The impact of transposable elements in environmental adaptation



Transposable elements (TEs) play an important role in the responsive capacity of their hosts in the face of environmental challenges. The variety of mechanisms by which TEs influence the capacity of adaptation of the host is as large as the variety of TEs and host genomes. For example, TEs might directly affect the function of individual genes, provide a mechanism for rapidly acquiring new genetic material and disseminate regulatory elements that can lead to the creation of stress-inducible regulatory networks. In this review, we summarize recent examples that are part of an increasing body of evidence suggesting a significant role of TEs in the host response to an ever-changing environment, both in prokaryote and in eukaryote organisms. We argue that in the near future, the increasing availability of genome sequences and the development of new tools to discover and analyse TE insertions will further show the relevant role of TEs in environmental adaptation.


Organisms are continuously challenged by their changing environments. Variation in climatic factors such as temperature and humidity, interactions with other organisms, resource availability, and presence of toxins or other chemicals, among other biotic and abiotic factors, are likely to produce new selective pressures on populations that can challenge their survival. Organisms can respond to these changing environmental conditions by shifting their geographical distribution, through phenotypic plasticity or undergoing adaptive evolution to the new local conditions (Chevin et al. 2010; Hoffmann & Sgrò 2011). Of these three mechanisms, adaptative evolution is argued to play the most important role in determining the fate of species challenged by changing environmental conditions (Visser 2008).

Adaptive evolution occurs by natural selection when individuals better able to survive and reproduce pass on more genes to the next generation. As a consequence, the genetic variants that confer a fitness advantage increase in frequency in the population. Mutation is the ultimate source of genetic variation and different types of mutations, such as point mutations or whole genome duplications, play a major role in adaptation. Transposable elements (TEs; see Box 1) are also likely to play a relevant role in adaptation because of their ability to generate mutations of great variety and magnitude, and their capacity to be responsive and susceptible to environmental changes (Biemont & Vieira 2006; Schmidt & Anderson 2006; Oliver & Greene 2009; Hua Van et al. 2011).

Box 1. Types of eukaryotic transposable elements

Transposable Elements (TEs) are move around in the genome by generating new copies of themselves. TEs are abundant, ancient, and active components of genomes. They are classified in class I and class II elements according to the presence or absence of an RNA transposition intermediate. Within each class, TEs are further subdivide in orders, based on their insertion mechanism, structure, and encoded proteins; in superfamilies, based on their replication strategy; and in families, based on sequence conservation (Wicker et al. 2007; Kapitonov & Jurka 2008).image_n/mec12170-gra-0001.png

Class I elements (retrotransposons) replicate using a RNA intermediate and a reverse transcriptase. Each complete replication cycle produces new TE copies. As a consequence, retrotransposons are often the major contributors to the repetitive fraction in large genomes. Types of Class I elements include long terminal repeat (LTR) elements and non-LTR elements, such as Long Interspersed Nuclear Elements (LINEs) and Short Interspersed Nuclear Elements (SINEs). LTR elements have partly overlapping open reading frames (ORFs), GAG and POL closely related to retroviral proteins, flanked in both ends by LTRs with promoter capability. LINE elements consist of a 5′ Untranslated Region (UTR) with promoter activity, two ORFs and a 3′ UTR with a poly-A tail, a tandem repeat or merely an A-rich region. SINEs are nonautonomous elements, they rely on LINEs for transposition, that originate from accidental retrotransposition of various polymerase III (Pol III) transcripts. Unlike retro-pseudogenes, SINEs possess an internal Pol III promoter allowing them to be expressed. Alus, the most common SINE in the human genome, consist of two CG-rich fragments, the left and right Alu, connected by an A-rich linker and ended in a poly-A tail.

Class II elements do not require a reverse-transcription step to integrate into the genome. DNA transposons encode a transposase that recognizes the terminal inverted repeats (TIRs) excises the TE out and then integrates the TE into a new site in the genome. The gap that is left at the position where the TE was originally inserted can be filled with a copy of the transposon by gap repair mechanisms. Alternatively, DNA transposons can increase in number by transposing during chromosome replication from a position that has already been replicated to another that has not been replicated yet. Miniature Inverted Repeats (MITEs) have no ORFs and also have TIRs. Two newly identified DNA transposons, Helitrons and Mavericks duplicate differently. Helitrons used a rolling-circle mechanism and do not have TIRs, while Mavericks, also known as polytons, probably replicate using a self-encoded DNA polymerase and have TIRs. Helitrons often carry gene fragments that have been captured from the host genome.

TEs can also be classified according to their self-sufficiency. TEs that are capable of producing the proteins necessary for their transposition are classified as autonomous elements, while TEs that depend on other TEs to transpose, such as SINEs and MITEs, are classified as nonautonomous elements. Nonautonomous elements are often deletion derivates of autonomous elements although sometimes they have only limited sequence similarity to their autonomous counterparts.

Transposable element-induced mutations range from subtle regulatory mutations to gross genomic rearrangements often having complex phenotypic effects. Box 2 includes a detailed description of the different types of mutations generated by TEs, actively by de novo insertion and retrotranposition, and passively by acting as substrates for ectopic recombination. As well as being vertically transferred, from parent to offspring, TEs can also be horizontally transferred, from one species to another, potentially causing the multitude of effects summarized in Box 2 in the new host species. Additionally, TEs can also act as vectors facilitating the horizontal transfer of new genetic content (Ochman et al. 2000; Frost et al. 2005). Whether they do or do not transfer genes, horizontal transfer of TEs is a source of raw genomic variation, and at times of biological innovation, that influences the ability of the organism to adapt to changes in its environment, and to colonize new ecological niches (Schaack et al. 2010).

Box 2. TEs generate a great variety of mutations

TEs can have a myriad of effects when they insert into new locations (Feschotte 2008; Goodier & Kazazian 2008; Gogvadze & Buzdin 2009). These effects vary depending on where exactly the TE inserts and on the sequence of the TE itself. When a TE inserts into the 5′ region of a gene, it can add new regulatory regions leading for example to gene overexpression (a) or can disrupt existing regulatory regions and inactivate the gene in a particular tissue or developmental stage (b). When a TE jumps into an exon it can disrupt the gene for example by altering the reading frame, or by introducing a stop codon (c). A TE that inserts in the 3′UTR of a gene can disrupt the regulatory sequences in that UTR and/or it can add new ones, for example it can add miRNA-binding sites (d). A TE can disrupt the 5′UTR of a gene leading to, for example, gene inactivation (e). When a TE inserts into an intron it can: (f) be incorporated as a new exon, (g) introduce a STOP codon leading to a truncated transcript, (h) introduce new splice sites creating new alternative spliced variants, (i) drive antisense transcription that could interfere with the sense transcript of the same gene, (j) spread epigenetic silencing leading to gene inactivation.image_n/mec12170-gra-0002.png

TEs are also involved in the duplication of genes and exons that may contribute to the generation of new genes (Marques et al. 2005; Xing et al. 2006). TE-encoded genes can be exapted to perform cellular functions (Volff 2006). Finally, TEs are also passive generators of mutations. TEs that belong to the same family of elements and are located in different regions of the genome can act as substrates for ectopic recombination events generating rearrangements such as inversions, translocations or duplications (Schwartz et al. 1998; Hill et al. 2000; Bailey et al. 2003).

TEs are also responsive and susceptible to environmental changes. Stress-activated TEs might generate the raw diversity that species require over evolutionary time to survive stressful situations. The first person to present this idea was Barbara McClintock through her extensive work in the maize transposons Ac and Ds (McClintock 1984). This idea seemed overly optimistic for other researchers that thought that activation of TEs is due to the disruption of the host mechanisms that suppress transposition in normal conditions. One of the clearest cases of TE activation due to the breaking down of repression mechanisms is hybrid dysgenesis in Drosophila. Hybrid dysgenesis is a sterility syndrome caused by very high rates of transposition of normally inactive TE families (Bingham et al. 1982; Bucheton et al. 1984; Petrov et al. 1995). Activation of TEs could be the consequence of the relaxation of epigenetic control induced by environmental changes (Slotkin & Martienssen 2007; Zeh et al. 2009; Rebollo et al. 2010). However, the many examples providing solid grounds for the activation of specific TEs in response to some specific stress conditions indicates that the link between TE activation and stress response is by far more complex than the simple release of regulation (Wessler 1996; Grandbastien et al. 1997; Capy et al. 2000; Schmidt & Anderson 2006; Fablet & Vieira 2011).

In this review, we investigate the evidence for the role of TEs in environmental adaptation. Because the literature on this topic is extensive, we do not attempt to review every known case of environment-related TE-induced adaptation, but rather focus on the most recent examples from diverse organisms that illustrate the variety of molecular mechanisms and phenotypic effects of TE-induced mutations. We start with site-specific insertions of TEs that result in adaptation to the environment. We then focus on the most recent evidence for environmental adaptation mediated by horizontal transfer of TEs. Finally, we review cases in which TEs are activated by, or in response to, environmental stresses.

TE-induced mutations involved in environmental adaptation

TE-induced mutations have been frequently associated with adaptation to the environment. Below, we briefly describe some of the most compelling examples of individual TE-induced environmental adaptations documented recently. These examples highlight the variety of molecular mechanisms and adaptive phenotypic effects of TEs, from bacteria to mammals.

Bacteria insertion sequences (IS) have long been associated with environmental adaptation. In early studies, it was unclear whether the IS element was the causal mutation responsible for the adaptive phenotypic change (e.g. Naas & Nordmann 1994; Schneider et al. 2000; de Visser et al. 2004). In recent years, however, a cause–effect relationship has been established between IS elements and adaptation to several environmental challenges such as adaptation to high osmolarity (Stoebel et al. 2009; Stoebel & Dorman 2010), tolerance to toxic organic solvents (Sun et al. 2009), metal-limited conditions (Chou et al. 2009) and nutrient-limited conditions (Gaffé et al. 2011). The molecular mechanisms underlying these IS-induced adaptive mutations are diverse, some insertions affect gene expression (up-regulation, down-regulation, and inactivation of nearby genes) while other insertions generate rearrangements leading to deletions. Although the same adaptive phenotypes may arise in strains lacking IS elements (Stoebel & Dorman 2010), the studies mentioned previously show that IS elements play an important role in environmental adaptation.

In plants, adaptation to local environments has been repeatedly associated with TE-induced mutations. For example, in soybean, the disruption by a TE insertion of GmphyA2, one of the two paralogs encoding phytocrom A, is associated with adaptation to high latitudes as showed by phenotypic experiments and allelic distribution analyses (Liu et al. 2008; Kanazawa et al. 2009). In Arabidopsis, light-regulation of gene expression is associated with FAR1 and FHY3 that have been co-opted from an ancient Mutator-like transposase (Lin et al. 2007). Lin et al. (2007) experimentally showed that these proteins increase gene expression by directly binding to the promoter regions of target genes. The authors argue that the domestication of FAR1 and FHY3 might have contributed to Arabidopsis adaptation to changing light environments. In wheat, several TE-induced mutations in vernalization genes are responsible for changes in the growth habit that enables wheat to adapt to a wide range of environments (Yan et al. 2006; Chu et al. 2011).

Adaptation to local environments is also linked to TE-induced mutations in Drosophila (González et al. 2008, 2010; González & Petrov 2009a). We carried out the first genome-wide screen for recent adaptive TE insertions in Drosophila melanogaster and we discovered several TE insertions involved in local adaptation (González et al. 2008, 2009b). In a follow-up study, we showed that a substantial proportion of the identified TE insertions are specifically adaptive to temperate environments, and that the frequency of some of these insertions correlates with environmental variables such as temperature and rainfall (González et al. 2010). We estimated that the already identified mutations only represent a subset of the total number of TE-induced adaptive mutations suggesting a widespread role of TEs in environmental adaptation in Drosophila.

Besides adaptation to local environments, TE insertions in Drosophila have also been involved in resistance to viral infection and resistance to insecticides. Resistance to viral infection has been associated with a TE insertion in the protein coding sequence of CHKov1 (Magwire et al. 2011). The TE insertion truncates CHKov1 creating four different altered transcripts, none of which contain all four exons of the wild-type gene. This insertion was previously shown to confer resistance to insecticides, although the authors already noted that the allele containing the insertion had been evolving in the populations for a long time before insecticides started to be used (Aminetzach et al. 2005). It turns out that the allele carrying the insertion would initially have played a role in defending flies against viral infection. However, flies carrying this particular TE insertion found themselves pre-adapted to the introduction of insecticides in the middle of last century. Magwire et al. (2011) also provide evidence that CHKov1 alleles carrying duplications of the gene region containing the insertion, resulted in further resistance to viral infection. Similar to the CHKov1 allelic series, the region containing a Cyp6g1 allele previously shown to confer resistance to pesticides (Daborn et al. 2002), has also suffered duplications and additional TE insertions that increased resistance to pesticides (Schmidt et al. 2010). These two examples support the view that alleles of large effect may sometimes reflect the accumulation of multiple mutations of small effect at key genes. Other than in Drosophila, a clear role for TE insertions in insecticide resistance has also been demonstrated in mosquitos (Darboux et al. 2007). The binary toxin produced by Bacillus sphaericus is used as an insecticide against the mosquito Culex pipiens. Resistance to this toxin is due to the insertion of a TE into the coding sequence of the toxin receptor. The insertion induces a new mRNA splicing event that creates a shorter transcript. This new transcript encodes an altered receptor unable to interact with the toxin resulting in resistance to this insecticide (Darboux et al. 2007).

Our last example connecting individual TE-induced mutations and environmental adaptation comes from paleogenomic studies in mammals (Santangelo et al. 2007; Franchini et al. 2011). Pomc, a gene involved in stress response and regulation of food intake and energy balance, has two functionally overlapping enhancers that originated from ancient unrelated TE insertions. In multicellular organisms, the presence of two enhancers capable of guiding similar patterns in spatiotemporal expression is common to several developmental genes. Rather than being redundant, the presence of the two enhancers is required to overcome the challenges imposed by critical environmental conditions such as changes in temperature (Frankel et al. 2010; Perry et al. 2010). Possibly, the presence of these two enhancers has been key to evolution of mammals through the periods of abrupt climate change. Given the abundance of TEs in mammalian genomes, the authors concluded that it is conceivable that sequential exaptation of TEs leading to analogous cell-specific enhancers could be a more generalized phenomenon than previously anticipated (Franchini et al. 2011).

Horizontal Transfer of TEs (HTT) and horizontal gene transfer (HGT) mediated by TEs

Besides being transferred from parent to offspring, TEs can also be horizontally transferred between species. A horizontally transferred TE (HTT) can generate in the new host species the same battery of mutations described for vertically transferred TEs (see Box 2). Additionally, TEs can also act as vectors facilitating the horizontal transfer of new genetic content (Ochman et al. 2000; Frost et al. 2005). This phenomenon has been extensively demonstrated in prokaryotes. In eukaryotes, although TEs are capable of capturing and transferring genes at a high frequency within a species (Jiang et al. 2004; Morgante et al. 2005; Schaack et al. 2010) they have not yet been found to transfer host genes between different species. Although horizontal gene transfer (HGT) can also occur independent of TE movement, in this review, we focus on TE-mediated HGT events.

HTT and HGT in prokaryote environmental adaptation

There is no doubt that prokaryotes increase their genetic variation by HGT (Ochman et al. 2000; Aminov 2011). This mechanism rapidly integrates ‘foreign’ DNA that gives the new host the opportunity to acquire new functions, and to colonize extremely diverse habitats (Wiedenbeck & Cohan 2011). This phenomenon is of such importance in bacteria that the vast majority of species-specific DNA sequences that differ between two given species have been the result of different events of horizontal transfer (Levin & Bergstrom 2000). The mechanisms by which TE- induced HGT can take place in prokaryotes are diverse and depend on which TE is involved. HGT events often involve operons and gene cassettes because horizontally transferred genes have a better chance to be functional in the new host genome if they are transferred with their flanking sequences. Box 3 briefly describes the main TE sequences often involved in HGT between prokaryote organisms. Additionally, a recent review is available to the readers interested in the mechanistic details of HGT in prokaryotes (Toussaint & Chandler 2012).

Box 3. Horizontal transfer in prokaryotes

The genetic content of an organism is received by vertical inheritance, leaving most organisms with a finite toolbox to face all eventualities along their life and limiting their possibilities to explore new ecological niches. Nevertheless, in some occasions, evolution provides an alternative mechanism for rapidly acquiring new genetic material: horizontal gene transfer. Below, we briefly described several types of prokaryotic transposons that have facilitated the horizontal transfer of genes.image_n/mec12170-gra-0003.pngimage_n/mec12170-gra-0004.png

Composite transposons. In a composite transposon, two Insertion Sequences (ISs) flank one or more genes such as Tn10 composed of two IS10 elements flanking the tetracycline resistance gene, or Tn5, two IS50 elements flanking a three-resistance gene operon: streptomycin, bleomycin and kanamycin (Ochman et al. 2000). Composite transposons can be mobilized between distantly related bacteria having a great impact on the adaptive capacity of the genome that hosts them. ISs are also involved in creating modular assemblies of genes, the simplest being concatenation within compound transposons. A good example is the 221 kb virulence megaplasmid of Shigella flexneri, pW100; (Buchrieser et al. 2000; Venkatesan et al. 2001). In this megaplasmid, ISs represents 46% of the DNA content including 26 full-length ISs and an extensive array of IS fragments indicative of ancestral rearrangements.

Insertion Sequence Common Regions (ISCR). ISCR are often associated with resistance and virulence genes. ISCR resemble ISs but lack terminal inverted repeats and are thought to transpose by a rolling-circle mechanism. ISCR impact on shuffling antibiotic resistance genes among bacteria is remarkable: they have been involved in horizontal transfer events of resistance genes of every single class (Toleman et al. 2006).

Conjugative transposons. Conjugative transposons encode their own ability to move from one bacterial cell to another via cell-to-cell contact. Conjugative transposons have a surprisingly broad host range, and they probably contribute as much as plasmids to the spread of antibiotic resistance genes in some genera of disease-causing bacteria. Many conjugative transposons can mobilize co-resident plasmids, and some of them can even excise and mobilize unlinked integrated elements.

Mobile Integrons (‘quantum leap’ evolution). Integrons are genetic elements able to acquire and rearrange open reading frames (ORFs) embedded in gene cassette units and convert them to functional genes by ensuring their correct expression. An Integron by itself is nonmobile and its basic functional units are the intI gene and the attI recombination site. intI encodes a site-specific tyrosine recombinase that recognizes the attI site (Collis et al. 1993; Collis & Hall 1995). intI is responsible for the integration and excision of the different genetic cassettes that compose the Integron. A promoter often embedded inside the intI gene or the attI sequence drives the expression of the Integron.

When Integrons are associated with transposons they can be mobilized in conjugative plasmids and can be transferred to individuals of the same or different species. Through their life in different genomes, integrons can acquire gene cassettes from different origin and be successful in different species thanks to the flexibility of the codon usage of the harboured genes. Intriguingly, most gene cassettes associated with mobile integrons are composed by antibiotic resistance genes (Naas et al. 2001), although a few genes of unknown function have also been identified (Cambray et al. 2010). Integrons have the capacity to harbour many gene cassettes as in the famous case of the Vibrio cholerae super-integron with 179 gene cassettes (Mazel 2006). The impact of the integration of a mobile unit with such high number of genes could be considered as a ‘quantum leap’ for the evolution of the new host.image_n/mec12170-gra-0005.png.

Several instances of TE-induced HGT are related to adaptation to different environmental conditions (Ochman et al. 2000; Hacker & Carniel 2001; Toleman et al. 2006; Cambray et al. 2010; Aminov 2011). In this section, we will focus on recent examples of HTT and HGT that play a role in (i) the acquisition of new catabolic and metabolic properties, (ii) detoxification, and (iii) pathogenicity and virulence.

  1. New catabolic and/or metabolic properties: We have chosen a recently described example that illustrates how the acquisition of new catabolic capacities has allowed a host bacterium to better adapt to harsh environmental conditions. Cupriavidus metallidurans is a β-proteobacterium adapted to live in environments that contain heavy metal pollution (Mijnendonckx et al. 2011). A recent genome-wide analysis revealed that there are 57 IS elements in this species, three of which show 100% identity with IS elements of a number of other bacteria: Ralstonia pickettii, Burkholderia vietnamiensis, Delftia acidovorans and Comamonas testosteroni. All these bacteria live in similar environments suggesting recent interactions and HTT events between these strains. These horizontally transferred TEs have been associated with genomic islands and with gene inactivation that affect the autotrophic growth capacity of C. metallidurans. Furthermore, one of the horizontally transferred TEs is also associated with stress response (Mijnendonckx et al. 2011). The previous example demonstrates the crucial role that TEs can have both directly and indirectly in the adaptive capacity of bacteria to harsh and polluted environments.
  2. Detoxification: The catabolic capacities of bacteria are not only directly linked to their own chances to survive in a changing environment, but also often contribute to the survival of other organisms that share the same, often contaminated, environment. Wei and collaborators showed that the gene methyl parathion hydrolase, mph, involved in the degradation of organophosphorus compounds, was part of a typical composite transposon (Tnmph; see Box 3) flanked by two IS6100 sequences in Pseudomonas sp (Wei et al. 2009). The Tnmph composite transposon was successfully transferred in the laboratory to a wide range of bacterial species, including some phylogenetically distant ones. These results suggest that Tnmph may contribute to the wide distribution of mph-like genes and the adaptation of bacteria to organophosphorus compounds (Wei et al. 2009). The possibility of manipulating bacteria that live in our everyday environments with composite transposons including genes like the mph, expands the already available possibilities to counteract some of the effects of contamination using microorganisms (Wu et al. 2012).
  3. Pathogenicity: Clostridium perfringens is a pathogenic bacterium that causes serious illness in different livestock animals. The different isolates of C. perfringens are classified based on which of four lethal toxins they produce (Sayeed et al. 2010). Type B isolates are the most virulent because they are able to produce two different toxins: beta-toxin and epsilon-toxin. Molecular characterization of type B isolates, demonstrated that these isolates contain not just one, but three different plasmids with virulence genes. The identification of IS elements (IS1151) as well as genes involved in conjugative transposition (tcp; see Box 3), strongly suggested that both circular and conjugative transposition may have been involved in the HGT of these large virulence platforms (Sayeed et al. 2010). This is another example of a case in which a series of HGT caused by HTT has resulted in a better adaptation. In this case, pathogenicity may increase the chances of wider spread and therefore increase the likelihood of survival of the host bacterium.

HTT and HGT in eukaryote environmental adaptation

Horizontal Transfer of TE events have been reported in eukaryotic species as diverse as Drosophila, yeast and fungi (Hall et al. 2005; Loreto et al. 2008; Gilbert et al. 2010; Schaack et al. 2010). These events may have evolutionary relevance only if the newly inserted TE is able to transpose, increase in copy number, or provide a new cellular function. The capacity to transpose and increase in copy number in a new invaded genome has been reported for Helitrons (Box 1) in several organisms including mammals, reptiles, fish, invertebrates and insect viruses (Thomas et al. 2010). There is also evidence for the generation of new cellular functions after an HTT event for P-elements in Drosophila (Pinsker et al. 2001) and SPIN elements in mouse (Pace et al. 2008). Therefore, HTT could be an important evolutionary force shaping eukaryotic genomes, although evidence for a specific role in environmental adaptation has yet to be found.

As in prokaryotes, HGT has had an important role in eukaryote genome evolution (Keeling & Palmer 2008; Syvanen 2012). The evidence for HGT in diverse eukaryotes is expanding rapidly in organisms such as nematodes (Haegeman et al. 2011) and fungi (Fitzpatrick 2012). Many of the reported HGT events are related to environmental adaptation. For example, the ability of distantly related unicellular eukaryotes to live in anaerobic environments (Loftus et al. 2005) or the transfer of antifreeze proteins in fish (Graham et al. 2008) are due to HGT events. Although there is no evidence yet of HGT mediated by TEs (Schaack et al. 2010), some authors predict that it will be soon discovered (Keeling & Palmer 2008). For example, Helitrons have a rolling-circle mechanism of transposition that makes them especially prone to take adjacent 3′ unrelated DNA along (Feschotte & Wessler 2001) and therefore are strong candidates to play a major role in HGT between eukaryotic species.

TE activation triggered by or in response to environmental stress

As we mentioned in the introduction, Barbara McClintock was the first to propose that the activation of TEs in response to stress induces mutations that could help the organism adapt to new environmental conditions (McClintock 1984). TEs would therefore play a key role in translating changes in the external environment into changes at the genomic level. Indeed, TEs respond directly to some specific stress situations and in some cases the specific TE sequences responsible for the stress response have been identified. This is the case of several Long Terminal Repeat (LTR) retrotransposons that contain cis-regulatory elements in their 5′ LTR that trigger transposon expression in response to a particular stimulus (Kumar and Bennetzen 1999). These regulatory sequences are similar to the well-characterized motifs required for the activation of stress-responsive genes (Grandbastien et al. 2005). The possibility of acquiring changes in the cis-regulatory elements entails the opportunity to respond to new and different environmental factors. Examples of TEs containing these cis-regulatory elements are abundant and are very well represented in the literature. In Box 4 we describe some of the classical examples, such as Tnt1 and Bare1 in tobacco and barley, respectively.

Box 4. The U3 Box of LTR Retrotransposons

The 5′ LTR works as a promoter containing the sequences that drive, specify, and signal for termination of transcription, and the capping signal. The LTR is subdivided in the U3, R and U5 domains. Different specific DNA elements in the U3 region (B boxes) have been identified in relation to specific molecules that signal for different stress responses, such as phytohormones and elicitors. See some examples in the table below and text and references for further details.image_n/mec12170-gra-0006.png.

Although the specific sequence that responds to stress has not been identified, for other LTR retroelements it has been shown experimentally that the LTR is sufficient in itself to activate TE transcription in response to stress. This, for example, is the case of the activation under nitrate starvation stress of the Blackbeard retrotransposon in the marine diatom Phaeodactylum tricornutum (Maumus et al. 2009). Because LTR elements are very abundant in this diatom genome, the authors suggest that their massive activation may probably contribute to major genome rearrangements that would allow this organism to respond rapidly to changing environmental conditions (Maumus et al. 2009). Furthermore, the authors show that the retroelement is hypomethylated in response to nitrate starvation, which provides a link between environmental stress and chromatin remodelling in diatoms.

Besides being present in the 5′LTR, transcriptional regulatory sequences are also located in the open reading frames of some LTR retrotransposons (Servant et al. 2008, 2012). This is the case for the LTR retrotransposon Ty1 of Saccharomyces cerevisiae. The transcription of the Ty1 retrotransposon is induced, among other specific stress conditions, by a shortage of adenylic nucleotides (Todeschini et al. 2005). A recent study by Servant and collaborators identifies the mechanism of activation of this TE (Servant et al. 2012). It turns out that severe adenine starvation activates the expression of the transcription factor TYE7. TYE7 binds to the E-boxes, located downstream of the transcription start site of the TYA gene, and alters Ty1 antisense transcription. As a consequence, there is an increase in sense Ty1 mRNA that leads to retrotransposition of this element and coactivation of the expression of genes adjacent to Ty1 insertions (Servant et al. 2008, 2012).

Other than LTR retrotransposons, class II elements such as Miniature Inverted-repeat Transposable Elements (MITEs) have also been shown to respond specifically to some stress conditions. This is the case, for example, for the mPing MITE in rice. In some rice strains mPing has amplified from c. 50 to 1000 copies (Naito et al. 2006). The analyses of the insertion sites in the strains that have undergone this burst of transposition revealed that under normal growth conditions mPing elements have a modest impact on the host because of highly evolved targeting mechanisms that minimize the effects on host gene expression (Naito et al. 2009). However, mPing is able to confer a stress-inducible state to the nearby genes regardless of whether the TE is inserted at their 5′ or the 3′ region, suggesting its potential to act as an enhancer element. Although a specific sequence inside the mPing element has not been defined, it is clear that mPing is able to provide the surrounding genes the capacity to respond to certain stress situations but not others (e.g. cold and salt but not drought). Because of the high copy number of mPing in rice genomes, its specific transcription and transposition could result in new gene regulatory networks of coordinated expression that would contribute to a fine-tuned response of this organism to specific stress factors. The creation of such regulatory networks in response to certain stresses could be a widespread phenomenon in nature since evidence for rapid and massive amplification of MITEs has been found in virtually all sequenced eukaryotic genomes and even in some prokaryote ones (Naito et al. 2009).

The case of mPing illustrates how the integration site of some TEs may confer stress-inducibility to nearby genes. However, the opposite is also true: some TEs specifically integrate close to stress-responsive genes. Tf1, an LTR retrotransposon from Schizosaccharomyces pombe, shows a tendency to integrate in a 500 bp window upstream of ORFs (Behrens et al. 2000; Bowen et al. 2003). Guo & Levin (2010), further demonstrate that in different activation experiments the newly integrated Tf1 elements insert close to RNAPol II promoters but interestingly, there was no correlation with the level of transcription of the targeted promoters. Instead, Tf1 had a strong preference for promoters that are induced by specific stress conditions, such as genes induced by cadmium and heat. The targeting of Tf1 to stress-induced promoters represents a unique response that may function to specifically alter expression levels of stress response genes (Guo & Levin 2010).

Activation of TEs is not always directly triggered by a specific stress but the effects that such stress causes in other cellular mechanisms allow a rapid activation of some particular TE copies (Dai et al. 2007; Coros et al. 2009). An interesting example to illustrate this kind of secondary response is the activation of the Ty5 retrotransposon in Sacharomyces subject to starvation stress. Ty5 in Sacharomyces preferentially integrates into heterochromatic regions. This pattern of integration is directed by the interaction between the Ty5 integrase targeting domain (TD) and the heterochromatic protein Sir4 when Ty5 is phosphorylated (Zhu et al. 2003; Dai et al. 2007). When Sacharomyces is faced with starvation, numerous signal transduction pathways, among them the protein kinase A pathway, are affected. When the TD of Ty5 is not phosphorylated, there is no interaction with Sir4 and the pattern of integration of this retrotransposon changes radically. Under such conditions Ty5 becomes a potent endogenous mutagen that integrates randomly throughout the genome, including into gene-rich regions. This change in the pattern of integration of Ty5 is observed in response to some specific stress conditions (e.g. starvation stress) and not others (e.g. heat-shock, DNA damage, osmotic shock or oxidative stress). The regulation of Ty5 phosphorylation by stress, demonstrates that TEs provide the cell with a prewired mechanism to reorganize the genome in response to environmental challenge (Dai et al. 2007).

Finally, we will highlight two of the several recent examples from the literature indicating that noncoding and small interfering RNAs are also another possible path by which TEs respond to stress (Hilbricht et al. 2008; Mariner et al. 2008; Lv et al. 2010; Yan et al. 2011; McCue et al. 2012). Possibly one of the best-studied stress responses in eukaryotes is the one triggered by heat-shock. However, the exact mechanisms by which most organisms subject to a heat-shock manage to repress the transcription of most genes are still unknown. Mariner and collaborators discovered one mechanism of response to heat-shock involving TEs in humans (Mariner et al. 2008). Alu elements function as cell stress genes: different stress conditions cause an increase in the expression of Alu RNAs, which rapidly decreases upon recovery from stress (Häsler & Strub 2006). Alu RNA has been implicated in regulating several aspects of gene expression such as alternative splicing, RNA editing, translation and miRNA expression and function (Häsler & Strub 2006; Häsler et al. 2007). In humans, Alu elements but not other Pol III transcribed genes are activated by heat stress. Mariner et al. (2008) demonstrated that the mRNA of the Alu element block transcription by binding RNApol II and entering the repressor complexes that will be loaded onto the promoters of the repressed genes. Interestingly, in mouse cells the SINE B2 element activated upon heat-shock is also able to repress transcription of many genes using a similar mechanism. Although the B2 SINE derived from tRNA from mouse, and the human Alu derived from 7SL-like precursor, do not have sequence identity or similar RNA secondary structures, their similar effects on the host heat–shock response suggest that these two SINE elements have converged to the same biological function.

An additional example reveals how siRNAs generated by a retrotransposon confer the capacity to respond to desiccation to the callus of the plant Craterostigma plantagineum (Hilbricht et al. 2008). CDT-1 was first identified as a plant desiccation tolerant gene and later recognized as being a TE, although it is still pending classification. Hilbright and collaborators reported that while no translation from this element is needed for the desiccation tolerance response, the transcription and the posterior production of related siRNAs from CDT-1 is essential to induced expression of desiccation-inducible genes.

Overall, the examples described previously strongly suggest a role of TEs in the ability of the host to respond to changes in the environment. The evidence that only some specific TE families, and not all the TEs in the genome, are activated in response to stress and the evidence that these TEs respond to some specific stress conditions and not others, strongly suggest that activation of TEs by stress is not only a byproduct of genome deregulation. The consequences of TE activation in response to stress are diverse. Stress-activated TEs: (i) contribute to major genomic rearrangements (Maumus et al. 2009), (ii) confer nearby genes the capacity to respond to stress (Guo & Levin 2010; Servant et al. 2012), which may lead to the creation of new regulatory networks (Naito et al. 2009; Ito et al. 2011) and (iii) alter the genome randomly through insertion of the newly generated copies (Dai et al. 2007). Therefore, containing a certain number of potentially active TEs may increase the genome ability to cope with environmental changes.

Concluding remarks

Given the opportunistic nature of evolution, the capacity of TEs to generate mutations of great variety and magnitude suggests that TEs are important players in genome evolution. Some authors may consider that the capacity of TEs to create genetic diversity that might result beneficial for the host genome has not been exploited often, nor has it necessarily been subject to positive selection. In this review, we argue that there are many examples that provide solid grounds for the beneficial effect of TEs in host genome evolution in general and in host environmental adaptation in particular. Note that several of the works summarized in this review (e.g. González et al. 2008; Naito et al. 2009; Franchini et al. 2011) strongly suggest that the particular cases described may represent the tip of the iceberg. Moreover, identifying TE insertions involved in environmental adaptation depends ultimately on our ability to identify a given nucleotide sequence as a TE or a TE remnant. As such, we are still likely underestimating the role of TEs in environmental adaptation just because of our limitations to identify TE insertions. We anticipate that in the next years increased availability of genome sequences, the development of new tools to accelerate the discovery of TE insertions (Fiston-Lavier et al. 2011; Flutre et al. 2011; Makalowski et al. 2012) and the increased knowledge about which genes and traits are relevant for adaptation will further support the prevalent role of TEs in environmental adaptation.


We thank Anna-Sophie Fiston-Lavier, Lain Guio, Ruth Hershberg, Lidia Mateo, Dmitri A. Petrov and Alfredo Ruiz for critically reading the manuscript. This work was supported by a grant from the Spanish Ministry of Science and Innovation BFU2009-08318/BMC awarded to E.C. and by a Ramon y Cajal grant (RYC-2010-07306), a Marie Curie CIG grant (PCIG-GA-2011-293860) and a National Programme for Fundamental Research Projects grant (BFU-2011-24397) awarded to J. G.

E.C. is a functional evolution researcher interested in the role of TEs in the evolution of eukaryote genomes. J.G. leads the Evolutionary and Functional Genomics research group which focuses on elucidating the molecular process and the functional consequences of adaptation.