Gelsolin

>> Proposed Mechanism of Action
>> Critical Care Environment

Quick links:
-> Summary
-> Gelsolin and plasma gelsolin biochemistry and actions
-> Gelsolin in patients
-> Gelsolin in animal models
-> References

Summary.

Gelsolin is an endogenous actin-binding molecule that exists in two forms, an intracellular and extracellular form referred to as plasma gelsolin. These two molecules differ only in a 23 amino acid “tail” in plasma gelsolin. Gelsolin has been show to sever filamentous actin to globular actin and “cap” filamentous actin to prevent further polymerization. These activities are critical for remodeling the cell’s cytoskeleton which is an integral process by which cell mobility is achieved. In addition, gelsolin appears to inhibit apoptosis by stabilizing mitochondria. Gelsolin has a highly conserved molecular structure from Drosophila to man. No evidence of antibodies against gelsolin has been seen in patient receiving multiple transfusions or in multiparus women.

Plasma gelsolin is a secreted protein that circulates in the extracellular fluids of humans at concentrations averaging 250 mg/L. Tissue injury of diverse types leads to prolonged reductions in plasma gelsolin levels. Following tissue injury, such as is encountered in severe trauma, burn, major surgery and bone marrow transplant patients, gelsolin levels decline to 0-25% of the normal level. This decline precedes and, therefore, predicts the development of many of the morbidities experience by these high risk patients such as adult respiratory distress syndrome (ARDS), acute lung injury (ALI), multiple organ dysfunction syndromes (MODS) and sepsis. Additionally, the degree of the fall in gelsolin levels in these patients is inversely associated with the duration of assisted ventilation, the duration of intensive care unit stay, the duration of overall hospital stay and death. Similar plasma gelsolin reductions are seen in animal models and precede lung permeability changes and inflammation. Infusion of recombinant plasma gelsolin ameliorates these effects and dramatically improves survival in these animal models.

It is known that in addition to actin, gelsolin binds bioactive inflammatory mediators including lipopolysaccharide (LPS), lysophosphatidic acid (LPA), diadenosine phosphate, Aß peptide and platelet-activating factor (PAF). It has been proposed that plasma gelsolin acts as a molecule sink that binds many of the inflammatory mediators to prevent the local inflammatory processes from spreading systemically. In severe illness such as major trauma, the plasma gelsolin drops, possibly by binding to the actin that is exposed as cells breakdown and is released into the intravascular space. This actin is bound by gelsolin and removed. The decreased levels of gelsolin and the resultant loss of its “buffering” activity may explain promotion of secondary tissue injury and its inhibition by gelsolin replacement.

In addition, treatment of mice with plasma gelsolin prevents lethal complications of endotoxin injections and significantly delays mortality in the cecal ligation-puncture model.

CBC is assessing the utility of measuring of gelsolin to predict poor outcomes in a variety of patients who are at risk of developing serious morbidity as a result of their underlying disease. Based on the animal studies, the potential for preventing the development of sepsis and its consequences by replenishment of depressed gelsolin levels with infusion of recombinant human plasma gelsolin (rhu-pGelsolin) is being explored.

Gelsolin and plasma gelsolin biochemistry and actions

Gelsolin Proteins

Intracellular gelsolin was originally discovered in 1979 and was characterized as a cellular actin binding protein that was involved in the remodeling of cellular actin filaments. These changes to the cell’s cytoskeleton are associated with cell shape changes and movement. Gelsolin is unique among mammalian proteins in having a secreted isoform of its cellular form. The secretory gelsolin, called plasma gelsolin, circulates in human and rodent blood at concentrations of 250 ± 50 mg/L. Both cellular and plasma isoforms originate from the same gene on chromosome 9, and alternative mRNA metabolism explains their different deployments. Plasma gelsolin is a bona-fide secreted protein with a processed signal sequence. It is otherwise identical to the cellular form with the exception of a 23 amino acid stretch at the molecule’s amino terminus, designated the “plasma extension.” Therefore the plasma form of gelsolin is slightly larger (84 kDa) than the cellular variant (82kDa). Cellular gelsolin has a six-fold sequence repeat structure highly conserved among gelsolins of vertebrate species. The plasma extension, however, is more variable between different organisms. Many tissues secrete plasma gelsolin. Since muscle is the bulkiest human tissue, it is the major source of plasma gelsolin.

Gelsolin-Actin Interaction

Gelsolin binds both monomeric and filamentous actin, although it prefers the latter. This binding requires micromolar calcium concentrations and is very tight, with a nanomolar dissociation constant. When gelsolin binds actin filaments, it ruptures them at the binding site by breaking the noncovalent bonds holding actin monomers together in the polymer. Following this severing reaction, gelsolin remains tightly bound to one end of the polarized actin filament, the end conventionally defined as “barbed,” the end that rapidly exchanges monomers.

Gc globulin: Interaction with Gelsolin.

A second extracellular actin-binding protein is Gc globulin (also known as the plasma vitamin D-binding protein). This polymorphic glycoprotein is immunogenic in humans and only binds monomeric actin as 1:1 complexes. It is a member of the albumin superfamily, and its site of synthesis is the liver. Plasma gelsolin and Gc cooperate to depolymerize actin efficiently. Gelsolin rapidly shortens actin filaments by severing, generating ends from which actin monomers dissociate. These monomers then tightly bind Gc and therefore cannot repolymerize. Limited evidence indicates that Gc and not gelsolin actually clears circulating actin by delivering actin monomers to hepatic macrophages.

Plasma gelsolin

Cellular gelsolin binds phosphoinositides, and this binding relates to the regulation of intracellular actin remodeling. When subsequent work identified that gelsolin also binds lysophosphatidic acid (LPA), a potent extracellular agonist, a new hypothesis arose concerning the role of plasma gelsolin in inflammatory homeostasis. In local injury, activated platelets and leukocytes generate LPA. Actin exposed by cell damage soaks up plasma gelsolin, and its local depletion allows the LPA to exert its effects on defense and repair. In catastrophic injury, more drastic gelsolin depletion permits LPA and possibly other mediators to impact distant organs, especially the lung. LPA infusions markedly increase the permeability of rat lungs, and concomitant infusion of human recombinant plasma gelsolin blocks this effect (HL Yin, University of Texas, Southwestern, unpublished results). LPA may not be the only protective target of plasma gelsolin. Additional evidence indicates that gelsolin binds aß peptide, the pathogentic agent causing Alzheimer’s disease and to platelet-activating factor (PAF) and to diadenosine polyphosphates, mediators implicated in inflammation. In the case of PAF, gelsolin strongly and selectively inhibits PAF-induced platelet activation.

Development of recombinant human plasma gelsolin (rhu-pGlesolin)

In 1993 it was shown that plasma gelsolin diminished the viscosity of thick airway secretions from patients with cystic fibrosis presumably because actin filaments contribute to this viscosity (3). Since Genentech had successfully launched an inhaled protein mucolytic, DNAse I (Pulmozyme), Biogen in-licensed plasma gelsolin technology. By 1997, Biogen was able to produce recombinant plasma gelsolin in E. coli. It was found that E. coli does not properly pair sulfhydryl groups during secretion but that treatment of the secreted product with oxidized glutathione leads to the correct orientation (4). Instillation of the gelsolin preparation into rodents and Cynomologous monkeys, both intravenously and by inhalation, revealed no toxicities. Human volunteers receiving recombinant plasma in a formulation containing Tween by inhalation had no toxic reactions. Biogen then undertook a small (16-patient) phase II dose-escalation trial of inhaled gelsolin in cystic fibrosis in Canada (where Pulmozyme was not approved). Although two patients receiving the highest doses showed improvements in airway resistance (FEV1), and the phase II trial was too small to conclude with confidence that gelsolin is ineffective as a mucolytic in cystic fibrosis, the overall result was considered insufficient to warrant further development. As a result, the cell lines and intellectual property reverted back to Brigham and Women’s Hospital who licensed them to CBC.

Gelsolin in patients

Measurements of plasma gelsolin in patients have consistently shown dramatic falls during significant inflammation resulting from diverse sources. In 25 patients with acute respiratory distress syndrome (ARDS), plasma gelsolin were depressed in all patients and the levels were about a third of that seen in normal controls (5). Interestingly, 18 of the 19 patients who were tested had detectable circulating actin, while no control serum had actin in their blood. This study also found that gelsolin was depressed to approximately 50% of the normal levels in 6 patients with bacterial pneumonia. Decreased plasma gelsolin levels were measured in 18 Nigerian children and 11 Gambian patients with Plasmodium falciparum malaria and these levels increased after treatment (14). In 23 patients with trauma, gelsolin levels were approximately 25% that seen in normal controls (6).

In many diseases, not only is gelsolin depleted, the degree of depletion inversely is associated with the severity of disease. Gelsolin levels were measured in patients with acute liver failure (n = 18), chronic hepatitis (n = 17), cirrhosis (n = 17), pancreatitis (n = 10), acute myocardial infarction (n = 10), myonecrosis (n = 12) and septic shock and compared to 25 healthy volunteers (2). Significant reductions in gelsolin compared to normal controls were seen in patients with acute liver failure, myocardial infarction, sepsis and myonecrosis and there was an inverse correlation between the gelsolin concentration and the severity of illness measured by the magnitude of disease specific serum enzymes.

Similarly, in 65 patients with severe trauma, the admission gelsolin levels were depressed and the gelsolin level was inversely correlated with the duration of mechanical ventilation the duration of ICU stay and the total hospital stay (7). Gelsolin levels less than 2 standard deviations below the mean of the control group (low gelsolin levels) were associated with prolonged mechanical ventilation and a stay in the intensive care unit >13 days. Six of the 10 patients who developed ARDS had low admission gelsolin levels compared with only 7 of the 55 patients without ARDS (p =0.0028). The mean gelsolin levels were 193 and 400 mg/L in patients with and without ARDS, respectively (p = 0.0001) and 398 mg/L in survivors versus 259 mg/L for patients who expired (p =0.0001). Ten of the 13 patients (77%) with gelsolin levels at the time of admission more than 2 SD below the control mean had “bad outcomes,” defined as mechanical ventilation for >13 days in the Trauma Intensive Unit, ARDS, and/or death.

Plasma gelsolin levels were measured weekly from 24 patients undergoing allogeneic hematopoietic stem cell transplantation (8). Mean gelsolin levels in the 9 patients with rapidly fatal idiopathic pneumonia syndrome (IPS) were significantly lower than those in patients without this complication by week 3 after transplantation (101 ± 61 mg/L versus 221 ± 54 mg/L; P = 0.0002). Seven (88% of the 8 patients with gelsolin levels <100 mg/L in the first month after transplantation died from IPS within 3 months; conversely, gelsolin levels fell to <100 mg/L in 7 (78%) of the 9 patients who died from IPS within 3 months of HSCT (P = 0.0007). These findings suggest that gelsolin levels shortly after allogeneic hematopoietic stem cell transplantation can predict the later development of fatal IPS.

Plasma gelsolin concentrations were assessed serially for 5 days in 31 patients admitted to the Surgical ICU. Low plasma gelsolin levels were associated with increased risk of death occurring in the ICU. Plasma gelsolin levels lower than 61 mg/L predicted longer ICU stay, prolonged ventilator dependence and increased overall in-hospital mortality (9).

Gelsolin in animal models

Gelsolin knockout mice have significantly higher airway protein concentrations than littermate heterozygote controls, consistent with increased permeability of the pulmonary vasculature in animals lacking gelsolin (13). Knockout animals also accumulated much more airway protein than control animals following ischemic injury. These results are complicated by the fact that the knockout mice lack both cytoplasmic and plasma gelsolins, but the findings are consistent with plasma gelsolin playing a protective role to prevent proteinaceous fluid leakiness in the lung.

Gelsolin levels were measured in three mouse models of pulmonary oxidant injury- immunotargeting of pulmonary endothelium with an H2O2-generating enzyme, continuous exposure to hyperoxia (FiO2 >0.95) and high dose thoracic X-ray radiation. The degree of lung injury was inversely related to the mice treated with glucose oxidase-conjugated antibodies against platelet endothelial cell adhesion molecule-1 (p <0.0001). By 60-72 hours of hyperoxic exposure, gelsolin levels had precipitously dropped in all mice with major lung damage (p < 0.0001) and there was a inverse relationship between the gelsolin level and the degree of injury (r = -0.72; 95% confidence interval: -0.81 to -0.59). Gelsolin levels had modest but progressive fail in the irradiated mice over the first 3 days (p = 0.012) despite the development of only microscopic lung damage (10). Intravenous infusion of human recombinant plasma gelsolin, at concentrations calculated to return the levels to normal, only slightly raised the gelsolin concentrations, although human gelsolin was detectable in the mouse blood with species-specific anti-gelsolin antibodies. Nevertheless, this modest rise in plasma gelsolin correlated with significant decreases in airway neutrophil and protein contents in the treated animals compared to bovine serum albumin-treated controls (11).

In another study, Sprague-Dawley rats were randomized to undergo a 40% body surface area thermal injury (Burn) or manipulation without burn (Sham) (12). Plasma gelsolin and Gc-globulin concentrations were determined at various times during the first 6 days of injury. Other animals were randomized to receive either recombinant human gelsolin (0.078, 0.78, or 7.8 mg) or albumin (7.8 mg) before and 8 h after Burn or Sham. Twenty-four hours later, pulmonary microvascular permeability was assessed by measuring the capillary filtration by use of an isolated, perfused lung model. Plasma gelsolin levels of burn-injured rats decreased to 10% of normal levels within 12 h and remained below normal levels for up to 6 days post-injury. Gc-globulin values also fall, but to a lesser extent and only transiently. Treatment of burned animals with intravenous infusions of recombinant human gelsolin prevented the increase in pulmonary microvascular permeability that accompanies this injury. These findings are consistent with the hypothesis that plasma gelsolin depletion contributes to the pathophysiology of pulmonary microvascular dysfunction during inflammation.

The effect of rhu-pGelsolin in sepsis was assessed in two mouse models – cecal ligation and puncture (CLP) and endotoxin challenge (15). In both models, plasma gelsolin fell to 25–50% of normal and was associated with the presence of circulating actin within 6 hrs of septic challenge. Repletion of plasma gelsolin leads to solubilization of circulating actin aggregates. There was a significantly reduction in mortality in the endotoxemic mice (survival rates were 88% in the gelsolin group vs. 0% in the saline group, p < 0.001) and in CLP-challenged mice (survival rates were 30% in the gelsolin group vs. 0% in the saline group, p = 0.001). Plasma gelsolin repletion also shifted the cytokine profile of endotoxemic mice toward anti-inflammatory (plasma interleukin-10 levels were 205 ± 108 pg/mL in the gelsolin group vs. 39 ± 29 pg/mL in the saline group, p = 0.02).

References