High titers of infectious SARS-CoV-2 in corpses of patients with COVID-19

Objectives The prolonged presence of infectious SARS-CoV-2 in deceased patients with COVID-19 has been reported. However, infectious virus titers have not been determined. Such information is important for public health, death investigation, and handling corpses. The aim of this study was to assess the level of SARS-CoV-2 infectivity in the corpses of patients with COVID-19. Methods We collected 11 nasopharyngeal swabs and 19 lung tissue specimens from 11 autopsy cases with COVID-19 in 2021. We then investigated the viral genomic copy number by real-time reverse transcription-polymerase chain reaction and infectious titers by cell culture and virus isolation. Results Infectious virus was present in six of 11 (55%) cases, four of 11 (36%) nasopharyngeal swabs, and nine of 19 (47%) lung specimens. The virus titers ranged from 6.00E + 01 plaque-forming units/ml to 2.09E + 06 plaque-forming units/g. In all cases in which an infectious virus was found, the time from death to discovery was within 1 day and the longest postmortem interval was 13 days. Conclusion The corpses of patients with COVID-19 may have high titers of infectious virus after a long postmortem interval (up to 13 days). Therefore, appropriate infection control measures must be taken when handling corpses.


Introduction
COVID-19 has caused more than 6.4 million deaths as of September 2022, and the causative pathogen, SARS-CoV-2, continues to circulate worldwide. Drommi et al. reported that nasopharyngeal swabs from 13 of 180 cadavers in the morgue were polymerase chain reaction (PCR)-positive [1] , and Ventura reported that nasopharyngeal swabs taken from a cadaver kept in the morgue for 87 days were PCR-positive [2] . Therefore, when autopsies are performed on bodies that have or are suspected to have been infected with this novel coronavirus, detecting SARS-CoV-2 and evaluating its infectivity is important to better determine the cause of death [ 3 , 4 ]. During autopsies, it is also important to reduce the risk of infection for personnel involved in the examination [ 5 , 6 ]. Knowing the infectivity of viruses that remain in corpses is a key public health issue for pathologists, forensic pathologists, medical examiners, clinical physicians, and nurses, in addition to those that handle corpses.
Heinrich et al. detected subgenomic RNA, which indicates viral replication, in specimens collected from the pharynx of corpses of patients with COVID-19 as long as 35.8 hours after death [7] . Plenzig et al. reported that infectious SARS-CoV-2 was present in the lungs of two corpses with postmortem times of 4 and 17 days, respectively [8] . Zacharias et al. reported that seven of 11 lung tissue swabs were culture positive, with a postmortem interval (PMI) ranging from 14 to 68 hours [9] . Heinrich et al. further reported that 20% of 128 SARS-CoV-2 RNA-positive corpses retained the infectious virus up to 14 days after death [10] .
These reports describe the presence of subgenomic RNA, demonstrating virus replication in the samples or the presence of live virus. However, the amount of infectious SARS-CoV-2 in corpse tissues has not been determined. Therefore, here, we evaluated infectious virus titers of samples taken from the corpses of patients with COVID-19 autopsied in Japan between January and October 2021. We further examined the relationship between the infectivity of the corpse and the viral RNA load on the autopsy date, the time from COVID-19 diagnosis to death, the time in the mortuary, and the PMI.

Samples
A total of 30 specimens (11 nasopharyngeal swabs and 19 lung tissues) were collected from forensic and pathological autopsies of corpses of patients with COVID-19 in Japan between January and October 2021. Three lung tissue specimens from the 11 cases were not tested. The information on the age, sex, body mass index (kg/m 2 ), medical history, antiviral medications (if any), place of death, date of death (month), decomposition status at the time of autopsy, cause of death, days from COVID-19 diagnosis to death, and PMI for each case is shown in Table 1 . The date of COVID-19 diagnosis is defined here as the date of the first positive result by PCR testing or as the date of death unless diagnosed before death.

Sample collection
Nasopharyngeal swabs were collected just before the autopsy. The swabs were placed in culture medium (SUGIYAMA-GEN Co., Ltd. Tokyo, Japan) and kept on ice during the autopsy and then stored at -80 °C when the autopsy was completed. An approximately 1-2-cm 2 pulmonary tissue sample was dissected with scissors and tweezers that were prewiped with 70% ethanol, kept on ice during the autopsy, and stored at -80 °C after the autopsy completion. The time interval between death and sample collection is shown in PMI (B + C) of Table 1 . Frozen tissues were homogenized and a clarified supernatant was used for virus isolation.

RNA extraction
RNA was extracted from a 200-μl nasopharyngeal swab culture medium sample or approximately 10 mg of pulmonary tissue sample by using the Maxwell R RSC Viral Total Nucleic Acid Purification Kit (Promega, Madison, Wisconsin, USA) in 50 μl of elution buffer.

Virus quantification by real-time reverse transcription-PCR
Real-time reverse transcription-PCR of the N1 and N2 viral genomic regions was performed using 1 μl of the RNA preparation. The probe and primer sequences and reaction conditions were as previously published by Adachi et al. [11] and Shirato et al. [12] . Endogenous hGAPDH-messenger RNA (mRNA) was used as an internal control, as previously described by Katano et al. [13] . The detection limit of the real-time reverse transcription-PCR was 5 copies/reaction.

Cell culture and virus isolation
Cell culture and virus isolation were performed according to the method of Matsuyama et al. [14] . VeroE6/TMPRSS2 (JCRB1819) cells were obtained from the National Institutes of Biomedical Innovation, Health and Nutrition, Japan. The cells were maintained in Dulbecco's modified Eagle medium (DMEM) containing 10% fetal calf serum (FCS) and antibiotics at 37 °C with 5% CO 2 .
A 24-well plate containing a VeroE6/TMPRSS2 cell culture monolayer at a low crowding density (70-90% confluence) was prepared. After the medium was discarded, 100 μl of the undiluted sample or 10-fold dilutions of the sample were added to the cells and incubated for 1 hour at 37 °C. Then, 0.5 ml of DMEM with 5% FCS, 2.5 μg/ml amphotericin B (Sigma-Aldrich Japan G.K., Tokyo, Japan), and 50 μg/ml gentamicin sulfate (Nacalai Tesque, Kyoto, Japan) was added and incubated at 37 °C for 1 week until a cytopathogenic effect was observed.

Virus titration assay
Confluent Vero E6/TMPRSS2 cells in 12-well plates were infected with 100 μl of undiluted or 10-fold dilutions (10 −1 -10 −5 ) of the sample. After incubation for 1 hour at 37 °C, the cells were washed once and overlaid with 1% agarose solution in DMEM with 5% FCS. The plates were incubated for 3 days and then fixed with 10% neutral buffered formalin. After removal of the agar, the plaques were counted.

Biosafety statement
All experiments with SARS-CoV-2 viruses were performed in the University of Tokyo's enhanced biosafety level 3 containment laboratories, which are approved for such use by the Ministry of Agriculture, Forestry, and Fisheries, Japan.

Statistical analysis
Fisher's exact probability tests using IBM SPSS Statistics, Version 25.0J for Windows (SPSS Inc., Chicago, IL, USA) were performed for the presence of infectious viruses and the viral genomic copy number at the date of autopsy in nasopharyngeal swabs (n = 11) and lung tissues (n = 19). The significance level was set at P -value < 0.05 (two-tailed). PMI, postmortem interval. a PMI(B + C) represents the time interval between death and sample collection. Colored areas indicate specimens with infectious viruses. The color coding is based on the amounts of virus; dark orange represents the sixth power of 10, light orange the fourth power of 10, pink the third power of 10, light yellow the second power of 10, and light green the first power of 10.

Viral genomic copy number and infectious virus titers
Infectious virus was detected in six (cases 2, 3, 4, 6, 10, and 11) of 11 cases and 13 (four nasopharyngeal swabs and nine lung tissue specimens) of 30 specimens ( Table 2 ). Figure 1 shows the relationship between viral genomic copy number on the autopsy date, time (days) from the COVID-19 diagnosis to the autopsy date, and the presence or absence of infectious virus. The infectious virus titers ranged from 6.00 E + 01 plaque-forming units (PFU)/ml to 6.00 E + 03 PFU/ml in nasopharynx swabs and from 3.89 E + 02 PFU/g to 2.09 E + 06 PFU/g in lung tissues. The viral genomic copy number for the specimens ranged from the third power of 10 to the seventh power of 10, except in case 1. Of the 13 specimens containing the infectious virus, the lowest viral genomic copy number (3840 copies/μl) was found in the left lung tissue of case 11, which had an infectious titer of 1.27E + 04 PFU/g. Of the 17 specimens containing the noninfectious virus, the highest viral genomic copy number (7.23E + 06 copies/μl) was found in the left lung tissue of case 5.

Statistical analysis of viral genomic copy number and infectivity
A comparison of the numbers of infectious virus-containing and noncontaining-specimens with a viral genomic copy number The numbers around the bubble indicate the infectious titer in PFU/ml or PFU/mg; the size of the bubble represents the magnitude of the infectious titer. The case number and specimen site are indicated in parentheses below the numbers, with N for nasopharyngeal swab, L for left lung, and R for right lung. The case number and specimen site are also indicated around the cross mark. PFU, plaque-forming units.

Table 3
Result of statistical analysis of infectivity and viral genomic copy number. of more than or less than 10 0,0 0 0 copies/μl showed no significant difference for either the nasopharyngeal or lung specimens ( Table 3 ).

Antemortem conditions of the corpses and infectivity
Of the two hospitalized cases (cases 2 and 5), infectious virus was found in case 2, in which the patient had been hospitalized for cirrhosis, contracted COVID-19 in the hospital, and died of severe pneumonia. This patient could not be treated with remdesivir due to renal dysfunction and had been taking dexamethasone, as well as heparin, for 3 days before death. In case 5, the patient was hospitalized 2 days after a positive PCR test, received remdesivir 2 days later, and died 4 days later. After death, the patient was stored in the morgue (4 °C) and an autopsy was performed 4 days after her death. In this case, the viral genomic copy number was in the million-copy range for the nasopharyngeal swab and the left and right lungs, but none of the specimens contained the infectious virus.
No infectious virus was detected in case 1, for which the viral genomic copy number was low, but this body was found in a pond 4 days after discharge when COVID-19 was already cured; the cause of death was drowning.
Of the eight nonhospitalized patients, the virus was isolated in five cases (cases 3, 4, 6, 10, and 11) and no infectious virus was detected in the other three cases (cases 7, 8, and 9).
In two cases (cases 3 and 4), the patients had been recuperating in hotel rooms without clinical treatment and were found to contain the infectious virus. Among the three cases (cases 6, 7, and 8) of patients that were recovering at home, the corpse in case 6 contained infectious virus whereas the other two cases did not. The patient in case 6 had a body mass index of 38.6 ( i.e ., was obese) and died at home 2 days after COVID-19 diagnosis. In contrast, the patient in case 7 died 9 days after diagnosis and was found 3 days after death, and the patient in case 8 died 3 days after diagnosis and was found 4 days after death in a highly decomposed state.
In the three cases (cases 9, 10, and 11) of patients found dead at home, the corpses in cases 10 and 11 were found to contain the infectious virus; however, both were in poor health before death and were found to be positive by postmortem PCR testing. The corpse in case 9 was in a highly decomposed state and the infectious virus was not found.

Postmortem state of the corpses and infectivity
Infectious virus was detected in three (cases 2, 4, and 6) of the four cases (cases 2, 4, 6, and 8), for which 4 days or less had elapsed from the antemortem diagnosis to death.
The two cases (cases 10 and 11) of three cases (cases 9, 10, and 11) in which the infection was detected by postmortem PCR testing were stored in a refrigerator the day after death; the left and right lungs of both corpses showed high infectious titers. These two cases were not decomposed.
In the two cases (cases 8 and 9) in which the infectious virus was not detected, the corpses had been left at room temperature before discovery and were highly decomposed.
Virus was isolated from six of the seven cases in which the time from death to discovery was within 1 day. In these six cases, the period of refrigeration ranged from 2 to 12 days and the PMI ranged from 3 to 13 days. Case 5 was the only case in which the virus was not isolated; the patient in this case died 4 days after starting COVID-19 treatment with remdesivir.

Discussion
Here, we determined the amount of virus in corpses of patients with COVID-19 and found that the infectious virus was present in large amounts (up to 2.09E + 06 PFU/g) in corpse lung tissue and in refrigerated corpses even 13 days after death. Grassi et al. reported that of 29 PCR-positive autopsy cases using postmortem nasopharyngeal swabs, viral mRNA was detected in seven of 22 posthospitalization deaths and six of seven nonhospitalization deaths [15] . Similar to previous reports [ 16 , 17 ], our findings indicate that individuals involved in autopsies or examinations of corpses must consider the risk of infection with SARS-CoV-2 and take measures to protect themselves from infection to reduce that risk. Our findings are crucial for public health worldwide.
In our analysis, there were no statistically significant differences in viral genomic copy number (10 0,0 0 0 copies/μl) or infectivity in either the nasopharynx or lungs. However, when the viral copy number in the nasopharyngeal swabs was less than 10 0,0 0 0 copies/μl ( Table 3 ), the infectious virus was not detected. Previously, it has been reported that there is a statistically significant difference between the minimum cycle threshold value of RNA and the presence of viral mRNA [15] . We plan to further examine the relationship between viral genomic copy number and infectivity.
The viral genomic copy number for case 1 was very low, as shown in Table 2 . In this case, a large mass (about 3 × 3 × 3 cm 3 ) was collected from the lung during autopsy. The specimen was then frozen at -80 °C and the mass was subsequently thawed when the RNA was extracted. Although the lung used for extraction weighed approximately 10 mg, it is possible that the viral mRNA was degraded when the specimen was thawed. In addition, because this was a case of drowning in a pond the day after discharge from the hospital, it is possible that the viral genomic copy number was not measured accurately due to the condition of the lungs. When examining the infectious titer of the virus, it is important to consider various factors, such as the timing of specimen collection, collection method, specimen storage method, and use of culture medium for transport, etc . In the future, we plan to examine conditions related to specimen collection for such emerging infectious diseases.
In our study, the infectious viruses were detected when the diagnosis date was within 4 days of the date of death and the body had not yet decomposed. Infectious viruses were also detected when the discovery date was within 1 day of the date of death. There appears to be a state in which the virus survives in the early stages of infection when the host's life has been terminated and the virus remains infectious. Infectious viruses were also detected when the refrigeration period was as long as 12 days, consistent with previous reports [ 18 , 19 ]. In our limited number of samples, we found that when death occurs because of rapidly worsening COVID-19 symptoms or within a short period after SARS-CoV-2 infection, the probability that infectious virus remains in the corpse is high. If antemortem information is not available, postmortem SARS-CoV-2 PCR testing should be performed. If SARS-CoV-2 PCRpositive corpses are found within a few days after death and they are kept in a cold environment, they should be handled with caution due to likely presence of the infectious virus. When COVID-19 first began to spread, different countries published their own guidelines for the handling of cadavers [20] , but autopsies of bodies infected or suspected of being infected with SARS-CoV-2 should be performed in accordance with the postmortem guidance of the World Health Organization [21] and the Centers for Disease Control and Prevention [22] .
In our study, the titer in lung tissue was higher than that in the nasopharyngeal swab fluid. Therefore, great care should be taken when handling the lungs during autopsy and formalin fixation should be performed immediately. Also, appropriate infection control measures should be implemented during the entire funeral and burial processes to protect the workers handling corpses. To this end, ensuring that all those involved in the handling corpses are promptly provided with the necessary supplies for infection protection is critical. Furthermore, measures, such as prioritizing vaccination, in this population of workers should be considered.

Limitations
The limitation of this study is the small number of the corpses of patients with COVID-19 examined. This is because the cases included in this study were COVID-19-related deaths for which pathological or forensic autopsies were performed between January and October 2021, and very few COVID-19 autopsies were performed in Japan during this time. Therefore, our research team could not be selective with respect to the cases included; we analyzed every case that occurred during this period.
In addition, the antemortem information was insufficient in some of the cases.
Generalizations based on the results of the statistical analysis should be made with caution.

Conclusion
To the best of our knowledge, this study is the first to show infectious virus titers in the corpses of patients with COVID-19. SARS-CoV-2 remains infectious with cold storage, regardless of the PMI, for at least 13 days. Therefore, those involved in the autopsies or examinations of corpses must consider the risk of infection with SARS-CoV-2 and take appropriate measures to protect themselves from infection to reduce that risk.

Declaration of competing interest
The authors have no competing interests to declare.

Funding
This study was conducted under the title of "Research on Evaluation of the Infectivity of coronavirus disease-2019 in Human Remains" (Project No. 20HA2008), which was funded by the Health, Labour and Welfare Administration's Research Grant for the Promotion of Emerging and Re-emerging Infectious Diseases and Immunization Policy in 2020 and 2021.

Ethical approval
The study protocol was approved by the institutional review boards of Chiba University Graduate School of Medicine, The Uni-versity of Tokyo Graduate School of Medicine, and the Institute of Medical Science.