Based on placental cell gene expression data and computer simulations, Brazilian researchers concluded that the virus may use the placental proteins DPP4 and CTSL as an alternative point of entry into human cells (image: Norman Barker / Johns Hopkins School of Medicine)

Published on 03/19/2021

By Karina Toledo  |  Agência FAPESP – Researchers at São Paulo State University (UNESP) in Brazil have identified two highly expressed genes in human placenta that encode proteins that may serve as an “alternative” mechanism for the novel coronavirus SARS-CoV-2 to invade human cells: dipeptidyl peptidase 4 (DPP4) and cathepsin L (CTSL).

The study was supported by FAPESP and is reported in an article available as a preprint (not yet peer-reviewed) on bioRxiv.

As the authors explain in the article, throughout pregnancy few cells in placenta express angiotensin-converting enzyme 2 (ACE-2) and transmembrane protease serine 2 (TMPRSS-2), the main molecules used by SARS-CoV-2 to infect human cells. Nevertheless, studies by other groups have confirmed the presence of the virus in placenta, using both transmission electron microscopy and RT-qPCR (quantitative reverse transcription polymerase chain reaction), the gold standard laboratory test to diagnose COVID-19 by identifying the virus’s RNA. The virus has also been detected in newborns, although whether it was transmitted in the womb or during delivery is unknown. 

“With this evidence as our starting-point we decided to investigate the possible existence of an alternative mechanism for infection of placental cells, albeit less efficient than that mediated by ACE-2. The presence of SARS-CoV-2 in placenta could lead to intrauterine transmission of the virus and the associated impairment of fetal development,” Luis Antonio Justulin, a professor in the university’s Botucatu Institute of Biosciences (IBB-UNESP).

In IBB-UNESP’s Department of Structural and Functional Biology, Justulin is leading a project to investigate how different conditions experienced in the womb and in early childhood can affect lifelong health, in accordance with the concept of the developmental origins of health and disease (DOHaD). For example, the group have shown experimentally that the offspring of female rats fed a low-protein diet during pregnancy and lactation are significantly more likely to develop prostate cancer as they age (read more at agencia.fapesp.br/28399). In the case of SARS-CoV-2, there have so far been no reports of newborns with complications due to maternal infection, but studies have shown that the virus can cause vascular alterations in placenta which could theoretically impair fetal uptake of oxygen and nutrients.

To achieve a better understanding of how the virus interacts with placental cells, the group led by Justulin decided to analyze gene expression data for different types of placental cells at different moments during gestation. They used data produced by previous studies, available from public repositories such as the Gene Expression Omnibus (GEO), and cross-referenced their findings for placenta with data from P-HIPSTer (Pathogen-Host Interactome Prediction using STructurE similaRity). 

This public database uses an algorithm that exploits both sequence- and structure-based information to infer interactions between viral and human proteins. The resulting “human-virus interactome atlas” maps possible networks of interactions between human proteins and proteins in several viruses to help identify potential therapeutic targets.

This strategy enabled the researchers to find out which genes expressed during pregnancy encode proteins with the most potential to interact with the main proteins in a range of coronaviruses, including SARS-CoV-2, SARS-CoV (which caused the 2002-04 SARS outbreak) and MERS-CoV (which causes Middle East respiratory syndrome).

“We performed two kinds of analysis,” Justulin said. “First we looked at the interactions between viral proteins and all placental proteins. Then we examined these interactions considering which genes are expressed to a greater or lesser extent during pregnancy.” 

To find out specifically which placental cells expressed the genes for the proteins believed to interact with the virus, the group used datasets from two other public databases, COVID-19 Cell Atlas and Human Cell Atlas Data Portal. Both store single-cell RNA sequencing data from which scientists can obtain the transcriptome of each cell in a tissue sample instead of having to analyze the sample as a whole. 

This part of the analysis used bioinformatics and was performed in collaboration with Robson Carvalho (IBB-UNESP) and Célia Regina Nogueira, researchers in the Botucatu Medical School’s Department of Clinical Medicine, and IBB-UNESP PhD candidate Sarah Santiloni Cury.

Main findings

The analyses and simulations showed that trophoblasts in the chorionic villi – the fetal part of the placenta – co-express large amounts of DPP4 and CTSL throughout the gestational period. 

“DPP4 has been linked to infection by MERS-CoV in previous research, and CTSL is reportedly associated with SARS-CoV,” Justulin said. “We analyzed several samples of fetal tissue and found high levels of these two proteins expressed in the fetus as well.”

The study also showed that expression of the genes DAAM1 and PAICS changes significantly during pregnancy (rising and falling respectively in successive trimesters). Both encode proteins that can interact with viral proteins according to the computer simulations performed at IBB-UNESP.

According to the authors, all this is initial evidence of the existence of a non-canonical route for infection of placental cells by SARS-CoV-2, requiring confirmation and further investigation in future research. For example, the group plan to analyze gene expression in the placenta of pregnant COVID-19 patients.

“We also plan to infect placental cells in vitro and see what happens in terms of gene expression,” Justulin said. “This type of experiment requires a biosafety III laboratory and we’ll need to partner with a group in an institution that has such a facility.”

The study described in the paper posted to bioRxiv was also supported by FAPESP via a research grant awarded to Carvalho and a PhD scholarship awarded to Flávia Bessi Constantino, first author of the paper.

“At this difficult time for the world it’s gratifying to be able to contribute in any way to our understanding of these viral infection mechanisms,” Constantino said. “Because our research line focuses on the initial stages of development we were intrigued by the possibility that the virus infects placenta and causes complications for development of the embryo. That’s how the idea of doing the study came up.”

The article “Prediction of non-canonical routes for SARS-CoV-2 infection in human placenta cells” can be read at: www.biorxiv.org/content/10.1101/2020.06.12.148411v1.full.pdf.

 

Source: https://agencia.fapesp.br/33564