key: cord-0975316-kurronhp
authors: Zeng, Q.; Huang, G.; Li, Y.-z.; Xu, G.; Dong, S.-y.; Zhong, T.-y.; Chen, Z.-t.; Xu, Y.
title: Loss expansion of SARS-CoV-2 specific immunity is a key risk factor in fatal patients with COVID-19
date: 2020-08-01
journal: nan
DOI: 10.1101/2020.07.29.20164681
sha: bb48837b4608479cafedb8c054f6aa8911541b19
doc_id: 975316
cord_uid: kurronhp

The dynamic immunological characteristics of COVID-19 patients are essential for clinicians to understand the disease progression. Our data showed that the immune system and function have gradually remodeled and declined with age from 16-91 years old in 25,239 healthy controls. Analyzing the relationship between the number of lymphocytes and age showed that lymphocytes and subsets tended to decline with age significantly, whereas, the number of natural killer cells tended to increase with age significantly. SARS-CoV-2 specific immunity has declined with age in fatal cases. Furthermore, SARS-CoV-2 specific immunity is associated with survival time in fatal cases. The loss expansion of SARS-CoV-2 specific immunity could be expanded in vitro. A concurrent decline in SARS-CoV-2 specific cellular and humoral immunity and prolonged SARS-CoV-2 exposure predicted fatal outcomes. Our findings have provided a basis for further analysis of SARS-CoV-2 specific immunity and understanding the pathogenesis of fatal COVID-19 patients.

Coronavirus disease 2019 (COVID-19) pandemic leads to severe illness, life-threatening 68 complications, and death, especially in high-risk groups such as elderly people and individuals 69 with comorbidities [1] [2] [3] [4] [5] [6] [7] . However, information on immunological characteristics in the assessment 70 of COVID-19 for frontline clinicians is limited. The objective of this study is to explore the 71 dynamic immunological characteristics as the disease progresses. 72

In January, 2020, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was 73 identified in samples of bronchoalveolar lavage fluid from patients in Wuhan, China, and was 74 al. 18 also observed SARS-CoV-2-specific humoral and cellular immunity in 14 convalescent 111 patients. However, to the best of our knowledge, SARS-CoV-2-specific CD4 and CD8 T cell 112 responses in fatal cases of COVID-19 were not reported. Furthermore, understanding the key 113 immune mechanisms to recover from novel SARS-CoV-2 exposure is an urgent need. 114

It is important to understand the key risk factors of critical patients who died. We hypothesize that 115 lethal COVID-19 disease would be associated with immune dysregulation and loss expansion of 116 SARS-CoV-2-specific immunity. Here, we analyzed dynamic immunological characteristics in 117 fatal COVID-19 patients in a unique longitudinal cohort of samples. 118

The immune system and function have gradually remodeled and declined with age in 25,239 120 healthy controls 121

The relationship between the number of lymphocytes and subsets and age showed that 122 lymphocytes and subsets except natural killer (NK) cells tended to decline with age significantly 123 ( Figure 1A -E, Spearman R = -[range, 0.08978-0.1950], P < 0.0001). However, the relationship 124 between NK cells and age showed that the number of NK cells tended to increase with age ( Figure  125 1F, Spearman R = 0.1188, P < 0.0001) significantly, suggesting that NK cells, a component of the 126

innate immune system, expanded with age. 127 SARS-CoV-2 specific immunity has declined with age in fatal COVID-19 patients 128 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint

The relationship between SARS-CoV-2 specific immunity and age showed that SARS-CoV-2 129 specific immunity tended to decline with age ( Figure 2A-C) . However, the relationship between 130 SARS-CoV-2 virus load and age showed that the virus load tended to increase with age ( Figure  131 2D). 132

The ability to expand SARS-CoV-2 specific interferon gamma (IFNγ)+CD4+ T cells (Figure 2A , 133 Spearman R = -0.8020, P < 0.0001) and IFNγ+CD8+ T cells ( Figure 2B , Spearman R = -0.8028, 134 P < 0.0001) was significantly inversely related to age. It suggested that the elderly was low in 135 expansion ability of SARS-CoV-2 specific T cell immunity, so it was easy to turn into severe 136 illness, which was an important risk factor for mortality in the elderly. SARS-CoV-2 specific 137 humoral immunity ( Figure 2C , Spearman R = -0.2671, P = 0.0254) was also inversely related to 138 age. However, this relationship was weaker than that of SARS-CoV-2 specific T cell immunity. 139 SARS-CoV-2 viral load was directly proportional to age, which was significant ( Figure 2D , 140 Spearman R = 0.4780, P < 0.0001). It suggested that the elderly were easily infected by the virus 141 or the patient had a higher viral load in the body. 142 SARS-CoV-2 specific immunity is associated with survival time in fatal COVID-19 patients 143

The relationship between SARS-CoV-2 specific immunity and survival time showed that as the 144 survival time increases, SARS-CoV-2 specific immunity tended to rise. However, the relationship 145 between virus load and survival time showed that the virus load tended to decrease as the survival 146 time increases (Figure 3) . 147

The correlation between SARS-CoV-2 specific IFNγ+CD4+ T cells ( Figure 3A , Spearman R = 148 0.7655, P < 0.0001) and IFNγ+CD8+ T cells ( Figure 3B , Spearman R = 0.7623, P < 0.0001) and 149 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint 8 survival time was relatively strong, and significantly correlated with survival time, suggesting that 150 patient survival time depended on the expansion ability of SARS-CoV-2 specific T cell immunity. 151

The aggregated data for 4 weeks showed that SARS-CoV-2 specific humoral immunity was 152 related to survival time significantly ( Figure 3C , Spearman R = 0.7924, P < 0.0001). There was no significant difference in the frequency of SARS-CoV-2 specific IFNγ+CD4+ T cells 158 ( Figure 4A ) between the first week and the second week. However, there were significant 159 differences between the first week and the third or fourth week, suggesting that SARS-CoV-2 160 specific IFNγ+CD4+ T cells began to expand in the third week significantly ( Figure 4A ). There 161 was no significant difference in the frequency of SARS-CoV-2 specific IFNγ+CD8+ T cells 162 between the first week and the second, third, or fourth week, indicating that the expansion of 163 specific CD8 T cells were lost ( Figure 4B ). There was a significant difference between the titer of 164 specific IgG antibodies in the first week and the second week. Compared to the first week, the IgG 165 titer in week 2-4 continued to show significant differences, suggesting that SARS-CoV-2 specific 166 antibodies began to expand in the second week, but the antibody titer was low ( Figure 4C ). There 167 was a significant difference between the viral load from week 1 to week 4 ( Figure 4D ), suggesting 168 that SARS-CoV-2 specific immunity could clear SARS-CoV-2 replication in fatal cases. However, 169

the SARS-CoV-2 specific immunity was unable to clear all SARS-CoV-2 in fatal cases. The mean 170 viral load was 6.7 x 10 4 copies/mL at week 3, indicating the presence of persistent SARS-CoV-2 171 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. The loss expansion of SARS-CoV-2 specific immunity in fatal cases could be expanded in 174 vitro 175

From the above results, it can be known that the key factor was the loss expansion of 176 SARS-CoV-2 specific humoral and cellular immunity and the virus cannot be eliminated from the 177 body in fatal cases. Since the expansion of SARS-CoV-2 specific immunity started from the first 178 2-3 weeks, we chose PMBCs in the third week for the expansion experiment in vitro. Our data 179

showed that these PBMCs could be expanded in vitro. After the expansion, SARS-CoV-2 specific 180

the SARS-CoV-2 specific immunity of these 21 fatal patients lost their expansion at week 3, but 182 the expansion of SARS-CoV-2 specific cellular immunity could be recovered in vitro experiment 183 ( Figure 5) . 184

A concurrent decline in SARS-CoV-2 specific cellular and humoral immunity and prolonged 185 SARS-CoV-2 exposure predicted fatal outcomes 186

There was significant correlation between SARS-CoV-2 specific IFNγ+CD4+ T cells ( Figure 6A , 187 Spearman R = 0.4537, P < 0.0001) and IFNγ+CD8+ T cells ( Figure 6B , Spearman R = 0.3721, P 188 = 0.0002) and SARS-CoV-2 specific humoral immunity. SARS-CoV-2 viral load significantly, suggesting that SARS-CoV-2 specific T cell immunity plays 192 a leading role in virus clearance. However, the frequency of SARS-CoV-2 specific T cells was 193 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint significantly lower than that of survival patients with COVID-19 ( Figure 9A , 9B), suggesting that 194 loss expansion of SARS-CoV-2 specific cellular immunity in fatal cases. 195 SARS-CoV-2 specific humoral immunity inversely correlated with SARS-CoV-2 viral load 196 significantly, suggesting that SARS-CoV-2 specific humoral immunity played a role in virus 197 clearance ( Figure 8) . However, the R number of SARS-CoV-2 specific humoral immunity was 198 smaller than that of SARS-CoV-2 specific T cell immunity (Figure 7, Figure 8 ), suggesting that 199 SARS-CoV-2 specific T cell immunity might play a major role in virus clearance in fatal cases. In 200 addition, the IgG titer was significantly lower than that of survival patients with COVID-19 201 ( Figure 9C ), suggesting that loss expansion of SARS-CoV-2 specific humoral immunity in fatal 202

However, the SARS-CoV-2 specific immunity was unable to clear SARS-CoV-2 in fatal cases. All 204 patients were detected positive viral load before death, suggesting that prolonged SARS-CoV-2 205 exposure predicted fatal outcomes ( Figure 4D) . 206

The extent of lymphopenia in patients admitted to the intensive care unit correlates with 208 COVID-19 severity and mortality [23] [24] [25] [26] . Therefore, studies of the immune system and function in 209 healthy controls are important to understand whether the immune system and function is related to 210 age. Our data showed that the immune system and function have gradually remodeled and 211 declined with age in 25,239 healthy controls. virus-specific memory CD8 T cells efficiently produced IFNγ, tumor necrosis factor α (TNF-α), 215 etc. and reduced lung viral load. Next, we tried to understand whether SARS-CoV-2 specific 216 immunity is associated with age in those fatal patients with COVID-19. Our data suggested that 217 SARS-CoV-2 specific immunity has declined with age in COVID-19 patients. 218

Levels of lymphocytes and lymphocyte subsets are of great importance to keep the immune 219 system working. Usually viral infection, immunodeficiency diseases, and other infectious diseases 220 lead to abnormal changes in the levels of lymphocyte subsets 16 . Although SARS-CoV-2 has been 221 identified as the causative agent of COVID-19, the mechanism by which SARS-CoV-2 impacts 222 the human immune system is still unclear 16 . From the above results, it can be known that the key factor was the loss expansion of 232 SARS-CoV-2 specific humoral and cellular immunity and the virus cannot be eliminated from the 233 body in fatal cases. We tried to understand whether the loss expansion of SARS-CoV-2 specific 234 immunity in fatal cases could be expanded in vitro. Since the expansion of SARS-CoV-2 specific 235 immunity started from the first 2-3 weeks, we chose PMBCs in the third week for the expansion 236 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint experiment in vitro. Our data showed that these PBMCs could be expanded in vitro. After the 237 expansion, SARS-CoV-2 specific IFNγ+CD4+ T cells and IFNγ+CD8+ T cells increased 238 significantly. It suggested that the SARS-CoV-2 specific immunity of these fatal patients lost their 239 expansion, but the expansion of SARS-CoV-2 specific cellular immunity could be recovered in 240 vitro experiment. 241

Our data indicated that a concurrent decline in SARS-CoV-2 specific cellular and humoral 242 immunity and prolonged SARS-CoV-2 exposure predicted fatal outcomes. First, the immune 243 system and function have gradually remodeled and declined with age in healthy controls. 244

Therefore, the elderly are susceptible to the SARS-CoV-2. Second, our data showed that 245 SARS-CoV-2 specific immunity has declined with age in COVID-19 patients. Therefore, the 246 elderly easily turned from mild to severe SARS-CoV-2 infections. Those results have explained 247 that the elderly is a risk factor for poor outcomes. Third, our data further showed that 248 SARS-CoV-2 specific immunity is associated with survival time in COVID-19 patients. The 249 correlation between SARS-CoV-2 specific IFNγ+CD4+ T cells and IFNγ+CD8+ T cells and 250 survival time was relatively strong, and significantly correlated with survival time, suggesting that 251 patient survival time depended on the expansion ability of SARS-CoV-2 specific T cell immunity. 252

The aggregated data for 4 weeks showed that SARS-CoV-2 specific humoral immunity was 253 related to survival time. SARS-CoV-2 viral load was inversely proportional to survival time 254

significantly. It suggested that the higher the viral load of SARS-CoV-2, the shorter the survival 255 time of patients. Therefore, controlling the replication of the virus might prolong the survival time 256 of patients. Finally, we observed that a concurrent decline in SARS-CoV-2 specific cellular and 257 humoral immunity and prolonged SARS-CoV-2 exposure predicted fatal outcomes. There was 258 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. 396 397 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint 21 398 399 Figure 6 A B 400 401 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. Figure S1 . A representative flow cytometry gating strategy. 426 427 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint

A novel coronavirus from patients with pneumonia in China

This retrospective study was conducted at Chinese PLA General Hospital, Peking Union Medical informed consent was obtained from healthy controls and waived to COVID-19 patients due to the 439 rapid emergence of this infectious disease. Identification of hypertensive patients was achieved by 440 reviewing and analyzing available electronic medical records and patient care resources. Clinical 441 outcomes (discharge, mortality, and recovery, etc.) were monitored and all clinical recovery 442 hypertensive patients meet the following criteria: body temperature returned to normal for more 443 than 3 days, respiratory symptoms improved significantly, and lung imaging showed significant 444 improvement. 445

Whole blood was centrifuged for 15 min at 1800 rpm to separate the cellular fraction and plasma. 448The plasma was then carefully removed from the cell pellet and stored at -20°C. Peripheral blood 449 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted August 1, 2020. were cultured for 7 days in the presence of 20 U/mL of recombinant IL-2 (R&D Systems, USA). 465

PBMCs were first stimulated with or without SPs (2 μg /mL) for 3 h at 37°C, then brefeldin A (10 467 μg/mL, Sigma-Aldrich, USA) was added to cultures to enable intracellular proteins to accumulate 468 in all stimulations. PBMCs were stimulated RPMI medium containing 10% FCS as a negative 469 control. After incubation for a total of 6 h, the cells were washed, fixed, permeabilized using 470 fixation/permeablization solution kit (BD Biosciences, USA) and blocked with FcR blocking 471 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted August 1, 2020. pyrocarbonate-treated water, and 2 μL of RNA template. The RT-PCR assay was performed 493 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted August 1, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint 29 standard deviations of the normal controls. 516

All samples were analyzed by FACSVerse™ flow cytometry (BD Bioscience, USA). Two blood 518 samples in two tubes (100 μL each) were stained according to the manufacturer's instructions. 519Then, red-cell lysis buffer (1 mL) was added to each tube, the samples were incubated for 10 min 520and washed with Sorvall cell washer (Thermo Fisher Scientific, USA). Cells were blocked with 521 

Categorical variables were described as frequency rates and percentages, and continuous variables 533 were described using mean and median values. Means for continuous variables were compared 534 using independent group t tests when the data were normally distributed; otherwise, the 535Mann-Whitney test was used. Data (non-normal distribution) from repeated measures were 536 compared using the generalized linear mixed model. Proportions for categorical variables were 537 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint 30 compared using the χ 2 test, although the Fisher's exact test was used when the data were limited. 538All statistical analyses were performed using SPSS (Statistical Package for the Social Sciences 539Inc., version 13.0). For unadjusted comparisons, a two-sided P value less than 0.05 was 540 considered statistically significant. The analyses have not been adjusted for multiple comparisons 541 and, given the potential for type I error, the findings should be interpreted as exploratory and 542 descriptive. Because the cohort of patients in our study was not derived from random selection, all 543 statistics are deemed to be descriptive only. 544 545 546 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted August 1, 2020. . https://doi.org/10.1101/2020.07.29.20164681 doi: medRxiv preprint