- Research Article
- Open Access
A reverse genetic approach identifies an ancestral frameshift mutation in RP1 causing recessive progressive retinal degeneration in European cattle breeds
- Pauline Michot1, 2,
- Sabine Chahory3,
- Andrew Marete1, 4,
- Cécile Grohs1,
- Dimitri Dagios3,
- Elise Donzel3,
- Abdelhak Aboukadiri1,
- Marie-Christine Deloche1, 2,
- Aurélie Allais-Bonnet2, 5,
- Matthieu Chambrial6,
- Sarah Barbey7,
- Lucie Genestout8,
- Mekki Boussaha1,
- Coralie Danchin-Burge9,
- Sébastien Fritz1, 2,
- Didier Boichard1 and
- Aurélien Capitan1, 2Email author
© The Author(s) 2016
- Received: 2 May 2016
- Accepted: 26 July 2016
- Published: 10 August 2016
Domestication and artificial selection have resulted in strong genetic drift, relaxation of purifying selection and accumulation of deleterious mutations. As a consequence, bovine breeds experience regular outbreaks of recessive genetic defects which might represent only the tip of the iceberg since their detection depends on the observation of affected animals with distinctive symptoms. Thus, recessive mutations resulting in embryonic mortality or in non-specific symptoms are likely to be missed. The increasing availability of whole-genome sequences has opened new research avenues such as reverse genetics for their investigation. Our aim was to characterize the genetic load of 15 European breeds using data from the 1000 bull genomes consortium and prove that widespread harmful mutations remain to be detected.
We listed 2489 putative deleterious variants (in 1923 genes) segregating at a minimal frequency of 5 % in at least one of the breeds studied. Gene enrichment analysis showed major enrichment for genes related to nervous, visual and auditory systems, and moderate enrichment for genes related to cardiovascular and musculoskeletal systems. For verification purposes, we investigated the phenotypic consequences of a frameshift variant in the retinitis pigmentosa-1 gene segregating in several breeds and at a high frequency (27 %) in Normande cattle. As described in certain human patients, clinical and histological examination revealed that this mutation causes progressive degeneration of photoreceptors leading to complete blindness in homozygotes. We established that the deleterious allele was even more frequent in the Normande breed before 1975 (>40 %) and has been progressively counter-selected likely because of its associated negative effect on udder morphology. Finally, using identity-by-descent analysis we demonstrated that this mutation resulted from a unique ancestral event that dates back to ~2800 to 4000 years.
We provide a list of mutations that likely represent a substantial part of the genetic load of domestication in European cattle. We demonstrate that they accumulated non-randomly and that genes related to cognition and sensory functions are particularly affected. Finally, we describe an ancestral deleterious variant segregating in different breeds causing progressive retinal degeneration and irreversible blindness in adult animals.
- Ingenuity Pathway Analysis
- Deleterious Mutation
- Frameshift Mutation
- Genomic Selection
- Pupillary Light Reflex
Domestication has had a dramatic effect on the genomes of plant and animal species. Reduction of environmental pressure combined with rapid growth of populations after strong demographic bottlenecks have resulted in relaxation of purifying selection and accumulation of deleterious mutations [1–5]. In the last 150 years, this phenomenon termed “the cost of domestication” has been particularly amplified in cattle because of the creation of breeds from a limited number of founder animals, overuse of a few elite sires with artificial insemination (AI) and intensive selection on specific traits. As a consequence most bovine breeds experience regular outbreaks of recessive genetic defects. With the advent of high-throughput genotyping and next-generation sequencing, efficient methods have been developed to identify the underlying mutations in record time and with a limited number of available cases [6, 7]. However, such approaches rely on the observation of affected animals with distinctive symptoms. It can be anticipated that the genetic defects reported so far represent only the tip of the iceberg and that many recessive mutations resulting in embryonic mortality or in non-specific symptoms, which can be confounded with those of common diseases, remain to be discovered. In addition to the influence of genetic drift and hitch-hiking, the frequency of some deleterious mutations, which would be detrimental in the wild, may have been involuntarily increased by artificial selection on behavior, coat color, morphological or production traits. This is the case for example for double-muscling, which causes dystocia [8, 9], and for a series of mutations under balancing selection [10, 11].
The increasing number of available whole-genome sequences (WGS) has recently opened new research avenues such as reverse genetics to investigate recessive defects. This strategy seems particularly suitable in cattle for which the sequencing of the most influential AI bulls of each breed (e.g. 1000 bull genomes project ) enables the identification of the vast majority of the non-private deleterious mutations that segregate in these populations. Furthermore, the inclusion of a subset of these polymorphisms into single nucleotide polymorphism (SNP) chips that are used for genomic selection should facilitate the detection of homozygotes (or of a deficit in homozygotes) for deleterious alleles among the tens of thousands of animals genotyped each year. In parallel, crossing genotyping data with pedigree information should enable the detection of severely affected homozygotes among animals that are born from at risk matings, which would not have been genotyped for genomic selection purposes. Finally, bovine populations provide an important number of cases available for sampling and experimental study to evaluate the functional consequences of the mutation, which is hardly possible in humans.
The purpose of this study was twofold: (i) to characterize the genetic load of 15 beef and dairy breeds using whole-genome sequencing data from the 1000 bull genomes consortium  and (ii) to prove that widespread harmful mutations remain to be detected in our cattle populations by characterizing the effect of a frameshift mutation in the retinitis pigmentosa-1 (RP1) gene, which segregates in Normande cattle and other European breeds.
Blood and ear biopsies were collected by veterinarians or by agricultural technicians licensed by the French Departmental Breeding Establishments [Etablissements Départementaux de l’Elevage (EDE)] during routine ear tagging, sampling for annual prophylaxis, paternity testing and genotyping for genetic defects or genomic selection. Ophthalmologic examinations and electroretinograms were approved after ethical evaluation by the ComERC committee (Ethical Committee for Clinical Research at the French Veterinary School of Maisons Alfort (ENVA) (Saisine n°14-01-2015) and performed under sedation controlled by a veterinarian specialized in cattle.
Invasive procedures were performed post-mortem after slaughter for meat production. Experiments reported in this work comply with the ethical guidelines of the French National Institute for Agricultural Research (INRA). All the samples and data analyzed were obtained with the permission of breeders, breeding organizations and research group providers.
Details on animals used for each analysis are presented in Additional file 1: Table S1.
Filtering of variants from whole-genome sequence data and prediction of their phenotypic consequences
Variants were selected from whole-genome sequence data of 1147 bulls from the 1000 bull genome project (for details on variant calling see Daetwyler et al. ). Briefly, raw reads were filtered and trimmed on chastity and quality score, then aligned on the UMD3.1 bovine reference sequence assembly using BWA . SNPs and InDel were called from pooled bam files using SAMtools 0.1.18 mpileup . Variants were then annotated using Ensembl Variant Effect Predictor . Frequencies and allele counts were calculated across and within breeds using vcftools “freq” and “count” options . Filtering consisted in selecting biallelic variants which (i) were predicted to cause a loss of protein function (i.e. affecting initiator codons, splice acceptor or donor sites, or causing a frameshift, a stop loss or gain, or a missense with a SIFT score of 0 ), (ii) had a calling quality (QUAL) above 30, (iii) presented a mapping quality (MQ) score of 59 or 60, (iv) had less than 5 % of animals with missing genotypes, and (v) had a minor allele frequency (MAF) higher than 5 % for at least one breed with a minimum of 20 individuals in the dataset (which means that alleles observed only once were not considered). It should be noted that variants with a SIFT score less than 0.05 are generally considered deleterious. In this study, we chose to retain only missense variants with a SIFT score of 0 to reduce possible artifacts. Furthermore, including missense deleterious variants with a SIFT score between 0.01 and 0.05 would have resulted in considering approximately one fourth of the total number of bovine genes, thus preventing subsequent gene enrichment analysis. In addition, each variant was manually checked to eliminate artifacts due to (i) adjacent substitutions within the same codon which are not accounted for in variant annotation, (ii) errors of annotations after comparing gene annotations from the UCSC and Ensembl genome browsers (http://genome.ucsc.edu, http://www.ensembl.org) annotations, (iii) repeated sequences (downloaded at http://genome.ucsc.edu, accession 21/10/2015). Only variants with a known official gene symbol were considered in the subsequent analyses.
To anticipate the phenotypic consequences of the mutations, annotations were completed by information on genetic syndromes associated with mutations within the same genes in humans (Online Mendelian Inheritance in Man, OMIM; http://www.omim.org) and mouse (Mammalian Phenotypes; http://www.informatics.jax.org) (see Additional file 2: Table S2).
Gene set enrichment analysis
Gene enrichment analysis was performed using Ingenuity Pathway Analysis software (http://www.ingenuity.com/products/ipa/, ). We focused on “top canonical pathways” with a p value lower than 0.01 and “diseases and bio functions” annotations with a p value lower than 0.05. Annotations related to cancer and the general pathways entitled “skin lesion” and “liver lesion” were not considered since their results suffer from a bias. Pathways related to drug metabolism, which were not relevant for this study, were also eliminated. In addition, a unique keyword was assigned to each significantly enriched function annotation, with particular attention paid to the attribution of keywords related to subcellular portions, cell types and organs rather than to general processes. When possible, keywords appearing only once were regrouped with higher order items (e.g. cell type changed for organ, or process changed for the category defined by IPA) or with the predefined IPA “categories”. Frequency of keywords was used to set the size of the words in the word cloud representation.
Twenty-three pure and crossbred Normande cows from the INRA experimental facility of Le Pin-au-Haras (Normandy, France) with genotypes available for the frameshift mutation in the retinitis pigmentosa-1 (RP1) gene were selected for ocular examination (see Additional file 1: Table S1). These consisted in four homozygous mutants, nine heterozygous and ten homozygous wild type animals. All these animals were in good health conditions and with no signs of systemic disease at the time of the study. Genotypes were not disclosed to the veterinarian to exclude any bias of personal interpretation. Examinations were performed indoors under ambient light. Visual performance was evaluated by the menace response test and dazzle and pupillary light reflexes (direct and indirect) were assessed with a Finoff transilluminator. Slit-lamp biomicroscopy (Kowa SL-15, Kowa Company) was performed before and after pupillary dilation using one drop of 1 % tropicamide. Fundi were examined by indirect ophthalmoscopy (Heine Omega 100, Heine Optotechnik, GmbH & CoKG) with 28-D and 20-D lenses.
Electroretinogram tests were performed on two 5.5-years old cows, one homozygous wild type and one homozygous mutant, with a Retiport (Roland Consult, Brandenburg, Germany), under sedation (Xylazine 0.04 mg/kg IM) and after pupillary dilation (tropicamide eyedrops) and blocking of the auriculopalpebral nerve by subcutaneous injection of lidocaine. Topical tetracaïne eyedrops were used to anesthetize the ocular surfaces and corneas were lubricated by topical application of sodium hyaluronate 1.2 % during the test. The following responses were recorded: rod response before and after dark adaptation for 20 min, following a dim white stimulus (0.02–0.03 cd/m2/s), mixed response following four bright white flashes (2–3 cd/m2/s) at a rate of 0.1 Hz and cone response following four bright white flashes (2–3 cd/m2/s) at a rate of 5 Hz.
Genotyping of Normande cattle that were reported to the French National Observatory of Bovine genetic abnormalities for progressive loss of vision
Twenty-eight Normande cows that were reported to the French National Observatory of Bovine genetic Abnormalities (ONAB) with signs of progressive loss of vision and blindness were genotyped for the RP1 frameshift mutation. Genomic DNA was extracted from blood or ear biopsies using a standard phenol–chloroform protocol and genotyped by PCR and Sanger sequencing for the Chr14 g.23995411_23995412insA mutation. PCR primers were designed from the UMD3.1 bovine genome assembly with Primer3 software ) to span the insertion (left: TGCACAGGAAACCATATTGC and right: TTGCCCTAGTTGTGACATGC). Reactions were performed using the Go-Taq Flexi DNA Polymerase (Promega) according to the manufacturer’s instructions on a Mastercycler pro thermocycler (Eppendorf). The resulting amplicons were purified and bidirectionally sequenced by Eurofins MWG (Germany) using conventional Sanger sequencing. Polymorphisms were detected with the novoSNP software .
Estimation of the allelic frequency of the RP1 frameshift mutation
The RP1 frameshift mutation was included in the Illumina EuroG10K custom SNP chip, which is routinely used for genomic selection in France. Thus, in addition to the 1000 genome dataset, genotypes for this mutation were available for 53,279 Holstein, 40,548 Montbéliarde, 12,106 Normande, 1634 Abondance, 1005 Red Pied Lowland, 698 Tarentaise, 579 Simmental, 507 Vosgienne, and 296 Brown Swiss animals.
Post-mortem ocular examination and histological analysis
The eyes of two homozygous mutant cull cows (aged 8 years) that displayed a severe phenotype and two control cows (one 6-year-old heterozygous Normande and one 8-year-old homozygous wild-type Holstein) were collected post-mortem at the slaughterhouse (SVA Trémorel, France). One eye was dissected on site to perform a visual examination of the eye’s fundus and to collect the retina and choroid. Samples were immediately frozen in liquid nitrogen and stored at –80 °C until DNA extraction. The second eye was injected with 2.5 ml formaldehyde and fixed by a 24-h incubation in the same solution. The retina and choroid were subsequently dehydrated in a graded ethanol series, cleared with xylene and embedded in paraffin. Microtome sections. (5 µm, Leica RM2245) were stained with haematoxylin, eosin and saffron (HES). Digital images were obtained with the NanoZoomer 2.0- HT slide scanner (Hamamatzu).
Association with recorded traits
At the time of the analysis, Illumina EuroG10K SNP chip genotype data were available for 7439 Normande animals, which had their sire genotyped with the Illumina BovineSNP50 chip. The Illumina EuroG10K SNP chip comprises the RP1 frameshift mutation as well as more than 10,000 common SNPs with the Illumina BovineSNP50 chip. Using Fimpute  we were able to attribute a genotype for the RP1 polymorphism to 48,715 additional animals that were previously genotyped with the BovineSNP50 chip. The complete dataset comprised 11,986 Normande cows with phenotype information on three coat colour phenotypes (proportion of white areas on the body; proportion of white areas on the face, and brindling intensity) and on 28 traits that are routinely recorded for genetic evaluations (milk yield, fat content, protein content, fat percent, protein percent, cell score, clinical mastitis, milking speed, stature, chest width, body depth, width at pin bone, rump angle, rear legs side view, rear legs rear view, back muscle, fillet muscle, rear muscle, fore udder attachment, rear udder height, udder balance, teat orientation, front teat distance, udder support, udder depth development, interval between calving and first insemination, fertility at insemination of lactating cows, fertility at insemination of heifers). Associations between the RP1 frameshift polymorphism and traits were tested using GCTA . Phenotypes were adjusted for environmental effects which were estimated in the national genetic evaluation procedure and assumed to reflect the genetic effect of the animal and a random residual effect. Therefore, the analysis model included only an overall mean, a polygenic effect, the effect of the genotype at the RP1 frameshift polymorphism, and a residual. The polygenic effect was estimated by using a genomic relationship matrix that was derived from 43,801 SNPs on the Illumina BovineSNP50 chip. Finally, a Bonferroni correction that consisted in dividing the p value by the total number of tests performed was applied to account for multiple-testing.
Across-breed identity-by-descent analysis around the RP1 frameshift mutation
Identity-by-descent (IBD) analysis was performed to (i) test for the existence of one versus multiple mutation events in the different breeds and (ii) estimate the date of the origin of the mutation(s). For that purpose, phased genotypes for a 1.3-Mb region (Chr14:23474270-24643266; corresponding to the smallest IBD homozygous region detected in the genome of one homozygous mutant Normande AI bull, named Diametre (FR5388012666) were extracted for 35 heterozygous and three homozygous carrier animals identified among the 1147 animals from run 4 of the 1000 bull genomes project. Phasing was performed within the framework of the 1000 bull genome project using BEAGLE [12, 23].
Within this homozygous region of Diametre’s genome, 9448 SNPs with the highest quality score (QUAL = 999) were selected and considered as reference haplotypes. For each animal, the rate of homozygous genotypes in opposition with the chosen reference genotypes was calculated for sliding windows of 100 SNPs. Then, the number of individuals that had at least 5 % of inconsistencies with the reference haplotype was counted and attributed to the position of the 51th SNP in each window. This level of 5 % of inconsistencies was chosen to account for the low sequence coverage of certain animals and for the putative occurrence of de novo mutations over time in the vicinity of the old frameshift mutation. The IBD block around the frameshift mutation was finally defined by windows for which none of the carriers displayed 5 % or more of inconsistencies with the haplotypes of Diametre. For control purposes, the same process was applied to a set of 38 non-carrier animals that were randomly selected among individuals belonging to the same breeds as the carriers.
Estimation of the age of the RP1 frameshift mutation according to the size of the IBD segment shared among breeds
We considered that two animals that shared an IBD segment of size c (c being the size in Morgan) inherited this segment from a common ancestor that lived 1/(2 c) generations ago. We assumed that, on average, 1 cM corresponds to 1,000,000 bp and that generation intervals range from 5 to 7 years, depending on the breeding system (natural mating population or modern breeding schemes).
Analysis of the changes in frequency of the RP1 frameshift mutation in the Normande breed
To study the changes in allelic frequency of the RP1 frameshift mutation in the Normande breed, first we developed a haplotype test using 15,515 animals (1077 homozygous carrier, 6363 heterozygous and 8075 homozygous wild type animals) that were genotyped for this variant with the Illumina EuroG10K custom SNP chip and had been phased and imputed for the Illumina BovineSNP50 markers within the framework of the French genomic selection . The haplotype was fixed to 50 SNPs between SNPs ARS-BFGL-BAC-12159 (Chr14 position 22587081 bp) and ARS-BFGL-NGS-36089 (Chr14 position 25698286). We identified 691 haplotypes among which 12.45 % were associated to the frameshift mutation, 83.79 % were not associated with it and 3.76 % were classified as undetermined (i.e. detected in both homozygous carriers and non-carriers). When applied to all the Normande cattle phased Illumina BovineSNP50 genotyped data, 97.3 % of the haplotypes were assigned a status (27.26 % were associated to the frameshift mutation, 83.79 % were associated to the wild type allele, 1.03 % undetermined) and 2.47 % were classified as not documented due to lack of haplotype information among the animals genotyped with the EuroG10K chip. From these haplotype-allele associations, we estimated the genotypes for 1375 phased Normand AI bulls (born between 1975 and 2015) for the RP1 frameshift polymorphism. Allelic frequencies were calculated over time for sliding windows of 7 years (i.e. on average one generation) after removing haplotypes without information.
During domestication, deleterious mutations have accumulated in non random sets of genes
A total of 1923 genes carried a deleterious mutation of which 566 counted two or more. A screening of phenotype databases revealed that 908 genes (1144 variants) were associated to at least one mammalian phenotype in laboratory animals (MGI database) and 375 (corresponding to 395 variants) with an inherited syndrome in humans (OMIM database). From our own interpretation, almost two-thirds of these syndromes described in mouse and humans presented a phenotype that would have been difficult to detect by the different national observatories for genetic defects in cattle (i.e. those affecting metabolism, immunity, cognition) (see Additional file 2: Table S2).
In this selection, we also retrieved five variants that were previously reported to cause major phenotypes in cattle. These comprise mutations that have been favored by artificial selection (i.e. p.Q204X mutation in MSTN for double muscling in Charolais ), or with a severe but invisible phenotype (i.e. p.R12X and p.R55X nonsense mutations in SCL37A2 and CWC15 for embryonic mortality in Montbeliarde  and Jersey , respectively), or with a mild phenotype that is present in several breeds (i.e. p.R238X mutation in FMO3 for trimethylaminuria or “fishy-off flavor” of milk  and p.W80X mutation in BCO2 for the “yellow color” of milk and fat ). These examples validate that such variants which are deleterious to the protein function may exist and segregate at moderate to high frequencies in cattle breeds.
In addition, we retrieved important genes for neuro-cognitive functions which are associated with behavioral disorders in humans such as mental retardation, schizophrenia, bipolar disorder or autism. Among other examples, we can cite genes coding for glutamate receptors (GRIK2 and GRM7), glutamate being the most important neurotransmitter in the brain, semaphorins which are involved, among other functions, in axon guidance (SEMA3A, SEMA3B, SEMA4A, SEMA4D and SEMA5A), calcium voltage channel subunits (CACNA1C and CACNB2), a receptor for neuroregulin 1 (ERBB4), the neurexin-3-alpha protein which has an important role in neuronal function (NRXN3), and a post synaptic protein (SYNGAP1) [36–45]. Interestingly, only four of the 407 genes from this cluster co-localized with selective sweeps in cattle, chicken, rabbit and/or pig (i.e. GRIK2 in rabbit, SEMA3 in pig and chicken, and ERBB4 and CACNA1C in cattle) [46–49].
Therefore, whereas genes that are involved in sensory functions and cognition represent obvious targets of domestication [5, 46], it is unlikely that the variants reported here were positively selected during domestication or subsequent selection processes. More likely, our results indicate that, in a domestic context, such mutations were more tolerated than mutations affecting other systems which are of primary importance for production, reproduction and survival.
Analyzing the frequency of the keywords that were assigned to each IPA annotation also revealed two additional clusters related to cardiovascular (12.8 %), and muscle and skeletal systems (12.3 %), which might be associated with positive selection. These two clusters comprised genes that are associated with selective sweeps and/or production traits such as MSTN for double-muscling, CCNL1 for reduced birth weight, THADA for body weight variation, GOLGA4 for stature, and LCORL for stature and skeletal development in cattle [49–51] and FANCA for the cardiac system and NR3C2 for blood pressure in horse . Note that neuro-muscular interactions were also underlined by two IPA canonical pathways (p value <0.01), i.e. the “Agrin interactions at neuromuscular junctions” and “nNOS signaling in skeletal muscles”, which are both involved in neuro-muscular diseases and muscle dystrophies. Finally, two other canonical pathways, the “complement system” (part of the innate immune system of an organism) pathways and the “t-RNA charging” (regrouping the key enzymes of protein translation mechanisms in mitochondria) are relevant because of their involvement in two functions that are subject to important selection pressure i.e. immunity and energetic metabolism via mitochondrial function [52, 53] (see Additional file 3: Table S7).
A frameshift mutation in the RP1 gene causes progressive blindness in cattle
A good candidate for phenotype characterization
Frequency of the RP1 frameshift allele among breeds represented in run4 of the 1000 bull genomes project
Frequency of the RP1 frameshift allele in % (number of genomes per breed with available genotype)
Beef Booster Composite
In humans and mouse, similar truncation mutations in the RP1 gene, which encodes a microtubule associated protein that is essential for the organization of the outer segments of the photoreceptors in the retina, have been reported to cause autosomal dominant and recessive retinitis pigmentosa [54–57]. Retinitis pigmentosa is a form of inherited degenerative retinal disorder that is characterized by progressive death of photoreceptor cells. Symptoms typically start with loss of night vision due to degeneration of rod-photoreceptors, followed by degeneration of cone-photoreceptors leading to loss of central vision and eventually to complete blindness .
Genotype frequencies for the RP1 frameshift variant among 23 Normande founder bulls and 28 animals reported to ONAB for loss of vision
Normande bulls in the 1000 bull genomes dataset (n = 23)
Loss of vision phenotype (n = 28)
8.7 % (n = 2)
50.0 % (n = 14)
43.5 % (n = 10)
35.7 % (n = 10)
47.8 % (n = 11)
14.3 % (n = 4)
Clinical and histological tests revealed symptoms of retinal degeneration in homozygous mutants
Results of the ocular tests for the Normande dairy cattle examined
Normal vision and eyes fundus
2 (<2 years)
7 (6.1 years)
7 (5.7 years)
Mild unilateral or bilateral focal retinal degeneration with preserved vision
2 (5.8 years)
3 (7.2 years)
Bilateral retinal degeneration with marked visual deficit
1 (4.5 years)
Bilateral retinal degeneration with blindness
1 (5.5 years)
Number of animals
Electroretinogram results i.e. values of the amplitudes and the culminating times of a- and b-waves in one control and one affected animal
Culminating time (msec)
Finally, to characterize this phenotype at the tissue level, we collected the retinas of two additional homozygous carriers (aged 8 years) and two control cows (one 6-year-old heterozygous Normande and one 8-year-old homozygous wild-type Holstein) after slaughter (Fig. 3). In concordance with previous analyses on the eyes’ fundus and retinal function, histological analyses revealed a total absence of photoreceptor outer segments along with a marked thinning and disorganization of the outer nuclear layer with very few remaining nuclei.
Taken together, these results provide strong support that the RP1 frameshift mutation causes a recessive loss of vision in bovine cattle. The phenotype observed is similar to the description in humans with a late onset of the disease due to progressive degeneration of the photoreceptors. Very few genetic conditions that affect eyesight have been reported in cattle  and, to our knowledge, this is the first time that a mutation causing retinal degeneration is reported in this species. Indeed, while, in the past, several cases of progressive retinal degeneration were reported in Holstein cows, their genetic etiology has not been confirmed so far [60, 61].
IBD analysis reveals a unique and ancestral mutation event
Genotype frequencies for the RP1 frameshift polymorphism from the EuroG10K genotyping results
Red pied lowland
The observation of this old variant at low to moderate frequencies in numerous bovine breeds could be explained by a combination of genetic drift and absence of or a very limited negative counter-selection due to the late onset of the defect. Nevertheless, the high frequency of this mutation observed in Normande cattle (27.7 % in the genotyped population for genomic selection) was particularly striking and led us to perform additional investigations to test for positive selection (either directly or mediated by hitch-hiking) in this breed.
Results of association studies between the RP1 frameshift mutation and 31 traits routinely evaluated for 11,986 Normande cows
p value (Bonferroni)
Front teat distance
Milk protein content
Milk fat content
In this work, we have drawn a list of putative deleterious mutations which are not rare (frequency higher than 5 %) in at least one of 15 investigated European bovine breeds. We showed that these variants, which likely represent a substantial part of the genetic load of domestication in cattle, did not accumulate randomly. Our results reveal that mutations in genes, which are involved in cognition or sensorial functions for which little or no selection pressure exists in domesticated animals, were more tolerated than mutations that affect other systems, which are of primary importance for production, reproduction and survival. Among these variants, we described an ancestral frameshift mutation in RP1 which segregates in numerous breeds and causes progressive retinal degeneration. To our knowledge, this is the most ancient and widespread mutation causing a recessive genetic defect in cattle reported to date. This example illustrates that our approach can help to unravel variants that are yet to be discovered and are the cause of unselected but debilitating phenotypes in domestic animals. We are confident that the phenotypic characterization of a number of the variants reported here will offer interesting results in the near future.
AC conceived and coordinated the study. AC, PM, SC and DB designed the study. SC, DD and ED performed clinical examinations. PM and AB performed histological analyses. PM, AC and MB analyzed WGS data. PM and AC performed gene enrichment analysis. PM, AC, SB and MC participated to sampling. MC, SB and CDB provided phenotype or pedigree information. CG, MCD and AAB performed DNA extraction and genotyping by PCR-sequencing. LG performed quality control of the SNP chip genotyping data and provided DNA samples from carriers of the RP1 mutation. PM, A and SF performed phasing and haplotype testing. AM performed imputation and association studies. PM, AC, SC and DB wrote the manuscript. All authors read and approved the final manuscript.
The authors are grateful to the breeders of the “GAEC de l’Araucaria” for their hospitality, Maelle Philippe (EVOLUTION) for providing samples, Albéric Valais (Normande breed Organization) for providing phenotype information, and the partners of the 1000 bull genomes consortium for the excellent collaboration. This study is part of the BOVANO project (ANR-14-CE19-0011) funded by the French Agence Nationale de la Recherche and Apisgene. P. Michot is recipient of a PhD. Grant from ALLICE and Apis Gène.
The authors declare that they have no competing interests.
Availability of data and material
The datasets supporting the conclusions of this article are included within the article and its additional files.
P. Michot is recipient of a PhD. Grant from ALLICE and Apis Gène. This study is part of the BOVANO project (ANR-14-CE19-0011) funded by the French Agence Nationale de la Recherche and Apisgene.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Lu J, Tang T, Tang H, Huang J, Shi S, Wu C. The accumulation of deleterious mutations in rice genomes: a hypothesis on the cost of domestication. Trends Genet. 2006;22:126–31.View ArticlePubMedGoogle Scholar
- Cruz F, Vila C, Webster MT. The legacy of domestication: accumulation of deleterious mutations in the dog genome. Mol Biol Evol. 2008;25:2331–6.View ArticlePubMedGoogle Scholar
- Nabholz B, Sarah G, Sabot F, Ruiz M, Adam H, Nidelet S, et al. Transcriptome population genomics reveals severe bottleneck and domestication cost in the African rice (Oryza glaberrima). Mol Ecol. 2014;23:2210–27.View ArticlePubMedGoogle Scholar
- Koenig D, Jimenez-Gomez JM, Kimura S, Fulop D, Chitwood DH, Headland LR, et al. Comparative transcriptomics reveals patterns of selection in domesticated and wild tomato. Proc Natl Acad Sci USA. 2013;110:2655–62.View ArticleGoogle Scholar
- Schubert M, Jónsson H, Chang D, Der Sarkissian C, Ermini L, Ginolhac A, et al. Prehistoric genomes reveal the genetic foundation and cost of horse domestication. Proc Natl Acad Sci USA. 2014;111:E5661–9.View ArticlePubMedPubMed CentralGoogle Scholar
- Charlier C, Coppieters W, Rollin F, Desmecht D, Agerholm JS, Cambisano N, et al. Highly effective SNP-based association mapping and management of recessive defects in livestock. Nat Genet. 2008;40:449–54.View ArticlePubMedGoogle Scholar
- Michot P, Fantini O, Braque R, Allais-Bonnet A, Saintilan R, Grohs C, et al. Whole-genome sequencing identifies a homozygous deletion encompassing exons 17 to 23 of the integrin beta 4 gene in a Charolais calf with junctional epidermolysis bullosa. Genet Sel Evol. 2015;47:37.View ArticlePubMedPubMed CentralGoogle Scholar
- Grobet L, Martin LJ, Poncelet D, Pirottin D, Brouwers B, Riquet J, et al. A deletion in the bovine myostatin gene causes the double–muscled phenotype in cattle. Nat Genet. 1997;17:71–4.View ArticlePubMedGoogle Scholar
- Arthur PF, Makarechian M, Price MA. Incidence of dystocia and perinatal calf mortality resulting from reciprocal crossing of double-muscled and normal cattle. Can Vet J. 1988;29:163–7.PubMedPubMed CentralGoogle Scholar
- Fasquelle C, Sartelet A, Li W, Dive M, Tamma N, Michaux C, et al. Balancing selection of a frame-shift mutation in the MRC2 gene accounts for the outbreak of the crooked tail syndrome in Belgian Blue cattle. PLoS Genet. 2009;5:e1000666.View ArticlePubMedPubMed CentralGoogle Scholar
- Sartelet A, Druet T, Michaux C, Fasquelle C, Géron S, Tamma N, et al. A splice site variant in the bovine RNF11 gene compromises growth and regulation of the inflammatory response. PLoS Genet. 2012;8:e1002581.View ArticlePubMedPubMed CentralGoogle Scholar
- Daetwyler HD, Capitan A, Pausch H, Stothard P, van Binsbergen R, Brøndum RF, et al. Whole-genome sequencing of 234 bulls facilitates mapping of monogenic and complex traits in cattle. Nat Genet. 2014;46:858–65.View ArticlePubMedGoogle Scholar
- Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25:1754–60.View ArticlePubMedPubMed CentralGoogle Scholar
- Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The sequence alignment/map format and SAMtools. Bioinformatics. 2009;25:2078–9.View ArticlePubMedPubMed CentralGoogle Scholar
- McLaren W, Pritchard B, Rios D, Chen Y, Flicek P, Cunningham F. Deriving the consequences of genomic variants with the Ensembl API and SNP Effect Predictor. Bioinformatics. 2010;26:2069–70.View ArticlePubMedPubMed CentralGoogle Scholar
- Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, et al. The variant call format and VCFtools. Bioinformatics. 2011;27:2156–8.View ArticlePubMedPubMed CentralGoogle Scholar
- Kumar P, Henikoff S, Ng PC. Predicting the effects of coding nonsynonymous variants on protein function using the SIFT algorithm. Nat Protoc. 2009;4:1073–81.View ArticlePubMedGoogle Scholar
- Calvano SE, Xiao W, Richards DR, Felciano RM, Baker HV, Cho RJ, et al. A network-based analysis of systemic inflammation in humans. Nature. 2005;437:1032–7.View ArticlePubMedGoogle Scholar
- Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol. 2000;132:365–86.PubMedGoogle Scholar
- Weckx S, Del-Favero J, Rademakers R, Claes L, Cruts M, De Jonghe P, et al. novoSNP, a novel computational tool for sequence variation discovery. Genome Res. 2005;15:436–42.View ArticlePubMedPubMed CentralGoogle Scholar
- Sargolzaei M, Chesnais JP, Schenkel FS. A new approach for efficient genotype imputation using information from relatives. BMC Genomics. 2014;15:478.View ArticlePubMedPubMed CentralGoogle Scholar
- Yang J, Lee SH, Goddard ME, Visscher PM. GCTA: a tool for genome-wide complex trait analysis. Am J Hum Genet. 2011;88:76–82.View ArticlePubMedPubMed CentralGoogle Scholar
- Browning BL, Browning SR. A unified approach to genotype imputation and haplotype-phase inference for large data sets of trios and unrelated individuals. Am J Hum Genet. 2009;84:210–23.View ArticlePubMedPubMed CentralGoogle Scholar
- Boichard D, Guillaume F, Baur A, Croiseau P, Rossignol MN, et al. Genomic selection in French dairy cattle. Anim Prod Sci. 2016;52:115–20.View ArticleGoogle Scholar
- Boitard S, Boussaha M, Capitan A, Rocha D, Servin B. Uncovering adaptation from sequence data: lessons from genome resequencing of four cattle breeds. Genetics. 2016;203:433–50.View ArticlePubMedGoogle Scholar
- Grobet L, Poncelet D, Royo LJ, Brouwers B, Pirottin D, Michaux C, et al. Molecular definition of an allelic series of mutations disrupting the myostatin function and causing double-muscling in cattle. Mamm Genome. 1998;9:210–3.View ArticlePubMedGoogle Scholar
- Fritz S, Capitan A, Djari A, Rodriguez SC, Barbat A, Baur A, et al. Detection of haplotypes associated with prenatal death in dairy cattle and identification of deleterious mutations in GART, SHBG and SLC37A2. PLoS One. 2013;8:e65550.View ArticlePubMedPubMed CentralGoogle Scholar
- Sonstegard TS, Cole JB, VanRaden MP, Van Tassell CP, Null DJ, Schroeder SG, et al. Identification of a nonsense mutation in CWC15 associated with decreased reproductive efficiency in Jersey cattle. PLoS One. 2013;8:e54872.View ArticlePubMedPubMed CentralGoogle Scholar
- Lunden A, Marklund S, Gustafsson V, Aderdersson L. A nonsense mutation in the FMO3 gene underlies fishy off-flavor in cow’s milk. Genome Res. 2002;12:1885–8.View ArticlePubMedPubMed CentralGoogle Scholar
- Berry SD, Davis SR, Beattie EM, Thomas NL, Burrett AK, Ward HE, et al. Mutation in bovine-carotene oxygenase 2 affects milk color. Genetics. 2009;182:923–6.View ArticlePubMedPubMed CentralGoogle Scholar
- Robertson NG, Skvorak AB, Yin Y, Weremowicz S, Johnson KR, Kovatch KA, et al. Mapping and characterization of a novel cochlear gene in human and in mouse: a positional candidate gene for a deafness disorder, DFNA9. Genomics. 1997;46:345–54.View ArticlePubMedGoogle Scholar
- Cohen-Salmon M, Mattei MG, Petit C. Mapping of the otogelin gene (OTGN) to mouse chromosome 7 and human chromosome 11p14.3: a candidate for human autosomal recessive nonsyndromic deafness DFNB18. Mamm Genome. 1999;10:520–2.View ArticlePubMedGoogle Scholar
- Yariz KO, Duman D, Seco CZ, Dallman J, Huang M, Peters TA, et al. Mutations in OTOGL, encoding the inner ear protein otogelin-like, cause moderate sensorineural hearing loss. Am J Hum Genet. 2012;91:872–82.View ArticlePubMedPubMed CentralGoogle Scholar
- Wang A, Liang Y, Fridell RA, Probst FJ, Wilcox ER, Touchman JW, et al. Association of unconventional myosin MYO15 mutations with human nonsyndromic deafness DFNB3. Science. 1998;280:1447–51.View ArticlePubMedGoogle Scholar
- Verpy E, Masmoudi S, Zwaenepoel I, Leibovici M, Hutchin TP, Del Castillo I, et al. Mutations in a new gene encoding a protein of the hair bundle cause non-syndromic deafness at the DFNB16 locus. Nat Genet. 2001;29:345–9.View ArticlePubMedGoogle Scholar
- Motazacker MM, Rost BR, Hucho T, Garshasbi M, Kahrizi K, Ullmann R, et al. A defect in the ionotropic glutamate receptor 6 gene (GRIK2) is associated with autosomal recessive mental retardation. Am J Hum Genet. 2007;81:792–8.View ArticlePubMedPubMed CentralGoogle Scholar
- Kandaswamy R, McQuillin A, Curtis D, Gurling H. Allelic association, DNA resequencing and copy number variation at the metabotropic glutamate receptor GRM7 gene locus in bipolar disorder. Am J Med Genet B Neuropsychiatr Genet. 2014;165:365–72.View ArticlePubMed CentralGoogle Scholar
- Shifman MI, Selzer ME. Differential expression of class 3 and 4 semaphorins and netrin in the lamprey spinal cord during regeneration. J Comp Neurol. 2007;501:631–46.View ArticlePubMedPubMed CentralGoogle Scholar
- Kantor DB, Chivatakarn O, Peer KL, Oster SF, Inatani M, Hansen MJ, et al. Semaphorin 5A is a bifunctional axon guidance cue regulated by heparan and chondroitin sulfate proteoglycans. Neuron. 2004;44:961–75.View ArticlePubMedGoogle Scholar
- Mosca-Boidron AL, Gueneau L, Huguet G, Goldenberg A, Henry C, Gigot N, et al. A de novo microdeletion of SEMA5A in a boy with autism spectrum disorder and intellectual disability. Eur J Hum Genet. 2015;24:838–43.View ArticlePubMedPubMed CentralGoogle Scholar
- Ferreira MA, O’Donovan MC, Meng YA, Jones IR, Ruderfer DM, Jones L, et al. Collaborative genome-wide association analysis supports a role for ANK3 and CACNA1C in bipolar disorder. Nat Genet. 2008;40:1056–8.View ArticlePubMedPubMed CentralGoogle Scholar
- Breitenkamp AF, Matthes J, Nass RD, Sinzig J, Lehmkuhl G, Nürnberg P, et al. Rare mutations of CACNB2 found in autism spectrum disease-affected families alter calcium channel function. PLoS One. 2014;9:e95579.View ArticlePubMedPubMed CentralGoogle Scholar
- Stefansson H, Petursson H, Sigurdsson E, Steinthorsdottir V, Bjornsdottir S, Sigmundsson T, et al. Neuregulin 1 and susceptibility to schizophrenia. Am J Hum Genet. 2002;71:877–92.View ArticlePubMedPubMed CentralGoogle Scholar
- Brown SM, Clapcote SJ, Millar JK, Torrance HS, Anderson SM, Walker R, et al. Synaptic modulators Nrxn1 and Nrxn3 are disregulated in a Disc1 mouse model of schizophrenia. Mol Psychiatry. 2011;16:585–7.View ArticlePubMedGoogle Scholar
- Jeyabalan N, Clement JP. SYNGAP1: mind the Gap. Front Cell Neurosci. 2016;10:32.View ArticlePubMedPubMed CentralGoogle Scholar
- Carneiro M, Rubin CJ, Di Palma F, Albert FW, Alfoldi J, Barrio AM, et al. Rabbit genome analysis reveals a polygenic basis for phenotypic change during domestication. Science. 2014;345:1074–9.View ArticlePubMedGoogle Scholar
- Rubin CJ, Megens HJ, Martinez Barrio A, Maqbool K, Sayyab S, Schwochow D, et al. Strong signatures of selection in the domestic pig genome. Proc Natl Acad Sci USA. 2012;109:19529–36.View ArticlePubMedPubMed CentralGoogle Scholar
- Rubin CJ, Zody MC, Eriksson JR, Meadows J, Sherwood E, Webster MT, et al. Whole-genome resequencing reveals loci under selection during chicken domestication. Nature. 2010;464:587–91.View ArticlePubMedGoogle Scholar
- Qanbari S, Pausch H, Jansen S, Somel M, Strom TM, Fries R, et al. Classic selective sweeps revealed by massive sequencing in cattle. PLoS Genet. 2014;10:e1004148.View ArticlePubMedPubMed CentralGoogle Scholar
- Ramey HR, Decker JE, McKay SD, Rolf MM, Schnabel RD, Taylor JF. Detection of selective sweeps in cattle using genome-wide SNP data. BMC Genomics. 2013;14:382.View ArticlePubMedPubMed CentralGoogle Scholar
- Flori L, Fritz S, Jaffrezic F, Boussaha M, Gut I, Heath S, et al. The genome response to artificial selection: a case study in dairy cattle. PLoS One. 2009;4:e6595.View ArticlePubMedPubMed CentralGoogle Scholar
- Mayilyan KR. Complement genetics, deficiencies, and disease associations. Protein Cell. 2012;3:487–96.View ArticlePubMedPubMed CentralGoogle Scholar
- Diodato D, Ghezzi D, Tiranti V. The mitochondrial aminoacyl tRNA synthetases: genes and syndromes. Int J Cell Biol. 2014;2014:787956.View ArticlePubMedPubMed CentralGoogle Scholar
- Pierce EA, Quinn T, Meehan T, McGee TL, Berson EL, Dryja TP. Mutations in a gene encoding a new oxygen-regulated photoreceptor protein cause dominant retinitis pigmentosa. Nat Genet. 1999;22:248–54.View ArticlePubMedGoogle Scholar
- Audo I, Mohand-Saïd S, Dhaenens CM, Germain A, Orhan E, Antonio A, et al. RP1 and autosomal dominant rod-cone dystrophy: novel mutations, a review of published variants, and genotype–phenotype correlation. Hum Mutat. 2011;33:73–80.View ArticlePubMedGoogle Scholar
- Liu Q, Collin RW, Cremers FP, den Hollander AI, van den Born LI, Pierce EA. Expression of wild-type Rp1 protein in Rp1 knock-in mice rescues the retinal degeneration phenotype. PLoS One. 2012;7:e43251.View ArticlePubMedPubMed CentralGoogle Scholar
- El Shamieh S, Boulanger-Scemama E, Lancelot ME, Antonio A, Démontant V, Condroyer C, et al. Targeted next generation sequencing identifies novel mutations in RP1 as a relatively common cause of autosomal recessive rod-cone dystrophy. Biomed Res Int. 2015;2015:485624.View ArticlePubMedPubMed CentralGoogle Scholar
- Hartong DT, Berson EL, Dryja TP. Retinitis pigmentosa. Lancet. 2006;368:1795–809.View ArticlePubMedGoogle Scholar
- Murgiano L, Jagannathan V, Calderoni V, Joechler M, Gentile A, Drögemüller C. Looking the cow in the eye: deletion in the NID1 gene is associated with recessive inherited cataract in Romagnola cattle. PLoS One. 2014;9:e110628.View ArticlePubMedPubMed CentralGoogle Scholar
- Bradley R, Terlecki S, Clegg FG. The pathology of a retinal degeneration in Friesian cows. J Comp Pathol. 1982;92:69–83.View ArticlePubMedGoogle Scholar
- Stehmann SM, Rebhun WC, Riis RC. Progessive retinal atrophy in related cattle. The Bovine Practitioner. 1987;20:195–7.Google Scholar