Pathophysiology and mechanism of long COVID: a comprehensive review.

BACKGROUND: After almost 2 years of fighting against SARS-CoV-2 pandemic, the number of patients enduring persistent symptoms long after acute infection is a matter of concern. This set of symptoms was referred to as "long COVID", and it was defined more recently as "Post COVID-19 condition" by the World health Organization (WHO). Although studies have revealed that long COVID can manifest whatever the severity of inaugural illness, the underlying pathophysiology is still enigmatic. AIM: To conduct a comprehensive review to address the putative pathophysiology underlying the persisting symptoms of long COVID. METHOD: We searched 11 bibliographic databases (Cochrane Library, JBI EBP Database, Medline, Embase, PsycInfo, CINHAL, Ovid Nursing Database, Journals@Ovid, SciLit, EuropePMC, and CoronaCentral). We selected studies that put forward hypotheses on the pathophysiology, as well as those that encompassed long COVID patients in their research investigation. RESULTS: A total of 98 articles were included in the systematic review, 54 of which exclusively addressed hypotheses on pathophysiology, while 44 involved COVID patients. Studies that included patients displayed heterogeneity with respect to the severity of initial illness, timing of analysis, or presence of a control group. Although long COVID likely results from long-term organ damage due to acute-phase infection, specific mechanisms following the initial illness could contribute to the later symptoms possibly affecting many organs. As such, autonomic nervous system damage could account for many symptoms without clear evidence of organ damage. Immune dysregulation, auto-immunity, endothelial dysfunction, occult viral persistence, as well as coagulation activation are the main underlying pathophysiological mechanisms so far. CONCLUSION: Evidence on why persistent symptoms occur is still limited, and available studies are heterogeneous. Apart from long-term organ damage, many hints suggest that specific mechanisms following acute illness could be involved in long COVID symptoms. KEY MESSAGESLong-COVID is a multisystem disease that develops regardless of the initial disease severity. Its clinical spectrum comprises a wide range of symptoms.The mechanisms underlying its pathophysiology are still unclear. Although organ damage from the acute infection phase likely accounts for symptoms, specific long-lasting inflammatory mechanisms have been proposed, as well.Existing studies involving Long-COVID patients are highly heterogeneous, as they include patients with various COVID-19 severity levels and different time frame analysis, as well.

[Access reorganization to the scientifc resources for the Belgian health federal institutions: creation of VDIC and the Vesalius.be portal]. / La réorganisation de l'accès aux ressources scientifiques à destination des institutions fédérales belges de santé: creation du VDIC et du portail Vesalius.be.

For several years, the Belgian federal civil servant has had insufficient access to the documentary resources in the field of the public health. A virtual structure gathering the collections of all of the federal resource centres and organizing the work of their teams was implemented in 2005: the Vesalius Documentation Information Centre (VDIC). A portal providing access to the scientific resources was officially launched in 2006. It offers the civil servant a simple and direct access to more than 2,000 journals in full text, a centralized catalogue of monographies and a selection of relevant databases. The portal had immediately a great success, as it broadly met the needs of its users. The VDIC portal impacts positively on the quality of information exchange and consequently improves the services rendered to the citizen. This portal is also directly intended for him: it provides for example access to the partners' scientific publications or to resources selected for their quality, thereby enabling the citizen to get more involved in his own health.

Functional characterization of neurotensin receptors in human cutaneous T cell lymphoma malignant lymphocytes.

Cutaneous T cell lymphomas are a clonal proliferation of CD4+ T lymphocytes primarily involving the skin. Mycosis fungoides is an epidermotropic CD4+ cutaneous T cell lymphoma, and a more aggressive form, Sezary syndrome, occurs when the malignant cells become nonepidermotropic. The role of neuropeptides in the growth and chemotaxis capacity of cutaneous T cell lymphoma cells remains unknown. In this report, we found that cutaneous T cell lymphoma cells, similarly to normal resting or activated peripheral lymphocytes, were able to bind neurotensin. We used an interleukin-2-dependent cutaneous T cell lymphoma malignant T cell line derived from cutaneous T cell lymphoma lesions in order to study the role of neurotensin in the proliferation and migration of these malignant cells. First, we determined that the malignant cells expressed neurotensin receptors on their cell membrane. Functional results indicated that neurotensin did not stimulate the growth of the cell line. In contrast, this neuropeptide inhibited the proliferation of the tumor cells in response to exogenous interleukin-2. Furthermore, we found that neurotensin enhanced both spontaneous and chemoattractant-induced migration of the malignant cells. This suggests that neurotensin in skin can play a role in the disease by locally limiting the growth of the cutaneous T cell lymphoma tumor cells in response to cytokines and by enhancing their chemotaxis capacity.

Cloning and expression of mouse-brain calmodulin as an activator of Bordetella pertussis adenylate cyclase in Escherichia coli.

Cloning of higher eukaryotic genes has seldom been performed by complementation of a defective prokaryotic function. This is especially true in the case of functions that are normally absent from the prokaryotic host. We demonstrate here that it is possible to identify by complementation the cDNA from mouse brain, which encodes calmodulin (CaM) synthesis, in spite of the fact that the recipient strain, Escherichia coli, does not normally harbour a CaM function. A three-component cloning procedure in which a gene product requiring CaM for activity, adenylate cyclase from the pathogen Bordetella pertussis, was used to screen a cDNA library for cAMP synthesis in E. coli. The nucleotide sequence of the corresponding cDNA is also reported.

Molecular cloning of a levocabastine-sensitive neurotensin binding site.

A search for sequences homologous to the neurotensin receptor cDNA in a rat hypothalamic library has identified a novel neurotensin receptor (NTR-2). The 1539 bp cDNA encodes a 416 amino acid protein and shows highest homology to the previously cloned neurotensin receptor (NTR-1) (64% homology and 43% identity). Binding and pharmacological studies demonstrate that NTR-2 expressed in COS cells recognizes neurotensin (NT) with high affinity as well as several other agonists and antagonists. However, a fundamental difference was found; unlike NTR-1, NTR-2 recognizes, with high affinity, levocabastine, a histamine H1 receptor antagonist previously shown to compete with NT for low-affinity binding sites in brain.

Primary structure and functional expression of mouse pituitary and human brain corticotrophin releasing factor receptors.

Corticotrophin-releasing factor (CRF) is the principal hypothalamic factor governing the pituitary-adrenal axis, but the wide extra-pituitary distribution of CRF and its receptors suggest a major role for this neuropeptide in the integration of the overall physiological and behavioral responses of an organism to stress. We have cloned a CRF receptor complementary DNA (cDNA) by expression in COS-7 cells of a cDNA library from the AtT20 mouse pituitary tumour cell line. The cloned mouse cDNA was then as a probe to isolate a human CRF receptor cDNA from a human brain cDNA library. The mouse and human cDNAs both encode 415 amino acid proteins that are 97% identical, containing seven putative transmembrane domains characteristic of G protein-coupled receptors. The CRF receptor shows homology with the receptors for growth hormone-releasing factor, vasoactive intestinal peptide, secretin, parathyroid hormone, and calcitonin. COS-7 cells transfected with the mouse CRF receptor cDNA bind radiolabelled ovine CRF with high affinity and respond specifically to CRF by accumulation of intracellular cAMP. A 2.7 kb mRNA coding for the CRF receptor could be detected in AtT20 cells and human cortex tissue. PCR analysis also detected the receptor transcript in human pituitary, brainstem, and testis.

ORL1, a novel member of the opioid receptor family. Cloning, functional expression and localization.

Selective PCR amplification of human and mouse genomic DNAs with oligonucleotides encoding highly conserved regions of the delta-opioid and somatostatin receptors generated a human DNA probe (hOP01, 761 bp) and its murine counterpart (mOP86, 447 bp). hOP01 was used to screen a cDNA library from human brainstem. A clone (named hORL1) was isolated, sequenced and found to encode a protein of 370 amino acids whose primary structure displays the seven putative membrane-spanning domains of a G protein-coupled membrane receptor. The hORL1 receptor is most closely related to opioid receptors not only on structural (sequence) but also on functional grounds: hORL1 is 49-50% identical to the murine mu-, delta- and kappa-opioid receptors and, in CHO-K1 cells stably transfected with a pRc/CMV:hORL1 construct, ORL1 mediates inhibition of adenylyl cyclase by etorphine, a 'universal' (nonselective) opiate agonist. Yet, hORL1 appears not to be a typical opioid receptor. Neither is it a somatostatin or sigma (N-allylnormetazocine) receptor. mRNAs hybridizing with synthetic oligonucleotides complementary to mOP86 are present in many regions of the mouse brain and spinal cord, particularly in limbic (amygdala, hippocampus, septum, habenula, ...) and hypothalamic structures. We conclude that the hORL1 receptor is a new member of the opioid receptor family with a potential role in modulating a number of brain functions, including instinctive behaviours and emotions.

Cloning and expression of a complementary DNA encoding a high affinity human neurotensin receptor.

A human neurotensin receptor (hNTR) cDNA was cloned from the colonic adenocarcinoma cell line HT29. The cloned cDNA encodes a putative peptide of 418 amino acids with 7 transmembrane domains. The amino acid sequence of the hNTR is 84% identical to the rat NTR [Neuron, 4 (1990) 847-854]. Transfection of this cDNA into COS cells results in the expression of receptors with pharmacological properties similar to those found with HT29 cells. Northern blot analysis using the hNTR cDNA probe indicated a single transcript of 4 kb in the brain, the small intestine and blood mononuclear cells.

Molecular cloning of a human beta 3-adrenergic receptor cDNA.

We report the molecular cloning of a beta 3-adrenergic receptor [beta 3-AR] cDNA from human brown adipose tissue. The cDNA-encoded protein is identical to the previously cloned beta 3-AR but with 6 additional amino acids at the C-terminus. The C-terminus is shared by the beta 3 receptors expressed in human neuroblastoma cells [SK-N-MC] [Mol. Pharmacol. 42 (1992) 964-970]. Furthermore, using a polymerase chain reaction strategy we have cloned and sequenced the beta 3-AR introns. Sequence analysis demonstrates that the human beta 3-AR gene comprises at least 3 exons and 2 introns and that the most abundant beta 3-AR transcripts encode a protein with an exon 3-derived C-terminus. Interestingly, although a similar organization has been found in rodent genes, the rat beta 3-AR transcripts encode a receptor with an exon 2-derived C-terminus.

Spatial and temporal dynamics of Puumala hantavirus infection in red bank vole (Clethrionomys glareolus) populations in Belgium.

Dynamics of hantavirus infection and population densities in rodents were investigated from 1996 to 1999 in southern Belgium. Evidence of Puumala infection was restricted to Clethrionomys glareolus. Although the serotype was not determined, antibodies against hantavirus were also found in eight Apodemus sylvaticus. In fall 1996, the seroprevalence in C. glareolus was high (20.1%, 37 of 184) and the infection was widely distributed in the area studied whereas a focal occurrence of positive rodents and lower seroprevalence rates were recorded in spring 1997 (14.3%, six of 42), fall 1997 (6. 6%, 11 of 166), spring 1998 (6.4%, three of 47) and fall 1998 (6.7%, 11 of 165). A pullulation of rodents was observed in spring 1999 and was associated with a markedly higher seroprevalence in C. glareolus (47.7%, 189 of 396). In all seasons, infection rates in adults were higher than in juveniles and subadults. No significant difference of prevalence was recorded between males and females. In two trapping sites, the temporary disappearance of positive animals after a crash in rodent populations suggests that a threshold in density is necessary for the maintenance of the enzootic cycle.

Genetic characterization of Puumala hantavirus strains from Belgium: evidence for a distinct phylogenetic lineage.

Puumala hantavirus (PUUV) sequences were recovered from red bank voles (Clethrionomys glareolus) trapped between 1996 and 1998 in four localities of southern Belgium: Thuin, Montbliart, Momignies and Couvin. In addition, three PUUV isolates originating from bank voles trapped in the 1980s in southern (Montbliart) and northern (Turnhout) Belgium were genetically characterized. Analysis of the complete S and partial M segment sequences showed that the Belgian PUUV strains constitute a genetic lineage, distinct from other known PUUV lineages from Europe and Japan. This lineage also includes a wild strain (Cg-Erft) originating from a neighbouring area of Germany. Within the Belgian lineage, geographical clustering of genetic variants was observed. In the Montbliart site, the range of diversity between the most temporally distant strains (from 1986 and 1996-1998) was higher than between those from 1996 and 1998, suggesting slight genetic drift via accumulation of neutral or quasi-neutral substitutions with time.

Functional, biochemical and molecular biological evidence for a possible beta(3)-adrenoceptor in human near-term myometrium.

The possible existence of a beta(3)-adrenoceptor (beta(3)-AR) in human near-term myometrium was investigated by in vitro functional and biochemical studies and analysis of mRNA expression. SR 59119A and SR 59104A and CGP 12177 (two selective agonists and a partial agonist, respectively, of the beta(3)-AR), salbutamol and terbutaline (beta(2)-AR agonists) each produced a concentration-dependent relaxation of the myometrial spontaneous contractions. There were no differences in pD(2) values for the relaxing potencies of terbutaline, salbutamol, CGP 12177 and SR 59119A. The rank order for their relaxing efficacies was SR 59119A>SR 59104A>terbutaline approximately salbutamol approximately CGP 12177 (E(max)=52+/-7%, 42+/-12% and approximately 30% respectively). Propranolol, a beta(1)- and beta(2)-AR antagonist, and ICI 118551, a beta(2)-AR antagonist (both at 0.1 microM), did not affect the SR 59119A-induced relaxation whereas SR 59230A, a selective beta(3)-AR antagonist (1 microM), significantly reduced the maximal relaxing effect of SR 59119A. SR 59119A and salbutamol induced a significant increase in cyclic AMP levels that was antagonized by SR 59230A but not by propranolol for SR 59119A, and by propranolol but not by SR 59230A for salbutamol. The beta(3)-AR mRNA was positively expressed in myometrium preparations in a reverse transcription polymerase chain assay. The results presented provide the first evidence for the existence of the beta(3)-AR subtype in human near-term myometrium and suggest that the effects of SR 59119A might be mediated through an increase in cyclic AMP level.

In vitro functional evidence of different neurotensin-receptors modulating the motor response of human colonic muscle strips.

1. The newly developed non-peptide neurotensin (NT)-receptor antagonists SR 48692 and SR 142948 were used to challenge NT responses of human colonic circular smooth muscle strips in vitro. The presence of NT1 and NT2 receptor transcripts in this tissue was tested by reverse transcriptase polymerase chain reaction (RT - PCR) analysis. 2. NT potently and dose-dependently contracted muscle strips, with significant regional differences in potency and efficacy between the transverse and distal colon: EC50, 3.6 and 7.5 nM; the maximal effect was 70 and 55% of 0.1 mM carbachol. Colonic responses to NT in both segments were virtually the same in the presence of atropine (1 microm), levocabastine (10 microM) or tetrodotoxin (1 microM). 3. SR 142948 (10 nM - 1 microM) competitively antagonized NT responses in the transverse and distal colon with similar affinities: pA2 values 8.71 and 8.45, slopes 0.98 and 0.99. SR 48692 (10 nM - 10 microM) antagonized the NT response competitively in the distal colon (pA2 6.55, slope 0.79) and non-competitively in the transverse colon (pA2 8.0, slope 0.51). Neither compound had any agonist effect. 4. The fact that the specific antagonists prevented NT-evoked atropine- and tetrodotoxin-insensitive mechanical responses of colonic muscle strips is highly consistent with the presence in these tissues of non-neuronal NT receptors, whose heterogeneity in the transverse segment is supported by the non-competitive antagonism of SR 48692. The finding of NT1 receptor transcript in both transverse and distal colon suggests its identity with the lower affinity site disclosed functionally by SR 48692 in these segments.

Neurotensin is an antagonist of the human neurotensin NT2 receptor expressed in Chinese hamster ovary cells.

The human levocabastine-sensitive neurotensin NT2 receptor was cloned from a cortex cDNA library and stably expressed in Chinese hamster ovary (CHO) cells in order to study its binding and signalling characteristics. The receptor binds neurotensin as well as several other ligands already described for neurotensin NT1 receptor. It also binds levocabastine, a histamine H1 receptor antagonist that is not recognised by neurotensin NT1 receptor. Neurotensin binding to recombinant neurotensin NT2 receptor expressed in CHO cells does not elicit a biological response as determined by second messenger measurements. Levocabastine, and the peptides neuromedin N and xenin were also ineffective on neurotensin NT2 receptor activation. Experiments with the neurotensin NT1 receptor antagonists SR48692 and SR142948A, resulted in the unanticipated discovery that both molecules are potent agonists on neurotensin NT2 receptor. Both compounds, following binding to neurotensin NT2 receptor, enhance inositol phosphates (IP) formation with a subsequent [Ca2+]i mobilisation; induce arachidonic acid release; and stimulate mitogen-activated protein kinase (MAPK) activity. Interestingly, these activities are antagonised by neurotensin and levocabastine in a concentration-dependent manner. These activities suggest that the human neurotensin NT2 receptor may be of physiological importance and that a natural agonist for the receptor may exist.

Acridine-linked oligonucleotide probes for the short dAT-rich motifs in the 3'-untranslated region of cytokine genes.

We have investigated the use of oligonucleotide probes for identifying cDNA clones containing the short dAT-rich motifs found in the 3'-untranslated region of cytokine genes. To obtain sufficiently stable duplexes between the octameric probes used to identify genes containing the sequence dTATTTATT and its complement, it was necessary to couple an intercalating agent, an acridine derivative (acr), to the 5'-positions of the probes. The resulting octamers 5'-acr-dAATAAATA and, particularly, 5'-acr-dTATTTATT were successfully used to distinguish the complementary sequences in cDNA from internal, single point mismatched sequences. Southern blot analyses of plasmids containing IL-1 beta and IL-8 gave positive results with the 3' degenerate probe, 5'-acr-dTATTTATTN, clearly showing that the very short probe approach can be used in this type of analysis. Subsequently, in slot blot analyses we found that, even without the degenerate nucleotide, N, plasmids bearing cytokine sequences with at least 7 contiguous matched nucleotides could be unambiguously identified with 5'-acr-dTATTTATT. Unfortunately, because of the ubiquity of these dAT-rich sequences in bacterial DNA, it was not possible to use these probes for direct colony screening. In contrast to the results obtained with DNA, at the RNA level, with IL-1 beta mRNA bound to nitrocellulose, the hybrid formed with 5'-acr-dAATAAATA was very unstable, even in 1M LiCl solution at 2 degrees C; however, in the same salt solution the slightly longer acridine-coupled probes 5'-acr-dAATAAATAGGG and 5'-acr-dAAAGAACAA remained hybridized to their complementary sequences up to about 18 degrees C.

Molecular cloning of the MCP-3 chemokine gene and regulation of its expression.

We have isolated a cDNA (NC28) transcribed from a mRNA which is transiently induced in U937 promonocytic cells by PMA and super-induced by cycloheximide. NC28 cDNA encodes a new member of the chemokine family, MCP-3, recently purified from MG-63 osteosarcoma cells by Van Damme et al. [1]. The MCP-3 protein sequence shows 74% identity with human monocyte chemoattractant protein 1 (MCP-1) and, like MCP-1, recombinant MCP-3 protein shows chemotactic activity for monocytes but not for neutrophils. However the secreted MCP-3 protein differs from MCP-1 in being N-glycosylated. The 3' noncoding regions of MCP-3 and MCP-1 mRNAs are more diverged (44%), allowing specific cDNA probes to be made, and indicating that the two genes are evolutionarily distant. Sequence comparisons of the 3' noncoding regions suggest that MCP-3 may be the human homologue of the mouse MARC gene [2], and that MCP-1 and MCP-3 genes arose by a gene duplication event before the mammalian radiation. Both MCP-1 and MCP-3 mRNAs are expressed by PBMC, principally by monocytes, with MCP-1 mRNA being expressed at levels 2-4 times that of MCP-3 mRNA. However, while MCP-1 mRNA is also expressed at high levels in fibroblast or astrocytoma cell lines after IL-1 and TNF stimulation, MCP-3 mRNA is expressed only at very low levels in these cells. The cellular origin of MCP-3 is thus more restricted than that of MCP-1. In our experiments on PBMC, LPS is not a consistent inducer of MCP-1 and MCP-3 mRNAs. In some experiments, it actually decreases levels of these two mRNAs, while concomitantly increasing IL-6 and TNF-alpha mRNA levels. Levels of MCP-1 and MCP-3 mRNAs in PBMC are both increased by IFN-gamma, although IL-6 mRNA is not induced. They are also increased by PHA-P and are decreased, in most cases, by IL-13 [3]. MCP-1 and MCP-3 mRNAs are thus co-ordinately regulated in monocytes in response to a number of inducing or inhibitory agents, in a manner differing in several respects from that of other monokines such as IL-6.

[Problem-based learning, description of a pedagogical method leading to evidence-based medicine]. / Le Problem-Based Learning, description d'une pédagogie conduisant à la médecine factuelle.

Problem-Based Learning is an educational method which uses health care scenarios to provide a context for learning and to elaborate knowledge through discussion. Additional expectations are to stimulate critical thinking and problem-solving skills, and to develop clinical reasoning taking into account the patient's psychosocial environment and preferences, the economic requirements as well as the best evidence from biomedical research. Appearing at the end of the 60's, it has been adopted by 10% of medical schools world-wide. PBL follows the same rules as Evidence-Based Medicine but is student-centered and provides the information-seeking skills necessary for self-directed life long learning. In this short article, we review the theoretical basis and process of PBL, emphasizing the teacher-student relationship and discussing the suggested advantages and disadvantages of this curriculum. Students in PBL programs make greater use of self-selected references and online searching. From this point of view, PBL strengthens the role of health libraries in medical education, and prepares the future physician for Evidence-Based Medicine.

Identification of TIAR as a protein binding to the translational regulatory AU-rich element of tumor necrosis factor alpha mRNA.

In monocyte/macrophages, the translation of tumor necrosis factor alpha (TNF-alpha) mRNA is tightly regulated. In unstimulated cells, translation of TNF-alpha mRNA is blocked. Upon stimulation with lipopolysaccharides, this repression is overcome, and the mRNA becomes efficiently translated. The key element in this regulation is the AU-rich element (ARE). We have previously reported the binding of two cytosolic protein complexes to the TNF-alpha mRNA ARE. One of these complexes (complex 1) forms with extracts of both unstimulated and lipopolysaccharide-stimulated macrophages and requires a large fragment of the ARE containing clustered AUUUA pentamers. The other complex (complex 2) is only detected after cell activation, binds to a minimal UUAUUUAUU nonamer, and is composed of a 55-kDa protein. Here, we report the identification of the RNA-binding protein TIAR as a protein involved in complex 1. The RNA sequence bound by TIAR and the cytoplasmic localization of this protein in macrophages argue for an involvement of TIAR in TNF mRNA posttranscriptional regulation.

Inhibition by SR 59119A of isoprenaline-, forskolin- and VIP-induced relaxation of human isolated bronchi.

In the human isolated bronchus (HIB) it has been shown that beta(3)-adrenoceptor stimulation fails to induce relaxation of airway smooth muscle. It has however been reported in human ventricular endomyocardial biopsies that beta(3)-adrenoceptor stimulation induced a marked negative inotropic effect which could be linked to Gi protein activation. The aims of this study were: (1) to determine in HIB (internal diameter 1-2 mm) whether the selective beta(3)-adrenoceptor agonist SR 59119A (N[7-methoxy-1,2,3, 4-tetrahydronaphthalen-(2R)methyl]-(2R)-2-hydroxy-2-(3-chloroph eny l)e thanamine hydrochloride) was able to inhibit adenylate-cyclase-mediated airway smooth muscle relaxation induced by isoprenaline, forskolin or vasoactive intestinal peptide (VIP) and (2) to investigate the role of the Gi protein in this interaction. SR 59119A (0.1 microM and 1 microM) induced a shift to the right of concentration response curve for isoprenaline (-0. 15+/-0.06 and -0.54+/-0.21 log unit, P<0.05 and P<0.01 respectively), forskolin (-0.12+/-0.02 and -0.30+/-0.05 log unit, P<0.001), and VIP (-0.42+/-0.12 log unit, P<0.01 with SR59119A 10(-6)M). The inhibitory effect of SR 59119A was (1) abolished by an incubation of HIB with pertussis toxin (1 microg/ml, during 15 h in Krebs-Henseleit solution, at 21 degrees C), which is known to inactivate the Gi protein and (2) increased after an incubation of HIB with the pro-inflammatory cytokine IL-1beta (10 ng/ml, during 15 h in Krebs-Henseleit solution, at 21 degrees C), which is known to up-regulate Gi protein expression. Our results suggest that the selective beta(3)-adrenoceptor agonist SR59119A might inhibit the cAMP-dependent relaxation of human isolated bronchus through Gi protein-mediated inhibition of adenylate cyclase.