| Abstract |
Introduction
During the first years of life, children are constantly confronted to a number of infections. Clinically evident infections have been a subject of research for many years, enabling development of successful prevention programs (e.g., vaccination programs). Nonetheless, children subclinically acquire a number of other infections whose role in health and disease has been little examined. It has been suggested that a normal course of infections in early life protects against the development of childhood leukemia (the Greaves hypothesis) and of chronic inflammatory disorders (the hygiene or “old friends” hypothesis). Nonetheless, these hypotheses are based on epidemiological studies which use proxy variables such as age of first school attendance, to denote exposure to common infections. Very limited prospective data exist on characterizing this repertoire of common infections in childhood by use of biological measures.
The primary objective of this study has been the description of the seroepidemiology of Polyomaviruses and Herpesviruses in early childhood. Polyomaviruses are DNA viruses (fourteen known so far), which are largely acquired subclinically in early life. Studies in adults show high seroprevalence rates but the exact age of acquisition is not known. Herpesviruses are also largely acquired in childhood with the exception of HSV-2, which is predominantly acquired through sexual contact later in life. Emerging data suggest that recent changes in family structure, childrearing, and hygiene practices have modified the epidemiology of Herpesviruses. This might have important public health implications as the timing of exposure has a critical role in the outcome of infection with Ηerpesviruses. These facts highlight the importance of studying the seroepidemiology of Polyomaviruses and Herpesviruses in early life in the general population.
The next objective has been to explore the role of Polyomaviruses and Herpesviruses infections in child’s neurodevelopment and growth. The developing brain is extremely sensitive to the neuroinflammation induced by viruses. Among Polyomaviruses, JCPyV is known to be neurotropic and is the etiologic agent of progressive multifocal leukoencephalopathy, a fatal demyelinating disease of the central nervous sytem occurring mainly in immonocompromised individuals. Moreover, Lintas et al. found enhanced frequencies of Polyomaviruses (BKPyV, JCPyV, SV40) in postmortem neocortical tissues of autistic patients, raising questions for an association between Polyomaviruses and autism. EBV, CMV and HSV-1, members of the Herpesviruses family, are also neurotropic viruses and under certain circumstances can infect the central nervous system and cause acute encephalitis/meningitis and long-term neurological sequelae (e.g., sensorineural hearing loss due to congenital CMV infection). The potential association of Polyomaviruses and Herpesviruses with child’s growth, obesity and cardiometabolic traits is less evident but is based on an emerging hypothesis of an infectious origin of obesity. In particular, adenovirus 36 has been associated with obesity both in animal and human studies. Moreover a metabolic dysfunction has been shown to accompany some viral infections such as CMV, HSV-1 and HSV-2. Interestingly, an increasing number of studies report a link between pathogen burden, obesity, cardiometabolic traits and disease, but no evidence exists in childhood.
Lastly, an objective of the present study has been to describe the seroepidemiology of Helicobacter pylori, a very common chronic bacterial infection in humans which is associated with increased risk for gastric cancer. A wealth of evidence strongly implicates Helicobacter pylori infection in the pathogenesis of extragastric diseases such as idiopathic thrombocytopenic purpura, unexplained iron-deficiency anemia and vitamin B12 deficiency. Moreover, an increasing number of studies report an association between Helicobacter pylori and neuropsychiatric diseases such as Parkinon’s and Alzheimer. It is likely that an association with neurodevelopment is evident already in early life, but there are no data. Thus, we hypothesized that Helicobacter pylori infection during the first years of life but also maternal Helicobacter pylori infection during pregnancy may unfavorably affect offspring’s neurodevelopment through micronutrients deficiencies (e.g folate, B12). Fetal brain is extremely vulnerable to such micronutrients defects. To the best of our knowledge, no study has examined differences in neurodevelopment between children of Helicobacter pylori seropositive versus seronegative mothers.
Specific Objectives
• To describe the seroprevalence of Polyomaviruses, Herpesviruses and Helicobacter pylori in early life using repeated samples collected at birth (cord blood), three years and four years of age.
• To identify the factors – sociodemographic, perinatal, variables denoting social interactions, lifestyle, hygiene practices- that determine the acquisition of Polyomaviruses and Herpesviruses up to four years of age.
• To explore the association between seropositivity to single and multiple Polyomaviruses and Herpesviruses up to age four with neurodevelopmental assessment of children at the same age.
• To explore the association between Helicobacter pylori infection early in life including fetal life with neurodevelopment of children at age four.
• To explore the association between seropositivity to single and to multiple Polyomaviruses and Herpesviruses up to age four with obesity and cardiometabolic traits at age four and six.
Methods
This study uses data from the Rhea study a birth cohort that recruited 1,606 pregnant women from February 2007 to January 2008 (www.rhea.gr). The inclusion criteria were confirmed pregnancy, residency within the study area, aged 16 years or older, and good understanding of the Greek language. Recruitment occurred mostly before 15 weeks of gestation. Mothers were contacted again at 24 weeks of gestation, at birth, and for child’s follow-up at 9th month, 18th month, 4 years, and 6 years of age. When members of the cohort were 3 years of age, we invited a randomly selected subsample for a serological assessment (n = 109).
Blood samples collected at birth (cord blood) and at 3 and 4 years of age were processed following standard procedures and then stored at −80°C. Serum aliquots of 100 μL were shipped on dry ice to the German Cancer Research Center, Heidelberg, Germany, for serological analysis. Immunoglobulin G seroreactivity against the viral capsid protein 1 of ten Polyomaviruses (BKPyV, JCPyV, KIPyV,WUPyV,MCPyV, HPyV6, HPyV7, TSPyV, HPyV9, HPyV10), four EBV antigens (ZEBRA, EBNA-1, EA-D, VCA p18), five CMV antigens (pp52, pp65, pp150, pp28, CM2), one HSV-1 antigen (gG), one HSV-2 antigen (mgGunique), four HHV-8 antigens (LANA, v-cyclin, K8.1, ORF-65) and twelve H.pylori proteins (GroEL, UreA, HP0231, NapA, HP0305, HpaA, CagA, HyuA, catalase, VacA, HcpC, Omp) was measured by fluorescent bead-based multiplex serology (1:1,000 dilution). Of note, immunoglobulin G levels in cord blood reflect the mother’s immunoglobulin G levels. Seropositivity for a given antigen was defined as median fluorescence intensity values greater than the antigen- specific cut off. EBV and CMV seropositivity was defined as seropositivity for at least two virus specific antigens. Seropositivity for a given Helicobacter pylori protein was defined as seroreactivity greater than the protein-specific cut-off which was re-evaluated using the published reference panel. The aggregate number of different Polyomaviruses that the children were seropositive to was referred as Polyomaviruses burden. Similarly, we calculated the Herpesviruses burden.
Information on potential determinants of seroprevalence of polyomaviruses and herpesviruses is based on repeated questionnaires from birth to 4 years of age. Children’s cognitive and motor development was assessed by two trained psychologists with the McCarthy Scales of Children’s Abilities (MSCA). Additional information on children’s behavior was obtained via maternal report on standardized child behavior scales, which were administered at the 4 years of age follow-up. The Attention-Deficit/Hyperactivity Disorder Test (ADHDT) is designed to identify and evaluate ADHD in ages 3–23 years. The parent version of the Strengths and Difficulties Questionnaire (SDQ) is a behavioral screening instrument designed to assess strengths and difficulties of children aged 3–16 years. We measured body mass index, waist circumference and skinfold thickness at four anatomical sites at four and six years of age. At age six hand-to-leg bioelectrical impendence was also measured using a Bodystat 1500 machine. Data on metabolic traits including serum lipids, leptin and adiponectin levels were also available.
Descriptive analyses of the study population characteristics, exposures and outcomes were conducted. Multiple linear, logistic and Poisson regression models were used to explore the associations accordingly.
Results
1) Seroprevalence of Polyomaviruses ranged from 38.5% to 99.8% in cord blood and from 20.9% to 82.3% at age 4. Seroprevalence of EBV, CMV, HSV-1, HSV-2 and HHV-8 was 99.4%, 74.9%, 26.2%, 8.0%, and 1.6% in cord blood and 52.5%, 25.8%, 3.6%, 1.4%, and 0% at age 4, respectively. Determinants of seropositivity at age 4 were cord seropositivity (JCPyV, HPyV7, HPyV10, CMV), vaginal delivery (HPyV10), breastfeeding (CMV), younger age at day-care entry (BKPyV, KIPyV, WUPyV, TSPyV, HPyV10, HPyV9, EBV, CMV), and swimming pool attendance (BKPyV, KIPyV, WUPyV, HPyV10). Television viewing, parental stress, and hygiene practices were inversely associated with the seroprevalence of Polyomaviruses and Herpesviruses.
2) Based on crosss-ectional data at age four, we investigated the association between seropositivity to single or multiple Polyomaviruses and Herpesviruses with child’s neurodevelopment and behavior among 674 children. Seropositivity to BKPyV, a potential neurotropic virus, was associated with higher score in ADHDT inattention subscale. Moreover, children seropositive to ≥8 polyomaviruses had lower score in ADHDT inattention subscale and lower score in SDQ hyperactivity–inattention subscale versus children seropositive to ≤3 Polyomaviruses.
3) H.pylori seroprevalence in cord blood, representing maternal status, was 41.5% and in children four years of age was 6.5%. Children of Helicobacter pylori seropositive mothers compared to those of seronegative mothers, had lower score in all scales of the MCSA excluding motor scale. In addition, seropositivity in cord blood specifically to GroEl and NapA – two of the 12 Helicobacter pylori proteins investigated – was associated with statistically significant lower scores in almost all scales of MCSA. At age four, Helicobacter pylori seropositive children performed worst in neurodevelopment assessment compared to their seronegative counterparts although no association reached statistically significant level.
4) Based on cross-sectional data of 674 children participating at the 4 years of age follow-up and prospective data of 440 children at age six, we investigated the association of seropositivity to single or multiple Polyomaviruses and Herpesviruses with child’s obesity and cardiometabolic traits at the corresponding ages. BKPyV seropositivity was associated with lower BMI, waist circumference, sum of skinfolds and lower leptin levels at age four and with lower BMI, waist circumference and %body fat at age six. On the other hand, CMV seropositivity was associated with higher BMI at age four and six and higher sum of skinfolds at age six. Moreover, children with “2-3 Herpesviruses infections” versus those with “0 Herpesviruses infections” had higher BMI SD-score, waist circumference and sum of skinfolds at age four.
Conclusions
In summary, findings of the present thesis describe the natural history of acquisition of ten Polyomaviruses (BKPyV, JCPyV, KIPyV, WUPyV, HPyV6, HPyV7, TSPyV, MCPyV, HPyV9 and HPyV10), five Herpesviruses (EBV, CMV, HSV-1, HSV-2, HHV-8) and Helicobacter pylori in a population of young healthy children which are prospectively followed in the Rhea mother-child cohort in Crete. To the best of our knowledge, there is no previous report gathering such detailed information on seroprevalence to those common infections. We have shown that, by age 4, children were commonly seropositive to Polyomaviruses and Herpesviruses, with the exceptions of HHV-8, HSV-1 and HSV-2. Seroprevalence to Helicobacter pylori was low and in accordance with reports from other western populations. The resent analysis provides important insights into the factors that determine the acquisition of such common infections by age four. For example, younger age at day-care entry importantly contributed in the epidemiology of BKPyV, KIPyV, WUPyV, TSPyV, HPyV10, HPyV9,EBV, and CMV while there was as strong socioeconomic position patterning of Polyomaviruses seroprevalence at age four. Such findings are important allowing inference on potential routes of transmission for those viruses among young children.
Intriguing are the findings regarding the association of Polyomaviruses and Herpesviruses infections with child’s neurodevelopment. The inverse association between the number of Polyomaviruses infections acquired up to age four with symptoms of inattention and hyperactivity might indicate not only potential interaction between the immune system and the developing brain but also a different pattern of acquisition of such common infections among children with such behaviours. From a public health perspective, it is particularly important the finding that children of Helicobacter pylori seropositive mothers had worst neurodevelopmental performance at age four, raising questions on whether treatment for Helicobacter pylori could prevent such an outcome. Finally, we observed very strong associations for BKPyV and CMV with obesity and cardiometabolic traits in childhood. Similar findings have not been reported in children and should be replicated in other populations.
|
Seroepidemiology of polyomaviruses, herpesviruses and helicobacter pylori in early life and associations with health outcomes in childhood the RHEA mother child cohort in Crete Greece
