Viruses and the Placenta: the Essential Virus ﬁrst View

Review Article

Viruses and the placenta: the essential virus ﬁrst view

LUIS P. VILLARREAL

Center for Virus Research, Department of Molecular Biology and Biochemistry, University of California Irvine, Irvine, CA, USA

Villarreal LP. Viruses and the placenta: the essential virus first view. APMIS 2016; 124: 20–30. A virus first perspective is presented as an alternative hypothesis to explain the role of various endogenized retroviruses in the origin of the mammalian placenta. It is argued that virus–host persistence is a key determinant of host survival and the various ERVs involved have directly affected virus–host persistence. Key words: Virus evolution; host evolution; placenta; endogenous retroviruses. Luis P. Villarreal Center for Virus Research, Department of Molecular Biology and Biochemistry, University of California Irvine, Irvine, CA 92617 USA. e-mail: [email protected]

THE PROBLEM – WHY SHOULD ERVS include the presence of intercisternal A-type parti- CONTRIBUTE TO THE COMPLEX cles (IAPs) (3), presence in human oocytes (4), pres- PLACENTAL NETWORK? CONFRONTING ence in early preimplanted embryo (5), antivial THE ACCEPTED VIEWS activity of human sera (6), the presence of reverse transcriptase (RT) inhibitors (7, 8), and the pres- The emergence of mammalian vivipary and the pla- ence of ERV3 mRNA (9, 10). For a summary of centa presents many biological and behavioral these early observations see (11). In 1997, I pro- issues that challenge theories of evolution, see (1). posed some general reasons why virus should be These biological and immunological dilemmas are involved in the origin of vivipary (12). ERVs asso- associated with the emergence of the ‘foreign’ mam- ciated with mammalian reproductive biology are malian placenta (expressing paternal genes). In lineage specific and their acquisition is associated addition, the very first cell type to differentiate in with the origin of each lineage (13, 14). But it was mammalian embryo is the trophectoderm which the discovery of the involvement of ERV envelope will generate the placenta, thus major alterations to proteins (such as syncytins) in reproductive biology programs of early developmental are also needed. that has really engaged the interest of many evolu- The placenta will mediate the blood (and immune) tionary biologist in this virus–host relationship. exchange between mother and her non-self embryo Overall, they have adopted a now well accepted and contribute to very complex biological and perspective that retroviruses have repeatedly pro- behavioral changes needed for live birth. All these vided env genes which have proved useful (were changes require complex and network based regula- exapated/domesticated) for the various functional tory changes to the genetic programs that had and structural requirements of a placenta (15–18). mostly been present in ancestral egg laying mam- And once the early env-mediated placenta emerged, mals. This represents a major transition in the evo- fitter (better) versions of placentas via newer ERVs lution of complexity that has been difficult to followed. This is the currently accepted perspective explain by traditional concepts. Over the years, it on ERV involvement in the origin of the placenta has become increasingly evident that endogenized and it presents a ‘host come first’ perspective. retroviruses (ERVs) have been intimately and dee- Here, the fortuitous virus is simply providing a ply involved in the placenta of all mammalian lin- convenient and diverse source of useful env genes. eages (2). These historic retrovirus observations But, egg-laying animals (especially avians) are highly successful and diverse, so why viruses might Received 4 June 2015. Accepted 26 October 2015 mediate such a drastic change in reproductive host

20 VIRAL ORIGIN OF THE PLACENTA biology remains an open question. In this essay, I direction of increasing complexity. In this way, present a virus first perspective that offers an alterna- viruses present an omnipresent and ancient issue. tive hypothesis for virus involvement in the origin of Hence, we must always consider how any virus the placenta. action on host will affect virus–host survival in their respective virosphere or virus habitat (e.g., reproductive tissue). A most significant develop- A VIRUS FIRST PERSPECTIVE JUSTIFIED: ment would be the emergence of a stable persistent VIRUS PERSISTENCE AND DISTINCT relationship between virus and host as this repre- SOCIAL (NETWORK) FEATURES sents a virus–host symbiosis that now protects the host form the same and often other lytic (disease It has been 10 years since I published my book causing) viruses. Persistence is difficult to attain which was first to present the evolution of life not the indirect result of survivor of runaway (self- from a virus first perspective (19). If the main the- ish) replicons. Persistence inherently requires self- sis of that book can be stated in simple terms it is regulating and self-opposing functions. Thus, even that we must first consider the virus–host relation- ‘defective’ (and parasitic) components of viruses ship with better understand evolution of the host. (and transposons) can express virus-specific regula- From this perspective, we can then see that viruses tory (opposing) molecules (including ncRNA), were involved in most all major transitions of host clearly promote virus–host persistence, and biology in evolution. This will likely seem an over- respond to oppose lytic virus infection. Thus, the stated or even preposterous position to most read- presence of incomplete viral elements in host gen- ers. How could genetic parasites (viruses) be omes are not simply the remnants of past viral providing such fundamental capacity for host evo- infection and disease (virus sweeps), but should be lution? And why would they do so? But we have considered as the savior of the host lineage by pro- come to recently realize that viruses are omnipre- viding the capacity for self regulation and persis- sent so all life must survive in its virosphere habi- tence of viruses that can still threaten related tat. And such survival often involves virus species. Virus persistence provides a large selective themselves since virus, their defective and various advantage in the virosphere. It also presents a per- other genetic parasites (mostly called transposons) spective that is essentially the converse of the cur- can and often do provide virus resistance systems. rent view in evolutionary biology: viral persistence These viral colonizers then can also be used to is a big determinant of host survival with strong provide new sources of host complexity (such as effects on host group survival as well (via virus the placenta). Thus, to understand the deep role communication) (22). It defines a relationship virus plays, we must always consider a virus first between virus and host and between host them- perspective for the evolution of complexity in the selves that does not adhere to the predator–prey host. Essentially, the concept is that viruses are theory (23, 24). Nor does it adhere to the red- fully competent agents and editors of all host sys- queen hypothesis. Persistent virus is usually highly tems of instruction (DNA, RNA, epigenetic, trans- prevalent, silent, often genetically stable, co-evol- lational etc.) (20). Thus, they provide the host with ving with the host and usually transmitted from new sources of instruction systems (not errors). In parent (old) to offspring (young) or in close coor- addition, they promote network formation by pro- dination to host reproductive biology. Establishing viding coherent societies (quasispecies populations) persistence is difficult and can be thought of as of agents able to edit host code content (and add resulting from a successful hacking of host identity new identity) in a diffuse, distributed manner, networks to insert new code and promote survival which promotes the creation of and editing of host of a new more complex virus–host identity. Since regulatory networks. Thus, a viral role in the ori- persistence is always regulated, it is a mostly silent gin of the placental regulatory network can be state in which reactivation is tightly linked to host expected (21). Viruses possess all the advantages of (reproductive) biology with big consequences to evolution relative to host: extreme genetic adapt- host and virus fitness. This makes it much more ability, extreme diversity, extreme numbers, difficult to study. Asserting the core importance of extreme rates of genetic exchange, tolerance for persistent virus to host survival thus presents a big ‘unfit’ variation, and the ability to reassemble from break from historic views in evolutionary biology cryptic or ‘dead’ parts. They can transition and adds a process of selection that stems from between the chemical and living world. Thus, I am the horizontal transmission of persistent virus. As asserting that most initial genetic and selective the virosphere provides no ‘virus-free’ habitat for events that transform host regulatory complexity any life form, all living forms have adapted to are usually ‘pushed’ by virus action in a general their own viral habitat.

APPLYING THE VIRUS FIRST PERSPECTIVE Indeed, the QS-based feature of RNA viruses in TO PLACENTAL ORIGINS particular, allows us to think about the involvement of a virus ‘consortia’ as natural editors of host Let us now conceptually reconsider the placenta genetic content (31). Thus, the presence of such dis- from this virus perspective. Accordingly, ‘exapted’ tributed virus-derived (often defective) information viral env genes were not initially a convenient is not the residual product of errors, but the pro- source of genetic errors for promoting a more effi- duct of a QS-based colonization that directly cient placenta, but instead they were successful col- affected virus persistence, virosphere survival, host onizers that allowed the host lineage to control competition, and has also modified host identity various persisting viruses prevalent in their repro- systems. Host evolution is then free to adapt these ductive systems. The absence of these ERVs (envs) new viral network systems for host reproduction would leave these host susceptible to these same and survival. Successful virus colonization of host and/or other viruses (25). In this light, the presence thus promotes new complex host group and indi- of an impervious eggshell would preclude the colo- vidual identity that strongly affects competition nization of the shell membrane by active and self with related, but, uncolonized host populations and protective virus. But by creating a virus accessible leads to a modified virosphere. and rich tissue (trophectoderm; exposed after zona pellucida loss) which is needed for host reproduction, it promotes virus-based network solutions AN EXAMPLE OF ONGOING POPULATION- (e.g., genetic reprogramming, immune suppression, BASED ENDOGINAZATION: KOALA transformation, membrane fusion) to biologically RETROVIRUS difficult problems needed for vivipary to emerge (12). In addition, the new virus–host symbiont has Let us now outline some evidence that is most rele- acquired a very significant advantage compared the vant to this virus first scenario: that is, virosphere uncolonized host that remain susceptible to the dis- survival via persisting new virus information. The ease induced by related viruses. By modifying host Koala’s of Australia provide a particularly recent, identity (and immunity), these ERVs have thus set relevant, and informative story, see (32). Koalas the stage needed to promote virus persistence as have recently undergone an epidemic of retrovirus well as a new host reproductive strategy. This (KoRV, a gamma retrovirus)-mediated leukemia. reproduction strategy must in turn promote the Survivors, however, are undergoing endogenization reproductive success of the new virus–host combi- by an array of this same virus [which itself appears nation. However, the ERV-host relationship is to originated from an virus of rodents or bats (33)]. often dynamic and can continue to be susceptible Survivors do not die from leukemia, but they can to subsequent (competing, displacing) ERV colo- generally still produce the virus, which is now held nization as the persistent/acute virus habitat further in check by the endogenized (proviral) versions. evolves [as an exemplar, see JSRV (26)]. Wild mice Thus, they have established persistent infections show similar strain (mating) specific ERV-cancer with low disease. This endogenization is occurring biology (27, 28). This virus first scenario can thus by a complex process involving geographically (and provide an answer as to ‘why’ various distinct, but, tissue specific) distinct populations of both exoge- host lineage-specific viruses have been involved in neous and endogenous viruses involving an increas- their placentas and promote more complex host ingly large diversity of ERVs at low copy level (34). reproductive biology. These viruses thus resemble a Clearly, the endogenized KoRV must modify the competent, but, diverse gang of highly effective net- exogeneous KoRV-induced immune cell dis-regula- work hackers that seek to add new (viral) instruction (leukemia) that would otherwise occur. Indeed, tions. Because viruses can also disperse and interact wild Koalas with endogenized KoRV no longer as populations [quasispecies: QS, see (29)], viruses make antibodies to KoRV (33). Thus, virus infor- can modify distributed regulatory networks, not mation has become ‘one’ (symbiotic) with host and simply a specific loci or individual host. Such net- must be involved in virus control. But not all Koala work editing need not occur by a serial set of indi- populations have been equally affected by the epi- vidual events involving master individual type virus demic. Populations isolated in some islands have selection (30). They can instead involve quasis- much less disease and no endogenization. However, pecies-based and defective virus-based processes it would not be difficult to predict what might hap- that occur via population based virus colonization pen if the persistently infected mainland Koalas (see the Koala-virus example below). Such diversity now come in close contact with these isolated popu- makes these agents prone to multifunctional-, con- lations: survival of the persistently infected. This ditional-, and context-dependent interactions. new Koala KoRV virosphere requires that these

ERVs must remain in Koala genome along with population-based ‘virus persistence’, for example their capacity to cause tumors in non-endogenized via maternal antibodies transmitted through the populations. Thus, inducing lethal tumors (viral placenta at birth (40). Generally, such persistently harm) is not a breeding artifact but an important infected wild colonies are healthy. Yet the introduc- phenotype that can be transmitted to compel unin- tion of such wild mice into uncolonized ‘virus-free’ fected Koalas to either die or become one (persis- breeding colonies will usually result in reproductive tent) with KoRV. This situation promotes collapse of the entire virus-free colony. Thus, the reproductive isolation via the survival of KoRV history of persisting virus in a specific population endogenized Koalas. can have measurable and large survival consequences. Along these lines, we can consider the recent Ebola virus epidemic. Human male survivors GAMMA ERVS IN BATS, INTERACTION appear to persistently produce viruses in reproduc- WITH OTHER VIRUSES, AND tive tissue (41). Clearly, such persistently infected REPRODUCTIVE ISOLATION and sexually transmissible humans pose a major risk to all extant human populations. In contrast to Gamma retroviruses viruses (in contrast to len- rodents and bats, humans host a lot of persistent tiviruses) have been very successful in endogenizing DNA virus infections (polyomavirus, papillo- vertebrate species (as sources of ERVs). Gamma mavirus, adenovirus, herpes virus). Many human- retroviruses are mostly transmitted from old to specific viruses can also be found in the reproduc- young, often via reproductive tissue. Indeed, the tive organs. Herpes 6/7 and HSV-2 are especially reproductive tracts appear to generally provide a present in such tissues (42) and able to cause fatal ‘virus-rich’ and also a ‘virus-mixed’ habitat. Thus, encephalitis in unprotected newborns (via non- we might also anticipate that the placenta will need immune mothers) (43). Interestingly, these HVs per- to provide general mixed virus resistance and that sist via microRNAs that modify host apoptosis (44) such resistance will often be mediated by resident and can act cooperatively (45). HHV 6 can also or endogenized virus, as has been reported (35). integrate into chromosomal (centromere) DNA and Gamma-retro virus endogenization has also allow genomic maternal to fetal transmission (46). occurred in bats (36). Interestingly, ERVs seems to And the presence of such prevalent viruses in have also been involved in (helped) the endogeniza- human reproductive tissue can have major conse- tion of filoviruses (Ebola and Marburg) that has quences to other viruses, such as HIV-1. Indeed, in also occurred in bats (37). Given the capacity of the S. African epidemic, heterosexual transmission bats to host many persisting RNA viruses that are of HIV-1 depends heavily on co-infection with highly pathogenic to other species, their significant HSV-2 (47). A similar situation applies HIV-1 and genome colonization by the gamma retroviruses papillomavirus-induced cancers (48). Some viruses and rolling circular DNA virus defectives (express- can inhibit HIV (49). Interestingly, HERVK serum ing stem-loop miRNAs) is particularly interesting immunity can also affect HIV (50). Thus, the repro- (38). Such persistence by potentially lethal virus is ductive tract provides a virus-rich and mixed-virus not simply due to the fortuitous containment of an habitat. infection, but must have resulted from the establishment of a virus ‘addiction module’ in that the same defective virus must resist similar virus. But if the VIRAL IDENTITY AND IMMUNE NETWORKS host were to lose this (defective) virus information, VIA PARASITE DERIVED STEM-LOOP RNA it too would become susceptible to lethal infection. This, I suggest, defines a general issue that applies The importance of small non-coding RNAs for to all species and their viruses. It also can explain DNA virus persistence has recently become clear the emergence of sexually incompatible populations (51). But small non-coding RNA regions (with (due to incompatible persistent viruses). It is fur- stem-loops) are also the main identifying and regu- thermore likely that this issue also relates to sexual latory elements for most if not all RNA virus. incompatibilities seen via methylation (39). Virus Indeed, the definition of a gamma retrovirus persistence (addiction) is not specific to ERVs. In depends on such a stem-loop element. Also, within wild mouse colonies, for example, many viruses can retroviral LTRs and various other crucial control establish prevalent and highly stable persistent elements, stem-loop RNA are essential for identifi- infections, including MVM, MPV, Theilers virus, cation and regulatory function. Thus, RNA–RNA LCMV, MHV. Although some of these viruses are interactions via stem-loop regions promote the also capable of causing disease (even in wild establishment of RNA-based regulatory networks. colonies), they are often held in check by Other parasitic retro-agents (LINEs/SINES, alu’s)

© 2016 APMIS. Published by John Wiley & Sons Ltd 23 VILLARREAL can also be transcribed to produce non-coding major alteration to antiviral systems occurred in stem-loop RNAs. This suggests the possibility for early mammalian embryos. Also, lncRNA appears an extensive and mixed system of RNA-based regu- to be involved in early embryo programming, but lation all deriving from parasitic agents (52). From such RNAs are mostly derived from retrotrans- the host perspective small non-coding RNAs are posons (57). Other expressed functional repeat mostly thought to control host–virus (retroposon) RNAs are also derived from retrotransposon (58). interaction (53). Indeed, many human microRNA’s Indeed, ERVs themselves appear to be directly target retroviruses and ERVs (54). involved in fetal imprinting (59). Subsequently, embryo implantation involves reverse transcriptase activity that is derived from retroposons (60). In MULTIFUNCTIONAL NETWORK ISSUES FOR addition, the placenta clearly depends on the various THE PLACENTA SOLVED BY ERVS ERV env (syncytin) genes for both structural and functional needs (61). And at least ten lineages of Let us now further consider a virus first (virus-origin) mammals have acquired their own version of envs perspective for the origin of the placenta. Accord- for placental function. But the regulatory regions for ingly, virus should: (i) be involved the origin of the these syncytins is complex and composed of mixtures trophectoderm (first embryonic cell to differentiate), of LTRs and other regulatory regions derived from (ii) promote embryo implantation, (iii) promote com- other retroviruses (62). Indeed, the placenta does not plex placenta functions (including the cellular inter- seem to emerge from the acquisition of a lot of new face and invasion that feeds the embryo), (iv) genes, but instead appears to result from a more regulate the mother’s (host) immune response, (v) complex regulation of mostly previously existing communicate to reprogram the mother’s (host) phys- genes via a the emergence of a regulatory network iology and behavior to support the embryo during that was derived to a large degree from ERV-LTR pregnancy and after birth. These may seem like elements (63). These LTRs may be providing enhan- impossible and overly diverse tasks for viruses to cer-based gene regulation (64). In addition, LTR reg- help solve. This is on top of the fact that prior egg- ulation of insulin (65), poly-A control (66), NOS3 based reproduction must have already been working expression (67) have all been reported in the pla- well. But let us recall the general competence of virus centa. The acquisition of such regulatory complexity to regulate all systems of host control, including all in the placenta along with its coherence clearly pre- genetic, epigenetic, transforming systems via a pro- sent a big problem. How can we explain the origin cess involving transmissible-, diffuse-, ERVs-, and and integration of this network? A selective process ncRNA-based regulation. Such new regulations can involving serial individual fittest type selection be forcefully superimposed onto the host. Viruses are (exapted genes) does not account well for how net- good for this. In the next section (on Motherhood work coherence (cooperation) is attained. Indeed, I behavior and virus), a related complex issue of virus– think such step-wise selection is not possible given host reproduction reprogramming in the context of the successive and long durations needed for differ- parasitoid wasps is also presented, but involving dis- ential offspring survival and also the clear similarity tinctly different DNA viruses. With respect to the of viral elements to be distributed in the network in placenta, there is indeed evidence of viral (and antivi- order for the placental to be formed and function. ral) involvement in all of the above issues. Retroviral However, if instead we invoke a process similar to and retroposon RNA is highly expressed and regu- what is occurring in Koala ERV endogenization, we lated in the early embryo. And although DNA see evidence for population (network)-based colo- methylation is thought to restrict retrovirus and nization en mass. For this to occur via the placenta, retroposons, stem cells (and the placenta) are open the placenta must have been initially involved in viral to ERVs as their DNAs are hypomethylated (55). and antiviral control, as has been reported (35). lncRNA, siRNA and RNAi are all involved in early Indeed, both network emergence and antiviral status embryo regulation but are also either derived from should associate with ERV acquisition. And other retroposons or thought to regulate ERVs and retro- human viruses, such as HIV, HSV, could also be posons. The RNAi system in invertebrate animals affected (68). and plants is a core innate immune regulator of virus infection. Yet, its antiviral function was mostly lost in jawed vertebrates along with the emergence of the PLACENTAL VARIATION AND interferon system and adaptive immunity. Interest- CORRESPONDING ERVS ingly, it retains activity in the early embryo and is needed for early development (56), but is not appar- Biologically, the placenta varies greatly between ent in most somatic tissue. This strongly suggests a species (69), especially its invasiveness (70). The

24 © 2016 APMIS. Published by John Wiley & Sons Ltd VIRAL ORIGIN OF THE PLACENTA selective pressure and molecular basis for such vari- tive to this situation, we could expect that an initial ation has always been curious and difficult to and new ERV colonization resulted in a more explain. However, if we consider the involvement of stable persistent relationship between virus and host distinct viral ecologies and colonization histories in reproductive tissue, but that such a state will often placental origins, we might better explain such pla- involve the emergence of a new antiviral state and cental variability. Endogenous viruses (often defec- will occur via diffuse (Koala-like) network-like tive but some expressing env or gag) can clearly mechanisms. The resulting virus–host combination provide restriction factors that limit exogenous can still be subjected to further successful ERV col- virus susceptibility (71, 72), especially env (73). Such onizations that similarly further alter antiviral restrictions, however, are highly species, strain, and states and regulatory networks. In this light, the virus specific. The sheep retrovirus (JSRV) seems to recent report regarding the expression patterns of provide the best model for how a virus is able to HERV-K in human reproductive tissue (early both infect as exogenous disease causing virus yet embryos and placental cytotrophoblasts) is espe- be essentially required for reproduction as an endo- cially interesting (78). This ERV (env) does not genized virus (74). Such a relationship can establish function as a syncytin, so env gene exaption is not a dynamic, ongoing change to host ERV composi- a possible explanation for its presence. Also, these tion with degradation of older displaced copies sim- particular HERV Ks are relatively new and exoge- ilar to that as seen in primates (75). This JSRV nous additions to the human (but not chimpanzee) model, I suggests, captures the essence of the virus– genome, previously thought to provide no gene host dynamic in reproductive tissue. Reproductive function to humans due to polymorphisms. But transmission of virus becomes key. Indeed, various now it appears that this HERV K has provided other animals show reproductive virus transmission virus restriction factors (eng, gag) that are also and ERV changes, such as drosophila (76). How- important for embryo function and it has also ever, it is clear that many species are unable to pro- induced a more general antiviral state via the duce an exogenous virus from endogenous copies, IFITM1-specific interferon response that more gen- such as primates. Clearly, there are distinct variabil- erally inhibits other virus replication. This general ity in species-specific virus–host composition and response is operating thorough the HERVK- ecology. In many situations, I propose that it is encoded rec gene (a rev-like RNA transport pro- likely that other viruses of the reproductive tissue tein) that interacts with stem-loop viral RNA are also involved in the placental-ERV relationship. regions. In these specific embryos, however, about Thus, the ERVs expressed in placental tissues may 1/3 of cellular mRNAs have 30 UTRs that can bind need to be evaluated in the context of additional this rec and promote ribosome occupancy. Thus, virus mixtures and could have a more generalized HERVK has promoted the emergence of a new reg- antiviral affect. ulatory translational network in these human cells as well as a generalized antiviral response. Indeed, it has been previously observed that HERVK can ERV-DERIVED SYNCYTINS AND also interact with other viruses. For example, HIV GENERALIZED ANTIVIRAL ACTIVITIES infection activates many human-specific HERV-Ks found at centromeres (79). In addition, HCMV has Recently, the presence of syncytin-like ERV env in been reported induces HERVK (80), as has EBV marsupial reproductive tissue has been reported (81). The relationship between ERV envs and host (77). Marsupials have simple short-lived placentas is thus complex (82). But their ability to induce in which embryos implant for periods of about 1 general antiviral immunity as well as edit existing week, then the embryo is rejected from the interface host regulatory networks seems established. and must feed off of the pouch secretions. This simple placenta is much less invasive and long-lasting then that of mammals and it also does not promote MOTHERHOOD, BEHAVIOR, AND VIRUS? the exchange of blood (and antibodies) between mother and embryo. Yet even in this simplified pla- In an early paper I compared the role of ERVs in centa it seems an ERV env gene were needed to mammalian reproduction to that of polydnaviruses solve the interface problem posed by embryo in the reproduction of parasitoid wasp (12). In both implantation. Why might a virus also provide a motherhood and wasp embryo parasitization of good solution to this simplified biological situation? caterpillars, the host (caterpillar or mother) must Many ERVs that produce env genes in reproductive be able to support an embryo that is foreign. Both tissue are dynamic and changing on an evolution- these situations have some surprisingly similar sets ary time scale (75). If we adopt a virus first perspec- of biological issues to overcome, including immune

© 2016 APMIS. Published by John Wiley & Sons Ltd 25 VILLARREAL suppression of host, altered genetic program to sup- making a cocoon for them and guarding them port (feed) the embryo and altered development against other parasitoid wasp species, before the and behavior of the host. And in both situations, caterpillar dies. The mechanism by which this dra- endogenized viruses provide solutions to these com- matic behavior is induced is not understood. It plex problems. In the parasitoid wasp, the super- seems likely, however, that the polydnavirus is coiled closed circular DNA’s that are packaged into involved. Given the paucity of polydnaviral ORFs, VLPs are now accepted to have been clearly virus I would guess that regulatory RNAs are also likely derived (83). This complex set of distributed viral to be involved. In mammals, mothers must also genomes have become endogenized, are mostly undergo major behavioral changes. Indeed, some defective and are expressed exclusively in female increase in maternal brain size occurs during preg- wasp reproductive tissue. Relatively few viral ORFs nancy. It has been proposed that a general link to are expressed from these circular DNAs. Indeed, brain size is due to mother–oﬀ spring bonding (85). most DNA segments have repeated sequences It is also apparent that genomic imprinting essential within them. Interestingly, a diﬀerential microRNA for maternal brain development (86). Many of these response is seen in response to parasitization (84). changes are thought to be mediated by the pla- What is particularly fascinating about the para- centa, thus trophectoderm and the placenta are sitoid wasp is that in some situations, the para- likely sources of maternal behavioral control. sitized host caterpillar becomes immobile after the Unlike the fertilized parasitoid wasp egg, which is wasp parasites exit its body. The caterpillar, how- surrounded by polydnaviral VLP layer, the sur- ever, is induced to protect these wasp larvae by rounding placenta of mammalian embryo provides

Fig. 1. The RNA gangen hypothesis: group identity and cooperativity of an RNA collective that requires opposite functions for the genesis of life (social behavior of agents). Reprinted with permission from Villarreal, Luis P. 2015. ‘Force for Ancient and Recent Life: Viral and Stem-Loop RNA Consortia Promote Life’. Annals of the New York Academy of Sciences 1341 (1): 25–34.

26 © 2016 APMIS. Published by John Wiley & Sons Ltd VIRAL ORIGIN OF THE PLACENTA much of the endogenous virus (ERVs) required for stage for who will survive in the virosphere and what reproduction. A question that arises is if these may follow regarding virus–host selection. In sum- human ERVs might also be involved in controlling mary, viruses are competent in all biological codes the behavioral changes of motherhood. Is the pla- and various forms of communication. And since centa using ERVs in some way that alters the viruses can often function as diffuse populations, mother’s physiology and behavior? As the mecha- they are capable ‘hackers’ of complex network sys- nism by which maternal brains are modified by the tems not only able to reprogram a network but also placenta are not understood, we cannot answer this to provide novel solutions, often via mixed and question. Yet it is clear that the placenta does use defective and counteracting viruses (via quasis- ERV products (env mediated budding) to communi- pecies). In their capacity to promote persistence, cate with other maternal tissues. The human viruses also promote the infectious acquisition of cytotrophoblasts produce exosomes that have incor- systems of identity and immunity. Because viruses porated both syn-1 and syn2 (87). In addition, syn- are transmissible, they affect the relationships (com- 1 containing blood borne exosomes can regulate munication) not just within individuals but also to the immune response (88). Also these placental extended groups. This is the most powerful role. exosomes incorporate miRNA (89). Indeed, I have recently proposed that a quasispecies consortia (Gangen) of transmissible stem-loop RNA’s can better account for the origin of ribo- TRANSMISSIBLE SMALL RNAS, ERV zymes and the identity and communication networks REGULATION AND MOTHERHOOD of RNA world organisms (97). See Fig. 1. From the BEHAVIOR: EVERYTHING FROM VIRUS origin of life to the evolution of humans, viruses seem to have been involved. Thus, the large scale Given that RNAi (dicer) is active in preimplantation expansion ERV LTRs and other stem-loop RNA embryos (56) and the ancestral role of miRNA in elements (e.g., alu’s) in the recent evolution of the silencing retroposons in preimplantation embryos human brain, might also indicate a viral role. So (90), ERV activity seems highly regulated by ncRNA powerful and ancient are viruses, that I would sum- and specific to the placenta. Indeed, maternal plasma marize their role in life as ‘Ex Virus Omnia’ (from has high levels miRNAs (91). And the placenta is a virus everything). major site of secretion of exosome-containing micro RNAs (92, 93). Given the ability of miRNAs to control anxiety (94), a crucial maternal behavior, it thus REFERENCES seems plausible these env (syncytin) expressing exosomes are involved in regulating maternal behavior. 1. Loke YW. Life’s Vital Link: The Astonishing Role of Along these lines, the large human-specific C19MC the Placenta. Oxford, UK: Oxford University Press, miRNA cluster is one of the sets of miRNAs 2013: 276. expressed in exosomes and this cluster is also 2. Churakov G, Kriegs JO, Baertsch R, Zemann A, Bro- imprinted in placenta (95). But this primate-specific sius J, Schmitz J. Mosaic retroposon insertion pat- C19MC cluster is also expressed in fetal brain and its terns in placental mammals. Genome Res 2009;19:868–75. induced overexpression strongly associated with 3. Seman G, Levy BM, Panigel M, Dmochowski L. pediatric brain tumors, see (96). Thus, there seems to Type-C virus particles in placenta of the cottontop exist a clear pathway for the ERV mediated placen- marmoset (Saguinus oedipus). J Natl Cancer Inst tal control via ncRNA’s of maternal brain growth 1975;54:251–2. and behavior. The relationship of a mother to her 4. Larsson E, Nilsson BO, Sundstrom€ P, Widehn S. offspring is often considered in the context of Morphological and microbiological signs of endoge- – nous C-virus in human oocytes. Int J Cancer mother offspring conflict. Clearly, a parasitiod lar- 1981;28:551–7. vae is in a similar conflict with its caterpillar host. 5. Calarco PG. Intracisternal A particles in preimplanta- But as I have outlined above, in both situations tion embryos of feral mice (Mus musculus). Intervirol- endogenous viruses were involved in the origin (and ogy 1979;11:321–5. resolution) of these embryo–host relationships. 6. Maeda S, Yonezawa K, Yachi A. Serum antibody However, the survival advantage for the persisting reacting with placental syncytiotrophoblast in sera of patients with autoimmune diseases–a possible relation virus involved, is seldom considered. Viruses have to type C RNA retrovirus. Clin Exp Immunol long been dismissed as simple selfish agents, not cen- 1985;60:645–53. tral to evolution. And their persistence has been trea- 7. Nelson JA, Levy JA, Leong JC. Human placentas ted as a trivial matter. Here, I argue that the virus contain a specific inhibitor of RNA-directed DNA perspective should instead be considered first. For polymerase. Proc Natl Acad Sci U S A 1981;78:1670– the virus–host relationship (e.g., persistence) sets the 4.

8. Leong JC, Wood SO, Lyford AO, Levy JA. Purifica- glycoprotein-expressing proviruses interfere with infection of a specific inhibitor of reverse transcriptase tions. J Virol 1981;40:745–51. from human placenta. Int J Cancer 1984;33:435–9. 26. Black SG, Arnaud F, Palmarini M, Spencer TE. 9. Kato N, Pfeifer-Ohlsson S, Kato M, Larsson E, Ryd- Endogenous retroviruses in trophoblast differentiation nert J, Ohlsson R, et al. Tissue-specific expression of and placental development. Am J Reprod Immunol human provirus ERV3 mRNA in human placenta: 2010;64:255–64. two of the three ERV3 mRNAs contain human cellu- 27. Gardner MB, Chiri A, Dougherty MF, Casagrande J, lar sequences. J Virol 1987;61:2182–91. Estes JD. Congenital transmission of murine leukemia 10. Larsson E, Venables P, Andersson AC, Fan W, Rigby virus from wild mice prone to the development of S, Botling J, et al. Tissue and differentiation specific lymphoma and paralysis. J Natl Cancer Inst expression on the endogenous retrovirus ERV3 1979;62:63–70. (HERV-R) in normal human tissues and during 28. Gardner MB, Rasheed S. Retroviruses in feral mice. induced monocytic differentiation in the U-937 cell Int Rev Exp Pathol 1982;23:209–67. line. Leukemia 1997;11(Suppl 3):142–4. 29. Domingo E, Sheldon J, Perales C. Viral quasispecies 11. Harris JR. Placental endogenous retrovirus (ERV): evolution. Microbiol Mol Biol Rev. 2012;76:159–216. structural, functional, and evolutionary significance. 30. Villarreal LP, Witzany G. Rethnking quasispecies the- BioEssays 1998;20:307–16. ory: from fittest type to cooperative consportia. World 12. Villarreal LP, Villareal LP. On viruses, sex, and moth- J Biol Chem 2013;23:71–82. erhood. J Virol 1997;71:859–65. 31. Witzany G. Natural genome-editing competences of 13. Lee A, Nolan A, Watson J, Tristem M. Identification viruses. Acta Biotheor 2006;54:235–53. of an ancient endogenous retrovirus, predating the 32. Tarlinton RE. Koala retrovirus endogenisation in divergence of the placental mammals. Philos Trans R action. In: Witzany G, editor. Viruses: Essential Soc Lond B Biol Sci 2013;368:20120503. Agents of Life [Internet]. The Netherlands: Springer, 14. Murphy WJ, Pringle TH, Crider TA, Springer MS, 2012[cited 2013 Feb 9]. p. 283–91. Available from: Miller W. Using genomic data to unravel the root of http://link.springer.com/chapter/10.1007/978-94-007- the placental mammal phylogeny. Genome Res 4899-6_14 2007;17:413–21. 33. Fiebig U, Keller M, Moller€ A, Timms P, Denner J. 15. Blond JL, Beseme F, Duret L, Bouton O, Bedin F, Lack of antiviral antibody response in koalas infected Perron H, et al. Molecular characterization and pla- with koala retroviruses (KoRV). Virus Res cental expression of HERV-W, a new human endoge- 2015;16:30–4. nous retrovirus family. J Virol 1999;73:1175–85. 34. Ishida Y, Zhao K, Greenwood AD, Roca AL. Prolif- 16. Blond JL, Lavillette D, Cheynet V, Bouton O, Oriol eration of endogenous retroviruses in the early stages G, Chapel-Fernandes S, et al. An envelope glycopro- of a host germ line invasion. Mol Biol Evol tein of the human endogenous retrovirus HERV-W is 2015;32:109–20. expressed in the human placenta and fuses cells 35. Delorme-Axford E, Donker RB, Mouillet J-F, Chu T, expressing the type D mammalian retrovirus receptor. Bayer A, Ouyang Y, et al. Human placental tro- J Virol 2000;74:3321–9. phoblasts confer viral resistance to recipient cells. Proc 17. Dupressoir A, Lavialle C, Heidmann T. From ances- Natl Acad Sci U S A 2013;110:12048–53. tral infectious retroviruses to bona fide cellular genes: 36. Cui J, Tachedjian M, Wang L, Tachedjian G, Wang role of the captured syncytins in placentation. L-F, Zhang S. Discovery of retroviral homologs in Placenta 2012;33:663–71. bats: implications for the origin of mammalian gam- 18. Haig D. Retroviruses and the placenta. Curr Biol maretroviruses. J Virol 2012;86:4288–93. 2012;22:R609–13. 37. Taylor DJ, Dittmar K, Ballinger MJ, Bruenn JA. 19. Villarreal LP. Viruses and the Evolution of Life [Inter- Evolutionary maintenance of filovirus-like genes in net]. Washington, DC: ASM Press, 2005: xv, 395 p. bat genomes. BMC Evol Biol 2015;11:336. Available from: http://www.loc.gov/catdir/toc/ 38. Platt RN, Vandewege MW, Kern C, Schmidt CJ, ecip0419/2004013977.html Hoffmann FG, Ray DA. Large numbers of novel 20. Witzany G. Biocommunication and Natural Genome miRNAs originate from DNA transposons and are Editing. New York: Springer, 2009. coincident with a large species radiation in bats. Mol 21. Chuong EB, Rumi MAK, Soares MJ, Baker JC. Biol Evol 2014;31:msu112. Endogenous retroviruses function as species-specific 39. Schutt S, Florl AR, Shi W, Hemberger M, Orth A, enhancer elements in the placenta. Nat Genet [Inter- Otto S, et al. DNA methylation in placentas of internet]. 2013 [cited 2013 Feb 13]; Available from: http:// species mouse hybrids. Genetics 2003;165:223–8. www.nature.com/ng/journal/vaop/ncurrent/full/ 40. Gustafsson E, Blomqvist G, Bellman A, Holmdahl R, ng.2553.html Mattsson A, Mattsson R. Maternal antibodies protect 22. Villarreal LP. Persistence pays: how viruses promote immunoglobulin deficient neonatal mice from mouse host group survival. Curr Opin Microbiol hepatitis virus (MHV)-associated wasting syndrome. 2009;12:467–72. Am J Reprod Immunol 1996;36:33–9. 23. Villarreal LP, Defilippis VR, Gottlieb KA. Acute and 41. Rodriguez LL, Roo AD, Guimard Y, Trappier SG, persistent viral life strategies and their relationship to Sanchez A, Bressler D, et al. Persistence and genetic emerging diseases. Virology 2000;272:1–6. stability of Ebola virus during the outbreak in Kikwit, 24. Villarreal LP. Virus-host symbiosis mediated by per- Democratic Republic of the Congo, 1995. J Infect Dis sistence. Symbiosis 2007;44:1–9. 1999;179(Suppl 1):S170–6. 25. Robinson HL, Astrin SM, Senior AM, Salazar FH. 42. Balachandra K, Chimabutra K, Supromajakr P, Wasi Host Susceptibility to endogenous viruses: defective, C, Yamamoto T, Mukai T, et al. High rate of reacti-

vation of human herpesvirus 6 in children with dengue 58. Matylla-Kulinska K, Tafer H, Weiss A, Schroeder R. hemorrhagic fever. J Infect Dis 1994;170:746–8. Functional repeat-derived RNAs often originate from 43. Abel L, Plancoulaine S, Jouanguy E, Zhang S-Y, retrotransposon-propagated ncRNAs. Wiley Interdis- Mahfoufi N, Nicolas N, et al. Age-dependent mende- cip Rev RNA 2014;5:591–600. lian predisposition to herpes simplex virus type 1 59. Sekita Y, Wagatsuma H, Nakamura K, Ono R, Kagami encephalitis in childhood. J Pediatr 2010;157:623–9. M, Wakisaka N, et al. Role of retrotransposon-derived e1. imprinted gene, Rtl1, in the feto-maternal interface of 44. Gupta A, Gartner JJ, Sethupathy P, Hatzigeorgiou mouse placenta. Nat Genet 2008;40:243–8. AG, Fraser NW. Anti-apoptotic function of a micro- 60. Liu JL, Yang ZM. Non-coding RNAs and embryo RNA encoded by the HSV-1 latency-associated tran- implantation. Front Biosci Elite Ed 2011;3:1092–9. script. Nature 2006;442:82–5. 61. Mangeney M, Renard M, Schlecht-Louf G, Boual- 45. Feederle R, Haar J, Bernhardt K, Linnstaedt SD, laga I, Heidmann O, Letzelter C, et al. Placental Bannert H, Lips H, et al. The members of an Epstein- syncytins: genetic disjunction between the fusogenic Barr virus microRNA cluster cooperate to transform and immunosuppressive activity of retroviral envel- B lymphocytes. J Virol 2011;85:9801–10. ope proteins. Proc Natl Acad Sci U S A 2007;104: 46. Cheema A, Katta J, Velez AP, Medveczky M, Med- 20534–9. veczky PG, Quilitz R, et al. Encephalitis and inherited 62. Perot P, Bolze P-A, Mallet F. From viruses to genes: HHV-6. Infect Dis Clin Pract 2012;20:419–21. syncytins. In: Witzany G, editor. Viruses: Essential 47. Tan DHS, Kaul R, Walsmley S. Left out but not for- Agents of Life[Internet]. Netherlands: Springer, 2012 gotten: should closer attention be paid to coinfection [cited 2012 Dec 18]. p. 325–61. Available from: http:// with herpes simplex virus type 1 and HIV? Can J link.springer.com/chapter/10.1007/978-94-007-4899- Infect Dis Med Microbiol 2009;20:e1–7. 6_17 48. Auvert B, Lissouba P, Cutler E, Zarca K, Puren A, 63. Emera D, Wagner GP. Transposable element recruit- Taljaard D. Association of oncogenic and nononco- ments in the mammalian placenta: impacts and mech- genic human papillomavirus with HIV incidence. J anisms. Brief Funct Genomics 2012;11:267–76. Acquir Immune Defic Syndr 1999 2010;53:111–6. 64. Pavlicev M, Hiratsuka K, Swaggart KA, Dunn C, 49. Kanak M, Alseiari M, Balasubramanian P, Addanki Muglia L. Detecting endogenous retrovirus-driven tis- K, Aggarwal M, Noorali S, et al. Triplex-forming sue-specific gene transcription. Genome Biol Evol microRNAs form stable complexes with HIV-1 pro- 2015;7:1082–97. virus and inhibit its replication. Appl Immunohis- 65. Bieche I, Laurent A, Laurendeau I, Duret L, Giovan- tochem Mol Morphol 2010;18:532–45. grandi Y, Frendo J-L, et al. Placenta-specific INSL4 50. Jones RB, Garrison KE, Mujib S, Mihajlovic V, expression is mediated by a human endogenous retro- Aidarus N, Hunter DV, et al. HERV-K–specific T virus element. Biol Reprod 2003;68:1422–9. cells eliminate diverse HIV-1/2 and SIV primary iso- 66. Goodchild NL, Wilkinson DA, Mager DL. A human lates. J Clin Invest [Internet]. 2012 Nov 12 [cited 2012 endogenous long terminal repeat provides a Nov 28]; Available from: http://www.jci.org/articles/ polyadenylation signal to a novel, alternatively view/64560#sd spliced transcript in normal placenta. Gene 1992;121: 51. Seo GJ, Fink LHL, O’Hara B, Atwood WJ, Sullivan 287–94. CS. Evolutionarily conserved function of a viral 67. Huh JW, Ha HS, Kim DS, Kim HS. Placenta- microRNA. J Virol 2008;82:9823–8. Restricted Expression of LTR-Derived NOS3. Placenta 52. Mattick JS, Taft RJ, Faulkner GJ. A global view of [Internet]. 2008; Available from: http://www.ncbi.nlm. genomic information–moving beyond the gene and the nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed master regulator. Trends Genet 2010;26:21–8. &dopt=Citation&list_uids=18474398 53. Yeung ML, Benkirane M, Jeang KT. Small non-cod- 68. Patterson BK, Behbahani H, Kabat WJ, Sullivan Y, ing RNAs, mammalian cells, and viruses: regulatory O’Gorman MR, Landay A, et al. Leukemia inhibitory interactions? Retrovirology 2007;4:74. factor inhibits HIV-1 replication and is upregulated in 54. Hakim ST, Alsayari M, McLean DC, Saleem S, placentae from nontransmitting women. J Clin Invest Addanki KC, Aggarwal M, et al. A large number of 2001;107:287–94. the human microRNAs target lentiviruses, retro- 69. Reiss D, Zhang Y, Mager DL. Widely variable viruses, and endogenous retroviruses. Biochem Bio- endogenous retroviral methylation levels in human phys Res Commun 2008;369:357–62. placenta. Nucleic Acids Res 2007;35:4743–54. 55. Gimenez J, Montgiraud C, Oriol G, Pichon J-P, Ruel 70. Elliot MG, Crespi BJ. Genetic recapitulation of human K, Tsatsaris V, et al. Comparative methylation of pre-eclampsia risk during convergent evolution of ERVWE1/Syncytin-1 and other human endogenous reduced placental invasiveness in eutherian mammals. retrovirus LTRs in placenta tissues. DNA Res Philos Trans R Soc B Biol Sci 2015;370:20140069. 2009;16:195–211. 71. Ikeda H, Sato H, Odaka T. Mapping of the Fv-4 56. Svoboda P, Flemr M. The role of miRNAs and mouse gene controlling resistance to murine leukemia endogenous siRNAs in maternal-to-zygotic repro- viruses. Int J Cancer 1981;28:237–40. gramming and the establishment of pluripotency. 72. Goff SP. Retrovirus restriction factors. Mol Cell EMBO Rep 2010;11:590–7. 2004;16:849–59. 57. Kapusta A, Kronenberg Z, Lynch VJ, Zhuo X, Ram- 73. Malfavon-Borja R, Feschotte C. Fighting fire with say L, Bourque G, et al. Transposable elements are fire: endogenous retrovirus envelopes as restriction major contributors to the origin, diversification, and factors. J Virol 2015;89:4047–50. regulation of vertebrate long noncoding RNAs. PLoS 74. Arnaud F, Caporale M, Varela M, Biek R, Chessa B, Genet 2013;9:e1003470. Alberti A, et al. A paradigm for virus-host coevolution:

sequential counter-adaptations between endogenous 86. Keverne EB. Importance of the matriline for geno- and exogenous retroviruses. PLoS Pathog 2007;3: mic imprinting, brain development and behaviour. 1716–29. Philos Trans R Soc B Biol Sci [Internet]. 2013 Jan 5 75. Esnault C, Cornelis G, Heidmann O, Heidmann T. [cited 2015 May 14];368(1609). Available from: Differential evolutionary fate of an ancestral primate http://www.ncbi.nlm.nih.gov/pmc/articles/ endogenous retrovirus envelope gene, the EnvV Syn- PMC3539356/ cytin, captured for a function in placentation. PLoS 87. Vargas A, Zhou S, Ethier-Chiasson M, Flipo D, Genet 2013;9:e1003400. Lafond J, Gilbert C, et al. Syncytin proteins incorpo- 76. Pelisson A, Sarot E, Payen-Groschene G, Bucheton rated in placenta exosomes are important for cell A. A novel repeat-associated small interfering RNA- uptake and show variation in abundance in serum mediated silencing pathway downregulates comple- exosomes from patients with preeclampsia. FASEB J mentary sense gypsy transcripts in somatic cells of the 2014;28:3703–19. Drosophila ovary. J Virol 2007;81:1951–60. 88. Tolosa JM, Schjenken JE, Clifton VL, Vargas A, 77. Cornelis G, Vernochet C, Carradec Q, Souquere S, Barbeau B, Lowry P, et al. The endogenous retro- Mulot B, Catzeflis F, et al. Retroviral envelope gene viral envelope protein syncytin-1 inhibits LPS/ captures and syncytin exaptation for placentation in PHA-stimulated cytokine responses in human blood marsupials. Proc Natl Acad Sci U S A 2015;112: and is sorted into placental exosomes. Placenta E487–96. 2012;33:933–41. 78. Kammerer€ U, Germeyer A, Stengel S, Kapp M, Den- 89. Ouyang Y, Mouillet J-F, Coyne CB, Sadovsky Y. ner J. Human endogenous retrovirus K (HERV-K) is Review: Placenta-specific microRNAs in exosomes – expressed in villous and extravillous cytotrophoblast good things come in nano-packages. Placenta [Inter- cells of the human placenta. J Reprod Immunol net]. [cited 2014 Jan 24]; Available from: http://www. 2011;91:1–8. sciencedirect.com/science/article/pii/ 79. Contreras-Galindo R, Kaplan MH, He S, Contreras- S0143400413007959 Galindo AC, Gonzalez-Hernandez MJ, Kappes F, 90. Svoboda P, Stein P, Anger M, Bernstein E, Hannon et al. HIV infection reveals wide-spread expansion of GJ, Schultz RM. RNAi and expression of retrotrans- novel centromeric human endogenous retroviruses. posons MuERV-L and IAP in preimplantation mouse Genome Res [Internet]. 2013 May 8 [cited 2013 May embryos. Dev Biol 2004;269:276–85. 16]; Available from: http://genome.cshlp.org/content/ 91. Miura K, Miura S, Yamasaki K, Higashijima A, early/2013/05/08/gr.144303.112 Kinoshita A, Yoshiura K, et al. Identification of preg- 80. Bergallo M, Galliano I, Montanari P, Gambarino S, nancy-associated MicroRNAs in maternal plasma. Mareschi K, Ferro F, et al. CMV induces HERV-K Clin Chem 2010;56:1767–71. and HERV-W expression in kidney transplant recipi- 92. Chim SSC, Shing TKF, Hung ECW, Leung T, Lau T, ents. J Clin Virol 2015;68:28–31. Chiu RWK, et al. Detection and characterization of 81. Sutkowski N, Conrad B, Thorley-Lawson DA, Huber placental microRNAs in maternal plasma. Clin Chem BT. Epstein-Barr virus transactivates the human 2008;54:482–90. endogenous retrovirus HERV-K18 that encodes a 93. Luo S-S, Ishibashi O, Ishikawa G, Ishikawa T, superantigen. Immunity 2001;15:579–89. Katayama A, Mishima T, et al. Human villous tro- 82. Grow EJ, Flynn RA, Chavez SL, Bayless NL, phoblasts express and secrete placenta-specific micro- Wossidlo M, Wesche DJ, et al. Intrinsic retroviral RNAs into maternal circulation via exosomes. Biol reactivation in human preimplantation embryos and Reprod 2009;81:717–29. pluripotent cells. Nature [Internet]. 2015 Apr 20 [cited 94. Muinos-Gimeno~ M, Espinosa-Parrilla Y, Guidi M, 2015 May 18];advance online publication. Available Kagerbauer B, Sipila€ T, Maron E, et al. Human from: http://www.nature.com/nature/journal/vaop/ microRNAs miR-22, miR-138-2, miR-148a, and miR- ncurrent/full/nature14308.html 488 are associated with panic disorder and regulate 83. Bezier A, Annaheim M, Herbiniere J, Wetterwald C, several anxiety candidate genes and related pathways. Gyapay G, Bernard-Samain S, et al. Polydnaviruses Biol Psychiatry 2011;69:526–33. of braconid wasps derive from an ancestral nudivirus. 95. Noguer-Dance M, Abu-Amero S, Al-Khtib M, Science 2009;323:926–30. Lefevre A, Coullin P, Moore GE, et al. The primate- 84. Gundersen-Rindal DE, Pedroni MJ. Larval stage specific microRNA gene cluster (C19MC) is imprinted Lymantria dispar microRNAs differentially expressed in the placenta. Hum Mol Genet 2010;19:3566–82. in response to parasitization by Glyptapanteles flavi- 96. Morales Prieto DM. Markert UR. MicroRNAs in coxis parasitoid. Arch Virol 2010;155:783–7. pregnancy. J Reprod Immunol 2011;88:106–11. 85. Weisbecker V, Goswami A. Brain size, life history, 97. Villarreal LP. Force for ancient and recent life: viral and metabolism at the marsupial/placental dichotomy. and stem-loop RNA consortia promote life. Ann N Y Proc Natl Acad Sci U S A 2010;107:16216–21. Acad Sci 2015;1341:25–34.