Ribosomal RNA gene copy number and nucleolar-size polymorphisms within and among chicken lines selected for enhanced growth

Insights into ribosomal DNA dominance and magnification through characterization of isogenic deletion alleles

The major loci for the large primary ribosomal RNA (rRNA) genes (35S rRNAs) exist as hundreds to thousands of tandem repeats in all organisms and dozens to hundreds in Drosophila. The highly repetitive nature of the ribosomal DNA (rDNA) makes it intrinsically unstable, and many conditions arise from the reduction in or magnification of copy number, but the conditions under which it does so remain unknown. By targeted DNA damage to the rDNA of the Y chromosome, we created and investigated a series of rDNA alleles. We found that complete loss of rDNA leads to lethality after the completion of embryogenesis, blocking larval molting and metamorphosis. We find that the resident retrotransposons-R1 and R2-are regulated by active rDNA such that reduction in copy number derepresses these elements. Their expression is highest during the early first instar, when loss of rDNA is lethal. Regulation of R1 and R2 may be related to their structural arrangement within the rDNA, as we find they are clustered in the flanks of the nucleolus organizing region (NOR; the cytological appearance of the rDNA). We assessed the complex nucleolar dominance relationship between X- and Y-linked rDNA using a histone H3.3-GFP reporter construct and incorporation at the NOR and found that dominance is controlled by rDNA copy number as at high multiplicity the Y-linked array is dominant, but at low multiplicity the X-linked array becomes derepressed. Finally, we found that multiple conditions that disrupt nucleolar dominance lead to increased rDNA magnification, suggesting that the phenomena of dominance and magnification are related, and a single mechanism may underlie and unify these two longstanding observations in Drosophila.

Combined Assay of rDNA and SatIII Copy Numbers as an Individual Profile of Stress Resistance, Longevity, Fertility and Disease Predisposition

The ribosomal DNA and pericentromeric satellite repeats are two important types of moderately repeated sequences existing in the human genome. They are functionally involved in the universal stress response. There is accumulating evidence that the copy number variation (CNV) of the repeat units is a novel factor modulating the stress response and, thus, has phenotypic manifestations. The ribosomal repeat copy number plays a role in stress resistance, lifespan, in vitro fertilization chances, disease progression and aging, while the dynamics of the satellite copy number are a sort of indicator of the current stress state. Here, we review some facts showing that a combined assay of rDNA and SatII/III abundance can provide valuable individual data ("stress profile") indicating not only the inherited adaptive reserve but also the stress duration and acute or chronic character of the stress. Thus, the repeat count could have applications in personalized medicine in the future.

Asexuality Associated with Marked Genomic Expansion of Tandemly Repeated rRNA and Histone Genes

How does asexual reproduction influence genome evolution? Although is it clear that genomic structural variation is common and important in natural populations, we know very little about how one of the most fundamental of eukaryotic traits-mode of genomic inheritance-influences genome structure. We address this question with the New Zealand freshwater snail Potamopyrgus antipodarum, which features multiple separately derived obligately asexual lineages that coexist and compete with otherwise similar sexual lineages. We used whole-genome sequencing reads from a diverse set of sexual and asexual individuals to analyze genomic abundance of a critically important gene family, rDNA (the genes encoding rRNAs), that is notable for dynamic and variable copy number. Our genomic survey of rDNA in P. antipodarum revealed two striking results. First, the core histone and 5S rRNA genes occur between tandem copies of the 18S-5.8S-28S gene cluster, a unique architecture for these crucial gene families. Second, asexual P. antipodarum harbor dramatically more rDNA-histone copies than sexuals, which we validated through molecular and cytogenetic analysis. The repeated expansion of this genomic region in asexual P. antipodarum lineages following distinct transitions to asexuality represents a dramatic genome structural change associated with asexual reproduction-with potential functional consequences related to the loss of sexual reproduction.

Structure of the intergenic spacers in chicken ribosomal DNA

BACKGROUND

Ribosomal DNA (rDNA) repeats are situated in the nucleolus organizer regions (NOR) of chromosomes and transcribed into rRNA for ribosome biogenesis. Thus, they are an essential component of eukaryotic genomes. rDNA repeat units consist of rRNA gene clusters that are transcribed into single pre-rRNA molecules, each separated by intergenic spacers (IGS) that contain regulatory elements for rRNA gene cluster transcription. Because of their high repeat content, rDNA sequences are usually absent from genome assemblies. In this work, we used the long-read sequencing technology to describe the chicken IGS and fill the knowledge gap on rDNA sequences of one of the key domesticated animals.

METHODS

We used the long-read PacBio RSII technique to sequence the BAC clone WAG137G04 (Wageningen BAC library) known to contain chicken NOR elements and the HGAP workflow software suit to assemble the PacBio RSII reads. Whole-genome sequence contigs homologous to the chicken rDNA repetitive unit were identified based on the Gallus_gallus-5.0 assembly with BLAST. We used the Geneious 9.0.5 and Mega software, maximum likelihood method and Chickspress project for sequence evolution analysis, phylogenetic tree construction and analysis of the raw transcriptome data.

RESULTS

Three complete IGS sequences in the White Leghorn chicken genome and one IGS sequence in the red junglefowl contig AADN04001305.1 (Gallus_gallus-5.0) were detected. They had various lengths and contained three groups of tandem repeats (some of them being very GC rich) that form highly organized arrays. Initiation and termination sites of rDNA transcription were located within small and large unique regions (SUR and LUR), respectively. No functionally significant sites were detected within the tandem repeat sequences.

CONCLUSIONS

Due to the highly organized GC-rich repeats, the structure of the chicken IGS differs from that of IGS in human, apes, Xenopus or fish rDNA. However, the chicken IGS shares some molecular organization features with that of the turtles, which are other representatives of the Sauropsida clade that includes birds and reptiles. Our current results on the structure of chicken IGS together with the previously reported ribosomal gene cluster sequence provide sufficient data to consider that the complete chicken rDNA sequence is assembled with confidence in terms of molecular DNA organization.

Sequence properties of certain GC rich avian genes, their origins and absence from genome assemblies: case studies

Background

More and more eukaryotic genomes are sequenced and assembled, most of them presented as a complete model in which missing chromosomal regions are filled by Ns and where a few chromosomes may be lacking. Avian genomes often contain sequences with high GC content, which has been hypothesized to be at the origin of many missing sequences in these genomes. We investigated features of these missing sequences to discover why some may not have been integrated into genomic libraries and/or sequenced.

Results

The sequences of five red jungle fowl cDNA models with high GC content were used as queries to search publicly available datasets of Illumina and Pacbio sequencing reads. These were used to reconstruct the leptin, TNFα, MRPL52, PCP2 and PET100 genes, all of which are absent from the red jungle fowl genome model. These gene sequences displayed elevated GC contents, had intron sizes that were sometimes larger than non-avian orthologues, and had non-coding regions that contained numerous tandem and inverted repeat sequences with motifs able to assemble into stable G-quadruplexes and intrastrand dyadic structures. Our results suggest that Illumina technology was unable to sequence the non-coding regions of these genes. On the other hand, PacBio technology was able to sequence these regions, but with dramatically lower efficiency than would typically be expected.

Conclusions

High GC content was not the principal reason why numerous GC-rich regions of avian genomes are missing from genome assembly models. Instead, it is the presence of tandem repeats containing motifs capable of assembling into very stable secondary structures that is likely responsible.

Dosage effects of human ribosomal genes (rDNA) in health and disease

Human ribosomal RNA genes encoding a pre-transcript of the three major ribosomal RNA (18S, 5.8S, and 28S rRNA) are tandemly repeated in human genome. Their total copy number varies from 250 to 670 per diploid genome with a mean of approximately 420 copies, but only a fraction of them is transcriptionally active. The functional consequences of human ribosomal RNA gene dosage are not widely known and often assumed to be negligible. Here, we review the facts of rRNA gene dosage effects on normal growth and aging, stress resistance of healthy individuals, and survivability of patients with chromosomal abnormalities, as well as on the risk and severity of some multifactorial diseases with proven genetic predisposition. An original hypothesis that rRNA gene dosage can be a modulating factor involved in the pathogenesis of schizophrenia and rheumatoid arthritis is put forward.

Deep landscape update of dispersed and tandem repeats in the genome model of the red jungle fowl, Gallus gallus, using a series of de novo investigating tools

BACKGROUND

The program RepeatMasker and the database Repbase-ISB are part of the most widely used strategy for annotating repeats in animal genomes. They have been used to show that avian genomes have a lower repeat content (8-12 %) than the sequenced genomes of many vertebrate species (30-55 %). However, the efficiency of such a library-based strategies is dependent on the quality and completeness of the sequences in the database that is used. An alternative to these library based methods are methods that identify repeats de novo. These alternative methods have existed for a least a decade and may be more powerful than the library based methods. We have used an annotation strategy involving several complementary de novo tools to determine the repeat content of the model genome galGal4 (1.04 Gbp), including identifying simple sequence repeats (SSRs), tandem repeats and transposable elements (TEs).

RESULTS

We annotated over one Gbp. of the galGal4 genome and showed that it is composed of approximately 19 % SSRs and TEs repeats. Furthermore, we estimate that the actual genome of the red jungle fowl contains about 31-35 % repeats. We find that library-based methods tend to overestimate TE diversity. These results have a major impact on the current understanding of repeats distributions throughout chromosomes in the red jungle fowl.

CONCLUSIONS

Our results are a proof of concept of the reliability of using de novo tools to annotate repeats in large animal genomes. They have also revealed issues that will need to be resolved in order to develop gold-standard methodologies for annotating repeats in eukaryote genomes.

The highly pathogenic H7N3 avian influenza strain from July 2012 in Mexico acquired an extended cleavage site through recombination with host 28S rRNA

BACKGROUND

A characteristic difference between highly and non-highly pathogenic avian influenza strains is the presence of an extended, often multibasic, cleavage motif insertion in the hemagglutinin protein. Such motif is found in H7N3 strains from chicken farm outbreaks in 2012 in Mexico.

METHODS

Through phylogenetic, sequence and structural analysis, we try to shed light on the role, prevalence, likelihood of appearance and origin of the inserted cleavage motifs in these H7N3 avian influenza strains.

RESULTS

The H7N3 avian influenza strain which caused outbreaks in chicken farms in June/July 2012 in Mexico has a new extended cleavage site which is the likely reason for its high pathogenicity in these birds. This cleavage site appears to have been naturally acquired and was not present in the closest low pathogenic precursors. Structural modeling shows that insertion of a productive cleavage site is quite flexible to accept insertions of different length and with sequences from different possible origins. Different from recent cleavage site insertions, the origin of the insert here is not from the viral genome but from host 28S ribosomal RNA (rRNA) instead. This is a novelty for a natural acquisition as a similar insertion has so far only been observed in a laboratory strain before. Given the abundance of viral and host RNA in infected cells, the acquisition of a pathogenicity-enhancing extended cleavage site through a similar route by other low-pathogenic avian strains in future does not seem unlikely. Important for surveillance of these H7N3 strains, the structural sites known to enhance mammalian airborne transmission are dominated by the characteristic avian residues and the risk of human to human transmission should currently be low but should be monitored for future changes accordingly.

CONCLUSIONS

This highly pathogenic H7N3 avian influenza strain acquired a novel extended cleavage site which likely originated from recombination with 28S rRNA from the avian host. Notably, this new virus can infect humans but currently lacks critical host receptor adaptations that would facilitate human to human transmission.

Architecture and organization of chicken microchromosome 16: order of the NOR, MHC-Y, and MHC-B subregions

Here we present a high-resolution cytogenomic analysis of chicken microchromosome 16. We established the location of the major histocompatibility complex (MHC)-B and -Y subregions relative to each other and to the nucleolus organizer region (NOR) encoding the 18S-5.8S-28S ribosomal DNA. To do so, we employed multicolor fluorescence in situ hybridization using large-insert bacterial artificial chromosome clones with fully sequenced inserts or repetitive sequence probes specific for the subregion of interest. We show that the MHC-Y and -B regions are located on the same side of the NOR, rather than opposite ends, as previously proposed. On the q arm, the MHC-Y is closely adjacent to the NOR, whereas the MHC-B is distal near the q-terminus. A relatively large GC-rich region separates the 2 MHC subregions and includes a specialized structure, a secondary constriction. We propose that the GC-rich large physical distance is the basis for the lack of genetic linkage between the NOR and MHC-B and between the MHC-Y and -B. An integrated model for GGA 16 is presented that incorporates gene complex order in the context of key architectural features including p and q arms, primary (centromere) and secondary constrictions, telomeres, as well as AT- and GC-rich regions.

Use of sperm plasmid DNA lipofection combined with REMI (restriction enzyme-mediated insertion) for production of transgenic chickens expressing eGFP (enhanced green fluorescent protein) or human follicle-stimulating hormone

Linearized p-eGFP (plasmid-enhanced green fluorescent protein) or p-hFSH (plasmid human FSH) sequences with the corresponding restriction enzyme were lipofected into sperm genomic DNA. Sperm transfected with p-eGFP were used for artificial insemination in hens, and in 17 out of 19 of the resultant chicks, the exogenous DNA was detected in their lymphocytes as determined by PCR and expressed in tissues as determined by (a) PCR, (b) specific emission of green fluorescence by the eGFP, and (c) Southern blot analysis. A complete homology was found between the Aequorea Victoria eGFP DNA and a 313-bp PCR product of extracted DNA from chick blood cells. Following insemination with sperm lipofected with p-hFSH, transgenic offspring were obtained for two generations as determined by detection of the transgene for human FSH (PCR) and expression of the gene (RT-PCR and quantitative real-time PCR) and the presence of the protein in blood (radioimmunoassay). Data demonstrate that lipofection of plasmid DNA with restriction enzyme is a highly efficient method for the production of transfected sperm to produce transgenic offspring by direct artificial insemination.

Gene mapping of 28S and 5S rDNA sites in the spined loach Cobitis taenia (Pisces, Cobitidae) from a diploid population and a diploid-tetraploid population

We compare the chromosomal 28S and 5S rDNA patterns of the spined loach C. taenia (2n = 48) from an exclusively diploid population and from a diploid-polyploid population using 28S and 5S rDNA probe preparation and labelling, and fluorescence in situ hybridization (FISH). The 5S rDNA was located in two to three chromosome pairs, and separated from the 28S loci for the males and one female (F1) from the diploid population. Loaches from a diploid-polyploid population, and one female (F2) from the diploid population were characterized by at least one chromosome pair with 5S and 28S overlapping signals. The fishes differed mainly in their number of 28S rDNA loci, located on 3-6 chromosomes. All individuals from both populations were characterized by one acrocentric chromosome bearing a 28S rDNA signal on the telomeres of its long arm. The number of major ribosomal DNA in the karyotype of C. taenia by FISH was always higher than the number of Ag-NORs. Our data confirm the extensive polymorphism of NORs in both populations, as already has been observed in closely related Cobitis species, and less polymorphic 5S rDNA pattern. However, this preliminary result highlights the need for a wider scale study.

Telomeres in the chicken: genome stability and chromosome ends

Telomeres are the complex nucleoprotein structures at the termini of linear chromosomes. Telomeric DNA consists of a highly conserved hexanucleotide arranged in tandem repeats. Telomerase, a ribonucleoprotein of the reverse transcriptase family, specifies the sequence of telomeric DNA and maintains telomere array length. Numerous studies in model organisms established the significance of telomere structure and function in regulating genome stability, cellular aging, and oncogenesis. Our overall research objectives are to understand the organization of the telomere arrays in chicken in the context of the unusual organization and specialized features of this higher vertebrate genome (which include a compact genome, numerous microchromosomes, and high recombination rate) and to elucidate the role telomeres play in genome stability impacting cell function and life span. Recent studies found that the chicken genome contains three overlapping size classes of telomere arrays that differ in location and age-related stability: Class I 0.5 to 10 kb, Class II 10 to 40 kb, and Class III 40 kb to 2 Mb. Some notable features of chicken telomere biology are that the chicken genome contains ten times more telomeric DNA than the human genome and the Class III telomere arrays are the largest described for any vertebrate species. In vivo, chicken telomeres (Class II) shorten in an age-related fashion and telomerase activity is high in early stage embryos and developing organs but down-regulates during late embryogenesis or postnatally in most somatic tissues. In vitro, chicken cells down-regulate telomerase activity unless transformed. Knowledge of chicken telomere biology contributes information relevant to present and future biotechnology applications of chickens in vivo and chicken cells in vitro.

Patterns of ribosomal gene variation in elite commercial chicken pure line populations

The nucleolus organizer region (NOR) encodes the tandemly repeated 18S, 5.8S and 28S ribosomal (r) RNA genes. The NORs of broiler and layer commercial chicken pure lines were studied to establish the type and extent of genetic variation at this important locus. The parameters studied were gene copy number, repeat size, and diversity of NOR-types. The populations were organized into three groups for analysis including brown-egg broiler (13 lines), brown-egg layer (six lines), and white-egg layer (eight lines). The ribosomal gene copy number average of the white-egg layer populations was significantly lower (329 genes) than that of the brown-egg layers (372 genes); the brown-egg broiler ribosomal gene average was intermediate (350 genes). The white-egg layer populations exhibited a ribosomal repeat unit average size of 36 kb, significantly different from the brown-egg layer and brown-egg broiler average repeat unit size of 32.5 and 33.9 kb, respectively. NOR array size was similar among the three groups (6 mb). The brown-egg broiler populations exhibited polymorphic NOR patterns, intra- and interline, whereas the white-egg layer populations were essentially monomorphic for NOR-type; brown-egg layers exhibited an intermediate level of NOR diversity. Some NOR array characteristics may be a function of breed origin as brown-egg commercial populations, both broilers and layers, have similar breed origins and exhibited similarities for predominant repeat unit size as compared with white-egg layer populations. However, the finding that brown-egg broiler lines typically exhibit a greater number of segregating NOR-types than brown-egg layer lines suggests that the selection schemes of broiler vs. layer pure line populations may also have influenced the degree of variation at this gene complex.