Andrzej Mackiewicz

Glycoforms of serum α1-acid glycoprotein as markers of inflammation and cancer

Abstractα1-acid glycoprotein (AGP) is a serum acute phase glycoprotein which possesses five N-linked complex type heteroglycan side chains which may be present as bi-, tri- and tetraantennary structures. Depending upon the content of biantennary structure on AGP, up to four glycoforms of AGP are present in serum. These glycoforms can be easily estimated in body fluids by means of crossed affinity-immunoelectrophoresis (CAIE) with the lectin, Concanavalin A (Con A). Con A selectively binds biantennary structures; the more biantennary structures on AGP, the stronger the binding. In acute inflammation, a relative increase of AGP glycoforms with biantennary units is observed - a type I glycosylation change. In some chronic inflammatory states there is an relative decrease of AGP glycoforms with biantennary heteroglycans — a type II glycosylation change. Moreover, in certain other states such as pregnancy, estrogen administration or liver damage, type II glycosylation changes are also seen. A detailed analysis of the clinical applications of the assessment of AGP glycoforms in sera of patients with rheumatic diseases, AIDS and various types of cancers is presented. Accumulated data shows that AGP glycoforms may be very useful in the detection of intercurrent infections in the course of rheumatoid arthritis, systemic lupus erythematosus, or myeloblastic leukaemia, and in the detection of secondary infections in human immunodeficiency virus infected individuals. AGP glycoforms are also very useful in differentiation between various forms of trophoblastic disease and are helpful in monitoring the treatment of these patients. Finally, AGP glycoforms provide valuable information for differentiation between primary and secondary liver cancer.

Microheterogeneity forms of alpha1-acid glycoprotein as indicators of rheumatoid arthritis activity

The microheterogeneity of alpha 1-acid glycoprotein (AGP) has been studied in the sera of 48 patients with rheumatoid arthritis and of 12 healthy individuals. For each rheumatoid patient the disease activity has been assessed and each patient has been assigned to one of four activity grades: I, inactive; II, mildly active; III, moderately active; and IV, severe. Concanavalin A-affinity electrophoresis revealed three microheterogeneity variants of AGP: non-reactive with Con A, weakly reactive with Con A and strongly reactive with Con A. The relative amounts of AGP-variants observed in the healthy donors were similar to those observed in the patients with activity grade I, but differed significantly from patients with grades II, III and IV. The differences between the AGP-patterns of patients with activity grades II, III and IV were also statistically significant.

Changes in glycosylation of acute-phase proteins in health and disease: Occurrence, regulation and function

The pathophysiological variations in different glycoforms of acute-phase glycoproteins in serum most likely result from changes in the glycosylation process during their biosynthesis in the parenchymal cells of the liver. Biosynthesis in other cells or tissues may contribute, but in general appears to play a minor role. Inflammatory cytokines appear to regulate the process, but glycosylation changes are independent of protein synthesis. In addition, other humoral factors such as corticosteroids and growth factors are involved. The interplay of these factors is determined by the stage of the disease (e.g rheumatoid arthritis), the physiological situation (e.g. pregnancy), or directly or indirectly by extraneous factors such as drugs (e.g. ethanol). Information about the functional implications of the changes is limited, but some reports suggest that for α1-acid glycoprotein the changes might affect the operation of the immune system.

Interleukin‐6‐Type Cytokine‐induced Changes in Acute Phase Protein Glycosylationaa

The plasma levels and the glycosylation of acute-phase proteins (APP) are subject to marked changes during acute and chronic inflammation. The pathophysiological variations in different glycoforms of APP in serum most likely result from changes in the glycosylation process during their biosynthesis in the parenchymal cells of the liver. This is suggested from in vitro studies with isolated hepatocytes and hepatoma cell lines. Inflammatory cytokines appear to regulate the changes in glycosylation independent from the rate of synthesis of the APP. In addition, other humoral factors like corticosteroids and growth factors are involved. The interplay of these factors is determined by the stage of the disease (as in rheumatoid arthritis) or the physiological situation (as in pregnancy). The changes in glycosylation of specific APP might affect the operation of the immune system.

Severe rheumatoid arthritis prohibits the pregnancy-induced decrease in alpha3-fucosylation of alpha1-acid glycoprotein.

Patients suffering from rheumatoid arthritis (RA) may experience a temporary reduction of disease symptoms during pregnancy. As indicated by the occurrence of RA-disease symptoms during pregnancy, three categories of patients were defined, namely, remission, relapse and unchanged. In all three categories changes in the plasma level and glycosylation of α1-acid glycoprotein (AGP) were determined longitudinally in comparison to those occurring in pregnancy of healthy women. In healthy pregnancy, we observed: (i) a peak in the plasma concentration at week 18 and a minimum at week 30; (ii) a continuous increase in the degree of branching of the glycans during the entire pregnancy period, and (iii) a decrease in the degree of α3-fucosylation of AGP-glycans with a minimum occurring at week 25. Comparable pregnancy-induced changes in glycosylation were found for two other acute-phase proteins α1-protease inhibitor (PI) and α1-antichymotrypsin (ACT). Increased oestrogen levels, known to occur during pregnancy, may be one of the factors that induce these changes, because the increased branching and decreased α3-fucosylation is in agreement with our earlier findings regarding an involvement of this hormone in the regulation of acute phase protein glycosylation in oestrogen-treated males as well as females. In all three clinical categories in RA, pregnancy also induced a continuous increase in the degree of branching of the glycans of AGP. However, similar changes in concentration and fucosylation were only found during remission of the disease symptoms. In the relapse and unchanged categories in RA, the degree of fucosylation and the plasma concentration of AGP remained constant throughout pregnancy. This indicates a relationship between changes in α3-fucosylation of AGP and RA disease activity.

Determination of lectin-sugar dissociation constants by agarose affinity electrophoresis

Agarose crossed affinity electrophoresis (aff-EP) was employed for the determination of lectin-sugar dissociation constants (Ki). In the first dimension of the aff-EP increasing amounts of sugar (alpha-methyl-D-mannoside) were added to a given concentration of lectin (concanavalin A). Then the electrophoresis was run with alpha 1-acid glycoprotein, alpha 1-antitrypsin and alpha-fetoprotein as markers of lectin-sugar interactions. Mathematical equations for determination of the mechanisms and constants of lectin-sugar-glycoprotein interactions were developed. The mean value of the concanavalin A-alpha-methyl-D-mannoside dissociation constant calculated according to the introduced equations was 0.28 mM. In this system it was also possible to determine lectin-glycoprotein dissociation constants (K). The observed influence of the sugar on lectin-glycoprotein binding might be due to hydrophobic interactions since the addition of nonionic detergent caused reversal of this phenomenon.

Soluble human interleukin-6-receptor modulates interleukin-6-dependent N-glycosylation of α1-protease inhibitor secreted by HepG2 cells

Interleukin‐6 (IL‐6) induces changes in gene expression and the N‐glycosylation pattern of acute‐phase proteins in hepatocytes. IL‐6 exerts its action via a cell surface receptor complex consisting of an 80 kDa IL‐6 binding protein (gp80) and a 130 kDa glycoprotein (gp130) involved in signal transduction. A genetically engineered gp80‐derived soluble human IL‐6‐receptor (shIL‐6‐R) significantly enhanced the IL‐6 effect on N‐glycosylation changes (revealed by reactivity with the lectin—concanavalin A) of a1‐protease inhibitor (PI) secreted by human hepatoma cells (HepG2). Stable transfection of IL‐6‐cDNA into HepG2 cells (HepG2‐IL‐6) resulting in constitutive secretion of 2 μg of IL‐6 per 106 cells in 24 h led to a down‐regulation of surface‐bound gp80 and subsequent homologous desensitization or HepG2‐IL‐6 cells towards IL‐6. Soluble human IL‐6‐R functionally substituted membrane‐bound gp80 resulting in a reconstitution of responsiveness of HepG2‐IL‐6 cells.

Transforming growth factor Β1 influences glycosylation of α1-protease inhibitor in human hepatoma cell lines

We have previously shown that changes in acute-phase protein glycosylation result from alterations occurring within hepatocytes as a result of regulation by cytokines, that the glycosylation patterns of proteins secreted by Hep 3B and Hep G2 cells respond differently to the crude mixtures of cytokines found in conditioned medium from LPS-stimulated monocytes, and that interleukin-6 (IL-6) causes increased concanavalin A (Con A) binding ofαl protease inhibitor in Hep 3B cells and decreased Con A binding of this protein in Hep G2 cells. In the present study we found that transforming growth factorΒl (TGF-Β), like IL-6, led to secretion of forms ofαl-protease inhibitor with increased Con A binding in Hep 3B cells, and that IL-6 and TGF-Β in combination were additive. In contrast, in Hep G2 cells, TGF-Β had an effect opposite to that produced by IL-6, leading to secretion of forms ofαl-protease inhibitor with increased Con A binding. When employed in combination with IL-6, TGF-Β abolished the effect of that cytokine. These studies indicate that TGF-Β influences glycosylation of al-protease inhibitor in two human hepatoma cell lines in a manner that can be differentiated from that of IL-6. The identification of TGF-Β as a second defined cytokine capable of influencing glycoprotein glycosylation and the demonstration that the effect of one cytokine can be modulated by another cytokine support the view that changes in glycosylation of plasma proteins are mediated by combinations of cytokines.

C-reactive protein and α1-acid glycoprotein in monitoring of patients with acute arterial occlusion

C-reactive protein (CRP) and alpha 1-acid glycoprotein (AGP) levels were studied in sera of 75 patients with acute arterial occlusion. Depending on the degree of ischaemia, patients were divided into two groups: grade I--26 patients--and grade II--49 patients. All patients were treated surgically; 42 embolectomies, 17 endarterectomies and 16 bypasses were performed. After surgery in 19 patients various complications were observed. The concentration of both proteins at the time of admission was higher in the serum of patients with grade II than with grade I ischaemia. Similarly the concentration of both proteins was significantly higher in the sera of the patients admitted after 8 h than in patients admitted within 8 h of the onset of ischaemia. CRP and AGP levels were significantly higher in the serum of patients with ischaemia of the lower limb than in those with ischaemia of the upper limb. In all patients 2-3 days after surgery a significant increase in serum CRP and AGP was observed. In uncomplicated cases on days 7-10 the values of both proteins decreased below the level observed at the time of admission. However, in patients who experienced postoperative complications high levels of both serum proteins (especially CRP) were found on days 7-10. Complications were detected with a sensitivity and specificity of 84 and 95%, respectively, using a CRP level of 49 mg l-1 as the cut-off point.

Role of IL-6 in Acute Phase Protein Glycosylationa

Following tissue injury or infection, the concentration of a number of plasma proteins, referred to as acute phase proteins (APP), changes substantially. In addition to these quantitative changes, qualitative alterations in the carbohydrate moieties of APP are often seen in certain disease states. Differences in the number of branches of a glycoprotein’s heteroglycan antennary structures are designated major microheterogeneity and may be determined by lectin (Concanavalin A = [Con A]) affinity electrophoresis (AFF-EP). Two types of major microheterogeneity in the sera of patients seem to be distinguishable: type I, observed in a number of patients with acute inflammatory states, in which increased Con A reactivity is seen; and type 11, seen thus far in several groups of patients with chronic inflammatory processes, in which decreased Con A reactivity occurs. Recently, using an in vitro system comprising two different human hepatoma cell lines (Hep 3B and Hep G2), we have demonstrated that glycosylation changes of APP occur at the biosynthetic level within hepatocytes, that they require cytokines for initiation, and that they are regulated independently of mechanisms regulating the expression of APP genes.”’ Thus, the glycosylation changes of APP may be considered an independent phenomenon in the course of illness. In the present study, we have examined the effect of E. coli-derived human recombinant interleukin-6 (IL-6) preparations (generous gift of T. Kishimoto and T. Hirano, Japan) on the glycosylation profile of three APPs [a 1-protease inhibitor (PI) and ceruloplasmin (Cp), which are two positive APPs, and alphafetoprotein (AFP), which is a negative APP] synthesized by Hep 3B and Hep G2 cells. Two positive controls were used: a conditioned medium (CM) from LPS-activated peripheral blood monocytes derived from healthy donors and a CM obtained from nonactivated monocytes of active rheumatoid arthritis patients (RA-CM). After 72 h of incubation with IL-6, CM, or RA-CM, changes in the rates of PI, Cp, and AFP synthesis by Hep 3B and Hep G2 cells were studied, as were the changes in their glycosylation profile.