Secondary infections in mechanically ventilated patients with COVID-19: An overlooked matter?

OBJECTIVE: The susceptibility to infection probably increases in COVID-19 patients due to a combination of virusand drug-induced immunosuppression. The reported rate of secondary infections was quite low in previous studies. The objectives of our study were to investigate the rate of secondary infections, risk factors for secondary infections and risk factors for mortality in COVID-19 critically ill patients. METHODS: We performed a single-center retrospective study in mechanically ventilated critically ill COVID-19 patients admitted to our Critical Care Unit (CCU). We recorded the patients' demographic data; clinical data; microbiology data and incidence of secondary infection during CCU stay, including ventilator-associated pneumonia (VAP) and nosocomial bacteremia (primary and secondary). RESULTS: A total of 107 patients with a mean age 62.2 ± 10.6 years were included. Incidence of secondary infection during CCU stay was 43.0% (46 patients), including nosocomial bacteremia (34 patients) and VAP (35 patients). Age was related to development of secondary infection (65.2 ± 7.3 vs. 59.9 ± 12.2 years, p=0.007). Age ≥ 65 years and secondary infection were independent predictors of mortality (OR=2.692, 95% CI 1.068-6.782, p<0.036; and OR=3.658, 95% CI 1.385- 9.660, p=0.009, respectively). The hazard ratio for death within 90 days in the ≥ 65 years group and in patients infected by antimicrobial resistant pathogens was 1.901 (95% CI 1.198- 3.018; p= 0.005 by log-rank test) and 1.787 (95% CI 1.023-3.122; p= 0.036 by log-rank test), respectively. CONCLUSIONS: Our data suggest that the incidence of secondary infection and infection by antimicrobial resistant pathogens is very high in critically ill patients with COVID-19 with a significant impact on prognosis.

A novel interplay between membrane and nuclear melatonin receptors in human lymphocytes: significance in IL-2 production.

Human lymphocyte melatonin, through membrane and nuclear receptors binding, acts as an activator in IL-2 production. Antagonism of membrane melatonin receptors using luzindole exacerbates the drop of the IL-2 production induced by PGE(2) in peripheral blood mononuclear and Jurkat cells. This paper studies the melatonin membrane and nuclear receptors interplay in PGE(2)-diminished IL-2 production. The decrease in IL-2 production after PGE(2) and/or luzindole administration correlated with downregulation in the nuclear receptor RORalpha. We also highlighted a role of cAMP in the pathway, because forskolin mimicked the effects of luzindole and/or PGE(2) in the RORalpha expression. Finally, a significant RORalpha downregulation was observed in T cells permanently transfected with inducible MT(1) antisense. In conclusion, we show a novel connection between melatonin membrane receptor signalling and RORalpha expression, opening a new way to understand melatonin regulation in lymphocyte physiology.

Nocturnal increases in the triiodothyronine/thyroxine ratio in the rat thymus and pineal gland follow increases of type II 5'-deiodinase activity.

The type II 5'-deiodinase (5'D-II) is regulated by the light-dark cycle in some tissues in which the enzyme is present. This prompted us to investigate putative influences of light-dark cycle on thyroid hormone concentrations in these tissues. The results revealed the following facts: (a) Deiodinase activity in the rat thymus exhibits a nyctohemeral profile with peak values late at night and basal values during the day. The thyroid hormone concentrations in the thymus also show a 24 h rhythm with an increase in the triiodothyronine/thyroxine (TT3/TT4) ratio at night. (b) The content of thyroid hormones in the pineal gland exhibits, like in the thymus, nyctohemeral variations with increase values in the TT3/TT4 ratio during the dark period coinciding with the maximal enzyme activity. (c) Other tissue, like the anterior pituitary, in which 5'D-II, activity does not exhibit a diurnal variation, the concentration of thyroid hormones does not show modifications. In conclusion, the nocturnal increase of 5'D-II activity produces an increase of T3 concentration and a decrease of T4 concentration in both thymus and pineal gland. Therefore, these diurnal changes in 5'D-II activity is a mean by which the cell can regulate the intracellular availability of the most active thyroid hormone T3.

Expression of type II thyroxine 5'-deiodinase from rat harderian gland in Xenopus laevis oocytes.

The presence of isoenzymes mediating the conversion of thyroxine to 3,5,3'-triiodothyronine has been studied according to characteristic kinetics and physiological regulation. In this paper, we report the expression of type II 5'-deiodinase (5'D) activity in oocytes of Xenopus laevis. Oocytes injected with total RNA extracted from rat Harderian gland, and then incubated up to five days demonstrated a progressive increase in 5'D activity, reaching a maximal value at 24 h; then, 5'D activity remained almost stable for an additional period of four days. Characteristics of the enzyme activity expressed by oocytes included its inhibition by iopanoic acid, but not by propylthiouracil, and its increase during beta-adrenergic agonist treatment and hypothyroidism. The expressed activity manifests characteristics typical of the type II isoenzyme. Deiodinating activity in oocytes also exhibited diurnal variations. In this study, 5'D activity expressed in oocytes exhibited low values when animals were killed during the day, and high values when animals were killed at night. Maximal values were reached 3-4 h before the nocturnal peak of 5'D activity in Harderian gland crude homogenates. Results suggest that the in vivo activation of 5'D by isoproterenol, hypothyroidism, or dark exposure may be caused by an increase in the synthesis and/or maturation of the RNA expressing the enzyme.

Melatonin is responsible for the nocturnal increase observed in serum and thymus of thymosin alpha1 and thymulin concentrations: observations in rats and humans.

This paper shows that melatonin regulates both thymosin alpha1 and thymulin production as well as the expression of the prothymosin alpha gene. The results revealed the following facts: (a) The concentrations of thymosin alpha1 in both serum and thymus of rat showed a nyctohemeral profile with peak values late at night and basal values during the day. The concentrations of thymulin in rat serum also showed a 24-h rhythm with an increase in their values at night. This rhythmical character for thymosin alpha1, and thymulin was also found in the human serum. (b) Rats injected with melatonin during the day exhibited a significant increase in the concentrations of both peptides. Moreover, continuous light exposure on the animals at daytime and pinealectomy cause a decrease in thymosin a1 and thymulin concentrations with regards to those found in control rats. (c) Melatonin regulates the expression of the prothymosin alpha gene, analyzed by Northern blot. These results suggest that melatonin may be involved in the regulation of immune functions by increasing the thymic peptides production.

Type II thyroxine 5'-deiodinase in the rat thymus.

In the present study we have shown type II thyroxine 5'-deiodination (5'D) in the rat thymus. The enzyme activity was identified in crude extract homogenates by measuring the 125I released from [3',5'-125I]thyroxine which is used as a substrate of the reaction. The release of 125I is dependent on protein tissue concentration, time, temperature and pH, and is saturable by increasing the substrate concentration, indicating its enzymatic nature. Characteristics of the enzyme activity also include a low Km (9.1 nM), its dependence on dithiothreitol, and its inhibition by iopanoic acid, but not by propylthiouracil. Experiments to investigate the cellular location of the enzyme in the thymic gland showed that the enzyme is present in both stromal cells and thymocytes. At the subcellular level, 5'D activity was associated with cellular membranes. Thyroid status appears to regulate 5'D activity in rat thymus. Hypothyroidism caused an increase in thymus 5'D activity. The Km value remained unchanged (9.1 vs 10.5 nM) during hypothyroidism, but Vmax increased significantly from 17.7 fmol/mg protein per h in euthyroid rats to 53.5 fmol/mg protein per h in hypothyroid rats. 5'D activity was also modulated by catecholamines through beta-adrenergic receptors because isoproterenol, but not methoxamine or clonidine, could activate the enzyme. Because these characteristics define the type II iodothyronine-deiodinating pathway in other tissues, we suggest that the rat thymus also shares this pathway.

Involvement of nuclear receptors in the enhanced IL-2 production by melatonin in Jurkat cells.

This report shows that melatonin enhances IL-2 production by Jurkat cells via a nuclear receptor-mediated mechanism. Jurkat cells express nuclear (RZR alpha, ROR alpha 1, and ROR alpha 2) and membrane (mt1) melatonin receptors, and melatonin binds to Jurkat nuclei and membranes with the same affinity described for human peripheral blood mononuclear cells (PBMCs). Melatonin enhances IL-2 production by Jurkat cells activated by either phytohemagglutinin (PHA) or phorbol myristate acetate (PMA). PHA activation of Jurkat cells does not change the profile of melatonin receptor expression; on the contrary, PMA activation negatively regulates the mt1 receptor. In the absence of the membrane receptor, melatonin still activates the IL-2 production. These results show that the expression of the nuclear melatonin receptor is sufficient for melatonin to activate IL-2 production by Jurkat cells.

Characterization of binding sites for beta-adrenergic agonists and vasoactive intestinal peptide in the rat harderian gland.

Vasoactive intestinal peptide (VIP) receptors and beta-adrenergic receptors were investigated in rat Harderian gland membranes using 125I-VIP and 125I-cyanopindolol (125I-CYP), respectively, as ligands. The receptor bindings were rapid, reversible, saturable, specific, and dependent on time, temperature, and membrane concentration. The stoichiometric data suggested the presence of two classes of VIP receptors with Kd values of 0.36 and 65.37 nM and binding capacities of 323 and 39,537 fmol VIP/mg protein, respectively. The interaction showed a high degree of specificity, as suggested by competitive displacement experiments with several peptides structurally or not structurally related to VIP as follows: VIP > helodermin > rGRF > PHI > > secretin. Glucagon, somatostatin, insulin, and pancreastatin were ineffective at concentrations up to 1 microM. However, the stoichiometric data suggest the presence of one class of binding sites for 125I-CYP. The Kd for the single site was 290 pM with a binding capacity of 32 pmol/L. The pharmacological characterization of 125I-CYP binding to membranes showed that only isoproterenol, a beta-adrenergic agonist, and norepinephrine, an alpha beta-adrenergic agonist, was as effective as propranolol in inhibiting 125I-CYP binding to Harderian gland membranes. However, alpha 1- and alpha 2-adrenergic agonists and blockers such as methoxamine, prazosin, clonidine, and yohimbine were shown to be ineffective. These results demonstrate the presence of specific VIP and beta-adrenergic receptors in the Harderian gland and suggest a role for VIP and beta-adrenergic agonists in the physiology of this gland.

Interaction of vasoactive intestinal peptide (VIP) with rat lymphoid cells.

Receptors for vasoactive intestinal peptide (VIP) have been characterized in rat lymphoid cells. The interaction of [125I] VIP with blood mononuclear cells was rapid, reversible, specific and saturable. At apparent equilibrium, the binding of [125I] VIP was competitively inhibited by native VIP in the 0.01-100 nM range concentration. The binding data were compatible with the existence of two classes of receptors: a high-affinity class with a Kd = 0.050 +/- 0.009 nM and a low binding capacity (2.60 +/- 0.28 fmol/10(6) cells), and a low-affinity class with a Kd = 142 +/- 80 nM and a high binding capacity (1966 +/- 330 fmol/10(6) cells). Secretin, glucagon, insulin and somatostatin did not show any effect at a concentration as high as 100 nM. With spleen lymphoid cells, stoichiometric studies were performed. The binding data were compatible with the existence of two classes of receptors: a high-affinity class with a Kd = 0.100 +/- 0.033 nM and a low binding capacity (4.60 +/- 1.07 fmol/10(6) cells), and low-affinity class with a Kd = 255 +/- 110 nM and high binding capacity (2915 +/- 1160 fmol/10(6) cells). With thymocytes, no binding was obtained under different conditions.

Vasoactive intestinal peptide stimulates type II thyroxine 5'-deiodinase and N-acetyltransferase activities in dispersed pineal cells of euthyroid and hypothyroid rats.

The effect of vasoactive intestinal peptide (VIP) on thyroxine type II 5'-deiodinase (5'-D) and N-acetyltransferase (NAT) activities were studied using pineal cells of euthyroid and hypothyroid rats. VIP activated 5'-D activity in a dose-dependent manner in both euthyroid and hypothyroid animals. However, basal and VIP stimulated activity was higher in pinealocytes from hypothyroid than in cells from euthyroid rats. VIP was also able to stimulate NAT activity but hypothyroidism did not induce modifications in its activity. Both 5'-D and NAT activities were stimulated not only by VIP, but also by isoproterenol, a beta-adrenergic receptor agonist, and forskolin, a potent activator of adenylate cyclase activity. The results suggest that VIP may be involved in the physiological regulation of pineal 5'-D activity.

In vivo activation of pineal N-acetyltransferase but not type II thyroxine 5'-deiodinase by phenylephrine in young rats.

The regulation by alpha- and beta-adrenergic agonists of pineal N-acetyltransferase (NAT) and type II thyroxine 5'-deiodinase (5'-D) in rats at either 2 or 6 weeks of age was studied. The pattern of stimulation was different because NAT activity could be clearly activated by an alpha-adrenergic agonist, phenylephrine, only in 2-week-old rats. However, isoproterenol, a beta-adrenergic agonist, was able to stimulate NAT activity in rats at both 2 and 6 weeks of life. On the other hand, phenylephrine was always ineffective in stimulating 5'-D activity, while isoproterenol clearly activated it at both ages. These results strongly suggest a role for alpha-adrenergic receptors, in addition to beta-adrenergic receptors, in regulating rat pineal NAT activity during development.

Beta- and alpha-adrenergic mechanisms are involved in regulating type II thyroxine 5'-deiodinase in rat thymus.

The role of adrenergic receptors in regulation of rat thymus type II thyroxine 5'-deiodinase (5'-D) activity was investigated. Our results show that norepinephrine, an alpha- and beta-adrenergic agonist elicited an increase in thymus 5'-D activity. Isoproterenol, beta-adrenergic agonist, also increased the enzyme activity, being less effective than norepinephrine. Moreover, alpha-adrenergic agonists, i.e., methoxamine, an alpha1-agonist, and clonidine, an alpha2-agonist, did not increase 5'-D activity. The effect of isoproterenol was potentiated by methoxamine, but the potentiating effect was observed only at doses of isoproterenol which induce submaximal activation of the enzyme. Administration of propranolol, beta-adrenergic blocker, and prazosin, an alpha1-adrenergic blocker, inhibited significantly the activation of the enzyme induced by norepinephrine. However, yohimbine, an alpha2-adrenergic blocker, had small effect. These results show, in hypothyroid rats, a clear regulation by adrenergic mechanisms of 5'-D activity in the thymus, where alpha- and beta-adrenergic receptors might be involved.

Different experimental conditions which regulate type II 5'-deiodinase mRNA in rat Harderian gland.

In the present study, we describe the modifications in the expression of type II 5'deiodinase activity (5'D) in Xenopus laevis oocytes by injection of polyadenylated (poly A) mRNA from hypothyroid rat Harderian gland. The time-course study showed that the expression of the enzyme was dependent on time. Thus, enzyme activity was observed in oocytes 6 and 12 hours after the injection with poly A mRNA, reaching a maximal value at 24 hours. The activity was partially inhibited by 6-n-propyl-thiouracil, completely inhibited by iopanoic acid and exhibited a higher affinity for the T4 (Km=1.5 nM) than rT3 (Km=20 nM). The expression of the enzyme was modified in different experimental conditions: (a) exhibited diurnal variations with maximal peak values at night, (b) was inhibited by light at night and, (c) was activated by isoproterenol. On the other hand, we have also identified, for the first time, the size of mRNA capable of inducing 5'D in rats.

Interaction of a bovine thymic peptide extract with vasoactive intestinal peptide (VIP) receptors.

Bovine thymic peptide extract (1-100 micrograms/ml) is shown to completely inhibit the binding of [125I]VIP to rat blood mononuclear cells, lymphoid cells of spleen, and liver plasma membranes. In the three models, the bovine thymic peptide extract inhibits [125I]VIP binding with a potency that is 4000-7000 times lower than that of the native VIP, on a weight basis. In rat liver plasma membranes, the bovine thymic peptide extract stimulates adenylate cyclase with a maximal efficiency that is similar to that of VIP. At maximal doses, VIP and thymic peptide extract do not exert an additive effect on adenylate cyclase, suggesting that the activation of the enzyme by the bovine thymic peptide extract occurs through VIP receptors. Finally, no VIP-like immunoreactivity was detected in the thymic peptide extract using an antiserum raised against mammalian VIP. All these data suggest the presence in the bovine thymic peptide extract of a new substance which behaves as a VIP agonist in rat.

Vasoactive intestinal peptide (VIP) binding to solubilized material from rat liver plasma membranes.

A non-ionic detergent such as Lubrol-PX extracts in soluble form the VIP-binding structures of rat liver plasma membranes. Detergent-solubilized proteins bind specifically [125I]VIP and the complex tracer-protein is identified by the use of Sepharose 6B columns. The interaction is only possible in the absence of detergent (below 0.001%) and is inhibited by native peptide. A molecular weight of about 80,000 was estimated for VIP-binding proteins by reference to a series of globular markers of proteins. Binding to VIP soluble proteins is specific and dependent on time as studied by the Hummel and Dreyer (Biochim. Biophys. Acta 63:530-532, 1962) assay.

Melatonin biosynthesis in the thymus of humans and rats.

Melatonin is an indoleamine widely distributed in the evolution that shows a great functional versatility, playing an important role as a transmitter of photoperiodic information and exhibiting antioxidant, oncostatic, anti-aging and immunomodulatory properties. In vertebrates, this molecule is produced by the pineal gland and other extrapineal sites. The present study was carried out to investigate the presence of melatonin in thymus and the possibility of an endogenous melatonin synthesis in this organ, in which T cells are matured. In this work, we demonstrate in humans and rats that thymus contains melatonin, expresses the mRNAs encoding N-acetyltransferase and hydroxyindol-O-methyltransferase, the two key enzymes of the melatonin synthesis, and has this biosynthetic machinery activated. In addition, rat thymocytes cultured for 24 h exhibited high levels of melatonin. The results presented here suggest that human and rat thymuses are able to synthesize melatonin, which could have intracrine, autocrine and paracrine functions.

The effect of experimental hypothyroidism on phosphofructokinase activity and fructose 2,6-bisphosphate concentrations in rat heart.

Experimental hypothyroidism was induced in rats by the administration of NaClO4. Hearts from normal and hypothyroid rats were homogenized, and the extracts were assayed for phosphofructokinase-1 and phosphofructokinase-2 activity and fructose 2,6-bisphosphate concentrations. Hypothyroidism was associated with a drastic loss of phosphofructokinase-1 activity. A hyperbolic relationship between plasma thyroxine concentrations and phosphofructokinase-1 activity was found. As treatment with NaClO4 progressed, the decrease in blood thyroxine was faster than the decrease in enzyme activity. After prolonged hypothyroidism (a decrease in thyroxine of more than 10-fold), a 4-fold decrease in phosphofructokinase-1 activity was observed. In this metabolic condition 2-fold decreases in phosphofructokinase-2 activity and in fructose 2,6-bisphosphate were observed. A similar decrease in phosphofructokinase-1 activity in a partially purified preparation was found. The addition of L-thyroxine in the diet had little effect on phosphofructokinase-1 activity. However, exposure of minced pieces of hearts of hypothyroid rats to tri-iodothyronine for 5 h resulted in a clear increase in phosphofructokinase-1 activity, which was partially prevented by the simultaneous addition of cycloheximide. These results could account for the decrease in carbohydrate metabolism in heart from hypothyroid rats.

Continuous light exposure modifies the nocturnal increase in rat thymus type II thyroxine 5'-deiodinase.

In the present study we show that thymus type II thyroxine deiodinase activity exhibits a nyctohemeral profile, with basal values during the day and high values at night. This rhythmic character is dependent on neuroadrenergic input since exposure to continuous light at night completely abolished the nocturnal rise of the enzyme activity. However, treatment with isoproterenol under light exposure at night restored it.

Interaction of vasoactive intestinal peptide (VIP) with human peripheral blood lymphocytes: specific binding and cyclic AMP production.

VIP binding sites and cyclic AMP production by the peptide have been studied in human blood mononuclear cells before and after selective depletion of or enrichment for T-lymphocytes, B-lymphocytes-K-NK cells and monocytes. The specifically bound 125I-labelled VIP correlated significantly with the presence of B-lymphocytes and/or cells of K-NK system. The stoichiometric data were compatible with the existence of two classes of binding sites. T-lymphocytes and monocytes did not show binding of the tracer. The cyclic AMP production stimulated by VIP correlated significantly with the presence of B-lymphocytes and/or K-NK cells.

Thyroxine type II 5'-deiodinase activity in pineal and Harderian gland is enhanced by hypothyroidism but is independent of serum thyroxine concentrations during hyperthyroidism.

1. This paper studies the effect of thyroid status on 5'-D activity in pineal gland, Harderian gland, brown adipose tissue (BAT), pituitary gland, brain frontal cortex (BFC), and cerebellum. 2. Hypothyroidism clearly increased diurnal 5'-D activity in Harderian gland, BAT, pituitary gland, BFC, and cerebellum. In pineal gland, diurnal values of 5'-D activity were not affected by hypothyroidism. 3. Hypothyroidism in adult rats clearly enhanced nocturnal increase of 5'-D activity in pineal and Harderian gland. Congenital hypothyroidism also enhanced the nocturnal increase of 5'-D activity in pineal gland. 4. Hyperthyroidism inhibited 5'-D activity in pituitary gland, BFC, and cerebellum. A small inhibition, although significant, was found in BAT. 5. In pineal and Harderian gland, hyperthyroidism did not inhibit either the basal diurnal values of the enzyme or the nocturnal increase of its activity. 6. Results suggest that, in tissues where 5'D-activity is regulated by adrenergic mechanisms, mostly pineal gland and Harderian gland, the enzyme activity is independent of serum T4 concentrations during hyperthyroidism.