Single‐cell RNA sequencing reveals transcriptional profiles of monocytes in HBV‐infected pregnant women during mid‐pregnancy

Abstract There is a growing body of evidence that innate immunity also plays an important role in the progression of hepatitis B virus (HBV) infection. However, there is less study on systematically elucidating the characteristics of innate immunity in HBV‐infected pregnant women. We compared the features of peripheral blood mononuclear cells in three healthy pregnant women and three HBV‐infected pregnant women by single‐cell RNA sequencing. 10 DEGs were detected between groups and monocytes were the main expression source of most of the DEGs, which involved in the inflammatory response, apoptosis and immune regulation. Meanwhile, qPCR and ELISA were performed to verify above genes. Monocytes displayed immune response defect, reflecting poor ability of response to IFN. In addition, eight clusters were identified in monocytes. We identified molecular drivers in monocytes subpopulations.TNFSF10+ monocytes, MT1G+ monocytes and TUBB1+ monocytes were featured with different gene expression pattern and biological function.TNFSF10+ monocytes and MT1G+ monocytes were characterized by high levels of inflammation response.TNFSF10+ monocytes, MT1G+ monocytes and TUBB1+ monocytes showed decreased response to IFN. Our results dissects alterations in monocytes related to the immune response of HBV‐infected pregnant women and provides a rich resource for fully understanding immunopathogenesis and developing effective preventing HBV intrauterine infection strategies.

hepatitis B immunoglobulin can block transmission substantially, about 5%-10% of newborns are still infected with HBV, possibly via intrauterine infection. 4,5 The progression of HBV infection is closely associated with the immune response. 6 There is increasing evidence that innate immunity also plays an important role in the progression of HBV infection. 6,7 Therefore, it is very important to study innate immunity in HBVinfected pregnant women to guide the development of therapies aimed at blocking intrauterine transmission. Monocytes are an important part of innate immunity. After HBV infection, activated monocytes produce and secrete a large number of inflammatory cytokines and chemokines to recruit inflammatory cells and clear target cells in an active and synergistic manner. Antigens are presented to T cells by MHC II molecules, resulting in the activation and proliferation of T cells, triggering an antiviral response. 6,8 During chronic HBV infection, cytokines play a critical role in immune regulation and inflammation.
They inhibit viral replication and influence the persistence of HBV infection. Interferons α, β and γ have essential roles in the innate immune response against CHB. Chronic HBV infection suppresses the production of IFN and further reduces cellular responses to IFN, affecting the activation of other cellular pathways and mechanisms. 6,9,10 However, few studies have evaluated the functional association between HBV infection in pregnant women and monocytes. In addition, activated monocytes cannot be effectively used as therapeutic targets, mainly due to their substantial heterogeneity and complex functions, requiring the accurate isolation of monocyte subsets and localisation of activated or dysfunctional subsets. 11 With the development of single-cell RNA sequencing technology, it has become possible to accurately isolate different monocyte subpopulations and reveal their functional differences. [12][13][14] To evaluate differences in immune function between HBVinfected pregnant women and healthy pregnant women and to guide the development of therapies blocking mother-to-child transmission, we used single-cell RNA sequencing to obtain the transcriptional profiles of peripheral blood mononuclear cells (PBMCs) from three HBV-infected pregnant women and three healthy pregnant women, with a focus on alterations in monocytes. Our results showed that innate immune response was weakened in HBV-infected pregnant women. A functional enrichment analysis revealed monocyte subclusters that played a role in HBV-infected pregnant women. These findings improve our understanding of the dynamics of monocyte subsets in HBV-infected pregnant women and provide therapeutic targets for reducing mother-to-child transmission of HBV.

| Human subjects
Peripheral blood samples were collected from pregnant women in their second trimester who were admitted to the Department of Obstetrics

| Isolation of peripheral blood mononuclear cells
Five millilitres of peripheral blood from all subjects were collected into EDTA anticoagulated tubes. PBMCs were separated using Ficoll-Paque PREMIUM 1.084 g/L sterile solution (GE Healthcare, Uppsala, Sweden) by density gradient centrifugation. Then, 90% FBS + 10% DMSO was added to the cryopreservation tube, followed by storage at −80°C in a refrigerator.

| Single-cell RNA sequencing
Single-cell RNA-seq libraries were prepared using Chromium Single Cell 30 Reagent v2 Kits according to the manufacturer's protocol. PBMC suspensions were loaded on the Chromium Single Cell Controller Instrument (10× Genomics) to generate single-cell gel beads in emulsions (GEMs). About 15,000-20,000 cells were added to each channel with a targeted cell recovery of 10,000 cells.
Captured cells were lysed, and the released RNAs were barcoded by reverse transcription in individual GEMs. Then, these libraries were sequenced on the Illumina sequencing platform (HiSeq X Ten), and 150 bp paired-end reads were generated.

| Single-cell RNA-seq data preprocessing
The Cell Ranger pipeline (version 3.1.0) provided by 10× Genomics was used to demultiplex cellular barcodes, map reads to the genome and transcriptome using the STAR aligner and down-sample reads as required to generate normalized aggregate data across samples, producing a matrix of gene counts versus cells. The unique molecular identifier (UMI) count matrix was obtained using the R package Seurat 15 (version 3.0). To remove low-quality cells and likely multiplet captures, which are a major concern in microdroplet-based experiments, we filtered out cells with UMI/gene numbers >±2 times the standard deviation of the mean, assuming a Gaussian distribution of values. Following the visual inspection of the distribution of cells by the fraction of mitochondrial genes expressed, we further discarded low-quality cells in which a certain percentage of counts belonged to mitochondrial genes. Library size normalisation was performed using Seurat with the filtered matrix to obtain the normalized counts. The most highly variable genes across single cells were identified using the method described in Macosko et al. 16 Briefly, the average expression and dispersion were calculated for each gene, and genes were placed into several bins based on expression. A principal component analysis (PCA) was performed to reduce the dimensionality of the log-transformed gene-barcode matrices of highly variable genes. Cells were clustered based on a graph-based clustering approach and were visualized in two-dimensional space using tSNE. The likelihood ratio to simultaneously evaluate changes in mean expression and in the percentage of cells with expression was used to identify significant DEGs between clusters. Here, the R package SingleR, 17 a novel computational method for unbiased cell type recognition based on scRNA-seq data, was used to independently infer the cell of origin of single cells and identify cell types.

| Differentially expressed genes
DEGs were identified using the Seurat package. 18 p < 0.05 and |log-2fold change| > 1.5 (or |log2fold change| < 0.67) were set as the thresholds for significant differential expression. GO and KEGG pathway enrichment analyses of DEGs were performed using R based on the hypergeometric distribution.

| GO analysis and KEGG pathway enrichment analysis
The Cluster profiler 19 of R Package was used to analyse the biological functional genes of the top 100 DEGs with the largest expression in each cluster or cell type, and the biological functional gene set was the focus of the study. Pathways with significant enrichment were selected according to p value, and the significant value of p was 0.05.

| Gene set variation analysis
GSVA (Version 1.30.0) 20 was used for differential pathway analysis.
To evaluate the activity of gene sets among clusters, the average expression and log2fold change of gene sets for each cluster were first determined. Then the LIMMA package (Version 3.38.3) was used to compare the pathway activity between the two groups. p < 0.05 was set as the threshold for significance.

| Cell stimulation
Monocytes were plated in 12-well plates (Costar, Corning Incorporated, Corning, NY, USA) at 5 × 10 5 cells/well, unless indicated otherwise. Cells were stimulated with normal maternal serum and hepatitis B maternal serum for 48 h at 37°C. Cells were harvested for RNA isolation and qPCR.

| qPCR
Total mRNA was isolated from cells using the RNA Fast 200 Kit The primers are shown in Table S2. The reaction volume was set to 20 μL and conditions were as follows: 95°C for 10 min, followed by 40 cycles of 15 s at 95°C, 30 s at 60°C and 30 s at 72°C. Data were analysed by the 2 −ΔΔCt method, and relative mRNA expression levels are reported as fold differences.

| Enzyme-linked immunosorbent assay
IL-1β concentrations in the cell culture supernatant and maternal serum were measured using commercial enzyme-linked immunosorbent assay (ELISA) kits for IL-1β (Hengyuan, Shanghai, China; HS1973-Hu), according to the manufacturer's instructions. The lower limit of detection was 1 pg/mL.

| Statistical analysis
Statistical analyses were performed using GraphPad Prism 8.0 (La Jolla, CA, USA). Data are shown as means ± standard error of the mean. Unpaired Student's t-tests were used to compare means between groups. For all analyses, p < 0.05 was considered significant.

| Single cell atlas of PBMCs in HBV-infected pregnant women and healthy pregnant women
We collected fresh PBMCs derived from three HBV-infected pregnant women (CASE group) and three healthy (CON group) pregnant women according to the inclusion and exclusion criteria. The gestational week of the HBV-infected pregnant women was between 22 and 23 weeks, and all of them were at the stage of HBeAg-positive chronic infection. No antiviral therapy were used before pregnancy and before blood sampling in this study (Table S1). After single-cell RNA sequencing and aggregating all sample data using Cell Ranger, 15 Figure 1A). We showed heat maps of the top 10 specifically expressed genes for each cluster and five cell types ( Figure 1B,C). By further comparison, we observed that there was no significant difference in the proportions of these five cell types between the two groups, even though monocytes and pDc slightly increased in case group ( Figure 1D,E).

| Differentially expressed genes in PBMCs were enriched in monocytes
To explore the differences between different groups, we used DEGs method to analyse the enrichment of PBMC differential genes in HBV-infected pregnant women and healthy pregnant women. The results showed that there were 10 differentially expressed genes IL-1β, a protein encoded by IL1B, is a proinflammatory factor secreted by monocytes, which is involved in inflammatory response against HBV. We detected IL-1β protein levels in the cell culture supernatant. Quantitative analysis showed a significant increasing expression in the case group compared with the control group (p < 0.05). Furthermore, levels of IL-1β in serum was also detected in two groups. IL-1β levels were significantly higher in HBV-infected women than in healthy pregnant women (p < 0.05; Figure 2E).

| Functional defects in monocytes of HBVinfected pregnant women
Since there was not differences of the cell frequency of monocytes between groups, we further analysed the transcriptional alteration in monocytes. A differential expression analysis and GO analysis of monocytes between HBV-infected pregnant women and healthy pregnant women were performed and only nine DEGs were identified. Compared to control group, the upregulated DEGs in monocytes were enriched in biological processes in case group, such as the negative regulation of growth, cellular zinc ion homeostasis and immune regulation. More importantly, functions related to the immune system and innate immune response were downregulated ( Figure 3A). It is necessary to investigate which aspects function of monocytes existed changes.
Next, we compared the expression levels of antiviral interferonstimulated genes 13 in monocytes.IFITM1, IFITM2, IFITM3, IFI6, IRF1, BST2 and PSMB8 were clearly lower than those in monocytes in HBV-infected pregnant women, suggesting that the antiviral activity of monocytes from HBV-infected pregnant women was weakened ( Figure 3B).
Our results indicated that, monoctyes of HBV-infected pregnant women showed enhanced ability of immune regulation, whereas the immune response was defective, reflecting in the response to interferon and to virus defence, which may be involved in the mechanism of HBV intrauterine infection.

| Heterogeneity of monocyte clusters
However, it was difficult to deeply understand the changes of innate immunity in HBV-infected pregnant women at the overall level of monocytes, which might be related to the heterogeneity of monocytes. Next, we performed dimensionality reduction clustering analysis on monocytes to identify specific transcriptional signatures. We In addition, a one-to-many GSVA was then performed to reveal the characteristics of pathways enriched in immune responses in each cluster ( Figure 4H). As for cytokine-cytokine receptor signalling pathway, Cluster 5 showed significant enrichment, whereas cluster 2 and cluster 7 showed lower expression. Therefore, we further focused on these three clusters to excavate the alterations of monocytes of HBV-infected pregnant women.

| Functional changes of monocyte subsets in HBV-infected pregnant women
To estimate biologic processes to dysregulated genes in monocyte subsets, we evaluated DEGs between HBV-infected and healthy pregnant women within each specific subpopulation. An analysis of DEGs between the two groups suggested that IL1B and IER3 were highly expressed only in monocytes, while metallothionein gene expression levels were relatively higher in monocytes. Thus, we propose monocytes play an important role in HBV-infected pregnant women with a high viral load. IL1B encodes the cytokine IL-1β, which is produced by monocytes in response to infection and is closely related to the innate immune response and immune regulation. 28 After HBV infection, monocytes are activated and secrete the inflammatory mediator IL-1β, enhancing the inflammatory response. The mRNA and protein levels of IL-1β were higher in patients with chronic hepatitis B than in healthy controls, and IL-1β levels were positively correlated with disease severity. 29 Our results were consistent with previous results, and we also verified that IL1B genes not only participate in the detoxification of heavy metals, regulation of zinc homeostasis and scavenging of free radicals, but also play important roles in immune regulation and liver diseases. 33,34 The expression level of MT is correlated with disease progression in chronic hepatitis B 35 with augment in early hepatitis B and decrease in advanced hepatitis B, cirrhosis and liver cancer. The elevated expression levels of MT1 and MT2 suggested that these genes may be related to immune activation and innate responses of monocytes to HBV infection. Compared with other white blood cells, monocytes showed the highest MT expression at the protein and mRNA levels and the ability to bind to metal zinc. The anti-apoptotic phenotype of monocytes is associated with an elevation in MT in HIV. 36 The results of our study showed that metallothionein was expressed at relatively high levels in monocytes of HBV-infected pregnant women, including high expression levels in almost every monocyte cluster.
The expression levels of MT1X, MT2A and MT1E were also increased in the monocyte line THP-1 stimulated by serum of HBV-infected pregnant women. We speculated that monocytes may induce MT gene expression during persistent HBV infection, and MT overexpression in monocytes may be involved in resistance to apoptosis.
Compared with healthy pregnant women, a function enrichment analysis showed that monocytes show immune response defects, as evidenced by a decreased ability to respond to interferon. In healthy pregnancy, monocytes exhibits anti-inflammatory activity. 37 On the contrary, our results showed many genes related to 'cytokine activity', pregnancy. 26 In this study, TUBB1+ monocytes enriched mRNA splicing pathway. We hypothesized that mRNA splicing pathway may play a part role in inhibition of monocytes. Hence, restoring the biological functions of these three clusters is a potential strategy to enhance the antiviral capacity in HBV-infected pregnant women and may reduce the risk of vertical transmission of HBV from mother to child.
There are some limitations in this study. On one hand, it is necessary to augment the sample size, which can better support our research results. On the other hand, this study is mostly descriptive analysis, which shows alteration of transcription profiles and biological processes of monocytes in HBV-infected pregnant women.
Data mining of single-cell RNA sequencing is insufficient, and further research on the mechanism of innate immune response is still needed. In the next step, the specific mechanism will be verified in monocytes of HBV-infected pregnant women by experimental methods based on differential expressed genes and molecular pathways found in this study, in order to deepen the understanding of the immune changes of HBV-infected pregnant women. The interaction between monocytes and other cells in PBMC will also be the direction of our further research.
In summary, our study reveals the transcriptional profiles of monocytes in HBV-infected pregnant women during the middle trimester of pregnancy and expands our understanding of the alterations of monocytes subsets. Furthermore, we will deeply explore interaction between monocyte clusters and other immune cells to uncover underlying mechanism of monocytes in HBV pregnant women.

ACK N O WLE D G E M ENTS
We thank all of the patients and healthy controls for participating in the study and for permitting us to use their blood for this research.
We thank the Translational Medicine Center of the First Affiliated Hospital of Xi'an Jiaotong University in China for providing the laboratory. This study was funded by The National Natural Science Fund, China (81771615).

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The single-cell RNA sequencing data generated in this study can be found in the SRA database [SRA accession: PRJNA645249].