Sequences of four mouse histone H3 genes: implications for evolution of mouse histone genes

A subset of replication-dependent histone mRNAs are expressed as polyadenylated RNAs in terminally differentiated tissues

Histone proteins are synthesized in large amounts during S-phase to package the newly replicated DNA, and are among the most stable proteins in the cell. The replication-dependent (RD)-histone mRNAs expressed during S-phase end in a conserved stem-loop rather than a polyA tail. In addition, there are replication-independent (RI)-histone genes that encode histone variants as polyadenylated mRNAs. Most variants have specific functions in chromatin, but H3.3 also serves as a replacement histone for damaged histones in long-lived terminally differentiated cells. There are no reported replacement histone genes for histones H2A, H2B or H4. We report that a subset of RD-histone genes are expressed in terminally differentiated tissues as polyadenylated mRNAs, likely serving as replacement histone genes in long-lived non-dividing cells. Expression of two genes, HIST2H2AA3 and HIST1H2BC, is conserved in mammals. They are expressed as polyadenylated mRNAs in fibroblasts differentiated in vitro, but not in serum starved fibroblasts, suggesting that their expression is part of the terminal differentiation program. There are two histone H4 genes and an H3 gene that encode mRNAs that are polyadenylated and expressed at 5- to 10-fold lower levels than the mRNAs from H2A and H2B genes, which may be replacement genes for the H3.1 and H4 proteins.

Patterns of Gene Conversion in Duplicated Yeast Histones Suggest Strong Selection on a Coadapted Macromolecular Complex

We find evidence for interlocus gene conversion in five duplicated histone genes from six yeast species. The sequences of these duplicated genes, surviving from the ancient genome duplication, show phylogenetic patterns inconsistent with the well-resolved orthology relationships inferred from a likelihood model of gene loss after the genome duplication. Instead, these paralogous genes are more closely related to each other than any is to its nearest ortholog. In addition to simulations supporting gene conversion, we also present evidence for elevated rates of radical amino acid substitutions along the branches implicated in the conversion events. As these patterns are similar to those seen in ribosomal proteins that have undergone gene conversion, we speculate that in cases where duplicated genes code for proteins that are a part of tightly interacting complexes, selection may favor the fixation of gene conversion events in order to maintain high protein identities between duplicated copies.

Autoprotection against acetaminophen toxicity in cultured rat hepatocytes: the effect of pretreatment and growth factors

To evaluate the effect of acetaminophen pretreatment and growth factors on acetaminophen hepatotoxicity in cultured rat hepatocytes, rat hepatocytes in primary culture were exposed to acetaminophen 8 mM after pretreatment with either acetaminophen 1 mM, treatment with growth factors (EGF and HGF), or no treatment. Growth response was measured by changes in DNA, [3H]thymidine incorporation and mRNA of growth related proteins, cell damage by leakage of LDH to the medium and changes in ATP, and protection against toxicity by changes in glutathione, cytochrome p450 and the expression of glutathione-S-transferase and Cyp1A2. Pretreatment with acetaminophen induced growth response, weaker than that of growth factors, but pretreatment and growth factors reduced cell damage equally effectively. Glutathione and glutathione-S-transferase increased more by growth factors than by pretreatment, but both conditions reduced Cyp1A2 to near zero. Pretreatment and growth factors protect against acetaminophen toxicity by suppressing the expression of Cyp1A2, thereby reducing the production of the intermediate N-acetyl-p-benzoquinone imine (NAPQI). Suppression of Cyp1A2 expression by pretreatment is assumed to be due to a growth-stimulating effect of low concentrations of acetaminophen.

Purifying selection and birth-and-death evolution in the histone H4 gene family

Histones are small basic proteins encoded by a multigene family and are responsible for the nucleosomal organization of chromatin in eukaryotes. Because of the high degree of protein sequence conservation, it is generally believed that histone genes are subject to concerted evolution. However, purifying selection can also generate a high degree of sequence homogeneity. In this study, we examined the long-term evolution of histone H4 genes to determine whether concerted evolution or purifying selection was the major factor for maintaining sequence homogeneity. We analyzed the proportion (p(S)) of synonymous nucleotide differences between the H4 genes from 59 species of fungi, plants, animals, and protists and found that p(S) is generally very high and often close to the saturation level (p(S) ranging from 0.3 to 0.6) even though protein sequences are virtually identical for all H4 genes. A small proportion of genes showed a low level of p(S) values, but this appeared to be caused by recent gene duplication. Our findings suggest that the members of this gene family evolve according to the birth-and-death model of evolution under strong purifying selection. Using histone-like genes in archaebacteria as outgroups, we also showed that H1, H2A, H2B, H3, and H4 histone genes in eukaryotes form separate clusters and that these classes of genes diverged nearly at the same time, before the eukaryotic kingdoms diverged.

Liver gene regulation in rats following both 70 or 90% hepatectomy and endotoxin treatment

BACKGROUND

The metabolic state effect of liver failure on liver gene regulation was evaluated in a rat model.

METHODS

Following 70 or 90% hepatectomy and lipopolysaccharide or vehicle treatment at intervals up to 24 h, the liver remnants were analyzed for mRNA levels for acute-phase, liver-specific and growth-related proteins.

RESULTS

After 70% hepatectomy mRNA for alpha 1-acid glycoprotein, alpha 2-macroglobulin, thiostatin and fibrinogen, haptoglobin increased three- to sevenfold (P < 0.05), and mRNA for cyclin D and histone 3 increased seven- and 15-fold (P < 0.05), respectively. After lipopolysaccharide injection and 70% hepatectomy were done, mRNA for acute-phase proteins raised significantly (P < 0.05), more to five to 20-fold, while mRNA for growth-related proteins raised significantly (P < 0.05) less to three- to fourfold. After 90% hepatectomy, acute-phase protein mRNA increased five- to ninefold (P < 0.05) more than after 70% hepatectomy, while mRNA for histone 3 and cyclin D did not increase within 24 h, which indicates a delayed growth after 90% hepatectomy. In 90% of hepatectomized rats treated with lipopolysaccharide, acute-phase protein mRNA raised three- to sixfold (P < 0.05) less than after vehicle treatment.

CONCLUSION

In endotoxemia from liver failure, the synthesis of acute-phase proteins is upregulated by gene regulation at the expense of that for regeneration, which may be an appropriate response for immediate survival. In severe liver failure, endotoxin may interfere with the appropriate gene regulation.

Heterochromatin dynamics in mouse cells: interaction between chromatin assembly factor 1 and HP1 proteins

Mechanisms contributing to the maintenance of heterochromatin in proliferating cells are poorly understood. We demonstrate that chromatin assembly factor 1 (CAF-1) binds to mouse HP1 proteins via an N-terminal domain of its p150 subunit, a domain dispensable for nucleosome assembly during DNA replication. Mutations in p150 prevent association with HP1 in heterochromatin in cells that are not in S phase and the formation of CAF-1-HP1 complexes in nascent chromatin during DNA replication in vitro. We suggest that CAF-1 p150 has a heterochromatin-specific function distinct from its nucleosome assembly function during S phase. Just before mitosis, CAF-1 p150 and some HP1 progressively dissociate from heterochromatin concomitant with histone H3 phosphorylation. The HP1 proteins reassociate with chromatin at the end of mitosis, as histone H3 is dephosphorylated.

The mouse histone H1 genes: gene organization and differential regulation

There are six mouse histone H1 genes present in the histone gene cluster on mouse chromosome 13. These genes encode five histone H1 variants expressed in somatic cells, H1a to H1e, and the testis-specific H1t histone. Two of the genes that have not been assigned previously to the five somatic H1 subtypes have been identified as encoding the H1b and H1d subtypes. Three of the H1 genes, H1a, H1c and H1t, are present on an 80 kb segment of DNA that contains nine core histone genes. Two others, H1d and H1e, are present in a second patch, while the H1b gene is at least 500 kb away in a patch containing 14 core histone genes. The histone H1 genes are differentially expressed. All five genes for the somatic histone H1 proteins are expressed in exponentially growing cells. However, the levels of H1a, H1b and H1d mRNAs are greatly reduced in cells that are terminally differentiated or arrested in G0, while the H1c and H1e mRNAs continue to be expressed. In addition to the major RNA that ends at the stem-loop, the H1c gene expresses a longer, polyadenylated mRNA in differentiated cells, although in varying amounts. None of the other histone H1 genes encodes detectable amounts of polyadenylated mRNAs.

Expression of liver functions following sub-lethal and non-lethal doses of allyl alcohol and acetaminophen in the rat

BACKGROUND/AIMS

To relate severity of intoxication with allyl alcohol and acetaminophen to modulated hepatic gene expression of liver functions and regeneration.

METHODS

Rats fasted for 12 h received acetaminophen 3.5 or 5.6 g per kg body weight, or allyl alcohol 100 or 125 microl by gastric tube, doses producing no and about 30% mortality, respectively, within 2 days. In the morning 2, 6, 12, 24, and 36 h after intoxication, RNA was extracted from liver tissue. By slot blot hybridization mRNA levels were determined for acute phase proteins, enzymes involved in ammonia elimination and urea synthesis, and for proteins related to liver regeneration.

RESULTS

After allyl alcohol, mRNA of "positive" acute phase proteins was higher than after acetaminophen and increased with the dose, whereas after acetaminophen it decreased with the dose. The mRNA of the urea cycle enzymes and glutamine synthetase was uniformly reduced by allyl alcohol, whereas that of most urea cycle enzymes was above the controls after the non-lethal, but not after the sub-lethal, dose of acetaminophen. The mRNA of glutamine synthetase was significantly more reduced by acetaminophen than by allyl alcohol. The mRNA of cell-cycle dependent proteins was greatly reduced after both toxins, more after the higher dose.

CONCLUSIONS

The study shows that acetaminophen intoxication inhibits or fails to induce the expression of acute phase proteins in contrast to allyl alcohol intoxication. Allyl alcohol suppressed the expression of urea cycle enzymes, whereas that of the rate limiting enzymes carbamoylphosphate synthase and argininosuccinate synthetase was increased by the non-lethal but not by the sub-lethal dose of acetaminophen. The expression of the cell-cycle dependent proteins was more suppressed after the sub-lethal than after the non-lethal dose of both toxins. The data support the view that a fatal outcome of the intoxications depends more on the ability to regenerate than on the maintenance of liver-specific functions.

Assembly of CENP-A into centromeric chromatin requires a cooperative array of nucleosomal DNA contact sites

We investigated the requirements for targeting the centromeric histone H3 homologue CENP-A for assembly at centromeres in human cells by transfection of epitope-tagged CENP-A derivatives into HeLa cells. Centromeric targeting is driven solely by the conserved histone fold domain of CENP-A. Using the crystal structure of histone H3 as a guide, a series of CENP-A/histone H3 chimeras was constructed to test the role of discrete structural elements of the histone fold domain. Three elements were identified that are necessary for efficient targeting to centromeres. Two correspond to contact sites between histone H3 and nucleosomal DNA. The third maps to a homotypic H3-H3 interaction site important for assembly of the (H3/H4)2 heterotetramer. Immunoprecipitation confirms that CENP-A self-associates in vivo. In addition, targeting requires that CENP-A expression is uncoupled from histone H3 synthesis during S phase. CENP-A mRNA accumulates later in the cell cycle than histone H3, peaking in G2. Isolation of the gene for human CENP-A revealed a regulatory motif in the promoter region that directs the late S/G2 expression of other cell cycle-dependent transcripts such as cdc2, cdc25C, and cyclin A. Our data suggest a mechanism for molecular recognition of centromeric DNA at the nucleosomal level mediated by a cooperative series of differentiated CENP-A-DNA contact sites arrayed across the surface of a CENP-A nucleosome and a distinctive assembly pathway occurring late in the cell cycle.

The protein that binds the 3' end of histone mRNA: a novel RNA-binding protein required for histone pre-mRNA processing

Replication-dependent histone mRNAs are not polyadenylated but end in a conserved 26-nucleotide structure that contains a stem-loop. Much of the cell cycle regulation of histone mRNA is post-transcriptional and is mediated by the 3' end of histone mRNA. The stem-loop binding protein (SLBP) that binds the 3' end of histone mRNA is a candidate for the factor that participates in most, if not all, of the post-transcriptional regulatory events. We have cloned the cDNA for the SLBP from humans, mice, and frogs, using the recently developed yeast three-hybrid system. The human SLBP is a 31-kD protein and contains a novel RNA-binding domain, which has been mapped to a 73-amino-acid region of the protein. The cloned SLBP is the protein bound to the 3' end of histone mRNA as antibodies specific for the SLBP remove all specific binding activity from nuclear and polyribosomal extracts. These depleted extracts do not cleave histone pre-mRNA efficiently, demonstrating that the SLBP is required for efficient histone pre-mRNA processing.

Expression of liver-specific functions in rat hepatocytes following sublethal and lethal acetaminophen poisoning

AIM

In order to study the short-term effect of moderate and severe reduction of liver function by acetaminophen poisoning of different severity on gene expression for liver-specific functions, rats were given 3.75 and 7.5 g per kg body weight acetaminophen intragastrically. The lower dose is associated with low mortality; after the higher dose, most rats die at between 12 and 24 h.

METHODS

In the morning, 1 1/2, 3, 6, 9, and 12 h after the injection, the rats were killed and RNA was extracted from liver tissue. By slot-blot hybridization mRNA steady-state levels were determined for enzymes involved in metabolic liver functions, i.e. ureagenesis, gluconeogenesis, and drug metabolism, for acute phase proteins, "house-keeping" proteins, and for proteins related to liver regeneration. Results were expressed as per cent of the level in similarly fasted, untreated rats of the same stock

RESULTS

After the smaller dose of acetaminophen, most of the examined mRNA levels were increasing during the experimental period, being two- to four-fold elevated in relation to control after 6 to 12 h. Rats receiving the lethal dose either showed no or a later and smaller increase, and in several cases a fall towards the end of the experiment. The greatest differences were seen for mRNA of arginase, beta-fibrinogen, alpha 1-acid glycoprotein, alpha-tubulin, histone 3, TGF beta, and cyclin d, i.e. proteins associated with acute phase response and liver cell replication and maintenance.

CONCLUSIONS

It is concluded that reversible intoxication with acetaminophen induces an adaptive modulation of mRNA expression of liver functions and regeneration which is lacking after severe intoxication. This adaptation, with emphasis on acute phase response and regeneration, may be crucial for recovery after acetaminophen intoxication. If this also applies to the intoxication in man, estimates of the corresponding variables may be clues to the prognosis of acetaminophen-induced fulminant hepatic failure.

Characterization of the mouse histone gene cluster on chromosome 13: 45 histone genes in three patches spread over 1Mb

The histone gene cluster on mouse chromosome 13 has been isolated and characterized. Using overlapping YAC clones containing histone genes from chromosome 13, a contig of approximately 2 Mb has been defined. It contains 45 histone genes, organized in three patches containing tightly clustered genes. An 80-kb patch (patch III) containing 12 histone genes is near one end of the contig, and a similar-sized patch (patch I) containing 15 histone genes is near the other end of the contig, located at least 500 kb from the central patch (patch II) of histone genes. The entire cluster contains six histone H1 genes, including the testis-specific histone H1t gene that maps to the middle of the cluster. All nine histone H3 genes in this cluster have been sequenced, and their level of expression determined. Each histone H3 gene is distinct, with five genes encoding the H3.2 protein subtype and four genes encoding the H3.1 protein. They are all expressed, with each histone H3 gene accounting for a small proportion of the total histone H3 mRNA.

Expression of messenger RNA for liver functions following 70% and 90% hepatectomy

AIMS/METHODS

The effect of moderate and severe reduction of the functional liver mass on gene expression for liver functions was studied in rats following 70% and 90% hepatectomy. At intervals up to 24 h after operation rats were killed and RNA was extracted from the remaining liver tissue. By slot-blot hybridization mRNA steady-state levels were determined for enzymes involved in metabolic 'liver-specific' functions, acute phase proteins, 'house-keeping', and growth-related proteins. Results were expressed as per cent of levels in a pool from fed control rats of the same gender and age.

RESULTS

Among 'liver-specific' metabolic functions only expression of gluconeogenesis, represented by phosphoenol carboxykinase mRNA, was augmented initially, followed by a fall to very low values after 90% hepatectomy. The drug metabolizing system represented by CYP2B1/2 mRNA was reduced to half of the control values. Expression of urea synthesis, as reflected by carbamoylphosphate synthetase mRNA, showed a gradual decline after 90% hepatectomy, in contrast to rising levels of argininosuccinate lyase and arginase mRNA, possibly serving polyamine rather than urea synthesis. The mRNA level of the acute phase protein alpha 1-acid glycoprotein showed a smaller and later rise in 90% than in 70% hepatectomized rats, whereas that of alpha 2-macroglobulin only increased after 90% hepatectomy like the 'house-keeping' beta-actin mRNA. A rise in histone 3, which coincides with mitosis, was only seen after 70% hepatectomy, indicating that after 90% hepatectomy the response to growth-stimulating factors is weak or delayed, supported by a delayed rise in cyclin d and low levels of growth hormone receptor mRNA.

CONCLUSIONS

It is concluded that attempts by gene regulation to adapt liver functions to a reduction of the liver mass depend on the amount of liver tissue lost. When the loss is nearly fatal, compensation for normal metabolic functions may be abandoned for efforts to regenerate, which, however, may be delayed or after all be too weak.

Structure of a cluster of mouse histone genes

The structure of a 25 kilobase region of mouse DNA containing 6 functional histone genes and an H2a pseudogene has been determined. The sequences and levels of expression of the H3 and H2b gene as well as the sequence of the H2a pseudogene have been determined.

A solitary human H3 histone gene on chromosome 1

A solitary histone H3 gene encoding a novel H3 protein sequence has been isolated. This H3 gene maps to chromosome 1 (1q42), whereas we have shown previously that the majority of the human histone genes form a large cluster on chromosome 6 (6p21.3). In addition, a small cluster has been described at 1q21. The clustered histone genes are expressed during the S-phase of the cell cycle, hence their definition as replication-dependent histone genes. In contrast, expression of replacement histone genes is essentially cell-cycle independent; they are solitary genes and map outside the major clusters. The newly described H3 gene maps outside all known histone gene clusters and varies by four amino acid residues from the consensus mammalian H3 structure. In contrast to other solitary histone genes, this human H3 gene shows the consensus promoter and 3' flanking portions that are typical for replication-dependent genes.

Human CENP-A contains a histone H3 related histone fold domain that is required for targeting to the centromere

Centromeres are the differentiated chromosomal domains that specify the mitotic behavior of chromosomes. To examine the molecular basis for the specification of centromeric chromatin, we have cloned a human cDNA that encodes the 17-kD histone-like centromere antigen, CENP-A. Two domains are evident in the 140 aa CENP-A polypeptide: a unique NH2-terminal domain and a 93-amino acid COOH-terminal domain that shares 62% identity with nucleosomal core protein, histone H3. An epitope tagged derivative of CENP-A was faithfully targeted to centromeres when expressed in a variety of animal cells and this targeting activity was shown to reside in the histone-like COOH-terminal domain of CENP-A. These data clearly indicate that the assembly of centromeres is driven, at least in part, by the incorporation of a novel core histone into centromeric chromatin.

An alternative pathway of histone mRNA 3' end formation in mouse round spermatids

During mammalian spermiogenesis, the post-meiotic stage of spermatogenesis, histones are replaced by protamines on the DNA. Despite this histone elimination, novel polyadenylated histone transcripts were detected in mouse round spermatids. Sequence analysis of a spermatid-specific H2a cDNA clone indicated that it was derived from a mRNA of a replication-dependent histone gene even though its transcript was not polyadenylated in somatic and earlier spermatogenic cells. In round spermatids, both the hairpin and purine-rich elements in the 3' untranslated region of the somatic pre-mRNA were retained in the mature poly(A)+ mRNA transcripts. Polyadenylation occurred downstream of the purine-rich element and was not preceded by the somatic AATAAA polyadenylation signal sequence. While polyadenylated histone transcripts from replication-dependent genes have been observed previously in somatic cells, characteristics of this type of 3'-end formation in mammalian round spermatids were unique. In particular, a specific replication-dependent H2a gene was transcribed either as a polyadenylated or non-polyadenylated transcript in these cells, suggesting that the type of transcript present was dependent on the RNA sequence. Finally, both poly(A)- and poly(A)+ mRNAs were found on polyribosomes from round spermatids, indicating that histones were being translated in these cells and that the polyadenylation status of these transcripts did not affect their translatability.

Isolation and characterization of two replication-dependent mouse H1 histone genes

Mice contain at least seven nonallelic forms of the H1 histones, including the somatic variants H1a-e and less closely related variants H1 degrees and H1t. The mouse H1 degrees and H1c (H1var.1) genes were isolated and characterized previously. We have now isolated, sequenced and studied the expression properties of two additional mouse H1 genes, termed H1var.2 and H1var.3. Extensive amino acid and nucleotide sequence comparisons were made between the two genes and other mammalian H1 histone genes. A high degree of nucleotide sequence identity was seen between the H1var.2, rat H1d and human H1b genes, even well beyond the coding region, indicating that these genes are likely homologues. Unlike the previously characterized mouse H1var.1 gene which produces both nonpolyadenylated and polyadenylated mRNAs, the H1var.2 and H1var.3 genes produce only typical, replication dependent, nonpolyadenylated mRNAs.

Variable effects of the conserved RNA hairpin element upon 3' end processing of histone pre-mRNA in vitro

We have studied the requirements for efficient histone-specific RNA 3' processing in nuclear extract from mammalian tissue culture cells. Processing is strongly impaired by mutations in the pre-mRNA spacer element that reduce the base-pairing potential with U7 RNA. Moreover, by exchanging the hairpin and spacer elements of two differently processed H4 genes, we find that this difference is exclusively due to the spacer element. Finally, processing is inhibited by the addition of competitor RNAs, if these contain a wild-type spacer sequence, but not if their spacer element is mutated. Conversely, the importance of the hairpin for histone RNA 3' processing is highly variable: A hairpin mutant of the H4-12 gene is processed with almost wild-type efficiency in extract from K21 mouse mastocytoma cells but is strongly affected in HeLa cell extract, whereas an identical hairpin mutant of the H4-1 gene is affected in both extracts. The hairpin defect of H4-12-specific RNA in HeLa cells can be overcome by a compensatory mutation that increases the base complementarity to U7 snRNA. Very similar results were also obtained in RNA competition experiments: processing of H4-12-specific RNA can be competed by RNA carrying a wild-type hairpin element in extract from HeLa, but not K21 cells, whereas processing of H4-1-specific RNA can be competed in both extracts. With two additional histone genes we obtained results that were in one case intermediate and in the other similar to those obtained with H4-1. These results suggest that hairpin binding factor(s) can cooperatively support the ability of U7 snRNPs to form an active processing complex, but is(are) not directly involved in the processing mechanism.

The histone mRNA 3' end is required for localization of histone mRNA to polyribosomes

The final step in mRNA biosynthesis is transport of the mRNA from the nucleus to the cytoplasm. Histone genes from which the 3' stem-loop has been deleted are transcribed to give RNAs with heterogeneous 3' ends. These RNAs are localized in the nucleus and are stable. Addition of the histone 3' processing signal either on short (< 250 nts) or long (> 1000 nts) transcripts restores 3' processing and transport of the mRNA to the cytoplasm. In addition chimeric histone-U1 snRNA genes which produced RNAs with either histone or U1 3' ends were analyzed. Transcripts which ended with U1 snRNA 3' ends were not efficiently localized to polyribosomes. However, transcripts containing the same sequences including the snRNA 3' end followed by the histone 3' end were present in the cytoplasm on polyribosomes. Taken together these results suggest that the histone 3' end is required for export of histone mRNA to the cytoplasm and association of the mRNA with polyribosomes.

Isolation and characterization of developmentally regulated sea urchin U2 snRNA genes

Genes encoding the U2 snRNA have been isolated from the sea urchins, Strongylocentrotus purpuratus and Lytechinus variegatus. Representatives of tandemly repeated gene sets have been isolated from both sea urchin species and a unique U2 gene has also been isolated from L. variegatus. The sequence of the U2 snRNA encoded by the tandemly repeated genes differs in two nucleotides between S. purpuratus and L. variegatus. The unique U2 gene from L. variegatus encodes the same U2 RNA as the tandemly repeated genes. There is a change in the U2 genes expressed between morula and pluteus embryos as judged by a change in the U2 RNA sequence in S. purpuratus embryos. The tandemly repeated genes were expressed at a higher rate in blastula than in gastrula stage relative to the single-copy gene, when the two genes were injected into sea urchin zygotes.

Multiple processing-defective mutations in a mammalian histone pre-mRNA are suppressed by compensatory changes in U7 RNA both in vivo and in vitro

To study the role of base-pairing between the mammalian U7 snRNA and the highly variable histone downstream element (HDE) during the 3'-end maturation of mammalian histone pre-mRNAs, we mutated the HDE of the mouse H2A-614 gene and assayed processing in HeLa cells both in vivo and in vitro. Either a 9-nucleotide deletion or a block substitution of pyrimidines for 6 purines within the HDE abolished all 3'-end processing. Compensatory changes were introduced into a synthetic human U7 gene, whose transcripts assemble into Sm snRNPs in vivo. Suppression of the 6-purine substitution as well as a 3-purine substitution within the HDE was obtained in vivo by coexpressing the corresponding U7 suppressor RNAs and in vitro by using nuclear extracts prepared from HeLa cells containing U7 suppressor genes. Our results not only provide genetic evidence for base-pairing between the U7 snRNP and the HDE of mammalian histone pre-mRNAs but reveal an unexpected tolerance to drastic changes in the nature of the base-paired region.

A common transcriptional activator is located in the coding region of two replication-dependent mouse histone genes

There is a region in the mouse histone H3 gene protein-encoding sequence required for high expression. The 110-nucleotide coding region activating sequence (CRAS) from codons 58 to 93 of the H3.2 gene restored expression when placed 520 nucleotides 5' of the start of transcription in the correct orientation. Since identical mRNA molecules are produced by transcription of the original deletion gene and the deletion gene with the CRAS at -520, effects of the deletions on mRNA stability or other posttranscriptional events are completely ruled out. Inversion of the CRAS sequence in its proper position in the H3 gene resulted in only a threefold increase in expression, and placing the CRAS sequence 5' of the deleted gene in the wrong orientation had no effect on expression. In-frame deletions in the coding region of an H2a.2 gene led to identification of a 105-nucleotide sequence in the coding region between amino acids 50 and 85 necessary for high expression of the gene. Additionally, insertion of the H3 CRAS into the deleted region of the H2a.2 gene restored expression of the H2a gene. Thus, the CRAS element has an orientation-dependent, position-independent effect. Gel mobility shift competition studies indicate that the same proteins interact with both the H3 and H2a CRAS elements, suggesting that a common factor is involved in expression of histone genes.

A common transcriptional activator is located in the coding region of two replication-dependent mouse histone genes

There is a region in the mouse histone H3 gene protein-encoding sequence required for high expression. The 110-nucleotide coding region activating sequence (CRAS) from codons 58 to 93 of the H3.2 gene restored expression when placed 520 nucleotides 5' of the start of transcription in the correct orientation. Since identical mRNA molecules are produced by transcription of the original deletion gene and the deletion gene with the CRAS at -520, effects of the deletions on mRNA stability or other posttranscriptional events are completely ruled out. Inversion of the CRAS sequence in its proper position in the H3 gene resulted in only a threefold increase in expression, and placing the CRAS sequence 5' of the deleted gene in the wrong orientation had no effect on expression. In-frame deletions in the coding region of an H2a.2 gene led to identification of a 105-nucleotide sequence in the coding region between amino acids 50 and 85 necessary for high expression of the gene. Additionally, insertion of the H3 CRAS into the deleted region of the H2a.2 gene restored expression of the H2a gene. Thus, the CRAS element has an orientation-dependent, position-independent effect. Gel mobility shift competition studies indicate that the same proteins interact with both the H3 and H2a CRAS elements, suggesting that a common factor is involved in expression of histone genes.

An intact histone 3'-processing site is required for transcription termination in a mouse histone H2a gene

A transcription termination site has been characterized between the mouse histone H2a-614 and H3-614 genes. There is a poly(A)- RNA present in small amounts in the nucleus which ends 600 nucleotides 3' to the H2a-614 gene. Nuclear transcription studies demonstrate that transcription extends at least 600 nucleotides 3' to the gene but is greatly reduced 700 nucleotides 3' to the gene. If all or part of the normal 3'-processing signal, consisting of the stem-loop and the U7 small nuclear ribonucleoprotein binding site, is deleted, transcription then continues past the putative termination site and RNAs which end at the 3' end of the downstream H3-614 gene accumulate. Insertion of a 150-nucleotide fragment containing the termination site between the histone 3' end and downstream polyadenylation sites reduces usage of polyadenylation sites 85 to 90%. Taken together these results suggest there is a transcription termination site which requires an intact histone 3'-processing signal to function.

An H3-H4 histone gene pair in the marine copepod Tigriopus californicus, contains an intergenic dyad symmetry element

Histone genes are one of the most widely studied multigene families in eucaryotes. Over 200 histone genes have been sequenced, primarily in vertebrates, echinoderms, fungi and plants. We present here the structure and genomic orientation of an H3-H4 histone gene pair from the marine copepod, Tigriopus californicus. These histone gene sequences are the first to be determined for the class Crustacea and among the first to be determined for protostomes. The H4 and H3 genes in Tigriopus are shown to be adjacent, to have opposite polarity, and to contain a 26 bp region of dyad symmetry centrally located within the spacer region between the two genes. A similarly located dyad element has been found in yeast which contributes to the coordinated cell cycle control of the adjacent histone genes. The Tigriopus H3-H4 histone gene pair is clustered with one H2A and two H2B histone genes on a 15 kb genomic Bam H1 fragment. The H4 gene sequence predicts an H4 protein with an unusual serine to threonine substitution at the amino terminal residue. The H3 gene sequence predicts an H3 protein which is identical to the vertebrate H3.2 histone.

An intact histone 3'-processing site is required for transcription termination in a mouse histone H2a gene

A transcription termination site has been characterized between the mouse histone H2a-614 and H3-614 genes. There is a poly(A)- RNA present in small amounts in the nucleus which ends 600 nucleotides 3' to the H2a-614 gene. Nuclear transcription studies demonstrate that transcription extends at least 600 nucleotides 3' to the gene but is greatly reduced 700 nucleotides 3' to the gene. If all or part of the normal 3'-processing signal, consisting of the stem-loop and the U7 small nuclear ribonucleoprotein binding site, is deleted, transcription then continues past the putative termination site and RNAs which end at the 3' end of the downstream H3-614 gene accumulate. Insertion of a 150-nucleotide fragment containing the termination site between the histone 3' end and downstream polyadenylation sites reduces usage of polyadenylation sites 85 to 90%. Taken together these results suggest there is a transcription termination site which requires an intact histone 3'-processing signal to function.

Introns in histone genes alter the distribution of 3' ends

Chimeric genes were constructed which contained either a histone or globin promoter, a human alpha-globin coding region as a cDNA or containing one or both intervening sequences, and the 3' end of a mouse histone H2a gene. The genes were introduced into mouse L cells or Chinese Hamster Ovary cells. The genes containing at least one intervening sequence produced two mRNAs in about equal amounts, one which ended at a cryptic polyadenylation site 33 nucleotides 3' to the normal histone mRNA 3' end and one which ended at the normal histone 3' end. In contrast, the same construct containing a globin cDNA yielded mRNA ending only at the correct histone 3' end. Similar proportions of polyadenylated and non-polyadenylated mRNA were obtained when the cryptic polyadenylation signal was replaced with the globin polyadenylation signal. More than 90% of the transcripts were accurately spliced. All of the unspliced transcripts had histone 3' ends.

Molecular dissection of mutations at the heterozygous thymidine kinase locus in mouse lymphoma cells

The mouse lymphoma L5178Y TK+/- 3.7.2C cell line allows quantitation of induced TK(+/-)----TK-/- mutations at the heterozygous thymidine kinase (Tk) locus. TK-/- mutant colonies show a bimodal size distribution, reflecting a difference in the growth rates of the two size classes that is hypothesized to result from different degrees of genetic damage. The two homologous chromosomes 11 containing the alleles of the Tk gene in L5178Y 3.7.2C TK+/- cells are distinguishable at the cytogenetic level. We find, in addition, that the two alleles are distinguishable at the molecular level because of an Nco I restriction fragment length polymorphism at the 3' end of the gene. In a set of 51 large-colony and 48 small-colony TK-/- mutants induced by ionizing radiation or by chemical mutagens, we find that 78, including all except one of the small-colony mutants, have lost the Tk+ allele and that some of these have two to four copies of the remaining Tk- allele. Nineteen of the large-colony TK-/- mutants that do not show Tk+ allele loss show no other structural changes detectable at the level of Southern blot analysis. One shows a partial deletion. The variety of mutagenic lesions recorded at the heterozygous Tk locus may be representative of events observed in human malignancy and may include a wider range of mutagenic events than can be observed at hemizygous test loci.

The efficiency of 3'-end formation contributes to the relative levels of different histone mRNAs

Sequences at both the 5' and 3' ends of mouse histone genes contribute to the expression of individual genes. The 3' sequences required for high expression of the mouse H2a-614 gene are the same as the sequences required for 3'-end formation. When these sequences were substituted for the 3' end of the poorly expressed H2a-291 gene, expression of the H2a-291 gene was increased fivefold. A 65-nucleotide fragment containing the H2a-614 3' processing signal increased expression of the H2a-291 gene when it was placed in the proper orientation downstream of the H2a-291 3' end. The only mRNAs that accumulated from this gene ended at the H2a-291 3' end, which suggests that the transcript is sequentially processed. In an in vitro processing system, the different histone 3' ends showed different processing efficiencies, which correlated with their expression in cells. These results suggest that the efficiency of processing is important in determining the steady-state levels of individual mouse histone mRNAs.

Each of the conserved sequence elements flanking the cleavage site of mammalian histone pre-mRNAs has a distinct role in the 3'-end processing reaction

To study the substrate requirements for the histone 3'-end processing reaction of mammalian histone pre-mRNAs, we created a set of mutations in the sequences flanking the processing site of a mouse H3 gene. We found that deletion of the downstream purine-rich element hypothesized to interact with U7 small nuclear RNA abolishes in vitro 3'-end processing. Somewhat surprisingly, however, mutations in the hairpin loop element which destabilize or destroy the secondary structure reduce but do not abolish 3'-end processing. This is in apparent contrast to results obtained for the sea urchin system, where both sequence elements appear to be absolutely required for 3'-end formation.

The efficiency of 3'-end formation contributes to the relative levels of different histone mRNAs

Sequences at both the 5' and 3' ends of mouse histone genes contribute to the expression of individual genes. The 3' sequences required for high expression of the mouse H2a-614 gene are the same as the sequences required for 3'-end formation. When these sequences were substituted for the 3' end of the poorly expressed H2a-291 gene, expression of the H2a-291 gene was increased fivefold. A 65-nucleotide fragment containing the H2a-614 3' processing signal increased expression of the H2a-291 gene when it was placed in the proper orientation downstream of the H2a-291 3' end. The only mRNAs that accumulated from this gene ended at the H2a-291 3' end, which suggests that the transcript is sequentially processed. In an in vitro processing system, the different histone 3' ends showed different processing efficiencies, which correlated with their expression in cells. These results suggest that the efficiency of processing is important in determining the steady-state levels of individual mouse histone mRNAs.

Each of the conserved sequence elements flanking the cleavage site of mammalian histone pre-mRNAs has a distinct role in the 3'-end processing reaction

To study the substrate requirements for the histone 3'-end processing reaction of mammalian histone pre-mRNAs, we created a set of mutations in the sequences flanking the processing site of a mouse H3 gene. We found that deletion of the downstream purine-rich element hypothesized to interact with U7 small nuclear RNA abolishes in vitro 3'-end processing. Somewhat surprisingly, however, mutations in the hairpin loop element which destabilize or destroy the secondary structure reduce but do not abolish 3'-end processing. This is in apparent contrast to results obtained for the sea urchin system, where both sequence elements appear to be absolutely required for 3'-end formation.

A region in the coding sequence is required for high-level expression of murine histone H3 gene

Replication-dependent histone genes are expressed at high rates in S phase to provide the histone proteins required for chromosomal replication. Two genes, an H2a and H3 gene, located on chromosome 3 in the mouse and cloned together in a single 3-kilobase (plasmid MM614) restriction fragment are highly expressed. By transfecting mouse histone gene constructs into Chinese hamster ovary cells, we have identified a 110-nucleotide region within the coding sequence of the H3.2-614 gene that is required for high-level expression. Deletion of this region reduces expression of the gene by 20-fold. Additionally, the histone-coding region activates the human alpha-globin promoter, which is normally not expressed well in Chinese hamster ovary cells. Similar results with deletion constructions involving the H2a-614 gene suggest that an intragenic region plays an important role in transcription of these genes.

Histone gene switching in murine erythroleukemia cells is differentiation specific and occurs without loss of cell cycle regulation

We investigated the expression characteristics of the fully replication-dependent (FRD) and the partially replication-dependent (PRD) histone gene variants by measuring changes in steady-state mRNA levels during hexamethylene bisacetamide (HMBA)-induced differentiation of murine erythroleukemia (MEL) cells. Between 24 and 60 h after induction, there was a dramatic switch in histone gene expression, such that the ratio of PRD to FRD transcripts increased severalfold over that found in uninduced MEL cells. We demonstrated that this gene switching was not simply a partial or complete uncoupling of PRD gene expression from DNA synthesis. PRD and FRD transcript levels were regulated coordinately upon treatment of uninduced or induced MEL cells with inhibitors of DNA synthesis, protein synthesis, or both. Using several criteria, we were unable to detect any difference in PRD and FRD gene expression under any conditions except in cells undergoing differentiation. MEL cells were arrested at a precommitment stage of differentiation by induction with HMBA in the presence of dexamethasone (DEX). If DEX was subsequently removed, DNA synthesis resumed, the cells underwent commitment, and histone gene switching was observed. In contrast, if both DEX and HMBA were removed, DNA synthesis still resumed, but commitment did not occur and no gene switching was observed. These results imply that histone gene switching is intimately related to the differentiation process.

Histone gene switching in murine erythroleukemia cells is differentiation specific and occurs without loss of cell cycle regulation

We investigated the expression characteristics of the fully replication-dependent (FRD) and the partially replication-dependent (PRD) histone gene variants by measuring changes in steady-state mRNA levels during hexamethylene bisacetamide (HMBA)-induced differentiation of murine erythroleukemia (MEL) cells. Between 24 and 60 h after induction, there was a dramatic switch in histone gene expression, such that the ratio of PRD to FRD transcripts increased severalfold over that found in uninduced MEL cells. We demonstrated that this gene switching was not simply a partial or complete uncoupling of PRD gene expression from DNA synthesis. PRD and FRD transcript levels were regulated coordinately upon treatment of uninduced or induced MEL cells with inhibitors of DNA synthesis, protein synthesis, or both. Using several criteria, we were unable to detect any difference in PRD and FRD gene expression under any conditions except in cells undergoing differentiation. MEL cells were arrested at a precommitment stage of differentiation by induction with HMBA in the presence of dexamethasone (DEX). If DEX was subsequently removed, DNA synthesis resumed, the cells underwent commitment, and histone gene switching was observed. In contrast, if both DEX and HMBA were removed, DNA synthesis still resumed, but commitment did not occur and no gene switching was observed. These results imply that histone gene switching is intimately related to the differentiation process.

Differential expression of individual members of the histone multigene family due to sequences in the 5' and 3' regions of the genes

Histone proteins are encoded by a multigene family. The H3.2(614) and H2a(614) genes are present as single copies which are expressed at high levels, accounting for 30 to 40% of the H3 and H2a mRNAs, respectively, in different types of mouse cells. The other genes which have been isolated each contribute only a very small amount to the total type-specific mRNA pool. We demonstrate here that the differences in the level of expression of these genes are partly due to differences in their transcription rates. To investigate the sequences responsible for these differences in expression among the members of each family, we carried out DNA-mediated gene transfer experiments with both intact and chimeric histone genes. The 5' region of a highly expressed gene [H3.2(614) or H2a(614)] was attached to the 3' region of a histone gene which was expressed at low levels (H3-221 or H2a-291) and vice versa. The results show that sequences in both the 5' and 3' regions of the H3.2(614) and H2a(614) genes contribute to their high level of mRNA production by two independent mechanisms. The effect of the 3' sequences on mRNA accumulation has been narrowed to a 65-base-pair region including the 3'-terminal palindrome and downstream signal implicated in mRNA processing.

Structural and functional characterization of mouse U7 small nuclear RNA active in 3' processing of histone pre-mRNA

Oligonucleotides derived from the spacer element of the histone RNA 3' processing signal were used to characterize mouse U7 small nuclear RNA (snRNA), i.e., the snRNA component active in 3' processing of histone pre-mRNA. Under RNase H conditions, such oligonucleotides inhibited the processing reaction, indicating the formation of a DNA-RNA hybrid with a functional ribonucleoprotein component. Moreover, these oligonucleotides hybridized to a single nuclear RNA species of approximately 65 nucleotides. The sequence of this RNA was determined by primer extension experiments and was found to bear several structural similarities with sea urchin U7 snRNA. The comparison of mouse and sea urchin U7 snRNA structures yields some further insight into the mechanism of histone RNA 3' processing.

Nucleotide sequence of a cDNA clone encoding mouse transition protein 1

We have determined the nucleotide sequence of cDNA clones encoding mouse transition protein 1 (TP1), a basic nuclear protein involved in nuclear condensation during spermiogenesis. The nucleotide sequence predicts that transition protein 1 in rats and mice differs by only one amino acid. The rate of substitution of nucleotides in the coding region of mouse and rat transition protein 1 mRNA is close to the average of many proteins in rats and mice, and the usage of degenerate codons is typical of the mouse. The identification of this cDNA clone, in conjunction with previous work (Kleene et al. (1983) Dev. Biol. 98, 455-464; Hecht et al. (1986) Exp. Cell Res. 164, 183-190), demonstrates that the mRNA for mouse transition protein 1 accumulates during the haploid phase of spermatogenesis.

Differential expression of individual members of the histone multigene family due to sequences in the 5' and 3' regions of the genes

Histone proteins are encoded by a multigene family. The H3.2(614) and H2a(614) genes are present as single copies which are expressed at high levels, accounting for 30 to 40% of the H3 and H2a mRNAs, respectively, in different types of mouse cells. The other genes which have been isolated each contribute only a very small amount to the total type-specific mRNA pool. We demonstrate here that the differences in the level of expression of these genes are partly due to differences in their transcription rates. To investigate the sequences responsible for these differences in expression among the members of each family, we carried out DNA-mediated gene transfer experiments with both intact and chimeric histone genes. The 5' region of a highly expressed gene [H3.2(614) or H2a(614)] was attached to the 3' region of a histone gene which was expressed at low levels (H3-221 or H2a-291) and vice versa. The results show that sequences in both the 5' and 3' regions of the H3.2(614) and H2a(614) genes contribute to their high level of mRNA production by two independent mechanisms. The effect of the 3' sequences on mRNA accumulation has been narrowed to a 65-base-pair region including the 3'-terminal palindrome and downstream signal implicated in mRNA processing.

The stem-loop structure at the 3' end of histone mRNA is necessary and sufficient for regulation of histone mRNA stability

Chimeric genes were made by fusing mouse histone genes with a human alpha-globin gene. The genes were introduced into mouse L cells and the stability of the chimeric mRNAs was measured when DNA synthesis was inhibited. An mRNA containing all the globin coding sequences and the last 30 nucleotides of the histone mRNA was degraded at the same rate as histone mRNA.

Structural and functional characterization of mouse U7 small nuclear RNA active in 3' processing of histone pre-mRNA

Oligonucleotides derived from the spacer element of the histone RNA 3' processing signal were used to characterize mouse U7 small nuclear RNA (snRNA), i.e., the snRNA component active in 3' processing of histone pre-mRNA. Under RNase H conditions, such oligonucleotides inhibited the processing reaction, indicating the formation of a DNA-RNA hybrid with a functional ribonucleoprotein component. Moreover, these oligonucleotides hybridized to a single nuclear RNA species of approximately 65 nucleotides. The sequence of this RNA was determined by primer extension experiments and was found to bear several structural similarities with sea urchin U7 snRNA. The comparison of mouse and sea urchin U7 snRNA structures yields some further insight into the mechanism of histone RNA 3' processing.

Synonymous nucleotide substitution rates of beta-tubulin and histone genes conform to high overall genomic rates in rodents but not in sea urchins

Sea urchin and rodent genomes have been posited to evolve rapidly as indicated by divergences in single copy nuclear DNA sequences. We have examined whether the synonymous substitution rates of three highly conserved genes, beta-tubulin, histone H4, and histone H3, adhere to these high genomic substitution rates by comparing sequences between two sea urchins, Strongylocentrotus purpuratus and Lytechinus pictus, and between rodents and humans. Whereas the rate of change between the 3' untranslated regions of the beta-tubulin cDNA of S. purpuratus (Sp-beta 1), sequenced in this study, and of L. pictus (Lp-beta 3) was consistent with the overall rate of change estimated from previous DNA hybridization results between these species, the synonymous substitution rates for the carboxyl domains of these beta-tubulins, as well as for the late histones H4 and H3, were significantly depressed. In contrast, synonymous nucleotide substitution rates between rodents and between rodent and human for the carboxyl domain proper of identical beta-tubulin isotypes and for histone H4 and H3.1 did not differ from the overall rate of change for the rodent genomes. Moreover, an analysis of paralogous human and mouse beta-tubulin sequences supported the conclusion that the synonymous substitution rates in the mouse were higher than those in the human. Differences in constraint on evolutionary change were not evident strictly from the conserved amino acid sequences and base compositions of these genes. Other constraining influences seemed more relevant to the departure of the synonymous substitution rates of the sea urchin beta-tubulin and histone coding regions from the average genomic rate.

Identification of the human U7 snRNP as one of several factors involved in the 3' end maturation of histone premessenger RNA's

In eukaryotic cells, the conversion of gene transcripts into messenger RNA's involves multiple factors, including the highly abundant small nuclear ribonucleoprotein (snRNP) complexes that mediate the splicing reaction. Separable factors are also required for the 3' end processing of histone pre-mRNA's. The two conserved signals flanking the 3' cleavage site are recognized by discrete components present in active HeLa cell extracts: the upstream stem loop associates with a nuclease-insensitive factor, while binding to the downstream element is mediated by a component having the properties of a snRNP. The sequence of the RNA moiety of the low abundance human U7 snRNP suggests how the relatively degenerate downstream element of mammalian pre-mRNA's could be recognized by RNA base-pairing.

The stem-loop structure at the 3' end of histone mRNA is necessary and sufficient for regulation of histone mRNA stability

Chimeric genes were made by fusing mouse histone genes with a human alpha-globin gene. The genes were introduced into mouse L cells and the stability of the chimeric mRNAs was measured when DNA synthesis was inhibited. An mRNA containing all the globin coding sequences and the last 30 nucleotides of the histone mRNA was degraded at the same rate as histone mRNA.

Both conserved signals on mammalian histone pre-mRNAs associate with small nuclear ribonucleoproteins during 3' end formation in vitro

Pre-mRNA substrates containing sequences from human and mouse histone genes are accurately processed in a HeLa cell nuclear extract to generate mature 3' termini. When in vitro processing reactions containing either human histone H3 or mouse histone H3 transcripts are treated with RNase T1 and probed with antibodies specific for the Sm protein determinants or for the trimethylguanosine cap structure unique to the U RNAs present in small nuclear ribonucleoproteins, RNA fragments that encompass the site of 3' end formation on the pre-mRNA transcript are selectively recovered. Several different interactions are detected: at time zero, the protected region contains the upstream conserved hairpin loop structure; at later times during the reaction, protection extends beyond the site of 3' end formation to include the downstream conserved sequence element and the 5' cap of the transcript is bound as well. Possible interactions between Sm small nuclear ribonucleoproteins and these conserved sequence elements in histone pre-mRNAs are discussed.