key: cord-0765016-5wlvp2kv
authors: Madden, Patrick; Thomas, Yanique; Blair, Robert; Samer, Sadia; Doyle, Mark; Midkiff, Cecily; Doyle-Meyers, Lara; Becker, Mark; Arif, Shoaib; McRaven, Michael; Simons, Lacy; Carias, Ann; Martinelli, Elena; Lorenzo-Redondo, Ramon; Hultquist, Judd; Villinger, Francois; Veazey, Ronald; Hope, Thomas
title: An immunoPET probe to SARS-CoV-2 reveals early infection of the male genital tract in rhesus macaques
date: 2022-04-08
journal: Res Sq
DOI: 10.21203/rs.3.rs-1479315/v1
sha: 73d3d052e0951a4785692bcf209fe6816d99b515
doc_id: 765016
cord_uid: 5wlvp2kv

The systemic nature of SARS-CoV-2 infection is highly recognized, but poorly characterized. A non-invasive and unbiased method is needed to clarify whole body spatiotemporal dynamics of SARS-CoV-2 infection after transmission. We recently developed a probe based on the anti-SARS-CoV-2 spike antibody CR3022 to study SARS-CoV-2 pathogenesis in vivo. Herein, we describe its use in immunoPET to investigate SARS-CoV-2 infection of three rhesus macaques. Using PET/CT imaging of macaques at different times post-SARS-CoV-2 inoculation, we track the 64Cu-labelled CR3022-F(ab’)2 probe targeting the spike protein of SARS-CoV-2 to study the dynamics of infection within the respiratory tract and uncover novel sites of infection. Using this method, we uncovered differences in lung pathology between infection with the WA1 isolate and the delta variant, which were readily corroborated through computed tomography scans. The 64Cu-CR3022-probe also demonstrated dynamic changes occurring between 1- and 2-weeks post-infection. Remarkably, a robust signal was seen in the male genital tract (MGT) of all three animals studied. Infection of the MGT was validated by immunofluorescence imaging of infected cells in the testicular and penile tissue and severe pathology was observed in the testes of one animal at 2-weeks post-infection. The results presented here underscore the utility of using immunoPET to study the dynamics of SARS-CoV-2 infection to understand its pathogenicity and discover new anatomical sites of viral replication. We provide direct evidence for SARS-CoV-2 infection of the MGT in rhesus macaques revealing the possible pathologic outcomes of viral replication at these sites.

The COVID-19 pandemic has exposed the broad systemic impact that can be caused by 43 infection with a respiratory virus. Disease associated with SARS-CoV-2 infection starts with 44 respiratory pathologies and subsequently can extend to other organ systems. There is now 45 ample evidence that SARS-CoV-2 can disseminate and replicate in tissues beyond the 46 respiratory tract. A clear example is infection in the gastrointestinal (GI) tract 1 . GI symptoms, 47 including diarrhea, have been reported by individuals with mild COVID-19, and hospitalized 48 patients have exhibited more severe symptoms such as ischemia and GI bleeds 2, 3,4 . In addition, 49

it is now well established that virus is shed through the GI tract in most infected individuals and 50 wastewater screening has become an important tool for disease surveillance 5, 6 . Although less 51 studied, many other tissues have been found to harbor SARS-CoV-2. Multiple groups have 52

shown the presence of viral RNA in cardiac, renal, and brain tissues 7-10 . There is also some 53 evidence of virus in the male genital tract (MGT) 11, 12 . Furthermore, symptoms associated with 54 all these organ systems have been regularly reported 13 . Likewise, other RNA viruses have 55 documented early dissemination to distal tissues that manifest infection-related pathology over 56 positive for spike and nucleocapsid primarily found in the interstitial space (Fig 7Q and R) . 283

Staining of penile tissue from an uninfected macaque ( Fig 7S) To better compare the probe signal in the lungs of JF82 at the 1-week (Fig 9A-B ) and 2-302

week (Fig 9C-D) timepoints, the 3D reconstruction of the JF82 lungs was isolated from the PET 303 datasets and projected over the CT reconstruction of the skeleton (Fig 9A-D) . Both the 1-and 304 16 2-week lung signals are apparent and localized with a level of signal comparable to the previous 305 animal (IN22) infected with the Delta variant. An evaluation of the data set revealed a PET 306 signal overlying a region of opacity in the lower lobe of the left lung as designated with the 307 asterisks in several of the panels (Fig 9A-B , 9G-L). The lung rotation series shown for week 1 308 ( Fig 9E) and week 2 ( Fig 9F) reveal a major signal associated with the dorsal side of the left lung 309 at both timepoints and less signal associated with the right lung in the week 1 scan. Evaluation 310 of signal from coronal sections of the week 1 PET/CT overlay ( Fig 9G) and CT alone ( Fig 9I)  311 reveals an overlap of the probe signal with an opaque region consistent with focal pneumonia. 312

It is notable that both the PET and CT signal associated with this spot in the left lung are gone in 313 the week 2 scan (Fig 9H and J) . This is consistent with reports that the lung pathology observed 314 in the rhesus macaque model is most apparent after 1 week of infection and can wane by week 315 2 25 . To better illustrate the change between week 1 and week 2, the week 1 scan in red and the 316

week 2 scan in blue were overlaid with the week 1 CT signal (Fig 9K and L) . Microscopic analysis 317 revealed pulmonary infiltrates were still present in the alveolar space at necropsy (Fig 1E) . 318

However, no infected cells were detected with immunofluorescence using an anti-SARS-CoV-2 319 antibody in FFPE tissue. These findings are suggestive of a resolving infection which is 320 supported by the histopathology (Fig 1E) and viral RNA levels (Fig 1F and G) . 321

We next evaluated the signal associated with the MGT of JF82 as illustrated in Fig 10,  322 which presents the front and near side view (~45˚) of the abdominal area of the week 1 ( Fig 10A  323 and D), week 2 (10B and E), and overlay (10C and F). The overlay (Fig 10C and F ) reveals the 324 dynamics of the probe signal in the MGT of JF82 in the first 2 weeks of SARS-CoV-2 infection. 325

The white signal in the overlay reveals that the probe signal is maintained in the prostate, the 326 vasculature at the base of the spermatic cord, and the base of the testes. To gain additional 327 insights into the MGT associated signal at the 2 time points, we isolated the MGT volumes and 328 3D projected the signal (Fig 10G-J) . In the week 1 scan (Fig 10G-H) , a signal associated with the 329 root of the penis is also apparent in addition to the signal associated with the vasculature at the 330 base of the spermatic cord and the base of the testes. In the week 2 scan, the signal becomes 331 more diffuse, spreading throughout the penis and testes, and extending into the spermatic 332 cord, especially into the right spermatic cord. To better visualize the signals associated with the 333 different tissues, we isolated the volumes containing the penile signal for the week 2 scan . The signal distribution throughout the penis at week 2 is readily apparent and distinct 335 from the signal associated with the spermatic cord. It is notable that the probe signal associated 336 with the MGT becomes better distributed and more pronounced in the week 2 scan, consistent 337 with a spreading infection in the MGT between week 1 and week 2. In contrast, a focus of 338 infection in the right lung (Fig 9) of the same animal is observed in the week 1 scan and 339 resolved in the week 2 scan. 340

Histopathology of testicular tissue from JF82 revealed multifocal regions of degenerate 341 seminiferous tubules characterized by a complete loss of germ cells and spermatids (Fig 10N) . 342

These regions also have evidence of edema as revealed by increased spaces between individual 343 seminiferous tubules. Degenerate seminiferous tubules occasionally contained macrophages 344 with phagocytosed spermatids, and the adjacent interstitium was infiltrated by low numbers of 345 lymphocytes and plasma cells. To further characterize the degenerative changes noted on H&E, 346 immunofluorescence for CD206 -a mannose receptor present on monocytes, macrophages 27 , 347 and mature spermatids 28-30 -and caspase 3 -a cellular marker of apoptosis -was performed. 348

Degenerate seminiferous tubules were readily identified by the marked decrease in CD206 349 expression (due to loss of mature spermatids) and increased expression of caspase 3 compared 350 to adjacent, nondegenerate, tubules (Fig 10O, Q-S) . Evidence of intra-tubule macrophages was 351 readily apparent (Fig. 10R, S) . SARS-CoV-2 infected cells can be identified in the JF82 testes with 352 triple staining for NSP8, nucleocapsid, and SARS-CoV-2 anti-sera (Fig 10T and U) . to LP14. This is consistent to similar viral loads between WA1 and the Delta variant in a recent 363 report 26 . 364

The probe signal associated with the MGT was not anticipated, but apparent in all 3 365 animals. This reveals that infection of MGT is consistently seen in the rhesus macaque model of 366 SARS-CoV2 IN/IT challenge. A comparison of the isolated MGT PET signal from the four whole-367 body PET scans of 3 SARS-CoV-2 infected rhesus macaques (Fig 11E-H) presented as a rotation 368 series further reveals the dynamics of SARS-CoV-2 after infection. The difference in MGT signal 369 between the animal infected with WA1 (LP14) and the animals infected with the Delta variant 370 19 (IN22 and JF82) at week 1 is much less pronounced than that seen in the lungs. In all week 1 371 MGT scans, the signal is asymmetrically distributed with diffuse signal throughout the testicles, 372 an increased signal at the base and top of the testes, and a signal associated with the root of 373 the penis (Fig 11E-G) . The MGT PET signal is visibly increased in the week 2 scan relative to the 374 week one scan (Fig 11G, H) for JF82 indicating further spread of infection into the MGT at that 375 time. Another obvious difference between the week 1 whole-body PET scans is a variable signal 376 associated with the heart and liver as shown in Fig 11I- K. LP14 had a diffuse PET signal 377 throughout the liver ( Fig 11I) . In contrast, the IN22 PET signal (Fig 11J) was primarily localized 378 with the right side of the liver and in JF82 the PET signal ( Fig 11K) was localized to the base of 379 the liver. Additionally, the extent of labeling of the heart is variable with JF82 and IN22 having a 380 greater signal than LP14 (Fig 11L) . 381

To take advantage of the quantitative aspects of PET detection, we isolated the total 382 standard uptake value (SUV) of the PET signal associated with the CT defined volumes as 383 plotted in Fig 11L. The 3 animals scanned 1 week (W1) post SARS-CoV-2 challenge are 384 presented together for the whole-body (WB) scan signal and all evaluated tissues. The single 385

week 2 (W2) PET signal of JF82 is presented for comparison. The WB values for all scans are 386 clustered revealing the reproducible nature of evaluation of the PET signal. An increase in the 387 total SUV between week 1 and week 2 for the MGT and testes volumes is consistent with the 388 increase of signal suggested by visual inspection of the isolated tissues (Fig 10, Fig 11G and H) . 389

Another relevant tissue with PET signal observed in all 3 animals is the prostate. Probe 390 labeling of the prostate first became apparent in the early PET scan (3 hr. after injection) of 391 IN22 where it was among the strongest, non-kidney associated signals (Fig 5A and B) . 392 20 Therefore, we revaluated the PET/CT data sets for the 3 animals. We were able to detect a 393 signal associated with the prostate in all animals (marked by white asterisks, Fig 12A- degeneration was noted in seminiferous tubules of JF82 (Fig. 10N ). Panels demonstrate normal 408 spermatogenesis or lack thereof in JF82 ( Fig. 12N , O, R, S). 409

In addition, all animals had a PET signal located at the top of each testicle, where the 410 spermatic cord connects with testicle. A dissection of the rhesus macaque testicular anatomy is 411 shown in Fig 13A- C. The spermatic cord contains the vasculature supplying blood to the 412 testicles, the vas deferens which transports mature sperm produced in the testes, and the 413 cremaster muscle (Fig 13A) . The position in natural context of the macaque penis, testes, and 414 21 spermatic cord are shown in Fig 13B. A magnified view of the vasculature of the pampiniform 415 plexus is shown in Fig 13C. As shown in the IN22 MGT PET/CT series (Fig 13D-G) , there is a 416 major signal associated with the top of the testicles, especially the right testicle. A further 417 examination of this signal within the PET/CT data set demonstrates it is located just above the 418 testicle in the yellow volume (Fig 13D-G) . This yellow volume is overlapping with a vasculature 419 structure consistent with the pampiniform plexus visualized by CT (Fig 10G and B) . The signal 420 associated with the pampiniform plexus and spermatic cord is seen in the 1-week scan of all 421 animals in front ( The PET signal includes several parameters in each anatomical area, each of which provide 427 different insights into the distribution of the probe within the tissue. For example, the total SUV 428 ( Fig 11L) , provides insights into the overall signal in each tissue/animal. However, the total SUV 429 does not account for variability in tissue size and shape. The mean SUV provides insights into 430 the relative intensity of PET signal in each tissue ( Fig 14A) . This comparison of mean SUV reveals 431 the relative intensity of the signal across tissues, with the prostate having consistently high 432 signal in all animals. In contrast, the total SUV/whole body Total SUV for the different tissues 433 illustrates the percentages of the total whole-body signal in each tissue without consideration 434 of the size of each tissue relative to the other ( Fig 14B) . All 7 PET parameters for each tissue are 435 shown in the heatmap (Fig 14C) , which reveals that different tissues have unique signal 436 22 characteristics. From this analysis, it is evident that the prostate signal of IN22 and JF82 have 437 the highest mean SUV and standard deviation with relatively a high median SUV being also seen 438 in the prostate of the 3 rd animal LP14. For example, the prostate signal of IN22 and JF82 has the 439 highest mean SUV and standard deviation with relatively a high median SUV being also seen in 440 the prostate of the 3 rd animal LP14, indicating a more clustered signal compared to the tissues 441 with the highest total SUV. 442

To facilitate the use of our PET data to gain insights into our data set we utilized 443

Principal Component Analysis (PCA) followed by hierarchical clustering to examine the data 444 from all scans of the 3 animals. To cluster tissues according to their SUV signal characteristics, 445

we applied PCA including all variables obtained from the SUV measurements (i.e., Total SUV, 446

Total SUV Whole Body ratio, Mean SUV, Median SUV, Standard Deviation SUV, Max SUV, and 447

Kurtosis) for each tissue and animal. After this analysis, we observed 4 different clusters that 448 correspond to tissues that display very distinct SUV signals. The clustering captures the prostate 449 signal described in the heat map where they are part of cluster 4 (dark blue group). Of note, the 450 major right lung signal of IN22 also clusters within this group, while the prostate signal of LP14 451 does not ( Fig 14D) . The evident PET signal observed in the lungs of the Delta variant animals 452 IN22 and JF82 are part of the cluster 3 (green group) ( Fig 14D) revealing their similarity with its 453 small foci of high signal and increased kurtosis value (indicating data heavy tails or more 454 outliers) consistent with the signal variability. The higher kurtosis is also associated with penile 455 and testes signal of all the animals in the red and light blue clusters 1 and 2. This analysis allows 456 us to begin to appreciate the nature of the PET signal associated with the different tissues, 457 different animals, different viruses, and identify outliers. Further validation of the PET/CT probe's ability to identify areas of infection at sites 485 other than the lungs and its utility in understanding COVID-19 pathogenesis was the revelation 486 of a reproducible infection of the MGT. In all three animals, the probe was associated with the 487 prostate, penis, pampiniform plexus, and testicles. We have been able to identify SARS-CoV-2 488 infected cells in the testicles of all 3 animals. Comparing the longitudinal PET scans of JF82 489 reveals that while the lung pathology and signal wane between week 1 and 2, the signal of the 490 MGT, and more specifically in the testicles increases in week 2. Consistent with increased 491 testicular infection at week 2 indicated in the PET scan, H&E staining of the JF82 testes revealed 492 a unique pathology consisting of denuded stretches of the twisted and intertwined 493 seminiferous tubules (presenting as a tube cluster). Spermatids were absent in these 494 degenerate regions and the remaining Sertoli cells were undergoing apoptosis, seen through 495 caspase 3 staining (Fig. 10S ). Local inflammation, or orchitis, was suggested by immune 496 infiltrates (Fig. 10R ). Both the PET signal and pathology is greater in the left testicle of JF82. 497

Staining macrophages and mature spermatids for CD206 and all cells for caspase 3 activation 498 reveals a severe, acute response within short stretches of the seminiferous tubules where the 499 Sertoli cells are undergoing apoptosis due to inflammasome activation (caspase 3 activation), 500 and no spermatids are present consistent with an ongoing acute infection (Fig. 10N-S) . We 501 have detected infected cells within the JF82 testes ( (Fig. 10T , U) and the relationship between 502 25 the infected cells and testicular pathology is ongoing. Similar decreases in the cellular content 503 of the seminiferous tubules and sloughing of Sertoli cells and spermatids into the lumen have 504 been reported in multiple studies of autopsy tissues from COVID-19 related fatalities 35-40 . 505 Our results suggest that SARS-CoV-2 rapidly and efficiently infects multiple tissues of the 506 male genital tract (MGT) early during infection in rhesus macaques. The complex vasculature 507 and known ACE2 expression of the tissues of the MGT make it a potential target of the virus 11, 508 [41] [42] [43] . The SARS-CoV-2 infection of the testicles has been reported in mouse and hamster 509 respiratory challenge models 44-46 . Likewise, the testicles are also a target of Ebola and Zika virus 510 during systemic infection 47 . We observed a similar distribution of PET-probe signal in all 3 511 animals in the week 1 scan with labeling of the prostate, root of the penis, the top the testicles 512 and a second region of labeling at the base of the testicles (Fig. 11E-H) . The signal above the 513 testicles localizes to the pampiniform plexus and vasculature of the spermatic cord while the 514 signal at the base of the testes is less clear, but appears to be associated with the cauda 515 epididymis, the highly vascularized tail of the epididymis that serves as the storage site for 516 mature sperm. A further dissection of the testicles before the organ scan should facilitate a 517 detailed localization of the PET signal associated with the MGT in future studies. 518

Although these studies were done with a rhesus macaque model, it is reasonable to 519 suggest that these observations may also apply to humans infected with SARS-CoV-2 because of 520 several clinical observations relating to male sexual health and fertility. It is highly relevant in 521 this extrapolation to consider that we have identified 4 distinct tissues where SARS-CoV-2 522 infection could impact male sexual health and fertility: SARS-CoV-2 infection of the prostate, 523 penis, pampiniform plexus, and testicles. The infection of the MGT and associated pathology 524 26 has been suggested by several publications and clinical studies. The prostate is known to be 525 ACE2 positive. 48 Interest in SARS-CoV-2 infection of the prostate has focused on two areas. First 526 is the potential impact on treatment of benign prostate hyperplasia 49 and prostate cancer 50 527 with androgen deprivation therapy on the severity of COVID-19 51 and secondly, the potential 528 impact of SARS-CoV-2 infection on prostate cancer treatments. It is notable that prostate 529 cancers are known to express high levels of transmembrane serine protease 2, TMPRSS2 52 , 530 which is known to activate the SARS-CoV-2 spike protein to its optimal fusogenic potential 53 . 531

Multiple studies have explored this space and it does not appear that SARS-CoV-2 infection is 532 associated with an increase in prostate cancer 54 . In contrast, another study suggests that 533 infection with SARS-CoV-2 may be associated with an increase in prostate specific antigen (PSA) 534 detection in plasma 55 . Future studies are needed to confirm whether the robust signal of the 535 SARS-CoV-2 PET/CT probe reflects a high-level infection of the human prostate and its 536 subsequent impact on male sexual health and fertility 56, 57 . SARS-CoV-2 infection of the penis is 537 potentially associated with the vasculature of the corpus cavernosum, which expresses high 538 levels of ACE2 in the rhesus macaque and human penile tissue ( Fig 7S) 43, 58 . Because the corpus 539 cavernosum plays a key role in erectile function, the inflammation caused by SARS-CoV-2 540 infection of the penile vasculature is hypothesized to lead to erectile dysfunction (ED). This has 541 indeed been reported to be linked to 59, 60 . In addition, treatments for ED such as 542 Viagra and Cialis are known to affect the renin-angiotensin-aldosterone-system where ACE2 543 functions as a part of the physiologic regulation of blood flow associated with normal erectile 544 function 61 . 545 27 A potential impact of COVID-19 infection on the pampiniform plexus might be suggested 546 by several case reports of COVID-19 associated thrombosis located in the pampiniform plexus 62-547 65 . Additionally, the signal distribution of the left and right testes is distinct, with the signal of 548 the right testicle being more focused at the top of the testicle while the signal on the left 549 testicle being more distributed in the spermatic cord. This is reminiscent of the condition 550 known as varicocele, which manifests as varicose veins of the scrotum, and is prominent in the 551 left testicle relative to the right testicle 66 . This is due to the left testicle receiving its blood flow 552 from the left renal vein which exposes it to higher blood pressure and slower blood flow 67 . This 553 difference could be insightful if it is confirmed in more animals. 554

The potential infection of the testicles by SARS-CoV-2 could be highly impactful on male 555 fertility, potentially decreasing sperm count and semen quality 47, 68-70 . It is known that SARS-556

CoV-2 infection in humans is associated with oligo-and azoospermia and a transient decrease 557

in fertility after infection 36, 38, 66, 68, 71, 72 . One study found that fertility amongst infected men 558 was reduced and returned to baseline 3-6 months after SARS-CoV-2 infection 73 . This decrease in 559 fertility was not seen in infected women or men who received a SARS-CoV-2 vaccination. We 560 find that the pathology associated with the testicles in the week 2 necropsy is extreme, with 561 apparent ablation of sperm production within short regions of the seminiferous tubules and 562 with accompanying immune infiltration consistent with an emerging COVID-19 associated 563 infection. We believe the immunoPET technique described here will be an important addition 582 to the toolkit for studying and understanding SARS-CoV-2 pathogenesis. The availability of an 583 immunoPET probe to SARS-CoV-2 in the clinical setting has the potential to reveal the 584 underlying role of disseminated viral infection in long COVID and could guide therapeutic 585 interventions to resolve SARS-CoV-2 related sequalae which could be a major health concern 586 for the lifetimes of those infected during the COVID-19 pandemic. PET/CT guided necropsies were performed in three sequential phases each separated by a 659 PET/CT followed by a period of analysis and sampling (1. whole-body, 2. organ, and 3. tissue). 660

First, whole-body scans were acquired prior to sending the animal to necropsy. Following 661 euthanasia, all major organ systems were removed (pluck, gastrointestinal tract, liver, spleen, 662 kidneys, urinary bladder, testicles, penis, prostate, seminal vesicles, nasal turbinate, lymph 663 nodes, carotid artery, cervical spinal cord, and brain) placed in a clear, plastic, sealable 664 container and sent back to PET/CT for an "organ scan". After the organ scans were 665 reconstructed, "hot" regions of each major organ (as seen on the organ scan) were sampled 666 and placed in cryomolds. The final "tissue" scan was acquired by placing the cryomolds in a 667 clear, plastic, sealable container and scanning them with the PET/CT. Following acquisition of 668 the tissue scan, cryomolds were filled with OCT and frozen on dry ice. All remaining tissue (not 669 placed in cryomolds) was placed in zinc-formalin fixative. All samples were stored for 5 days 670 before being removed from containment. 671

Acquired PET/CT whole-body images were analyzed using the MIM Software (MIM Software  674 Inc., Cleveland, OH). The PET and CT scans were reconstructed using the software and PETCT 675 fusions were created to analyze regions of interest through axial, sagittal, and coronal views. 676

The PET scans were presented in calculated Standardized Uptake Values, and all images were 677 set to the same scale (0-20 SUVbw). The PET scale was selected based on the overall signal 678 intensity of the PET scans, and the CT scale for optimal visibility of the tissues. Regions of 679 interest (ROI) were isolated using a combination of the Region Grow function and manual 680 contouring on a representative scan, then these regions were copied on subsequent scans of 681 the same animal using a specialized developed workflow. This workflow allows the software to 682 use the CT scans to map the selected ROI and locate that exact volume in subsequent scans. 683

Manual adjustments were then used to counter any changes in the animals' orientation 684 between scans. The areas within these regions were then extracted from the full scans, and the 685 anatomical regions were analyzed in both 2D cross-sections and 3D projections. 3D views are 686 maximum intensity projections of isolated ROI in the PET scans. Fisher Scientific), vortexed, and pulse centrifuged. The RT-qPCR reaction was subjected to RT-715 qPCR at a program of uracil-DNA glycosylase incubation at 25°C for 2 minutes, room 716 temperature incubation at 50°C for 15 minutes, and an enzyme activation at 95°C for 2 minutes 717 followed by 40 cycles of a denaturing step at 95°C for 3 seconds and annealing at 60°C for 30 718 seconds. Fluorescence signals were detected with a QuantStudio 6 Sequence Detector (Applied 719 Biosystems, Foster City, CA). Data were captured and analyzed with Sequence Detector 720 Software version 1.3 (Applied Biosystems). Viral copy numbers were calculated by plotting Cq 721 values obtained from unknown (i.e., test) samples against a standard curve that represented 722 known viral copy numbers. The limit of detection of the viral RNA assay was 10 copies per 723 reaction volume. A 2019-nCoV-positive control (catalog number 10006625; IDTDNA) was 724 analyzed in parallel with every set of test samples to verify that the RT-qPCR master mix and 725 reagents were prepared correctly to produce amplification of the target nucleic acid. A non-726 template control was included in the qPCR to ensure that there was no cross-contamination 727 between reactions. 728 OCT embedded tissue blocks were sectioned between 10-15 µM and 2-4 sections were 729 placed into RNAse free tubes. RNA extraction was carried out using a RNeasy Plus Mini Kit 730 (#74124, Qiagen), according to manufacturer's protocol. Briefly, 350ul of lysis buffer with 731 dithiothreitol (DTT) was added to the tubes containing sections. The tubes were vortexed 732 briefly then frozen at -20C. Once thawed the samples were again vigorously vortexed then 733 centrifuged for 3 minutes. Supernatant was removed and applied to the gDNA Eliminator spin 734 column. The flow-through was mixed with 350ul of 70% ethanol and added to a RNeasy spin 735 column. The spin column was washed three times with buffers RW1 and RPE. RNA was then 736 eluted using 30-50ul of RNAse free water. All steps prior to addition of lysis buffer were carried 737 out in a BSL3 facility. 738 739

Tissue samples were collected in Zinc formalin (Anatech) and fixed for a minimum of 72 hours 741 before being washed and dehydrated using a Thermo Excelsior AS processor. Upon removal 742 from the processor, tissues were transferred to a Thermo Shandon Histocentre 3 embedding 743 station where they were submersed in warm paraffin and allowed to cool into blocks. From 744 these blocks, 5um sections were cut and mounted on charged glass slides, baked overnight at 745 60 o C, and passed through Xylene, graded ethanol, and double distilled water to remove paraffin 746 and rehydrate tissue sections. A Leica Autostainer XL was used to complete the 747 deparaffinization, rehydration and routine hematoxylin and eosin stain. Slides were digitally 748 imaged with a NanoZoomer S360 (Hamamatsu) and subsequently examined by a board-749 certified, veterinary pathologist using HALO software (Indica Labs). 750 751 Fluorescent Immunohistochemistry of FFPE tissues 752 5um sections of Formalin-fixed, paraffin-embedded (FFPE) tissues were mounted on charged 753 glass slides, baked for two hours at 60 o C, and passed through Xylene, graded ethanol, and 754 double distilled water to remove paraffin and rehydrate tissue sections. A microwave was used 755 for heat induced epitope retrieval. Slides were heated in a high pH solution (Vector Labs H-756 3301), rinsed in hot water, and transferred to a heated low pH solution (Vector Labs H-3300) 757

where they were allowed to cool to room temperature. Sections were washed in a solution of 758 phosphate-buffered saline and fish gelatin (PBS-FSG) and transferred to a humidified chamber, 759 for staining at room temperature. Lung sections were blocked with 10% normal goat serum 760 (NGS) for 40 minutes, followed by a 60-minute incubation with the anti-SARS primary antibody 761 diluted in NGS. Slides were washed and transferred back to the humidified chamber for a 40-762 minute incubation with the secondary antibody, also diluted in NGS. Sequential staining of FFPE 763 testes, for CD206 and Caspase 3, was done as described above with 1% normal donkey serum 764 37 (NDS) being used in place of NGS for blocking and antibody dilutions. DAPI was used to label the 765 nuclei of each section. Slides were mounted using a homemade anti-quenching mounting 766 media containing Mowiol (Calbiochem #475904) and DABCO (Sigma #D2522) and imaged at 767 OCT embedded tissue were cryosectioned in a BSL3 facility between 10 and 15 µM. One or two 772 sections of each tissue were placed on glass microscope slides. Tissues were placed into an 773 airtight container containing 4% PFA in PIPES buffer. The container was sealed and thoroughly 774 decontaminated before being removed from the BSL3 facility. The remainder of the staining 775 took place outside of the BSL3. Tissue sections were treated with L-lysine (0.1%, SigmaAldrich) 776 to reduce background and non-specific interactions before being blocked using 3% BSA 777 (Invitrogen, ThermoFisher) for 30 minutes at room temperature. Staining with primary 778 antibodies was carried out at 37 o C and secondary antibodies at room temperature. each for 779 1hr. Table of primary and secondary antibodies used is below. All slides were stained with 780 38 Hoechst (1:25,000, ThermoFisher, USA) for 15 minutes and washed with PBS between all steps. 781

Dako fluorescent mounting medium (#S302380-2, Agilent, USA) was used to mount cover slips 782 which were sealed with nail polish. A DeltaVision Ultra inverted microscope (Cytivia, USA) was 783 used to obtain images using the 10x, 20x, 60x, and 100x lenses. Images were deconvolved, 784 stitched, and projected using the softWoRx software (Applied Precision, USA). 785 

Residual SARS-CoV-2 viral antigens detected in GI and hepatic tissues 967 from five recovered patients with COVID-19

Gastrointestinal 969 manifestations in COVID-19

Ischemic 972 gastrointestinal complications of COVID-19: a systematic review on imaging presentation

Gut in COVID 19-is it worth noticing

Wastewater surveillance: an effective and adaptable 979 surveillance tool in settings with a low prevalence of COVID-19

Wastewater by Amplicon Sequencing and Using the Novel Program SAM Refiner

Epub 20210819

SARS-CoV-2 is localized in 988 cardiomyocytes: a postmortem biopsy case

Evidence for residual SARS-CoV-2 in 991 glioblastoma tissue of a convalescent patient

Neuroinvasion of SARS-CoV-2 in human and mouse brain

Histopathological findings and viral tropism in UK patients with severe fatal 1003 COVID-19: a post-mortem study

Implications of testicular ACE2 and the renin-angiotensin 1006 system for SARS-CoV-2 on testis function

Impact of COVID-19 1009 on Male Fertility

Addressing the 1012 post-acute sequelae of SARS-CoV-2 infection: a multidisciplinary model of care

Orchitis: a complication of severe 1020 acute respiratory syndrome (SARS)

Effective prophylaxis of COVID-19 in rhesus macaques using a combination of two parentally-1025 administered SARS-CoV-2 neutralizing antibodies

Acute Respiratory Distress in Aged, SARS-CoV-2-Infected 1031

African Green Monkeys but Not Rhesus Macaques. The American Journal of Pathology

SARS-CoV-2 infection protects against rechallenge in rhesus macaques

Cellular events of acute, resolving or 1044 progressive COVID-19 in SARS-CoV-2 infected non-human primates

Acute Respiratory Distress in Aged, SARS-CoV-2-Infected 1051

African Green Monkeys but Not Rhesus Macaques

Whole-body immunoPET reveals active SIV dynamics in viremic and antiretroviral 1056 therapy-treated macaques

Localization of infection in neonatal rhesus macaques after oral viral 1061 challenge

PET/CT targeted tissue sampling reveals virus specific dIgA 1066 49 can alter the distribution and localization of HIV after rectal exposure

Development of an In 1072 Vivo Probe to Track SARS-CoV-2 Infection in Rhesus Macaques. Front Immunol

Cellular events of acute, resolving or 1078 progressive COVID-19 in SARS-CoV-2 infected non-human primates

Infection and pathogenesis of the Delta variant of SARS-CoV-2 in Rhesus 1082 macaque. Virologica Sinica

Testicular activin and follistatin levels are 1085 elevated during the course of experimental autoimmune epididymo-orchitis in mice

The 1093 role of mannose receptor on HIV-1 entry into human spermatozoa

1096 Fertilization potential in vitro is correlated with head-specific mannose-ligand receptor 1097 expression, acrosome status and membrane cholesterol content

Evaluation of the mRNA-1273 Vaccine against SARS-CoV-2 in Nonhuman

Visualization of early events in 1113 mRNA vaccine delivery in non-human primates via PET-CT and near-infrared imaging

Condition WHOCCDWGoP-C-. A 1116 clinical case definition of post-COVID-19 condition by a Delphi consensus

Cooke 1120 G. Global surveillance, research, and collaboration needed to improve understanding and 1121 management of long COVID

Testicular pathology in fatal COVID-19: A descriptive autopsy study

COVID-19 disrupts spermatogenesis 1130 through the oxidative stress pathway following induction of apoptosis

Pathological Findings in the Testes of COVID-1135 19 Patients: Clinical Implications

COVID-19 disrupts the blood-testis barrier through the induction of 1140 inflammatory cytokines and disruption of junctional proteins

Histopathology and Ultrastructural Findings of Fatal COVID-19 Infections 1145 on Testis

Pathological and molecular examinations of postmortem testis biopsies 1149 reveal SARS-CoV-2 infection in the testis and spermatogenesis damage in COVID-19 patients. 1150 Cellular and Molecular Immunology

COVID-19 and male reproductive system: pathogenic 1152 features and possible mechanisms

Single-cell transcriptome analysis of the novel coronavirus (SARS-CoV-2) associated gene ACE2 1156 expression in normal and non-obstructive azoospermia (NOA) human male testes

Impacts of COVID-19 and SARS-CoV-2 1160 on male reproductive function: a systematic review and meta-analysis protocol

Severe acute respiratory 1165 syndrome coronavirus 2 (SARS-CoV-2) infections by intranasal or testicular inoculation induces 1166 testicular damage preventable by vaccination in golden Syrian hamsters

Epub 20210617

Live 1175 imaging of SARS-CoV-2 infection in mice reveals that neutralizing antibodies require Fc function 1176 for optimal efficacy

Comparative analysis of viral infection outcomes in human seminal fluid from prior viral 1180 epidemics and Sars-CoV-2 may offer trends for viral sexual transmissibility and long-term 1181 reproductive health implications

Is 1187 COVID-19 a risk factor for progression of benign prostatic hyperplasia and exacerbation of its 1188 related symptoms?: a systematic review

TMPRSS2 and COVID-19: 1191 Serendipity or Opportunity for Intervention?

Androgen deprivation therapy and 1195 excess mortality in men with prostate cancer during the initial phase of the COVID-19 1196 pandemic

Prognostic and predictive molecular biological markers in prostate cancer -1200 significance of expression of genes PCA3 and TMPRSS2

Gong 1203 J. COVID-19 and androgen-targeted therapy for prostate cancer patients

The clinical impact of androgen 1207 deprivation therapy on SARS-CoV-2 infection rates and disease severity

Variation of Serum PSA Levels in COVID-19 Infected Male Patients with 1211

A Prospective Cohort Studys

Sex differences in COVID-19: the role of 1215 androgens in disease severity and progression

SARS-CoV-2 in the 1218

The World Journal of Men's Health

Diminazene protects corpus cavernosum 1222 against hypercholesterolemia-induced injury

COVID-19 Endothelial Dysfunction Can Cause 1226

Erectile Dysfunction: Histopathological, Immunohistochemical, and Ultrastructural Study of the 1227 Human Penis

Mask 1229 up to keep it up": Preliminary evidence of the association between erectile dysfunction and 1230 COVID-19

The war against the SARS-CoV2 infection: Is it better to 1232 fight or mitigate it? Med Hypotheses

Venous thrombosis of 1235 the pampiniform plexus after coronavirus infection (COVID-19): A case report

An unusual and atypical 1239 presentation of the novel coronavirus: A case report and brief review of the literature

Testicular vein thrombosis mimicking 1243 epididymo-orchitis after suspected Covid-19 infection

Testicular pain as an 1247 unusual presentation of COVID-19: a brief review of SARS-CoV-2 and the testis

Epub 20200723

Evaluation of testicular spermatogenic 1251 function by ultrasound elastography in patients with varicocele-associated infertility

Haemodynamic aspects of left-sided 1255 varicocele and its association with so-called right-sided varicocele

COVID-19 and human spermatozoa-Potential risks for infertility and sexual 1258 transmission?

Serni 1262 S. Semen impairment and occurrence of SARS-CoV-2 virus in semen after recovery from COVID-1263 19

The impact of COVID-19 on the male reproductive 1265 tract and fertility: A systematic review

Impaired spermatogenesis in COVID-19 patients

Could 1271 SARS-CoV-2 infection affect male fertility and sexuality?

A prospective cohort study 1275 of COVID-19 vaccination, SARS-CoV-2 infection, and fertility

Acute Respiratory Syndrome (SARS)

COVID-19 and 1281 male fertility: Taking stock of one year after the outbreak began

Testosterone in COVID-19: 1284 An Adversary Bane or Comrade Boon

Roozbeh 1287 J. Potential mechanisms of SARS-CoV-2 action on male gonadal function and fertility: Current 1288 status and future prospects

Prospective two-arm study of the testicular function in 1292 patients with COVID-19

A and B) Whole-body PET/CT scans of LP14 8 days 1166 post-infection. Front view (A) and rotated 45° (B) both shown. PET signal is display as SUV

Side (G) and front (H) view. (I and J) Isolated 1170 lung PET volumes used in G and H are shown independently, side view (I) and front view (J). (K 1171 and L) Single axial z-slice images of respiratory tract PET/CT signals are shown, each image 1172 represents a single z-plane from scan shown in C. (M and N) Fluorescent microscopy images of 1173 LP14 lung tissue blocks. Red is SARS-CoV-2 anti-sera and blue is Hoechst nuclear stain

/CT images highlighting the lower 1179 abdomen of LP14 from the whole-body scan. Front (A), rotated 45° (B), and side (C) views are 1180 all shown. Right and left labeled in front view. Red arrow in B and C shows location of baculum 1181 in CT scan. (D and E) PET signal with CT overlay removed to highlight signal in MGT, front (D) 1182 and side (E) views shown. (F, H, and J) Isolated 3D volume of MGT from whole body scan 1183 overlaid with CT images. Side (F)

PET signal from 1186 single z-plane of organ scan used in (L). (N) Image of LP14 testis used in (L). (O) 3D volume of 1187 PET signal from organ scan of single testis in previous panels. 1188 1189 52 1190 1191 Figure 4. Immunofluorescence of LP14 Testes. (A) Fluorescent microscopy image of LP14 1192 seminiferous tubules. ACE2 staining shown in red, Hoechst nuclear staining shown in blue. Inset 1193 shows zoom in of single tubule to better view ACE2 staining in Sertoli and myoid cells. Scale 1194 bars 50 µM. (B) Fluorescent microscopy image of LP14 testis shows infected cells. SARS-CoV-2 1195 anti-sera staining in green, background fluorescent in red, and Hoechst nuclear staining in blue. 1196 Scale bars 25 µM. (C) Microscopy images of two tubules containing infected cells (top and 1197 bottom rows). Red is SARS-CoV-2 anti-sera, green is smooth muscle actin

A and B) Whole-body PET/CT scans of IN22 1205 obtained 3-hours after probe administration. Front view (A) and side (B) both shown

Whole-body PET/CT scans of IN22 obtained 21-hours after probe administration. Front view (C) 1207 and side (D) both shown. Right and left labeled in front views (A and C). (E) Organ tray post 1208 necropsy and (F) PET/CT image. (G) Second organ tray post necropsy and (H) PET/CT image

A and B) Show lung volumes from 3-hour scan, side (A) and 1216 front (B) views shown. (C and D) Show lung volumes from 21-hour scan, side (C) and front (D) 1217 views shown. (E and F) Isolated lung PET volumes for each scan are shown independent of CT. 1218 (G) Sagittal z-slice from CT showing lungs. (H) PET signal overlaid on z

Transverse z-slice through torso from CT. (J) PET signal overlaid on z-slice from I. (K) Image of 1220 respiratory tract after necropsy. (L) Inset showing overt lung pathology in right lower lobe

/CT signal from M overlaid onto 1222 image from K. (O) PET/CT signal with CT contrast increased to observe pathology in lower right 1223 lung lobe. (P) H&E image of lung tissue showing areas of expanded alveolar space and 1224 inflammatory infiltrate (arrows)

White arrows 1226 indicate SARS-CoV-2 positive cells. (R) H&E image of lung tissue showing macrophages and 1227 neutrophils (arrowheads) and type II pneumocytes (arrows). (S) Immunofluorescence image 1228 showing infected cells of the alveoli (arrows) and lining the alveolar septa (arrowheads)

Fluorescent microscopy image of IN22 lung tissue. Spike shown in green, nucleocapsid shown in 1231 red, background in white, and Hoechst nuclear stain in blue. Scale bar 100 µM

Additional fluorescent microscopy images of lung tissue showing foci of infected cells. U and W 1233 are shown in low magnification Y. Green is SARS-CoV-2 anti-sera and blue is Hoechst nuclear 1234 stain. Scales bar 500 µM. (X-Z) Validation of dual antibody staining utilizing spectral imaging. (X) 1235 shows microscopy of J2 antibody with redX secondary and rabbit anti-NC monoclonal and Cy5 1236 secondary. (Y) Shows area within green square in X. White arrows point to regions of interest 1237 that are cell associated. Grey arrows indicate control regions of spectral evaluation. The areas 1238 evaluated by spectral imaging

A and B) PET/CT images highlighting the lower 1243 abdomen of IN22 obtained 3-hours after probe administration. Front view (A) and side (B) both 1244 shown. (C and D) PET/CT images highlighting the lower abdomen of IN22 obtained 21-hours 1245 after probe administration. Front view (C) and side (D) both shown. (E and F) Isolated 3D 1246 volume of MGT from 3-hour scan overlaid with whole-body CT

G and H) Isolated 3D volume of MGT from 21-hour scan overlaid with whole-body CT

Front (I) and side (J) views shown. (K and L) 1250 3D volume of MGT overlaid onto CT images from previous panels. (M) Image of testicles after 1251 necropsy. (N) PET signal from organ scan of testicles. (O) Overlay of PET signal onto image from 1252 panel M. (P) Overlay of PET signal onto CT signal from same scan. White asterisks mark location 1253 of pampiniform plexus in all previous panels. (Q and R) Fluorescence microscopy of SARS-CoV-2 1254 infected cells in testicular tissue from IN22. Red is SARS-CoV-2 spike, green is SARS-CoV-2 1255 nucleocapsid, white is background, and blue is Hoechst nuclear stain. Insets show channels 1256 independently, larger image is all channels merged. Scale bars 10 µM (S) Fluorescent 1257 microscopy image of corpus cavernosum tissue from an uninfected animal showing ACE2 1258 staining in red

White is 1260 dsRNA antibody J2, red is SARS-CoV-2 nucleocapsid, green is background, and blue is Hoechst 1261 nuclear stain. (U, V, and W) Show individual cells highlighted in T

PET/CT scan of JF82 from 1-week post-infection. Front view (A) and side (B) both shown

Front view (C) and side (D) both 1297 shown. Right and left labeled in front views (A and C). (E and F) Overlay of the week 1 scan 1298 62 (shown in red) and the week 2 scan (shown in blue). Front (E) and side (F) views both shown. 1299 (G) Organ tray post necropsy and (H) PET/CT image. (I) Second organ tray post necropsy

Overlay of PET signal onto photograph of tissue cassettes

A) and front (B) views shown. White asterisk indicates location of lung pathology 1309 highlighted below. (C and D) Show lung volumes from 2-week scan, side (C) and front (D) views 1310 shown. (E and F) Isolated lung PET volumes for each scan are shown independent of CT

H) Overlay of single z-image of PET signal onto single z-image of CT in the lungs at week 1 (G) 1312 and week 2 (H). (I and J) Single z-image of CT used in G and H shown independent of PET signal 1313 for week 1 (I) and week 2 (J). (K) Overlay of week 1 (shown in red) and week 2 (shown in blue) 1314 PET signal. (L) Overlay from K shown with CT image to localize PET signal

White asterisk indicates location of left lung 1316 pathology in all panels

Male genital tract signal in JF82. (A and B) Front view PET/CT images highlighting the 1356 lower abdomen of JF82 at 1-week (A) and 2-weeks (B). (C) Overlay of week 1 (shown in red) and 1357 week 2 (shown in blue) PET signal. (D and E) Side view of same images shown in A and B

Isolated MGT volume from week 2 with isolated penile and 1361 testicular signal. Back (K), rotated 45° (L), and side (M) views shown. (N) H&E stain of JF82 1362 testicular tissue. Degenerate seminiferous tubules highlighted in black oval. Scale bar 500 µM. 1363 (O) Fluorescent microscopy shows a similar area of degenerate tubules. Green is CD206, red 1364 caspase 3, and white nuclear stain. Scale bar 500 µM. (P) Higher magnification image of 1365 degenerate tubules. Intraluminal macrophages (arrowheads) and Sertoli cells (arrows) are 1366 shown inside tubules. Scale bar 50 µM (Q) Higher magnification of image in O shows Sertoli 1367 cells (arrow) staining with caspase 3 and macrophages (arrowhead) staining with CD206. Scale 1368 bar 50 µM. (R) Fluorescent microscopy image showing degenerate tubule full of intraluminal 1369 macrophages

Red is caspase 3, 1371 green is CD206, and blue is nuclear stain. Scale bar 200 µM. (T and U) Fluorescent microscopy 1372 images of infected cells in testicular tissue of JF82. White is SARS-CoV-2 anti-sera, red is NSP8, 1373 green is nucleocapsid, and blue is Hoechst nuclear stain. Insets show individual channels, larger 1374 image is merge. Scale bars 10 µM

IN22 (F), JF82 week 1 (G), and 1406 JF82 week 2 (H). (I-K) Front and side views of whole-body scans, white lines indicate volumes 1407 taken for heart and lungs for LP14 (I), IN22 (J), and JF82 week 1 (K). Insets show each image 1408 without white outlines. (L) Total SUVs for whole-body scans and each individual volume isolated 1409 displayed in graph. Animals are indicated by icon shape and volumes by color. 1410 1411 1412 70 1413 1414 Figure 12. Comparison of prostate and penile signal between animals

Asterisks mark location of 1417 prostate. Insets show sagittal z-slice of each animal highlighting prostate signal

G-I) PET/CT volume of penis for IN22. Front (G), rotated 45° (H), and side (I) views shown

M) PET/CT signal of IN22 penis after necropsy. (N) PET/CT signal of JF82 penis 1421 after necropsy. (O) PET/CT signal overlaid onto an image of tissue cassettes (P) containing 1422 penile tissue of IN22. (Q) PET/CT signal overlaid onto an image of tissue cassettes (R) containing 1423 penile tissue of JF82. (S-U) H&E images of testicular tissue from each animal. LP14 (S) and IN22 1424 (T) shows normal spermatogenesis and tissue architecture. IN22 (U) shows degenerate tubules 1425 (asterisks) interspersed among healthy tubules

A) Labeled dissection showing 1430 anatomical structure of a macaque testis. (B) CT image of testes and associated image showing 1431 the matching anatomical features of the spermatic cord. (C) Inset highlighting the location and 1432 73 appearance of the pampiniform plexus

N) Sagittal z-slice of PET/CT 1438 of LP14 showing right testis. (O) Frontal z-slice of PET/CT, colored lines correspond to sagittal 1439 slices shown in N, P, and R. (P) Sagittal z-slice of PET/CT showing penile tissue. (Q) Frontal z-slice 1440 highlighting the testicular tissue in white ovals. (R) Sagittal z-slice of PET/CT showing left testis 1441 of LP14. White ovals highlight signal associated with testes and not pampiniform plexus

Heat map showing clustering of tissues and parameters measured from the PET data