot spots with the gamma-crystallin gene cluster.

Signals for the selection of a splice site in pre-mRNA. Computer analysis of splice junction sequences and like sequences

To evaluate the importance of the surrounding nucleotide sequence in the selection of a splice site for mRNA, we have carried out computer studies of eukaryotic protein genes whose entire nucleotide sequences were available. A splice site-like sequence that has a significant homology to the consensus splice junction sequences is frequently found within an intron and exon. It is found that the higher the homology of a candidate donor site sequence to the nine-nucleotide consensus sequence, the higher is its probability of being a donor site. For most of the donors, the stability of presumed base-pairing with U1-RNA is higher than that of donor-like sequences, if any, in the adjacent exon and intron. However, homology of a candidate acceptor sequence to the 15-nucleotide consensus is a poor criterion of an acceptor site. The presence of a sequence that could serve as a branch-point 18 to 37 nucleotides before an acceptor does not seem to be critical in distinguishing it from an acceptor-like sequence. For genes of human, rat, mouse and chicken, respectively, nucleotide frequencies around splice junctions of many genes have been calculated. They seem to be different at some positions around a donor site from species to species. The acceptors for these vertebrates have longer pyrimidine-rich regions than the previous consensus sequence. The newly derived nucleotide frequencies were used as the standard to calculate the weighted homology score of a candidate splice site sequence in a gene of the four species. This weighted homology score of the 40 to 60-nucleotide intron-exon sequence is a much better criterion of an acceptor. These results suggest that the most important signal in the selection of a splice resides in the surrounding nucleotide sequence. It is also suggested that the surrounding nucleotide sequence alone is not generally sufficient for the selection.

The alpha 1 and alpha 2 domains of H-2 class I molecules interact to form unique epitopes

Mouse class I antigens are the major targets of cytolytic T lymphocytes in both major histocompatibility complex (MHC)-restricted and allogeneic responses. Considerable evidence has recently accumulated demonstrating that MHC class I molecules encoded by genes whose alpha 1 and alpha 2 coding exons were interchanged are not recognized by T lymphocytes specific for parental class I products. Along with the loss of T-cell reactivity, there is a loss of recognition by some, but not all monoclonal antibodies. In this communication we report that the loss of reactivity by monoclonal antibodies is accompanied by the gain of new epitopes caused by the interaction of alpha 1 and alpha 2 domains. These epitopes are immunodominant. They are the major determinant recognized by polyclonal antisera raised by immunization with L cells transfected with exon-shuffled class I genes. Four new monoclonal antibodies have been produced which recognize at least two separate epitopes caused by the interaction of the alpha 1p and alpha 2d domains.

Complete nucleotide sequence of the chicken alpha A-crystallin gene and its 5' flanking region

The chicken alpha A-crystallin gene and 2.6 kb of its 5' flanking sequence have been isolated and characterized by electron microscopy and sequencing. The structural gene is 4.5 kb long and contains two introns, each approx. 1 kb in length. The first intron divides codons 63 and 64, and the second intron divides codons 104 and 105, as in rodents. There is little indication that the insert exon of rodents (an alternatively spliced sequence) is present in complete form in the chicken alpha A-crystallin gene; small stretches of similarity to this sequence were found throughout the gene. The 5' flanking sequence of the chicken alpha A-crystallin gene shows considerable sequence similarity with other mammalian alpha B-crystallin genes. In addition, one consensus sequence (GCAGCATGCCCTCCTAG) present in the 5' flanking region of the chicken alpha A-crystallin gene was found in the 5' flanking region of most reported crystallin genes.

Linkage between the beta B2 and beta B3 crystallin genes in man and rat: a remnant of an ancient beta-crystallin gene cluster

Human and rat genomic clones containing beta B2- and/or beta B3-crystallin sequences have been isolated and characterized. Both in the human and the rat genome the single-copy beta B3-crystallin gene is linked to a beta B2-crystallin gene. In both species the linked genes, separated by 20 kb in the human and 11 kb in the rat genome, are oriented head-to-tail with respect to transcription. A single copy of the beta B2-crystallin gene is present in the rat genome, in the human genome two copies of this gene are found. The second human copy could as yet not be linked to the beta B2/beta B3-crystallin gene cluster.

Differential regulation of gamma-crystallin genes during mouse lens development

Using gene-specific probes derived from four mouse gamma-crystallin cDNAs, we have examined the regulation of different members of the mouse gamma-crystallin gene family during lens development. Our analysis revealed that, while the different gamma-crystallin genes appear to be coordinately activated during embryogenesis, the steady-state levels of their corresponding transcripts are differentially regulated, resulting in variations in the relative abundance of individual species at different stages of development. This complex pattern of gene regulation presumably accounts for one of the mechanisms determining the spatial distribution of different gamma-crystallins within the lens.

Lens-specific promoter activity of a mouse gamma-crystallin gene

Crystallins are the major water-soluble proteins in vertebrate eye lenses. These lens-specific proteins are encoded by several gene families, and their expression is differentially regulated during lens cell differentiation. Here we show that a cloned mouse gamma-crystallin promoter is active in lens explants derived from 14-day-old chicken embryos but inactive in a variety of cells of non-lens origin. We also show that sequences required for proper utilization of this promoter are contained between nucleotide positions -392 and +47 relative to the transcription initiation site; deletion of sequences from positions -392 to -171 completely abolishes promoter activity. Since chickens do not have gamma-crystallin genes, the expression of a mouse gamma-crystallin promoter in chicken lens cells suggests that different classes of crystallin genes may be regulated by common lens tissue-specific mechanism(s) independent of species.

The triosephosphate isomerase gene from maize: introns antedate the plant-animal divergence

We have cloned and characterized a cDNA and genomic DNA for the triosephosphate isomerase expressed in maize roots. The gene is interrupted by eight introns. If we compare this gene with that for the protein in chicken, which has six introns, we see that five of the introns are at identical places, one has shifted by three codons, and two are totally new. This great matching leads us to conclude that the introns were in place before the plant-animal divergence, and that the parental gene had at least eight introns, two of which were lost in the line that leads to animals.

Concerted and divergent evolution within the rat gamma-crystallin gene family

The nucleotide sequences of six rat gamma-crystallin genes have been determined. All genes have the same mosaic structure: the first exons contain a relatively short (25 to 44 base-pair) 5' non-coding region and the first nine base-pairs of the coding sequence, the second exons encode protein motifs I and II, while protein motifs III and IV are encoded by the third exons. The third exons also contain a 60 to 67-base-pair long 3' non-coding region. In the gamma 1-2 gene, the splice acceptor site of the third exon has been shifted three base-pairs upstream. Hence, the protein product of this gene is one amino acid residue longer. The first introns, though varying in length from 85 to 100 base-pairs, are conserved in sequence. The second introns vary considerably in length (0.9 X 10(3) to 1.9 X 10(3) base-pairs) and sequence. The second exons of the genes show concerted evolution and have undergone multiple gene conversions. In contrast, the third exons show divergent evolution. From the sequences of the third exons, an evolutionary tree of the gene family was constructed. This tree suggests that three of the present genes derive directly from the genes that originated from a tandem duplication of a two-gene cluster. Two duplications of the last gene of the four-gene cluster then yielded the other three genes. Region a' of the third exon, encoding protein motif III, is variable, while the region encoding protein motif IV (b') is constant. We postulate that this variability in region a' is due to a period of radiation after each gene duplication. A comparison of the rat sequences with those of orthologous sequences from other species shows that the variation in region a' is now preserved. Hence, it might specify the specific functional property of each gamma-crystallin protein within the lens.

Intron insertions and deletions in the beta/gamma-crystallin gene family: the rat beta B1 gene

The rat beta B1-crystallin gene is 13.6 kilobases long and contains six exons. The coding region of the gene is divided over five exons. Each functional entity of the protein is encoded by a separate exon except for the carboxyl-terminal extension, which shares the last exon with the fourth protein motif. Exon 2, encoding the amino-terminal extension of the protein, contains two direct repeats with an overall homology of 68% to the rat brain identifier sequence. A copy of the brain identifier sequence is also found in the 3'-flanking region of the gene. The start site of the mRNA was located by S1 nuclease mapping and analysis of the RNA sequence. The 5' end of the gene was shown to be a 27-base-pair noncoding exon, which is separated from the translation start site by 1.36 kilobases of intronic DNA. The 5'-flanking sequence of the beta B1 gene is highly homologous to that of a gamma-crystallin gene.

Structural and evolutionary relationships among five members of the human gamma-crystallin gene family

We have characterized five human gamma-crystallin genes isolated from a genomic phage library. DNA sequencing of four of the genes revealed that two of them predict polypeptides of 174 residues showing 71% homology in their amino acid sequence; the other two correspond to closely related pseudogenes which contain the same in-frame termination codon at identical positions in the coding sequence. Two of the genes and one of the pseudogenes are oriented in a head-to-tail fashion clustered within 22.5 kilobases. All three contain a TATA box 60 to 80 base pairs upstream of the initiation codon and a highly conserved segment of 44 base pairs in length immediately preceding the TATA box. The two genes and the two pseudogenes are similar in structure: each contains a small 5' exon encoding three amino acids followed by two larger exons that correspond exactly to the two similar structural domains of the polypeptide. The first intron varies from 100 to 110 base pairs, and the second intron ranges from 1 to several kilobases, rendering an overall gene size of 1.7 to 4.5 kilobases. At least one of the two pseudogenes appears to have been functional before inactivation, suggesting that their identical mutation was generated by gene conversion.

Complete structure of the hamster alpha A crystallin gene. Reflection of an evolutionary history by means of exon shuffling

The eye lens contains a structural protein, alpha crystallin, composed of two homologous primary gene products alpha A2 and alpha B2. In certain rodents, still another alpha crystallin polypeptide, alpha AIns, occurs, which is identical to alpha A2 except that it contains an insertion peptide between residues 63 and 64. In this paper we describe the complete alpha A crystallin gene that has been cloned from DNA isolated from Syrian golden hamster. Evidence is provided that the alpha A gene is present as a single copy in the hamster genome. The detailed organization of the gene has been established by means of DNA sequence analysis and S1 nuclease mapping, revealing that the gene consists of four exons. The first exon contains the information for the 68 base-pair long 5' non-coding region as well as the coding information for the first 63 amino acids. The second exon encodes the 23 amino acid insertion sequence, the third exon codes for amino acid 87 to 127 of the alpha AIns chain, whereas the last exon encodes the C-terminal 69 amino acids and contains the information for the 523 base-pair long 3' non-coding region. The second exon is bordered by a 3' splice junction (A X G/G X C), which deviates from the consensus for donor splice sites (A X G/G X T). This deviation is found in both hamster and mouse. An internal duplication was detected in the first exon by using a DIAGON-generated matrix for comparison. By means of similar DIAGON-generated matrices it was confirmed that the amino acids coded for by the third and fourth exons are homologous to the small heat-shock proteins of Drosophila, Caenorhabditis and soyabean. The implications of the differential splicing and the evolutionary aspects of the detected homologies are discussed.

Lens-specific promoter activity of a mouse gamma-crystallin gene

Crystallins are the major water-soluble proteins in vertebrate eye lenses. These lens-specific proteins are encoded by several gene families, and their expression is differentially regulated during lens cell differentiation. Here we show that a cloned mouse gamma-crystallin promoter is active in lens explants derived from 14-day-old chicken embryos but inactive in a variety of cells of non-lens origin. We also show that sequences required for proper utilization of this promoter are contained between nucleotide positions -392 and +47 relative to the transcription initiation site; deletion of sequences from positions -392 to -171 completely abolishes promoter activity. Since chickens do not have gamma-crystallin genes, the expression of a mouse gamma-crystallin promoter in chicken lens cells suggests that different classes of crystallin genes may be regulated by common lens tissue-specific mechanism(s) independent of species.

Versatile Escherichia coli-Bacillus shuttle vectors derived from runaway replication plasmids related to CloDF13

Versatile cloning vectors were constructed employing a runaway replication mutant of the bacteriocinogenic plasmid CloDF13. These vectors can, under conditions where protein synthesis is not inhibited, be amplified in Escherichia coli to high levels by elevating the temperature and are therefore useful for the production of large quantities of DNA and protein. Since the constructed shuttle vectors, which harbour at least six unique restriction endonucleases sites, replicate in E. coli, Enterobacter cloacae, Staphylococcus aureus and a variety of Bacilli, they are applicable for the genetic engineering of both gram-negative and gram-positive bacteria.

Structural and evolutionary relationships among five members of the human gamma-crystallin gene family

We have characterized five human gamma-crystallin genes isolated from a genomic phage library. DNA sequencing of four of the genes revealed that two of them predict polypeptides of 174 residues showing 71% homology in their amino acid sequence; the other two correspond to closely related pseudogenes which contain the same in-frame termination codon at identical positions in the coding sequence. Two of the genes and one of the pseudogenes are oriented in a head-to-tail fashion clustered within 22.5 kilobases. All three contain a TATA box 60 to 80 base pairs upstream of the initiation codon and a highly conserved segment of 44 base pairs in length immediately preceding the TATA box. The two genes and the two pseudogenes are similar in structure: each contains a small 5' exon encoding three amino acids followed by two larger exons that correspond exactly to the two similar structural domains of the polypeptide. The first intron varies from 100 to 110 base pairs, and the second intron ranges from 1 to several kilobases, rendering an overall gene size of 1.7 to 4.5 kilobases. At least one of the two pseudogenes appears to have been functional before inactivation, suggesting that their identical mutation was generated by gene conversion.

Rat lens beta-crystallins are internally duplicated and homologous to gamma-crystallins

The nucleotide sequence of two cloned rat lens beta-crystallin cDNAs pRL beta B3-2 and pRL beta B1-3 has been determined. pRL beta B3-2 contains the complete coding information for a beta-crystallin, designated beta B3, of 210 amino acid residues. pRL beta B1-3 is incomplete at its 5' end; the 5' codogenic information which is not present in this cDNA clone was deduced from the cloned gene. pRL beta B1-3 codes for a beta-crystallin polypeptide, designated beta B1, whose full length is 247 amino acid residues. Considerable sequence homology is noted between the amino- and carboxy-terminal halves of each protein. The two rat beta-crystallins show a substantial sequence homology with each other (60%) as well as with the published sequences of rat gamma-crystallin (37%) and bovine and murine beta-crystallins (55 and 45%). All these proteins have a two-domain structure which, like the bovine gamma II-crystallin, might be folded into four remarkably similar protein motifs. Our data further indicate that the beta-crystallins can be subdivided into two groups which are evolutionarily related. Both groups are, although more distantly, also related to the gamma-crystallins.

Comparison of the crystallin mRNA populations from rat, calf and duck lens. Evidence for a longer alpha A2-mRNA and two distinct alpha B2-mRNAs in the birds

Total cytoplasmic poly(A)-containing RNA from rat, calf and duck lens was fractionated by electrophoresis in methylmercury hydroxide-containing agarose gels. RNA electrophoresed in parallel lanes was either transferred onto nitrocellulose and hybridized with total cDNA synthesized on the initial mRNA or was recovered from individual gel fractions for in vitro translation in a reticulocyte cell-free system. This allowed the identification and size-characterization of individual mRNA species encoding alpha-, beta-, gamma- and delta-crystallin polypeptides. The 14 S mRNA fraction of rat lens comprises two alpha A2-mRNAs of approximately 1250 and 1350 nucleotides and the alpha AIns-mRNA with a size similar to that of the largest alpha A2-mRNA. The calf lens 14 S mRNA fraction harbors a heterogeneous population of alpha A2-mRNA. In the same fraction another mRNA encoding a polypeptide, designated X, has been found sharing no homology with alpha A sequences. The duck lens alpha A2-mRNA appears to be 400-450 bases longer than the rat and calf lens alpha A2-mRNAs. Furthermore, in contrast to the single alpha B2-mRNA in rat and calf lens, two alpha B2-mRNAs have been identified in duck lens, one, the major species, similar in size to the alpha B2-mRNA in rat and calf lens (800 bases), and the other species 700 nucleotides longer. The large size differences among the alpha A2- and alpha B2-mRNAs most likely reside in their 3'-untranslated sequences.

Characterization of the rat gamma-crystallin gene family and its expression in the eye lens

Rat genomic clones, which together contain all of the rat genomic gamma-crystallin sequences, have been characterized. Five gamma-crystallin genes are located on a contiguous DNA region, 63 X 10(3) base-pairs long. These genes, named (5') gamma 1-1, gamma 1-2, gamma 2-2 and gamma 3-1 (3'), are all oriented head to tail. A sixth gamma-crystallin gene, named the gamma 4-1 gene, could not be linked to the gamma-crystallin gene cluster with our present set of genomic clones. Mapping experiments using single copy sequences which form the extreme 5' or 3' region of the gene cluster showed that, if the gamma 4-1 gene is located on the same chromosome, then it must be separated from the gene cluster by at least 25 X 10(3) base-pairs of DNA. All gamma-crystallin genes have a similar mosaic structure. They contain a large (0.9 X 10(3) to 1.88 X 10(3) base-pairs) intron in the middle of the gene and are further interrupted close to the 5' end of the gene. The length of the first exon varies from about 40 to about 50 base-pairs. The complementary DNA clone pRL-gamma-3 used in this study is a copy of the transcript of the gamma 3-1 gene, while the second complementary DNA clone, pRL-gamma-2, is most likely a copy of the transcript of the gamma 2-1 gene. It is further shown that rat lens messenger RNA protects fragments from the 3' ends of the four other gamma-crystallin genes against degradation by S1 nuclease, hence all six gamma-crystallin genes present in the rat genome must be transcribed in the lens. Repetitive sequences were found to be present between and around the gamma-crystallin genes. Mapping with cloned repetitive sequences showed that three different repeats, designated A, B and C, occur more than once in the gamma-crystallin gene cluster. Repeat C is also found in the gamma 4-1 region. A repetitive region 3' to the gamma 3-1 gene contains members of all three repeat families.

Lens-specific expression of the chloramphenicol acetyltransferase gene promoted by 5' flanking sequences of the murine alpha A-crystallin gene in explanted chicken lens epithelia

We have developed a system using explanted embryonic chicken lens epithelia to express foreign recombinant genes containing crystallin DNA regulatory sequences introduced by calcium phosphate transfection. Optimal results were obtained with lens epithelia from 14-day embryos transfected 1 day after explantation and assayed 3 days later. When DNA sequences (-364 to +45) of the murine alpha A-crystallin gene were inserted in the pSVO-CAT expression vector of Gorman et al. [Gorman, C. M., Moffat, L. F. & Howard, B. H. (1982) Mol. Cell. Biol. 2, 1044-1051] in the same orientation as in the crystallin gene, they promoted chloramphenicol acetyltransferase (CAT; EC 2.3.1.28) activity in the transfected epithelia. Sequences 87 to 364 base pairs upstream from the murine gene cap site were required for CAT gene expression. These crystallin gene regulatory sequences did not promote CAT expression in primary cultures of embryonic chicken fibroblasts or other nonlens cells. By contrast, the long terminal repeat of Rous sarcoma virus and the early promoter of simian virus 40 promoted CAT activity in lens and nonlens cells. Our experiments thus demonstrate that the explanted embryonic chicken lens epithelium is an advantageous recipient for identifying lens-cell-specific regulatory sequences of crystallin genes and implicate a DNA region upstream of the "TATA box" for regulation of the murine alpha A-crystallin gene. These experiments also suggest that explanted epithelia from other tissues may be useful for studying the expression of foreign genes.

Nucleotide sequence of a chicken delta-crystallin gene

We have determined the complete nucleotide sequence of one of the two non-allelic delta-crystallin genes in the chicken, arbitrarily designated delta-gene 1, using a genomic clone (lambda g delta 106) containing the entire gene sequence. By comparison of the genomic sequence and the delta-crystallin cDNA sequence previously determined, we have identified exon sequences in the genomic sequence. Thus, the presence of 17 exons and 16 introns in the gene has been clarified. The delta-crystallin polypeptide deduced from the exon sequences consists of 465 amino acids which is larger, by 19 amino acid residues, than the polypeptide deduced from the cDNA sequence previously reported. Re-examination of the cDNA sequence using the same cDNA clone previously used shows that the present exon sequences are correct and the molecular weight of the deduced delta-crystallin polypeptide is 50,615 daltons instead of the previously reported value of 48,447 daltons. In addition, some structural features of the delta-crystallin gene including putative expression signals are discussed.

The crystallin gene families

Recent work from our laboratory on the structure and the genetic organization of the lens beta- and gamma-crystallin gene families is reviewed briefly. In the rat six different gamma-crystallin genes are present which all have an identical distribution of exons and introns, namely a small intron after the third translation codon and a larger one within the coding region for the connecting peptide which links the two domains of the gamma-crystallins. We find five rat genes physically linked and located on a DNA segment of only 50 kilobases, whereas the sixth gene is more distant. The polypeptide sequences, as deduced from DNA sequence analysis, of these six rat and two human gamma-crystallin genes are compared and discussed in terms of structural and evolutionary aspects. The gene coding for rat beta B1a-crystallin appears to be a single-copy gene of much larger size than the gamma-crystallin genes. The beta B1 gene is not physically linked to the other beta-crystallin genes, even though the various beta genes are evolutionarily related and in that sense constitute a gene family. In contrast to the gamma-crystallin genes, the beta B1 gene has an intron not only between the domain sequence but also between the motif sequences. In addition, the exon coding for the N-terminal extension of the protein is separated by an intron from the first protein motif sequence. We anticipate that structural and genetic investigations on lens crystallin genes and their expression might provide a framework for revealing the basis of (some) hereditary disorders in the visual system.

The molecular structures and interactions of bovine and human gamma-crystallins

Knowledge of the three-dimensional structure of bovine gamma II-crystallin has provided the basis for building molecular models using computer graphics of two human gamma-crystallins, the sequences of which have recently been determined. The tertiary structures of these gamma-crystallins are predicted to be highly conserved. They have extensive networks of interacting charges on their surfaces, which may contribute to their thermodynamic stability and partially define the degree of water retention in the lens. The human crystallins appear to be more hydrophobic than the bovine molecule. All have arrangements of cysteine thiols which may be important as electron sinks and reserve redox potential in the normal lens but which may contribute to protein aggregation in cataract.

Crystallin genes: templates for lens transparency

Analysis of recombinant DNAs provides new information on the basis of crystallin evolution and diversity. All crystallin genes contain introns. Two similar, tandemly linked chicken delta-crystallin genes, which probably arose by gene duplication, contain at least 16-17 introns. In the beta-crystallins three introns are situated between exons encoding the structural motifs of the protein, thus relating gene and protein structure. The structurally similar beta- and gamma-crystallins are coded by separate gene families which apparently arose by successive duplications of a common ancestral gene. The N-termini (5' end of gene) of the beta-crystallins appear to have diverged, while the 3' ends have been conserved. In the single murine alpha A-crystallin gene, coding information (the insert exon) for the alpha Ains peptide is contained within an intron. Alternative RNA splicing of this gene gives both the alpha A2 and the alpha Ains crystallin mRNAs. Thus, molecular genetics is providing a deeper appreciation of evolutionary events and is serving to redefine the crystallins in terms of their genes. Since the crystallins are so abundant in the lens, greater understanding of their polypeptide and gene structure should contribute to our understanding of and ability to treat cataract.

Two human gamma-crystallin genes are linked and riddled with Alu-repeats

A human genomic cosmid clone, pHcos gamma-1, has been isolated containing two closely linked gamma-crystallin genes, oriented in the same direction. The sequence of these genes and their 5' and 3' flanking regions has been determined. The coding regions of both genes are interrupted by two introns. The first introns (94 and 100 bp, respectively) are located in the 5' region of the genes. The second introns (2.82 and 0.95 kb, respectively) divide the genes into two halves, each encoding a structural domain of the gamma-crystallin protein. The coding regions of the two genes show 80% homology. Due to a mutation in the splice acceptor site of the second intron of the first gene, the coding region of its third exon is 3 bp longer than that of the second gene. In the flanking regions several conserved sequence elements were found, including those elements that are known to be necessary for the correct expression of eukaryotic genes. The flanking and intronic regions of the genes contain 'simple sequence' DNA and Alu repeats. The Alu repeats are usually clustered, contain truncated elements, and are often located near simple sequence DNA.

Gamma-crystallin family of the mouse lens: structural and evolutionary relationships

The heterogeneity inherent among gamma-crystallins of the mouse lens was investigated by sequence analysis of three gamma-crystallin-specific cDNAs. Comparison of the nucleotide sequence of these cDNAs and one previously reported by us revealed that the four gamma-cDNAs share 80-90% homology in nucleotide sequence. The entire 3' half of the coding region shows more variability than the 5' half, whereas the greatest variability is observed in the 3' untranslated region where numerous base substitutions, deletions, and insertions seem to have occurred. Alignment of the amino acid sequences of the four mouse gamma-crystallins according to the known four structural motifs of the major calf gamma-crystallin, gamma-II, suggests that all four mouse polypeptides are structurally very similar to calf gamma-II. However, most of the mouse polypeptides differ from gamma-II by the absence of one amino acid residue, resulting in a shorter connecting peptide between the two globular domains of the protein. Primary sequence alignment also revealed that the four mouse gamma-crystallins are most divergent in the third structural motif of the polypeptide. The significance of these differences in terms of the structure and function of the gamma-crystallins in the mouse lens is discussed.

Intron splicing: a conserved internal signal in introns of animal pre-mRNAs

Splicing of introns of yeast pre-mRNAs requires an internal conserved sequence T-A-C-T-A-A-C that is located 20-55 nucleotides from the 3' intron boundary. Sequences differing only in certain positions from this yeast signal have now been identified in the corresponding internal region of pre-mRNA introns of a variety of animal genes. A computer program that searches for homologues to a consensus structure and calculates the accuracy of match of each homologue is used to locate these sequences. We list here the signals found by this search in introns of sea urchin, mouse, rat, and human genes and give the consensus for each species. We also give the consensus found for Drosophila and chicken and duck signals. We then discuss the accumulating evidence that these internal signals are required for splicing in animals. It is also noted that a single-stranded region of small nuclear RNA U2 contains sequences complementary both to the proposed mammalian internal signal and to the neighboring CT-A-G at the 3' intron boundary. A role for U2 ribonucleoprotein in intron splicing is thus suggested.

Evidence against gamma-crystallin DNA or RNA sequences in the chicken

A cloned cDNA (pM gamma 1 Crl) encoding about two-thirds of a gamma-crystallin polypeptide from the murine lens was used as a hybridization probe to search for the presence of gamma-crystallin-like RNA or DNA sequences in the chicken. The 15-day-old chicken lens did not contain any RNA sequences homologous to the murine gamma-crystallin cDNA, as judged by Northern blot hybridization. An approximate 2.3 Kbp (kilobase pair) Bam HI fragment from the chicken genome hybridized to the murine gamma-crystallin cDNA in Southern blots. Cloning and sequencing of this genomic fragment, however, did not reveal any homology with gamma- or beta-crystallin sequences. A stretch of 22 dG: dC nucleotides was present in the cloned DNA fragment and possibly this hybridized to the cloned gamma-crystallin cDNA via its dG:dC nucleotide tails introduced during the cloning procedure. These data support those of McDevitt and Croft (1977) indicating that the chicken lens lacks gamma-crystallin and provide evidence against gamma-crystallin mRNA or a gamma-crystallin gene or pseudogene in the chicken. The absence of gamma-crystallin mRNA in the embryonic chicken lens cells makes them potentially useful recipients for investigating the expression of cloned gamma-crystallin genes.

Isolation and characterization of beta- and gamma-crystallin genes from rat genomic cosmid libraries

Two libraries, together containing about 10(6) colonies, have been constructed by cloning different size fractions of a partial Sau3A digest of rat genomic DNA in the cosmid vector pTM. Upon screening with two cDNA clones, one containing alpha A2-crystallin and one containing beta B1-crystallin sequences, 14 cosmid clones were isolated which were beta B1-crystallin-specific; none was found which contained alpha A2-crystallin sequences. The inserts of the beta B1 clones, which range from 35 to 45 kb in length, contain overlapping DNA segments covering more than 60 kb of rat genomic DNA. The composite BamHI restriction map of this region shows a single beta B1-crystallin gene, which is interrupted by several intronic sequences. Five recombinants hybridizing with two different rat lens gamma-crystallin cDNA clones were also isolated from these libraries. Four of these contain 31- to 41-kb inserts, whereas the fifth recombinant contains a 12.2-kb insert. Hybridization analysis with 5' and 3'-specific cDNA fragments indicates that altogether these inserts contain six gamma-crystallin genes, three of which are located on one insert of only 31 kb.

Analysis of the mouse gamma-crystallin gene family: assignment of multiple cDNAs to discrete genomic sequences and characterization of a representative gene

Blot hybridization analysis of mouse DNA with gamma-crystallin-specific cDNAs has detected the presence of a multigene family comprised of at least four related genes. The detailed structure of one of these genes, mouse gamma 4-crystallin (M gamma 4.1), and its corresponding cDNA has been determined. The gene spans approximately 2.6 kilobases (kb) and contains two introns. The gene predicts a polypeptide of 174 amino acids that shares extensive sequence homology with gamma-crystallin polypeptides of other species. The two similar structural domains of the protein correspond exactly to the second and third exons of the gene, supporting an exon-duplication model of gene evolution. The similarity in structure of this gene to that recently reported for a gamma-crystallin gene of the rat (1) suggests that a common structure may exist for all gamma-crystallin genes of the two species. Moreover, a highly conserved region, 50 nucleotides in length, immediately precedes the TATA box of both the mouse and rat genes, suggesting that this sequence may be important in gene regulation.