key: cord-0741045-gzpf8auo
authors: Bloise, Enrrico; Zhang, Jianhong; Nakpu, Jittanan; Hamada, Hirotaka; Dunk, Caroline E.; Li, Siliang; Imperio, Guinever E.; Nadeem, Lubna; Kibschull, Mark; Lye, Phetcharawan; Matthews, Stephen G.; Lye, Stephen J.
title: Expression of SARS-CoV-2 cell entry genes, ACE2 and TMPRSS2, in the placenta across gestation and at the maternal-fetal interface in pregnancies complicated by preterm birth or preeclampsia
date: 2020-08-25
journal: Am J Obstet Gynecol
DOI: 10.1016/j.ajog.2020.08.055
sha: 8233409fe352e7f164b7f5c8180003edd70b0e3e
doc_id: 741045
cord_uid: gzpf8auo

Abstract Background While there is some evidence that SARS-CoV-2 can invade the human placenta, limited data exist on the gestational-age dependent expression profile of the SARS-CoV-2 cell entry mediators, ACE2 and TMPRSS2 at the human maternal-fetal interface. There is also no information as to whether the expression of these mediators is altered in pregnancies complicated by pre-eclampsia (PE) or preterm birth (PTB). This is important since the expression of decidual and placental ACE2 and TMPRSS2 across gestation may impact susceptibility of pregnancies to vertical transmission of SARS-CoV-2. Objectives To investigate the expression pattern of specific SARS-CoV-2 cell entry genes, ACE2 and TMPRSS2, in the placenta across human pregnancy and in paired samples of decidua and placenta in pregnancies complicated by PTB or PE compared to term, uncomplicated pregnancies. Study Design Two separate cohorts of patients, totalling 87 pregnancies were included. The first cohort comprised of placentae from first (7-9 weeks), second (16-18 weeks), third-trimester preterm (26-31 weeks) and third-trimester term (38-41 weeks) pregnancies (n=5/group), whereas, the second independent cohort, included matched decidua and placentae from pregnancies from term, uncomplicated pregnancies (37-41 weeks; n=14) as well as pregnancies complicated by PTB (26-37 weeks, n=11) or PE (25-37 weeks n=42). Samples were subjected to qPCR and next-generation sequencing (NGS)/RNAseq for ACE2 and TMPRSS2 mRNA expression quantification, respectively. Results In the first cohort, the SARS-CoV-2 cell entry genes ACE2 and TMPRSS2 exhibited a gestational-age dependent expression profile, i.e. ACE2 and TMPRSS2 mRNA was higher (p<0.05) in the first trimester compared to second trimester, PTB and term placentae (p<0.05) and exhibited a negative correlation with gestational age (p<0.05). In the second cohort, RNAseq demonstrated very low/undetectable expression levels of ACE2 in PTB, PE and term decidua and in placentae from late gestation. In contrast, TMPRSS2 was expressed in both decidual and placental samples but did not change in pregnancies complicated by either PTB or PE. Conclusions The increased expression of these SARS-CoV-2 cell entry associated genes in the placenta during the first trimester compared to later stages of pregnancy suggest the possibility of differential susceptibility to placental entry to SARS-CoV-2 across pregnancy. Even though there is some evidence of increased rates of PTB associated with SARS-CoV-2 infection, we found no increase in mRNA expression of ACE2 or TMPRSS2 at the maternal-fetal interface.

The authors report no conflicts of interest. 24 25 Role of the funding source 26 EB is supported by Coordenação de Aperfeiçoamento Pessoal de Nível Superior (CAPES, 27 finance Code 001, Capes-Print fellowship) 28 SJL is supported by Canadian Institutes of Health Research (CIHR) grant (FDN-143262) . ACE2 and TMPRSS2 are highly expressed in the human placenta in early pregnancy 58 but their expression decreases significantly with advancing gestation. Expression of 59 these genes at the maternal-fetal interface did not change in pregnancies 60 complicated by either preterm birth or preeclampsia. 61

The decrease in expression of placental ACE2 and TMPRSS2 with advancing 63 gestational age suggests the potential for differential risk of placental infection 64 across pregnancy. Pregnancies complicated by preterm birth or preeclampsia are not 65 understood. However, to date, there has been limited evidence of its maternal to fetal 115 transmission, in contrast to other viruses 6-9 . The most likely path for the virus to access to 116 the fetus would be via the placenta, however, little is known about the tropism of the 117 SARS-CoV-2 for the decidua and placenta. Emerging reports suggest that SARS-CoV-2 118 can invade the human placenta 10 and the occurrence of a second trimester miscarriage in a 119 patient with symptomatic coronavirus disease, exhibiting SARS-CoV-2 positivity in the 120 placental submembranes and cotyledons; associated with mixed inflammatory infiltrates 121 and funisitis 11 . Furthermore, potential vertical transmission in pregnant women with 122 COVID-19 has been reported. In one case of second trimester preterm delivery, the 123 amniotic fluid and infant tested positive (by PCR) for SARS-CoV-2 12 and one neonate 124 exhibited elevated IGM antibody levels 2 hours after birth 13 . In another case, a third 125 trimester (35 +5 weeks gestation) neonate was born to a mother exhibiting clincal symptons 126 and being tested positive (by PCR) for SARS-CoV-2 genes. The neonate (blood and non-127 bronchoscopic bronchoalveolar lavage fluid-first day of life), placenta and clear amniotic 128 fluid (collected prior to rupture of membranes during cesarean section) also tested positive 129 for SARS-CoV-2 genes. Placental pathological examination, identified diffuse peri-villous 130 fibrin deposition with associated infarction and acute and chronic intervillositis. Of 131 importance, SARS-CoV-2 N-protein immunostaining in the placenta, revealed strong 132 immunosignals concentrated in the syncytial layer 14 . This suggests that the syncytial layer 133 is enriched with SARS-CoV-2 cell entry receptors, and highlights the need to investigate 134 potential routes and associated mechanisms of placental SARS-CoV-2 infection and 135 vertical transmission. 136

Compared to the general population, pregnant women are particularly susceptible to 137 specific viral infections including, cytomegalovirus (CMV), herpes simplex (HSV) and zika 138 (ZIKV) viruses, and exhibit greater complications and mortality rates associated with 139 varicella, rubeola and H1N1 infections. Importantly, the cause of this susceptibility is 140 poorly defined 6 , but it may be related to the immunological adaptations inherent to 141 pregnancy or to the tissue distribution of cell entry viral mediators at the maternal-fetal 142 interface 6 . Indeed, it is estimated that approximately one-third of pregnant women died 143 The best described mediators of SARS-CoV-2 cell entrance, are angiotensin-152 converting enzyme 2 (ACE2) and the transmembrane protease serine 2 (TMPRSS2) protein 153 receptors 22-24 . SARS-CoV-2 cell entry comprises the binding of the N-terminal portion of 154 its spike (S) protein attached to the "corona" like viral envelope, to a pocket of the cell 155 membrane ACE2 receptor. In an important second step, TMPRSS2 cleaves and detaches 156 the S1 from the S2 portion to allow a conformational rearrangement of the viral membrane 157 and subsequent fusion and entry of the virus into the targeted cell 22,25 . Tissue identification 158 and the expression dynamics of these two cell-entry associated proteins are crucial for a 159 better understanding of the SARS-CoV-2 cell tropism and COVID-19 pathogenesis, 160 treatment and prevention. 161

Evidence suggests that placental ACE2 expression decreases from early to late 162 pregnancy 26 , however, even though the expression of TMPRSS2 has been demonstrated in 163 the human placenta 27,28 , the mRNA expression profile of ACE2 and TMPRSS2 in the 164 human placenta across pregnancy (simultaneously comparing first, second and third 165 trimester) and whether mRNA expression of these SARS-CoV-2 associated cell entry 166 proteins are dysregulated in the decidua and placenta from pregnancies complicated by 167 PTB or pre-eclampsia (PE) is unclear. We hypothesized that the placenta expresses ACE2 168 and TMPRSS2, that encode proteins mediating infection of cells within the human 169 maternal-fetal interface, in a gestational-age 170 dependent manner, and that changes in the expression of these genes at the maternal-fetal 171 interface may be associated with pregnancies complicated by PTB and/or PE. Information 172 on SARS-CoV-2 expression dynamics at the maternal-fetal interface across pregnancy, and 173 in cases of obstetric complications, may provide increased understanding of the potential 174 for placenta and fetal infection and thus support management of patients who present with 175 SARS-CoV-2 infection during pregnancy. 176 177

This is a cross-sectional study, involving two different cohorts of patients, totalling 180 87 pregnancies. In the first cohort, we assessed the developmental expression profile of 181 specific SARS-CoV-2 associated cell-entry genes, ACE2 and TMPRSS2, in human 182 placental tissue from 1) first trimester (7-9 weeks; n=5) or 2) second trimester (16-18 183 weeks; n=5) elective terminations, 3) third-trimester spontaneous onset PTB (26-35 weeks; 184 n=5) 2 delivered vaginally and 3 by c-section indicated for fetal distress (2) or bleeding 185 from fibroid (1), and 4) term delivery (38-41 weeks; n=5) following elective c-section 186 delivery prior to labor (4) or following vaginal delivery (1) Americas Inc. CA). Pair-end RNA sequencing was conducted at read-length of 100bp and 244 50 million reads per library. The workflow was conducted on the same platform by the 245 same research team. We analyzed but found no significant variance due to batch effect. 246 247

Sequencing data were analyzed using a high-performance clustering computing system 249 (Galen; Sinai Health System, Toronto). Data quality of raw fastqc files was examined by 250 FastQC (v0.11). After trimming off contaminated reads, Bowtie2 (v2.3), an ultrafast and 251 memory-efficient tool, was used to align sequencing reads to the human reference genome 252 (hg38). Finally, binary alignment/mapping files were built up for the downstream analysis. 253

Uniquely mapped reads were summarized to feature counts using GenomicAlignments 254 (v1.23). Normalized RNA reads for TMPRSS2 expression were extracted by default 255 settings of plotCounts (normalized counts plus a pseudocount of 0.5 as log2 scales) using 256 the outcomes of the DESeqDataSet function provided by DEseq2 package (v1.27). TPM 257 was calculated by the calculateTPM function from R package scater (v1.17) using mapped 258 raw reads and effective gene length of the transcripts. The gene expressions of ACE2 and 259 TMPRSS2 were identified by the ensemble ID ENSG00000130234 and 260 ENSG00000184012, respectively. In this study, we found no evidence of a significant 261 batch effect, therefore no batch correction was assigned. 262 263

Statistical analysis was performed using Prism (version 7.0; GraphPad Software, 265

Inc., San Diego, CA). qPCR data were assessed for normal distribution using D'Agostino 266 and Pearson or the Shapiro-Wilk test; outliers were identified using "QuickCalcs" Outlier 267 calculator program (version 7.0; GraphPad Software, Inc., San Diego, CA). Gene 268 expression in the ontogenetic study was analyzed using one-way ANOVA followed by 269

Tukey's multiple comparisons test. Next-generation sequencing/NGS statistical analyses 270 were performed by R software (v3.6) and RStudio (v1.3). Multiple comparisons of decidual 271 and placental from PE, preterm and term pregnancies were conducted using the Kruskal-272 Wallis test and followed by pairwise Wilcoxon rank sum tests. The Spearman correlation 273 was used to evaluate the linear relationship between gestational age and given gene. 274

Statistical significance was assumed when p<0.05. 275

The placental expression of ACE2 and TMPRSS2 mRNA was relatively high in the 280 first trimester of pregnancy (7-9 weeks) but exhibited a significant decrease (p<0.05) in 281 samples collected during the second trimester (16-18 weeks), during the third trimester, 282 preterm (26-31 weeks) and at term (Fig, 1A & C) . Spearman correlation analysis, identified 283 a negative correlation between placental ACE2 and TMPRSS2 mRNA expression (p<0.05) 284 with advancing gestational age (Fig, 1, B & D) . 285

We accessed unpublished RNAseq data generated from paired placental and 289 decidual samples in order to investigate the expression of ACE2 and TMPRSS2 in a large 290 cohort of patients experiencing a PTB, or diagnosed with PE, as well as in patients at term 291 undergoing an elective c-section but otherwise uncomplicated term delivery. All samples 292 from the RNAseq cohort exhibited 0-1 ACE2 read count (raw values). Of the 134 samples, 293 only 9 exhibited a single ACE2 read count, when the mean coverage of RNA sequencing 294 ranged around 60 million reads. The remainder of the samples did not exhibit any ACE2 295 mapping out (data not shown). By evaluating the TPM reads counts (Fig. 2 A&B) , we 296 conclude that the expression of ACE2 in decidua and placenta is undetectable by RNA Seq 297 (Fig. 2 C&D) . In contrast, TMPRSS2 was expressed in both decidual and placental samples 298 ( Fig. 2A-B) , although expression of this gene did not change across the patient groups (Fig. 299 3A & B) or with labor status (Fig, 1A & C) . 300 301 Comments 302

In the present study, we determined, for the first time, the mRNA expression of two 304 key SARS-CoV-2 cell entry associated proteins in placentae from first, second and third-305 trimester pregnancies and in pregnancies complicated by PTB or PE. We found that 306 placental expression of the genes that promote SARS-CoV-2 cell entry, ACE2 and 307 TMPRSS2, is downregulated as gestation progresses. Expression of ACE2 or TMPRSS2 at 308 the decidual interface (placenta and decidua) did not change in pregnancies complicated by 309 PTB (irrespective of labor status) or PE. CoV-2 12 . The lower or absence of expression of ACE2 from mid-pregnancy onwards is also 321 consistent with the limited evidence of vertical transmission of SARS-CoV-2 during 322 pregnancy; as Dashraath 17 note, the majority of reports relate to women who acquired 323 SARS-CoV-2 in the third trimester. 324

We did not observe altered placental or decidual expression of ACE2 or TMPRSS2 325 in pregnancies complicated by PTB (whether spontaneous onset or iatrogenic). This is 326 consistent with clinical observations that while rates of preterm birth are increased, this is 327 largely due to PPROM (there were no cases of PROM in our study) or iatrogenic 328 indications 7 . While PE is often associated with placental inflammation 33 , it was not 329 associated with any changes in ACE2 or TMPRSS2 mRNA expression in the decidual or 330 placenta. 331 332

Placental ACE2 mRNA and protein expression have been previously investigated. 334 ACE2 is an important component of the renin-angiotensin system (RAS), where it converts 335 angiotensin II (Ang II) in Ang 1-7, an antagonist of Ang II that acts via Mas G-coupled 336 protein receptor regulation 26,34 . Ang II regulates placental vascular tone and is thought to 337 participate in the pathogenesis of gestational hypertension and PE 35,36 . In the first trimester 338 placenta, ACE2 is abundantly immunolocalized to the syncytiotrophoblast and villous 339 stroma, with lower levels in cytotrophoblasts 26 . This pattern of localization suggests that 340 SARS-CoV-2 present in the maternal circulation has the potential to enter the maternal 341 blood-bathed syncytiotrophoblast and infect the placenta via ACE2 binding. In fact, a case 342 report depicted SARS-CoV-2 particles being predominantly present in the 343 syncytiotrophoblast of a second-trimester pregnancy complicated with PE and placenta 344 abruption 37 , whereas SARS-CoV-2 N-protein immunoreactivity was also concentrated in 345 the syncytial layer in a case of third trimester SARS-CoV-2 vertical transmission 14 . 346

However, and most importantly, ACE2 was not immunolocalized in the fetal vascular 347 endothelium of the villous stroma 26 , which theoretically, could prevent the SARS-CoV-2 348 penetration via ACE2 binding into the fetal circulation. Furthermore, SARS-CoV-2 349 syncytiotrophoblast entry has the potential to induce a potent inflammatory response and 350 functionally disrupt the syncytiotrophoblast barrier, by negatively impacting nutrient and 351 drug transport efficiency, hormonal output and cellular turn over. These possibilities 352 clearly require further investigation. 353 354

It is possible that there are confounding factors associated with the gestational-age 356 differences observed. In our early pregnancy cohort, we were unable (due to institution 357 ethical policies) to collect further clinical information of the elective pregnancy 358

terminations. As such, confounding factors may include maternal body mass index (BMI) 359 status, fetal sex, ethnicity and the presence of unknown maternal infective and or 360 inflammatory states and or endocrine and hypertensive disorders. 361

Using (NGS)/RNAseq, we were able to concomitantly screen the expression profile of 362 ACE2 and TMPRSS2 in a large number of matched decidua and placentae from pregnancies 363 complicated by PTB or PE. We observed very low levels of decidual and placental ACE2 in 364 all groups investigated in later gestation. This is somewhat similar to our findings using 365 qPCR showing detectable but nevertheless lower, placental ACE2 mRNA levels in later 366 stages of pregnancy in healthy patients. The divergent detection of ACE2 mRNA in our 367 cohorts is likely due to the differential sensitivity of techniques. In this context, data 368 extracted from public datasets deposited at ArrayExpress or newly generated scRNA-seq, 369 identified minimal levels of ACE2 and TMPRSS2 in first, second and third trimester 370 placentae 28 . This is in agreement with our second and third trimester findings, but in 371 disagreement with our first trimester results. Differences in the first trimester findings may 372 be attributed to tissue collection protocols, patient inclusion criteria and or differences in 373 the techniques used for gene expression assessment in these studies. Addititional studies 374 are clearly required. 375

Additionally, the interpretation of our data requires caution, because there may be 376

other, yet undefined mechanisms linking SARS-CoV-2 obstetric outcomes in normal and in 377 pathological conditions. Furthermore, future studies should investigate the protein 378 expression pattern, localization and function of TMPRSS2 in the decidua and placenta, as and placenta-specific (B) expressions were calculated by Kruskal-Wallis test, followed by 526 Wilcoxon Rank test. P values were listed when significant differences were detected at 527 p<0.05. n=42 (preeclampsia), n=11 (preterm) and n=14 (term) respectively. 528

Clinical features of patients infected with 2019 405 novel coronavirus in Wuhan

SARS-CoV-2: a storm is raging

COVID-19 and the cardiovascular 409 system: implications for risk assessment, diagnosis, and treatment options

Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-413 19 in the New York City Area. JAMA. 2020

COVID-19-415 associated Acute Hemorrhagic Necrotizing Encephalopathy: CT and MRI 416 Features

Risks associated with viral infections during pregnancy

Coronavirus disease 2019 (COVID-19) in pregnant 420 women: A report based on 116 cases

What are the risks of COVID-19 infection in pregnant women?

Clinical characteristics and intrauterine vertical 424 transmission potential of COVID-19 infection in nine pregnant women: a 425 retrospective review of medical records

Visualization of SARS-CoV-2 virus 427 invading the human placenta using electron microscopy

Second-Trimester Miscarriage in a Pregnant 430 Woman With SARS-CoV-2 Infection

Preterm delivery in pregnant woman with critical COVID-19 pneumonia and 433 vertical transmission. Prenat Diagn

Possible Vertical Transmission of SARS-CoV-2 From 435 an Infected Mother to Her Newborn

Transplacental transmission of 437 SARS-CoV-2 infection

Expert consensus for managing pregnant women 439 and neonates born to mothers with suspected or confirmed novel coronavirus 440 (COVID-19) infection

Pregnancy and perinatal outcomes of 442 women with severe acute respiratory syndrome

Coronavirus disease 2019 (COVID-19) 445 pandemic and pregnancy

Middle East Respiratory Syndrome 447

MERS-CoV) infection during pregnancy: Report of two cases & 448 review of the literature

A case-controlled study comparing clinical 450 course and outcomes of pregnant and non-pregnant women with severe acute 451 respiratory syndrome

Risks of Novel Coronavirus Disease (COVID-19) in 453 Pregnancy; a Narrative Review

COVID-19 in pregnancy was 455 associated with maternal morbidity and preterm birth

SARS-CoV-2 Cell Entry 458 Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease 459 Inhibitor

Structural and Functional Basis of SARS-CoV-2 461 Entry by Using Human ACE2

A pneumonia outbreak associated with a new 463 coronavirus of probable bat origin

The pivotal link between ACE2 465 deficiency and SARS-CoV-2 infection

The expression and 467 localization of the human placental prorenin/renin-angiotensin system 468 throughout pregnancy: roles in trophoblast invasion and angiogenesis? 469 Placenta

The SARS-CoV-2 receptor ACE2 expression 471 of maternal-fetal interface and fetal organs by single-cell transcriptome study. 472 PLoS One

Gestational age-dependent gene expression 476 profiling of ATP-binding cassette transporters in the healthy human placenta

Effect of oxygen on multidrug resistance in the 479 first trimester human placenta

Breast Cancer Resistance Protein 481 (BCRP/ABCG2) Inhibits Extra Villous Trophoblast Migration: The Impact of 482 Bacterial and Viral Infection

Analysis of relative gene expression data using real-484 time quantitative PCR and the 2(-Delta Delta C(T)) Method

Cross-Talk between Oxidative Stress and Inflammation in Preeclampsia

-7) counterregulates angiotensin II signaling in human 491 endothelial cells

A novel mechanism of angiotensin II-regulated 493 placental vascular tone in the development of hypertension in preeclampsia

The relevance of the Renin-Angiotensin 496 system in the development of drugs to combat preeclampsia

SARS-CoV-2 Infection of the Placenta

Revisiting the housekeeping genes of human 501 placental development and insufficiency syndromes