Eur J Clin Microbiol Infect Dis DOI 10.1007/s10096-017-3111-8

REVIEW

The pathogenesis of microcephaly resulting from congenital infections: why is my baby’s head so small?

L. D. Frenkel1,2 & F. Gomez3 & F. Sabahi 4

Received: 29 August 2017 /Accepted: 17 September 2017 # Springer-Verlag GmbH Germany 2017

Abstract The emergence of Zika-virus-associated congenital review, we integrate all these findings to create a unified hy- microcephaly has engendered renewed interest in the patho- pothesis of the pathogenesis of congenital microcephaly in- genesis of microcephaly induced by infectious agents. Three duced by these infectious agents. of the original “TORCH” agents are associated with an appre- ciable incidence of congenital microcephaly: cytomegalovi- rus, rubella virus, and Toxoplasma gondii. The pathology of Introduction congenital microcephaly is characterized by neurotropic infec- tious agents that involve the fetal nervous system, leading to Microcephaly has become an issue of increased interest since brain destruction with calcifications, microcephaly, sensori- the recognition that it is a consequential manifestation of con- neural hearing loss, and ophthalmologic abnormalities. The genital Zika virus infection [1–3]. Microcephaly is generally inflammatory reaction induced by these four agents has an defined as a head circumference ≤ 2 standard deviations below important role in pathogenesis. The potential role of “strain the mean for gestational age [4]. Zika virus associated micro- differences” in pathogenesis of microcephaly by these four cephaly and other clinical manifestations of vertical transmis- pathogens is examined. Specific epidemiologic factors, such sion from mother to fetus during gestation are similar to those as first and early second trimester maternal infection, and the caused by three of the original “TORCH” agents (Toxoplasma manifestations of congenital infection in the infant, shed some gondii, rubella virus, and cytomegalovirus) [3, 5–7]. Most light on the pathogenesis. Immune aspects of normal pregnan- microcephaly associated with congenital Zika virus infection cy and their role in congenital infections is examined. In this and with other congenital infections (Tables 1 and 2)isa reflection of neurotropism for fetal central nervous system Lawrence D. Frenkel, Fernando Gomez and Farzaneh Sabahi contributed (CNS) cells [6], with massive destruction of neural tissue dur- equally to this work. ing the early development of the CNS of the fetus, which predominantly occurs during the first and early second trimes- * L. D. Frenkel ters of pregnancy [3, 8, 38, 39]. This neural tissue destruction [email protected] is most frequently associated with congenital cytomegalovirus (CMV), rubella virus, Toxoplasma gondii, and Zika virus in- 1 Departments of Pediatrics and Microbiology, University of Illinois fections, and less commonly with other pathogens. CNS man- College of Medicine, Rockford, IL, USA ifestations are often associated with other manifestations in- 2 Department of Pediatrics, Division of Immunology, Allergy, and cluding intrauterine growth retardation (IUGR) [10, 19, 40], Infectious Disease, The Children’s Hospital at Saint Peter’s ophthalmologic anomalies [29, 41, 42] (including University Hospital, New Brunswick, NJ, USA microphthalmia) [43], neuro-developmental abnormalities, 3 Department of Specialty Medicine, Rocky Vista University School of and sensorineural hearing loss [4, 10, 20]. It is important to Osteopathic Medicine, Parker, CO, USA recognize clinical manifestations associated with congenital 4 Department of Virology, Faculty of Basic Medical Sciences, Tarbiat infection, other than microcephaly, such as seizures, sensori- Modares University, Tehran, Iran neural hearing loss, and ophthalmologic abnormalities. Prompt diagnosis and treatment of the above-noted Eur J Clin Microbiol Infect Dis

Table 1 Comparison of major pathogens associated with congenital microcephaly

Microorganism Risk of symptomatic CNS manifestations Mode of transmission Vaccine availability References congenital infectiona (%) at birthb (%)

Zika virus 8–10 80c Mosquito and sexual In clinical trials [8, 9] Cytomegalovirus 1–15 10–50 Sexual and oral In clinical trials [10–14] Rubella virus 20–50 10–20 Respiratory Available [3, 15, 16] Toxoplasma gondii 10–20 5–10 Ingestion None available [17, 18] a Presenting at birth with primary maternal infection b Microcephaly or structural defects of the CNS c As demonstrated on neuroimaging manifestations, including “silent” seizures (not apparent on Among the pregnancies with positive laboratory evidence of observation and diagnosed on electroencephalogram), and ap- maternal infection, possible Zika virus-associated birth defects propriate medical intervention for other manifestations of were reported in 8%, 5%, and 4% during the first, second, and symptomatic congenital infection may improve the quality third trimester infections respectively [9]. of life and preserve function [8, 44]. Most infections with this virus are transmitted by the bite of an Aedes mosquito [47, 48]. It is less frequently transmitted sexually [49, 50]. The full appreciation of the Zika virus epi- The pathogens demic in the Americas was delayed and complicated because its clinical manifestations in adults are similar to those caused by Zika virus dengue and chikungunya viruses; the three viruses can cross- react serologically, and 80% of maternal infections are asymp- The epidemiology of Zika virus infection was recently tomatic [45, 51]. The major manifestations of symptomatic Zika reviewed by Calvet, et al. [45]. Serologic studies of the Yap virus infection in adults include macular or papular rash, pruri- Island epidemic suggested that the attack rate of individuals tus, headache, arthralgia, myalgia, non-purulent conjunctivitis, living on the island was above 70%, with less than 20% of the and low-grade fever [19, 45, 46]. The association of Zika virus inhabitants having symptomatic disease [46]. However, a re- infection with Guillain–Barré syndrome is likely, but the mech- cent review of Zika virus pregnancy registry data [8]reported anism remains to be further defined [52, 53]. that only 38% of the pregnant women were asymptomatic and Estimates suggest that as many as 10% of pregnant women 61% were symptomatic. The incidence of birth defects was with acute Zika virus infection deliver infants with serious similar in symptomatic and asymptomatic pregnant women. congenital anomalies, usually including microcephaly [8, 19,

Table 2 Microcephaly-associated pathogens: shared reported findings*

Zika virus CMV Rubella virus T. gondii

Microcephaly† ++++ IUGR† ++++ CNS calcifications† ++++ Sensorineural hearing loss† ++++ Chorioretinal inflammation†, atrophy, or scars ++++ Hydrocephalus, hydranencephaly or ventriculomegaly ++– + Malformed gyri +++– Cortical dysplasia ++–– Cerebellar hypoplasia or aplasia ++–– Encephalitis or meningoencephalitis ++++ Microphthalmia ++++ Optic nerve atrophy ++++ Cataracts + – ++ Hepatic dysfunction – ++ + Organism details Flavivirus (ss + RNA) Herpesvirus (dsDNA) Togavirus (ss + RNA) Protozoan (intracellular) References [4, 19–28][29–32][29–34][29–31, 35–37]

* Differences are discussed elsewhere in this article † classic findings for TORCH syndrome agents Eur J Clin Microbiol Infect Dis

54, 55] and other signs of severe CNS damage (developmental the first or second trimesters of pregnancy rather than the third delay, seizures, spasticity, and arthrogryposis) [7, 56], eye ab- trimester, there appears to be a greater chance of more severe normalities (cataracts, microphthalmia, and chorioretinitis) congenital CMV infection [10]. Maternal CMV infection is as- [21, 42], and intrauterine growth retardation [7, 21, 22, 48]. sociated with a 30% chance of congenital infection and as much In one study, 7% of these congenitally infected infants with as a 15% chance of clinically apparent manifestations at birth microcephaly had sensorineural hearing loss [4]. Currently (symptomatic congenital CMV), with up to 50% of these infants available data suggest that the incidence of symptomatic or manifesting microcephaly [11, 12, 14, 64]. asymptomatic congenital infection is similar with symptom- Unlike the usual case for rubella virus [65], and probably for atic or asymptomatic maternal infection [8, 9]. Zika virus infections, congenital CMV infection occurs in in- Thirteen infants with evidence of central nervous system fants born both to women who are seronegative or who are damage on neuro-imaging, but without microcephaly at birth, seropositive prior to pregnancy [12, 64]. This is believed to born to mothers with documented Zika virus infection (none represent recurrent (reactivation of latent virus) or reinfection during the third trimester), have been described [57]. Two (superinfection with a new strain) of CMV in the pregnant host others, born to mothers with Zika virus infection during the third [12]. The frequency of transmission of CMV from mother to trimester, have also been described [58]. Data from infants born fetus when the mother was seropositive prior to pregnancy is with normal head size and laboratory evidence (quantitative estimatedtobe1%[12]. The rate of transmission of CMV from polymerase chain reaction, IgM capture enzyme-linked immu- mothers who were CMV seropositive and the severity of clinical nosorbent assay) of congenital Zika virus infection at birth, re- manifestations seem to be similar to those for CMV seronega- vealed that microcephaly developed after birth. Half had evi- tive mothers who experience primary infection during the preg- dence of first or second-trimester maternal infection. However, nancy, although these data remain to be confirmed [12]. they all had CNS abnormalities on neuroimaging at birth, con- The major manifestations seen with symptomatic congenital sistent with severe congenital Zika virus infection [57]. CMV include [66] CNS abnormalities (microcephaly with Prolonged shedding of Zika virus in a congenitally infected scattered punctuate intracranial calcifications and brain atrophy, infant has been reported [23, 59]; this may be more common, neurodevelopmental abnormalities) [67], intrauterine growth re- as is the case with congenital CMV and rubella. tardation [14], ophthalmologic abnormalities (chorioretinitis, retinal atrophy, and microphthalmia, often unilateral) [43, 68], Cytomegalovirus evidence of disseminated organ involvement (thrombocytope- nia, petechiae, purpura hepatosplenomegaly, jaundice, and The epidemiology and immunology of this infection are com- hyperbilirubinemia) [10, 69], and sensorineural hearing loss plicated by issues of symptomatic and asymptomatic congenital which can first develop in early childhood [10]. Of note is the infection occurring in mothers with primary and non-primary fact that infants infected from a maternal source often shed infection, and by late manifestations of congenital infection in CMV in their urine for a prolonged period of time [70]. asymptomatic infants at birth [11, 12, 60]. In the immune- competent non-pregnant host, the vast majority of CMV infec- Rubella virus tions are subclinical. In spite of a robust host immune response, once a normal host is infected by CMV, the virus is never elim- Rubella virus is transmitted by respiratory secretions and usually inated, causing a persistent asymptomatic infection [60]. results in clinically apparent but mild, self-limited disease, in However, in the immune-suppressed individual and fetus, the children and adults. Rubella virus disease is manifest, in sero- virus may demonstrate uncontrolled replication, dissemination, negative individuals, with a homogenous macular rash, fever, organ damage, and death [60]. A variety of CMV-induced neu- pharyngitis, arthralgia, and adenopathy, often with large post- rological disorders have been described in immune- auricular lymph nodes [71]. Rubella is also known to cause mild compromised individuals [61]. In older children and adults, it encephalitis in the apparently normal host [72]. Prior to the can on rare occasions manifest as heterophile-negative mono- availability of rubella vaccine, huge epidemics of rubella disease nucleosis [13]. It is a sexually transmitted pathogen [13], and is occurred, and subsequently large numbers of infants were born also transmitted via salivary secretions, especially between chil- with congenital anomalies. Rubella infection does not occur in dren in day-care situations and from these infected children to seropositive individuals and consequently congenital rubella is a their CMV-seronegative mothers [62]. manifestation of primary maternal infection [71]. CMV is the most common cause of congenital infection in The congenital rubella syndrome was first described by Greg the United States, occurring in 0.5–1% of pregnancies, as well in 1941 [41]. Greg’s original report focused on the association of as being a leading cause of sensorineural hearing loss and men- congenital cataracts, microphthalmia (both often unilateral), tal retardation [10, 12]. Maternal infection with CMV during congenital heart defects, and the fact that the mothers of these gestation is generally asymptomatic or mild [10, 63]asistrue infants had rubella virus infection early in their pregnancies. In for Zika virus infections. When primary infection occurs during the report by Miller et al., the frequency of congenital rubella Eur J Clin Microbiol Infect Dis infection was more than 80% during the first 12 weeks, 54% at Toxoplasma gondii 13–14 weeks, and 25% at the end of the second trimester [15]. Exposure to maternal infection in the first 12 weeks of gestation Toxoplasma gondii is one of the most common parasitic in- resulted in fetal loss or symptomatic congenital infection in fections worldwide [78]. T. gondii is a parasite which is orally almost all cases, and symptomatic infection in 35% of those ingested or congenitally transmitted to the human host [35]. fetuses infected at 13 to 16 weeks gestation. No congenital Oral transmission of T. gondii is through exposure to cat feces defects attributable to rubella were found with infections that or from eating under-cooked meat. The time during gestation occurred after 16 weeks gestation [15, 41, 72]. when maternal infection is acquired, inoculum size, and the Less commonly, congenital infection may be asymptomat- genetic background of the individual host, all influence the ic; although stigmata of severe congenital infection, in these transmission and manifestations of T. gondii infection in infants, may develop later in the life of the infected infant [16, immune-competent human hosts and congenitally infected in- 72]. Symptomatic congenital infection is manifested by hear- fants [73, 79]. In the immune-competent and non-pregnant ing loss in 90%, cardiac anomalies in 50%, mental retardation hosts, acute toxoplasmosis is asymptomatic or unrecognized in 40%, microcephaly in 10%, and cataracts and other oph- in 80% of individuals [35]. The most common manifestations thalmologic anomalies in 40% [72]aswellaspurpura,pete- in the other 20% of immune-competent individuals are lymph- chiae, jaundice, bone lesions, and pneumonia [71]. As is the adenopathy and fatigue without fever. Occasionally, individ- case with Zika virus and CMV, congenital rubella infection uals may manifest acute infection with a heterophile-negative may occur after subclinical or clinically apparent maternal mononucleosis picture (clinically similar to that caused by infection, although most maternal infections resulting in con- CMV), or even myocarditis or polymyositis [80]. genital rubella are symptomatic [41, 71]. The US incidence estimates for congenital T. gondii infec- In congenital rubella infection, as is also true with CMV, tions range from 0.8 to 20 per 10,000 live births [35, 73]. seemingly uninfected infants may demonstrate delayed hearing Congenital infection is usually a result of primary maternal loss, and mental as well as other defects, after several years [16, T. gondii infection during or immediately preceding pregnancy 72]. The development of delayed stigmata also seems to be the [35, 80]. However, reactivation infection in immune- case with untreated congenital toxoplasmosis [73]. Delayed find- compromised women and infection with a second serotype have ings are thought to be a manifestation of persistent infection as been reported [73, 79]. Evidence suggests that congenital infec- well as the persistent inflammatory response [72]. It is not yet tion with T. gondii is a result of maternal parasitemia leading to known if additional delayed manifestations will also be true for infection of the placenta and then the fetus [35]. Both symptom- Zika virus congenitally infected infants, although delayed micro- atic and asymptomatic maternal infections may result in con- cephaly and hearing loss have now been reported after maternal genital infections; however, most infections in pregnant women Zika virus infection [57] (Sampaio Boaventura V, Azevedo A, de areasymptomatic[80]. Although more primary maternal infec- Andrade N, et al. Prospective evaluation of hearing function after tions occurring during the third trimester result in congenitally presumably congenital Zika virus infection. Presented at the First infected infants (65% as opposed to 25% and 54% in the first International Conference on Zika Virus, Washington DC, 2017). and second trimesters respectively) [35], maternal primary in- It is generally agreed that fetal involvement is a direct reflec- fection during the first and second trimesters of pregnancy tion of primary maternal rubella viremia followed by placental (weeks 10 to 24) is more likely to result in manifestations of infection, with subsequent acute and chronic infection of fetal severe symptomatic congenital infection, including CNS in- tissues leading subsequently to tissue destruction in the involved volvement [73, 80]. This dichotomy between rates of transmis- organs [16, 38, 71, 74]. The complex role of fetal immune and sion of primary maternal infection to the fetus and symptomatic inflammatory responses remains unresolved. Viral shedding fetal infection is thought to reflect differences in placental cell from both rubella virus [71] and CMV severe congenitally in- susceptibility to infection by T. gondii organisms [81]. The pla- fected infants can persist for as many as several years [11, 16, cental cytotrophoblasts in the anchoring villi are more suscepti- 43, 72], as may be the case for Zika virus [4]. Maternal factors ble to infection and replication of the parasite than are the have been demonstrated to suppress rubella virus-specific cell- syncytiotrophoblast cells [81]. mediated immunity during pregnancy as measured by phytohe- Congenital toxoplasmosis is subclinical in approximately magglutinin (PHA) and mixed lymphocyte-culture responsive- 75% of infected infants and, as noted, is inversely related to ness [75]. Mothers of severely symptomatic congenitally CMV- the gestational age at which maternal infection is acquired infected infants have also been demonstrated to have CMV- [35]. The most common manifestations of severe congenital specific impairment of cell-mediated immunity during pregnan- infection include developmental delays, intracranial calcifica- cy as measured by lymphocyte-mediated cytotoxicity [76]. tions, seizures, hydrocephalus, chorioretinitis, anemia, and Fortunately, the incidence of congenital rubella infection in the jaundice. Microcephaly is noted in 5–15% of severely symp- US, since the acceptance of routine infant rubella immunization, tomatic congenitally infected infants, and microphthalmia in is almost zero [77]. 1–2%. The unique phenomenon of acute chorioretinitis Eur J Clin Microbiol Infect Dis manifesting in individuals between 15 and 20 years of age Studies of congenital CMV have clearly demonstrated the with mild or asymptomatic congenital infection, remains not isolation of multiple and diverse strains, including all 7 gN, 5 well explained; as does the predilection for the macula, basal gB, and 2 gH genotypes, of CMV from congenitally infected ganglia, and periventricular areas of the brain in severe con- infants but failed to demonstrate one specific genotype associ- genital toxoplasmosis [73]. Ocular toxoplasmosis is more of- ated with symptomatic congenital infection [88]. Thirty-nine ten bilateral in congenital disease than in acquired disease percent of the infants shedding CMV in their saliva, urine, or [82]. In other, often subclinically, congenitally infected indi- blood were infected by more than one genotype [88]. As yet, viduals who manifest acute chorioretinitis during adolescence, there is no clear difference in fetal outcomes of infants congen- it is thought that subtle changes in their cell-mediated immune itally infected with different strains or more than one strain of status may allow an ocular T. gondii cyst to reactivate and CMV [88]. Other investigators have demonstrated that multiple rupture [73]. Of particular interest is the fact that microcephaly CMV genotypes are found at the maternal-fetal placenta inter- may be unrecognized in congenital toxoplasmosis at birth and face [89]. In the case of congenital CMV, it is not clear whether be recognized between 12 and 24 months of age. Some of different or multiple maternal CMV strains are related to these infants, with delayed recognition of microcephaly, were reactivated virus or to reinfection with a different strain [14, 64]. thought to have a subclinical congenital infection. This sug- Thirteen serotypes of rubella virus forming two clades are gests that antimicrobial treatment may be important in limiting recognized [90]. The incidence of congenital rubella in Japan is the ongoing destruction of CNS tissue. The efficacy of treating comparable to that in the United States [91]. Phylogenetic anal- congenital toxoplasmosis is supported by the excellent results ysis of rubella virus strains from an outbreak in Spain, from of collaborative studies [83]. 2004 to 2005, were analyzed and only one genotype (strain) was isolated. All the samples were from unvaccinated individ- uals; two had symptomatic congenital rubella [92]. The last “Strain” differences major epidemic of rubella in the United States occurred in 1964 and almost certainly was caused by one strain of rubella Historically, pathogen “strain” differences associated with the virus, with an estimated 30,000 cases of congenital rubella syn- occurrence and severity of congenital infections have not re- drome [16]. Two rubella vaccines are generally used in the US ceived the attention which they seem to deserve. Originally, and elsewhere and have been very successful in terminating Zika virus was distributed throughout much of Africa and rubella epidemics and thus congenital rubella infection. This, Asia. Utilizing phylogenetic analysis of nucleotide and amino along with the lack of strain differences in rubella viruses within acid sequences of open reading frames, two geographically a geographic region, suggests that any strain differences are not distinct lineages of strains (clades) have been demonstrated: associated with the development of congenital rubella. the African and Asian [22, 84]. After about 50 years of circu- Some data suggests that the manifestations and severity of lation, the Asian clade was responsible for an epidemic in toxoplasmosis are related, in part, to T. gondii serotypes [73]. 2007, on Yap Island, Federated States of Micronesia (one of Three main serotypes are recognized: types I, II, and III [93]. the Pacific islands). Viral evolution has been suggested as an Other data implicates atypical strains of T. gondii, largely from explanation for the apparent changes in some serious manifes- SouthAmerica,inmoreseverecongenitalinfectionandma- tations of Zika virus disease as it spread from Africa, and ternal reinfection [94]. In an analysis of 82 French isolates, perhaps to Asia, and from there to Yap Island and then to type II strains were found in over 90%; type I and atypical the Americas [85]. These changes include an increase in cases strains were not found. Six cases exhibited fetal death, 16 of congenital microcephaly and Guillain–Barré syndrome. exhibited severe stigmata at birth, and six had no clinical ev- There is strong conservation among all Zika virus strains over- idence of congenital toxoplasmosis [93]. In a study in the US, all, with less than 12% divergence at the nucleotide level [22]. looking at serotypes of 193 congenitally infected infants, 39% Using additional sequences from Columbia, Mexico, Panama, were type II and 61% were non-type II. The non-type II con- and Martinique, Adiga proposed a clade, separate from that in genitally infected infants had more severe disease [95]. Brazil [86]. The group also described a common mutation in the so-called Latin American clade, at the C-terminal end of the NS5 gene which is involved in RNA-dependent RNA Pathogenesis and pathology of microcephaly polymerase activity. The functional genomics of the Zika viral proteins was recently further reviewed by Wang et al. [87], Contributors to congenital microcephaly include environmen- who noted the ability of Zika virus to infect human placental tal factors such as exposure to infectious agents and toxins in macrophages and cytotrophoblasts as well as several CNS cell utero as well as numerous genetic disorders [96]. The patho- types. The significance of these findings remains to be deter- physiology of microcephaly associated with maternal infec- mined, but they suggest that Zika virus “strains” may be re- tions varies with the pathogen but shares some similarities. All lated to differences in manifestations of Zika viral infection. of the etiologies of microcephaly lead to decreased brain tissue Eur J Clin Microbiol Infect Dis mass (volume) and thereby contribute to a decrease in head described are decreased numbers of thick flat gyri size, since it is the increase in brain mass which normally (pachygyria) or the absence of gyri (lissencephaly). The de- expands the soft fetal skull. crease in brain matter is accompanied by increased intracranial There are several groups of genetic abnormalities that to- fluid that can express itself as ventriculomegaly, hydrocepha- gether contribute to the entity referred to as primary congenital lus, hydranencephaly (absence of cerebral hemispheres of var- microcephaly [97]. In general, these genetic abnormalities iable degree), and/or increased fluid surrounding the brain. In lead to abnormally encoded proteins involved in replication, some cases, the overall increased amount of fluid in the cranial death, or migration of neuronal cells. The decreased replica- vault can lead to, at least transient, enlargement of the head tion, abnormal cell migration, or increased neuronal cell death [20]. As the fluid levels recede, most likely as a consequence during development of the brain leads to a decreased number of increased development of the CNS vasculature, there may of neurons in the cerebrum and thus to an overall decrease in be a collapse of the cranial vault that can lead to microcephaly, brain tissue volume, which is accompanied by microcephaly. with overriding of the cranial bones and the appearance of a Infectious agents can also affect replication, death, or mi- somewhat superiorly flattened cranium with overlying redun- gration of neuronal cells. The infectious agents most common- dant, rugose, scalp skin. ly associated with microcephaly are neurotropic intracellular Microscopically, as a group, the predominant findings are: pathogens. These pathogens are neurotropic to both fetal and cortical thinning, cortical dysplasia (abnormal neural migra- adult CNS cells, but the degree of damage caused by these tion), foci of confluent necrosis (cystic change), neural cell agents is greater in the developing brain. In general, these degeneration with phagocytosis (sometimes associated with pathogens can enter the central nervous system via a variety microglial nodules), foci of mineralization or calcification, of methods including entry via infected peripheral neurons, and foci with perivascular lymphocytic inflammatory infil- endothelial cell infections, and by migration of infected in- trates. There may also be accompanying mononuclear inflam- flammatory cells to the CNS, but the methods used for CNS mation of the meninges. Neurotropic infectious agents can be entryvariesbypathogen[61, 98, 99]. Once in the CNS, the identified in neurons via direct observation of the pathogen, pathogens can infect a variety of cells, including neuronal classic cytopathic viral effects, or identification of the patho- precursor cells, leading to a decrease in cortical neuron num- gen by other means, such as immunohistochemistry or in situ bers and decreased brain mass. nucleic acid hybridization studies. Comparison of the pathologic findings of intrauterine infec- Prominent neuronal apoptosis and necrosis with axonal tions by Zika virus, cytomegalovirus, rubella virus, and degeneration occur with intrauterine Zika virus infections, T. gondii reveals some shared clinical manifestations (Table 2). and can be accompanied by occasional neuron-shaped calci- Neurologic findings are quite commonly shared, as is intrauter- fications in the cortex and subcortical white matter [20, 23, ine growth retardation. With decreased intrauterine fetal move- 122]. Viral inclusions are not seen [122]. Inflammation tends ment for any reason (including intrauterine infections with neu- to be mild and composed primarily of CD3+ T-lymphocytes ral damage and oligohydramnios) the infant can develop joint and a few macrophages distributed primarily around vessels, contractures. Arthrogryposis is well described in congenital but additional lymphocytes can also be found scattered Zika virus infection [19, 20, 24, 57, 122], Club foot is described throughout the brain tissue [20, 23, 122, 125]. Meningitis in congenital rubella syndrome [33]. Shared findings also in- can co-exist [20]. Activated microglia and gliosis are common clude sensorineural hearing loss and ocular deformities. findings [20, 23, 122, 125]. There is abnormal formation of Hepatomegaly accompanied by elevated transaminases, cerebral structures, with abnormalities of the gyri described hyperbilirubinemia (with or without jaundice), splenomegaly, (such as lissencephaly, pachygyria, agyria, polymicrogyria), petechiae, and anemia may also occur in infants secondary to holoprosencephaly, and dysgenesis and agenesis of the corpus intrauterine infections by some of these pathogens. callosum [19, 20, 25, 122, 125]. Also described are cerebral All of these infections can induce variable degrees of in- atrophy with subventricular and white matter tissue loss, areas flammatory reaction (mediated predominantly by lympho- of cystic change, cerebral atrophy, ventriculomegaly, cerebel- cytes and macrophages) as well as tissue loss [23, 33, lar hypoplasia, malformed or absent thalami, hypoplasia of the 122–124]. The tissue destruction caused by the pathogens pons and brainstem, and microcephaly, both proportionate and and associated inflammatory responses to the pathogens pre- disproportionate [19, 20, 23, 25, 122, 123, 125]. Glial- sents variably from microscopic to large areas of tissue loss. neuronal heterotopia has been reported in the brainstem, cere- Damage can be focal or more extensive, with thinning of the bral white matter, and the subarachnoid space [20, 125]. cerebral cortex or development of areas of hypoplasia or Calcifications are found in gray and white matter, in the pa- aplasia of the brain (e.g., cerebellum and corpus callosum). renchyma or in perivascular spaces, and can be diffusely dis- Occasionally, absence of neural tissue can cause aplasia of the tributed in the cerebrum with a subcortical (band-like distri- cerebellar vermis and cerebral schizencephaly (deep clefts or bution at the gray–white matter junction) or periventricular grooves in the cortex) in some of those infected. Also pattern, and can also be found in germinal matrix, thalamus, Eur J Clin Microbiol Infect Dis basal ganglia, cerebellum, and brainstem [25, 122, 125]. reaction to the pathogens; this can lead to additional tissue Spinal cord tract abnormalities occur [125], and neurogenic necrosis. Microglial nodules can occur, and areas of necrosis paraspinal muscle atrophy has been reported [20]. Other CNS are prone to calcification [36]. Intracranial calcifications, ven- pathology-related findings seen with intrauterine Zika virus tricular enlargement, and hydrocephalus are found in congen- infections include redundant rugose scalp skin, overlapping itally infected patients [35, 123, 128]. Calcifications tend to be cranial sutures, arthrogryposis, and ocular abnormalities such disseminated, periventricular, or confined to the basal ganglia as microphthalmia, retinal pigment mottling, chorioretinal at- [30, 128]. Microcephaly is seen in congenitally infected in- rophy, optic nerve atrophy, colobomas, etc. [21, 26, 123]. fants at birth but appears to develop more commonly after Findings described with congenital CMV infections in- birth [123]. Ocular lesions such as chorioretinitis, optic nerve clude changes consistent with meningoencephalitis [126]. atrophy, cataracts, microcornea, and microphthalmia may be Within the cerebrum, pathology is seen predominantly in the present [29, 35, 37, 123, 128]. ventricular or subventricular regions, basal ganglia, brainstem, In a review article, Tian-Ming Yuan et al. summarize numer- and diencephalon [124, 126, 127]. It has been demonstrated ous studies which highlight the fact that activation of inflamma- that CMV can infect cells within the CNS extensively, includ- tory responses to pathogens is an important process that contrib- ing endothelial cells, neurons, astrocytes, oligodendrocytes, utes to brain damage [129]. Intracellular pathogens induce reac- microglial cells, choroid plexus cells, and ependymal cells tions initially within the innate immune system via pattern- [61, 124]. CMValso appears to have the ability to infect neural recognition receptors that sense pathogen-associated molecular precursor cells [124]. The infection leads to mononuclear in- patterns such as Toll-like receptors located on cell surface or on flammation accompanied by apoptosis, tissue necrosis, and internal cell membranes (endosomes, endoplasmic reticulum, gliosis. The encephalopathy can be extensive and accompa- etc.) and subsequently in the adaptive immune system via nied by foci containing cells with CMV inclusion bodies [61, antigen-presenting cells (glial cells and macrophages) and T- 126, 127]. Tissue destruction leads to cerebral atrophy, micro- lymphocytes [130]. The end result of the inflammatory response cephaly, ventriculomegaly, parenchymal and ependymal is the destruction of neurons via apoptosis and tissue damage, cysts, microcephaly, abnormal development of cerebral gyri primarily seen as an increase in microglial cells or cystic areas of (polymicrogyria, lissencephaly, etc.), cerebellar lesions, and tissue loss. In addition, the inflammatory processes that occur calcifications [123, 124, 126, 127]. Calcifications are usually can lead to demyelination of the white matter. Eventually, areas periventricular, corresponding to the areas of ventricular and with tissue necrosis can undergo dystrophic calcification, which subventricular pathology, but can also be diffuse [30, 124, is visualized radiographically as calcifications within the brain 126, 127]. Other CNS pathology-related findings seen with tissue. The process of dystrophic calcification will not be seen if congenital CMV infections include sensorineural hearing loss, examination of the brain tissue occurs shortly after infection, as and ocular abnormalities, including chorioretinitis, optic atro- this process takes a prolonged time period to occur, and it may phy, and anophthalmia [29, 123, 124, 126]. Apoptosis of un- never occur if the inflammatory process leads to areas of necro- infected “bystander” cells has been described in the retina sis that end up devoid of necrotic cellular debris via effective infected with CMV [124]. phagocytic processes. Pathologic findings described in the CNS with intrauterine Toll-like receptors (TLRs) and major histocompatibility rubella virus infections are severe brain damage, complex class (MHC) I molecules have been demonstrated leptomeningitis, parenchymal, perivascular and to be present in the central nervous system neurons, microglia, periventricular tissue necrosis, and subarachnoid hemorrhage and astrocytes [131]. Expression of proteins from an intracel- [33]. Calcifications may be periventricular or diffuse [30]. lular pathogen in conjunction with MHC I molecules leads to Small head circumference, sensorineural hearing loss, and oc- activation of cytotoxic CD8+ lymphocytes and consequently ular lesions (chorioretinitis, microphthalmos, microcornea, the death of infected cells. Activation of the TLRs of an in- cataracts, corneal ulcers, optic atrophy, depigmentation of fected cell can help to start an inflammatory reaction and lead the retina, etc.) are also reported [29, 33, 123]. to the eventual death of the infected cells via apoptosis [132]. Features described in congenital toxoplasmosis include the Recently, it has been shown that Zika virus directly infects presence of cerebral inflammation, primarily periventricular embryonic human cortical neural progenitor cells in vitro and periaqueductal, consisting primarily of lymphocytes and [133]. Using a human cerebral organoid model [134], activa- macrophages sometimes accompanied by occasional plasma tion of TLR3 in neurons by Zika virus leads to dysregulated cells [36]. The inflammatory response appears to be in re- neurogenesis, apoptosis, and abnormal cell differentiation sponse to either cells containing tachyzoites, or ruptured [135]. A fission yeast model has demonstrated that specific bradyzoite containing toxoplasma cysts. Both the infected Zika virus-encoded proteins can directly cause cycle inhibi- cells and adjacent uninfected cells can undergo necrosis. tion and cell death via cellular oxidative stress-activated sig- Intact toxoplasma cysts induce no inflammatory reaction. naling pathways [136], raising the possibility that unique Zika Vasculitis can occur in areas involved by the inflammatory virus encoded proteins may also be major contributors to the Eur J Clin Microbiol Infect Dis observed severe and rapid neural tissue loss. Infection of cells the concept of active tolerance mechanisms against the fetus, by most RNA viruses activates TLR3 which recognizes while the mother still is able to elicit normal immune re- double-stranded RNA, and single-stranded RNA viruses can sponses against most pathogens. Pregnancy seems to be asso- activate TLR7 and TLR8 [137]. Other TLRs recognize addi- ciated with increased susceptibility and severity to pathogens tional viral, bacterial, and parasite components and may be such as toxoplasma, listeria, influenza virus, and herpes linked to the pathogenesis of neural cell death in those infec- varicella-zoster virus [145–148]. These pathogens all share tions. For example, TLR9 responds to dsDNAviruses, such as the requirement that intact cell-mediated immunity must be CMV, recognizing non-methylated viral CpG-containing maintained to protect the host from destructive infection. DNA [137], and profilin of the T. gondii eukaryote actin cy- Humoral or antibody-mediated immunity has been shown to toskeleton is recognized by TLR5 [138, 139]. It appears that be intact during pregnancy, and is more effective against ex- TLRs present in brain neurons, microglia, and astrocytes can tracellular pathogens. Antibodies facilitate opsonization and be triggered by parasite and viral components and may initiate ingestion by neutrophils and monocytes [148]. Pregnancy- similar inflammatory and apoptotic pathways that may lead to associated changes point to the down-regulation of cell- similar pathologic findings in infections due to intracellular mediated immune responses. Women with cell-mediated au- pathogens. toimmune disorders, such as rheumatoid arthritis, tend to ex- perience remission during pregnancy [149–151]. However, antibody-mediated autoimmune disorders, such as systemic The immunology of normal pregnancy lupus erythematosus, often intensify during pregnancy [148, 152]. A better understanding of the immunology of normal Maternal immune tolerance of the fetus is essential for a human pregnancy may help us understand factors that result in healthy outcome of pregnancy. This remarkable phenomenon congenital infection and affliction. was acknowledged by Peter Medawar in the 1950s [140]and A longitudinal study of normal pregnant women in three has significantly impacted research in the field. In his article, trimesters of pregnancy showed a significant decrease in three theories were proposed to explain the lack of immuno- CMV-specific T-helper response, interleukin (IL)-2, and inter- logical reaction towards the fetus during pregnancy: (a) the feron gamma (IFN-γ) expression in the first and third trimesters anatomical separation between the mother and the fetus, (b) of pregnancy [153]. Our laboratory has also documented a pro- immaturity of antigenic expression by the fetus, and (c) the gressive reduction of PHA and viral-specific cell-mediated im- immunological inertness of the mother. It has now been munity in pregnant CMV or herpes simplex virus seropositive proved that fetal cells are separated from the maternal immune women from the first trimester towards the third trimester of system. The point of contact is fetal extravillous trophoblast pregnancy [154]. Other studies reported a decreased Th1/Th2 cells, which have poor antigenicity due to lack of expression ratio as well as a decrease in natural killer (NK)-cell function of MHC I [except human leukocyte antigen (HLA)-C] and [155–158] and number and changes in regulatory T-cells (Tregs) MHC II molecules [141]. Class I HLA antigens are expressed and Th17-cells [111]. The healthy immune system needs to be on the surface of various types of normal mature cells and regulated to prevent undesired immune stimulation against self signify self, while class II HLA markers are only expressed as well as harmless non-self organisms or the fetus. on the surface of immunologically active cells. Class I HLA antigens are further divided to HLA-1a and HLA-1b antigens. HLA-1a consists of HLA-A, B, and C alleles, and contains The immunology of the maternal-fetal interface several hundred family members; while HLA-1b consists of HLA-E, F, and G and has very low allelic variation. In this section, we summarize the most important findings on Direct presentation of fetal cell HLA molecules does not the role of decidual innate and adaptive immunity. A number normally occur at the maternal-fetal interface (the decidua). of different types of maternal leukocytes reside at the The mucous membrane (endometrium) of the pregnant uterus maternal-fetal interface; located between the uterine mucosa is shown to be a bi-directional exchange surface for cells and and the extraembryonic tissues of the developing conceptus molecules. Various different subsets of maternal immune cells (Table 3). These leukocytes have diverse roles in implantation, (up to 50%) are found at this interface [142]. Fetal cells and placental development, control of infectious diseases and par- anti-fetal HLA antibodies are present in maternal circulation, turition [112, 159, 160]. First-trimester human decidual leu- and maternal cells are present in babies after birth. This phe- kocytes are primarily composed of NK-cells (~70%) and mac- nomenon is recognized as maternal-fetal microchimerism rophages (~20%). The proportion of T-cells is variable (~10– [143, 144]. The decidua is, therefore, an important site for 20%). Dendritic cells (DCs), B-cells and other cells are rare immune tolerance and various tolerogenic mechanisms are [112]. Maintaining a delicate balance between decidual im- described at this site. The research with regard to the initial mune cells and their interactions is essential for maternal im- theory of immunological inertness of the mother has added to mune tolerance towards the fetus. Eur J Clin Microbiol Infect Dis

Table 3 Immune cells at the maternal-fetal interphase

Cell type “Major” population Function References surface markers

Decidual natural killer cells - CD56 + bright, CD16- - Expression of angiogenic factors [100–103] - High level production of various cytokines and chemokines - Recognize HLA-C and HLA-G antigens - Interact with CD14+ APCs to expand Tregs and inhibit Th17-cells Decidual dendritic cells - Myeloid origin - Potent stimulators in MLRa [104] - DC-SIGN-, HLA-DR+ - Prime CD4+ T-cells towards Th2 by lower IL-12 production - BDCA1+, BDCA3+ - CD11c+, CD83+ Decidual macrophages - CD14+, DC-SIGN+ - Mainly “M2” immune-regulatory macrophages [105–110] -HighCD86+,highCD80+ - Induce an immunosuppressive phenotype - High HLA-DR+ - Phagocytosis of apoptotic cells - Lower pro-inflammatory IL-1 and higher anti-inflammatory IL-4 and IL-10 production - Involved in vascular remodeling, placenta formation, and parturition - Bind to HLA-G+ cells, thereby induce tolerance towards trophoblastic cells Decidual T regulatory cells - CD4+, CD25+, Foxp3+ - Mainly induced in the decidua (induced Tregs or iTregs) [111–116] - Expansion by fetal alloantigens - Immune tolerance towards fetal antigens Decidual Th17-cells - CD4+, Th17+ - Significant reduction in healthy pregnancy helps in tolerance induction [112, 117, 118] towards fetal antigens Decidual CD4 + HLA-G+ T-cells - CD4+, HLA-G+ - Immunosuppressive [119] - IL-10 production Decidual CD8+ T-cells - CD8+, CD45RO+ - Most abundant decidual T-cells in term pregnancy [118, 120, 121] - Highly differentiated activated effective memory (EM) phenotype (EM cells mainly) - Lower expression of perforin and granzyme B - Virus-specific decidual CD8+ T-cells are reported a Mixed leucocyte reactions Most human decidual NK-cells (dNKs) express immuno- blood and induced locally upon exposure to stimuli, such as suppressive molecules and have limited cytolytic activity, a transforming growth factor beta (TGF-β) and foreign anti- property that has been speculated to be in part due to their gens, to exert efficient suppressive activity [114]. In contrast interaction with HLA-C and HLA-G antigens on trophoblastic to Treg-cells, Th17 proinflammatory cells are significantly cells. Through their interaction with CD14+ antigen- decreased during healthy pregnancy. This causes increased presenting cells (APCs), dNK-cells play a role in the expan- immunosuppression, decreased immunostimulation, and low- sion of Treg-cells and inhibition of Th17 inflammatory lym- er inflammatory effects. These changes may be part of the phocytes [100–103]. reason for increased pathogenicity of certain infectious agents Decidual macrophages (dMACs) are the most abundant during normal human pregnancy [117]. APCs in the decidua throughout pregnancy [105]. These cells CD8+ T-cells are the most important cells that can directly are proposed to play a central role in maintaining a balance recognize allogeneic fetal (paternal) HLA molecules. In contrast between pro-inflammatory state and immune tolerance during to peripheral blood, CD8+ T-cells are the most abundant T-cells pregnancy [104–106]. in decidual tissue of term pregnancy. In addition, these CD8+ T- Compared to peripheral blood T-cells, decidual T-cells are cells are largely highly differentiated with activated effector more heterogeneous with major differences and specific func- memory phenotype; in contrast to the peripheral blood in which tions. Different T-cell subsets contribute in unique ways to most CD8+ T-cells are naive unprimed T-cells. These cells ex- pregnancy outcome; some protect the fetus from immune re- press less perforin and granzyme in comparison to their periph- jection, and others contribute to adverse pregnancy outcomes, eral counterparts [118]. The specificity of these CD8+ T-cells is such as spontaneous abortion. In general, there is limited re- debated. One hypothesis is that these cells are specific for fetal cruitment of activated T-cells to the decidua and harmful re- alloantigens. A more recent study showed that decidual CD8+ T active clones of cytotoxic T lymphocytes are eliminated by cells are virus-specific CD8+ memory T-cells, which may pro- clonal deletion [113, 144]. tect the fetus against the virus infections [120]. Decidual Tregs play a significant role in the maintenance of In addition to the local decidual immunosuppressive mech- the fetus by suppressing the maternal immune response. They anisms described above, which serve to protect the fetus from are mostly induced Tregs (iTregs) and are thought to originate rejection, the fetal immune cells need to tolerate non-inherited from CD4+ T cells specifically recruited from the maternal maternal antigens that cross the placenta. Fetal T lymphocytes Eur J Clin Microbiol Infect Dis are recognized at 10 weeks of gestation, and interestingly are production of the components of inflammation pathways. This produced by distinct hematopoietic stem cells from those in virus also manipulates the host cell intrinsic autophagy response adults [161, 162]. However, upon antigenic stimulation, they for better pathogen replication and disease establishment [167]. are easily differentiated to Treg-cells, comprising 15% of fetal The importance of innate immunity in protection against Zika peripheral lymphocytes. Overall, the immune suppression at virus is highlighted by the fact that various flaviviruses, includ- the maternal-fetal interface results in increased susceptibility ing Zika virus, have been shown to have developed a number of of the fetus to infectious agents [163]. strategies to circumvent the components of innate immune re- sponses throughout their co-evolution with their hosts [168–170]. By analogy to other flaviviruses, such as dengue The immunopathogenesis of congenital affliction virus, CD4+ and CD8+ T-cells are expected to play important roles against Zika virus. A recent study on four Zika virus- Data to fully explain the immunopathogenesis of congenital immune subjects detected specific CD4+ T-cell responses infections is somewhat limited. However, a number of factors against NS1 and E proteins of the virus [171]. seem to be in play. Since the fetus is essentially a graft of Recently, Zika virus has been shown to replicate in first- foreign tissue, several immune tolerance mechanisms, includ- trimester human maternal-decidual tissues in ex-vivo organ cul- ing down-regulation of innate and acquired immune re- tures [172]. In this model, efficient viral replication and spread to sponses, are involved in protecting the developing fetus. the adjacent cells was shown by a rapid increase in Zika virus Unfortunately, these same mechanisms may leave the mother load and high titers of free virus infectious particles. This may and fetus more susceptible to infection. The immune status of explain, in part, the rapid and severe degree of CNS tissue de- the pregnant individual seems to determine pathogen load at struction, leading to severe microcephaly. Mid-gestation mater- the maternal-fetal placental interface; with higher pathogen nal-decidual tissues also remained susceptible to the virus. Fetal- loads facilitating the invasion of the fetal circulation and dis- derived chorionic villi cells were most susceptible to Zika virus semination to fetal organs. It is of interest that primary mater- in the first trimester, and showed decreased viral titers as the fetal nal infection with all four of these pathogens is generally age increased. In addition, a genome-wide transcriptome analy- asymptomatic or mild. The recognition that primary maternal sis demonstrated an increase in components of innate immune infection is generally necessary for the development of Zika responses in the decidual tissues of the placenta [3]. When these virus, rubella virus, and T. gondii congenital infection sug- results were compared with those of human cytomegalovirus, gests that specific humoral or cell-mediated immunity is pro- Zika virus was shown not to increase immune-cell activation in tective. Possibly, genetically determined, variations in im- the maternal-fetal interface; instead, Zika virus upregulated pla- mune mechanisms from individual to individual may contrib- cental cell death and apoptosis. In addition, in contrast to CMV, ute to the development of congenital infection and symptom- type I and III interferons were stimulated by Zika virus [3]. This atic congenital infection. For example, more severe depression points to the importance of interferons in Zika virus infection. of maternal cell-mediated immunity may be a deciding factor The kinetics of replication and spread of Zika virus in the host in the genesis of congenital infection and determine its sever- compared to cytomegalovirus may be explained, in part, by the ity. The most important events in embryogenesis, such as the efficient cell-free mode of Zika virus spread in comparison to the development of the central nervous system, take place early in cell-associated mode of CMV spread [3]. In addition, kinetic gestation, and thus first-trimester and second-trimester mater- differences may explain the severity and character of microceph- nal infections more commonly lead to symptomatic congenital aly associated with congenital Zika virus infection. infection. In addition, it has been suggested that some patho- gens have a tropism for specific fetal cells and tissues. Cytomegalovirus

Zika virus Specific genes that determine cellular tropism and others that modify the immune responses to this virus have been identi- Limited peer-reviewed literature is currently available with re- fied [11]. As noted above, innate and adaptive immune re- gard to the pathogenesis of congenital Zika virus infection. As is sponses can limit but not prevent the spread and latency of the case with rubella virus, congenital infection appears to occur CMV [11, 76]. It is accepted that maternal viremia, placental only with primary maternal infection, suggesting that specific infection, and subsequent dissemination to fetal organs, in- humoral immunity to Zika virus is protective against congenital cluding the CNS, are all involved in the pathogenesis of con- infection. Primary Zika virus infection stimulates the production genital CMV. However, the complex role of immunologic of interferons and various interferon stimulatory genes factors remains largely unresolved [70]. CD8+ T lymphocytes [164–166]. These can promote an augmented inflammatory re- play a central role in the control of CMV replication [70]. sponse, which seems to be especially important at the maternal- However, both asymptomatic and symptomatic congenitally fetal interface. Zika virus has been shown to induce the infected infants showed similar CD8+ responses, suggesting Eur J Clin Microbiol Infect Dis that they were not sufficient to prevent congenital CMV infec- tion. NK cells are also believed to play a role in controlling CMV infection [70]. Infants with symptomatic congenital CMV have a lower NK activity than asymptomatic infants [173]. Cytomegalovirus infection is clearly different than rubella virus infection, and probably Zika virus infection, in that it does occur in women who were seropositive prior to their current pregnancy [11, 64]. This reactivation or reinfection is unpredictable in terms of which women can be so affected and in terms of in which trimester it occurs. Restriction enzyme analysis of CMV isolated from pregnancies, in which pairs (conceived and born at different times) of congenitally infect- ed siblings were studied, demonstrated that two of the pairs reflected reactivation infection—the viruses were identical— but in one pair the CMV strains were different, reflecting new CMV infection [174]. It is important to note, however, that the infants born to two of these subsequent pregnancies (reactivation) were asymptomatic, whereas the reinfection pregnancy produced a severe congenitally infected infant. In general, however, the impression that pregnancies in seropos- itive women produce less severely, or even subclinically, in- Fig. 1 Lymphocyte transformation in pregnant and non-pregnant fected infants must be confirmed [11]. subjects (Frenkel LD, Bhumbra NA, Neel K, Biekart E, Nankervus GA Some seropositive pregnant women have positive CMV (1983) Lymphocyte transformation in pregnant and non-pregnant subject. viral cultures intermittently during their pregnancies; this find- Presented at the Conjoint Meeting on Infectious Diseases, Montreal, Canada.) ing does not seem to be correlated with fetal infection [11]. Similarly, seropositive non-pregnant women also demonstrate intermittent CMV shedding. Of further interest is the fact that interferon secretion in severely symptomatic congenitally infect- over 40% of normal postpartum seropositive women with ed infants and their mothers have been reported [176]. A signif- normal pregnancies and infants also demonstrate CMV shed- icantly delayed lymphoproliferative response has been reported ding into their breast milk [11]. CMV congenitally infected in mothers who transmitted CMV to their babies compared to infants have positive urine cultures for months or even years non-transmitting mothers [177]. Furthermore, the transmitting after birth [43]. T-cells restrain viral replication and prevent mothers (i.e., those who delivered CMV infected infants) had disease, but do not eliminate the virus or preclude transmis- lower CD4+ and CD8+ T-cells compared to non-transmitting sion [175]. Human CMV is an ancient and ubiquitous patho- women [178, 179]. Taken together, various studies indicate that gen that elicits a robust immune response in the immune- defective function and delayed maternal CD4+ T-cell produc- competent host, and has evolved effective immune defense tion may be an important contributor to CMV transmission from strategies involving innate, humoral, and cell-mediated immu- mother to the fetus during gestation, and defective fetal CD4+ T- nity, which circumvent these host immune defenses. This al- cell response may contribute to CMV congenital infection and lows active CMV infection to be suppressed but does not disease. eliminate the virus from the infected host. In CMV-seropositive pregnant women, naive CD8+ effector As previously discussed, several, but not all, investigators T-cells are reduced by 50%. In pregnant seronegative women, have reported that CMV-seropositive women demonstrate de- no such change is observed [180]. During pregnancy, levels of pressed CMV-specific T-cell-mediated immunity during gesta- maternal CD8+ T-cell activation are normal. CMV-infected fe- tion [153, 154]. CMV-specific lymphocyte proliferation, IL-2, tuses showed a significant increase in activated CD8+ T-cells as and IFN-γ production are significantly reduced in the first and early as 22 weeks’ gestation. However, fewer CD8+ T-cells third trimesters of pregnancy (see Fig. 1). Apparently, in CMV from these infected fetuses were capable of releasing IFN-γ seropositive subjects, most of the memory T-cells are specific upon stimulation with CMVantigen [11, 70]. The definitive role for CMVantigens, and this increases as the seropositive subjects of CD8+ T-cells, with regard to congenital infection, remains age [175]. Specific impairment of cell-mediated immunity oc- unclear. However, induction of pro-inflammatory cytokines as- curs in mothers of infants with congenital CMV infection [76], sociated with fetal CMV infection is manifested in a shift to- but not all investigators have been able to reproduce this finding wards the pro-inflammatory Th1 profile and away from the [11]. Suppressed CMV-specific lymphocyte proliferation and normal anti-inflammatory Th2 pattern. In combination with Eur J Clin Microbiol Infect Dis direct cytopathic effects, this results in damage to, and impair- Toxoplasma gondii ment of, critical organ functions in the fetus [38]. CMV can down-regulate HLA-A, B, C, and even HLA-E The pathogenesis of congenital T. gondii is related to host ge- and HLA-G molecules on infected cells. CMV can escape netic susceptibility and host immune status as well as to genet- immune control by inhibiting the optimum presentation of ically determined pathogen factors [73]. Human studies have their peptides to CD8+ T lymphocytes. The loss of HLA mol- suggested that clinical manifestations and severity of congenital ecules on infected cells can make them susceptible to attack by toxoplasmosis are determined, in part, by parasitic genotype, NK-cells [181], facilitating cellular and tissue destruction. On especially atypical strains [185]. Innate, NK-cell function is the other hand, inadequate NK-cell function has been found to known to limit acute toxoplasma infection, but T-helper cell be associated with severe CMV infections in children [182]. and CD8+ cytotoxic T cell responses are necessary for resolu- dNK-cells may be helpful in the prevention of CMV transmis- tion of active infection [73]. A variety of regulatory cells and sion from mother to fetus during early pregnancy. Another cytokines are believed to protect against untoward tissue injury. factor in the pathogenesis of CMV infection of the CNS is However, IFN-γ and a host of innate responses also enhance the that CMV initiates a robust inflammatory response character- transplacental transmission of T. gondii via the upregulation of ized by the recruitment of activated resident and peripheral the intracellular adhesion molecule ICAM-1. T. gondii infected immune cells as well as expression of a large number of pro- macrophages and dendritic cells may infect trophoblasts, inflammatory cytokines and IFN-regulated genes [183], followed by congenital transmission of the parasites. IFN- γ which facilitates the destruction of CNS cells and tissues. seems to have an important role in the recruitment of T-cells into the brain after congenital infection. T-cell hyporesponsive- ness to T. gondii antigens has been observed in congenitally Rubella virus infected infants, followed by resolution later in childhood [186] as is the case with congenital CMV. Up-regulated inflam- The experience with rubella virus epidemiology, clinical dis- matory responses also amplify the chronic tissue destruction of ease, and the vaccine suggests that antibodies to this virus are susceptible tissues. It is hypothesized that the reactivation of effective in terminating acute infection and preventing future congenital T. gondii infection is under the control of Th1 T- maternal, and thus congenital, infection. The role of cell- cells and that new ocular lesions often occur in children 10 or mediated immunity in these regards remains less clear, in part more years of age, perhaps as toxoplasma-specific T-cell immu- due to the virtual elimination of rubella disease and congenital nity wanes [73]. As is the case with CMV, reinfection or reac- infection in the developed world. Cell-mediated immunity is tivation of T. gondii infection can give rise to a severe congen- thought to be generally critical to the elimination of infections itally infected infant in an otherwise normal mother, albeit an by intracellular pathogens. Examination of pathologic tissue apparently rare event [187]. from infants with symptomatic congenital rubella demon- strates direct cytopathic effects which are proposed to trigger apoptosis and inhibition of fetal cellular mitosis [184]. In con- Discussion genital rubella virus infection, this virus persists in the fetus. There is evidence of chronic infection with ongoing tissue Our hypothesis involves five concepts: effects, and the infected infant continues to shed virus for months to years [16]. Accompanying the viral shedding, there 1. Normal pregnancy leads to a generalized down- is an unusual persistence of fetal and neonatal synthesis of regulation of T-cell-mediated immunity to help preserve IgM class antibodies against rubella virus, for months. Other the fetal graft. studies have suggested that humoral immune function in con- 2. Multiple, less understood, factors (including primary genital infection is not completely normal and may be delayed maternal infection and a more profound immune suppres- in the infected fetus [16]. A number of studies have demon- sion than seen in the normal pregnant host) allow for strated a variety of defects in cell-mediated immunity in in- systemic maternal infection with viremia/parasitemia. fants with congenital rubella [16]. The validity of generalizing This, in turn, results in the delivery of a high pathogen these data from infants to fetal immune responses is unclear. In load to the placenta. addition, one study has documented decreased rubella virus- 3. Failure of maternal-fetal placental barriers allows ma- specific cell-mediated immunity during pregnancy [75]. Late ternal infections to cross the placenta and disseminate to stage auto- or hyper-immune effects in congenital rubella vi- susceptible developing organs in the fetus. rus infection may explain the delayed manifestations such as 4. The time of maternal infection (early gestation is when diabetes and thyroid disease. important embryonic events are occurring), and cellular tropism, together, determine the characteristics and de- gree of fetal tissue/organ injury, including the CNS. The Eur J Clin Microbiol Infect Dis

pathology of microcephaly is similar regardless of the MMWR Morb Mortal Wkly Rep 66:615–621. https://doi.org/10. causal pathogen, but implies maternal infection early in 15585/mmwr.mm6623e1 9. Honein MA, Dawson AL, Petersen EE et al (2017) Birth defects gestation with profound destruction of fetal CNS tissues. among fetuses and infants of US women with evidence of possible 5. The final step in congenital affliction is chronic de- Zika virus infection during pregnancy. JAMA 317:59. https://doi. structive inflammation thought to be the result of down- org/10.1001/jama.2016.19006 regulation of fetal specific immune defenses in the face of 10. Boppana SB, Ross SA, Fowler KB (2013) Congenital cytomega- lovirus infection: clinical outcome. Clin Infect Dis 57:S178–S181. amplified fetal innate immune responses that upregulate https://doi.org/10.1093/cid/cit629 inflammation. 11. Britt WJ (2016) Cytomegalovirus. In: Wilson CB, Nizet V, Maldonado Y et al (eds) Remington and Klein's infectious dis- eases of the fetus and newborn infant, 8th edn. Saunders, Philadelphia, pp 724–781 12. Britt WJ (2017) Congenital human cytomegalovirus infection and Acknowledgments We would like to thank Dr. Francina Towne for her the enigma of maternal immunity. J Virol 91:e02392–e02416. assistance in pre-publication review of a portion of this article. https://doi.org/10.1128/JVI.02392-16 13. Ho M (2008) The history of cytomegalovirus and its diseases. Compliance with ethical standards Med Microbiol Immunol 197:65–73. https://doi.org/10.1007/ s00430-007-0066-x Funding This work received no internal or external funding. 14. Boppana SB, Pass RF, Britt WJ et al (1992) Symptomatic congen- ital cytomegalovirus infection: neonatal morbidity and mortality. Pediatr Infect Dis J 11:93–98 Conflict of interest The authors declare they have no conflict of 15. Miller E, Cradock-Watson JE, Pollock TM (1982) Consequences interest. of confirmed maternal rubella at successive stages of pregnancy. Lancet 2:781–784 Ethical approval The authors certify that the ethical standards required 16. Reef SE, Plotkin SA (2016) Rubella. In: Wilson CB, Nizet V, by the journal have been reviewed and met. Maldonado Y et al (eds) Remington and Klein's infectious dis- eases of the fetus and newborn infant, 8th edn. Saunders, Informed consent Informed consent is deemed by the authors to be not Philadelphia, pp 894–932 applicable to this manuscript. 17. Koppe JG, Rothova A (1989) Congenital toxoplasmosis. A long- term follow-up of 20 years. Int Ophthalmol 13:387–390 18. Gilbert R, Dunn D, Wallon M et al (2001) Ecological comparison of the risks of mother-to-child transmission and clinical manifes- References tations of congenital toxoplasmosis according to prenatal treat- ment protocol. Epidemiol Infect 127:113–120 1. Schuler-Faccini L, Ribeiro EM, Feitosa IML et al (2016) Possible 19. Brasil P, Pereira JP, Moreira ME et al (2016) Zika virus infection in – association between Zika virus infection and microcephaly — pregnant women in Rio de Janeiro. N Engl J Med 375:2321 2334. Brazil, 2015. MMWR Morb Mortal Wkly Rep 65:59–62. https:// https://doi.org/10.1056/NEJMoa1602412 doi.org/10.15585/mmwr.mm6503e2 20. de Oliveira Melo AS, Aguiar RS, Ramos Amorim MM et al (2016) Congenital Zika virus infection — beyond microcephaly. 2. Rubin EJ, Greene MF, Baden LR (2016) Zika virus and micro- – cephaly. N Engl J Med 374:984–985. https://doi.org/10.1056/ JAMA Neurol 73:1407 1416. https://doi.org/10.1001/ NEJMe1601862 jamaneurol.2016.3720 21. de Paula Freitas B, de Oliveira Dias JR, Prazeres J et al (2016) 3. Klase ZA, Khakhina S, Schneider ADB et al (2016) Zika fetal Ocular findings in infants with microcephaly associated with pre- neuropathogenesis: etiology of a viral syndrome. PLoS Negl sumed Zika virus congenital infection in Salvador, Brazil. JAMA Trop Dis 10(8):e0004877. https://doi.org/10.1371/journal.pntd. Ophthalmol 134(5):529–535. https://doi.org/10.1001/ 0004877 jamaophthalmol.2016.0267 4. Leal MC, Muniz LF, Ferreira TSAA et al (2016) Hearing loss in 22. Petersen LR, Jamieson DJ, Powers AM, Honein MA (2016) Zika infants with microcephaly and evidence of congenital Zika virus virus. N Engl J Med 374:1552–1563. https://doi.org/10.1056/ — – infection Brazil, November 2015 May 2016. MMWR Morb NEJMra1602113 – Mortal Wkly Rep 65:917 919. https://doi.org/10.15585/mmwr. 23. Driggers RW, Ho C-Y, Korhonen EM et al (2016) Zika virus mm6534e3 infection with prolonged maternal viremia and fetal brain abnor- 5. Nahmias AJ, Walls KW, Stewart JA et al (1971) The ToRCH malities. N Engl J Med 374:2142–2151. https://doi.org/10.1056/ complex-perinatal infections associated with toxoplasma and ru- NEJMoa1601824 – bella, cytomegol- and herpes simplex viruses. Pediatr Res 5:405 24. Chan JFW, Choi GKY, Yip CCY et al (2016) Zika fever and 406. https://doi.org/10.1203/00006450-197108000-00144 congenital Zika syndrome: an unexpected emerging arboviral dis- 6. Coyne CB, Lazear HM (2016) Zika virus — reigniting the ease. J Inf Secur 72:507–524. https://doi.org/10.1016/j.jinf.2016. TORCH. Nat Rev Microbiol 14:707–715. https://doi.org/10. 02.011 1038/nrmicro.2016.125 25. de Oliveira-Szejnfeld PS, Levine D, de Oliveira Melo AS et al 7. Alvarado MG, Schwartz DA (2017) Zika virus infection in preg- (2016) Congenital brain abnormalities and Zika virus: what the nancy, microcephaly, and maternal and fetal health: what we think, radiologist can expect to see prenatally and postnatally. Radiology what we know, and what we think we know. Arch Pathol Lab Med 281:203–218. https://doi.org/10.1148/radiol.2016161584 141:26–32. https://doi.org/10.5858/arpa.2016-0382-RA 26. de Miranda HA, Cavalcante Costa M, Monteiro Frazão MA et al 8. Shapiro-Mendoza CK, Rice ME, Galang RR et al (2017) (2016) Expanded spectrum of congenital ocular findings in micro- Pregnancy outcomes after maternal Zika virus infection during cephaly with presumed Zika infection. Ophthalmology 123:1788– pregnancy — U.S. territories, January 1, 2016–April 25, 2017. 1794. https://doi.org/10.1016/j.ophtha.2016.05.001 Eur J Clin Microbiol Infect Dis

27. de Araújo TVB, Rodrigues LC, de Alencar Ximenes RA et al 48. Karwowski MP, Nelson JM, Staples JE et al (2016) Zika virus (2016) Association between Zika virus infection and microceph- disease: a CDC update for pediatric health care providers. aly in Brazil, January to May, 2016: preliminary report of a case– Pediatrics 137:1–13 control study. Lancet Infect Dis 16:1356–1363. https://doi.org/10. 49. D’Ortenzio E, Matheron S, de Lamballerie X et al (2016) 1016/S1473-3099(16)30318-8 Evidence of sexual transmission of Zika virus. N Engl J Med 28. Stokowski LA, Dobyns WB (2016) More than microcephaly: 374:2195–2198. https://doi.org/10.1056/NEJMc1604449 congenital Zika syndrome. Medscape. http://www.medscape. 50. Oster AM, Brooks JT, Stryker JE et al (2016) Interim guidelines com/viewarticle/868966 for prevention of sexual transmission of Zika virus — United 29. Mets MB, Chhabra MS (2008) Eye manifestations of intrauterine States, 2016. MMWR Morb Mortal Wkly Rep 65:120–121. infections and their impact on childhood blindness. Surv https://doi.org/10.15585/mmwr.mm6505e1 Ophthalmol 53:95–111. https://doi.org/10.1016/j.survophthal. 51. Mlakar J, Korva M, Tul N et al (2016) Zika virus associated with 2007.12.003 microcephaly. N Engl J Med 374:951–958. https://doi.org/10. 30. Bale JF (2009) Fetal infections and brain development. Clin 1056/NEJMoa1600651 Perinatol 36:639–653. https://doi.org/10.1016/j.clp.2009.06.005 52. Dirlikov E, Major CG, Mayshack M et al (2016) Guillain-Barré 31. Bale JF (2011) Viral infections of the nervous system. In: syndrome during ongoing Zika virus transmission — Puerto Rico, Swaiman KF, Ashwal S, Ferriero DM, Schor NF (eds) January 1–July 31, 2016. MMWR Morb Mortal Wkly Rep 65: Swaiman's pediatric neurology: principles and practice, 5th edn. 910–914. https://doi.org/10.15585/mmwr.mm6534e1 – Elsevier Saunders, Philadelphia, pp 1262 1290 53. Parra B, Lizarazo J, Jiménez-Arango JA et al (2016) Guillain– 32. Lo C-P, Chen C-Y (2008) Neuroimaging of viral infections in Barré syndrome associated with Zika virus infection in – infants and young children. Neuroimaging Clin N Am 18:119 Colombia. N Engl J Med 375:1513–1523. https://doi.org/10. 132. https://doi.org/10.1016/j.nic.2007.12.002 1056/NEJMoa1605564 33. Desmond MM (1969) Congenital rubella encephalitis. Am J Dis 54. Oliveira Melo AS, Malinger G, Ximenes R et al (2016) Zika virus Child 118:30. https://doi.org/10.1001/archpedi.1969. intrauterine infection causes fetal brain abnormality and micro- 02100040032005 cephaly: tip of the iceberg? Ultrasound Obstet Gynecol 47:6–7. 34. St John MA, Benjamin S (2000) An epidemic of congenital ru- https://doi.org/10.1002/uog.15831 – bella in Barbados. Ann Trop Paediatr 20:231 235 55. Rasmussen SA, Jamieson DJ, Honein MA, Petersen LR (2016) 35. McAuley JB (2014) Congenital toxoplasmosis. J Pediatric Infect Zika virus and birth defects — reviewing the evidence for causal- – Dis Soc 3:S30 S35. https://doi.org/10.1093/jpids/piu077 ity. N Engl J Med 374:1981–1987. https://doi.org/10.1056/ 36. Frenkel JK (1974) Pathology and pathogenesis of congenital toxo- NEJMsr1604338 plasmosis. Bull N Y Acad Med 50:182–191 56. Costello A, Dua T, Duran P et al (2016) Defining the syndrome 37. Wilson CB, Remington JS, Stagno S, Reynolds DW (1981) associated with congenital Zika virus infection. Bull World Health Development of adverse sequelae in children born with subclini- Organ 94:406–406A. https://doi.org/10.2471/BLT.16.176990 cal congenital toxoplasma infection. Obstet Gynecol Surv 36: 57. van der Linden V,Pessoa A, Dobyns Wet al (2016) Description of 436–437 13 infants born during October 2015–January 2016 with congen- 38. Adams Waldorf KM, McAdams RM (2013) Influence of infection ital Zika virus infection without microcephaly at birth — Brazil. during pregnancy on fetal development. Reproduction 146:R151– MMWR Morb Mortal Wkly Rep 65:1343–1348. https://doi.org/ R162. https://doi.org/10.1530/REP-13-0232 10.15585/mmwr.mm6547e2 39. Pass RF, Fowler KB, Boppana SB et al (2006) Congenital cyto- 58. Soares de Souza A, Moraes Dias C, Braga FDCB et al (2016) megalovirus infection following first trimester maternal infection: Fetal infection by Zika virus in the third trimester: report of 2 symptoms at birth and outcome. J Clin Virol 35:216–220. https:// cases. Clin Infect Dis 63:1622–1625 doi.org/10.1016/j.jcv.2005.09.015 40. Spector DH (2014) IUGR and congenital cytomegalovirus infec- 59. Oliveira DBL, Almeida FJ, Durigon EL et al (2016) Prolonged – shedding of Zika virus associated with congenital infection. N tion. J Infect Dis 209:1497 1499. https://doi.org/10.1093/infdis/ – jiu092 Engl J Med 375:1202 1204 41. Gregg NM (1941) Congenital cataract following German measles 60. La Rosa C, Diamond DJ (2012) The immune response to human – in the mother. Trans Ophthalmol Soc Aust 3:35–46 CMV. Future Virol 7:279 293. https://doi.org/10.2217/fvl.12.8 42. Yepez JB, Murati FA, Pettito M et al (2017) Ophthalmic manifes- 61. Ludlow M, Kortekaas J, Herden C et al (2016) Neurotropic virus tations of congenital Zika syndrome in Colombia and Venezuela. infections as the cause of immediate and delayed neuropathology. – JAMA Ophthalmol 135:440. https://doi.org/10.1001/ Acta Neuropathol 131:159 184. https://doi.org/10.1007/s00401- jamaophthalmol.2017.0561 015-1511-3 43. Frenkel LD, Keys MP, Hefferen SJ et al (1980) Unusual eye ab- 62. Numazakinamio Y, Yano N, Morizuka T et al (1970) Primary normalities associated with congenital cytomegalovirus infection. infection with human cytomegalovirus: virus isolation from Pediatrics 66:763–766 healthy infants and pregnant women. Am J Epidemiol 91:410– 44. Reynolds MR, Jones AM, Petersen EE et al (2017) Vital signs: 417. https://doi.org/10.1093/oxfordjournals.aje.a121151 update on Zika virus–associated birth defects and evaluation of all 63. Carlson A, Norwitz ER, Stiller RJ (2010) Cytomegalovirus infec- U.S. infants with congenital Zika virus exposure — U.S. Zika tion in pregnancy: should all women be screened? Rev Obstet Pregnancy Registry, 2016. MMWR Morb Mortal Wkly Rep 66: Gynecol 3:172–179. https://doi.org/10.3909/riog0131 366–373. https://doi.org/10.15585/mmwr.mm6613e1 64. Boppana SB, Rivera LB, Fowler KB et al (2001) Intrauterine 45. Calvet GA, dos Santos FB, Sequeira PC (2016) Zika virus infec- transmission of cytomegalovirus to infants of women with tion: epidemiology, clinical manifestations, and diagnosis. Curr preconceptional immunity. N Engl J Med 344:1366–1371 Opin Infect Dis 29:459–466 65. Dudgeon JA (1969) Congenital rubella. Am J Dis Child 118:35. 46. Duffy MR, Chen T-H, Hancock WT et al (2009) Zika virus out- https://doi.org/10.1001/archpedi.1969.02100040037007 break on Yap Island, Federated States of Micronesia. N Engl J 66. Kylat RI, Kelly EN, Ford-Jones EL (2006) Clinical findings and Med 360:2536–2543 adverse outcome in neonates with symptomatic congenital cyto- 47. Vogel G (2016) Top mosquito suspect found infected with Zika. megalovirus (SCCMV) infection. Eur J Pediatr 165:773–778. Science. https://doi.org/10.1126/science.aaf5743 https://doi.org/10.1007/s00431-006-0172-6 Eur J Clin Microbiol Infect Dis

67. Boppana SB, Fowler KB, Britt WJ et al (1999) Symptomatic 85. de Bernardi Schneider A, Malone RW, Guo J-TT et al (2017) congenital cytomegalovirus infection in infants born to mothers Molecular evolution of Zika virus as it crossed the Pacific to the with preexisting immunity to cytomegalovirus. Pediatrics 104:55– Americas. Cladistics 33:1–20. https://doi.org/10.1111/cla.12178 60 86. Adiga R (2016) Phylogenetic analysis of the NS5 gene of Zika 68. Anderson KS, Amos CS, Boppana S, Pass R (1996) Ocular ab- virus. J Med Virol 88:1821–1826. https://doi.org/10.1002/jmv. normalities in congenital cytomegalovirus infection. J Am Optom 24615 Assoc 67:273–278 87. Wang A, Thurmond S, Islas L et al (2017) Zika virus genome 69. Yamamoto AY, Mussi-Pinhata MM, Isaac ML et al (2011) biology and molecular pathogenesis. Emerg Microbes Infect 6: Congenital cytomegalovirus infection as a cause of sensorineural e13. https://doi.org/10.1038/emi.2016.141 hearing loss in a highly immune population. Pediatr Infect Dis J 88. Ross SA, Novak Z, Pati S et al (2011) Mixed infection and strain 30:1043–1046 diversity in congenital cytomegalovirus infection. J Infect Dis 70. Burny W, Liesnard C, Donner C, Marchant A (2004) 204:1003–1007. https://doi.org/10.1093/infdis/jir457 Epidemiology, pathogenesis and prevention of congenital cyto- 89. McDonagh S, Maidji E, Ma W et al (2004) Viral and bacterial megalovirus infection. Expert Rev Anti-Infect Ther 2:881–894 pathogens at the maternal-fetal interface. J Infect Dis 190:826– 71. Dudgeon JA (1967) Maternal rubella and its effect on the foetus. 834. https://doi.org/10.1086/422330 Arch Dis Child 42:110–125. https://doi.org/10.1136/adc.42.222. 90. Abernathy E, Chen M, Bera J et al (2013) Analysis of whole 110 genome sequences of 16 strains of rubella virus from the United 72. Frenkel LD, Bellanti JA (1982) Immunology of measles, mumps States, 1961–2009. Virol J 10:32. https://doi.org/10.1186/1743- and rubella viruses. In: Nahmias AJ, O’Reilly RJ (eds) 422X-10-32 Immunology of human infection (comprehensive immunology) 91. Ueda K (2016) Epidemiology of rubella and congenital rubella part 2. Springer, New York, pp 135–136 syndrome in Japan before 1989. Vaccine 34:1971–1974. https:// 73. Peyron F, Wallon M, Kieffer F, Garweg J (2016) Toxoplasmosis. doi.org/10.1016/j.vaccine.2015.10.010 In: Wilson CB, Nizet V, Maldonado Y et al (eds) Remington and 92. Martinez-Torres AO, Mosquera MM, Sanz JC et al (2009) Klein's infectious diseases of the fetus and newborn infant, 8th Phylogenetic analysis of rubella virus strains from an outbreak edn. Saunders, Philadelphia, pp 949–1042 in Madrid, Spain, from 2004 to 2005. J Clin Microbiol 47:158– 74. Alford CA, Neva FA, Weller TH et al (1964) Virologic and sero- 163. https://doi.org/10.1128/JCM.00469-08 logic studies on human products of conception after maternal ru- 93. Ajzenberg D, Cogné N, Paris L et al (2002) Genotype of 86 bella. N Engl J Med 271:1275–1281. https://doi.org/10.1056/ Toxoplasma gondii isolates associated with human congenital NEJM196412172712501 toxoplasmosis, and correlation with clinical findings. J Infect 75. Thong YH, Steele RW, Vincent MM et al (1973) Impaired in vitro Dis 186:684–689. https://doi.org/10.1086/342663 cell-mediated immunity to rubella virus during pregnancy. N Engl 94. Elbez-Rubinstein A, Ajzenberg D, Dardé M et al (2009) J Med 289:604–606 Congenital toxoplasmosis and reinfection during pregnancy: case 76. Rola-Pleszczynski M, Frenkel LD, Fuccillo DA et al (1977) report, strain characterization, experimental model of reinfection, Specific impairment of cell-mediated immunity in mothers of in- and review. J Infect Dis 199:280–285. https://doi.org/10.1086/ fants with congenital infection due to cytomegalovirus. J Infect 595793 Dis 135:386–391 95. Xiao J, Yolken RH (2015) Strain hypothesis of Toxoplasma gondii 77. Papania MJ, Wallace GS, Rota PA et al (2014) Elimination of infection on the outcome of human diseases. Acta Physiol 213: endemic measles, rubella, and congenital rubella syndrome from 828–845. https://doi.org/10.1111/apha.12458 the Western hemisphere: the US experience. JAMA Pediatr 168: 96. Von der Hagen M, Pivarcsi M, Liebe J et al (2014) Diagnostic 148–155. https://doi.org/10.1001/jamapediatrics.2013.4342 approach to microcephaly in childhood: a two-center study and 78. Robert-Gangneux F, Darde M-L (2012) Epidemiology of and di- review of the literature. Dev Med Child Neurol 56:732–741. agnostic strategies for toxoplasmosis. Clin Microbiol Rev 25:264– https://doi.org/10.1111/dmcn.12425 296. https://doi.org/10.1128/CMR.05013-11 97. Faheem M, Naseer MI, Rasool M et al (2015) Molecular genetics 79. Lindsay DS, Dubey JP (2011) Toxoplasma gondii:thechanging of human primary microcephaly: an overview. BMC Med Genet paradigm of congenital toxoplasmosis. Parasitology 138:1829– 8:S4. https://doi.org/10.1186/1755-8794-8-S1-S4 1831. https://doi.org/10.1017/S0031182011001478 98. McGavern DB, Kang SS (2011) Illuminating viral infections in 80. Weiss LM, Dubey JP (2009) Toxoplasmosis: a history of clinical the nervous system. Nat Rev Immunol 11:318–329. https://doi. observations. Int J Parasitol 39:895–901. https://doi.org/10.1016/j. org/10.1038/nri2971 ijpara.2009.02.004 99. Elsheikha HM, Khan NA (2010) Protozoa traversal of the blood– 81. Robbins JR, Zeldovich VB, Poukchanski A et al (2012) Tissue brain barrier to invade the central nervous system. FEMS barriers of the human placenta to infection with Toxoplasma Microbiol Rev 34:532–553. https://doi.org/10.1111/j.1574-6976. gondii. Infect Immun 80:418–428. https://doi.org/10.1128/IAI. 2010.00215.x 05899-11 100. Vacca P, Cantoni C, Vitale M et al (2010) Crosstalk between de- 82. Jones JL, Holland GN (2010) Annual burden of ocular toxoplas- cidual NK and CD14+ myelomonocytic cells results in induction mosis in the United States. Am J Trop Med Hyg 82:464–465. of Tregs and immunosuppression. Proc Natl Acad Sci 107:11918– https://doi.org/10.4269/ajtmh.2010.09-0664 11923 83. McLeod R, Boyer K, Karrison T et al (2006) Outcome of treat- 101. Veenstra van Nieuwenhoven AL, Moes H, Jan Heineman M et al ment for congenital toxoplasmosis, 1981-2004: the national col- (2008) Cytokine production by monocytes, NK cells, and lympho- laborative Chicago-based, congenital toxoplasmosis study. Clin cytes is different in preeclamptic patients as compared with normal Infect Dis 42:1383–1394. https://doi.org/10.1086/501360 pregnant women. Hypertens Pregnancy 27:207–224 84. Haddow AD, Schuh AJ, Yasuda CY et al (2012) Genetic charac- 102. Wilczyński JR, Tchórzewski H, Banasik M et al (2003) terization of Zika virus strains: geographic expansion of the Asian Lymphocyte subset distribution and cytokine secretion in third lineage. PLoS Negl Trop Dis 6:e1477. https://doi.org/10.1371/ trimester decidua in normal pregnancy and preeclampsia. Eur J journal.pntd.0001477 Obstet Gynecol Reprod Biol 109:8–15 Eur J Clin Microbiol Infect Dis

103. Liu H-Y, Liu Z-K, Chao H et al (2014) High-dose interferon-γ 124. Cheeran MCJ, Lokensgard JR, Schleiss MR (2009) promotes abortion in mice by suppressing Treg and Th17 polari- Neuropathogenesis of congenital cytomegalovirus infection: dis- zation. J Interf Cytokine Res 34:394–403 ease mechanisms and prospects for intervention. Clin Microbiol 104. Gardner L, Moffett A (2003) Dendritic cells in the human decidua. Rev 22:99–126. https://doi.org/10.1128/CMR.00023-08 Biol Reprod 69:1438–1446. https://doi.org/10.1095/biolreprod. 125. Strafela P, Vizjak A, Mraz J et al (2017) Zika virus-associated 103.017574 micrencephaly: a thorough description of neuropathologic find- 105. Williams PJ, Searle RF, Robson SC et al (2009) Decidual leuco- ings in the fetal central nervous system. Arch Pathol Lab Med cyte populations in early to late gestation normal human pregnan- 141:73–81. https://doi.org/10.5858/arpa.2016-0341-SA cy. J Reprod Immunol 82:24–31 126. Bale JF (1984) Human cytomegalovirus infection and disorders of 106. Faas MM, Spaans F, De Vos P (2014) Monocytes and macro- the nervous system. Arch Neurol 41:310–320 phages in pregnancy and pre-eclampsia. Front Immunol 5:298. 127. Elliott GB, Elliott KA (1962) Observations on cerebral cytome- https://doi.org/10.3389/fimmu.2014.00298 galic inclusion disease of the foetus and newborn. Arch Dis Child 107. Abrahams VM, Kim YM, Straszewski SL et al (2004) 37:34–39 Macrophages and apoptotic cell clearance during pregnancy. 128. Lago EG, Baldisserotto M, Hoefel Filho JR et al (2007) Am J Reprod Immunol 51:275–282 Agreement between ultrasonography and computed tomography 108. Cupurdija K, Azzola D, Hainz U et al (2004) Macrophages of in detecting intracranial calcifications in congenital toxoplasmo- human first trimester decidua express markers associated to alter- sis. Clin Radiol 62:1004–1011. https://doi.org/10.1016/j.crad. native activation. Am J Reprod Immunol 51:117–122 2007.05.001 109. Houser BL, Tilburgs T, Hill J et al (2011) Two unique human 129. Yuan TM, Sun Y, Zhan CY, Yu HM (2010) Intrauterine infection/ decidual macrophage populations. J Immunol 186:2633–2642 inflammation and perinatal brain damage: role of glial cells and 110. Heikkinen J, Mottonen M, Komi J et al (2003) Phenotypic char- toll-like receptor signaling. J Neuroimmunol 229:16–25. https:// acterization of human decidual macrophages. Clin Exp Immunol doi.org/10.1016/j.jneuroim.2010.08.008 131:498–505 130. Kumar V, Abbas AK, Aster JC (2015) Robbins & Cotran patho- 111. Ernerudh J, Berg G, Mjösberg J (2011) Regulatory T helper cells logic basis of disease. Elsevier Saunders, Philadelphia, pp 186– in pregnancy and their roles in systemic versus local immune tol- 200 erance. Am J Reprod Immunol 66:31–43 131. Crack PJ, Bray PJ (2007) Toll-like receptors in the brain and their 112. Erlebacher A (2013) Immunology of the maternal-fetal interface. potential roles in neuropathology. Immunol Cell Biol 85:476–480. Annu Rev Immunol 31:387–411 https://doi.org/10.1038/sj.icb.7100103 113. Nancy P, Erlebacher A (2014) T cell behavior at the maternal-fetal 132. Salaun B, Romero P, Lebecque S (2007) Toll-like receptor’stwo- interface. Int J Dev Biol 58:189–198 edged sword: when immunity meets apoptosis. Eur J Immunol 37: 114. Teles A, Zenclussen AC, Schumacher A (2013) Regulatory T cells 3311–3318. https://doi.org/10.1002/eji.200737744 are baby′s best friends. Am J Reprod Immunol 69:331–339 133. Tang H, Hammack C, Ogden SC et al (2016) Zika virus infects 115. Tilburgs T, Roelen DL, van der Mast BJ et al (2008) Evidence for a human cortical neural progenitors and attenuates their growth. Cell selective migration of fetus-specific CD4+CD25bright regulatory Stem Cell 18:587–590. https://doi.org/10.1016/j.stem.2016.02. T cells from the peripheral blood to the decidua in human preg- 016 nancy. J Immunol 180:5737–5745 134. Lancaster MA, Renner M, Martin C-A et al (2012) Cerebral 116. Mjosberg J, Svensson J, Johansson E et al (2009) Systemic reduc- organoids model human brain development and microcephaly. tion of functionally suppressive CD4dimCD25highFoxp3+ Tregs Nature 501:373–379. https://doi.org/10.1038/nature12517 in human second trimester pregnancy is induced by progesterone 135. Dang J, Tiwari SK, Lichinchi G et al (2016) Zika virus depletes and 17-estradiol. J Immunol 183:759–769 neural progenitors in human cerebral organoids through activation 117. Darmochwal-Kolarz D, Kludka-Sternik M, Tabarkiewicz J et al of the innate immune receptor TLR3. Cell Stem Cell 19:258–265. (2012) The predominance of Th17 lymphocytes and decreased https://doi.org/10.1016/j.stem.2016.04.014 number and function of Treg cells in preeclampsia. J Reprod 136. Li G, Poulsen M, Fenyvuesvolgyi C et al (2017) Characterization Immunol 93:75–81 of cytopathic factors through genome-wide analysis of the Zika 118. Tilburgs T, Claas FHJ, Scherjon SA (2010) Elsevier trophoblast viral proteins in fission yeast. Proc Natl Acad Sci 114:E376–E385. research award lecture: unique properties of decidual T cells and https://doi.org/10.1073/pnas.1619735114 their role in immune regulation during human pregnancy. Placenta 137. Xagorari A, Chlichlia K (2008) Toll-like receptors and viruses: 31:S82–S86 induction of innate antiviral immune responses. Open Microbiol 119. Amodio G, Mugione A, Sanchez AM et al (2013) HLA-G ex- J2:49–59. https://doi.org/10.2174/1874285800802010049 pressing DC-10 and CD4+ T cells accumulate in human decidua 138. Gonzalez RMS, Shehata H, O’Connell MJ et al (2014) during pregnancy. Hum Immunol 74:406–411 Toxoplasma gondii-derived profilin triggers human toll-like re- 120. van Egmond A, van der Keur C, Swings GMJ et al (2016) The ceptor 5-dependent cytokine production. J Innate Immun 6:685– possible role of virus-specific CD8+ memory T cells in decidual 694. https://doi.org/10.1159/000362367 tissue. J Reprod Immunol 113:1–8. https://doi.org/10.1016/j.jri. 139. Yarovinsky F (2014) Innate immunity to Toxoplasma gondii in- 2015.09.073 fection. Nat Rev Immunol 14:109–121. https://doi.org/10.1038/ 121. van Lier RAW, ten Berge IJM, Gamadia LE (2003) Human CD8+ nri3598 T-cell differentiation in response to viruses. Nat Rev Immunol 3: 140. Medawar PB (1953) Some immunological and endocrinological 931–939 problems raised by evolution of viviparity invertebrates. Symp 122. Ritter JM, Martines RB, Zaki SR (2017) Zika virus: pathology Soc Exp Biol 7:3208–32 from the pandemic. Arch Pathol Lab Med 141:49–59. https:// 141. Adams KM, Nelson JL (2006) Bi-directional cell trafficking dur- doi.org/10.5858/arpa.2016-0397-SA ing pregnancy: long-term consequences for human health. In: Mor 123. Panchaud A, Stojanov M, Ammerdorffer A et al (2016) Emerging G (ed) immunology of pregnancy. Eurekah.com and Springer role of Zika virus in adverse fetal and neonatal outcomes. Clin Science + Media, Georgetown, pp 244–252 Microbiol Rev 29:659 –694. https://doi.org/10.1128/CMR.00014- 142. Geiselhart A, Dietl J, Marzusch K et al (1995) Comparative anal- 16 ysis of the immunophenotypes of decidual and peripheral blood Eur J Clin Microbiol Infect Dis

large granular lymphocytes and T cells during early human preg- 163. Prendergast AJ, Klenerman P, Goulder PJR (2012) The impact of nancy. Am J Reprod Immunol 33:315–322 differential antiviral immunity in children and adults. Nat Rev 143. Nelson JL (2002) Microchimerism: incidental byproduct of preg- Immunol 12:636–648. https://doi.org/10.1038/nri3277 nancy or active participant in human health? Trends Mol Med 8: 164. Quicke KM, Bowen JR, Johnson EL et al (2016) Zika virus infects 109–113 human placental macrophages. Cell Host Microbe 20:83–90. 144. Erlebacher A, Vencato D, Price KA et al (2007) Constraints in https://doi.org/10.1016/j.chom.2016.05.015 antigen presentation severely restrict T cell recognition of the al- 165. Hamel R, Dejarnac O, Wichit S et al (2015) Biology of Zika virus logeneic fetus. J Clin Invest 117:1399–1411 infection in human skin cells. J Virol 89:8880–8896. https://doi. 145. Harris JW (1919) Influenza occurring in pregnant women. J Am org/10.1128/jvi.00354-15 Med Assoc 72:978–980 166. Bayer A, Lennemann NJ, Ouyang Y et al (2016) Type III inter- 146. Siston AM (2010) Pandemic 2009 influenza a(H1N1) virus illness ferons produced by human placental trophoblasts confer protec- among pregnant women in the United States. J Am Med Assoc tion against Zika virus infection. Cell Host Microbe 19:705–712. 303:1517–1525 https://doi.org/10.1016/j.chom.2016.03.008 147. Sappenfield E, Jamieson DJ, Kourtis AP (2013) Pregnancy and 167. Olagnier D, Amatore D, Castiello L et al (2016) Dengue virus susceptibility to infectious diseases. Infect Dis Obstet Gynecol immunopathogenesis: lessons applicable to the emergence of – 2013:1–8 Zika virus. J Mol Biol 428:3429 3448. https://doi.org/10.1016/j. 148. Jamieson D, Theiler R, Rasmussen S (2006) Emerging infections jmb.2016.04.024 and pregnancy. Emerg Infect Dis 12:1638–1643 168. Quicke K, Suthar M (2013) The innate immune playbook for – 149. Robinson DP, Klein SL (2012) Pregnancy and pregnancy- restricting West Nile virus infection. Viruses 5:2643 2658. associated hormones alter immune responses and disease patho- https://doi.org/10.3390/v5112643 genesis. Horm Behav 62:263–271 169. Xie X, Shan C, Shi P-Y (2016) Restriction of Zika virus by host – 150. Elenkov IJ (2001) IL-12, TNF-alpha, and hormonal changes dur- innate immunity. Cell Host Microbe 19:566 567. https://doi.org/ ing late pregnancy and early postpartum: implications for autoim- 10.1016/j.chom.2016.04.019 mune disease activity during these times. J Clin Endocrinol Metab 170. Grant A, Ponia SS, Tripathi S et al (2016) Zika virus targets human 86:4933–4938 STAT2 to inhibit type I interferon signaling. Cell Host Microbe 19(6):882–890. https://doi.org/10.1016/j.chom.2016.05.009 151. Runmarker B, Andersen O (1995) Pregnancy is associated with a 171. Stettler K, Beltramello M, Espinosa DA et al (2016) Specificity, lower risk of onset and a better prognosis in multiple sclerosis. cross-reactivity, and function of antibodies elicited by Zika virus Brain 118:253–261 infection. Science 353:823–826. https://doi.org/10.1126/science. 152. Szekeres-Bartho J (2002) Immunological relationship between the aaf8505 mother and the fetus. Int Rev Immunol 21:471–495 172. Weisblum Y, Oiknine-Djian E, Vorontsov OM et al (2016) Zika 153. Sabahi F, Rola-Plesczcynski M, O’Connell S, Frenkel LD (1995) virus infects early- and midgestation human maternal decidual Qualitative and quantitative analysis of T lymphocytes during nor- – tissues, inducing distinct innate tissue responses in the maternal- mal human pregnancy. Am J Reprod Immunol 33:381 393 – ’ fetal interface. J Virol 91:e01905 e01916. https://doi.org/10.1128/ 154. Sabahi F, Frenkel LD, O Connell S (1993) Depression of cell- jvi.01905-16 mediated immunity in pregnancy. J Immunol 150(8):A101 173. Cauda R, Prasthofer EF, Grossi CE et al (1987) Congenital cyto- 155. Wegmann TG, Lin H, Guilbert L, Mosmann TR (1993) megalovirus: Immunological alterations. J Med Virol 23:41–49. Bidirectional cytokine interactions in the maternal-fetal relation- https://doi.org/10.1002/jmv.1890230106 ship: is successful pregnancy a Th2 phenomenon? Immunol 174. Stagno S, Pass RF, Dworsky ME, Alford CA (1982) Maternal – Today 14:353 356 cytomegalovirus infection and perinatal transmission. Clin 156. Veenstra van Nieuwenhoven AL, Bouman A, Moes H et al (2002) Obstet Gynecol 25:563–576 Cytokine production in natural killer cells and lymphocytes in 175. Sylwester AW, Mitchell BL, Edgar JB et al (2005) Broadly pregnant women compared with women in the follicular phase targeted human cytomegalovirus-specific CD4+and CD8+T cells – of the ovarian cycle. Fertil Steril 77:1032 1037 dominate the memory compartments of exposed subjects. J Exp 157. Saito S, Sakai M, Sasaki Y et al (1999) Quantitative analysis of Med 202:673–685. https://doi.org/10.1084/jem.20050882 peripheral blood Th0, Th1, Th2 and the Th1:Th2 cell ratio during 176. Starr SE, Tolpin MD, Friedman HM et al (1979) Impaired cellular normal human pregnancy and preeclampsia. Clin Exp Immunol immunity to cytomegalovirus in congenitally infected children – 117:550 555 and their mothers. J Infect Dis 140:500–505. https://doi.org/10. 158. Borzychowski AM, Croy BA, Chan WL et al (2005) Changes in 1093/infdis/140.4.500 systemic type 1 and type 2 immunity in normal pregnancy and pre- 177. Revello MG, Lilleri D, Zavattoni M et al (2006) eclampsia may be mediated by natural killer cells. Eur J Immunol Lymphoproliferative response in primary human cytomegalovi- 35:3054–3063 rus (HCMV) infection is delayed in HCMV transmitter mothers. 159. Trundley A, Moffett A (2004) Human uterine leukocytes and J Infect Dis 193:269–276. https://doi.org/10.1086/498872 pregnancy. Tissue Antigens 63:1–12 178. Lilleri D, Fornara C, Furione M et al (2007) Development of 160. Vega-Sanchez R, Gomez-Lopez N, Flores-Pliego A et al (2010) human cytomegalovirus–specific T cell immunity during primary Placental blood leukocytes are functional and phenotypically dif- infection of pregnant women and its correlation with virus trans- ferent than peripheral leukocytes during human labor. J Reprod mission to the fetus. J Infect Dis 195:1062–1070. https://doi.org/ Immunol 84:100–110 10.1086/512245 161. Haynes BF (1995) Early human T cell development: analysis of 179. Fornara C, Lilleri D, Revello MG et al (2011) Kinetics of effector the human thymus at the time of initial entry of hematopoietic functions and phenotype of virus-specific and γδ T lymphocytes stem cells into the fetal thymic microenvironment. J Exp Med in primary human cytomegalovirus infection during pregnancy. J 181:1445–1458. https://doi.org/10.1084/jem.181.4.1445 Clin Immunol 31:1054 –1064. https://doi.org/10.1007/s10875- 162. Mold JE, Venkatasubrahmanyam S, Burt TD et al (2010) Fetal and 011-9577-8 adult hematopoietic stem cells give rise to distinct T cell lineages 180. Lissauer D, Choudhary M, Pachnio A et al (2011) in humans. Science 330:1695–1699. https://doi.org/10.1126/ Cytomegalovirus sero positivity dramatically alters the maternal science.1196509 CD8+ T cell repertoire and leads to the accumulation of highly Eur J Clin Microbiol Infect Dis

differentiated memory cells during human pregnancy. Hum 184. Lee J-Y, Bowden DS (2000) Rubella virus replication and links to Reprod 26:3355–3365. https://doi.org/10.1093/humrep/der327 teratogenicity. Clin Microbiol Rev 13:571–587. https://doi.org/10. 181. Jun Y, Kim E, Jin M et al (2000) Human cytomegalovirus gene 1128/cmr.13.4.571-587.2000 products US3 and US6 down-regulate trophoblast class I MHC 185. Sobanski V, Ajzenberg D, Delhaes L et al (2013) Severe toxoplas- molecules. J Immunol 164:805–811. https://doi.org/10.4049/ mosis in immunocompetent hosts: be aware of atypical strains. jimmunol.164.2.805 Am J Respir Crit Care Med 187:1143–1145. https://doi.org/10. 182. Biron CA, Byron KS, Sullivan JL (1989) Severe herpesvirus infec- 1164/rccm.201209-1635LE tions in an adolescent without natural killer cells. N Engl J Med 320: 186. Guglietta S, Beghetto E, Spadoni A et al (2007) Age-dependent 1731–1735. https://doi.org/10.1056/NEJM198906293202605 impairment of functional helper T cell responses to 183. Slavuljica I, Kveštak D, Csaba Huszthy P et al (2014) immunodominant epitopes of Toxoplasma gondii antigens in con- – Immunobiology of congenital cytomegalovirus infection of the genitally infected individuals. Microbes Infect 9:127 133. https:// central nervous system—the murine cytomegalovirus model. doi.org/10.1016/j.micinf.2006.10.017 Cell Mol Immunol 12:180–191. https://doi.org/10.1038/cmi. 187. Valdès V, Legagneur H, Watrin V et al (2011) Toxoplasmose 2014.51 congénitale secondaire à une réinfection maternelle pendant la grossesse. Arch Pédiatrie 18:761–763. https://doi.org/10.1016/j. arcped.2011.04.011