Duration of SARS-CoV-2 mRNA vaccine persistence and factors associated with cardiac involvement in recently vaccinated patients

Autopsies

Tissues were collected from autopsies performed at Massachusetts General Hospital between January 2021 and February 2022. The tissues collected were bilateral axillary lymph nodes, mediastinal lymph nodes, liver, spleen, cardiac left ventricle, and cardiac right ventricle. Since the location of the injection site of vaccination was usually not known, the axillary lymph nodes were sampled and analyzed bilaterally in all cases. Inclusion criteria were the ability to collect fresh tissue, a clear history of either vaccination or no vaccination, and a post-mortem interval ≤ 60 h. Patients with a history of SARS-CoV-2 infection were excluded from the vaccine group and the non-vaccinated control group. Cardiac tissue from patients dying from SARS-CoV-2 was used as a positive control for SARS-CoV-2 PCR. The study was approved by the Hospital’s Human Subject’s Institutional Review Board (Mass General Brigham IRB). The requirement for patient consent was waived by the IRB, and patient consent was not obtained.

RNA extraction

Approximately 2 g of tissue was placed into a 50 mL conical tube containing 10 mL TRIzol reagent (Thermo Fisher Scientific, Waltham, MA) and kept at −80 °C. All samples were thawed at room temperature in TRIzol reagent for 10 min to sterilize any SARS-CoV-2 infected samples. Using a Hitachi HVC20 homogenizer (Kinematica AG, Switzerland), tissue was lysed on ice with 5 s durations at high speed, followed by 10+ s at rest.

RNA was extracted from the TRIzol homogenate according to the manufacturer’s protocol. Briefly, 200 µL chloroform was added per 1 mL TRIzol sample and vortexed at high speed for 15 s, and incubated at room temperature for 3 min. The volume of chloroform was increased 2× for samples with visibly high tissue content. After incubation, samples were centrifuged at 12,000g for 15 min at 4 °C. Phase-separated RNA was transferred to a clean 1.5 mL tube to which 500 µL of 2-propanol and 4 µL of glycogen were added and mixed by gentle pipetting, followed by 10 min incubation at room temperature. Samples were centrifuged at 12,000g for 10 min at 4 °C. The supernatant was aspirated, and pelleted RNA was washed 1–2 times with 75% (v/v) ethanol and spun down at 7500g for 5 min at 4 °C. After final aspiration, RNA pellets were briefly air dried, resuspended with 100–400 µL of endonuclease-free water, and placed in a 55 °C heating block for approximately 1 min.

RNA quality assessment

The yield for each extracted RNA sample was determined using a Nanodrop ND-2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA). The 280/260 ratio was an initial indicator of each sample’s purity. An EDTA-free protocol from the Joint Genome Institute (JGI)²⁵ was applied for all DNase treatment using RN easy kit (Qiagen, Louisville, KY) and 2U of DNAse I enzyme (New England Biolabs, Ipswich, MA). Two-step RT-qPCR assays of housekeeping genes, either with or without the reverse transcriptase (-RT), were performed to further confirm quality RNA yield. The mix with RT contained 0.5–1 µg of total RNA, Random Primers (Promega), Oligo(dT)15 Primer (Promega, Madison, WI), and Superscript III reverse transcriptase (Invitrogen). In triplicate, sample cDNA of 2 µL (10 ng) was added to a 20 µL volume containing SYBR Green Master Mix (Life Technologies, Carlsbad, CA) and 300 nM of each primer (IDT DNA, Coralville, IA, see Supplementary Table 5). Thermocycling settings were 10 min at 95 °C DNA Polymerase activation, followed by 40 cycles with 20 s at 95 °C, 20 s at 60 °C, and 15 s at 72 °C. Amplification C_t values and melting curve plots of ACTB liver cDNA and ACTB spleen samples (10 ng) were compared to their -RT samples (also 10 ng) in parallel. All other tissue cDNA was compared with -RT controls for GAPDH gene amplification. Extracted RNA used for vaccine detection RT-qPCR experiments had cDNA with C_t values 7 cycles or higher than -RT parallel mixes.

RT-qPCR assay for mRNA vaccine detection

The finalized RT-qPCR assay target for BNT162b2 mRNA sequence aligns in most of the central helix domain (CH), whereas for mRNA-1273 it lies within the heptad-repeat-2 domain (HR2) (Fig. 1a). For vaccine-positive controls, double-stranded DNA (dsDNA) fragments of BNT162b2 and mRNA-1273 vaccine sequences were purchased from Integrated DNA Technologies (Coralville, IA). The gBlock^TM sequences were selected to either include or be near both vaccines’ two Proline substitutions designed to help stabilize the spike protein (Supplementary Fig. 2)^26,27. For BNT162b2 detection, primers were designed using Primer3Plus and tested with SYBR Green assay on dsDNA fragments and non-template control (NTC) sample wells. Melting curve analysis was used to determine the best set of primers. Optimized primer concentrations are listed in Table 2. For mRNA-1273 vaccine detection, three sets of primers were tested using serial dilutions against a localized TaqMan probe and the final set selected for optimal primer efficiency score. The primer efficiency of the final BNT162b2 and mRNA-1273 primer/probes sets were 97.7% and 97.2%, respectively (Supplementary Fig. 3). After correcting from dsDNA material in the calibration curves to ssRNA vaccine (i.e. multiplied by 2), the limit of detection (LOD) for BNT162b2 was 6 ssRNA copies/reaction and the LOD for mRNA-1273 was 11 ssRNA copies/reaction.

RNA samples of 2 µL (100 ng total) were added to a final volume of 20 µL reaction mixture containing 5 µL TaqMan Fast Virus 1-step Master Mix (Life Technologies, Carlsbad, CA) in triplicate. Thermocycling was performed at 50 °C for 10 min for reverse transcription, followed by 20 s at 95 °C and then 45 cycles of 95 °C for 5 s and 60 °C for 30 s. All assays included dsDNA vaccine sequence fragments (positive controls) and NTC wells (negative controls). Five unvaccinated autopsies were tested with each vaccine assay as additional controls. Criteria for vaccine detection in the assays required samples to have a Ct value below 35 cycles in all three wells tested. As a further control, a SARS-CoV-2 detection assay for E gene was administered²⁸. Briefly, the same RNA concentration of 100 ng was combined with TaqMan Fast Virus 1-step Master Mix with E Sarbeco primers and probe (IDT DNA, Coralville, IA, see Supplementary Table 5). Incubation for reverse transcription was done at 50 °C for 10 min, followed by 95 °C for 20 s, 45 cycles of 95 °C for 3 s, and 58 °C for 30 s. All fluorescent-based assays were performed on a QuantStudio 3 (Thermo Fisher Scientific).

Validation of vaccine-positive samples

To validate either BNT162b2- or mRNA-1273-positive samples from the RT-qPCR assays, their cDNA products were used in 1–3 possible PCR reactions to target areas outside the positive control (dsDNA fragment) sequence. Table 3 lists the primer sets, location in the vaccine sequence, and amplicon length. Briefly, PCR reactions were performed with 25 µL reaction mix consisting of 50–100 ng sample cDNA, 1–1.25 µM each primer, and 12.5 µL GoTaq® Green Master Mix (Promega, Madison, WI) following manufacturer’s instructions. Reactions consisted of 40–43 amplification cycles of 30 s denaturation at 95 °C, 60 s annealing at 60 °C, and elongation time dependent of bp length at 72 °C (30 s per 500 bp). PCR products were resolved on 1.5% agarose gel stained with SYBR Safe DNA Gel Stain (Thermo Fisher Scientific). Bands of target bp length were excised and purified with QIAquick ® Gel Extraction Kit (Qiagen, Louisville, KY) and sequenced by Genewiz (Cambridge, MA) and analyzed with BLASTn for sequence alignment with the vaccine sequence²⁹.

Assessment of the timing of onset of myocardial injury

The age of healing myocardial injury in the specimens was estimated using standard pathologic features: myocardial necrosis with low amounts of neutrophils (days 1–2), large amounts of neutrophils (days 3–5), macrophage infiltration and myocyte removal (days 4–14), loose granulation tissue (days 14–21), late collagenous granulation tissue (days 21–28), and dense scar formation (more than 28 days)^30,31.

Immunohistochemistry

Tissue sections from formalin-fixed paraffin-embedded tissue were mounted on positively charged glass slides. Deparaffinization and rehydration of sections were done by treating slides with Xylene (2 treatments 5 min each) and graded ethanol (100, 90, 70, 50%, and water) solutions, respectively. Following rehydration, antigen retrieval was carried out by the heat-induced epitope retrieval (HIER) method using a low pH antigen retrieval solution (Biocare Medical, Reveal Decloaker, Pacheco, CA) in a pressure cooker for 3 min. Slides were then slowly cooled and rinsed in water. Endogenous peroxidases and alkaline phosphatases were blocked by treating sections with BLOXALL® (Vector Laboratories, Burlingame, CA) for 20–30 minutes followed by two washes in tris-buffered saline with Tween (TBST) buffer. Slides were then incubated with 2.5% normal horse serum (Vector Laboratories) to block non-specific binding.

To detect the macrophage marker CD68, tissue sections were incubated overnight with a mouse monoclonal antibody (anti-CD68 Monoclonal Antibody, clone KP1, eBioscience^TM, Life Technologies, Carlsbad, CA) at a dilution of 1/250 at 40 °C. Primary antibodies were detected by treating slides with a commercial secondary detection reagent (ImmPRESS® HRP Horse Anti-Mouse IgG PLUS Polymer Kit, Peroxidase, Vector Laboratories, Burlingame, CA) for 20 min followed by washing with TBST wash solution. Finally, antigens were detected with a brown Horse Radish Peroxidase (HRP) substrate (ImmPACT® DAB Substrate Kit, Peroxidase,Vector Laboratories) by incubating experimental slides for 1 min. Counterstaining was done with Meyer’s hematoxylin and slides were dehydrated by graded treatment with alcohol (50, 70, 90, and 100% respectively) followed by removal of alcohol with two treatments in xylene (3 min each). Permanent coverslips were mounted using Cytoseal XYL Mounting Medium (Richard-Allan Scientific® Cytoseal XYL Mounting Medium, Thermo Scientific, Waltham, MA). The macrophages were manually counted in the ten 400x high power fields (HPF) containing the most macrophages using an Olympus BX53 microscope and expressed as the average number of macrophages per HPF.

For spike protein detection, slides were incubated with the antibody (SARS-CoV-2 Spike Protein S1 rabbit Polyclonal Antibody, Invitrogen, Waltham, MA, catalogue # PA5-114528) at 1/1000 dilution overnight. Primary antibodies were detected by treating slides with a commercial secondary detection reagent (ImmPRESS®-AP Horse Anti-Rabbit IgG Polymer Detection Kit, Alkaline Phosphatase, Vector Laboratories, Burlingame, CA) for 20 min followed by washing with TBST solution. Antigens were detected with a red alkaline phosphatase substrate (Vector Laboratories) by incubating experimental slides for 1 min.

Statistical analysis

Groups were compared using t-test, Fisher exact test, or chi-square test as appropriate. Quantstudio Design & Analysis Software v1.52 was used for amplification graphing and melting curve plot analysis. All other graphs and analysis were performed using GraphPad Prism 9 software (GraphPad Prism Software, San Diego, CA). P values less than 0.05 were considered significant. All statistical tests were two-sided.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.