<p> 1High frequency detection of Toxoplasma gondii DNA in human neonatal </p><p>2tissue from Libya.</p><p>3Sameena Z.H. Haqa, Muftah S. Abushahamaa,b, Omar Gerwasha,b, Jacqueline M. Hughesa, </p><p>4Elizabeth A. Wrighta, Mohamed S. Elmahaishib, Zhao-Rong Luna,c,d, Denise Thomassond and </p><p>5Geoff Hidea,d* </p><p>6aBiomedical Research Centre, School of Environment and Life Sciences, University of Salford, 7Salford, M5 4WT, UK 8bMisurata Central Hospital, PO Box 65 Misurata, Libya 9cCenter for Parasitic Organisms, State Key Laboratory of Biocontrol, School of Life Sciences, 10and Key Laboratory of Tropical Disease Control of the Ministry of Education, Zhongshan 11School of Medicine, Sun Yat-Sen University, Guangzhou 510275, China 12dEcosystems and Environment Research Centre, School of Environment and Life Sciences, 13University of Salford, Salford, M5 4WT, UK 14</p><p>15*Corresponding Author: Tel: +44 161 2953371; Email: [email protected] 16</p><p>17</p><p>1 1</p><p>2 18</p><p>19Abstract 20Background: 21Toxoplasma gondii is a parasite that causes significant disease in humans. Toxoplasmosis is 22normally asymptomatic, unless associated with congenital transmission, or in 23immunocompromised people. Congenital transmission generally occurs at low frequencies. 24In this study, we use PCR to investigate possible congenital transmission of T. gondii during 25pregnancy in a cohort of mothers from Libya. 26Methods: 27Two hundred and seventy two pregnant women (producing 276 neonates) were recruited to 28obtain umbilical cord tissue from their neonates at birth. DNA was extracted from umbilical 29cord tissue and tested for T. gondii DNA using two specific PCR protocols based on the sag 1 30and sag 3 genes. 31Results: 32Toxoplasma gondii DNA was detected in the umbilical cord DNA from 27 of the 276 33neonates giving a prevalence of 9.9% (95% CI: 6.8-13.9%). Compared with more commonly 34reported rates of congenital transmission of 0.1% of live births, this is high. There was no 35association of infection with unsuccessful pregnancy. 36Conclusions: 37This study shows a high frequency presence of T. gondii DNA associated with neonatal tissue 38at birth in this cohort of 276 neonates from Libya. Although PCR cannot detect living 39parasites, there is the possibility that this indicates a higher than usual frequency of 40congenital transmission. 41 42Keywords</p><p>43Congenital infection, Toxoplasma gondii, Libya, miscarriage, neonatal, PCR </p><p>3 2</p><p>4 44 45Introduction</p><p>46Toxoplasmosis is caused by the protozoan parasite Toxoplasma gondii and is an important</p><p>47human and animal disease with a worldwide distribution.1 The prevalence of infection in the</p><p>48human population is estimated at 30% globally. However prevalence varies with values of</p><p>4910 -20% recorded in the UK, China and US but exceeds 40% in some countries for example,</p><p>50within South America or continental Europe.2 Human infection is acquired through contact</p><p>51with oocysts passed in faeces by the definitive host, the cat, or by consuming the tissue</p><p>52cyst(s) found in uncooked or undercooked meat. Congenital transmission, from mother to</p><p>53baby, is also considered a route of transmission – although is considered as rare in humans</p><p>54with occurrence in less than 1 in 1000 live births.1, 3-5 Toxoplasma infection is usually</p><p>55asymptomatic. However, in pregnant women a novel infection can cause congenital</p><p>56toxoplasmosis in the foetus, neonate and developing child and some of them may also</p><p>57develop brain diseases such as bipolar disorder and schizophrenia.4, 6-9 Studies on prevalence</p><p>58in pregnant women, for example in China,10 show that the frequency of infection usually</p><p>59follows the general population prevalence – in this example, averaging 10% in both cases.</p><p>60However, chronic infection of the mother usually does not result in transmission to the</p><p>61foetus and transmission is generally thought to be related to seroconversion during</p><p>62pregnancy5 or possibly, under some circumstances, reinfection of a chronic infection.11</p><p>63 In humans, an estimated 10% of prenatally acquired infections result in abortion (i.e.</p><p>64spontaneous miscarriage) or neonatal death.12, 13 Furthermore, an additional 10-23% of</p><p>65infected newborns will show signs and symptoms present at birth such as retinochoroiditis,</p><p>66intracranial calcifications and hydrocephalus.12, 14, 15 It is these abnormalities that</p><p>5 3</p><p>6 67significantly reduce the quality of life of children that survive prenatal infection of the</p><p>68mother.3, 5 </p><p>69 While studies on congenital transmission in humans suggest that it may occur rarely</p><p>70in the absence of infection during pregnancy, some recent studies in other species have,</p><p>71controversially, proposed that congenital infection may be occurring much more frequently</p><p>72than previously thought. A number of studies, conducted over many years, using</p><p>73experimental infection of rodents has shown that congenital infection can occur</p><p>74frequently.16-18 However, the frequency of this mode of transmission in natural populations</p><p>75of animals, or humans, is less clear. PCR based studies on the transmission of T. gondii in</p><p>76several mammalian species e.g. mice and woodmice17, 19 - 21 and sheep22 - 25 have suggested</p><p>77that congenital transmission may be occurring at a higher frequency than previously</p><p>78thought. Although higher frequencies of congenital transmission are reported in these</p><p>79studies, in many cases there seems to be no measurable disease effect on the neonates in</p><p>80these species. This might suggest that congenital infection may result in apparently healthy</p><p>81carriers. There is some debate, however, as to the generality and importance of these</p><p>82results – especially in sheep.26 - 28</p><p>83 There are some challenges in measuring congenital transmission rates in natural</p><p>84human populations in ways that are not detrimental to the mother or neonate. The</p><p>85standard approach is to measure the seroprevalence using a variety of serological</p><p>86approaches.1 However, some studies have suggested that more sensitive techniques, such</p><p>87as PCR or parasite detection by bioassay, are capable of detecting parasites in seronegative</p><p>88individuals (e.g. rodents29, 30, pigs31 and humans32). In the PCR based studies on sheep,22, 23</p><p>89the measurement of congenital transmission was conducted utilising umbilical cord tissue as</p><p>7 4</p><p>8 90a proxy for neonatal tissue. In those studies, comparison of PCR detection of Toxoplasma</p><p>91DNA in umbilical cord tissue with internal tissues (brain) from infected and uninfected</p><p>92aborted lambs showed very strong agreement. These results indicated that PCR positive</p><p>93umbilical cord tissue was a very good predictor for infection of the brain . This offers the</p><p>94possibility of using a similar approach to investigate congenital transmission in humans.</p><p>95 Little is known about congenital or general T. gondii infection in humans in Libya. A</p><p>96study of 4,280 people from Tripoli was conducted in 1987 and showed prevalence of 51.6%,</p><p>9743.3% and 43.7% in adult males (1032/2000), adult females (130/300) and school children</p><p>98(865/1980) respectively.33 While another study, conducted specifically on pregnant women</p><p>99in Benghazi (n=369), found that 47.4% (175/369) were infected with T. gondii.34 To our</p><p>100knowledge, no studies have been conducted on congenital transmission in humans in Libya.</p><p>101The objectives of this study were, first, to use PCR detection of T. gondii DNA as a tool to</p><p>102investigate the presence of T. gondii in umbilical cord tissue at birth to gain an assessment</p><p>103of possible congenital transmission and, secondly, to investigate the relationship between</p><p>104presence of T. gondii DNA and clinical or demographic history.</p><p>9 5</p><p>10 105Materials and Methods </p><p>106Sample collection and ethical approval</p><p>107A total of 272 mothers were recruited during antenatal consultations at Misurata Central</p><p>108Hospital, Libya, in the period prior to the Libyan crisis (2011) when civil war broke out. The</p><p>109participating mothers originated from several cities across the West of Libya (Misurata</p><p>110n=235, Zliten n=17, Tawarga n=9, Al Khums n=6, Sirte n=2, Qaminis n=1, Tripoli n=1,</p><p>111unknown location n= 1). Ethical procedures followed local Libyan regulations at the Hospital</p><p>112and both informed patient consent and sample collection was overseen by one of the</p><p>113authors (M.S.E., Consultant Obstetrician) but followed protocols approved for a parallel</p><p>114study in the UK which was awarded ethical approval by the NHS National Research Ethics</p><p>115Service (ref O5/Q1409/105). University of Salford ethical approval was given by the</p><p>116Research Governance and Ethics Committee, RGEC04/16, and the Research Ethics Panel,</p><p>117REPN09/102. Informed consent was obtained from mothers during antenatal sessions to</p><p>118allow collection of discarded umbilical cord tissue, allow completion of a questionnaire and</p><p>119allow information to be obtained from the clinical notes concerning the current birth. Both</p><p>120written information (translated into Arabic) and verbal explanation were provided to ensure</p><p>121understanding and subjects were coded to ensure anonymity with respect to samples.</p><p>122Subjects were free to withdraw at any stage in the consent, collection or post collection</p><p>123process. At collection, samples containing 1 cm3 of tissue (i.e. umbilical cord) were collected</p><p>124aseptically and washed using sterile saline to remove any maternal fluids. Using sterile</p><p>125scissors, they were sheared into small pieces and placed into tubes containing 400 µl of lysis</p><p>126buffer (0·1 M Tris-HCl (pH 8·0), 0·2 M NaCl, 5 mM EDTA, 0·4% SDS), 4 µl of Proteinase K (50</p><p>127mg/ml) and the contents were mixed thoroughly using a vortex for a few seconds. The tube</p><p>11 6</p><p>12 128was then incubated overnight at 56 oC to assist with cell lysis and the crude DNA extract was</p><p>129then stored at -20 oC until further purification. Further DNA purification was carried out</p><p>130using phenol/chloroform/isoamyl alcohol as previously described.22 Purified DNA was</p><p>131tested by PCR using primers to the mammalian tubulin gene to ensure the suitability of the</p><p>132DNA for PCR as previously described.22 Appropriate protocols were used at each stage of the</p><p>133DNA purification and any subsequent PCR amplifications to prevent cross-contamination as</p><p>134described previously.21, 23, 25, 35 </p><p>135PCR detection of T. gondii</p><p>136Detection of T. gondii was carried out using a nested polymerase chain reaction (PCR)</p><p>137amplification of both surface antigen gene 1 (sag 1) as described previously24, 36 and surface</p><p>138antigen gene 3 (sag 3)37 with subsequent modifications.38 Pure T. gondii DNA from RH,</p><p>139Martin, 17695 and COR strains were used as positive controls in all PCR experiments. All</p><p>140samples were tested a minimum of 3 times with a PCR to detect the sag 1 gene and</p><p>141considered positive when three successful PCR amplifications were achieved. Occasional or</p><p>142sporadic positive reactions were retested and considered negative if they could not be</p><p>143amplified to satisfy the three successful amplification criteria. Samples found to be positive</p><p>144using the sag 1 PCR criteria were also confirmed using sag 3 PCR which amplifies from an</p><p>145independent gene locus. Samples negative for sag 1 PCR were also confirmed as negative</p><p>146by sag 3 PCR.</p><p>147Statistical analysis</p><p>148Statistical associations were carried out using 2 x 2 contingency tables, with Fisher’s Exact</p><p>149Test. Correlation tests involving classes of data (e.g. age and weight classes) were carried</p><p>13 7</p><p>14 150out using a rank correlation test (Spearman’s). Comparisons of observed with expected data</p><p>151were carried out using chi2 analysis on EXCEL. P-values of less than 0.05 were used to</p><p>152indicate statistical significance.</p><p>15 8</p><p>16 153</p><p>154Results</p><p>155 Two hundred and seventy two Libyan mothers consented to participate in this study</p><p>156to investigate the frequency of detection of T. gondii DNA in foetal-derived umbilical cord</p><p>157tissue at birth. Maternal age ranges, at the time of the current birth, varied from 17-45</p><p>158years, with a median age of 27 years (interquartile range, 24 to 32 years). This cohort of</p><p>159mothers gave birth to 276 babies (with singletons (n=268) and twins (n=8)). Of the</p><p>160newborns, 51.1% (141/276) were male, 47.8% (132/276) were female and 1.1% (3/276) had</p><p>161unrecorded gender in the available clinical records. Furthermore, 5 pregnancies (1.8%</p><p>162(5/272) were unsuccessful resulting in intrauterine foetal death of the neonate (all</p><p>163singletons in this instance – 3 male and 2 females). The subsequent determination of cause</p><p>164of death of these neonates was not recorded in the clinical notes made available to the</p><p>165study team. Clinical records were used to determine historical childbirth data, including live</p><p>166and miscarried babies for each mother. Historical data from only 271 mothers could be</p><p>167obtained. This cohort of mothers had a varied history of pregnancies (including miscarried</p><p>168births) ranging between 1 and 15 (median, 3; interquartile range, 1 to 5; n= 271) births per</p><p>169mother. Sixty of the mothers had had one or more miscarriages (including the 5 recorded</p><p>170from the current cohort of births) totalling the loss of 90 potential neonates. The number of</p><p>171miscarriages per mother ranged from 0 (n=211) to between 1 and 5 (median, 1;</p><p>172interquartile range, 1 – 2; n=60). Four of the 5 current miscarriages recorded in this study</p><p>173were first time miscarriages while the other mother had also miscarried once previously, but</p><p>174had previously also had 6 successful births.</p><p>17 9</p><p>18 175 The risk of miscarriage with maternal age is presented in Table 1. The rate of</p><p>176miscarriage in current (1.8% (5/272)) and previous pregnancies was high with 21.7%</p><p>177(60/272) mothers experiencing miscarriage at some point. A strong positive correlation</p><p>178between the age class of the mothers and the proportion of mothers with miscarriages was</p><p>179observed (correlation co-efficient = 0.88, n = 7, P< 0.05).</p><p>180 T. gondii DNA was detected in 27 of the 276 foetal-derived umbilical cord samples</p><p>181(Table 2), representing the 272 pregnancies. This gave a detection prevalence of 9.9%</p><p>182(27/272) of pregnancies (95% CI: 6.8-13.9%). Of the successful pregnancies, resulting in the</p><p>183birth of live babies, 10.1% (27/267) of foetal-derived umbilical cord tissue showed positive</p><p>184detection of T. gondii DNA. T. gondii DNA was not detected in the umbilical cord of any of</p><p>185the 5 miscarried neonates. Therefore, no association was found between presence of T.</p><p>186gondii DNA and unsuccessful pregnancies (P= 1.00). Of 273 newborns with available gender</p><p>187information, 9.2% (13/141) males and 10.6% (14/132) females were Toxoplasma positive</p><p>188and there was no significant effect of gender on T. gondii DNA presence (P = 0.84). To</p><p>189investigate whether a history of miscarriage was associated with presence of T. gondii DNA</p><p>190in the foetal umbilical cord tissue, clinical records were examined. Of the 27 tissue samples</p><p>191positive for T. gondii DNA, 6 were neonates from mothers that had suffered at least one</p><p>192historical miscarriage. There was no association between history of previous miscarriage</p><p>193and T. gondii positivity (P=0.81) nor total miscarriage history (i.e. including the current</p><p>194miscarriage) (P=1.00). The birthweight of the cohort of babies ranged from 1.3 kg to 5 kg. To</p><p>195investigate whether there was any association between T. gondii positivity and weight of</p><p>196baby at birth, newborns were categorised into weight classes, and there was no significant</p><p>19 10</p><p>20 197difference (χ2 = 10.4; d.f. = 6; P = 0.89) in frequency of T. gondii positivity in difference</p><p>198weight classes.</p><p>199</p><p>200Discussion</p><p>201 The results of the current study indicated that high frequencies of T. gondii DNA</p><p>202could be detected in the foetal-derived umbilical cord tissue samples of neonates born in</p><p>203Libya. One could hypothesise that there is a risk of congenital transmission of the parasite</p><p>204T. gondii occurring from mother to baby. If this is the case, then there is a significant</p><p>205implication. As the transmission rate is significantly higher than previously published</p><p>206estimates (less than 1 in 1,000 live births,1, 3-5) the implication is that if parasite transmission</p><p>207is occurring, there are potentially a considerable number of infected babies being born</p><p>208apparently asymptomatically (at least at birth). For example in this study, of 27 T. gondii PCR</p><p>209positive umbilical cord samples from the newborns, all were born without miscarriage and</p><p>210without any observable post-natal defects typically associated with T. gondii (e.g.</p><p>211hydrocephalus). In fact, some studies have evidenced that, in general, 80-90% of</p><p>212congenitally infected neonates show no clinical symptoms at birth.39, 40 Although none of the</p><p>213neonates in this study showed any detectable symptoms (100% (27/27)), with the relatively</p><p>214small sample size of T. gondii positive neonates, this is not significantly different from 80 –</p><p>21590% found in the previously reported studies. </p><p>216 One of the limitations of this study is that detection is based on PCR and this means</p><p>217that it is possible that it does not represent living parasites, but parasite DNA molecules.</p><p>218While this has to be acknowledged, PCR is used widely as a diagnostic tool for Toxoplasma1</p><p>21 11</p><p>22 219and a wide range of other protozoan parasitic diseases.41 The presence of parasite DNA</p><p>220does, however, indicate the presence, and proximity, of the parasite to the foetus during</p><p>221these pregnancies. It is, perhaps, surprising that none of the occurrences of miscarriage,</p><p>222seen in this study, appeared to be associated with the presence of T. gondii DNA. In this</p><p>223cohort of mothers, the current and historic levels of miscarriage were high. </p><p>224 An important contributing factor to newborn mortality has often been thought to be</p><p>225the activity of parasites and infectious agents. Toxoplasma is well known as an agent that</p><p>226causes miscarriage.5 In this study there was no relationship between infection and</p><p>227miscarriage. However, the sample size of miscarried neonates was small (n=5). In the case of</p><p>228unsuccessful pregnancies, none of the samples showed the presence of the parasite DNA in</p><p>229the foetal umbilical cord samples. Furthermore, T. gondii was detected at a higher</p><p>230frequency from the umbilical cord samples of successful pregnancies (10.1% (27/267)) than</p><p>231would be expected from previously published estimates on congenital transmission rates of</p><p>232around 0.1%. This raises the question as to whether live born babies might be symptomatic</p><p>233or asymptomatic carriers of the infection. In this study it was not possible to follow up</p><p>234individuals – this is an important aspect that could be considered in future studies since</p><p>235congenital infection is often associated with late onset clinical entities such as ocular</p><p>236toxoplasmosis. However, if congenitally infected children were frequently asymptomatic,</p><p>237and infection persisted through adulthood, it might offer the opportunity for the parasite to</p><p>238be passed down vertically or via serial congenital transmission in human populations.26</p><p>239 Few studies have been conducted on vertical transmission in humans but some</p><p>240evidence for this may be accumulated from animal species. Watson and Beverley42 studied</p><p>241sheep Toxoplasma and showed that infection of the placenta corresponded to infection of</p><p>23 12</p><p>24 242sacrificed live born lambs; thus demonstrating a link between maternal and foetal infection.</p><p>243Further recent studies on sheep have revealed that congenital transmission may be</p><p>244sufficient to explain the maintenance of T. gondii in natural populations of sheep without</p><p>245requiring new infections by oocysts secreted by cats.22 - 25 In the case of these sheep studies,</p><p>246it was demonstrated that PCR detection of T. gondii DNA in the umbilical cord was a good</p><p>247indicator of infection when aborted lamb brain tissue was compared in the same animal.</p><p>248Nevertheless, a challenge to be encountered in humans (as may be the case in sheep and</p><p>249other animals) is to understand whether these individuals are truly asymptomatic carriers of</p><p>250vertical transmission. In the case of the sheep studies24, 25 detailed family pedigrees were</p><p>251established which demonstrated the occurrence of familial transmission. Further work</p><p>252remains to be done to determine the role of vertical transmission of Toxoplasma in humans.</p><p>253 Only a few studies exist on T. gondii prevalence in Libya and they demonstrate high</p><p>254prevalences (40 – 50%) 33, 34 compared to many other areas of the world. Although updates</p><p>255on these baseline studies have not been done, the relatively high levels of detection of T.</p><p>256gondii in the foetal-derived tissues, reported in this study, are consistent with these high</p><p>257prevalence levels. Similar high prevalence levels of T. gondii have been reported in pregnant</p><p>258women from continental Europe: Spain 28.6%; Sweden 18%; Belgium 48%; Netherlands 35</p><p>259%; Greece 36% and Germany 63%.2 However, in these other studies only 0.2% congenital</p><p>260transmission was shown to be occurring,43 based on proven clinical cases of infection. In</p><p>261comparison the reported rate of 9.9% (27/272) reported in this study, based on PCR, is very</p><p>262high. The difference can be explained by differences in the detection and diagnostic systems</p><p>263used but interestingly may suggest higher levels of transmission of parasites than appear as</p><p>264clinically proven cases.</p><p>25 13</p><p>26 265 Unlike the current study, previous reports have generally used serology for the</p><p>266diagnosis of T. gondii in human tissues. Serology is the classical diagnostic modality</p><p>267consisting of several tests on which, virtually all Toxoplasma prevalence studies are carried</p><p>268out43 but a variety of different approaches are used. Most of these require the</p><p>269determination of antibody cut off prior to use and there is some variation in standardisation</p><p>270globally. The gold standard for testing for T. gondii is serology which has been used by</p><p>271experts in the field1 now routinely records the serological method used and antibody cut-</p><p>272offs when reporting prevalence studies.4 This enables a judgement of the relative validity of</p><p>273results. In the current study only molecular methods were used to detect for congenital</p><p>274transmission in babies. A possibility exists that, if congenital transmission is being detected</p><p>275in this study, differences in the approaches to detection of the parasite may account for</p><p>276these unusually high transmission frequencies. In sheep, Mason et al.,44 demonstrated that</p><p>277seronegative neonatal lambs could be PCR positive and the two approaches (serology vs</p><p>278PCR) showed poor agreement when comparisons were made. This is suggestive that the</p><p>279differences in measurements of congenital transmission could be due to differences in the</p><p>280methodology of measurement. Further work may still be required to establish robust</p><p>281protocols for investigating congenital transmission frequency.</p><p>282 Clinical cases of congenital infection have been observed to occur from chronically</p><p>283infected mothers in some rare instances.5 These may either be caused by reactivation of</p><p>284cysts or reinfection.11, 45, 46, 47 It has been suggested26, 48 that the natural state of</p><p>285immunosuppression entered into by the mother during pregnancy could be hijacked by the</p><p>286parasite. In this case it could use the opportunity to reactivate from tissue cysts, as</p><p>287tachyzoites, which then cross the placenta to infect the foetus. Although most studies</p><p>27 14</p><p>28 288report little evidence for reactivation, some studies suggest this might be a possibility.49, 50, 51</p><p>289Further work would be required to investigate this question.</p><p>290 There are a number of shortcomings of this study which we recognise have an</p><p>291influence on the strength with which the results can be interpreted. Firstly, while data on</p><p>292historical and current miscarriages were collected from records, serological data on</p><p>293screening/diagnosis of mothers for T. gondii infection was not conducted. It was therefore,</p><p>294not possible to determine whether mothers had a chronic infection or sero-converted</p><p>295during pregnancy. In the case of those babies with detectable T. gondii DNA, it is therefore</p><p>296not possible to determine how or at what stage the mothers became infected. It is not</p><p>297possible, therefore, to address chronic reactivation versus infection during pregnancy with</p><p>298this dataset. Furthermore, it was not possible to get access to blood from the babies and,</p><p>299therefore, not able to confirm diagnosis by serology. As discussed previously, based on PCR</p><p>300evidence alone it is not formally possible to conclude that the babies had active infections.</p><p>301Ideally, future studies, should involve early recruitment of mothers and monitoring of</p><p>302infection throughout pregnancy by collection of serum samples. Suitable neonatal samples</p><p>303(e.g. cord blood) for T. gondii testing would need to be collected to try to link the PCR</p><p>304positivity with infection in the newborns. Finally, a long term follow up study of the</p><p>305newborns could provide important clinical data with respect to the consequences of</p><p>306infection. Unfortunately, at the time of writing the political difficulties in Libya are likely to</p><p>307make such future studies challenging.</p><p>308Conclusions</p><p>309 This study shows a high frequency of detection of T. gondii DNA in neonates from</p><p>310births in Libya although in this study there was no association with miscarriage. While we 29 15</p><p>30 311recognise that there are caveats associated with our interpretations, it raises the possibility</p><p>312that the foetus could be potentially exposed to T. gondii infection with higher frequency</p><p>313than previous studies suggested. It also raises questions about the methods used to detect</p><p>314transmission to the foetus, the importance and mechanism of congenital transmission and</p><p>315the potential generation of asymptomatic carriers of vertical transmission in humans. Future</p><p>316studies are required to explore these important questions further.</p><p>317Authors’ contributions</p><p>318SZHH, MSA, OG, DT and GH conceived the study. MSE conducted the clinical assessment of</p><p>319patients in Libya. MSA, OG and MSE conducted sample collection in Libya. DT, MSE and GH</p><p>320considered ethical issues and MSE implemented local ethical processes in Libya. Laboratory</p><p>321work was conducted by MSA, OG, JMH, EAW and DT. Data analysis and interpretation was</p><p>322conducted by SZHH, MSA, OG, ZRL and GH. SZHH and GH drafted the manuscript but all</p><p>323authors contributed to the writing of it. SZHH, ZRL and GH critically revised the manuscript</p><p>324for intellectual content. All authors read and approved the final manuscript. GH is the</p><p>325guarantor of the paper.</p><p>326Acknowledgements</p><p>327We would like to thank all of the anonymous subjects for volunteering to take part in this</p><p>328study and Professor Judith Smith for provision of reference T. gondii strains. We thank Dr</p><p>329Khaled Elmahaishi for his involvement in the sampling in the hospital in Misurata, Libya.</p><p>330Preliminary unverified data from this work was reported at the International Conference on</p><p>331Parasitology in Melbourne in August 2010 and reported in the proceedings of that</p><p>332conference.</p><p>31 16</p><p>32 333Funding</p><p>334This work was supported by The People’s Bureau of the Great Socialist People’s Libyan Arab</p><p>335Jamahiriya London [grant numbers 3548, 4225]; the University of Salford and the British</p><p>336Society for Parasitology.</p><p>337Potential conflicts of interest</p><p>338All authors report no potential conflicts of interest.</p><p>339Ethical approval</p><p>340Ethical procedures followed local Libyan regulations at the Misurata Hospital. Informed</p><p>341patient consent, patient anonymity and sample collection was overseen by MSE, consultant</p><p>342obstetrician. Protocols used were the same as used in an ethically approved parallel study in</p><p>343the UK (NHS National Research Ethics Service (ref O5/Q1409/105)). University of Salford</p><p>344ethical approval was given by the Research Governance and Ethics Committee, RGEC04/16,</p><p>345and the Research Ethics Panel, REPN09/102. </p><p>33 17</p><p>34 346References </p><p>3471. Dubey JP. Toxoplasmosis of animals and humans. 2nd Edn. CRC Press, Boca Raton, Florida,</p><p>348USA. 2010.</p><p>3492. Pappas G, Roussos N, Falagas ME. Toxoplasmosis snapshots: Global status of Toxoplasma </p><p>350gondii seroprevalence and implications for pregnancy and congenital toxoplasmosis. Int J </p><p>351Parasitol 2009; 39:1385-94.</p><p>3523. Tenter AM, Heckeroth AR, Weiss LM. T. gondii: from animals to humans. Int J Parasitol </p><p>3532000; 30:1217-58.</p><p>3544. Dubey JP, Jones JL. T. gondii infection in humans and animals in the United States. Int J </p><p>355Parasitol 2008; 38:1257-78.</p><p>3565. Peyron F, Wallon M, Kieffer F, Garweg J. Toxoplasmosis. In Wilson, Nizet, Maldonado,</p><p>357Remington and Klein. Infectious diseases of the fetus and newborn infant. 8 th ed. Elsevier</p><p>358Saunders, Philadelphia. 2015.</p><p>3596. Montoya JG, Liesenfield O. Toxoplasmosis. Lancet 2004; 363:1965-1976.</p><p>3607. Mortensen PB, Pedersen CB, McGrath JJ et al. Neonatal antibodies to infectious agents</p><p>361and risk of bipolar disorder: a population-based case-control study. Bipolar Disord 2011;</p><p>362624-9. </p><p>3638 Tedla Y, Shibre T, Ali O et al. Serum antibodies to Toxoplasma gondii and Herpesvidae</p><p>364family viruses in individuals with schizophrenia and bipolardisorder: a case-control study.</p><p>365Ethiop Med J. 2011; 211-20. </p><p>3669. Torrey EF, Bartko JJ, Lun ZR, Yolken RH. Antibodies to Toxoplasma gondii in patients with</p><p>367schizophrenia: a meta-analysis. Schizophr Bull 2006; 33:729-36.</p><p>35 18</p><p>36 36810. Gao XJ, Zhao JZ, He ZH et al. Toxoplasma gondii infection in pregnant women in China. </p><p>369Parasitol 2012; 139:139-47.</p><p>37011. Valdes V, Legagneur H, Watrin V et al. Congenital toxoplasmosis due to maternal</p><p>371reinfection during pregnancy. Arch Pediatr 2011; 761-3</p><p>37212. Remington JS. Desmonts G. Toxoplasmosis. In: Remington JS, Klein JO. Infectious disease</p><p>373of the fetus and newborn infant. Philadelphia, USA, WB Saunders. 1990.</p><p>37413. Orriss GD. Animal Disease of Public Health Importance. Emerg Inf Dis 1997; 3: 497-502.</p><p>37514. Koskiniemi M, Lappalainen M, Hedman K. Toxoplasmosis needs evaluation: overview</p><p>376and proposals. Am J Dis Childhood 1989; 143:724-8.</p><p>37715. Gilbert RE, Dunn DT, Lightman S et al. Incidence of symptomatic toxoplasma eye disease:</p><p>378aetiology and public health implications. Epidemiol Infect 1999; 123:283-9.</p><p>37916. Beverley JKA. Congenital transmission of Toxoplasmosis through successive generations </p><p>380of mice. Nature 1959; 183:1348-9.</p><p>38117. Owen MR, Trees AJ. Vertical transmission of T. gondii from chronically infected house </p><p>382Mus musculus and field Apodemus sylvaticus mice determined by polymerase chain </p><p>383reaction. Parasitol 1998; 116:299-304.</p><p>38418. Gao JM, Yi SQ, Wu MS et al.. Investigation of infectivity of neonates and adults from </p><p>385different rat strains to Toxoplasma gondii Prugniaud shows both variation which correlates </p><p>386with iNOS and Arginase-1 activity and increased susceptibility of neonates to infection. Exp </p><p>387Parasitol 2015; 149:47-53.</p><p>38819. Marshall PA, Hughes JM, Williams RH et al. Detection of high levels of congenital </p><p>389transmission of Toxoplasma gondii in natural urban populations of Mus domesticus. </p><p>390Parasitol 2004; 128:39-42. </p><p>37 19</p><p>38 39120. Thomasson D, Wright EA, Hughes JM et al. Prevalence and co-infection of Toxoplasma </p><p>392gondii and Neospora caninum in Apodemus sylvaticus in an area relatively free of cats. </p><p>393Parasitol 2011; 138:1117-23.</p><p>39421. Bajnok J, Boyce K, Rogan MT et al. Prevalence of Toxoplasma gondii in localized</p><p>395populations of Apodemus sylvaticus is linked to population genotype not to population</p><p>396location. Parasitol 2015; 142:680-90. </p><p>39722. Duncanson P, Terry RS, Smith JE, Hide G. High levels of congenital transmission of </p><p>398Toxoplasma gondii in a commercial sheep flock. Int J Parasitol 2001; 31:1699-1703.</p><p>39923. Williams RH, Morley EK, Hughes JM et al. High levels of congenital transmission of </p><p>400Toxoplasma gondii in longitudinal and cross-sectional studies on sheep farms provides </p><p>401evidence of vertical transmission in ovine hosts. Parasitol 2005; 130:301-7.</p><p>40224. Morley EK, Williams RH, Hughes JM et al. Significant familial differences in the frequency</p><p>403of abortion and Toxoplasma gondii infection within a flock of Charollais sheep. Parasitol </p><p>4042005; 131:181-5. </p><p>40525. Morley EK, Williams RH, Hughes JM et al. Evidence that primary infection of Charollais </p><p>406sheep with Toxoplasma gondii may not prevent foetal infection and abortion in subsequent </p><p>407lambings. Parasitol 2008; 135:169-173.</p><p>40826. Hide G, Morley EK, Hughes JM et al. Evidence for high levels of vertical transmission in T.</p><p>409gondii. Parasitology 2009; 136:1877-85.</p><p>41027. Buxton D, Rodger SM, Maley SW, Wright SE. Toxoplasmosis: The possibility of vertical </p><p>411transmission. Small Ruminant Res 2006; 62:43-6.</p><p>41228. Innes EA, Bartley PM, Buxton D, Katzer F. Ovine toxoplasmosis. Parasitol 2009;</p><p>413136:1887-94. </p><p>39 20</p><p>40 41429. Dubey JP, Shen SK, Kwok OC, Thulliez P. Toxoplasmosis in rats Rattus norvegicus: </p><p>415congenital transmission to first and second generation offspring and isolation T. gondii from </p><p>416seronegative rats. Parasitol 1997; 115:9-14.</p><p>41730. Araujo JB, da Silva AV, Rosa RC et al. Isolation and multilocus genotyping of Toxoplasma</p><p>418gondii in seronegative rodents in Brazil. Vet Parasitol 2010; 174:328-31. </p><p>41931. Dubey JP, Hill DE, Rozeboom DW et al. High prevalence and genotypes of Toxoplasma</p><p>420gondii isolated from organic pigs in northern USA. Vet Parasitol 2012; 188:14-8. </p><p>42132. Fallahi S, Seyyed Tabaei SJ, Pournia Y et al. Comparison of loop-mediated isothermal</p><p>422amplification (LAMP) and nested-PCR assay targeting the RE and B1 gene for detection of</p><p>423Toxoplasma gondii in blood samples of children with leukaemia. Diagn Microbiol Infect Dis</p><p>4242014; 79:347-54. </p><p>42533. Khadre MA, El Nageh MM. Serological survey for toxoplasmosis in Tripoli in Libya. Trans</p><p>426Roy Soc Trop Med Hyg 1987; 81: 761-3.</p><p>42734. Kassem HH, Morsy TA. The prevalence of anti-Toxoplasma antibodies among pregnant</p><p>428women in Benghazi, Libya. J Egyptian Soc Parasitol 1991; 21: 69-74.</p><p>42935. Hughes JM, Williams RH, Morley EK et al. The prevalence of Neospora caninum and co-</p><p>430infection with Toxoplasma gondii by PCR analysis in naturally occurring mammal </p><p>431populations. Parasitol 2006; 132:29-36.</p><p>43236. Terry RS, Smith JE, Duncanson P, Hide G. MGE-PCR: a novel approach to the analysis of</p><p>433Toxoplasma gondii strain differentiation using mobile genetic elements. Int J Parasitol 2001;</p><p>43431:155-61.</p><p>43537. Grigg ME, Ganatra J, Boothroyd JC, Margolis TP. Unusual abundance of atypical strains</p><p>436associated with human ocular toxoplasmosis. J Infect Dis 2001; 184:633-9.</p><p>41 21</p><p>42 43738. Su C, Zhang X, Dubey JP. Genotyping of Toxoplasma gondii by multilocus PCR-RFLP</p><p>438markers: a high resolution and simple method for identification of parasites. Int J Parasitol</p><p>4392006; 36:841 – 8.</p><p>44039. Berrebi A, Assouline, C., Bessieres et al. Long-term outcome of children with congenital</p><p>441toxoplasmosis. Am J Obstet Gynaecol 2010; 203: 552e1-6.</p><p>44240. Gras L, Wallon M, Pollak A et al. Association between prenatal treatment and clinical</p><p>443manifestations of congenital toxoplasmosis in infancy: a cohort study in 13 European</p><p>444centres. Acta Paediatr 2005; 94:1721-31.</p><p>44541. Hide G and Tait A. Molecular epidemiology of African sleeping sickness. Parasitol 2009;</p><p>446136:1491-1500.</p><p>44742. Watson WA, Beverley JK. Ovine abortion due to experimental Toxoplasmosis. Vet Rec</p><p>4481971; 88:42-5.</p><p>44943. Rorman E, Zamir CS, Rilkis I, Ben-David H. Congenital toxoplasmosis-prenatal aspects of</p><p>450Toxoplasma gondii infection. Reproduct Toxicol 2006; 21:458-72.</p><p>45144. Mason S, Quinnell RJ, Smith JE. Detection of Toxoplasma gondii in lambs via PCR</p><p>452screening and serological follow-up. Vet Parasitol 2010; 169:258-63.</p><p>45345. Gavinet MF, Robert F, Firtion G et al. Congenital toxoplasmosis due to maternal</p><p>454reinfection during pregnancy. J Clin Microbiol. 1997; 35:1276-77. </p><p>45546. Araujo F, Slifer T, Kim S. Chronic infection with Toxoplasma gondii does not prevent</p><p>456acute disease or colonization of the brain with tissue cysts following reinfection with</p><p>457different strains of the parasite. J Parasitol 1997; 83:521-2. </p><p>43 22</p><p>44 45847. Dao A, Fortier B, Soete M et al. Successful reinfection of chronically infected mice by a</p><p>459different Toxoplasma gondii genotype. Int J Parasitol 2001;31:63-5. </p><p>46048. Hide G. Role of vertical transmission of Toxoplasma gondii in prevalence of infection.</p><p>461Expert Rev Anti Infect Ther 2016;14:335-44.</p><p>46249. Kodjikian L, Hoigne I, Adam O et al. Vertical transmission of toxoplasmosis from a</p><p>463chronically infected immunocompetent woman. Pediatr Infect Dis J 2004;23:272-4. </p><p>46450. Gigley JP, Bhadra R, Khan IA. CD8 T Cells and Toxoplasma gondii: A New Paradigm. J</p><p>465Parasitol Res 2011; 243796. </p><p>46651. Gigley JP, Bhadra R, Moretto MM, Khan IA. T cell exhaustion in protozoan disease.</p><p>467Trends Parasitol 2012; 28: 377-84.</p><p>45 23</p><p>46</p>