key: cord-1040549-mla3xde3
authors: Zhu, Hua; Zhang, Hanwen; Zhou, Nina; Ding, Jin; Jiang, Jinquan; Liu, Teli; Liu, Ziyu; Wang, Feng; Zhang, Qian; Zhang, Zhuochen; Yan, Shi; Li, Lei; Benabdallah, Nadia; Jin, Hongjun; Liu, Zhaofei; Cai, Lisheng; Thorek, Daniel L. J.; Yang, Xing; Yang, Zhi
title: Molecular PET/CT Profiling of ACE2 Expression In Vivo: Implications for Infection and Outcome from SARS‐CoV‐2
date: 2021-06-26
journal: Adv Sci (Weinh)
DOI: 10.1002/advs.202100965
sha: e324d437e78d9424a99d4d90f6a15bd0e368e49f
doc_id: 1040549
cord_uid: mla3xde3

Rapid progress has been made to identify and study the causative agent leading to coronavirus disease 2019 (COVID‐19) but many questions including who is most susceptible and what determines severity remain unanswered. Angiotensin‐converting enzyme 2 (ACE2) is a key factor in the infection process of severe acute respiratory syndrome coronavirus‐2 (SARS‐CoV‐2). In this study, molecularly specific positron emission tomography imaging agents for targeting ACE2 are first developed, and these novel agents are evaluated in vitro, in preclinical model systems, and in a first‐in‐human translational ACE2 imaging of healthy volunteers and a SARS‐CoV‐2 recovered patient (NCT04422457). ACE2 expression levels in different organs in live subjects are quantitatively delineated and observable differences are measured in the patient recovered from COVID‐19. Surprising sites of uptake in the breast, reproductive system and very low uptake in pulmonary tissues are reported. This novel method can add a unique tool to facilitate SARS‐CoV‐2 related research and improve understanding of this enigmatic disease. Molecular imaging provides quantitative annotation of ACE2, the SARS‐CoV‐2 entry receptor, to noninvasively monitor organs impacted by the COVID‐19.

The ongoing epidemic caused by severe acute respiratory syndrome CoV-2 (SARS-CoV-2) has emerged as a global health and economic crisis of unprecedented magnitude. [1] The World Health Organization (WHO) officially declared the novel coronavirus disease 2019 (COVID-19) a pandemic on March 11, 2020. [2] To date, more than 106.2 million cases and over 2.31 million deaths have been attributed to COVID-19 worldwide. [3] The majority of the infected exhibit mild or no symptoms. However, multiple organs are at risk of rapid and sustained damage and failure in those who have adverse manifestations of the disease. Dozens of studies, many ongoing, have attempted to clarify the susceptibility of COVID-19 and the risk factors contributing to its serious complications. [4, 5] While age and on SARS-CoV-2 infection and COVID-19 severity, we initiated this study to develop 64 Cu and 68 Ga-labeled peptides to specifically targeting ACE2. We have developed a class of ACE2-specific PET radioligands with excellent imaging and pharmacokinetic properties. Preclinical model systems were used to validate and optimize this approach to quantify ACE2 expression in vivo using 68 Ga-and 64 Cu-labeled analogs of the ACE2-targeting DX600 peptide:, 68 Ga-HZ20 and 64 Cu-HZ20, respectively ( Figure 1B) . The human-specific ACE2 inhibiting cyclic peptide DX600 was 68 Ga-HZ20 that was advanced to the clinic for PET/CT imaging and organ-based standardized uptake value (SUV analysis) of 20 healthy volunteers. A patient who had recovered from SARS-CoV-2 infection was included, and demonstrated greater 68 Ga-HZ20 SUV max (a standardized measure of tracer accumulation). Complementing pathological analyses, these intriguing results are expected to demonstrate that radiolabeled HZ20 analogs have substantial value in quantifying the ACE2 distribution in the entire body, and monitoring the transient upregulation of ACE2 expression to guide patient management and evaluate therapeutic intervention of COVID-19 patients.

DX600 is a high affinity ACE2 binding cyclic peptide with an intramolecular disulfide bond initially discovered from a phage display screen for ACE2 inhibition. [22] It shows high selectivity on ACE2 over ACE and is stable to ACE2 catalyzed hydrolysis. We hypothesized the potency of the ligand would not be significantly diminished through modification with a radio-metal chelator, such as 1-(1,3-carboxypropyl)-1,4,7-triazacyclononane-4,7-diacetic acid (NODAGA) for stable 64 Cu labeling or 1,4,7,10tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) for stable 68 Ga labeling, at the N-terminus of DX600 as it is distant from the key binding domain of CSPLRYYPWWKC. Chelate conjugated-DX600 was obtained in high purity (>98%) and characterized by mass spectrometry (Figure 1 and Figure S1 , Supporting Information), which are matching with the structure-based estimation. 64 Cu-HZ20 ( Figure 1B ) was prepared quantitatively with a radiochemical purity of >97% within 30 min. 68 Ga-HZ20 ( Figure 1B ) was produced in high radiochemical yield of 59.9 ± 3.9% (nondecay corrected, n = 10) with over 95% radiochemical purity. The specific activity of 68 Ga-HZ20 was determined to be 6.0 × 10 4 GBq mmol −1 (n = 3). 64 Cu/ 68 Ga-HZ20 with different decay half-lives (12.7 h vs 68 min) allowed us to conveniently carry out the further studies.

We initially tested the binding potency by using surface plasmon resonance (SPR). DX600 and HZ20 showed similar K d of 98.7 and 100.0 nm, respectively ( Figure 1D and Figure S2B , Supporting Information). The specificity of binding was assessed using a saturation binding assay with 64/Nat Cu-HZ20 on ACE2expressing and vehicle HEK293 cells at 4 and 37°C, respectively. 64/Nat Cu-HZ20 displayed a higher binding potency toward ACE2 (66 ± 1 nm) at 4°C than the parental DX600 ( Figure 1C ), a lower affinity (143 ± 1 nm) at 37°C with higher binding molecules (Figure S2A , Supporting Information), which indicated that the internalization of 64 Cu-HZ20 might exist at physiological conditions, and may be retained in target cells. Our further in vitro cell uptake studies showed that the radiolabeled ACE2-binding peptide displayed a highly specific accumulation ACE2-expressing cells In vitro characterization of ACE2 targeting radioligands. A) DX600 and radiolabeled DX600 analogs may interface the penetration of SARS-CoV-2 into host cells by specifically binding to ACE2 receptor. B) The structure ACE2 targeting peptides. C) Saturation binding assay of 64/Nat Cu-HZ20 over HEK293-hACE2 and HEK293-WT (non-transduced) at 4°C. D) Binding assay of HZ20. E) Time-dependent binding and internalization of 64 Cu-HZ20 over HEK293-hACE2 cells at 37°C. in a time dependent manner with significant retention ( Figure  S2A , Supporting Information), and 71 ± 4% of total bound 64 Cu-HZ20 by was internalized ( Figure 1E ).

To evaluate the distribution, kinetics and targeted uptake of HZ20 PET at human (h)-ACE2 sites in vivo, we evaluated two xenograft models of hACE2-expressing cell lines, an engineered HEK293T and the liver hepatocellular carcinoma HepG2, with high endogenous ACE2 expression. [18] Both radioligands enabled clear visualization of ACE2-expressing xenografts following tail vein administration, and in vivo specificity was tested with cold blocking and ACE2-models and non-hACE2 expressing xenografts (Figure 2 and Figure S2 , Supporting Information), which were also further verified by ex vivo autoradiography ( Figure 2C ). The percent injected activity per mL (%IA mL −1 ) of HZ20 accumulation in heart, liver, lung, kidney, muscle, and tumor at 1, 2, and 3 h were quantified by using 68 Ga-HZ20 PET imaging ( Figure S6 , Supporting Information). HepG2 tumor showed 0.56%IA mL −1 at 2 h, which was 96 times higher than that of muscle, and significantly blocked with co-injection of DX600. Similar to 68 Ga-HZ20, 64 Cu-HZ20 also showed the highest signal in the kidneys, high intensity in HEK293-hACE2 xenografts and moderate in liver (Figure 2A,B) . Kinetic analyses of this longer-lived tracer show persistent renal signal which coupled ACE2 immunohistochemistry suggest binding to the highly expressed murine ACE2 in this organ, for which radio-HZ20 ligands have greatly reduced affinity ( Figure S2 , Supporting Information). We further investigated 64 Cu-HZ20 using hACE2 transgenic mice. Compared to the distribution of 64 Cu-HZ20 in the wild-type strain, hACE2 transgenic mice showed significantly higher accumulation in the heart, lung, liver, and intestine, (Figure S3 , Supporting Information) which is consistent with the tissue profile of transgenic hACE2 distribution. [23] In vitro SPR and cell data and in vivo experiments clearly demonstrated the feasibility for 64 Cu/ 68 Ga-HZ20 to specifically monitor human ACE2 expression with high contrast ratios. The capability to perform later post-clearance imaging with cyclotronproduced 64 Cu (half-life 12.7 h) and logistical advantages are offset by the increased absorbed dose computed to the kidneys. The short-lived 68 Ga (half-life of 68 min) can be labeled into the DOTA chelator and is easily accessible through a network of widely available 68 Ge/ 68 Ga generators. [24] In this study, we performed cGMP production of 68 Ga-HZ20 in the hospital, 68 Ga-HZ20 was proved to be stable within 2 h ( Figure S1D , Supporting Information), and passed all quality control standards for clinical trial (Table S1, Supporting Information). Preclinical dosimetry was performed to determine a lead ligand for firstin-human evaluation, which showed that the shorter half-life of Gallium-68 was more amenable for human study (Figures S4-S6, Supporting Information). After an acute toxicity test proved 68 Ga-HZ20 to be well tolerated ( Figure S5 , Supporting Information), we initiated a clinical PET/CT imaging study in ten volunteers.

Five males and five females were enrolled and underwent multiple whole-body PET/CT examinations to compute the mean organ absorbed dose per unit of administered radioactivity. The effective dose of 68 Ga-HZ20 to male and female were calculated as 0.017 and 0.016 mSv MBq −1 , both of which are lower than the whole-body effective dose of 0.019 mSv MBq −1 as reported for the In vivo PET imaging of hACE2-positive pseudotumors with 64 Cu-HZ20 and 68 Ga-HZ20 and its ex vivo assays. A) Dual xenografts (left: HEK293-WT; right: HEK293-hACE2) imaged with 64 Cu-HZ20 at 18 h post injection. B) Quantitative analysis of dynamic 64 Cu-HZ20 imaging of the dual xenografts, kidneys, heart and liver. C) Ex vivo autoradiographic imaging of collected dual xenografts and H&E staining after PET imaging at 18 h post injection. D) Micro-PET/CT imaging of 68 Ga-HZ20 in ACE2 expressing HepG2 tumor-bearing mice at 2 h post-injection. The images were shown on the right, in comparison with the blocking control on the left. In the blocking group, 68 Ga-HZ20 was co-injected with 50 mg kg −1 of DX600. The upper and lower images are MIP images and cross-sectional images of the mice, respectively. E) Comparison of the SUV max of micro-PET imaging at 2 h postinjection. Heart, liver, lung, kidney, muscle, and tumor were selected for both group (n = 4) and statistical significance could be observed for the tumor in comparison to the control group, 0.099 ± 0.0065 versus 0.024 ± 0.0041 (****p < 0.0001). F) Immunohistochemistry results of representative mouse tissue.

ICRP for 18 F-Fluorodeoxyglucose ( 18 F-FDG, the most commonly used PET radiopharmaceutical).

With the radiation safety on 68 Ga-HZ20 determined, we enrolled an additional ten healthy volunteers and carried out a study in the total of 20 subjects (Table S2 , Supporting Information) with the aim to quantify the ACE2 distribution in different organs. 1.85-2.96 MBq kg −1 of 68 Ga-HZ20 was applied as the injection dose. PET data were acquired for all the subjects beginning immediately after administration of 68 Ga-HZ20, and seven scans were performed for all the subjects (5, 14, 23, 32, 41, 90 , and 180 min). The images of a representative male volunteer are shown in Figure S10 , Supporting Information. The standard metric for tracer accumulation is reported at each time point, which measures the bodyweight normalized maximum concentration in a voxel within the region of interest (the standardized uptake value maximum; SUV max ). Organs including the nasal mucosa, oropharynx, conjunctiva, heart muscle, breast, gallbladder, renal cortex, testis, seminal vesicle, corpus luteum, and digestive tract, showed both absorption and elimination phase, indicating ACE2-specific uptake of 68 Ga-HZ20 (Fig-ure 3A-C and Figure S8 , Supporting Information). Rapid elimination from brain, thyroid, lung, liver, spleen, adrenal gland, pancreas, and uterus, suggest relatively low ACE2 expression levels, and the radioactivity gradually clears from the blood pool (Figure 3C and Figure S8 , Supporting Information). Balancing the factors of ACE2-specific uptake and blood pool clearance we posit that static imaging at 90 min may optimally reflect the ACE2 distribution.

A static whole-body PET/CT scan at 90 min was selected for comparative data across patients and cohorts, and the average organ SUV max of 68 Ga-HZ20 are summarized ( Figure 3D ). As expected for a peptide-based ligand, poor blood-brain barrier penetration is noted resulting in little accumulation of activity in brain. The greatest accumulation of signal was observed in kidney (SUV max of 25.67±1.39 in renal cortex and 18.34 ± 1.42 in renal medulla), which may be caused by the co-effects of 68 Ga-HZ20 excretion and high kidney ACE2 expression.

The reproductive system showed a relatively high SUV max , including uterus (2.41 ± 0.34), ovary (2.12 ± 1.42), breast (1.78 ± 0.17) for females and testis (4.46 ± 0.30), penis (2.00 ± 0.25) for . 68 Ga-HZ20 distribution in volunteers. SUV max of each organ was analyzed through the Siemens workstation (Multi-Modality Workplace). The dynamic changes of the SUV max ratio of selected-organ-to-lung at seven time points (5, 14, 23, 32, 40, 90 , and 180 min) were listed. A) Oropharynx, nasal mucosa, and eye, which represents the organs exposed to virus entry. B) Breast, gallbladder, and testis. C) Renal cortex, spleen, liver, and pancreas. D) Rank ordering of organ ACE2 expression in different organs indicated by SUV max . The average SUV max from 20 healthy volunteers at 90 min scans were shown in the column (M: only for male, F: only for female). The SUV max of the organs from the recovered subject at 90 min scan are indicated with a star.

males. Other organs with high SUV max included nasal (1.91 ± 0.18), conjunctiva (1.71 ± 0.08), pancreas (1.63 ± 0.24), esophagus (1.63 ± 0.11), and liver (1.59 ± 0.14). There is continuing interest in the gastrointestinal system as a viral entry site, and indeed moderate radioactivity uptake was detected with the gallbladder and rectum having high SUV max (2.14 ± 0.16 and 1.46 ± 0.13). However, 68 Ga-HZ20 PET imaging showed only low SUV max in the healthy volunteer lung, even though lung is the critical organ for SARS-CoV-2 infection, which is consistent with the data from Human Protein Atlas database. [18] The representative whole body PET of a 38-year-old male (#009) and 34-year-old female (#012) at 90 min post-injection and immunohistochemical analysis of human tissues are shown in Figure 4 . For the male, the SUV max value showed the nasal mucosa (1.57), gallbladder (0.75), and small intestine (0.48) had moderate accumulation, and the renal cortex (23.88) and testis (5.57) showed high accumulation ( Figure 4A ). Analysis in the female showed the nasal mucosa (2.54), breast (2.72), and gallbladder (2.76) had moderate accumulation, while the renal cortex (38.77) and the corpus luteum of left ovary (6.85) showed high accumulation ( Figure 4B ). The high kidney uptake is in agreement with our preclinical and human immunohistochemical findings (Figure 4C ).

ACE2 distribution imaged by 68 Ga-HZ20 reflects sites implicated in the clinical manifestations of COVID-19 pathology could be clearly visualized with good contrast, Table 1 . [16, 18] The renal cortex, corpus luteum, and testis display high 68 Ga-HZ20 uptake (SUV max > 2.5). The breast, gallbladder, ovary, nasal mucosa, and esophagus showed medium uptake for the young female subject (SUV max 1.5-2.5), with relatively low uptake (SUV max <1.5) in other organs. Additionally, the high ACE2 expression in conjunctiva as reported, [25, 26] was also visible on PET/CT ( Figure S11 , Supporting Information).

Interestingly, transient ACE2 uptake in the corpus luteum was observed compared to the immature follicles and 68 Ga-HZ20 contrast was visualized in the ovaries of two young ovulating female volunteers (#005: top row, SUV max of 7.35 vs 1.95; #012: bottom row, SUV max of 6.85 vs 1.93, Figure 5 ). In order to [1] gallbladder, [2] and small intestine [4] showed moderate radioactivity accumulation, and the renal cortex [3] and testis [5] showed high accumulation. B) Radioactivity uptake in a 34-year-old female volunteer (# 012) at 90 min post-injection. The MIP and transverse image showed that the nasal mucosa, [1] breast [2] and gallbladder [3] showed moderate accumulation, and the renal cortex [4] and the corpus luteum of left ovary [5] showed high accumulation. C) Immunohistochemistry analysis of ACE2 expression in normal human organs (10×).

investigate if identification of previously unidentified and transient ACE2 expression changes may be present we conducted a delayed PET/MR (2 h after 68 Ga-HZ20 injection) examination was carried out immediately following PET/CT imaging to reveal the precise structure of ovarian foci ( Figure 5 ). The stroma of ovary showed slight uptake, and there was no uptake in immature follicles.

Clinical severity and mortality of COVID-19 has been more severe for men than women, in several studies [27, 28] and the aged population is particularly susceptible. [29, 30] To investigate the relationship of ACE2 across these populations we compared organ uptake by 68 Ga-HZ20 PET (Figure 6 , Figure S12 , Support-ing Information). A strong correlation was discovered between the ACE2 expression in the breast of female volunteers and with age [31, 32] (50-year as the cutoff), with the young group higher than the old group (2.21 ± 0.25 vs 1.35 ± 0.22, p = 0.0002, at 90 min, Figure 6 ). At this time, we continue to increase the sample size to confirm these and other observations, a limitation of this firstin-man study.

We next tested ACE2 PET in a 38-year-old male infected SARS-CoV-2 at 7 months post infection. Images at 90 min post-injection are shown in Figure 7 and all organs with high radioactivity uptake were visualized clearly with SUV max of each organ determined ( Figure 3D ). Evaluation of a larger number of recovered PET; and C,F) fusion images showed both ovary foci localized in the corpus luteum (red arrow). The stroma of ovary showed slight uptake, and there was no uptake in immature follicles (blue arrow). In order to identify the precise structure of ovarian foci, PET/MR examinations were carried out at 2 h post-injection of 68 Ga-HZ20 on the same day. The comparison of reproductive system organs between young (<50 years old) and old (≥50 years old) volunteers shows that the young group has a significant higher uptake in the breast than the old, but no significant difference was observed in ovaries and testes. Bottom row: The comparison of high uptake organs (gallbladder, nasal mucosa, and renal cortex) between male and female shows no significant difference. 

Possible related clinical manifestation of COVID-2019

High High High [34] [35] [36] Acute kidney injuries [45, 46] (renal proximal tubules)

Testis High High High [36] Seminiferous tubular injuries [47] (Sertoli cells and Leydig cells)

Corpus luteum High No data No data No data Gallbladder Medium High High [37] Gastrointestinal discomfort [ 45, 46] (membranes of gallbladder epithelium)

Intestine Medium High High [ 34, 35] Gastrointestinal discomfort and diarrhea [ 45, 46] (microvilli of the intestinal tract)

Medium No data High [38] Olfactory deficits [ 48, 49] Breast (young) Medium No (adipocytes, glandular cells, myoepithelial cells)

No data Breast milk samples from nine mothers were negative (50) Ovary Medium Medium (stroma cells) High [39] No data Oropharynx Low No data High [40] Sore throat [45] Heart muscle Low medium High [ 35, 41] Acute cardiac injuries [ 45, 46] (cytoplasm cardiomyocytes)

Seminal vesicle Low medium High [42] No data [51] (glandular cells)

Low No Low [43, 44] Pneumonia [ 45, 46] a) High 68 Ga-HZ20 uptake (SUV max > 2.5), medium uptake (SUV max 1.5-2.5), low uptake (SUV max < 1. patients is required to draw robust conclusions, here the gallbladder, testis, and many of the normal organs showed increased uptake versus the healthy volunteer pool. Conversely, renal cortex accumulation of the tracer was substantially lower (SUV max : 18.40 vs 25.67 ± 1.39, see Supporting Information videos for details). Acute kidney injury has been reported in about 9% patients hospitalized with COVID-19, [33] and these results were unexpected. While speculative, the differences in this patient suggest an organ-specific response, at the molecular expressionlevel, resulting from infection even at these extended times postinfection. Extension and verification of these results in a wider population of COVID-19 patients will provide additional insight.

The ACE2 expression level and organ-specific distribution may potentially reflect the susceptibility, severity, and prognosis of SARS-CoV-2 infection. A non-invasive imaging tool to explore expression characteristics of ACE2 in vivo may assist in understanding the pathogenesis of SARS-CoV-2 infection with the potential to assist in development of mitigation and treatment planning. A high affinity human ACE2 specific peptide, DX600, was selected and modified as a PET imaging agent for translational evaluation. Radiometal-chelator conjugation did not affect binding nanomolar potency or specificity of the ligand, which was rapidly internalized. Preclinical investigation of 68 Ga/ 64 Cu-HZ20 demonstrated ideal pharmacokinetic properties in models of human ACE2 expression, with rapid blood clearance, low background organ uptake and predominant renal clearance. High contrast of ACE2 pseudotumors was achieved within 2 h and within genetically engineered models. After safety and dosimetric evaluation, 68 Ga-HZ20 was applied for a first-in-human translational study with 20 healthy volunteers of different ages and both sexes. To fully characterize this molecular imaging tool, dynamic scans were performed for each volunteer. The results showed highly consistent pharmacokinetics among the 20 volunteers, with high ACE2 expressing organs (such as kidney, gallbladder, testis for male) reflected in both high SUV max and signal retention within the 3 h tested. The plateau of signal in expressing tissues and clearance from the blood pool afford convenient 90 min post-administration static images useful to profile ACE2 expression.

Imaging results were mostly consistent with the HPA database (http://www.proteinatlas.org) and previous immunohistochemical profiling of ACE2, including microvilli of the intestinal tract and renal proximal tubules, gallbladder epithelium, testicular sertoli cells, and leydig cells (Table 1) . [16, 34, 50] Intriguingly, we re- port a differential finding between these imaging results and histological HPA data in the ACE2 expression in the female breast. We observed that the ACE2 expression level in breast is age dependent, with medium for young group and low for old group (p = 0.0002, Figure 6 ), while HPA reported no ACE2 expression.

PET imaging of ACE2 non-invasively can provide real time and global receptor distribution quantitatively using the SUV max value of the uptake of 68 Ga-HZ20 correlating directly with ACE2 expression level, which offers novel information relevant for infection by SARS-CoV-2, and pathology of COVID-19. Recent reports have indicated that partial loss of the sense of smell or even total anosmia is early symptoms of SARS-CoV-2 infection. [47, 48] It was suggested that the virus may exploit goblet and ciliated cells in the nasal epithelia as entry portals, a plausible primary infection site in many patients. [51] Along the respiratory tract, the nasal mucosa showed higher 68 Ga-HZ20 uptake than oropharynx, and the lung showed the lowest uptake. It should be noted that the lower density of the lung tissue likely biases measured uptake values to underestimate tracer concentration. This observations are consistent with a recent study using high-sensitivity RNA in situ mapping that has shown ACE2 expression is highest in the nose with decreasing expression throughout the lower respiratory tract, paralleled by a striking gradient of SARS-CoV-2 infec-tion in proximal (high) versus distal (low) pulmonary epithelial cultures. [52] This gradient could be clearly visualized in the same volunteer from the PET image.

Surprisingly, 68 Ga-HZ20 uptake in the lung and heart was very low. Data from the Human Protein Atlas and others has shown ACE2 receptor expression variability in patients bronchial and lung tissues with indications that underlying conditions inducing inflammation [63] may have a role in expression levels. [53] An interesting finding from these comparisons across patients in the imaged cohort is that there are negligible differences between young and geriatric males or females. These data potentially underline the equivalent risk of infection across all adults, and that further evaluation can be accomplished with the tracer noninvasively.

Studies have found that 70% of patients infected with SARS-CoV accompanied by diarrhea. A recent case report described that ACE2 is highly expressed in stratified epithelial cells and absorptive intestinal cells in the upper esophagus, ileum, and colon. [54] These findings may support the possibility of fecal-oral transmission route, and also consistent with the uptake of 68 Ga-HZ20 in the oropharynx, intestine, and rectum observed in our imaging results ( Table 1) .

Several of our results highlight expression profiles in reproductive organs. Recent studies have reported testicular damage www.advancedsciencenews.com www.advancedscience.com caused by SARS-Cov-2 infection, [46, 55] and investigation by PET on reproductive function of recovered male patients, especially youth [35] is warranted. Besides ACE2 expression observed in the breast of young female, 68 Ga-HZ20 uptake was observed in ovaries (without a difference between young and old women). We also found high 68 Ga-HZ20 uptake in corpus luteum in two young women, but not in immature follicles. ACE2 expression in antral follicles, mature luteal cells, the theca and stromal cell layer in ewes has been previously described, [56] but to our knowledge this is the first observation of ACE2 in human corpus luteum with the direct comparison around surrounding organs. The ACE2 specific inhibition of the cyclic DX600 peptide has been established, however uptake at these unexpected sites may suggest non-specific interactions which can be definitively determined with future biopsy confirmation.

Acute kidney injury has also been commonly found in patients infected with SARS-CoV-2, [57] with likely long term impacts on overall health. In addition to the host's immune response, these acute sequelae of fighting the infection may also come from the direct attack of the virus on the target cells expressing ACE2. [58] RNA-Seq studies have shown that ACE2 is abundantly expressed in various renal proximal tubule cell subtypes; consistent with our observations. Limited by qualified resources for carrying out studies in patients during the SARS-CoV-2 infection, we were only able to image one recovered volunteer in which we report dramatic changes in observed ACE2 expression ( Figure 3D and Figure S9 , Supporting Information). We stress that conclusions cannot be drawn across healthy volunteers and the recovered patient. However, these data strongly motivate further investigations in a larger recovered patient pool.

In summary, we have developed a non-invasive imaging method using 68 Ga/ 64 Cu-HZ20 PET to interrogate the global expression profile of ACE2 in human, the key receptor for SARS-CoVs to infect human cells. To our knowledge, this is the first study to quantitatively profile ACE2 expression by PET imaging, and this approach demonstrates the capacity to measure organ express profiles at baseline and following COVID-19. The results from 20 healthy volunteers were consistent with pathological reports of receptor distribution, while preliminary imaging data from a SARS-CoV-2 infection recovered man may suggest ACE2 level changes. These preliminary results require further investigation in a wider patient population. With the capacity of quantitatively detecting stable and transient ACE2 expression non-invasively, this quantitative imaging method may have utility to evaluate differences across patients for SARS-CoV-2 infectivity; COVID-19 symptom severity and duration; and for evaluation of physiological effects from other emerging and novel coronaviruses.

Ethical Statement: Human study was approved by the Ethics Committee of Peking University Cancer Hospital (2020KT62) and registered in Chinese Clinical Trial Registry (ChiCTR2000033675, Date of registration: June 8, 2020) and ClinicalTrials.gov (NCT04422457, Date of registration: June 9, 2020). All clinical study was conducted following the latest guidelines of the Declaration of Helsinki. All animal studies were performed according General Procedures: All solvents and chemicals purchased from commercial sources were of analytical grade or better and were used without further purification. DX600, NODAGA-DX600, and DOTA-DX600 were custom synthesized by ChinaPeptides Co., Ltd (Shanghai, China) or CS-Bio (San Diego, California). Sep-Pak Accell Plus QMA and Sep-Pak C18-Light cartridges were purchased from Waters (Ireland). Acrodisc 25 mm syringe filter (0.22 µm) was purchased from Pall Corporation (USA). The product was analyzed by radio-high performance liquid chromatography (HPLC) (1200, Agilent, USA) equipped with detector (Flow-count, Bioscan, Washington. D.C., USA), using a C18 column (Eclipse Plus C18, 4.5 × 250 mm, 5µm, Agilent, USA). The product purity was also determined using Radio-TLC (AR 2000, Bioscan, USA) after radiolabeling. The PET/CT imaging studies of small animals were performed on the Mira PET/CT of PINGSENG Healthcare Inc. (Shanghai, China), or microPET R4 rodent scanner (Siemens) and analyzed by ASIProVM. The Clinical PET/CT scans were obtained on a Biograph mCT Flow 64 scanner (Siemens, Erlangen, Germany) with unenhanced low-dose CT. 68 Ga-HZ20 PET/MRI was performed on a hybrid 3.0T PET/MR scanner (uPMR790, UIH, Shanghai, China) in female volunteers.

Radiolabeling and Quality Control of Radiopeptides: 68 Ga-HZ20: 195 µL 1.0 m NaOAc solution containing 40 µg (1.17 × 10 −5 mmol) DOTA-DX600 was added into 3.0 mL of 68 GaCl 3 freshly eluted from 68 Ge-68 Ga generator (Isotope Technologies Garching, Germany) by 0.05 m of hydrochloric acid, with the radioactivity ranging from 370 to 1110 MBq. The final pH of the reaction was controlled as 4.2 and the mixture was heated at 95°C for 15 min. After the reaction completed which was monitored by radio-TLC, the mixture was loaded onto an activated Sep-pak C18 cartridge. The cartridge was first washed with 5 mL of water to remove the free 68 Ga, and then eluted with pure ethanol to obtain the product of 68 Ga-HZ20 as a 0.5 mL ethanol solution. The ethanol solution of the product was diluted with 10 mL 0.9% saline and filtered through a 0.22 µm filter (Merk, Darmstadt, Germany) to make the radiopharmaceutical used in this study. It contains 37-370 MBq of 68 Ga-HZ20 in ethanol-water (ethanol < 5%).

The solution was analyzed by radio-HPLC to assess the radiochemical purity. The HPLC was eluted with water-CH 3 CN system (Phase A: 0.1% TFA H 2 O; Phase B: 0.1% TFA CH 3 CN) using gradient elution (0-5 min 20% B; 5-10 min 20%-80% B; 10-12 min 80% B; 12-15 min 80%-20% B) at a flow rate of 1.0 mL min −1 . The radiolabeling yield was 59.9 ± 3.9% (non-decay corrected, n = 10) and the radiochemical purity was over 95%. In vitro stability study of 68 Ga-HZ20 in phosphate buffer saline was performed by adding 50 µL of 68 Ga-HZ20 to 450 µL of phosphate buffer saline and incubating at 37°C. At 1, 2, and 4 h time points, 10 µL aliquot was analyzed by radio-HPLC to assess the radiochemical purity.

64/nat Cu-HZ20: 64 Cu-labeled NODAGA-DX600 was prepared by dissolving 20-40 µg of peptide in 10-20 µL of trace-free water and 50 µL of 0.25 m ammonium-acetate buffer (pH 6.0), and adding 37-74 MBq 64 CuCl 2 solution (1-2 µL) followed by a 5 min incubation at 95°C. After incubation, the final product was analyzed with analytical HPLC (column: Kromasil 100-5-C18, 4.6 × 150 mm; flow rate: 1.0 mL min −1 ; mobile phase: 0.1% TFA in water and CH 3 CN; gradient: 0-10 min, 10%-50% CH3CN; 11-14 min, 95% CH 3 CN, 15 min, 10% CH 3 CN), and reformulated with PBS/BSA (1.0% bovine serum albumin) solution for further experiments. If needed, purification with HLB cartridge (Phenomenex; Torrance, CA). For binding studies, one equivalent of nat Cu(SO 4 ) 2 was further added into reaction mixture, and the final solution was incubated for another 5 min to generate structurally identical 64/Nat Cu-labeled HZ20 for saturation binding studies.

Surface Plasmon Resonance Assay: The binding affinity between DX600 (or DOTA-DX600) and ACE2 was determined by SPR assay with Nicoya Open SPR system (Nicoya Lifesciences Inc., Ontario, Canada), in comparison to DX600. Briefly, recombinant human ACE2 protein (50 µg mL −1 ; BP003061) was immobilized on the surface of the nanogold sensor chip, after the chip was activated by 1-ethyl-3-(3dimethylaminopropyl)carbodiimide/N-hydroxysuccinimide. Various dilutions of DX600 or DOTA-DX600 were added at a flow rate of 20 µL min −1 for 7 min and the SPR signal were collected. K D , ka and kd were calculated www.advancedsciencenews.com www.advancedscience.com using Trace Drawer Evaluation version 2.0 (Trace Software International, Saint-Romain, France).

Saturation Affinity of 64/Nat Cu-HZ20 Binding to hACE2: Saturation affinity studies were performed in HEK293-hACE2 cells and HEK293-WT using various concentrations of 64/Nat Cu-HZ20. Triplicate samples containing 0.28 × 10 6 cells and 0.01-2000 nmol of 64/Nat Cu-HZ20 in 0.25 mL cell culture medium were incubated at 37°C for 1 h and 4°C for 24 h, respectively. The cells were collected with glass microfiber filters, washed with 4 × 2 mL of ice-cold TBS (pH 7.4), and radio-assayed with a gamma counter. 64/Nat Cu-HZ20 uptake (molecules per cell) in the HEK293-hACE2 cells was plotted versus 64/Nat Cu-HZ20 concentration, and Kd values were estimated using a least-squares fitting routine (GraphPad Prism 8, San Diego, CA).

In Vitro Uptake of 64 Cu-HZ20: hACE2-positive cells (HEK293-hACE2) and HEK293-WT cells were cultured in DMEM HG cell medium and preseeded in 6-well plate the day before cell uptake experiments. Cells (0.1 × 10 6 in a total volume of 0.95 mL cell medium) in triplicate were mixed with ≈26 kBq of 64 Cu-HZ20 in 0.05 mL PBS buffer (final concentration: 2.5 nm), and the cell mixtures were incubated at 37°C for 1 h. Nonspecific uptake was determined by co-incubating with 10 µg of DX600. After incubation, the unbound radioactivity was collected and then the cells were washed twice with cold PBS. Glycine buffer (pH 2.8) was used to incubate the cells for 5 min over ice bath two times to remove the non-internalized radiotracer. After incubation, the cells were destroyed with 1 m NaOH and the residual was collected for measuring the internalized radiotracer. For kinetic uptake studies, the samples were incubated at 37°C for 15, 30, 60, and 120 min and the cells were measured with a gamma-counter.

PET Imaging in Animals: 68 Ga-HZ20: Four-week-old female athymic nude mice were purchased from SLAC Laboratory Animal Co. Ltd., China. 5 × 10 6 of HepG2 cells were inoculated into the right front shoulder region of the mice to build the xenograft tumor model with ACE2 expression. When the tumor volumes were estimated to be 450-800 mm 3 , the mice were used for the studies. Under isoflurane anesthesia, mice bearing HepG2 tumors (n = 4 per group) were injected intravenously with 100 µL of 3.7 MBq 68 Ga-HZ20 for small animal PET. Scans were performed at 60, 120, and 180 min after administration. Unlabeled DX600 (50 mg kg −1 body weight) and 3.7 MBq 68 Ga-labeled tracers were co-injected into mice bearing HepG2 tumors (n = 4 per group) for blocking study as a control group. Subsequently, animals were scanned at 60, 120, and 180 min postinjection. Regions of interest were defined over tumor, heart, liver, lung, kidney, and muscle and uptake computed relative to the injected activity.

Female immunocompromised R2G2 mice from Jackson Laboratories (Bar Harbor, Maine) were used for subcutaneous inoculation of dual pseudotumors of hACE2 and control (non-expressing) HEK293T cells. The cells (3 × 10 6 ) suspended in a 1:1 mixture of PBS and Matrigel were implanted on the shoulder area. Tumors were monitored and imaging commenced after reaching >200mm 3 .

Dual xenografts-bearing mice or transgenic mice were injected with about 3.0 MBq (0.3 nmole) of 64 Cu-HZ20 via tail vein injection. At 1, 4, and 20 h post-injection, PET imaging was performed for 15 or 30 min on microPET R4 rodent scanner (Siemens) with the tumors centered in the field of view, and the animal under 2% isoflurane anesthesia. PET images were reconstructed by an iterative 3D maximum a priori algorithm. The calibration factor of the PET scanner was measured with a mouse-sized phantom composed of a cylinder uniformly filled with an aqueous solution of 18 F with a known activity concentration. ROI analysis of the acquired images was performed using ASIProVM (Siemens) and the observed percent injected activity of tissue (%IA mL −1 ) was measured.

Following PET imaging, tumors were excised, embedded in optimalcutting-temperature mounting medium (OCT, Sakura Finetek), and frozen on dry ice prior to cutting a series of 10 µm frozen sections. Digital autoradiography was performed on a phosphor imaging plate at −20°C. Phosphor imaging plates were read at a pixel resolution of 40 µm with a Cyclone system (Perkin Elmer). Following autoradiography imaging, the same section was stained with H&E and whole-mount bright field images acquired in a similar manner.

The dual xenografts-bearing animals were randomly assigned into three cohorts and administrated ≈3.0 MBq (0.3 nmole) of 64 Cu-HZ20. At 1, www.advancedsciencenews.com www.advancedscience.com MultiModality workplace for analysis and comparison. SUV max is defined as:

where r is the maximum radioactivity activity concentration (kBq mL −1 ) measured by the PET scanner within the defined ROI, a′ is the decaycorrected amount of injected radiolabeled 68 Ga-HZ20 (kBq), and w is the weight of the patient (g). Statistics: Statistical analyses were performed with SAS software (V.9.4, SAS) and Prism (V8.0, GraphPad Software). The organ uptake data in the form of SUV max were grouped by gender and by age. To compare distributions among samples, the parametric continuous variables were expressed as mean ± SD. A mixed model was applied to compare the different groups for different time points, with study subjects as random effects. Independent sample t-tests were used to compare SUV max values between different groups. A p-value less than 0.05 was considered to be significant.

Supporting Information is available from the Wiley Online Library or from the author.

World Health Organization

COVID-19 Dashboard by the Center for Systems Science and Engineering

COVID-19 and the Use of Angiotensin-Converting Enzyme Inhibitors and Receptor Blockers

tion. The organs of interest were collected, rinsed, blotted, weighed, and counted with a -counter (Perkin Elmer Wizard 2 ). The total injected radioactivity per animal was determined from the measured radioactivity in an aliquot of the injectate. Data were expressed as percent of injected activity per gram of tissue (%IA g −1 ).Preclinical Immunohistochemistry: Harvested mouse tissues were formalin-fixed, paraffin-embedded, and sliced at 4 µm thickness. Normal human tissue blocks were acquired from the tissue bank at the Peking University Cancer Hospital Pathology Department Central Laboratory. Sections were incubated with 3% H 2 O 2 at RT for 10 min. Antigen was retrieved from the tissue in citric acid buffer (0.01 m, pH 6) with microwave heating for 2.5 min, followed by a 5 min cooling step. For normal human tissues, microwave heating was extended to several rounds of 3 min exposure. Tissues were blocked with goat antiserum for 30 min, stained with rabbit anti-ACE2 antibody (1:100 ab108252, Abcam) at 4°C overnight. The washed sections were then stained with goat anti-rabbit secondary (PV-6001, Beijing Zhongshan Jinqiao Biologicals) for 30 min at RT. Subsequently, the sections were developed with 3, 3′ diaminobenzidine tetrahydrochloride (DAB), dehydrated in an alcohol gradient, and sealed in neutral mounting media, and scanned (Aperio Versa 200, Leica).Clinical 68 Ga-HZ20 Release: 68Ga-HZ20 was prepared as above under GLP environment dispensing hot cell (NMC Ga-68, Tema Sinergie, S.p.A, Italy). The radiotracer met or exceeded release quality control and criteria (Table S1 , Supporting Information). Patients received tracer at a mean administered activity of 2.48 MBq kg −1 (±0.42 MBq kg −1 ).Imaging Study Subjects Enrollment: The informed consent was obtained from the volunteers for publication. The inclusion criteria for healthy subjects included: 1) older than 18 years, 2) the ability to provide informed written consent, 3) a medical history without any significant comorbidities, including physical examination, electrocardiogram, hematology, and biochemistry. The exclusion criteria included: 1) liver and renal function dysfunction, 2) pregnancy or current lactation. Following the criteria, four groups of subjects (five male with age below 50, four male with age above 50, five female with age below 50 and six female with age above 50) were enrolled in this study, for a total of 20 subjects. The age ranged from 32 to 72 with a mean of 51.1 ± 15.1 and the demographic data of all the volunteers were shown in Table S2 , Supporting Information.PET/CT Examination Procedures: No specific preparation was required for the subjects on the day of 68 Ga-HZ20 PET scanning. A low-dose CT scan (120 kV, 35 mA, slice 0.6 mm, matrix 512 × 512) was per-formed before the 68 Ga-HZ20 injection. Then, a whole-body dynamic PET scan was performed immediately after the intra-venous injection of 68 Ga-HZ20 in all subjects and continued for ≈40 min (five passes, round 10 min for each pass). All the subjects also underwent static whole-body PET/CT scan at 90 and 180 min post-injection. Dynamic whole-body PET/CT scans were performed on a Biograph mCT Flow 64 scanner (Siemens, Erlangen, Germany) with the setting of 120 kV, 146 mAs, slice 3 mm, matrix 200 × 200, full width at half maximum (FWHM) 5 mm, filter: Gaussian, field of view (FOV) 256 (head), 576 (body). The patient bed was set to continuously move at a speed of 2 mm s −1 to cover the entire body of each subject (from the top of the skull to the middle of the femur). Static whole-body PET/CT scan used a speed of 1 mm s −1 at 90 min and 0.8 mm s −1 at 180 min. 3D iterative reconstruction was applied for image reconstruction, with CT-based attenuation and scatter correction through standard vendor-based reconstruction. The 68 Ga activities were decay corrected to the time of injection and normalized to the total activity administered.PET/CT Image Analysis: MultiModality Workplace (Siemens, Erlangen, Germany) was used for data processing. To analyze the biodistribution of 68 Ga-HZ20, regions of interest (ROIs) were manually drawn on the largest transverse section of major organs/tissues, while avoiding major blood vessels. The normal organs/tissues selected for VOI analysis included: the brain, parotid glands, nasal mucosa, oropharynx, nasopharynx wall, thyroid, cardiac muscle, left heart ventricle, lungs, liver, gallbladder, pancreas, spleen, kidneys, red marrow, bone, stomach, small intestine, upper and lower colon, rectus, breast, uterus, ovary, prostate, testis, and muscle (quadriceps femoris). The maximum single-voxel standardized uptake value (SUV max ) were generated automatically from ROIs in

The authors declare no conflict of interest. Intellectual property protection has been filed by Washington University in St. Louis School of Medicine and Ha.Z and D.T.

Hu.Z., Ha.Z., and N.Z contributed equally to this work. Hu.Z., Ha.Z., D.T., X.Y., and Z.Y. conceived and designed the experiments. J.D., Ha.Z., and Z.L. radio-synthesized and characterized the agent. J.D., N.B., and Z.L. conducted preclinical research and quality control of radio-tracer. N.Z., Ha.Z., N.B., J.J., and D.T. performed the PET, PET/CT, and PET/MRI studies. N.B., and Q.Z. performed the immunohistochemistry and dose calculations. N.Z., Hu.Z., Ha.Z., N.B., T.L., H.J., and L. L. analyzed the data. X.Y., Hu.Z., D.T., and Ha.Z. co-wrote the paper. Z.L. and Z.Y. provided constructive discussion. All authors discussed the results and analysis and commented on manuscript.