Mice Immunization

The aim of the first experiment was to determine the effect of oral LPG administration on mouse humoral immune responses to COVID-19 when given 6 months after immunization. All ICR mice (n = 10) received intramuscular vaccine injections at weeks 0 and 4 post priming, and sera were collected between weeks 5 and 24 post priming for antibody assessment by enzyme-linked immunosorbent assays (ELISAs) and pseudovirus inhibition assays. At 24 weeks post vaccination, the mice were given water containing 1 g/L of ampicillin for 5 days to prevent colonization resistance (5), then they were randomly divided into two groups that received oral administration of either 200 μL of LPG (5 × 10⁹ CFU, n = 6) or phosphate-buffered saline (PBS, n = 4) once daily for 3 days. The antibody titers measured immediately before ampicillin administration were set as the baseline to be compared with measurements taken at various time points post LPG or PBS administration. Sera were collected on day 7, 14, 28, and 42 post oral administration of LPG or PBS for receptor-binding domain (RBD) binding antibody assessment by ELISA and neutralization antibody (nAb) quantification by pseudovirus inhibition assay.

The second experiment was conducted to determine the effect of oral LPG administration on humoral and cellular immune responses in mice when given immediately after immunization. All ICR mice were primed with an intramuscular injection of 100 μg of DNA-S at week 0. At week 12 after priming, the mice were given drinking water containing 1 g/L ampicillin for 5 days then boosted with 1 × 10¹¹ viral particles (vp) of AdC68-CoV/Flu via both intramuscular and intranasal routes. The mice were then randomly divided into two groups (n = 9/group) for oral administration of either 200 μL of LPG (5 × 10⁹ CFU) or PBS once daily for 3 days. The antibody titers measured immediately before ampicillin administration were set as the baseline to be compared with measurements taken at various time points post LPG or PBS administration. The mice (n = 3/group) were sacrificed on day 3, 7, 14 post administration of LPG or PBS. Sera and bronchoalveolar lavage (BAL) were collected on baseline and day 7 post administration of LPG or PBS for RBD binding antibody and nAb assessment. To assess RBD-specific T cell responses, splenocytes and BAL cells were isolated on day 3, 7, and 14 post oral administration of LPG or PBS and in vitro stimulated with 13 peptide pools (15-mer with 11 overlapped amino acids) covering the entire RBD sequence. The resulting IFN-γ secreting cells were quantified by ELISpot. Interleukin (IL)-6 and tumor necrosis factor (TNF)-α in BAL from mice sacrificed on day 3, 7, and 14 were quantified using Th1/2/17 Assay Kit (Cat#560485, BD Bioscience, Franklin Lakes, NJ, USA).

Oral LPG Administration of Mice for Transcriptional Analyses

To examine how LPG imprints the immune system. Two groups of ICR mice (20 mice per group) were given 1 g/L of ampicillin for 5 days followed by oral administration of 200 μL of LPG (5 × 10⁹ CFU) or PBS once daily for 3 days. Different tissues including spleen, mediastinal lymph node (LN), mesenteric LN, small intestine, and colon tissues were collected and stored in TRIzol at different time points after oral administration for RNA-sequencing.

Bacterial Culture

LPG isolated from the feces of healthy people was routinely grown in Man-Rogosa-Sharpe medium in a 37°C carbon dioxide incubator (Forma CO2, Thermo Fisher Scientific, Waltham, MA, USA). The logarithmic phase of LPG was washed and resuspended with sterile PBS for oral inoculation of mice.

Cell Lines and Viruses

The human cell lines used in this study included HEK 293A and HEK 293T. All cell lines were purchased from American Type Culture Collection (Manassas, VA, USA) and maintained in complete Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (PS) (D10) at 37°C in a 5% CO₂ incubator. The generation of wild-type SARS-CoV-2 pseudovirus was described in our previous publications (6, 7). In brief, HEK 293T cells were co-transfected with pNL4-3.Luc.R-E- (cat#3418, NIH AIDS Reagent Program) and pcDNA3 plasmid encoding SARS-CoV2-S using TurboFect reagent (cat#R0531, Thermo Fisher Scientific). At 24 h post transfection, the culture medium was refreshed with D10 medium for an additional 48-h incubation. The pseudovirus-containing supernatants were harvested by centrifugation then filtered through 0.45-μm filters and stored at −80°C as single-use aliquots. For each pseudovirus stock, a titration assay was performed using an aliquot to determine the dilution needed for the neutralization assay.

DNA Vaccine Construction

The DNA-S is a DNA vaccine candidate for which the codon-optimized DNA sequences encoding SARS-CoV-2 spike (S) protein (YP_009724390.1) were synthesized (Generay Biotech Co., Ltd., Shanghai, China), and cloned into the pcDNA3.1 vector to generate the pcDNA3.1-S plasmid.

Construction, Rescue, Propagation, and Purification of Recombinant Chimpanzee Adenoviruses

The AdC68-CoV/Flu is a chimpanzee adenoviral vaccine with a hemagglutinin stalk derived from influenza A/Shanghai/4664T/2013 strain and the RBD of wild-type SARS-CoV-2, which were synthesized and cloned into the E1 site of the recombinant AdC68 vector. The recombinant AdC68 is different from the wild-type AdC68 in that both the E1 and E3 genes are deleted. The E1 locus is essential for viral replication, so recombinant AdC68 viruses can only be produced in cell lines engineered to supply the E1 protein, such as HEK 293A. HEK 293A cells were seeded in 6-well plates for 24 h then were transfected with a PacI linearized corresponding AdC68 construct using Lipofectamine 2000 following the manufacturer's instruction (Invitrogen, Carlsbad, CA, USA). The rescued viruses were propagated in HEK 293A cells and purified using CsCl gradient ultracentrifugation in combination with desalting with Bio-Gel P-6 DG Media (Bio-Rad, Hercules, CA, USA).

RBD Binding Antibody Detection Using ELISA

RBD binding antibody titers in serum and BAL samples were measured by ELISA. In brief, sera and BAL samples were heat-inactivated at 56°C for 30 min before use. Next, 96-well ELISA plates were coated with 100 μL of 1 mg/mL recombinant SARS-CoV-2 RBD protein (Z03483-1, Genscript, Piscataway, NJ, USA) at 4°C overnight. After washing the plates with 300 μL PBS containing 0.5% Tween-20 (PBS-T) and blocking with 200 μL 5% non-fat milk in PBST (PBST/5% milk) for 2 h at room temperature (RT), a 2-fold dilution series (generally starting from 1:100) of mouse serum and BAL samples were added, followed by 3-h incubation at RT. For immunoglobulin G (IgG) measurement, a 1:5,000 dilution of horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (ZB-5305, Zsbio, Beijing, China) was applied in 100 μL PBST/5% milk. After 1 h incubation at RT, the plates were extensively washed with 300 μL PBS-T before addition of the substrate OPD (one SIGMAFAST OPD tablet [SLCC0308, Sigma, St. Louis, MO, USA] in 20 mL of deionized water). The reactions lasted for 5 min at RT and were terminated by adding 1M H₂SO₄, followed by reading at OD492 with a Synergy Microplate Reader (Bio-Tek, Winooski, VT, USA). ELISA endpoint titers were defined as the highest dilution that yielded an absorbance that was 2-fold greater than the background value.

Total IgG ELISA

Total serum IgG levels were measured using the IgG (Total) Mouse Uncoated ELISA kit (Cat# 88-50400, Thermo Fisher Scientific). Each sample was diluted 500-fold in assay buffer and run in duplicate with Southern Biotech TMB Stop Solution (Cat#0412-01) as the stop solution. Optical density values were measured at 450 nm and 570 nm on a VersaMax plate reader (Molecular Devices, Sunnyvale, CA, USA) and corrected by subtracting the measurement at 570 nm from the measurement at 450 nm. A four-parameter standard curve was used to calculate sample concentration values.

Pseudovirus Neutralization Assay

The protocol for SARS-CoV-2 pseudovirus generation was described previously (7). For assessment of serum neutralizing activity against SARS-CoV-2, serial 50-μL dilutions of heat-inactivated serum were made with D10 medium and mixed with equal volumes of diluted pseudovirus. After incubation at 37°C for 1 h, the serum-pseudovirus mixtures were transferred to a 96-well plate containing hACE2-293T cells, which were pre-seeded at 2 × 10⁴ cells per well and allowed to grow for 12 h. The plates were incubated for 48 h at 37°C in the presence of 5% CO₂, and luciferase activities were then measured using Bright-Glo™ Luciferase Assay System (Promega, Madison, WI, USA) on a luminometer (Promega GloMax 96). The neutralizing (50% inhibitory doses) titers were derived from the highest dilution resulting in a 50% reduction in relative light units (RLUs) relative to virus control wells after background (RLUs of no virus wells) subtraction.

IFN-γ Based ELISpot Assay

T cell responses were analyzed by using single-cell suspension of splenocytes and BAL by mouse IFN-γ ELISpot assay set (Cat#551083, BD Bioscience) following the manufacturer's protocol. In brief, 96-well ELISpot plates were pre-coated with anti-mouse IFN-γ antibodies (5 μg/mL, 100 μL) overnight at 4°C. After rinsing with RPMI-1640 medium (10–040-CVR, Corning, Corning, NY, USA) containing 10% FBS (BI, 04–001-1acs) and 1% PS (30–002-CI, Corning) (R10), plates were blocked with R10 for 2 h at RT before adding to each well 2 × 10⁵ of splenocytes or 2 × 10⁴ of BAL cells, followed by stimulation with the indicated 15-mer peptide(s) or pool(s). The peptides used for the stimulation work were synthesized in ChinaPeptides Co., Ltd. Each assay was performed in duplicate. Peptide stimulation lasted for 20 h after plates were placed in a humidified 5% CO₂ incubator at 37°C. The plates were washed before biotinylated anti-mouse IFN-γ antibody (2 μg/mL, 100 μL) was added for 2-h RT incubation. Next, streptavidin-conjugated HRP was added at 1:100 dilution and incubated at RT for 1 h. The plates were washed again and subjected to spot development with AEC substrate reagent (Cat#551951, BD Bioscience). The reaction was stopped by water rinsing, and the plates subsequently allowed to dry for 24 h in the dark. Plate images were captured with a Biospot plate reader (ChampSpot III, Beijing SageCreation Science Co., Ltd., Beijing, China) and analyzed for spot-forming cell counts. Quality control was implemented on the automatic counting data by removing pseudo-spots, such as those resulted from pollutants and filaments and other residues. R10 was used as a negative control and all the data were adjusted by subtracting the value of negative control, 50 SFCs/million was considered as cut off value.

BAL Cytokines

Flow cytometry assessments of serum levels of IL-6 and TNF-α were carried out using the BD Mouse Cytometric Bead Array kit for Th1/Th2/Th17 cytokine panel (Cat#560485, BD Bioscience).

RNA-Sequencing and Quantitative Real-Time Polymerase Chain Reaction

RNA-sequencing was performed to examine how LPG imprints the immune system. Total RNA of different tissues were extracted using the TRIzol agent following the manufacturer's instruction, and total RNA quality was determined using the Agilent 2100 (Agilent Technologies, Santa Clara, CA, USA). A total of 5 ng purified RNA was used for library construction with an Illumina® TruSeq® RNA Sample Preparation Kit v2 (Cat#RS-122-2001, Illumina, San Diego, CA, USA), and the library was sequenced on Illumina HiSeqTM2500 by Shanghai South Gene Technology Co., Ltd. (Shanghai, China). The clean data were obtained by removing adapter-containing reads and low-quality reads by FASTX-Toolkit (http://hannonlab.cshl.edu/fastx_toolkit/). These clean reads were mapped to the reference genes using Bowtie2 software and to the reference genome using HISAT software. Counts were converted to fragments per kilobase of exon per million mapped fragments (FPKM) values for normalization. Gene expression data were subjected to set enrichment analysis using gene set enrichment analysis (GSEA) and hierarchical cluster analysis with Multiexperiment Viewer (MeV 4.9).

Oral LPG Administration Immediately After SARS-CoV-2 Vaccination Significantly Boosts Both Effective Humoral and Cellular Responses Against COVID-19 in Mice

We next sought to determine whether LPG was capable of boosting vaccine-induced effective immune responses in the SARS-CoV-2 vaccination model. ICR mice were primed and boosted with vaccines followed by oral administration of LPG or PBS as described in detail in the Methods (Figure 2A). As shown in Figure 2B, serum RBD binding antibody titers at baseline (before ampicillin treatment) were comparable in the PBS and LPG groups with a geometric mean titer (GMT) of 9,406 (left panel). On day 7 after boost, the titers increased to GMT 129,016 in the PBS group, and a further 6.3-fold increase to GMT 819,200 was observed in the LPG group (P < 0.001; right panel). Similarly, a 2.3-fold promotion of nAb responses was observed in the LPG group compared with the PBS group, although significance was not reached (Figure 2C, right panel). We further assessed binding antibody and nAb titers in BAL collected on day 7 post oral administration. Binding antibody titers in the LPG group (GMT = 3,225) were ~10-fold higher than in the PBS group (GMT = 320; Figure 2D), as were nAb titers in the LPG group (GMT = 241) vs. the PBS group (GMT = 27; Figure 2E). We next examined RBD-specific T cell responses in the spleen and in BAL isolated from immunized mice on day 3, 7, and 14 post oral LPG administration by ELISpot assay, which captures IFN-γ after stimulation with pools of RBD-derived peptides. A total of 13 RBD peptide pools were used to fully cover the RBD sequence, each comprising five peptides (15-amino acid) that overlap by 11 amino acids. Although the boost vaccination induced similar T-cell responses in the LPG and PBS groups (data shown as day 3), they rapidly decreased in the PBS group and remained relatively stable in the LPG group (Figures 2F,G and Supplementary Figures 3A,B). Collectively, these results suggest that oral LPG administration boosts effective humoral and cellular responses against SARS-CoV-2 in vaccinated mice.

Open in a separate window

Figure 2

Oral LPG administration immediately after immunization improved humoral and cellular immune responses in mice. Experimental schedule (A). ICR mice were primed and boosted with vaccines followed by oral administration of LPG or PBS as described in the Methods. Serum antibody responses were assessed at baseline (immediately before antibiotic treatment) and day 7 post intragastric administration (D7) for RBD-specific binding antibodies by ELISA (B) and pseudovirus neutralization assays (C). BAL antibody responses were assessed on day 7 post intragastric administration for RBD-specific binding antibodies by ELISA (D) and pseudovirus neutralization assays (E). Assessments of RBD-specific T cell responses. T-cell responses in splenocytes (F) and BAL cells (G) were determined on day 3, 7, and 14 post intragastric administration. BAL: Bronchoalveolar lavage; D: Day; W: Week. Titer data are presented as geometric mean titer ± geometric standard deviation; ELISpot counts were expressed as mean ± s.e.m. Mann-Whitney tests were performed to analyze differences between experimental groups. **P < 0.01, ***P < 0.001, and ns, not significant.

Oral LPG Regulates Immune Responses Through Gut-Lung and Gut-Spleen Immune Axes

To examine how LPG imprints the immune system, we collected spleen, mediastinal LN, mesenteric LN, small intestine, and colon tissues to define their transcriptional landscapes with RNA-sequencing after oral LPG administration (Figure 3A). Surprisingly, three similar transcript profiles were identified among mesenteric LN, mediastinal LN, and spleen, including downregulated TNF-α (a representative marker of inflammation), apoptosis, and enhanced IFN responses in the LPG group compared to the PBS group (Figure 3B), suggesting that oral LPG can exert effects in both the mucosal and systemic immune systems to promote immune cell activation and differentiation. Interestingly, LPG actively down-regulated the TNF-α pathway, as evidenced by reduced expression of AP-1 transcription factor family members (FOS, FOSB, JUN, JUNB) (10) (Figure 3B, upper panel). It also dampened apoptosis pathways, shown by decreased expression of the apoptotic genes BCL2L11, BTG2, and BTG3 (11, 12) (Figure 3B, middle panel). This down-regulation began on day 0 and peaked on day 1 for the mesenteric LN and spleen and on day 3 for the mediastinal LN, indicating that LPG almost simultaneously exerted its effects on the mesenteric LN and spleen with a slight delay for the mediastinal LN. However, this down-regulation was not observed in small intestine and colon tissues, suggesting that this mechanism largely functions on immune cells rather than digestive tissues. To further corroborate this observation, we quantified IL-6 and TNF-α in BAL from mice sacrificed on day 3, 7, and 14. A ~20-fold reduction in IL-6 and 3-8-fold decrease in TNF-α were observed on day 3 and 7, respectively, in the LPG group compared with the PBS group (Figure 3C).

Open in a separate window

Figure 3

Oral LPG administration down-regulates inflammation and apoptosis but promotes IFN signaling in mouse lymphoid tissues. Experimental schedule (A). ICR mice (n = 20/group) were treated as described in the Methods, and different tissues were collected for RNA sequencing. GSEA revealed inhibition of TNF-α signaling (top) and apoptosis (middle) gene sets in lymphoid tissues in the LPG group compared with the PBS group, and enhancement of IFN response (bottom) gene sets in lymphoid and small intestine tissues in the LPG group compared with the PBS group (B). IL-6 and TNF-α in BAL were quantified from mice sacrificed on day 3, 7, and 14 (C). Heatmap shows log2 RPKM values of a selection of genes associated with B cells (D) and T cells (E) from the spleen and small intestine. BAL, bronchoalveolar lavage; ES, enrichment score; FDR-q, false discovery rate q value; NES, normalized enrichment score.

In contrast to down-regulation of TNF-α and apoptosis signaling, LPG up-regulated the IFN response pathway as manifested by increased expression of IFN-regulatory factor 7 (IRF7) and IFN-stimulated genes IFITM1/2/3 (13) in the small intestine, spleen, mesenteric LN, and less for the mediastinal LN, but not for colon tissue (Figure 3B, down panel). This up-regulation was triggered in both digestive tissues and lymph tissues with the exception of colon tissue. These results indicate that oral LPG administration can affect the lung immune response through the gut-lung and gut-spleen immune axes.

We thank the members of the JQX and JGX laboratory for advice and assistance during the study.