CTR1 is a therapeutic target when more blood vessels are needed

0
98

Insufficient oxygen to an area like the heart or legs, called hypoxia, is a cue to our bodies to make more blood vessels, and scientists have found some unusual partners are key to making that happen.

The endothelial cells that line existing blood vessels are essential to making new blood vessels, and they’ve found that two receptors on the surface of those cells come together then dive inside to enable the new growth, called angiogenesis, the Medical College of Georgia scientists report in the journal Nature Cell Biology.

As the name implies, vascular endothelial growth factor receptor 2, or VEGFR2, typically binds to VEGF, a signaling protein that enables new blood vessel growth, to go inside the cells. CTR1 enables copper, an essential mineral key to many important body functions, including angiogenesis, to also go inside.

Corresponding authors Dr. Masuko Ushio-Fukai, vascular biologist, and Dr. Tohru Fukai, vascular biologist and cardiologist, say that in the face of hypoxia, VEGF is naturally stimulated outside the cell, then in turn activates NADPH oxidase, a family of enzymes that generate reactive oxygen species, or ROS – in this case the good kind that enables cell signaling.

They’ve shown for the first time the step that happens next: ROS then quickly modifies CTR1, which prompts the VEGF receptor and this copper transporter to bind to each other on the cell surface and move inside the cell so the sustained VEGFR2 signaling that is essential to new blood vessel development happens.

When they knocked down the copper transporter, angiogenesis was severely impaired, Fukai says. They’ve also used the gene editing ability of CRISPR-Cas9 to make CTR1 unmodifiable and angiogenesis was again significantly reduced.

The scientists worked in models of blood vessel development in the highly vascularized retina and in peripheral artery disease in a limb.

While it’s long established that VEGF binding to its receptor causes angiogenesis, this is the first evidence of this binding of receptors for copper and VEGF that appears to be an early and important connection.

Their findings also make CTR1 a logical therapeutic target when more blood vessels are needed in the face of problems like ischemic heart disease, and potentially when the process becomes destructive in conditions like cancer and diabetic retinopathy, the scientists in MCG’s Vascular Biology Center say.

Ushio-Fukai’s lab has shown that treatment of endothelial cells with VEGF increases ROS levels in those cells. Her lab subsequently showed that ROS, a byproduct of our oxygen use, is very important for promoting VEGF-induced angiogenesis. Fukai’s lab has focused on the impact copper metabolism has on the body, including its transporter CTR1.

Now they’ve connected the dots between them. “Once ROS is generated, it modifies CTR1, which changes its function to become a binding partner for VEGFR2,” she says.

In this scenario at least, CTR1 does not take copper with it when it moves inside endothelial cells but in other parallel action it still does, which is another boon to angiogenesis.

Once the connections start generating new blood vessels, ROS levels go down and the CTR1 is free to return to the cell surface where it resumes its normal job of transporting copper inside, Ushio-Fukai says.

They also have evidence there is always some CTR1 left on the endothelial cell surface, which can continue to do its usual job of enabling copper to be used, in this case to also aid angiogenesis.

They also have indications that too much copper can be destructive. The Fukais have seen copper pile up inside cells in diabetes, where it instead impairs the ability to help make the healthy new blood vessels patients may need. They say another copper transporter, ATP7A, whose primary job is regulating copper levels inside cells, is likely a good treatment target in diabetes, where ATP7A levels are uncharacteristically low.

Copper is an essential trace mineral we need to consume in foods like nuts and whole grains which also is important to fundamentals like making red blood cells that carry oxygen. There is a lot of CTR1 in the gut to enable copper’s uptake and use throughout the body.

While VEGF and endothelial cells are both essential to angiogenesis, it’s other cell types, like platelets, immune cells called macrophages and even tumor cells that make VEGF.

As with copper, physiological levels of ROS are important to body functions like cell signaling, but high levels produced from unhealthy habits like smoking and the high-salt Western diet contribute to disease and conditions like hypertension and atherosclerosis in turn increase ROS.


Despite decades of efforts to understand its pathogenesis, heart disease remains a leading cause of mortality and places a growing burden on healthcare systems worldwide. Copper is an essential trace metal micronutrient that has been previously overlooked, but has recently gained attention and become an emerging player in the development of heart disease.

Although it is less abundant than other metals such as iron and zinc, copper is widely utilized as a catalytic or structural cofactor by enzymes and proteins that are highly relevant to cardiac physiology and pathology. These copper-binding proteins include cytochrome c oxidase (CCO), superoxide dismutase (SOD), metallothionein (MT), ceruloplasmin (CP), and lysyl oxidase (LOX), which regulate mitochondrial respiration, antioxidant defense, iron metabolism, and connective tissue crosslinking [1]. Increasing evidence has demonstrated that dysregulation of copper homeostasis causes heart disease.

Copper exists in two ionic forms in the body, namely, cuprous Cu+, which is dominant in the intracellular reductive environment, and cupric Cu2+, which is dominant in the ex- tracellular oxidative environment. In mammals, copper is exclusively absorbed from diets and water by enterocytes in the small intestine via the Cu+-specific copper transporter 1 (CTR1).

The reduction of Cu2+ to Cu+ is necessary for its entry into cells and is mediated by copper reductases, including members of the plasma membrane-bound six transmembrane epithelial antigen of the prostate (STEAP) family [2]. The latest study by Kurdistani et al. identified a novel copper reductase enzyme, the histone H3-H4 tetramer, which binds to Cu2+ and catalyzes its reduction to Cu+ in Saccharomyces cerevisiae yeast to maintain the function of the electron transport chain in mitochondria [3]. After entering interstitial fluid, copper initially binds to albumin or transcuprein, travels through the portal circulation, and is taken up by the liver via hepatic CTR1. Copper in the liver is then incorporated into CP, a major plasma copper-binding protein responsible for carrying >90% of copper in the circulation [4].

Copper-loaded CP delivers copper to extra-hepatic tissues, and excess cop- per returns to the liver for excretion into bile through hepatic copper-transporting ATPase 2 (ATP7B). Intracellular copper transportation and involved copper-binding proteins and chaperons are depicted in Figure 1.

Figure 1. Copper-binding proteins and intracellular copper transportation. Copper is exclusively absorbed by enterocytes in the small intestine via CTR1. CTR2 is a low-affinity copper importer that localizes to endosomes and lysosomes. Intracellular copper-binding proteins include COX, SCO, SOD, MT, CP, and LOX. ATP7A and ATP7B are copper exporters. Under normal conditions, ATP7A and ATP7B localize to TGN, where they supply copper to copper-dependent enzymes in the secretory pathway. When the cytosolic copper level rises, ATP7A or ATP7B interacts with DNCT4 and traffics to endosome-like vesicles and then to the plasma membrane, pumping excess copper into the extracellular space, or into bile in the case of the liver, to reduce intracellular copper levels. By contrast, when the intracellular copper level is low, ATP7A or ATP7B recycles to the TGN and transports copper from the cytoplasm into the Golgi. ATOX1: antioxidant 1 copper chaperone; ATP7A: copper- transporting ATPase 1; ATP7B: copper-transporting ATPase 2; CTR1: copper transporter 1; CTR2: copper transporter 2; CCO: cytochrome c oxidase; COX: cytochrome c oxidase copper chaperone; CP: ceruloplasmin, DNCT4: p62 subunit of dynactin; LOX: lysyl oxidase; MT: metallothionein; SCO: synthesis of cytochrome c oxidase; SOD: superoxide dismutase; TGN: trans-Golgi network.

Copper is vital for cellular functions, and excess copper is toxic. Therefore, the distribution and amount of bioavailable copper must be tightly controlled to meet metabolic requirements, while minimizing potential toxicity of excess copper. Impaired functions of copper transporters, defects in copper-dependent enzymes, and chronic copper deficiency cause heart diseases [5–10]. While the recommended daily allowance (RDA) of copper is
0.9 mg/day, the recommended optimal intake is 2.6 mg/day [11]. The copper requirement varies between individuals and depends on age, pregnancy, sex, health status, and other factors. For example, RDA is 340 µg/day for 1–3-year-old children and increases to a minimum of 1 mg/day during pregnancy [12]. Excessive daily zinc intake competes with copper for absorption by enterocytes in the small intestine and, therefore, decreases copper intake [13,14]. Diseased states, including hypertension, ischemic heart disease (IHD), heart

failure (HF), nephrotic syndrome, and celiac disease, often cause copper deficiency, which increases demands for daily copper intake [15–22].
Although a trace amount of copper is required daily, copper deficiency is common because the amount of copper in modern diets has decreased during the last several decades. Western-style diets enriched in saturated fat and simple sugars, particularly fructose, inhibit small intestinal copper absorption and, thus, contribute little to the daily dietary requirement for copper [11,23]. In addition, changes in farming methods have decreased the copper content of soil and, thus, in produce [24]. In fact, the National Health and Nutrition Examination Survey (NHANES III, 2003) revealed that more than 80% of 103,655 people studied in the US received a lower amount of copper than the RDA from their diet [25]. Similarly, dietary copper intake was lower in the National Diet and Nutrition Survey (NDNS) from 2000/01 [26] than in that from 1986/87 [27] in the UK.
In the past decades, studies from dozens of laboratories across the globe have revealed that copper plays a major role in maintaining normal heart morphology and function. A recent review summarizes the association between copper deficiency and IHD in de- tail [15]. Another summarizes the link between copper transporters and chaperones and cardiovascular disease [22]. In this review, we thoroughly summarize current knowledge about the roles of copper-dependent enzymes, copper transporters/chaperones, and copper deficiency in heart diseases, including cardiac hypertrophy, HF, IHD, and diabetes mellitus (DM) cardiomyopathy, focusing on detailing the molecular mechanisms and providing both preclinical and clinical evidence.

Copper Chaperones and Heart Physiological and Pathological Processes
CCO

Copper-dependent CCO is the mitochondrial respiratory chain complex IV, which contains copper and heme as required co-factors and plays a critical role in oxidative phosphorylation. Of the 13 subunits of mammalian CCO, three mitochondria-encoded DNA subunits (I, II, and III) contain copper and heme in their active sites and constitute the catalytic core of the oxidase complex. Several copper chaperones deliver copper to CCO, including CCO copper chaperone 11 (COX11), COX17, COX19, and COX23 and synthesis of CCO 1 (SCO1) and SCO2.

Thus, copper deficiency reduces CCO activity and the mitochondrial respiratory capacity in the heart, for which its functions heavily rely on intact mitochondrial respiration [8,28–34]. Copper deficiency causes cardiac hypertro- phy by impairing mitochondrial function and energy production, evidenced by increases in mitochondrial compensatory biogenesis and size and mitochondrial ultrastructural deteriorations, as well as a decrease in the number or loss of cristae [19,35].

Although the causative role of copper deficiency in decreasing CCO activity was documented as early as 1939 by Schttltze [36], the key players have only recently been identified based on human genetic studies and loss-of-function animal studies. Patients with the common E140K mutation in SCO2 were reported to die of HF due to fetal/infantile hypertrophic cardiomyopathy (HCM). This mutation is adjacent to the proposed SCO2 copper-binding motif C133xxxC137 (x denotes any amino acid) [10,37–40].

It converts negatively charged glutamate to positively charged lysine and, thus, displaces copper from the copper-binding domain and subsequently reduces CCO activity. Although patients with the homozygous E140K mutation exhibit a milder phenotype with later onset HCM and relatively slow progression of HF, patients with the heterozygous E140K mutation and other SCO2 missense point mutations rapidly develop HCM and HF after birth.

For example, two patients carrying Q53X and E140K mutations in SCO2 were diagnosed with HCM at 6 and 10 weeks of age, respectively, and died at 11 and 24 weeks of age, respectively [38]. By contrast, a patient with the homozygous E140K mutation in SCO2 was diagnosed with HCM at 13 months of age and rapidly progressed to moderate HCM at 21 months of age. Subcutaneous injections of copper-histidine to this patient starting when she was 23 months old for 2 months failed to improve HCM, and interventricular septal thickness increased to 1.5 cm by the end of the treatment [10]. However, compellingly, oral

copper administration at a dose of 140 µg/day starting when the patient was 25 months old remarkably ameliorated HCM, and interventricular septal thickness was reduced to
0.9 cm when she was 39 months old. Although this patient died from severe pneumonia at 42 months of age, she showed significant improvements of HCM and survived longer than all previously reported patients with the same mutation who died before 25 months of age [10].

The G132S mutation in the juxtamembrane region of SCO1 (a SCO2 paralog with complementary functions) is also associated with early onset HCM. This mutation re- duces protein stability of SCO1 by preventing its oligomerization and, thus, decreases CCO activity in skeletal muscle [41]. A patient with this mutation had detectable left ventricle hypertrophy at 2 months of age and died at 6 months of age due to HF with prominent cardiac concentric hypertrophy. Mice with cardiomyocyte-specific ablation of Sco1 show dilated cardiomyopathy with both copper deficiency and a reduction in CCO activity by 60% [42].

Consistently, copper deficiency also reduces protein expression of mitochondrial CCO subunits (MT-COs), which contributes to reduced CCO activity and subsequent cardiomyopathy. In rat neonates with copper deficiency induced by dietary copper restriction, MT-CO1 protein expression decreased by 40% starting at postnatal day (PND) 10, and MT-CO4 protein expression decreased by 20% starting at PND 21, paralleling decreased CCO activity in cardiac mitochondria in these rats [28]. Post-weaning copper repletion in these pups for 6 weeks failed to restore MT-CO1 expression and CCO activity in cardiac mitochondria, and copper repletion for 9 months failed to restore the copper level comparable to that of control rats [28,30]. These data highlight that copper is essential for cardiac development, and that reductions in cardiac CCO activity caused by copper deficiency during the perinatal period have a profound impact on heart functions that lasts into adulthood.

In addition to copper chaperones, the roles of their regulators in heart disease have also been described. CCO assembly factor 6 (COA6) participates in CCO complex assem- bly [43,44]. Thus, a COA6 mutation, W66R, in humans is associated with mitochondrial complex IV deficiency in the heart and subsequently causes cardiac hypertrophy [8]. Fi- broblasts from a patient with this loss-of-function mutation of COA6 showed an increase in protein turnover of mitochondrial complex IV subunits, including MT-CO1, 2, and 4. Compellingly, the addition of copper chloride to these fibroblasts for 7 days partially re- stored protein expression levels of complex IV and its subunits [8]. Further mechanistic studies revealed that COA6 interacts with SCO2 and is required to maintain COX2 protein stability.

The absence of COA6 results in rapid MT-CO2 protein turnover and concomitant reductions in CCO levels and activity in yeast [7]. A further sequencing study revealed that a conserved Cx9CxnCx10C (x denotes any amino acid) motif, which includes W66, in COA6 is crucial for copper-dependent mitochondrial respiration [8]. In addition, patients with mutations of this conserved Cx9CxnCx10C motif in COA6 protein, such as the missense mutation W59C or the nonsense mutation E87X, all had disrupted COA6 functions and sub- sequently developed severe mitochondrial respiratory chain disease (MRCD) with cardiac hypertrophy and HF that resulted in premature death at 1 year of age [45].

Consistently, coa6 knockdown in zebrafish embryos causes heart development defects that parallel those observed in humans [45], highlighting the importance of COA6 in cardiac development and pathology. Taken together, these data demonstrate the key roles of the CCO complex, its copper chaperones, and regulators of these chaperones in cardiac mitochondrial function and the development of copper deficiency-dependent cardiomyopathy. Targeting these factors may be a promising approach to treat cardiac hypertrophy and HF. The impacts of mutations/deletions of SCO1, COA6, and other copper-binding proteins in humans and preclinical animal models discussed in this article are listed in Table 1.

SOD
The copper- and zinc-containing SOD1 enzyme is critical for antioxidant defense by catalyzing dismutation of the deleterious superoxide radical (O2−) to molecular oxygen or hydrogen peroxide, which is in turn reduced to water by other enzymes [59]. The copper chaperone for SOD (CCS) binds to Cu+ via its N- or C-terminus, while it interacts with SOD1 via the central domain. Delivery of copper to SOD1 by the Cu-CCS complex permits formation of disulfide bonds in SOD1, which are required for its enzymatic activity and for the prevention of its misfolding, aggregation, and inactivation [60]. Mice with global Ccs deletion show markedly reduced SOD1 activity due to impaired copper incorporation into SOD1, supporting the important role of CCS in the regulation of SOD1 activity [61].

Compared with wild-type mice, Sod1-knockout mice show exacerbated oxidative stress. Although no cardiac function data were presented, Zhu et al. reported that due to excessive oxidative stress in the ischemic heart, Sod1-knockout mice are vulnerable to cardiac injury after the induction of acute myocardial ischemia by occlusion of the an- terior descending branch of the left coronary artery [46]. In addition, copper deficiency, via a reduction in SOD1 activity in endothelial cells, results in a reduction in nitric oxide (NO) and an elevation of superoxide anions, which in turn impairs endothelial function and adversely affects mouse embryonic heart development, resulting in a swollen heart and pericardial effusion [62]. Although copper-dependent SOD1 is critical for antioxi- dant defense, there are no reports of a link between SOD1 mutations and heart disease in humans.

Manganese-containing SOD2 (MnSOD) localizes to mitochondria and serves as the first line of defense against mitochondrial respiration-generated oxidative stress. Although SOD2 does not bind to copper, it indirectly regulates the activity of copper-containing SOD1. Sod2-deficient mice show increased release of superoxide anion radical derivatives and impaired SOD1 activity, which causes HF [63]. These mice exhibit myocardial damage, with enlarged mitochondria, loss of cristae, and fewer myofilaments, as well as lipid peroxidation and activation of apoptosis. A recent exon sequencing study revealed that the homozygous G181V missense mutation in SOD2 causes severe cardiomyopathy in human newborns, manifesting as severe biventricular dilation and a decreased left ventricular ejection fraction. A patient bearing this mutation died at 4 days of age. This mutation disrupts the mitochondrial superoxide scavenging activity of SOD2, and subsequently

results in the rapid development of HF and death [59]. These studies provided insights into the molecular mechanisms by which copper defects and superoxide induce cardiac injury. Copper-bound SOD3, which is primarily expressed in blood vessels and is extracel- lularly localized, has also been linked to heart disease. Activity and/or expression of copper-bound SOD3 was reported to be decreased in animal models and humans with hypertension, HF, and coronary heart disease [64–66]. Sod3 deletion strikingly increases the expression of collagen and matrix metalloproteinase-2 and -9 and production of super- oxide anions in mice subjected to transaortic constriction (TAC).

Accordingly, Sod3 deletion in mice worsens cardiac hypertrophy, left ventricular dilation, and fibrosis induced by pressure overload [48] and increases myocardial apoptosis, fibrosis, and inflammation induced by doxorubicin [49]. A SOD3 R231G variant, which processes less antioxidant ability, was reported to be positively associated with IHD, myocardial infarction, and HF in diabetic subjects [50–54]. The rs7655372 variant of the SOD3 gene was associated with a significantly increased risk of ischemic stroke in the Chinese Han population of Dali City [67]. By contrast, a protective T-allele of the rs2284659 variant in the promoter region of the SOD3 gene was negatively associated with the incidence of myocardial infarction and cardiovascular and all-cause mortality in both type 1 and type 2 diabetic patients [52]. Consistently, Oster et al. showed that cardiac copper levels positively correlated with the cardiac ejection fraction in 27 patients with coronary heart disease who underwent coronary artery bypass surgery [68]. This suggests that restoration of copper homeostasis in the heart is beneficial to heart disease, likely by improving antioxidant defense and
mitochondrial function.

MTs
MTs are cysteine-rich, low molecular weight proteins that bind to copper and serve as intracellular copper scavengers. The formation of copper-thiolate clusters in MTs sequesters excess copper in cells and thereby minimizes copper toxicity [69]. Thus, deletion of MT1/2 renders mice hypersensitive to copper toxicity [70]. Accordingly, MT1/2-knockout mice show severe cardiac dysfunction, oxidative stress, and cardiac fibrosis, which is further exacerbated during intermittent hypoxia, whereas mice overexpressing cardiac-specific MT-IIa are protected against intermittent hypoxia-induced cardiomyopathy [47]. Although genetic cardiac-specific MT-IIa overexpression in mice failed to prevent the initiation of cardiac hypertrophy induced by copper deficiency, it inhibited progression from cardiac hypertrophy to HF during copper deficiency. This is likely due to attenuation of cardiac lipid peroxidation and a reduction in natriuretic peptide A (ANP) production and, thus, in ANP-induced myocardial apoptosis [71]. These cardiac-specific MT-IIa-overexpressing mice are resistant to doxorubicin-induced cardiotoxicity and atrial contractions due to the prevention of the deterioration of mitochondrial morphology and reduction in creatine phosphokinase levels [72]. These data demonstrate the importance of MTs in cardiac function, oxidative stress, and apoptosis, as well as the protective role of cardiac MTs in the pathologic progression of cardiac hypertrophy to HF.

CP
CP carries more than 90% copper in plasma and, thus, is critical for maintaining activities of copper-dependent enzymes, including SOD1 and 3, and thus the removal of oxygen radicals. CP is also a ferroxidase that is important to mobilize iron. Thus, copper deficiency that reduces CP oxidative activity may indirectly impact heart disease via dysregulation of iron homeostasis [73]. In addition, CP is an oxidase for NO that converts NO to nitrite in vivo [74,75]. Circulating CP is negatively associated with NO bioavailability, presumably due to increased conversions of NO, and enhanced oxidative stress in turn adversely impacts heart function [56].
Although the detailed molecular mechanism is unclear, elevated circulating CP is associated with DM, obesity, dyslipidemia, atherosclerosis, IHD, and mortality [76–83]. Numerous clinical studies show that circulating CP positively correlates with the risk of

HF and mortality and is an independent and robust predictor of cardiovascular disease, HF, and mortality [56,78,79,84]. Two independent genome-wide association studies derived from the Cleveland Clinic GeneBank Study (4177 patients) and the Atherosclerosis Risk in Communities Study (ARIC study, 9240 patients) suggested that a single locus (rs1307255) on chromosome 3 in the CP gene increases CP levels [55,56]. However, both studies demon- strated no association between the rs1307255 variant and the incidence of HF. Nonetheless, these studies revealed that circulating CP levels are associated with major adverse cardio- vascular events, including myocardial infarction, stroke, HF, and all-cause mortality. An increase in one standard deviation of circulating CP (79 mg/L) was associated with a 14% increase in the risk of HF [56]. Greater efforts are needed to understand whether elevated circulating CP plays a causative and pathogenic role in heart disease and the associated molecular mechanisms.

LOX
The copper-dependent LOX enzyme is critical to catalyze lysine-derived crosslinking of collagen and elastin fibrils in the extracellular matrix. Inhibition of collagen and elastin crosslinking reduces the tensile strength and elastic properties of connective tissues, which results in a failure to maintain normal cardiac contraction and the development of concentric cardiac hypertrophy [85,86]. Therefore, reduced LOX activity in hearts of copper-deficient rats resulted in an abnormal connective tissue network and systolic dysfunction [87,88].

The collagen network of the myocardium is primarily composed of type I collagen, which is a large-diameter collagen fiber, and type III collagen, which is a relatively small-diameter collagen fiber. Collagen fibers have high tensile strength. Thus, small changes in the concen- tration, composition, and diameter of collagen and the degree of crosslinking profoundly affect the mechanical properties of the heart [89,90]. Tissues containing predominantly type I collagen and/or with a high degree of crosslinking are stiffer than tissues mainly contain- ing type III collagen and/or with less crosslinking. Copper deficiency reduces the type I-to-type III collagen ratio and, thus, collagen and elastin crosslinking, presumably due to decreased LOX activity. Consistently, in hearts of copper-deficient rats, the type III-to-type I collagen ratio was significantly increased by 5-fold after feeding a copper-deficient diet for 6 weeks and remained 2–3-fold higher from 8 to 12 weeks of feeding [91].

By contrast, an increased level of LOX is associated with excess crosslinking of type I and III collagens in patients with enhanced myocardial stiffness and HF [92,93]. The elevated type III-to-type I collagen ratio results in reduced stiffness and strength of collagen fibrils. These fibrils are also resistant to proteolytic enzymes, resulting in increased collagen deposition in the extracellular space, which initiates pathologic changes in the heart, including cardiac hypertrophy and myocardial fibrosis. Although no reports have linked LOX genetic variance to cardiomyopathy, a single nucleotide polymorphism in the LOX coding region, G473A, which causes the non-conservative R158E mutation in LOX protein, is linked to oral submucous fibrosis [94]. However, further loss- or gain-of-function animal studies and human genetic studies are needed to validate the role of LOX in cardiac fibrosis and to determine whether LOX is a target for the treatment of cardiomyopathy.

reference link: https://doi.org/10.3390/nu14030700


More information: Archita Das et al, Cysteine oxidation of copper transporter CTR1 drives VEGFR2 signalling and angiogenesis, Nature Cell Biology (2022). DOI: 10.1038/s41556-021-00822-7

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.