Activation of the complement system during immunotherapy with recombinant IL-2. Relation to the development of side effects

Therapy with high doses of rIL-2 is complicated by the occurrence of hypotensive reactions and the development of a vascular leakage syndrome (VLS). In four patients, who together received seven cycles of high doses of IL-2 (up to 12 x 10(6) U per m2 per day), and who developed these side effects, we observed an unexpected increase in plasma levels of C3a, indicating activation of the complement system. C3a levels markedly increased during IL-2 therapy from 4 nmol/liter (mean level) before therapy to 23 nmol/liter at the end of the cycle. Activation of C3 occurred via the classical pathway inasmuch as C4a levels also increased during therapy. Mean daily C3a levels correlated with signs of the VLS, such as daily weight gain (p less than 0.001) and albumin levels (inverse correlation, p less than 0.001). In five additional patients, who together received seven cycles of lower doses of IL-2 (2 x 10(6) U per m2 per day) and who did not develop a VLS, only moderate increases in C3a levels (up to 13 nmol/liter) were observed. The highest levels at the first day of the regimen (mean: 7 nmol/liter) occurred 8 h after the IL-2 infusion. Thus, administration of IL-2 induces a dose-dependent activation of the complement system in vivo, which appeared to be related to the development of side effects of this therapy, such as the VLS.     

Lack of C3 affects Th2 response development and the sequelae of chemotherapy in schistosomiasis

The role of the third component of complement (C3) during schistosome infection was investigated using mice deficient in C3. While no effect was observed 8 wk after infection on worm development or liver pathology, Ag-specific Th2-associated cytokine production (IL-13, IL-5, IL-6, and IL-10) was significantly reduced, and IFN-gamma production was enhanced in the absence of C3. IgG1 and IgE, but not IgG2a or IgM, Ab responses were also significantly impaired in infected C3(-/-) mice, suggesting that C3 may play a role in IL-4-mediated Th2 response enhancement during schistosome infection. Furthermore, C3-deficient mice could not effectively clear adult worms after praziquantel (PZQ) treatment and suffered increased morbidity due to the overproduction of proinflammatory mediators following drug administration. However, the ischemic liver damage that normally accompanies PZQ administration in infected wild-type mice was substantially reduced in treated C3-deficient mice, probably due to the absence of dead or dying worms in the livers of these animals. Together these results indicate that C3 enhances Th2 responses during schistosome infection, potentiates PZQ-mediated parasite clearance, and reduces chemotherapy-induced proinflammatory mediator production.     

Effect of complement activation with cobra venom factor on pulmonary vascular permeability

We examined the effect of acute complement activation on lung vascular permeability to proteins in awake sheep prepared with lung lymph fistulas. Complement was activated by cobra venom factor (CVF) infusion (400 U/kg for 1 h iv). Studies were made in two groups of sheep: 1) infusion of CVF containing the endogenous phospholipase A2 (PLA2) (n = 6); and 2) infusion of CVF pretreated with bromophenacyl bromide to inhibit PLA2 activity (n = 5). Intravascular complement activation transiently increased mean pulmonary arterial pressure (Ppa) and pulmonary vascular resistance (PVR) in both groups. Pulmonary lymph flow (Qlym) and lymph protein clearance (Qlym X lymph-to-plasma protein concentration ratio) were also transiently increased in both groups. Pulmonary vascular permeability to proteins was assessed by raising left atrial pressure and determining the lymph-to-plasma protein concentration ratio (L/P) at maximal Qlym. In both groups the L/P at maximal Qlym was not different from normal. In a separate group (n = 4), CVF-induced complement activation was associated with 111In-oxine granulocyte sequestration in the lungs. In vitro plasma from CVF-treated animals aggregated neutrophils but did not stimulate neutrophils to produce superoxide anion generation. Therefore, CVF-induced complement activation results in pulmonary neutrophil sequestration and in increases in PVR and lymph protein clearance. The increase in lymph protein clearance is due to increased pulmonary microvascular pressure and not increased vascular permeability to proteins.     

Human colonic epithelial cells detect and respond to C5a via apically expressed C5aR through the ERK pathway

Intestinal epithelial cells (IECs) exhibit numerous adaptations to maintain barrier function as well as play sentinel roles by expressing receptors for microbial products and antimicrobial peptides. The complement system is another important innate sensing and defense mechanism of the host against bacteria and increasing evidence shows that complement plays a role in colitis. The split component C5a is a potent proinflammatory molecule, and the C5a receptor (C5aR) CD88 has been reported on multiple cell types. Here, we examined the question of whether human colonic cell lines can detect activated complement via C5aR and what signaling pathway is critical in the subsequent responses. T84, HT29, and Caco2 cell lines all possessed mRNA and protein for C5aR and the decoy receptor C5L2. Polarized cells expressed the proteins on the apical cell membrane. C5a binding to the C5aR on human IECs activates the ERK pathway, which proved critical for a subsequent upregulation of IL-8 mRNA, increased permeability of monolayers, and enhanced proliferation of the cells. The fact that human IECs are capable of detecting complement activation in the lumen via this anaphylatoxin receptor highlights the potential for IECs to detect pathogens indirectly through complement activation and be primed to amplify the host response through heightened inflammatory mediator expression to further recruit immune cells.     

A comparison of the effects of aspirin and histamine of fluid balance in the recruited lung

Salicylate administration has been reported to increase the flow of protein-rich lymph from the lungs of animals, however, the mechanism of this response is unclear. In the present study we measured pulmonary hemodynamics and lung fluid and protein flux in anesthetized sheep, surgically prepared for the collection of lung lymph, in order to examine the possible effect of aspirin (ASP) on lung vascular permeability. ASP was given during recruitment of pulmonary microvascular surface area induced by sustained elevation of left atrial pressure (Pla) (Group 1) or continuous infusion of adenosine triphosphate (ATP) (Group 2). We compared the results of ASP administration to those found in similarly prepared animals given histamine (H) during like periods of increased Pla (Group 3) or ATP infusion (Group 4). ASP administration resulted in increased lymphatic protein clearance (Cp) in both Groups 1 and 2. In Group 1, following the characteristic increase in lung lymph flow (Q1) and fall in the ratio of lung lymph to plasma protein concentration (L/P) produced by Pla elevation, ASP administration resulted in a further increase in Q1 and a significant increase in L/P. The results found in ASP animals are qualitatively similar to those observed in Groups 3 and 4 after H. While we cannot specifically rule out a hemodynamic effect of the drug, our results suggest the increased protein flux observed following ASP administration was mediated at least in part through an increase in lung microvascular permeability.     

C1s-induced vascular permeability in C2-deficient guinea pigs

Normal guinea pigs that have been intradermally injected with C1s exhibit increased vascular permeability at the injection site. Guinea pigs that are genetically deficient in complement component C2 do not exhibit increased vascular permeability when given a similar injection. The C2-deficient guinea pigs respond normally to injections of bradykinin and kallikrein, suggesting that these animals can respond to kinins and have a normal kininogen pathway. When the C2-deficient guinea pigs are given guinea pig C2 before C1s injection, increased vascular permeability is observed. These results demonstrate a definite requirement for complement component C2 in the generation of C1s-induced vascular permeability.     

A comparison of the effects of aspirin and histamine on fluid balance in the recruited lung

Salicylate administration has been reported to increase the flow of protein-rich lymph from lungs of animals, however, the mechanism of this response is unclear. In the present study we measured pulmonary hemodynamics and lung fluid and lung fluid and protein flux in anesthetized sheep, surgically prepared for the collection of lung lymph, in order to examine the possible effect of aspirin (ASP) on lung vascular permeability. ASP was given during recruitment of pulmonary microvascular surface area induced by sustained elevation of left atrial pressure (Pla) (Group 1) or continuous infusion of adenosine triphosphate (ATP) (Group 2). We compared the results of ASP administration to those found in similarly prepared animals given histamine (H) during like periods of increased Pla (Group 3) or ATP infusion (Group 4). ASP administration resulted in increased lymphatic protein clearance (Cp) in both Groups 1 and 2. In Group 1, following the characteristic increase in lung lymph flow (Q1) and fall in the ratio of lung lymph to plasma protein concentration (L/P) produced by Pla elevation, ASP administration resulted in a further increase in Q1 and a significant increase in L/P. The results found in ASP animals are qualitatively similar to those observed in Groups 3 and 4 after H. While we cannot specifically rule out a hemodynamic effect of the drug, our results suggest the increased protein flux observed following ASP administration was mediated at least in part through and increase in lung microvascular permeability.     

Clinical trials of intrasplenic arterial infusion of interleukin-2 (IS-IL-2) to patients with advanced cancer

We tried a infusion of interleukin-2 (IL-2) of a relatively low dose via an intrasplenic arterial catheter connected to a chronometric infusion (IS-IL-2). Eighteen patients of colorectal cancer with metastases to the liver or lung or of unresectable hepatoma received a 24 hour continuous infusion with low dose recombinant of IL-2 (mainly 8 × 10⁵ JRU/day) for 25–40 days. All patients tolerated this protocol of the therapy and the main toxic effects were fever and general fatigue. Such serious toxicity as previously reported by high dose IL-2 therapy was not observed. Data of hepatic and renal functions were normal. IS-IL-2 therapy induced a high incidence of eosinophilia (12/18) and thrombocythemia (12/18). Peripheral natural killer (NK) and LAK activities were augmented in all patients and total white blood cell counts were increased during IS-IL-2 therapy. An increase in IL-2 receptor expression of peripheral blood mononuclear cells and significant rises in numbers of Leull (CD16)+, OKMl(CD11)+ and OKIal(HLA-DR)+ were observed. Of 18 patients 12 were evaluable for their response to therapy. Partial response (PR) was observed in one unresectable hepatoma and 11 demonstrated no change (NC) or progressive disease (PD). Six patients were not evaluable because of additional therapy (3 cases) or decreasing tumor cell markers having no measurable lesions (3 cases). Three patients of colorectal cancer from an unresectable group were presumed to have micrometastases to the liver as suggested by an elevated serum CEA level. After receiving IS-IL-2 therapy they demonstrated a decrease in the serum CEA level for more than 3 years after treatment. We conclude that continuous IS-IL-2 administration can result in an increase of their therapeutic efficacy of IL-2 administration and in a decrease its toxicity.     

Human recombinant interleukin 2-activated sheep lymphocytes lyse sheep pulmonary microvascular endothelial cells

Administration of lymphokine-activated killer (LAK) cells in combination with interleukin 2 (IL-2) has been effective in reducing tumor mass in humans, but has been accompanied by significant toxicity. We used a chronic awake sheep model to investigate the cause of the vascular leak syndrome associated with IL-2 administration. Sheep repeatedly infused with human recombinant IL-2 (hrIL-2) developed mild pulmonary hypertension, systemic hypotension, acidemia, hypoxemia, and increased flow of protein rich lung lymph. We hypothesized that LAK cells may damage lung endothelium in vivo and cause increased lung vascular permeability. Sheep peripheral blood and lung lymph lymphocytes incubated in vitro with hrIL-2 generated cytotoxic activity for human K-562 cells and sheep pulmonary microvascular endothelial cells. In addition, cytotoxic effector cells were isolated from the peripheral blood of a sheep which had received hrIL-2. These observations suggest that LAK cells possess the ability to damage endothelial cells and may contribute to an increased pulmonary vascular permeability observed following hrIL-2 infusion in sheep.     

Morphological analyses of interleukin-8 effects on rat ovarian follicles at ovulation and luteinization in vivo

The aim of the present study was to elucidate functions of the interleukin (IL)-8 at ovulation and luteinization in vivo. To compare the morphological differences between human chorionic gonadotropin (hCG) and IL-8 stimulation, scanning electron microscopy was employed to study rat ovarian vascular corrosion casts. Follicular growth and increased capillary vessel densities around the follicles were seen in vascular corrosion casts after IL-8 injection, similar to the result of hCG administration. This result indicated that exogenous IL-8 could play a role in the neovascularization during follicular development as an angiogenetic factor. Many fenestrations were observed in the vascular endothelium by hCG administration. In contrast, no fenestrations were observed with IL-8 injection, indicating that IL-8 may not be sufficient to increase the vascular permeability directly. Although germinal vesicle breakdown (GVBD) occurred at rates of 82% after the hCG injection, only 20% GVBD was observed after the IL-8 injection. The present study indicated that IL-8 might have important effects on rat follicles at ovulation and luteinization via vascularization in a similar manner to hCG. However, IL-8 was not effective on vascular permeability and oocyte maturation, which were different from hCG. Thus, we can conclude that IL-8 can participate in follicular development in part and may play important roles in ovulation and luteinization as one of some mediators induced by endogenous luteinizing hormone.     

Effect of complement depletion on lung fluid balance after thrombin

We examined the effect of complement depletion on lung fluid and protein exchange after thrombin-induced pulmonary thromboembolization. Sheep were prepared with lung lymph fistulas to assess pulmonary transvascular fluid and protein dynamics. Studies were made in three groups: in group I (n = 5) pulmonary thromboembolization (PT) was induced by an iv infusion of thrombin (55.0 +/- 12.9 NIH U/kg); in group II (n = 6) cobra venom factor (CVF) was given ip (94.5 +/- 18.8 U/kg/day) for 2 days to deplete complement, and then thrombin (66.4 +/- 37.0 NIH U/kg) was infused to raise pulmonary vascular resistance to the same level as in group I; in group III (n = 10) left atrial pressure (Pla) was increased by 10-15 Torr in normal animals by inflation of a Foley balloon catheter. In group I, thrombin infusion caused an increase in pulmonary lymph flow (Qlym) with a gradual increase in the lymph-to-plasma protein concentration ratio (L/P). In complement-depleted sheep, thrombin caused a transient increase in Qlym, which was associated with a decrease in L/P. In group I an increase in Pla further increased Qlym but without a change in L/P, indicating an increase in lung vascular permeability to proteins; whereas in the decomplemented-thrombin sheep raising Pla increased Qlym but decreased L/P. Results in the latter group were similar to those obtained in normal animals after left atrial hypertension (group III). Therefore the complement system participates in the increase in lung vascular permeability following thrombin-induced microembolization.     

Effects of dibutyryl cAMP on pulmonary air embolism-induced lung injury in awake sheep

To assess the role of intracellular adenosine 3',5'-cyclic monophosphate (cAMP), we tested the effects of dibutyryl cAMP (DBcAMP), an analogue of cAMP, on lung injury induced by pulmonary air embolism in awake sheep with chronic lung lymph fistula. We infused air (1.23 ml/min) in the pulmonary artery for 2 h in untreated control sheep. In DBcAMP-pretreated sheep DBcAMP was infused (1 mg/kg bolus and 0.02 mg.kg-1.min-1 constantly for 5 h); after 1 h from beginning of DBcAMP administration the air infusion was started. After the air infusion, pulmonary arterial pressure (Ppa) and lung lymph flow rate (Qlym) significantly increased in both groups. DBcAMP-pretreated sheep showed significantly lower responses in Qlym (2.7 X base line) compared with untreated control sheep (4.6 X base line); however, Ppa, left atrial pressure, and lung lymph-to-plasma protein concentration ratio were not significantly different between the two groups. Although plasma and lung lymph thromboxane B2 and 6-ketoprostaglandin F1 alpha concentrations increased significantly during the air infusion, DBcAMP-pretreated sheep showed significantly lower responses. Thus DBcAMP infusion attenuated pulmonary microvascular permeability induced by air embolism. We conclude that pulmonary vascular permeability is in part controlled by the intracellular cAMP level.     

Immunological effects of a single evening subcutaneous injection of low-dose interleukin-2 in association with the pineal hormone melatonin in advanced cancer patients.

Recent experiments have shown that a great variety of neurohormones can interact with IL-2 in the modulation of host antitumor immune response. On the basis of these data, a study was started to evaluate the effect of the pineal hormone melatonin (MLT) on IL-2-induced immune changes in cancer. The study included 30 advanced cancer patients. They were randomized to be treated with IL-2 at a dose of 3 million IU subcutaneously twice/daily (8.00 a.m. and 8.00 p.m.) for 6 days/week for 4 weeks, with IL-2 once/daily in the evening, with IL-2 once/daily plus MLT at 10 or at 50 mg/day. MLT was given orally at 8.00 p.m. Both IL-2 given twice daily and IL-2 given once daily and IL-2 given once daily in association with MTL induced a significant increase in mean number of lymphocytes, T lymphocytes, NK cells, CD25-positive cells and eosinophils, whereas the single administration of IL-2 alone was unable to determine a significant rise in the mean number of immune cells. Soluble IL-2 receptor and neopterin increase was significantly higher during IL-2 given twice/daily than during IL-2 plus MLT, while no difference was seen in TNF rise. This study would suggest that a single daily injection of low-dose IL-2 is able to efficiently activate the lymphocyte proliferation in cancer patients when it is given in association with the pineal hormone MLT.     

Absence of the complement anaphylatoxin C3a receptor suppresses Th2 effector functions in a murine model of pulmonary allergy

Asthma is a chronic inflammatory disease of the lung resulting in airway obstruction. The airway inflammation of asthma is strongly linked to Th2 lymphocytes and their cytokines, particularly IL-4, IL-5, and IL-13, which regulate airway hyperresponsiveness, eosinophil activation, mucus production, and IgE secretion. Historically, complement was not thought to contribute to the pathogenesis of asthma. However, our previous reports have demonstrated that complement contributes to bronchial hyperreactivity, recruitment of airway eosinophils, IL-4 production, and IgE responses in a mouse model of pulmonary allergy. To define the complement activation fragments that mediate these effects, we assessed the role of the complement anaphylatoxin C3a in a mouse model of pulmonary allergy by challenging C3aR-deficient mice intranasally with a mixed Ag preparation of Aspergillus fumigatus cell culture filtrate and OVA. Analysis by plethysmography after challenge revealed an attenuation in airway hyperresponsiveness in C3aR-deficient mice relative to wild-type mice. C3aR-deficient mice also had an 88% decrease in airway eosinophils and a 59% reduction in lung IL-4-producing cells. Consistent with the reduced numbers of IL-4-producing cells, C3aR-deficient mice had diminished bronchoalveolar lavage levels of the Th2 cytokines, IL-5 and IL-13. C3aR knockout mice also exhibited decreases in IgE titers as well as reduced mucus production. Collectively, these data highlight the importance of complement activation, the C3a anaphylatoxin, and its receptor during Th2 development in this experimental model and implicate these molecules as possible therapeutic targets in diseases such as asthma.     

C5a controls TLR-induced IL-10 and IL-12 production independent of phosphoinositide 3-kinase

The complement system is a classic central player in innate immunity. Most pathogens activate both complement and the toll-like receptor (TLR) pathway. Therefore, to provide a more comprehensive understanding of innate immunity, it is important to understand the crosstalk between these two systems. Mouse macrophages produce IL-12 and IL-10 in response to TLR ligands such as LPS, CpG, Poly I:C and Malp2. The TLR-induced IL-12 production was decreased, while that of IL-10 was increased by concurrent stimulation with a complement fragment C5a. Pharmacological studies have suggested that C5a regulates TLR4-induced IL-12 production in a phosphoinositide 3-kinase (PI3K)-dependent mechanism. In the present study, however, we found that the C5a-mediated changes can be observed in macrophages from mice lacking PI3K p85α or PI3K p110γ. The result indicates that the C5a action is PI3K-independent; neither class IA nor class IB PI3K subtype is involved in this regulation. The actions of C5a were sensitive to pertussis toxin and PD98059, suggesting a role of G protein-mediated activation of the Erk1/2 pathway.     

Complement-dependent acute-phase expression of C-reactive protein and serum amyloid P-component

The acute-phase response (APR) is regulated by TNF-alpha, IL-1beta, and IL-6 acting alone, in combination, or in concert with hormones. The anaphylotoxin C5a, generated during complement activation, induces in vitro the synthesis of these cytokines by leukocytes and of acute-phase proteins by HepG2 cells. However, there is no clear evidence for a role of C5a or any other complement activation product in regulation of the APR in vivo. In this study, using human C-reactive protein (CRP) transgenic mice deficient in C3 or C5, we investigated whether complement activation contributes to induction of the acute-phase proteins CRP and serum amyloid P-component (SAP). Absence of C3 or C5 resulted in decreased LPS-induced up-regulation of the CRP transgene and the mouse SAP gene. Also, LPS induced both the IL-1beta and IL-6 genes in normocomplementemic mice, but in complement-deficient mice it significantly induced only IL-6. Like LPS injection, activation of complement by cobra venom factor led to significant elevation of serum CRP and SAP in normocomplementemic mice but not in complement-deficient mice. Injection of recombinant human C5a into human CRP transgenic mice induced the IL-1beta gene and caused significant elevation of both serum CRP and SAP. However, in human CRP transgenic IL-6-deficient mice, recombinant human C5a did not induce the CRP nor the SAP gene. Based on these data, we conclude that during the APR, C5a generated as a consequence of complement activation acts in concert with IL-6 and/or IL-1beta to promote up-regulation of the CRP and SAP genes.     

Temporal desensitization of rat uteri for the decidual cell reaction is abolished by cholera toxin acting by a mechanism apparently not involving adenosine 3':5'-cyclic monophosphate

To determine if increased endometrial vascular permeability (a response which precedes decidualization) could be obtained in temporally nonsensitized uteri by treatments designed to increase endometrial adenosine 3':5'-cyclic monophosphate (cAMP) concentrations, cholera toxin (an activator of adenylate cyclase) was injected into the uterine lumen of immature rats treated to be at the equivalent of day 4, 5, or 6 of pseudopregnancy. In all experiments, the rats were pretreated with indomethacin to inhibit endogenous prostaglandin (PG) synthesis. Endometrial vascular permeability, determined using 125I-labeled bovine serum albumin, was assessed 8 h later. Cholera toxin increased endometrial vascular permeability to the same level in all groups. As determined by uterine weights 5 days after the intrauterine administration of cholera toxin or its vehicle, the toxin produced the same extent of decidualization in all groups. Cholera toxin had no detectable effect on uterine cAMP concentrations in rats sacrificed 15 min after intrauterine treatment. In contrast, intrauterine administration of PGE2 increased uterine cAMP concentrations at 15 min in all groups. These data suggest that the effects of cholera toxin and of PGE2 on endometrial vascular permeability and decidualization are not mediated by cAMP.     

Vascular hyperpermeability induced by tumor necrosis factor and its augmentation by IL-1 and IFN-gamma is inhibited by selective depletion of neutrophils with a monoclonal antibody

We investigated whether various recombinant cytokines induce vascular hyperpermeability when intradermally injected into rats. Only TNF did so. Of the other cytokines examined (IL-1, IL-2, granulocyte-CSF, IFN-alpha, IFN-beta, IFN-gamma) none had this effect. The increase in vascular permeability was dose dependent, and the peak response was observed at 90 min after TNF injection. When mixtures of TNF and various other cytokines (IL-1, IL-2, granulocyte-CSF, IFN-alpha, IFN-beta, IFN-gamma) were injected, only IL-1 and IFN-gamma augmented TNF-induced vascular hyperpermeability, the increase occurring in a dose-dependent manner. The induction of vascular hyperpermeability by TNF and its enhancement by IL-1 and IFN-gamma were inhibited by selective depletion of peripheral blood neutrophils with i.p. administration of an anti-rat neutrophil mAb, RP-3. Reconstitution of neutrophils to the depleted rats by in situ injection of these cells, restored TNF increased vascular permeability.     

C5a and C5a(desArg) enhance the susceptibility of monocyte-derived macrophages to HIV infection

Mononuclear phagocytes, which include circulating blood monocytes and differentiated tissue macrophages, are believed to play a central role in the sexual transmission of HIV infection. The ability of HIV to productively infect these cells may be influenced by action of exogenous or host-derived substances at the site of viral entry. Given the potent capacities of inflammatory mediators to stimulate anaphylatoxic and immunomodulatory functions in mucosa, the effects of complement-derived anaphylatoxins on the susceptibility of monocytes and monocyte-derived macrophages (MDM) to HIV-1 infection were examined. In our in vitro system, the susceptibility to infection was up to 40 times increased in MDM that had been exposed to C5a or C5a(desArg), but not to C3a or C3a(desArg), for 2 days before adding of virus. By contrast, the treatment with complement anaphylatoxins did not affect HIV replication in fresh monocytes. Stimulatory effect of C5a and its desArg derivative on HIV infection correlated with the increase of TNF-alpha and IL-6 secretion from MDM. All these functional effects of C5a and C5a(desArg) were reversible by treatment of cells with the mAb that functionally blocks C5aR. Taken together, these results indicate that C5a and C5a(desArg) may increase the susceptibility of MDM to HIV infection through stimulation of TNF-alpha and IL-6 secretion from these cells.