Genetic expression in interspecific hybrids

Genetic expression in interspecific hybrids

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

Referring to interspecific hybrids, I have the following two questions:-

Quoting from wikipedia:-

The offspring of an interspecific cross are very often sterile; thus, hybrid sterility prevents the movement of genes from one species to the other, keeping both species distinct. Sterility is often attributed to the different number of chromosomes the two species have, for example donkeys have 62 chromosomes, while horses have 64 chromosomes, and mules and hinnies have 63 chromosomes. Mules, hinnies, and other normally sterile interspecific hybrids cannot produce viable gametes, because differences in chromosome structure prevent appropriate pairing and segregation during meiosis, meiosis is disrupted, and viable sperm and eggs are not formed. However, fertility in female mules has been reported with a donkey as the father.

1) The sterility of hybrids prevents interspecific gene exchange and is necessary in case the hybrid has odd number of total chromosomes (the hybridized species has odd and even pairs of chromosomes individually), where equal meiotic division would not be possible, and is also necessary to avoid sex determination problems that may arise from such hybridizations genetically. What is the molecular (genetic) reason behind the sterility? In other words, why is the expression of primary sexual organs and its further proliferation to effect reproduction not possible, although the genes required for it are already present in the hybrids? Is this due to epigenetic mechanisms or is the reason inherent in the hybrid nature of the genome?

2) This might probably sound a bit naive, but why are these hybrids viable? They have 2 non-homologous haploid pairs of chromosomes which have a huge variety(compared to two homologous pairs) of different non-allelic genes. Why are all the essential functions not hindered by the presence of such considerably unrelated set of chromosomes?

Probably related to the question is the fact that certain allopolyploids are viable and become an entirely new species, but others are not. Is it because here are 2 sets of each combining species' chromosomes (and hence allowing proper meiotic pairing) or is there some other reason pertaining to the nature of the combined sets of chromosome?

Gene expression regulation in the context of mouse interspecific mosaic genomes

Accumulating evidence points to the mosaic nature of the mouse genome. However, little is known about the way the introgressed segments are regulated within the context of the recipient genetic background. To address this question, we have screened the testis transcriptome of interspecific recombinant congenic mouse strains (IRCSs) containing segments of Mus spretus origin at a homozygous state in a Mus musculus background.


Most genes (75%) were not transcriptionally modified either in the IRCSs or in the parent M. spretus mice, compared to M. musculus. The expression levels of most of the remaining transcripts were 'dictated' by either M. musculus transcription factors ('trans-driven' 20%), or M. spretus cis-acting elements ('cis-driven' 4%). Finally, 1% of transcripts were dysregulated following a cis-trans mismatch. We observed a higher sequence divergence between M. spretus and M. musculus promoters of strongly dysregulated genes than in promoters of similarly expressed genes.


Our study indicates that it is possible to classify the molecular events leading to expressional alterations when a homozygous graft of foreign genome segments is made in an interspecific host genome. The inadequacy of transcription factors of this host genome to recognize the foreign targets was clearly the major path leading to dysregulation.

Cis-regulatory evolution in prokaryotes revealed by interspecific archaeal hybrids

The study of allele-specific expression (ASE) in interspecific hybrids has played a central role in our understanding of a wide range of phenomena, including genomic imprinting, X-chromosome inactivation, and cis-regulatory evolution. However across the hundreds of studies of hybrid ASE, all have been restricted to sexually reproducing eukaryotes, leaving a major gap in our understanding of the genomic patterns of cis-regulatory evolution in prokaryotes. Here we introduce a method to generate stable hybrids between two species of halophilic archaea, and measure genome-wide ASE in these hybrids with RNA-seq. We found that over half of all genes have significant ASE, and that genes encoding kinases show evidence of lineage-specific selection on their cis-regulation. This pattern of polygenic selection suggested species-specific adaptation to low phosphate conditions, which we confirmed with growth experiments. Altogether, our work extends the study of ASE to archaea, and suggests that cis-regulation can evolve under polygenic lineage-specific selection in prokaryotes.

Conflict of interest statement

The authors declare that they have no competing interests.


Generation of stable H .…

Generation of stable H . volcanii × H . mediterranei hybrids. ( A…

Regulatory divergence between archaeal hybrids…

Regulatory divergence between archaeal hybrids is revealed by ASE analysis. ( A )…

Detection of lineage-specific selection and…

Detection of lineage-specific selection and differential fitness in low phosphate conditions. ( A…


The term hybrid is derived from Latin hybrida, used for crosses such as of a tame sow and a wild boar. The term came into popular use in English in the 19th century, though examples of its use have been found from the early 17th century. [1] Conspicuous hybrids are popularly named with portmanteau words, starting in the 1920s with the breeding of tiger–lion hybrids (liger and tigon). [2]

Animal and plant breeding Edit

From the point of view of animal and plant breeders, there are several kinds of hybrid formed from crosses within a species, such as between different breeds. [3] Single cross hybrids result from the cross between two true-breeding organisms which produces an F1 hybrid (first filial generation). The cross between two different homozygous lines produces an F1 hybrid that is heterozygous having two alleles, one contributed by each parent and typically one is dominant and the other recessive. Typically, the F1 generation is also phenotypically homogeneous, producing offspring that are all similar to each other. [4] Double cross hybrids result from the cross between two different F1 hybrids (i.e., there are four unrelated grandparents). [5] Three-way cross hybrids result from the cross between an F1 hybrid and an inbred line. Triple cross hybrids result from the crossing of two different three-way cross hybrids. [6] Top cross (or "topcross") hybrids result from the crossing of a top quality or pure-bred male and a lower quality female, intended to improve the quality of the offspring, on average. [7]

Population hybrids result from the crossing of plants or animals in one population with those of another population. These include interspecific hybrids or crosses between different breeds. [8]

In horticulture, the term stable hybrid is used to describe an annual plant that, if grown and bred in a small monoculture free of external pollen (e.g., an air-filtered greenhouse) produces offspring that are "true to type" with respect to phenotype i.e., a true-breeding organism. [9]

Biogeography Edit

Hybridisation can occur in the hybrid zones where the geographical ranges of species, subspecies, or distinct genetic lineages overlap. For example, the butterfly Limenitis arthemis has two major subspecies in North America, L. a. arthemis (the white admiral) and L. a. astyanax (the red-spotted purple). The white admiral has a bright, white band on its wings, while the red-spotted purple has cooler blue-green shades. Hybridisation occurs between a narrow area across New England, southern Ontario, and the Great Lakes, the "suture region". It is at these regions that the subspecies were formed. [10] Other hybrid zones have formed between described species of plants and animals.

Genetics Edit

From the point of view of genetics, several different kinds of hybrid can be distinguished. [11] [12] A genetic hybrid carries two different alleles of the same gene, where for instance one allele may code for a lighter coat colour than the other. [11] [12] A structural hybrid results from the fusion of gametes that have differing structure in at least one chromosome, as a result of structural abnormalities. [11] [12] A numerical hybrid results from the fusion of gametes having different haploid numbers of chromosomes. [11] [12] A permanent hybrid results when only the heterozygous genotype occurs, as in Oenothera lamarckiana, [13] because all homozygous combinations are lethal. [11] [12] In the early history of genetics, Hugo de Vries supposed these were caused by mutation. [14] [15]

Taxonomy Edit

From the point of view of taxonomy, hybrids differ according to their parentage. Hybrids between different subspecies (such as between the Dog and Eurasian wolf) are called intra-specific hybrids. [16] Interspecific hybrids are the offspring from interspecies mating [17] these sometimes result in hybrid speciation. [18] Intergeneric hybrids result from matings between different genera, such as between sheep and goats. [19] Interfamilial hybrids, such as between chickens and guineafowl or pheasants, are reliably described but extremely rare. [20] Interordinal hybrids (between different orders) are few, but have been made with the sea urchin Strongylocentrotus purpuratus (female) and the sand dollar Dendraster excentricus (male). [21]

Expression of parental traits Edit

When two distinct types of organisms breed with each other, the resulting hybrids typically have intermediate traits (e.g., one plant parent has red flowers, the other has white, and the hybrid, pink flowers). [22] Commonly, hybrids also combine traits seen only separately in one parent or the other (e.g., a bird hybrid might combine the yellow head of one parent with the orange belly of the other). [22]

Mechanisms of reproductive isolation Edit

Interspecific hybrids are bred by mating individuals from two species, normally from within the same genus. The offspring display traits and characteristics of both parents, but are often sterile, preventing gene flow between the species. [23] Sterility is often attributed to the different number of chromosomes between the two species. For example, donkeys have 62 chromosomes, horses have 64 chromosomes, and mules or hinnies have 63 chromosomes. Mules, hinnies, and other normally sterile interspecific hybrids cannot produce viable gametes, because differences in chromosome structure prevent appropriate pairing and segregation during meiosis, meiosis is disrupted, and viable sperm and eggs are not formed. However, fertility in female mules has been reported with a donkey as the father. [24]

A variety of mechanisms limit the success of hybridisation, including the large genetic difference between most species. Barriers include morphological differences, differing times of fertility, mating behaviors and cues, and physiological rejection of sperm cells or the developing embryo. Some act before fertilization others after it. [25] [26] [27] [28]

In plants, some barriers to hybridisation include blooming period differences, different pollinator vectors, inhibition of pollen tube growth, somatoplastic sterility, cytoplasmic-genic male sterility and structural differences of the chromosomes. [29]

Speciation Edit

A few animal species are the result of hybridization. The Lonicera fly is a natural hybrid. The American red wolf appears to be a hybrid of the gray wolf and the coyote, [31] although its taxonomic status has been a subject of controversy. [32] [33] [34] The European edible frog is a semi-permanent hybrid between pool frogs and marsh frogs its population requires the continued presence of at least one of the parent species. [35] Cave paintings indicate that the European bison is a natural hybrid of the aurochs and the steppe bison. [36] [37]

Plant hybridization is more commonplace compared to animal hybridization. Many crop species are hybrids, including notably the polyploid wheats: some have four sets of chromosomes (tetraploid) or six (hexaploid), while other wheat species have (like most eukaryotic organisms) two sets (diploid), so hybridization events likely involved the doubling of chromosome sets, causing immediate genetic isolation. [38]

Hybridization may be important in speciation in some plant groups. However, homoploid hybrid speciation (not increasing the number of sets of chromosomes) may be rare: by 1997, only 8 natural examples had been fully described. Experimental studies suggest that hybridization offers a rapid route to speciation, a prediction confirmed by the fact that early generation hybrids and ancient hybrid species have matching genomes, meaning that once hybridization has occurred, the new hybrid genome can remain stable. [39]

Many hybrid zones are known where the ranges of two species meet, and hybrids are continually produced in great numbers. These hybrid zones are useful as biological model systems for studying the mechanisms of speciation. Recently DNA analysis of a bear shot by a hunter in the North West Territories confirmed the existence of naturally-occurring and fertile grizzly–polar bear hybrids. [40]

Hybrid vigour Edit

Hybridization between reproductively isolated species often results in hybrid offspring with lower fitness than either parental. However, hybrids are not, as might be expected, always intermediate between their parents (as if there were blending inheritance), but are sometimes stronger or perform better than either parental lineage or variety, a phenomenon called heterosis, hybrid vigour, or heterozygote advantage. This is most common with plant hybrids. [41] A transgressive phenotype is a phenotype that displays more extreme characteristics than either of the parent lines. [42] Plant breeders use several techniques to produce hybrids, including line breeding and the formation of complex hybrids. An economically important example is hybrid maize (corn), which provides a considerable seed yield advantage over open pollinated varieties. Hybrid seed dominates the commercial maize seed market in the United States, Canada and many other major maize-producing countries. [43]

In a hybrid, any trait that falls outside the range of parental variation (and is thus not simply intermediate between its parents) is considered heterotic. Positive heterosis produces more robust hybrids, they might be stronger or bigger while the term negative heterosis refers to weaker or smaller hybrids. [44] Heterosis is common in both animal and plant hybrids. For example, hybrids between a lion and a tigress ("ligers") are much larger than either of the two progenitors, while "tigons" (lioness × tiger) are smaller. Similarly, the hybrids between the common pheasant (Phasianus colchicus) and domestic fowl (Gallus gallus) are larger than either of their parents, as are those produced between the common pheasant and hen golden pheasant (Chrysolophus pictus). [45] Spurs are absent in hybrids of the former type, although present in both parents. [46]

Anthropogenic hybridization Edit

Hybridization is greatly influenced by human impact on the environment, [47] through effects such as habitat fragmentation and species introductions. [48] Such impacts make it difficult to conserve the genetics of populations undergoing introgressive hybridization. Humans have introduced species worldwide to environments for a long time, both intentionally for purposes such as biological control, and unintentionally, as with accidental escapes of individuals. Introductions can drastically affect populations, including through hybridization. [12] [49]

Management Edit

There is a kind of continuum with three semi-distinct categories dealing with anthropogenic hybridization: hybridization without introgression, hybridization with widespread introgression (backcrossing with one of the parent species), and hybrid swarms (highly variable populations with much interbreeding as well as backcrossing with the parent species). Depending on where a population falls along this continuum, the management plans for that population will change. Hybridization is currently an area of great discussion within wildlife management and habitat management. Global climate change is creating other changes such as difference in population distributions which are indirect causes for an increase in anthropogenic hybridization. [47]

Conservationists disagree on when is the proper time to give up on a population that is becoming a hybrid swarm, or to try and save the still existing pure individuals. Once a population becomes a complete mixture, the goal becomes to conserve those hybrids to avoid their loss. Conservationists treat each case on its merits, depending on detecting hybrids within the population. It is nearly impossible to formulate a uniform hybridization policy, because hybridization can occur beneficially when it occurs "naturally", and when hybrid swarms are the only remaining evidence of prior species, they need to be conserved as well. [47]

Genetic mixing and extinction Edit

Regionally developed ecotypes can be threatened with extinction when new alleles or genes are introduced that alter that ecotype. This is sometimes called genetic mixing. [50] Hybridization and introgression, which can happen in natural and hybrid populations, of new genetic material can lead to the replacement of local genotypes if the hybrids are more fit and have breeding advantages over the indigenous ecotype or species. These hybridization events can result from the introduction of non-native genotypes by humans or through habitat modification, bringing previously isolated species into contact. Genetic mixing can be especially detrimental for rare species in isolated habitats, ultimately affecting the population to such a degree that none of the originally genetically distinct population remains. [51] [52]

Effect on biodiversity and food security Edit

In agriculture and animal husbandry, the Green Revolution's use of conventional hybridization increased yields by breeding "high-yielding varieties". The replacement of locally indigenous breeds, compounded with unintentional cross-pollination and crossbreeding (genetic mixing), has reduced the gene pools of various wild and indigenous breeds resulting in the loss of genetic diversity. [54] Since the indigenous breeds are often well-adapted to local extremes in climate and have immunity to local pathogens, this can be a significant genetic erosion of the gene pool for future breeding. Therefore, commercial plant geneticists strive to breed "widely adapted" cultivars to counteract this tendency. [55]

In animals Edit

Mammals Edit

Familiar examples of equid hybrids are the mule, a cross between a female horse and a male donkey, and the hinny, a cross between a female donkey and a male horse. Pairs of complementary types like the mule and hinny are called reciprocal hybrids. [56] Among many other mammal crosses are hybrid camels, crosses between a bactrian camel and a dromedary. [57] There are many examples of felid hybrids, including the liger.

The first known instance of hybrid speciation in marine mammals was discovered in 2014. The clymene dolphin (Stenella clymene) is a hybrid of two Atlantic species, the spinner and striped dolphins. [58] In 2019, scientists confirmed that a skull found 30 years earlier was a hybrid between the beluga whale and narwhal dubbed the narluga. [59]

Birds Edit

Cagebird breeders sometimes breed bird hybrids known as mules between species of finch, such as goldfinch × canary. [60]

Amphibians Edit

Among amphibians, Japanese giant salamanders and Chinese giant salamanders have created hybrids that threaten the survival of Japanese giant salamanders because of competition for similar resources in Japan. [61]

Fish Edit

Among fish, a group of about fifty natural hybrids between Australian blacktip shark and the larger common blacktip shark was found by Australia's eastern coast in 2012. [62]

Russian sturgeon and American paddlefish were hybridized in captivity when sperm from the paddlefish and eggs from the sturgeon were combined, unexpectedly resulting in viable offspring. This hybrid is called a sturddlefish. [63] [64]

Invertebrates Edit

Among insects, so-called killer bees were accidentally created during an attempt to breed a strain of bees that would both produce more honey and be better adapted to tropical conditions. It was done by crossing a European honey bee and an African bee. [65]

The Colias eurytheme and C. philodice butterflies have retained enough genetic compatibility to produce viable hybrid offspring. [66] Hybrid speciation may have produced the diverse Heliconius butterflies, [67] but that is disputed. [68]

In plants Edit

Plant species hybridize more readily than animal species, and the resulting hybrids are fertile more often. Many plant species are the result of hybridization, combined with polyploidy, which duplicates the chromosomes. Chromosome duplication allows orderly meiosis and so viable seed can be produced. [69]

Plant hybrids are generally given names that include an "×" (not in italics), such as Platanus × acerifolia for the London plane, a natural hybrid of P. orientalis (oriental plane) and P. occidentalis (American sycamore). [70] [71] The parent's names may be kept in their entirety, as seen in Prunus persica × Prunus americana, with the female parent's name given first, or if not known, the parent's names given alphabetically. [72]

Plant species that are genetically compatible may not hybridize in nature for various reasons, including geographical isolation, differences in flowering period, or differences in pollinators. Species that are brought together by humans in gardens may hybridize naturally, or hybridization can be facilitated by human efforts, such as altered flowering period or artificial pollination. Hybrids are sometimes created by humans to produce improved plants that have some of the characteristics of each of the parent species. Much work is now being done with hybrids between crops and their wild relatives to improve disease-resistance or climate resilience for both agricultural and horticultural crops. [73]

Some crop plants are hybrids from different genera (intergeneric hybrids), such as Triticale, × Triticosecale, a wheat–rye hybrid. [74] Most modern and ancient wheat breeds are themselves hybrids bread wheat, Triticum aestivum, is a hexaploid hybrid of three wild grasses. [30] Several commercial fruits including loganberry (Rubus × loganobaccus) [75] and grapefruit (Citrus × paradisi) [76] are hybrids, as are garden herbs such as peppermint (Mentha × piperita), [77] and trees such as the London plane (Platanus × acerifolia). [78] [79] Among many natural plant hybrids is Iris albicans, a sterile hybrid that spreads by rhizome division, [80] and Oenothera lamarckiana, a flower that was the subject of important experiments by Hugo de Vries that produced an understanding of polyploidy. [13]

A sterile hybrid between Trillium cernuum and T. grandiflorum [ citation needed ]

An ornamental lily hybrid known as Lilium 'Citronella' [81]

Sterility in a non-polyploid hybrid is often a result of chromosome number if parents are of differing chromosome pair number, the offspring will have an odd number of chromosomes, which leaves them unable to produce chromosomally-balanced gametes. [82] While that is undesirable in a crop such as wheat, for which growing a crop that produces no seeds would be pointless, it is an attractive attribute in some fruits. Triploid bananas and watermelons are intentionally bred because they produce no seeds and are also parthenocarpic. [83]

In humans Edit

There is evidence of hybridisation between modern humans and other species of the genus Homo. In 2010, the Neanderthal genome project showed that 1–4% of DNA from all people living today, apart from most Sub-Saharan Africans, is of Neanderthal heritage. Analyzing the genomes of 600 Europeans and East Asians found that combining them covered 20% of the Neanderthal genome that is in the modern human population. [84] Ancient human populations lived and interbred with Neanderthals, Denisovans, and at least one other extinct Homo species. [85] Thus, Neanderthal and Denisovan DNA has been incorporated into human DNA by introgression. [86]

In 1998, a complete prehistorical skeleton found in Portugal, the Lapedo child, had features of both anatomically modern humans and Neanderthals. [87] Some ancient human skulls with especially large nasal cavities and unusually shaped braincases represent human-Neanderthal hybrids. A 37,000- to 42,000-year-old human jawbone found in Romania's Oase cave contains traces of Neanderthal ancestry [a] from only four to six generations earlier. [89] All genes from Neanderthals in the current human population are descended from Neanderthal fathers and human mothers. [90] A Neanderthal skull unearthed in Italy in 1957 reveals Neanderthal mitochondrial DNA, which is passed on through only the maternal lineage, but the skull has a chin shape similar to modern humans. It is proposed that it was the offspring of a Neanderthal mother and a human father. [91]

Folk tales and myths sometimes contain mythological hybrids the Minotaur was the offspring of a human, Pasiphaë, and a white bull. [92] More often, they are composites of the physical attributes of two or more kinds of animals, mythical beasts, and humans, with no suggestion that they are the result of interbreeding, as in the centaur (man/horse), chimera (goat/lion/snake), hippocamp (fish/horse), and sphinx (woman/lion). [93] The Old Testament mentions a first generation of half-human hybrid giants, the Nephilim, [94] [95] while the apocryphal Book of Enoch describes the Nephilim as the wicked sons of fallen angels and attractive women. [96]


The main objective of this study was to document patterns of gene expression divergence in first (normal × dwarf) and second-generation hybrid crosses (backcross: [normal × dwarf] × normal], and compare them with pure normal and dwarf parental forms, at both embryonic and juvenile ontogenetic stages. More specifically, under the assumption that similar genes involved in the adaptive divergence of these species are also responsible for driving their reproductive isolation ( Nolte et al. 2009), we predicted more evidence of misexpression at the juvenile compared with the embryonic stage. In general, our results supported this prediction as we observed that few genes differed in average expression in hybrids compared with parentals at the embryonic stage, whereas many more did so at the juvenile stage. Secondly, we predicted more evidence of hybrid misexpression in backcross hybrids and the fact that nonadditivity was more prevalent in backcross compared with F1-hybrids supported this prediction. Lastly, extreme transgressivity of several key developmental genes was observed in backcross embryos. This emphasizes that, at the transcriptomic level, intrinsic hybrid misexpression may also play a role in explaining reproductive isolation of dwarf and normal whitefish. Below, we discuss the potential implications of those results, also considering the limitations of the data.

Patterns of Inheritance in Hybrids

Strikingly, patterns of inheritance were quite distinct between backcross and F1-hybrids, and d/a values were not correlated between them. Under an additive model of inheritance, we would also have expected F1-hybrid to be the midvalue of their parents and backcross to be closer to the normal phenotype (with which they share 75% of their genome). This was not the case, as more genes differentiated F1-hybrids to dwarf whereas, and to a lesser extent, more genes differentiated backcross to normal, a result also exemplified by the asymmetry in the direction of dominance in both F1-hybrids (normal dominance) and backcross (dwarf dominance). This idiosyncratic result cannot be ignored, yet is difficult to interpret beyond the fact that, as it has been emphasized many times before, gene expression is a complex phenotype, whose behavior is hard to predict and whose inheritance often does not follow simple Mendelian rules ( Rockman and Kruglyak 2006). Indeed, this asymmetry of gene expression divergence toward one parent is apparently quite common in F1-hybrids event though the underlying mechanistic reasons responsible for this trend are poorly understood (Drosophila: Ranz et al. 2004 Gibson et al. 2004 Mus: Rottscheidt and Harr 2007 and Salvelinus: Mavárez et al. submitted).

The Transcriptomic Basis of Ecological (Extrinsic) Reproductive Isolation Factors

Previous gene expression studies ( Derome et al. 2006 St-Cyr et al. 2008 Nolte et al. 2009) combined with physiological data ( Trudel et al. 2001) have shown that changes in the expression of metabolic genes are largely responsible for the physiological adaptation to distinct whitefish benthic (normal) and limnetic (dwarf) niches. Notably, a suite of six key metabolic genes (glyceraldehyde-3-phosphate dehydrogenase, Fructose–bisphosphate aldolase A, Beta-enolase, Trypsin-1 precursor, Cytochrome c oxidase polypeptide VIa, and Nucleoside diphosphate kinase) was identified as consistently divergent between normal and dwarf whitefish ( Nolte et al. 2009). We may then hypothesize that misexpression for those metabolic genes could contribute, to an atypical physiological phenotype and to an inferior, ecologically maladapted, individual. Here, we found that, in F1-hybrids juveniles, those genes mostly showed an intermediate pattern of expression ( supplementary table 1 , Supplementary Material online) and no transgressivity compared with parents. In backcross hybrids, two of those genes (G3PDH, FBPA A) showed additivity of expression, the rest being slightly nonadditive, whereas none revealed transgressivity.

According to the ecological theory of adaptive radiation, intermediate hybrid phenotypes may be selected against if no suitable ecological niche for them exists in nature ( Schluter 2000). Recent work in sticklebacks ( Gow et al. 2007) and cichlids ( van der Sluijs et al. 2008) has shown such environment driven natural selection may be key in explaining incipient population divergence. As such, reproductive isolation of lake whitefish could be at least partly seen as a by-product of divergent selection acting on metabolic genes. Admittedly, a clear demonstration of the association between the expression of such genes and phenotypic variation between dwarf and normal whitefish is lacking, and we are currently conducting quantitative trait loci, expression quantitative trait loci, and gene mapping studies toward this end (Renaut S, Nolte AW, Bernatchez L, unpublished data).

A Possible Role of Transgressivity in Reproductive Isolation

Gene expression in hybrids was generally more variable than parental, both at embryonic and juvenile stages. One might argue that the patterns of variance observed are confounded by a different number of families used in each treatment. Although this cannot be entirely ruled out, it is noteworthy that backcross individuals, who showed the highest level of variance in gene expression, consisted of a single female crossed to five males, relative to all other treatments, which consisted of many half-sib families. This family effect probably also explains the relatively large variance of dwarf whitefish, which comprised both half-sib families and many wild-caught natural families. It then seems unlikely that this factor would explain the general increased variance observed in the backcross group. Alternatively, recombination can release hidden variation and generate transgressive phenotypes ( Rieseberg et al. 1999, 2003 Mallet 2007). Populations are known to accumulate cryptic variation only revealed under certain genotypic or environmental conditions ( Le Rouzic and Carlborg 2007). Recently, Landry et al. (2007) have illustrated how in hybrids, the regulation of coevolved cis regulatory regions and trans transcription factors could be disrupted and lead to increased phenotypic novelties in hybrids. This may explain why many whitefish genes showed increased variance in expression and transgressivity in hybrids and yet were not differentially expressed between parentals. In fact, all the highly transgressive transcripts presented in table 4 were not differentially expressed in any comparisons.

Transgressive segregation may in some cases create fitter phenotypes (e.g., hybrid species resulting from selected “hopeful monsters”, Barton 2001 Mallet 2007). Conversely, it also underlies postzygotic isolation mechanisms such that transgressive hybrids often suffer a highly reduced survival ( Barton 2001 Coyne and Orr 2004). We propose that the overall patterns of transgressivity we observed, including misexpression of several key developmental genes, may contribute to abnormal hybrid development and increased embryonic mortality identified by Lu and Bernatchez (1998) and also Rogers et al. 2007 as a plausible postzygotic reproductive isolation mechanism. Namely, five of the nine transcripts identified as highly transgressive in embryos and involved in protein folding and mRNA translation are especially good candidates because knockdown mutants in D. rerio for those genes are known to show visible embryonic defect and almost invariably die prior to, or early after, hatching ( Amsterdam et al. 2004). This proportion (five of nine or 54%) was also significantly higher (P < 0.001, one tailed Fisher's Exact test) than the actual proportion of transcripts in the whole embryo data set that matched to essential genes identified in D. rerio (257 of 4,950 or 5%). In fact, the abnormal phenotypes described in the D. rerio study closely match our own observation that a large fraction of backcross eggs (35%) started to show visible defects (asymmetric axial body plan, small eyes, heart not beating, deformed tail) 15 days after our sampling and eventually die prior to hatching (Renaut S, unpublished data). A previous study on reproductive isolation in lake whitefish also showed that a large fraction of the backcross progeny died around the same developmental time ( Rogers and Bernatchez 2006). Of course, there is a leap between linking a knockdown mutation completely obliterating a gene product (as it is the case in the D. rerio study) and a simple increase in biological variation. Yet, it is noteworthy that these key developmental genes are the most transgressive of 4,950 surveyed in the embryo data set and were generally significantly more transgressive than expected by chance. Moreover, it is plausible that hybrid genetic combinations creating even greater misexpression for those genes may have caused early lethality (prior to our sampling) and thus may have reduced our ability to pick out such abnormal phenotypes. Consequently, the increased patterns of variance observed in hybrid embryos are likely to be conservative estimates.

Gene Expression Studies of Speciation

Most gene expression differentiation we observed was between normal and dwarf parental forms rather than between hybrids. These results contrast with many recent gene expression–speciation studies that have identified pervasive nonadditive patterns of gene expression in first generation hybrids (see recent reviews by Landry et al. 2007 Ortiz-Barrientos et al. 2007). Namely, our study brings new lights into gene expression studies of speciation for two main reasons. Firstly, the bulk of the work, done mostly in Drosophila, particularly in the melanogaster group ( Michalak and Noor 2003 Ranz et al. 2004 Landry et al. 2005 Haerty and Singh 2006 Moehring et al. 2007), and to a lesser extent in Xenopus ( Malone et al. 2007) and Mus ( Rottscheidt and Harr 2007) involves biological species that have diverged millions of years ago (e.g., D. simulansDrosophila mauritiana: 0.93 Ma, D. simulansD. melanogaster: 5.1 Ma ( Tamura et al. 2004), Xenopus laevis–Xenopus muelleri: >20 Ma ( Evans et al. 2004), Mus musculus subspecies: 0.3–1.0 Ma ( Boursot et al. 1996)]. Because genetic incompatibilities continue to accumulate over time even after complete reproductive isolation has been established, this complicates the identification of the loci that initially led to the divergence event ( Mallet 2006). In these studies, most hybrids are known to be poorly fit, sterile, or simply inviable (Ranz et al. 2004) rendering it difficult to disentangle whether gene expression misexpression is the cause rather than the consequence of hybrid inviability. In contrast, in young diverging lineages such as whitefish species pairs that diverged 12,000–15,000 years ago ( Bernatchez 2004), much less genetic divergence is expected due to frequent gene flow or recent common ancestry. The effects of hybridization may then be subtler and the genetic changes identified more likely to be involved in the very early steps of reproductive isolation. Secondly, even though the effect of early reproductive barriers may be more important in later hybrid generations ( Barton 2001), most recent studies have focused on first generation hybrids. Clearly, we identified different genes and patterns of inheritance in first and second-generation hybrids. Moreover, backcross hybrids have also revealed increased nonadditivity as well as transgressive expression of essential developmental genes. To our knowledge, no gene expression studies of speciation in natural systems have previously used genomewide gene expression data to compare the expression profile of second-generation hybrids with that of parental lineages.

CROP IMPROVEMENT | Doubled Haploid Production

Chromosome Elimination

During attempts to produce interspecific hybrids between cultivated barley and a related wild species Hordeum bulbosum a quite high frequency of haploid plants was found. The fertilization of barley ovules occurred using the H. bulbosum pollen and stimulated embryo development. However, during early development, the chromosomes from H. bulbosum were preferentially eliminated during cell division, leaving most embryos with only a single set of barley chromosomes. Elimination also occurred during endosperm development so that by 2 weeks after pollination, the embryos had to be rescued and grown into small plants on culture media. The keys to initially understanding that this was elimination and not parthenogenesis were the high frequencies of haploids, differences in response depending upon the ploidy level of the two parents, and similar results from reciprocal crosses ( Table 2 ).

Table 2 . Genome balance and elimination in progeny obtained from crosses between barley (Hordeum vulgare), designated V and H. bulbosum, designated B, at various ploidy levels and reciprocal crosses

Female parentMale parentConstitution of progeny plants

When crosses were made at the same ploidy level, e.g., 2x by 2x or 4x by 4x, elimination occurred to produce haploid barley plants. When 2x barley was crossed with 4x H. bulbosum, 3x triploid species hybrids were produced. Thus, chromosome balance was thought to be important in stability or elimination and it was later shown that this was most likely a balance of genes on specific chromosomes numbered 2 and 3. The exciting aspect of this system of haploid production is that nearly all barley genotypes would hybridize with selected diploid H. bulbosum plants which could be multiplied vegetatively and distributed. Thus, it would be suitable for barley breeding or improvement programs and allowed production of fairly large numbers of haploid plants. The efficiency of this method in barley has been greatly improved by selection of better H. bulbosum pollinators and improved crossing procedures.

Since the haploid plants were sterile, chromosome doubling procedures, usually colchicine treatment, were required as spontaneous doubling frequency was very low. Efficient chromosome doubling methods were developed so that 70–80% of the haploids were partially doubled and produced genetically homozygous seed.

Subsequent attempts to pollinate hexaploid bread wheats with H. bulbosum pollen showed that chromosome elimination occurs, resulting in rescued embryos developing into wheat haploids. However, most wheat genotypes could not be efficiently fertilized with H. bulbosum pollen due to the presence of KR (rye crossability) genes. Thus, while useful in producing some doubled haploids, the system was not suitable for breeding programs.

Chromosome elimination has been shown to be quite common among interspecific hybrids. Bread wheats pollinated with corn pollen and in combination with auxin treatment, induced good seed set. After embryo rescue, this system also produced wheat haploids across most genotypes. Thus, corn pollination of wheat is now quite commonly used for haploid production in wheat breeding and research programs. While not as successful, pollen from Sorghum, teosinte (Zea mexicana), pearl millet (Pennisetum glaucum), and Italian ryegrass (Lolium multiflorum) will also produce haploids in wheat. Corn pollination on oats (Avena sativa) results in oat haploids as well as some corn addition chromosomes in oats when elimination is not complete.

Chromosome elimination has been observed among many interspecific crosses in Hordeum as well as in some other genera and is an indication that chromosome elimination may occur quite often following wide hybridization. The rescue of haploid embryos is a key to detection while a good source of pollen is essential for developing a haploid production system.


Plant materials and hybridization

Three species including C. azalea (2n = 30), C. chekiangoleosa (2n = 30) and C. amplexicaulis (2n = 30) were used in this study (Fig. 1). According to morphological and molecular studies, both C. azalea and C. chekiangoleosa belong to the Sect. Camellia of Camellia, while C. amplexicaulis belongs to the Sect. Archecamellia of Camellia [32, 33]. Hybridizations were carried out by Palm Eco-Town Development Co. Ltd. in 2007 following the technique described by Gao et al. [34]. For all the hybridization experiments, C. azalea was served as the female parent, and the other two species were served as the male parents. All the plants in this study are grown in a same greenhouse of Palm Eco-Town Development Co. Ltd. at Guangzhou, China. To improve pollination efficiency, pollens from different individuals of the two paternal species were collected together, respectively. The mixed pollens were then used to pollinate the flowers of C. azalea plants. So, the F1 hybrids may be not from the identical parents, but their parents came from individuals of one wild population, respectively. Finally, two F1 hybrid series, C. azalea (♀) × C. chekiangoleosa (♂) and C. azalea (♀) × C. amplexicaulis (♂), representing intra-sectional and inter-sectional hybrids, were successfully obtained. Flower buds of the F1 hybrids and their parents at same stage were harvested and frozen in liquid nitrogen immediately, then transferred to − 80 °C refrigerator for storage.

RNA extraction and sequencing

Total RNA was extracted from the flower buds using the RNAprep Puree Plant Kit DP441 (TIANGEN, Beijing, China) according to the manufacturer’s instructions. For each species and hybrid, three biologic replicates (from three individuals, respectively) were set up as parallel experiments. Nanodrop 1000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA) was used to detect the quantity and quality of RNA. RNA-seq library was constructed for each sample. In total, 15 libraries were constructed, and then paired-end (2 × 150 bp) sequenced using Illumina HiSeq X-ten platform (Illumina Inc., San Diego, CA, USA) by Beijing Genomics Institute (BGI, Shenzhen, China) with the standard Illumina RNA-seq protocols.

Mapping transcriptome reads to the reference genome

Clean reads were obtained by removing reads with adapter contamination and ploy-Ns (≥ 5%) as well as low quality reads with over 20% of low-quality bases (Phred < 15). Since the reference genomes of the species in our study were not available, filtered reads from the parental libraries were first mapped to the genome of C. sinensis var. assamica [35] using STAR software [36] with default parameters, and only uniquely mapped reads were retained. According to the ITS sequences (Additional file 1: Table S3), the genetic distances between C. azalea, C. chekiangoleosa, C. amplexicaulis and C. sinensis was 0.044, 0.045 and 0.046, respectively. Though with genetic divergence at some extant, according to Vijayan ‘s study [33], they are still closely related species. So, C. sinensis is appropriate as reference for the RNA-seq analysis. Then, SAMtools [37] and VarScan [38] software were orderly used for SNP calling. SNP sites at which three replicates were consistent were marked. For allelic expression research, a tough problem deserving consideration is mapping bias. To relieve mapping bias, three pseudo-genomes, representing the female and the two male parents, were constructed by replacing the reference alleles in the C. sinensis genome with the corresponding alternative alleles at the SNP sites, respectively. Then, transcriptome reads from the parental libraries were realigned to their pseudo-genomes using the same parameters to obtain the final read counts at the SNP sites. When it came to the F1 hybrid series, reads from each library were mapped to the two pseudo-genomes of their parents, respectively. To relieve mapping bias, for each parental allele in the hybrids, we chose the maximum value of the two mapping results as the final reads count at each SNP site, and the sum of the two alleles as the total reads count at one site. To identify reliable SNP sites, quality controls were applied as follows: (i) the SNP sites in the two parents must be homozygous for difference (ii) each SNP site in the F1 hybrid must consist of only two alleles (one for the male parent, another for the female parent) (iii) the read count of the minor parental allele in the hybrid at each SNP site must be ≥2 and (iv) the total read count at each SNP site must be ≥20.

Gene expression quantitation

We wrote a R script to identify species-specific SNP sites from the mapping results. Finally, 37,078 SNPs, representing 7629 genes were identified from the cross of C. azalea × C. chekiangoleosa and 81,477 SNPs, representing 9566 genes, were identified from the cross of C. azalea × C. amplexicaulis. Transcript abundances of genes were evaluated as the normalized reads mapped per SNP site. Trimmed Mean of M-values (TMM) method [39] implemented in the edgeR package [40] was used for data normalization across libraries based on the assumption that most genes are not differentially expressed. Gene expression level was independently quantified for each cross, taking the biological replicates into consideration. The normalized gene expression for each cross is provided in the supporting information (Additional file 1: Tables S4 and S5). Cluster analysis was then carried out to examine the repeatability of the three biological replicates. As shown in Additional file 1: Figure S2, nearly all the biological replicates for each species and hybrid were clustered together (with R 2 > 0.90) except for amp1 and aza_che3, and these two samples were removed in the following analyses.

Classification of gene expression patterns

The edgeR package [40] was used for pairwise expression comparison, taking the three biological replicates into consideration. A fold change of 1.25 and the FDR < 0.05 were used as thresholds for differentially expressed gene (DEG) identification. DEGs between the hybrids and their parental species were further classed into eight clusters according to previous studies. Specifically, DEGs whose expression in the hybrid were higher than one of the parents but lower than another were classified as additivity (including additivity male > female and male < female) DEGs which were up/down-regulated in the hybrid compared with one of the parents but not differentially expressed with another were classified as male/female expression level dominance-up/down DEGs whose expression level in the hybrid were significantly higher/lower than both of the parents were classified as transgressivity (overdominance/underdominance).

Allelic expression patterns and cis- and trans-regulatory divergence assignment

Based on the species-specific SNP information, relative expression of the parental alleles in hybrids was evaluated. For each allele, the mean value of the three biological replicates was used for allelic expression as well as the subsequent regulatory divergence analysis. Expression divergence between the parental species is mainly caused by the combination of cis- and trans-regulatory changes, which could be quantified as log2 (parent1/parent2). In F1 hybrid, two parental alleles are exposed to a common trans-regulatory environment, and are equally affected by the trans-regulatory change. So, the log2-transformed radio of allelic expression in hybrid was used to quantify the degree of cis-effect: cis = log2 (F1Aparent1/F1Aparent2). Binomial exact test with FDR correction (FDR: 5%) was used to determine the significant cis-effect with a null hypothesis F1Aparent1 = F1Aparent2. Then trans-regulatory divergence was calculated as the difference between log2-transformed ratios of species-specific reads in the parents and the hybrids: trans = log2(parent1/parent2) - log2(F1Aparent1/F1Aparent2). Fisher’s exact test with FDR corrections (FDR: 5%) was used to identify the statistically significant trans-effects with a null hypothesis parent1/parent2 = F1Aparent1/ F1Aparent2. The relative proportion of total regulatory divergence attributable to cis-regulatory divergence (% cis) was calculated as (% cis) = [|cis|/(|cis| + |trans|)] × 100%, similarly, % trans. In addition, binomial exact test (FDR: 5%) was used to detect the significantly different expression between the two parental species. Regulatory divergence for different genes was then identified based on the results of binomial and Fisher’s tests as well as the direction of changes. According to previous studies [24], seven regulatory types were further identified (Additional file 1: Table S2). (i) cis only: the parental alleles were unequally expressed in the same ratio in F1 hybrid and between the two parents. (ii) trans only: the parental alleles were equally expressed in F1 hybrid but unequally expressed between the two parents. (iii) cis + trans: the parental alleles were unequally expressed both in F1 hybrid and between the two parents, but have the same direction (species with higher expression contributed the higher expressing allele in the F1 hybrid). (iv) cis × trans: the parental alleles are unequally expressed both in F1 hybrid and between the two parents, but have the opposite direction (species with higher expression contribute the lower expressing allele in the F1 hybrid). (v) Compensatory: the two parental alleles are equally expressed between the two parents but unequally in the F1 hybrid. (vi) conserved: the parental alleles are equally expressed both between the two parents and within the F1 hybrid. (vii) ambiguous: other situations not included in the above six categories.

Statistical test

Pearson correlation analysis was carried out to detect the relationship of gene expression between the hybrids and their parents. Wilcoxon’s rank-sum test was performed to compare the median parental expression divergence attributable to cis and trans-regulation. Kendall’s test was used to detect the relative contribution of cis- and trans-regulation to the divergent gene expression between different species. All the test statistics were calculated in R programe (v 3.3.2, CRAN). The main scripts used in this study are available in the supporting information.


Allopolyploidy, which involves genome doubling of an interspecific hybrid is an important mechanism of abrupt speciation in flowering plants 1, 2, 3, 4, 5, 6. Recent studies show that allopolyploid formation is accompanied by extensive changes to patterns of parental gene expression (“transcriptome shock”) 7, 8, 9, 10, 11, 12, 13, 14, 15 and that this is likely the consequence of interspecific hybridization rather than polyploidization [16]. To investigate the relative impacts of hybridization and polyploidization on transcription, we compared floral gene expression in allohexaploid Senecio cambrensis with that in its parent species, S. vulgaris (tetraploid) and S. squalidus (diploid), and their triploid F1 hybrid, S. x baxteri[17]. Major changes to parental gene expression were associated principally with S. x baxteri, suggesting that the polyploidization event responsible for the formation of S. cambrensis had a widespread calming effect on altered gene expression arising from hybridization [17]. To test this hypothesis, we analyzed floral gene expression in resynthesized lines of S. cambrensis and show that, for many genes, the “transcriptome shock” observed in S. x baxteri is calmed (“ameliorated”) after genome doubling in the first generation of synthetic S. cambrensis and this altered expression pattern is maintained in subsequent generations. These findings indicate that hybridization and polyploidization have immediate yet distinct effects on large-scale patterns of gene expression.

Author information


State Key Laboratory of Crop Genetics and Germplasm Enhancement, Cotton Hybrid R & D Engineering Center (the Ministry of Education), College of Agriculture, Nanjing Agricultural University, Nanjing, 210095, Jiangsu, China

Ting Zhao, Shouli Feng, Luyao Wang, Guandong Shang, Shisong Guo, Yuxin He, Baoliang Zhou & Xueying Guan

College of Agriculture and Biotechnology, Zhejiang University, Zhejiang, 210058, Hangzhou, China

Ting Zhao, Xiaoyuan Tao, Wei Ma & Xueying Guan

National Key Laboratory of Plant Molecular Genetics, National Plant Gene Research Center, Institute of Plant Physiology and Ecology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, 200032, China

Atz, J. W. (1962). Effects of hybridization on pigmentation in fishes of the genus Xiphophorus. Zoologica 47153.

Borowsky, R. L. (1978). The tailspot polymorphism of Xiphophorus (Pisces: Poeciliidae). Evolution 32886.

Borowsky, R. L., Herrera, R. J., Keawjam, R., and Hoffmann, R. N. (1979). Protein variability in fishes of the genus Xiphophorus. Manuscript.

Castro-Sierra, E., and Ohno, S. (1968). Allelic inhibition at the autosomally inherited gene locus for liver alcohol dehydrogenase in chicken-quail hybrids. Biochem. Genet. 1323.

Davis, B. J. (1964). Disc electrophoresis. II. Method and application to human serum proteins. Ann. N.Y. Acad. Sci. 121404.

Hitzeroth, H., Klose, J., Ohno, S., and Wolf, U. (1968). Asynchronous activation of parental alleles at the tissue-specific gene loci observed on hybrid trout during early development. Biochem. Genet. 1287.

Kallman, K. D. (1975). The platyfish, Xiphophorus maculatus. In King, R. C. (ed.), Handbook of Genetics, Vol. 4, Plenum, New York.

Scholl, A., and Anders F. (1973). Tissue-specific preferential expression of the Xiphophorus xiphidium allele for 6-phosphogluconate dehydrogenase in interspecific hybrids of platy-fish (Poeciliidae, Teleostei). In Schroder, J. H. (ed.), Genetics and Mutagenesis of fish Springer-Verlag, New York, pp. 301–313.

Whiteley, A. H., Hille, M., Lee, Y. R., Easton, D. P., Chamberlain, J. P., and Whiteley, H. R. (1975). Regulation of gene expression in inter species echinoid hybrids. In McMahon, D., and Fox, C. F. (eds.), ICN-UCLA Symposia on Molecular and Cellular Biology Developmental Biology, Vol. 2, W. A. Benjamin, Menlo Park, Calif.

Wright, D. A., and Moyer, F. H. (1966). Parental influences on lactate dehydrogenase in the early development of hybrid frogs in the genus Rana. J. Exp. Zool. 163215.

Mechanisms of Salinity Tolerance

Rana Munns and Mark Tester
Vol. 59, 2008


The physiological and molecular mechanisms of tolerance to osmotic and ionic components of salinity stress are reviewed at the cellular, organ, and whole-plant level. Plant growth responds to salinity in two phases: a rapid, osmotic phase that inhibits . Read More

Figure 1: Diversity in the salt tolerance of various species, shown as increases in shoot dry matter after growth in solution or sand culture containing NaCl for at least 3 weeks, relative to plant gr.

Figure 2: The growth response to salinity stress occurs in two phases: a rapid response to the increase in external osmotic pressure (the osmotic phase), and a slower response due to the accumulation .

Figure 3: The thermodynamics and mechanisms of Na+ and Cl− transport at the soil-root and stelar cell–xylem vessel interfaces in roots. Indicative cytosolic pH, ion concentrations, and voltages are de.

Figure 4: Differences in vacuolar concentrations of Na+ across roots of transpiring wheat plants growing in 150 mM NaCl. Concentrations were measured by quantitative and calibrated X-ray microanalysis.

Figure 5: Hypothetical relationships between salinity tolerance and leaf Na+ concentration for three different species, denoted by a, b, and c for rice, durum wheat, and barley. Within most species, t.

Figure 6: Relationships measured between salinity tolerance (biomass in salt as a % of biomass in control conditions) and leaf Na+ concentration in different wheat species. (a) Negative relationship f.

Watch the video: Genetic code (October 2022).