Glycobiology Advance Access originally published online on April 27, 2005
Glycobiology 2005 15(9):838-848; doi:10.1093/glycob/cwi067
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Increased sialylation and defucosylation of plasma proteins are early events in the acute phase response
Biochemistry and Cell Biology, ACTREC, Navi Mumbai 410 208, India
1 To whom correspondence should be addressed; e-mail: mchavan{at}notes.cc.sunysb.edu
2 Present address: Department of Biochemistry and Cell Biology, State University of New York at Stony Brook, Stony Brook, NY 11794
Received on January 21, 2005; revised on April 19, 2005; accepted on April 20, 2005
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Within hours of turpentine injection to stimulate the acute phase (AP) response in rats, the N-acetylneuraminic acid content of plasma proteins increases and that of fucose decreases, each by about 60%. The two changes are inversely related (r 5 20.97). The NeuAc/Gal ratio increases from the normal 0.75 to 1.0 on day 2 of the AP. Whereas 50% of the isolated oligosaccharides of normal plasma proteins are retarded on immobilized Ricinus communis agglutinin, those from day 2 AP plasma fail to do so. This indicates that NeuAc caps the normally Gal-terminated chains. a1-Acid glycoprotein (a positive AP protein), a1-macroglobulin (a non-AP protein), and a1-inhibitor3 (a negative AP protein) also show similar alterations in NeuAc/Gal ratio and decreases in Fuc. a2-Macroglobulin, which arises only during the AP, does not contain significant amounts of Fuc. Sambucus nigra agglutinin (a2,6-linked NeuAc-specific) binds a majority of plasma proteins, and binding is increased during the AP response. Maackia amurensis lectin (a2,3-linked NeuAc-specific) binds only three proteins in normal plasma and three additional proteins in AP plasma. The Fuc-specific Aleuria aurantia agglutinin and Lens culinaris agglutinin each detect five proteins in normal plasma. Their binding decreases during the AP response. These results show that: (1) sialylation and defucosylation of preexisting plasma proteins occur rapidly in the AP response; (2) sialylation caps the preexisting Gal-terminating oligosaccharides; and (3) the oligosaccharides of even the non-AP and negative AP proteins are modified. These changes are distinct from the elevation in the levels of protein-bound monosaccharides and the altered concanavalin A-binding profile the oligosaccharides of AP proteins acquire in diseases.
Key words: acute phase / glycosylation / plasma proteins
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Acute phase (AP) response (Gabay and Kushner, 1999
), also called innate immunity (Medzhitov and Janeway, 1997), occurs universally in animals and offers the first line of defense (Fearon, 1997). Extensive changes in the proteins of blood plasma constitute an important aspect of this response (Raynes, 1994; Fearon, 1997). Some proteins that normally do not occur in plasma, such as the C-reactive protein and serum amyloid A protein, are synthesized during the response, whereas others occurring normally, for example 
1-acid glycoprotein, are produced in larger amounts. Such proteins are referred to as the positive AP proteins. In contrast, the concentrations of some of the normal plasma proteins (PP) remain unchanged, although those of a few others actually undergo decrease. These are, respectively, called the non-AP and negative AP reactants. These changes in the levels of the proteins persist for the duration of the response (Gabay and Kushner, 1999). Although most individual proteins behave similarly in animals of different species, some differences do exist. Thus, although both 1-acid glycoprotein and 2-macroglobulin (2-M) occur normally in human plasma and their concentrations increase during the AP response, the latter is absent in the normal plasma (NP) of rats, and appears only (1-M) following the onset of the AP reaction. 1-Macroglobulin and 1-inhibitor3 (1-I3) are examples of the non-AP and negative AP proteins of rat plasma, respectively.
Most PP are glycoproteins. Many decades ago numerous studies reported increase in their levels in a variety of diseases, estimated and expressed as one of the component sugars (Winzler, 1955). It has now become clear that a form of the proteins appears in plasma during the AP that has novel concanavalin A (ConA)-binding characteristics (Turner, 1992). This change indicates qualitative alterations in the glycosyl moiety of the proteins. The change coincides with the duration of the AP suggesting that it may have a role in defense processes of the body (Gabay and Kushner, 1999).
The inflammatory cytokines, IL-1, TNF-, and IL-6 are the immediate stimulants of AP protein synthesis in the liver (Gabay and Kushner, 1999). Injection of an inflammogen, such as the oil of turpentine in animals, sets off a reproducible and powerful AP response. In the rat it lasts for 20 days (Koj et al., 1982) and offers a convenient system for the analysis of various aspects of the response. We have studied the glycosyl moiety of PP at intervals spanning the period of the AP response. In addition to total proteins, 1-acid glycoprotein, purified from plasma at various times after the stimulation of the response, has been studied. Do the putative changes in the glycosyl moiety occur in the positive as well as in the non-AP and negative AP proteins? Because the glycosylation apparatus in the liver is common to all glycoproteins, even the proteins belonging to the latter two categories, synthesized during this period, may be modified. If this modification has a role in defense functions, the non-AP and negative AP reactants may also participate in them. For this reason, the structurally-related proteinase inhibitors 1-M, 1-I3, and 2-M have also been studied.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Changes in total plasma protein-bound sugars during the acute phase response
A typical N-linked oligosaccharide, in this instance a biantennary chain, is shown in Figure 1a. The variant positions of N-acetylneuraminic acid (NeuAc) and fucose (Fuc) are indicated by (±). Monosaccharide analysis of the delipidated total PP was carried out for days 0 (normal), 1–12, 16, and 20. The values of sugars were initially obtained as nmol/mg protein, which were recalculated as percentage of the normal (day 0) value at each time point during the AP (Figure 1b). Sialic acid showed the largest quantitative variation with nearly 60% increase in its content on day 1. Protein-bound galactose (Gal) and mannose (Man) showed similar but smaller increases (20 and 25%, respectively) compared with NeuAc, although the rise in N-acetylglucosamine (GlcNAc) was the least. Fuc is present in the least amount among the protein-bound sugars, and its concentration fell further by about 60%. In all cases, except GlcNAc, the maximum variation was seen on day 1 of the AP response, after which the values tended to decrease till day 7, but then rose again on day 8. This rise (statistically significant over the day 7 values, P < .05), however, was smaller; subsequent to which the values declined to their normal levels by day 20. Even in the case of Fuc, whose content decreases during the AP, the level rose above the normal by day 8 before returning to normal. Thus there appear to be two peaks of variation in the protein-bound monosaccharides (except for GlcNAc) during the AP response, on days 1 and 8.
|
N-acetylgalactosamine (GalNAc) was present as a minor component of the sugars (
1% of GlcNAc), indicating only a small contribution of O-linked sugars to the pool of oligosaccharides. They are excluded from any further consideration.
The values of the protein-bound sugars, expressed per mg protein, have been corrected for the decrease occurring in plasma albumin during the AP. Being a negative reactant (Gabay and Kushner, 1999), the concentration of albumin in plasma decreases. Because the sugars are estimated on dried PP taken on weight basis, the results are susceptible to variations in the albumin content. Therefore, albumin concentration was determined as described in methods on three samples of plasma on days 0, 1–12, 16, and 20 of the AP response. A correction to the protein content was applied for each day in the following way: the albumin content on day 0 was 55.7 ± 0.85% (of total protein) and was 44.0 ± 0.60%, for example, on day 2. One mg protein on day 2 would thus have weighed 1.26 (55.7/44.0) mg if there were no decrease in albumin. The sugar contents were calculated for 1.26 mg rather than for 1 mg gravimetric weight of protein, on day 2.
The increase in the content of PP-bound sugars could be because of either the addition of sugars to the preexisting oligosaccharides, or the synthesis of new chains of different composition, or both. NeuAc is usually present on the antennae of N-linked oligosaccharides as the terminal monosaccharide, with Gal as the penultimate sugar (Figure 1a). Thus in a completely sialylated chain, the number of NeuAc residues would equal the number of Gal residues. The levels of both NeuAc and Gal increase in the AP reaction, but unequally (Figure 1b). In Figure 1c, the ratio of NeuAc to Gal is plotted as a function of progress of AP response. The ratio increases from 0.76 on day 0 to 0.98 on day 1. This means that in normal PP, three NeuAc are present for every four Gal residues, but soon upon the onset of the AP, practically every Gal is covered with NeuAc.
The Fuc content of PP decreases in the AP (Figure 1b), in contrast to other constituent monosaccharides, particularly sialic acid. Fuc and NeuAc, in fact, are found to vary inversely with respect to each other (r = –0.97) over the first 7-day period. The ratio of Fuc/NeuAc is plotted as a function of progress of the AP response (Figure 1d).
It may be emphasized that the results described here represent the total, that is, the newly synthesized as well as the preexisting, protein-bound oligosaccharides.
Monosaccharide composition of 1-acid glycoprotein during the acute phase
To determine whether the changes observed in the monosaccharide content of whole PP are seen at the level of individual proteins, 1-acid glycoprotein, a major AP reactant in the rat was purified from normal (day 0) and AP plasma (days 1–3, 7, 8, and 20). The purified, homogeneous, protein preparations (Figure 2a) were subjected to monosaccharide analysis. The levels of 1-acid glycoprotein (1-AGP)-bound Gal, GlcNAc, and Man did not vary significantly during the AP response (Figure 2b). However, the content of sialic acid increased by 40–45% in the first 3 days. Subsequently, it dropped slightly on day 7, rose again on day 8 and thereafter decreased steadily to return to the normal value by day 20 (Figure 2b). Fuc forms only about 2.5% of total sugars present in 1-AGP in NP. Its content dropped to one fourth the normal value on day 1, remained there until day 3, recovered slowly, and attained the normal value by day 20. The correlation coefficient between the NeuAc and Fuc values in the first three days equaled –.995.
|
When the variation in the NeuAc/Gal ratio was examined (Figure 2c), it was found to increase from the normal 0.66 to 0.90 on day 1 of the AP response. The higher ratio was maintained until day 3. It dropped slightly on day 7, rose again to nearly 0.9 on day 8 and then returned to normal on day 20. This shows that in the NP 1-AGP, out of three Gal residues one is uncapped; but most of them get sialylated on day 1 of the AP response. The plot of NeuAc/Gal (Figure 2c) resembles that of the NeuAc content (Figure 2b). This is obviously because there is little increase in Gal. The ratio of Fuc/NeuAc dropped from the normal 0.08 to 0.01 within a day of the onset of the response, remained low for 3 days and returned to normal by day 20. The steep decline in the ratio is because of rapid alterations in opposite directions in the contents of the two monosaccharides.
Thus, total PP and purified 1-acid glycoprotein show almost identical and biphasic variations in their monosaccharide compositions.
Ricinus communis agglutinin–Sepharose column chromatography of plasma protein-bound N-linked oligosaccharides
Sialic acid, as a rule, as noted above, is present as the terminal sugar on the antennae of the complex N-linked oligosaccharides, with Gal as the penultimate sugar (Figure 1a). Ricinus communis agglutinin (RCA) interacts with the antennal Gal, and the number of exposed Gal residues determines the nature of interaction of the oligosaccharides with the lectin: two exposed Gal residues per chain lead to binding of the oligosaccharide to RCA–Sepharose; and presence of one exposed Gal results in its retardation on the column (Merkle and Cummings, 1987). Because the protein-bound glycans are rapidly sialylated in the AP (Figure 1b), their Gal residues should no longer be available for interaction with RCA. To estimate the proportion of oligosaccharides of normal and AP PP terminating in sialic acid, oligosaccharides were released from the proteins and radiolabeled. They were fractionated on RCA–Sepharose into RCA-unbound, RCA-retarded, and RCA-bound (elutable with 0.1 M lactose) oligosaccharides (Merkle and Cummings, 1987). Figure 3a shows the elution profiles of oligosaccharides from normal and day 2 AP PP from RCA–Sepharose. Negligible amounts (less than 0.5%) of oligosaccharides from either normal or AP PP were bound to the column (i. e., eluted with 0.1 M lactose), indicating that practically no oligosaccharides exist in these proteins with two or more exposed Gal residues. The percentage of total oligosaccharides retarded on RCA–Sepharose on various days after induction of AP is plotted in Figure 3b. Almost 50% of oligosaccharides from NP were retarded, indicating that they have one Gal exposed. With the induction of the AP response, on days 1 and 2, the oligosaccharides were no longer retarded on the column, indicating that all their Gal residues are sialylated. There was some retardation on day 3, which fell again to near 0 by day 7/8, returning to normal value by day 20. The biphasic nature of the alterations seen in the monosaccharide analysis (Figure 1b) is evident in these results also.
|
Identification of proteins undergoing increased sialylation and defucosylation during the acute phase response
Because the total plasma protein-bound oligosaccharides show nearly complete sialylation and decreased fucosylation in the early phase of the APR (Figure 1b and c), it is clear that most oligosaccharides undergo the modifications. However, do some proteins contribute more to these changes than others? It is also not clear whether the postturpentine sialylation is via (2Æ6) or (2Æ3) linkage. To determine this, PP on western blots were probed with a panel of sialic acid- and Fuc-specific biotinylated lectins:
- Sambucus nigra agglutinin (SNA), which recognizes (2Æ6)-linked sialic acid;
- Maackia amurensis lectin (MAL), which binds to (2Æ3)-linked sialic acid;
- Aleuria aurantia agglutinin (AAA), which binds specifically to Fuc; and
- Lens culinaris agglutinin (LCA), which interacts with the trimannosyl core of the N-linked oligosaccharide bearing a Fuc residue on the core GlcNAc.
The blots developed using biotinylated SNA indicated an abundance of (2Æ6)-linked sialic acid in normal PP (Figure 4a). With the onset of the AP response, a general increase in the intensity was seen (by densitometry), indicating gain of (2Æ6)-linked sialic acid during the APR. The changes reverted to normal by day 20 (Figure 4a). In contrast, the (2Æ3)-linked sialic acid is not common in rat PP. In the case of NP, only three proteins, Mr 72k, 67k, and 60k were found to react with biotinylated Maackia amurensis agglutinin (Figure 4b). In addition to the increased intensity of these three bands by 10, 44, and 20%, respectively (determined by densitometric scanning), 2 additional bands of Mr 52k and 30k were observed on day 2 of the AP response, which persisted until day 8. After day 8, the intensity profile of the bands reverted to normal (Figure 4b). Another fainter band of approximately Mr 50k can also be discerned on days 2–12.
|
Biotinylated AAA detected five intense bands (Mr 75k, 66k, 52k, 46k, and 42k) in normal PP (Figure 5a). All five glycoproteins exhibited drastically reduced intensity (by 87, 82, 85, 80, and 90%, respectively) on the first day of the AP response, indicating loss of Fuc. The staining pattern of the five bands, however, differed somewhat from each other after day 2, possibly indicating differing turnover of their Fuc residues compared with the overall turnover of protein-bound Fuc (Figure 1b). The concentrations of their protein moieties were, however, unchanged as evident from Coomassie blue staining and quantitation by densitometry (not shown). By day 20, the Fuc levels returned to normal. Loss of Fuc was also observed when the blot was developed using biotinylated LCA (Figure 5b). AAA and LCA seem to react with the same PP.
|
Carbohydrate analysis of a positive AP, negative AP, and non-AP protein of rat plasma during the AP response
1-Macroglobulin, 1-I3, and 2-M were purified by methods described in the literature (Esnard and Gauthier, 1980; Schaeufele and Koo, 1982). 1-M, purified from normal and day 2 AP plasma, showed two subunits of Mr 155k and 42k by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS–PAGE) (not shown), as expected. 1-I3, besides the expected Mr 180k band, revealed two additional polypeptides, an intense one of Mr 100k and another of a lower molecular weight. The two polypeptides were found to react with the antibody raised against the Mr 180k peptide excised from the SDS-polyacrylamide gel (not shown), suggesting that they are derived from 1-I3. Less intensely stained bands of Mr ranging from 100k to 180k were also observed on the Western blot. 2-M was purified from day 2 AP plasma and was found to be homogeneous (not shown).
Monosaccharide analysis of purified proteins
The purified proteins were subjected to hydrolysis for the release of neutral sugars, amino-sugars, and sialic acid under the conditions described in Methods. The results appear in Table I.
|
With the onset of the AP response, the content of 1-M-bound sialic acid increased by 58%. The Fuc content of the protein from NP, which was only about 3% of total sugar, dropped to an undetectable level in the day 2 protein. The NeuAc/Gal ratio of 0.66 in the NP 1-M shifted to 0.98 in the AP protein (Table I). The differences in the other monosaccharides were statistically insignificant.
The sialic acid content of 1-I3 from the day 2 AP plasma showed a 44% increase over the protein from NP; although the Fuc content fell below the detectable level. The content of other monosaccharides did not differ significantly. The NeuAc/Gal ratio of 0.7 in 1-I3 of NP increased to about 1.0 in the preparation from day 2 AP plasma (Table I).
2-M does not occur in normal rat plasma. The protein from day 2 AP plasma did not contain detectable amounts of Fuc (data not shown). The ratio of NeuAc/Gal in this protein is almost 1.0, indicating that all Gal residues are capped (data not shown). The changes observed in the monosaccharide composition of the three types of AP proteins emphasize the observation made with total proteins: increased sialylation and defucosylation are two prominent phenomena occurring in early phases of the AP response.
RCA–Sepharose column chromatography of N-linked oligosaccharides of purified plasma proteins
Enzymatically released, radiolabeled, N-linked oligosaccharides from the five purified proteins were fractionated on the RCA–Sepharose column to estimate the percent of total oligosaccharides terminating in sialic acid. About 75% of total oligosaccharides of 1-M from NP were retarded on RCA–Sepharose, indicating the availability of at least one exposed Gal on these structures. However, all glycans of the day 2 AP 1-M appeared in the unbound fraction, suggesting their complete sialylation (data not shown). Likewise, 75% of oligosaccharides from normal 1-I3 were retarded on RCA–Sepharose, but none from the AP protein (Table II). Thus it appears that some of the protein-bound oligosaccharide chains that lacked sialic acid prior to onset of the AP response were modified to yield fully sialylated, nonfucosylated structures.
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Acute phase reaction is a preimmune response that is concerned with several defense-related functions: opsonization (Chang et al., 2002
; Hart et al., 2004), scavenging (Du Clos, 1996; Palaniyar et al., 2002; Van Amersfoort et al., 2003), growth regulation (Huang et al., 1984; Yi and Ruoslahti, 2001; Farrell, 2004), regulation of inflammation (Streetz et al., 2001; Ceciliani et al., 2002; Hiemstra, 2002), immunomodulation (Shiyan and Bovin, 1997; Tilg et al., 1997), proteinase inhibition (Hiemstra, 2002), and so on. Chronic inflammation is believed to predispose to several pathologies, including cardiovascular diseases (Jialal et al., 2004) and type II diabetes (Duncan et al., 2003; Spranger et al., 2003). Accordingly, AP proteins have been found to offer prognostic value for several diseases (Mahmoud and Rivera, 2002; Blake and Ridker, 2003; Rattazzi et al., 2003; Hashimoto et al., 2004; Vermeire et al., 2004; Yamamura et al., 2004; Casamassima et al., 2005; Ytting et al., 2005).
In most diseases the levels of all plasma protein-bound sugars are elevated (Winzler, 1955; Gabay and Kushner, 1999), and qualitative changes have also been reported in the oligosaccharide chains of the proteins (Turner, 1992). We believe that the modifications reported here in the early phases of AP response in rats are distinct from those just indicated because they differ in several important respects: (1) The changes seen here are in response to inflammation alone, whereas in the clinical setting, inflammation (a consequence of the pathology) would cooccur with the pathogenic agent, thus consisting of two distinct conditions. (2) The changes described here arise in less than 24 h, persist for two more days or so and then begin to revert. In the clinical studies, done almost invariably on hospital-based patients (suffering from the disease for at least a few days), the early alterations cannot be seen. (3) Most importantly, in this study, substantial rise occurs almost exclusively in NeuAc, with drastic decrease in Fuc. (4) The early change affects all plasma glycoproteins, whereas in diseases only the positive AP proteins are elevated.
In view of the above, possibly the only way the studies in humans could be compared with our results, would be to look at cases of, say, burns or fractures, provided their plasma protein-bound monosaccharides were estimated soon after hospital admission. We have carefully looked through the literature to see if any studies conform to these requirements. We came across two, both pertaining to levels of 1-acid glycoprotein-bound Fuc, estimated indirectly via AAA-staining. The first study (Ryden et al., 1999) found that four of six burns patients show an initial decrease (for 1 or 2 days) in the Fuc content, which then rises. The other study (De Graaf et al., 1993) found that all the three patients of burns, as well as those subjected to laparotomy for the removal of benign uterine tumors (an AP noninducer?), showed AAL reactivity that steadily increased over the period of the study, that is, without an initial decline. In view of the apparent contradictory results, it is difficult to draw any conclusions on the alterations in the early phase of AP response in humans. It is possible that the observations described in this article apply uniformly, but species specificity cannot be ruled out at the moment.
Concerning the turpentine-induced rat PP, the normal NeuAc/Gal ratio of about 0.75 indicates that approximately 25% of the N-linked oligosaccharide chains of PP terminate in Gal. On stimulation of the APR, the ratio changes almost to 1.0, showing that nearly all Gal residues are capped. The lack of retardation of isolated oligosaccharides from AP PP on RCA–Sepharose in the early part of the response supports this conclusion. The NeuAc/Gal ratio returns to 0.75 at the end of the APR. This behavior, observed with total proteins, is replicated by the oligosaccharides isolated from 1-acid glycoprotein at different stages of the APR. The oligosaccharides from purified 1-M (a non-AP protein) and 1-I3 (a negative AP protein) and 2-M (that appears only during the AP reaction, data not shown) also display similar characteristics. The capping of Gal residues thus is a general occurrence affecting all plasma glycoproteins during the AP.
The reversion of the NeuAc/Gal ratio to 0.75 on cessation of the AP suggests that the existence of 25% of Gal-terminating oligosaccharides in normal PP is a matter of regulation and not a nonspecific escape from sialylation during their synthesis. This implies a definite role for the uncapped Gal residues in normal blood plasma. This is surprising because desialylation is known to be a determinant for the removal of glycoproteins from plasma for their catabolism in the liver (Ashwell and Morell, 1974). During the APR, large quantities of C-reactive protein, capable of interacting with asialoglycoproteins (Kottgen et al., 1992), appear rapidly in plasma. It is likely that it will bind and neutralize the glycoproteins possessing oligosaccharides that terminate in Gal. In view of this, rapid sialylation of preexisting glycoproteins in the AP is understandable. However, fucosylated oligosaccharides are not substrates for sialylation (Kobata, 1992). Defucosylation would thus be mandatory for sialylation to commence. The apparent reciprocal relationship (r = –.97) between the increase in sialic acid and decrease in Fuc contents over the 7-day period of the APR is thus explicable. Also, an N-linked oligosaccharide bearing both sialic acid and Fuc on the antennal GlcNAc will generate the sialyl Lewis X (sLex) antigen, which is a ligand for P- and E-selectins (Larsen et al., 1992). A glycoprotein bearing this antigenic structure can compete with activated leucocytes to bind to selectin-bearing endothelial cells.
Much work has appeared in the literature on the appearance of sLex antigenicity during the AP reaction in patients. 1-AGP, the principal carrier of the antigen, and 1-antichymotrypsin and haptoglobin to lesser extent (Brinkman-van der Linden et al., 1998) are believed to prevent the homing of activated neutrophils to the activated endothelial cells. This response, occurring in somewhat later stages of the disease, possibly protects from the harmful overreaction to inflammation.
We suggest three possible mechanisms for the sialylation and defucosylation of the proteins during the initial phases of AP response, and later for their reversal: (1) Sialyl transferase (with the sugar donor) and fucosidase are secreted in the plasma, and the modifications occur in situ. (2) The proteins are acted upon by plasma membrane-bound enzymes of the liver or other (e.g., endothelial) cells. (3) The proteins are endocytosed by a receptor-mediated mechanism, modified intracellularly, and secreted back into plasma. Later, after three days in the response, the changes are reversed by a similar mechanism.
Inflammatory cytokines modulate the enhanced synthesis of AP proteins in the liver, and also apparently modify the glycosylation apparatus (Lombart et al., 1980). The levels of cytokines decline with the progress of the AP reaction, and the liver reverts to its normal functions. However, certain cytokines (e.g., IL-6) have been shown to be elevated transiently around day 8 of the APR (Wu et al., 1995), providing a possible explanation for the biphasic change in protein-bound sugars. The functional significance of the second peak remains unknown.
A mechanism that sialylates and defucosylates plasma glycoproteins, and a modified glycosylation apparatus in the liver during the AP (Lombart et al., 1980), should be expected to act indiscriminately on positive, neutral, or negative AP proteins. Greater sialic acid and negligible Fuc contents of 1-M and 1-I3 during the AP bear this out. Thus, although the quantities of the proteins do not increase, or may even decrease, the non- AP and negative AP proteins do undergo modification of their glycosyl moieties.
Although sialylation and defucosylation evidently affect the entire pool of protein-bound oligosaccharides in the plasma, there are also some differences. Probes with (2Æ6)- and (2Æ3)-linkage-specific sialic acid-binding lectins show that the content of the more prevalent 2Æ6-linked neuraminic acid is increased in the AP, affecting many, perhaps most, proteins; but sialylation with 2Æ3 linkage is restricted only to six proteins. Of these, three appear to be 2Æ3 sialylated postturpentine injection.
Fucose-specific lectins, AAA and LCA, detect only five proteins, obviously those that are relatively rich in the sugar. The intensity of binding to AAA/LCA diminishes greatly with the onset of APR and reappears as the animal recovers from the response. The Fuc moieties of different specific proteins of NP seem to turnover differently compared with the overall turnover of protein-bound Fuc. The amounts of the proteins (as seen by Coomassie blue staining and densitometry) bearing the Fuc residues, however, remain unchanged, indicating that the protein and the sugar are differently metabolized.
In conclusion, the results show that within 24 h of induction of the AP response in rats, the Gal-terminating oligosaccharides of total PP are capped with sialic acid. The sialylation proceeds simultaneously with defucosylation of oligosaccharides. The changes persist for about 3 days, and then slowly return to normal in 20 days, except for another small increase on day 8 of the response. The glycan moieties of non-AP and negative AP proteins are also modified during the response. We believe that this change represents glycoprotein modification during the AP reaction that is distinct from the rise seen in the quantitative levels of protein-bound sugars and the novel Con A-binding characteristics that the oligosaccharides acquire in pathological conditions.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Materials
Sodium borotritide (NaB3H4) (specific activity 11.2 Ci/mmol) and nitrocellulose were purchased from Amersham, Bucks, England. Peptidyl N-glycosidase F, SNA, MAL, and AAA were products of Boehringer Mannheim GmbH, (Mannheim, Germany). X-ray film was obtained from Kodak, New York, USA. Polyethylene glycol was from Fluka, Switzerland. Biobeads SM2 and Biogel P2 were procured from Biorad Laboratories, Richmond, CA, USA. ConA, DEAE–Sepharose 6B, Sephadex G25, and Sepharose 4B were acquired from Pharmacia, Uppsala, Sweden. Amberlite IRC-50 and NP-40 were from BDH Chemicals, England. All other fine chemicals and materials were obtained from Sigma Chemical Co., St. Louis, MO, USA. Laboratory reagents were procured locally and were of analytical grade.
Ricinus communis agglutinin was prepared as described earlier (Nicolson et al., 1974) in the laboratory and conjugated to Sepharose 4B using a published protocol (March et al., 1974). Three milligrams of lectin was conjugated per mL of Sepharose. LCA was purified as described earlier (Howard et al., 1971).
Methods
Induction of acute phase response in rats, collection of blood, and isolation of plasma.
Oil of turpentine (1 mL/200 gm body weight) was injected subcutaneously in adult inbred Wistar rats (200–250 gm body weight, irrespective of sex), to stimulate the AP response. Blood was collected from the animals, following anesthesia with chloroform, by cardiac puncture over EDTA (1.5 mg/mL blood) as the anticoagulant. Blood was centrifuged at 5000 xg for 10 min at 4°C. The plasma was isolated carefully; and if not used immediately, was frozen at –20°C. After injection of turpentine (considered day 0), blood was collected from the animals on days 1, 2, 3, 4, 5, 6, 7, 8, 12, 16, and 20, or as indicated. To minimize animal-to-animal variations we pooled equal volumes of blood plasma from three rats to generate every sample.
Preparation of rat plasma proteins.
Plasma was delipidated with chloroform : methanol (Finne and Krusius, 1979), dissolved in minimum volume of deionized water, dialyzed, and lyophilized. The lyophilized proteins were stored at –20°C over desiccant.
Enzymatic release of N-linked oligosaccharides from plasma proteins
N-linked oligosaccharides were released from plasma glycoproteins using peptidyl N-glycosidase F (PNGase F; Jackson, 1994). Five hundred micrograms (dry weight) of the glycoprotein sample were mixed with 5 µL of the denaturing buffer containing 1% (w/v) sodium dodecyl sulfate 0.5 M ß-mercaptoethanol, and 0.1M EDTA. After 30 min at room temperature, 40 µL of 0.2 M sodium phosphate, pH 7.4 were added and heated on the boiling water bath for 5 min. After cooling, 5 µL of 7.5% NP-40 and 1 unit of PNGase F were added. Following incubation at 37°C for 18 h, the reaction was terminated by addition of three volumes of chilled absolute alcohol. After 1 h on ice, the precipitated protein was centrifuged at 10,000 xg for 5 min at the room temperature. The supernatant containing the N-linked oligosaccharides was dried in the Speed-Vac. These conditions were found to be adequate for complete N-deglycosylation of thyroglobulin, transferrin, ovalbumin, and fetuin (data not shown).
Desalting of oligosaccharides
The oligosaccharides, dissolved in a small volume of deionized water, were purified by gel filtration on a Sephadex G25 column (2.5 x 50 cm), preequilibrated in water containing 7% n-propanol (Montreuil et al., 1994). The fractions testing positive by phenol-sulfuric acid were pooled. At the concentration of 0.75%, NP-40 elutes with the oligosaccharides because of micelle formation. It was removed by shaking with 1 mL of packed Biobeads SM2 overnight. The solution was removed and dried.
Radiolabeling of oligosaccharides by reduction with sodium borotritide
The N-linked oligosaccharides (50 nmol) were radiolabeled as described earlier (Takasaki et al., 1982).
Hydrolysis of plasma proteins for monosaccharide analysis
Dried PP were weighed and transferred to acid-washed glass vials and hydrolyzed by one of the following protocols (Hardy and Townsend, 1994):
- 2 M TFA for 5 h at 100°C for the release of neutral sugars;
- 6 N HCl for 5 h at 100°C for the release of amino sugars; or
- 0.1 N HCl for 1 h at 80°C for the release of sialic acid.
After hydrolysis, the sample was evaporated to dryness under vacuum using Speed-Vac and redissolved in water.
Monosaccharide analysis and quantitation on high-performance anion-exchange chromatography with pulsed amperometric detection
Monosaccharides were separated on a Dionex high performance liquid chromatography (HPLC) system with Carbo-Pac PA1 column (4 x 250 mm) and quantitated using the pulsed amperometric detector (Lee, 1990). For the analysis of neutral and aminosugars, the column was initially equilibrated in 16 mM NaOH at the flow rate of 1 mL/min. The standard containing 300 pmol each of Fuc, galactosamine, glucosamine, Gal, mannose and glucose, or the hydrolyzate, was injected. Prior to injection, 300 pmol of inositol was added to each sample as the external standard. The column was eluted isocratically with 16 mM NaOH and was regenerated between runs using 200 mM NaOH.
For the determination of sialic acid, the column was initially equilibrated in 5 mM sodium acetate in 100 mM NaOH at the flow rate of 1 mL/min. The standard containing 750 pmol each of NeuAc and N-glycolyl neuraminic acid, or the hydrolyzate, were injected and eluted using a gradient of 50 mM–250 mM sodium acetate in 100 mM NaOH. Seven hundred and fifty pmol of inositol were added to each sample, as the external standard. To avoid baseline instability, the pH of the column effluent was increased by pumping 300 mM NaOH at the rate of 1 mL/min before the eluate entered the detector flow cell.
The separated monosaccharides were quantitated by pulsed amperometry using Dionex pulse amperometric detector. The following potential parameters and pulse times were used for PAD: E1 = 0.05 V, t1 = 360 msec; E2 = –0.5 V, t2 = 120 msec; E3 = –0.7 V, t3 = 420 msec. The time constant was set to 3 sec. The chromatographic data were integrated using a software.
External standards were used to calibrate the response factors of the monosaccharides needed for the calculation of their amounts.
Purification of 1-acid glycoprotein, 1-macroglobulin, 2-macroglobulin, and 1-inhibitor3 from rat plasma
1-Acid glycoprotein was purified from the normal and AP rat plasma as described earlier (Charlwood et al., 1976).
1-Macroglobulin and 1-I3 were purified from normal and day 2 AP plasma. 2-M does not occur in normal rat plasma. It was isolated from day 2 AP plasma. The three proteins were purified by procedures detailed earlier (Esnard and Gauthier, 1980; Schaeufele and Koo, 1982).
RCA-affinity chromatography of N-linked oligosaccharides
Chromatographic separation of oligosaccharides on RCA–Sepharose (0.8 x 40 cm) was carried out at 25°C. The [3H]-labeled oligosaccharides, dissolved in minimum volume of Tris-buffered saline (TBS), were applied to the column. The column was washed with TBS and the unbound material was collected in 10 column volumes. Two milliliter fractions were collected at the rate of 12 mL/h. The bound oligosaccharides were eluted with 5 column volumes of TBS containing 0.1 M lactose in TBS. One hundred µL of each fraction was mixed with 5 mL of the scintillation cocktail (comprising of 60 gm naphthalene, 4 gm PPO, 0.2 gm POPOP, 20 mL methanol, 20 mL ethylene glycol per liter in dioxan), and counted in the ß-scintillation counter.
SDS–PAGE and staining of the gels
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed and the protein profile on the gel was visualized by staining with either Coomassie brilliant blue (0.02% Coomassie brilliant blue in 50% methanol and 10% acetic acid for 30 min, destaining in 50% methanol) or the silver reagent (Blum et al., 1987).
Quantitation of plasma albumin content by densitometry
Delipidated rat PP (100 µg by weight) were electrophoresed on 7% SDS–PAGE and the gel was stained with Coomassie blue. The albumin band was identified by using standard albumin and other molecular weight markers. The concentration of albumin was determined by the ratio of the area due to albumin and the sum of areas of all bands in the lane following densitometric scanning. Albumin concentration was determined in plasma samples on days 0, 1–12, 16, and 20 of the AP response. Three samples of plasma were taken for estimation for each day.
Western blotting
The transfer of proteins from the polyacrylamide gel to polyvinylidene fluoride nitrocellulose membrane was carried out at 70 volts for 3 h. The blot was stained with the antiserum as described (Harlow and Lane, 1988), using polyclonal IgG raised in rabbits as the primary antibody (diluted appropriately in 0.025 M Tris–HCl, 0.5% Triton, 0.5 M NaCl, and 1.66% skimmed milk powder). Biotinylation of lectins was carried by the procedure described earlier (Bayer and Wilchek, 1990). Blots were stained with biotinylated lectins (1µg/mL) (Brakel et al., 1990) and developed using 2 mM luminol, 0.45 mM p-iodophenol, and 0.05% hydrogen peroxide (Leong and Fox, 1990). Autoradiography was performed by superimposing the X-ray film on the blot for an appropriate time. The film was placed in the developer for 1–3 min, washed in water for 2 min and then cleared in the fixer. The film was further washed under running water. All steps were carried out in the dark room with an X-ray safe light.
Phenol-sulfuric acid assay for carbohydrates
The method described earlier was used (Chaplin, 1994).
Protein estimation
Protein was estimated as described earlier (Peterson, 1977), using bovine serum albumin as the standard.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
This work was supported by a grant (SP/SO/D-15/95) from the Department of Science and Technology, Government of India.
|
Abbreviations |
|---|
AAA, Aleuria aurantia agglutinin; AP, acute phase; ConA, concanavalin A; ECL, enhanced chemiluminescence; Fuc, fucose; Gal, galactose; GlcNAc, N-acetylglucosamine; HPAEC, high performance anion exchange chromatography; HPAEC–PAD, high-performance anion-exchange chromatography with pulsed amperometric detection; LCA, Lens culinaris agglutinin; MAL, Maackia amurensis lectin; NeuAc, N-acetylneuraminic acid; NP, normal plasma; PAD, pulse amperometric detector; PNGase F, peptidyl N-glycosidase F; PP, plasma proteins; PVDF, polyvinylidene fluoride, RCA, Ricinus communis agglutinin; sLex, sialyl Lewis X; SNA, Sambucus nigra agglutinin; TBS, Tris-buffered saline;
1-AGP,1-acid glycoprotein; 1-I3, 1-inhibitor3; 1-M, 1-macroglobulin; 2-M, 2-macroglobulin; GalNAc, N-acetylgalactosamine |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Ashwell, G. and Morell, A.G. (1974) The role of surface carbohydrates in the hepatic recognition and transport of circulating glycoproteins. Adv. Enzymol. Relat. Areas Mol. Biol., 41, 99–128.[ISI][Medline]
Bayer, E.A. and Wilchek, M. (1990) Protein biotinylation. In Wilchek, M. and Bayer, E.A. (eds), Methods Enzymol. Academic Press, London, Vol. 184, pp. 138–160.
Blake, G.J. and Ridker, P.M. (2003) C-reactive protein and other inflammatory risk markers in acute coronary syndromes. J. Am. Coll Cardiol., 41, 37S–42S.
Blum, H., Beier, H., and Gross, H.J. (1987) Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels. Electrophoresis, 8, 93–99.[CrossRef][ISI]
Brakel, C.L., Brower, M.S., and Garry, K. (1990) Streptavidin-Enzyme complexes in detection of antigens on western blots. In Wilchek, M. and Bayer, E.A. (eds.), Methods Enzymol. Academic Press, London, Vol. 184, pp. 184–436.
Brinkman-van der Linden, E.C., de Haan, P.F., Havenaar, E.C., and van Dijk, W. (1998) Inflammation-induced expression of sialyl LewisX is not restricted to alpha1-acid glycoprotein but also occurs to a lesser extent on alpha1-antichymotrypsin and haptoglobin. Glycoconj J., 15, 177–182.[CrossRef][ISI][Medline]
Casamassima, A., Picciariello, M., Quaranta, M., Berardino, R., Ranieri, C., Paradiso, A., Lorusso, V., and Guida, M. (2005) C-reactive protein: a biomarker of survival in patients with metastatic renal cell carcinoma treated with subcutaneous interleukin-2 based immunotherapy. J. Urol., 173, 52–55.[ISI][Medline]
Ceciliani, F., Giordano, A., and Spagnolo, V. (2002) The systemic reaction during inflammation: the acute-phase proteins. Protein Pept. Lett., 9, 211–223.[CrossRef][ISI][Medline]
Chang, M.K., Binder, C.J., Torzewski, M., and Witztum, J.L. (2002) C-reactive protein binds to both oxidized LDL and apoptotic cells through recognition of a common ligand: phosphorylcholine of oxidized phospholipids. Proc. Natl. Acad. Sci. U S A, 99, 13043–13048.
Chaplin, M.F. (1994) Carbohydrate Analysis. IRL Press, Oxford.
Charlwood, P.A., Hatton, M.W., and Regoeczi, E. (1976) The physicochemical and chemical properties of alpha 1-acid glycoproteins from mammalian and avian plasmas. Biochim. Biophys. Acta, 453, 81–92.[Medline]
De Graaf, T.W., Van der Stelt, M.E., Anbergen, M.G., and van Dijk, W. (1993) Inflammation-induced expression of sialyl Lewis X-containing glycan structures on alpha 1-acid glycoprotein (orosomucoid) in human sera. J. Exp Med., 177, 657–666.
Du Clos, T.W. (1996) The interaction of C-reactive protein and serum amyloid P component with nuclear antigens. Mol Biol. Rep., 23, 253–260.[CrossRef][ISI][Medline]
Duncan, B.B., Schmidt, M.I., Pankow, J.S., Ballantyne, C.M., Couper, D., Vigo, A., Hoogeveen, R., Folsom, A.R., and Heiss, G. (2003) Low-grade systemic inflammation and the development of type, 2 diabetes: the atherosclerosis risk in communities study. Diabetes, 52, 1799–1805.
Esnard, F. and Gauthier, F. (1980) Purification and physicochemical characterization of a new rat plasma proteinase inhibitor, alpha 1-inhibitor III. Biochim. Biophys. Acta, 614, 553–563.[Medline]
Farrell, D.H. (2004) Pathophysiologic roles of the fibrinogen gamma chain. Curr. Opin. Hematol., 11, 151–155.[CrossRef][ISI][Medline]
Fearon, D.T. (1997) Seeking wisdom in innate immunity. Nature, 388, 323–324.[CrossRef][Medline]
Finne, J. and Krusius, T. (1979) Structural features of the carbohydrate units of plasma glycoproteins. Eur J. Biochem., 102, 583–588.[ISI][Medline]
Gabay, C. and Kushner, I. (1999) Acute-phase proteins and other systemic responses to inflammation. N Engl. J. Med., 340, 448–454.
Hardy, M.R. and Townsend, R.R. (1994) High-pH anion-exchange chromatography of glycoprotein-derived carbohydrates. In Lennarz, W.J. and Hart, G.W. (eds), Methods Enzymol. Academic Press, London, Vol. 230, pp. 230–225.
Harlow, E. and Lane, D. (1988) Antibodies-A Laboratory Manual. Cold spring Harbor Laboratory, Cold Spring Harbor, NY.
Hart, S.P., Smith, J.R., and Dransfield, I. (2004) Phagocytosis of opsonized apoptotic cells: roles for ‘old-fashioned’ receptors for antibody and complement. Clin. Exp. Immunol., 135, 181–185.[CrossRef][ISI][Medline]
Hashimoto, S., Asao, T., Takahashi, J., Yagihashi, Y., Nishimura, T., Saniabadi, A.R., Poland, D.C., van Dijk, W., Kuwano, H., Kochibe, N., and Yazawa, S. (2004) Alpha1-acid glycoprotein fucosylation as a marker of carcinoma progression and prognosis. Cancer, 101, 2825–2836.[CrossRef][ISI][Medline]
Hiemstra, P.S. (2002) Novel roles of protease inhibitors in infection and inflammation. Biochem. Soc. Trans., 30, 116–120.[CrossRef][ISI][Medline]
Howard, I.K., Sage, H.J., Stein, M.D., Young, N.M., Leon, M.A., and Dyckes, D.F. (1971) Studies on a phytohemagglutinin from the lentil. II. Multiple forms of Lens culinaris hemagglutinin. J. Biol. Chem., 246, 1590–1595.
Huang, J.S., Huang, S.S., and Deuel, T.F. (1984) Specific covalent binding of platelet-derived growth factor to human plasma alpha, 2-macroglobulin. Proc. Natl Acad. Sci. U S A, 81, 342–346.
Jackson, P. (1994) High-resolution polyacrylamide gel electrophoresisof fluorophore-labeled reducing saccharides. In Lennarz, W.J. and Hart, G.W. (eds), Methods Enzymol. Academic Press, London, Vol. 230, pp. 250–265.
Jialal, I., Devaraj, S., and Venugopal, S.K. (2004) C-reactive protein: risk marker or mediator in atherothrombosis? Hypertension, 44, 6–11.
Kobata, A. (1992) Structures and functions of the sugar chains of glycoproteins. Eur. J. Biochem., 209, 483–501.[ISI][Medline]
Koj, A., Dubin, A., Kasperczyk, H., Bereta, J., and Gordon, A.H. (1982) Changes in the blood level and affinity to concanavalin A of rat plasma glycoproteins during acute inflammation and hepatoma growth. Biochem. J., 206, 545–553.[ISI][Medline]
Kottgen, E., Hell, B., Kage, A., and Tauber, R. (1992) Lectin specificity and binding characteristics of human C-reactive protein. J. Immunol., 149, 445–453.[Abstract]
Larsen, G.R., Sako, D., Ahern, T.J., Shaffer, M., Erban, J., Sajer, S.A., Gibson, R.M., Wagner, D.D., Furie, B.C., and Furie, B. (1992) P-selectin and E-selectin. Distinct but overlapping leukocyte ligand specificities. J. Biol. Chem., 267, 11104–11110.
Lee, Y.C. (1990) High-performance anion-exchange chromatography for carbohydrate analysis. Anal Biochem., 189, 151–162.[CrossRef][ISI][Medline]
Leong, M.M.L. and Fox, G.R. (1990) Luminescent detection of immunodot and western blots. In Wilchek, M. and Bayer, E.A. (eds), Methods Enzymol. Academic Press, London, Vol. 184, pp. 437–442.
Lombart, C., Sturgess, J., and Schachter, H. (1980) The effect of turpentine-induced inflammation on rat liver glycosyltransferases and Golgi complex ultrastructure. Biochim. Biophys. Acta, 629, 1–12.[Medline]
Mahmoud, F.A. and Rivera, N.I. (2002) The role of C-reactive protein as a prognostic indicator in advanced cancer. Curr. Oncol. Rep., 4, 250–255.[Medline]
March, S.C., Parikh, I., and Cuatrecasas, P. (1974) A simplified method for cyanogen bromide activation of agarose for affinity chromatography. Anal. Biochem., 60, 149–152.[CrossRef][ISI][Medline]
Medzhitov, R. and Janeway, C.A. Jr. (1997) Innate immunity: the virtues of a nonclonal system of recognition. Cell, 91, 295–298.[CrossRef][ISI][Medline]
Merkle, R.K. and Cummings, R.D. (1987) Lectin-affinity chromatography of glycopeptides. In Ginsburg, V. (ed.), Methods Enzymol. Academic Press, London, Vol. 138, pp. 232–259.
Montreuil, J., Bouquelet, S., Debray, H., Lemoine, J., Michalski, J.C., Spik, G., and Strecker, G. (1994) In Chaplin, M.F. and Kennedy, J.F. (eds.), Carbohydrate Analysis. Oxford University Press, London, pp. 181–293.
Nicolson, G.L., Blaustein, J., and Etzler, M.E. (1974) Characterization of two plant lectins from Ricinus communis and their quantitative interaction with a murine lymphoma. Biochemistry, 13, 196–204.[CrossRef][Medline]
Palaniyar, N., Nadesalingam, J., and Reid, K.B. (2002) Pulmonary innate immune proteins and receptors that interact with gram-positive bacterial ligands. Immunobiology, 205, 575–594.[CrossRef][ISI][Medline]
Peterson, G.L. (1977) A simplification of the protein assay method of Lowry et al. which is more generally applicable. Anal. Biochem., 83, 346–356.[CrossRef][ISI][Medline]
Rattazzi, M., Puato, M., Faggin, E., Bertipaglia, B., Zambon, A., and Pauletto, P. (2003) C-reactive protein and interleukin-6 in vascular disease: culprits or passive bystanders? J. Hypertens., 21, 1787–1803.[CrossRef][ISI][Medline]
Raynes, J.G. (1994) The acute phase response. Biochem. Soc. Trans., 22, 69–74.[ISI][Medline]
Ryden, I., Lundblad, A., and Pahlsson, P. (1999) Lectin ELISA for analysis of alpha(1)-acid glycoprotein fucosylation in