Glycobiology Advance Access originally published online on September 26, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Glycobiology vol 14 no 1 pp. 39-49, 2004
© Oxford University Press 2004; all rights reserved.
Expression of ß-galactoside 
2,6 sialyltransferase and of 2,6-sialylated glycoconjugates in normal human liver, hepatocarcinoma, and cirrhosis
2 Department of Experimental Pathology, F. Addarii Institute of Oncology, University of Bologna, Via S. Giacomo 14, 40126 Bologna, Italy; and 3 Pathology Unit, F. Addarii Institute of Oncology, University of Bologna, Bologna, Italy
Received on July 22, 2003; revised on September 2, 2003; accepted on September 2, 2003
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsReferences |
|---|
ß-Galactoside
2,6-sialyltransferase (ST6Gal.I) mediates the addition of 2,6-linked sialic acid to glycoproteins. ST6Gal.I is strongly expressed by the liver and is up-regulated in several cancers, but little is known of its regulation in human liver diseases. We have investigated the expression of ST6Gal.I and its product, the 2,6-sialylated lactosamine, in normal human liver, hepatocarcinoma (HCC), and cirrhosis. We found that both ST6Gal.I activity and mRNA can undergo up- or down-regulation in different HCC patients. At the mRNA level, the groups of specimens showing the highest expression were HCC of grade 2, HCC developed without preexisting cirrhosis, and HCC of male patients. The lectin from Sambucus nigra (SNA) reveals a significative overexpression of 2,6-sialylated glycoconjugates in HCC tissue homogenates and their intracellular accumulation in HCC histological sections, even though in a few cases the extent of 2,6-sialylation dramatically decreases. Transcription of the gene occurs through at least two different promoters, resulting in two differentially expressed mRNA species. RNA in situ hybridization reveals that the ST6Gal.I mRNA can be expressed at a quantitatively heterogeneous level among the neoplastic cells. Neither ST6Gal.I expression nor 2,6-sialylation are altered in cirrhosis. These data indicate that neoplastic transformation but not cirrhosis can alter the process of 2,6-sialylation of liver glycoproteins. Key words: cirrhosis / hepatocarcinoma / liver / Sambucus nigra agglutinin / sialyltransferases
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
The liver continuously synthesizes a huge amount of proteins, whose principal fates are to be secreted into the blood stream or inserted into the cell membranes. For both secreted and membrane proteins, glycosylation represents one of the most frequently occurring posttranslational modifications, and in some cases the sugar chains of glycoproteins play highly specific roles, such as the mediation of cell–cell and cell–matrix interactions and the binding of microorganisms and toxins (Varki, 1993
). In several pathological conditions and especially in cancer, glycosylation undergoes profound changes, which in some cases are specific for a given malignancy (Dall'Olio, 1996).
Sialic acids are sugars bearing a negative electric charge at physiological pH values, frequently terminating the sugar chains of glycoproteins and glycolipids, whose presence profoundly affects the chemicophysical as well as the biological properties of the glycoconjugates (Schauer, 2000). A determinate number of principal sialyl linkages has been described so far: sialic acid may be linked either through an 2,3- or an 2,6-bond to subterminal galactose; through an 2,6-bond to N-acetylgalactosamine (GalNAc); or through an 2,8-bond to another sialic acid, forming polysialic acid. The different sialyl linkages are elaborated by different members of the sialyltransferase family, a class of glycosyltransferases sharing the CMP–sialic acid as donor substrate but differing for the glycosidic structure on which they act and for the type of glycosidic linkage they form (Dall'Olio and Chiricolo, 2001; Harduin-Lepers et al., 2001).
ß-Galactoside 2,6-sialyltransferase I (ST6Gal.I according to the nomenclature proposed by Tsuji et al., 1996) has long been thought to be the only sialyltransferase able to catalyze the 2,6-sialylation of galactose (Weinstein et al., 1982a,b, 1987; Taatjes et al., 1988). A second ß-galactoside 2,6-sialyltransferase gene has been recently cloned (ST6Gal.II), but because this enzyme sialylates mainly oligosaccharides (Takashima et al., 2002; Krzewinski-Recchi et al., 2003), ST6Gal.I remains the only 2,6-sialyltransferase acting on glycoproteins so far identified. The gene encoding ST6Gal.I maps in chromosome 3 (q21–q28) (Wang et al., 1993), spans at least 145,000 bp of genomic DNA and is made up of at least 9 exons (Figure 1).
|
Studies performed in the past decade have revealed that at least three major mRNA species, sharing exons I–VI but differing in the 5'-untranslated regions (UTRs), can be generated in a tissue-specific manner through the use of different promoters and a differential assemblage of 5-UT sequences (Svensson et al., 1990
; Wang et al., 1990) (Figure 1). The first mRNA form, whose cDNA was cloned from a placenta cDNA library (Grundmann et al., 1990), contains the 5'-UT exons Y and Z (YZ form), is transcribed through the P3 promoter and is expressed by many tissues (Wang et al., 1993). The second species, characteristic of mature B-lymphocytes, contains the 5'-UT exon X (X form) (Stamenkovic et al., 1990) and is transcribed through the P2 promoter. The third type of transcript (H form), cloned from the hepatocarcinoma cell line HepG2, lacks exons Y, Z, or X but contains a short specific sequence in front of exon I (Aas-Eng et al., 1995; Lo and Lau, 1996); this sequence is not separated from exon I by introns in genomic DNA and consequently cannot be defined as an exon. The presence of the H transcript, which is the product of P1 promoter, although reported in normal and neoplastic colonic tissues (Dall'Olio et al., 2000) and cell lines (Dall'Olio et al., 1999) has never been reported in authentic human hepatic tissues.
ST6Gal.I is expressed at a dramatically different level among tissues, and the liver is the tissue displaying the highest level of expression (Paulson et al., 1989; Kitagawa and Paulson, 1994; Kaneko et al., 1995). Furthermore, ST6Gal.I is one of the most frequently up-regulated glycosyltransferases in human cancers (reviewed in Dall'Olio, 2000). In some malignancies, such as colon, the increased expression regards nearly 100% of the patients (Dall'Olio et al., 1989, 2000), whereas in other cancers, such as breast cancer, only a subset of patients display this modification (Recchi et al., 1998). Further interest in the expression of hepatic ST6Gal.I stems from the observation that in rat (Kaplan et al., 1983) and mouse (Dalziel et al., 1999) it behaves like an acute phase protein in that during inflammation it is dramatically up-regulated and a large amount of the cell-bound enzyme is released in the blood stream as a soluble, enzymatically active fragment.
Despite the importance of ST6Gal.I in the glycosylation machinery of the liver cells, very little is known on the regulation of human liver ST6Gal.I in normal and pathological conditions. Sometimes conflicting investigations in experimental systems have suggested that the expression of hepatic ST6Gal.I can be profoundly affected by neoplastic transformation (Miyagi et al., 1988; Jain et al., 1993; Pousset et al., 1997) and differentiation (Shah et al., 1992; Vertino-Bell et al., 1994). In this study, we investigate the expression of ST6Gal.I and of its cognate oligosaccharide structure, the sialyl 2,6-lactosaminyl epitope, in normal human liver, as well as in hepatocarcinoma (HCC) and cirrhosis.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Specimens
The features of the 21 HCC cases utilized in this study are reported in Table I. In 11 cases the tumor developed on preexisting cirrhosis, and in the remaining 10 cases cirrhosis was not present. Owing to the fact that the cirrhotic tissue cannot be referred to as normal, we defined normal as only the non-HCC tissue of noncirrhotic patients, whereas the non-HCC tissues of both cirrhotic and noncirrhotic patients is referred to as noncancer (see also Table II). The male to female ratio (61% to 39%) is grossly representative of HCC sex ratio. Fourteen out of 21 cases are grade 2 HCC; the other cases are of higher degree.
|
|
Expression of ST6Gal.I enzyme activity
The ST6Gal.I enzyme activity expressed by the 42 specimens is reported in Table I. The range of activity displayed by nonneoplastic liver is relatively narrow (from 0.44 to 1.82 nmoles h-1 mg prot-1), whereas that displayed by HCC tissues is much wider (from 0.10 to 2.51), although the mean values of noncancer and HCC tissues are very close (1.15 versus 1.20, respectively, Table II). Data reported in Table I reveal that ST6Gal.I activity may undergo dramatic up- or down-regulations as a consequence of neoplastic transformation in some patients, but in others it shows little or no changes. For example, in cancer tissue of case 8 the activity is about fourfold higher than in normal tissue, whereas in cancer tissue of case 4 it is nearly undetectable. It should be noted that the patients showing the most remarkable increase of activity (cases 1, 8, 16, and 17) are all males, and of the four patients showing the most remarkable decrease of activity (cases 4, 6, 10, 15), three are female. As a consequence, among HCC samples the average ST6Gal.I activity is about 30% lower in females than in males (0.99 versus 1.32), and among noncancer samples the activity is about 30% higher in females than in males (1.33 versus 1.03). Table II shows also that the ST6Gal.I activity is only marginally affected by the histological grade of the tumor (1.25 in G2 HCC versus 1.08 in G3–4), and it is not different in normal and cirrhotic tissues (1.14 versus 1.15). However, HCC developed on cirrhosis display a lower ST6Gal.I activity than HCC developed without underying cirrhosis (1.05 versus. 1.36).
Expression of 2,6-sialylated sugar chains in tissue homogenates
The 2,6sialyl-specific lectin from Sambucus nigra (SNA) is a widely used probe for the detection of 2,6-linked sialic acid, which is the product of ST6Gal.I (Shibuya et al., 1987). A representative SNA dot-blot analysis of the 42 liver specimens is provided in Figure 2, and the intensity of the dots (mean of at least three experiments) is reported in Table I. As shown, the reactivity of neoplastic tissues is in some cases markedly increased (cases 3, 11, 12, 21), but in a few cases it is dramatically decreased (patients 4 and 10). Statistical analysis (t test for paired samples) reveals that the SNA reactivity of HCC samples is significatively higher than that of noncancer samples (p = 0.014). Analysis of SNA reactivity by groups of specimens (Table II) suggests that the grade 3–4 specimens are slightly more reactive than grade 2 specimens (2.63 versus 2.16). Regarding the relationship between the expression of ST6Gal.I and that of 2,6-linked sialic acid, data reported in Table I indicate that the changes observed in SNA reactivity often correlate with the ST6Gal.I changes, but in some cases there is a dramatic discrepancy. The most striking example is provided by patient 15, in which the level of ST6Gal.I activity in HCC decreased, whereas the level of SNA reactivity strongly increased. A plot of the ST6Gal.I activity versus the SNA reactivity of the 42 specimens (Figure 2B) indicates the lack of a simple linear relationship between the two parameters (r = 0.057; p = 0.72); this is consistent with the notion that the regulation of 2,6-sialylation of glycoproteins is multifactorial and is not simply a function of the ST6Gal.I level.
|
To investigate the qualitative changes of SNA reactivity associated with neoplastic transformation, SNA-lectin blot analysis was performed on some representative cases (Figure 3). It should be noted that the pattern of staining is usually well conserved in noncancer and cancer samples in the region below 35 kDa. Nevertheless, specific bands appears or disappear in cancer tissues. For example, in sample 6N there are SNA-reactive bands of about 120, 68, 55, and 40 kDa, which are absent in sample 6T, whereas a strong band around 60 kDa is present in sample 6T but absent in its normal counterpart. In sample 8T, there is marked accumulation of high molecular weight (80–150 kDa) SNA-reactive bands that are not present in sample 8N. These data demonstrate a markedly heterogeneous expression of
2,6-sialylated glycoproteins in HCC specimens.
|
Expression of ST6Gal.I transcripts in surgical specimens
The expression of ST6Gal.I transcripts in surgical specimens was investigated by a semiquantitative multiplex reverse transcriptase polymerase chain reaction (RT-PCR) approach. After a preliminary investigation that ruled out the expression of the X form (data not shown), the study was focused on the YZ and H forms. An estimation of the total amount of ST6Gal.I transcripts and of the amount of the YZ and H forms was obtained through a comparison with the parallel amplification of known amounts of the YZ form and H form ST6Gal.I cDNAs. In Figure 4A is shown the multiplex amplification of the cloned YZ and H forms, mixed in the indicated proportions and used as a quantitative standard. The contemporary amplification of the two types of cDNA gives rise to two closely spaced but clearly distinguishable isoform-specific bands (indicated by the two upper arrows) and to one band resulting from the amplification of the coding region. In Figure 4B is the quantification of the three PCR bands in the three standard amplification reactions. It should be noted that the intensity of the YZ (white bar) and H products (black bar) grossly reflects the relative amount of template used. Consistent with the fact that the total amount of cDNA in the three samples is the same, the intensity of the coding region signal in the three samples is very similar (gray bar).
|
A representative example of the patterns obtained with the cDNAs from normal liver and HCC of some patients is provided in Figure 4C. In this case, an housekeeping gene (ß-actin) was coamplified to control for RNA quality. Beside a ß-actin and coding region signals of comparable intensity, all these specimens show the presence of the H transcript, indicating that this mRNA species is expressed by authentic normal and HCC liver tissues. The YZ form is expressed only in trace amounts by both N and T samples, with the remarkable exception of sample 4T, whose YZ expression is comparable with that of the H form. Interestingly, the same specimen show a nearly complete down-regulation of enzyme activity and SNA reactivity (Table I). However, a comparison between the intensity of the H and the coding region bands, reveals striking differences among specimens. In fact, although in sample 1T the intensity of the H and the coding region bands is similar, suggesting that nearly 100% of the transcript is of the H type, in sample 1N the low amount of H transcript does not appear to be sufficient to account for the intensity of coding region band detected. The contemporary absence of the YZ transcript and of the B-lymphocyte-specific X transcript (data not shown) could be suggestive of the presence of other mRNA species whose presence will be investigated by 5'-rapid amplification of cDNA ends (RACE) (see later discussion).
Table I reports the intensity of the coding region as well as of the YZ and the H bands, normalized for the internal ß-actin standard obtained with the 42 specimens. As shown, ST6Gal.I transcripts dramatically accumulate in HCC tissue of several patients (cases 8, 13, 15, 16, 17). However, in two patients a marked reduction is observed (cases 9 and 12), whereas in the other cases the level of transcript shows little or no changes. A comparison of the mean ST6Gal.I mRNA levels expressed by groups of specimens (Table II) indicates a tendency toward an accumulation in grade 2 (2.22 versus 1.60), but not in grade 3–4 HCC (1.58 versus 1.6) and in HCC developed without underlying cirrhosis (2.64 versus 1.43), but not in HCC developed on preexisting cirrhosis (1.44 versus 1.77). As observed for ST6Gal.I enzyme activity, the mRNA expression is also lower in female HCC tissues, compared with male HCC tissues (1.35 versus 2.41). The most remarkable increase of transcript expression is observed in patient 8, which shows also the most dramatic elevation of ST6Gal.I activity. Nevertheless a comparison of the ST6Gal.I activities with the respective level of mRNA also reveals marked inconsistencies. For example, in case 15 the transcript level is markedly increased in cancer tissue, but the enzyme activity is reduced. This reveals that the level of ST6Gal.I transcript may not directly correlate with the enzyme activity and suggest a multifactorial control. Multiplex RT-PCR analysis reveals that the H form is the major transcript expressed in all the 42 specimens examined (Table I). The expression of the YZ form is, in many cases, below the detection limit of this technique (50 fg cDNA); in other cases it is clearly detectable but expressed as a minor form; and in three cases (4T, 20N, and 21N) the H and YZ forms are expressed at a similar level. As already observed for sample 1N, in many cases the total level of ST6Gal.I transcript (coding region) does not appear to be supported by an adequate amount of either of the two forms. For this reason the nature of the sequences 5'-flanking exon I in sample 1N was investigated.
5'-RACE analysis of normal liver tissue
RACE allows the PCR amplification of the unknown terminal portions of a cDNA whose central part is known. The RNA from sample 1N was reverse-transcribed and a short oligonucleotide of known sequence (Generacer, Figure 5A) was linked to the 5' ends of the obtained cDNAs. The ST6Gal.I cDNAs were then PCR amplified, using Generacer primer, complementary to Generacer sequence as forward primer, and primer EIIR.5, complementary to exon II, which is shared by all known ST6Gal.I mRNAs, as reverse primer. The result of this PCR amplification is shown in Figure 5B. A single product of about 420 nt was obtained. This size is consistent with the presence of a very short sequence between Generacer and exon I. This product was cloned, and the several clones that were sequenced were all found to contain, beside exon I, the 5'-UT sequence of the H form. However, the different clones differed for the length of the hepatic sequence (Figure 5C), revealing that transcription through the P1 promoter can initiate in at least three different points (marked with an asterisk in Figure 5C). Clone 1 starts at position +11 (position 1 is the first nucleotide of the longest form), clones 2 and 3 start at position +8, and clone 4 starts at position +1. Moreover, clone 2 contained a T instead of A at position +10, revealing the existence of a possible polymorphism at this position, which generates a stretch of nine T residues: PCR amplification of the four cloned sequences with left primers HepL.1 and EIIL.2 (Figure 5D) reveals that although all the four clones were efficiently amplified with primer pair EIIL.2/EIIR.5, only clone 4 and at a much lesser extent clone 3 were amplified with primer pair HepL.1/EIIR.5. Because primer HepL.1 overlaps completely only with the longest form (clone 4), it is possible that the poor amplification yield given by some samples with HepL.1 reflects a higher proportion of shorter forms. Moreover, the difference between the amplification yields between clones 2 and 3 (which are of identical length but differ for the A/T substitution at position +10) suggests that also this difference (which creates a mismatch with primer HepL.1) contributes to a lower amplification efficiency with the H-specific primer in some samples. These data conclusively demonstrate the expression by authentic human normal liver of the H form and indicate that these transcripts are heterogeneous with respect to the length of their 5'-ends.
|
Expression of
2,6-linked sialic acid and of ST6Gal.I mRNA in histological sectionsIn Figure 6A–C, the SNA-staining patterns of histological sections from normal liver (A) and from HCC (B and C) are compared. In normal liver (A), SNA reactivity is mainly associated with sinusoids, whereas in HCC specimens this staining pattern is completely lost and the SNA reactivity appears as dense aggregates in the cytoplasm of the neoplastic hepatocytes (B, C). This pattern is shared by many of the HCC specimens examined and is consistent with an intracellular accumulation of
2,6-sialylated glycoconjugates probably because of an alteration of the apical/basolateral sorting mechanisms in cancer cells.
|
The tissue expression of ST6Gal.I mRNA, as determined by in situ hybridization, also reveals marked differences between normal and neoplastic tissues (D–I). In normal liver parenchyma (D), in situ hybridization provides a weak diffuse cytoplasmic staining evenly distributed in the tissue, whereas the cells of the bile ducts show a stronger level of expression (E). In cirrhotic tissue (F), the regenerating liver cells show a strong, uniform mRNA expression, and the surrounding connective tissue is unreactive. In HCC specimens, a variable degree of tissue heterogeneity is often evident (G–I). The specimen shown in Figure 6I shows an extreme heterogeneity, with some cells expressing very high levels of ST6Gal.I mRNA in close proximity with cells showing weak expression. Together, these data indicate that on neoplastic transformation, ST6Gal.I expression may undergo marked up- or down-regulation among the cells forming the tumor.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
In the present study we found that neoplastic transformation of hepatocytes can result in opposite effects on ST6Gal.I expression. In fact, we observed a different modulation of the enzyme in different groups of patients, and in situ hybridization shows that even within the same tumor the ST6Gal.I transcript can be up- or down-regulated by different groups of cells. These findings are not surprising in the light of the contradictory conclusions reached by the studies on experimental systems published in the past years. Beside studies reporting a positive relationship between ST6Gal.I and neoplastic transformation, others report an opposite conclusion. For example, a specific elevation of ST6Gal.I has been reported in a rat model of chemically induced hepatoma (Miyagi et al., 1988
), whereas an increase of ST6Gal.I activity and an accumulation of the mRNA were observed in HCCs spontaneously developed by transgenic mice expressing the large T antigen under the control of an hepatocyte-specific promoter (Pousset et al., 1997). On the other hand, ST6Gal.I expression was found to be strongly down-regulated, rather than increased, in a rat HCC cell line compared with normal tissue (Jain et al., 1993), and a positive relationship with organ differentiation is indicated by the fact that during ontogenesis of rat liver the level of ST6Gal.I mRNA, which is very low in newborns, dramatically increases (Vertino-Bell et al., 1994). The molecular bases of liver carcinogenesis are very heterogeneous in that many oncogenes and tumor suppressor genes may be involved (Ozturk, 1999); thus the opposite modulation of ST6Gal.I may reflect different pathways of oncogenic activation. In this light, it is interesting to note that the expression of the large T antigen, which induces HCCs overexpressing ST6Gal.I (Pousset et al., 1997), leads to the functional inactivation of the Rb protein and that the inactivation of Rb is a relatively frequent mutagenic event (20–25% of the cases) in liver carcinogenesis (Ozturk, 1999).
Although the number of cases examined in this article does not allow us to reach a definitive conclusion, we observed that ST6Gal.I expression in HCC is influenced by the grade of the tumor, by the underlying presence of cirrhosis, and by the sex of the patient. A lower ST6Gal.I expression by the less differentiated tumors, a tendency more evident at the mRNA level, is in agreement with a recently published article that reports a down-regulation of the ST6Gal.I enzyme protein in less differentiated HCC specimens (Cao et al., 2002). The fact that cirrhosis does not alter the expression of ST6Gal.I is unexpected because inflammatory conditions are well known to induce an up-regulation of this enzyme through a mechanism that, in mice, involves an IL-6-mediated transcription through the P1 promoter (Dalziel et al., 1999). Moreover, in HCC developed without preexisting cirrhosis the average ST6Gal.I mRNA is increased, but in those developed on a preexisting cirrhosis this tendency is not observed. This difference could be explained if it is hypothesized that different pathways of transformation are preferentially activated in the two conditions and that ST6Gal.I is modulated only by given pathways. ST6Gal.I expression, both at the enzyme activity and mRNA level, is lower in female than in male HCC specimens, although in noncancer specimens this tendency is less evident. In considering this observation, it should be kept in mind that ST6Gal.I can be modulated by steroids, such as dexamethasone, in rat hepatoma cells (Wang et al., 1989), suggesting that transcription of the gene might be affected by sexual steroid hormones.
In this study we have also demonstrated that in both normal and HCC tissues ST6Gal.I transcription occurs mainly through the P1 promoter, resulting in the H form as the major mRNA species. However, as occurs in colonic tissues (Dall'Olio et al., 1999, 2000), the P3 promoter can also be utilized by liver tissues and the contribution of the H and YZ transcripts to the total mRNA pool is variable. As previously observed in HepG2 cells (Lo and Lau, 1996), transcription through the P1 promoter can start in different points, giving rise to 5'-UTRs of different lengths. Even though the biological significance of a differential use of P1 and P3 promoters in human liver remains to be investigated, a recent study has shown that knockout mice specifically unable to express ST6Gal.I through the P1 promoter show a reduced ST6Gal.I expression beside a surprisingly normal level of 2,6-sialylation of serum glycoproteins (Appenheimer et al., 2003). Interestingly, on challenge with Salmonella typhimurium these mice show a greater accumulation of neutrophils in the peritoneal space and an increased bacterial burden in liver and spleen, suggesting that the P1-driven ST6Gal.I expression can play a role in antimicrobial defense mechanisms.
The expression of 2,6-sialylated glycoconjugates has been studied using SNA as a probe. The specificity of this lectin has been reported to include the 2,6-sialylated lactosamine (the ST6Gal.I product) as well as the sialyl-Tn (sialic acid 2,6-linked to N-acetylgalactosamine), which is the product of different sialyltransferases (Shibuya et al., 1987). However, histochemical studies have revealed that the contribution of sialyl-Tn to SNA reactivity of tissue sections is negligible (Murayama et al., 1997) and SNA can be considered a specific tool for the ST6Gal.I product. 2,6-Sialylated glycoconjugates show a significative tendency to accumulate in HCC tissue specimens, even though in a minority of patients we observed a lower expression, and histochemical studies reveal that the tissue distribution of these compounds in HCC can be markedly altered. In fact, in normal liver 2,6-sialylated glycoconjugates are mainly expressed along the sinusoids, but in HCC they often accumulate as dense aggregates in the cytoplasm of neoplastic hepatocytes. This latter observation is consistent with that obtained with another 2,6-sialyl-specific lectin, the CD22 molecule (Cao et al., 2002) and suggests an alteration of the mechanisms of intracellular transport in HCC. These data demonstrate that factors other than the ST6Gal.I level can influence the accumulation of 2,6-sialylated glycoconjugates and explain why we observed that the relationship between the expression of ST6Gal.I and that of 2,6-sialylated glycoconjugates is often loose. Although transfection studies have convincingly demonstrated that the expression of the ST6Gal.I cDNA is by itself sufficient to induce the appearance of 2,6-sialylated glycans on the cell surface (Dall'Olio et al., 1995; Lee et al., 1989), studies aimed at establishing the quantitative relationship between ST6Gal.I expression and 2,6-sialylation have revealed that such a relationship may be very loose not only in tissues (Dall'Olio et al., 2000; Kaneko et al., 1995) but also in cell lines transfected with the ST6Gal.I cDNA under the control of a constitutive promoter (Dall'Olio et al., 2001). In this strictly controlled artificial system, the amount of 2,6-linked sialic acid on the cell surface depends strongly on the cell line used, not only on the level of ST6Gal.I enzyme activity.
In conclusion, our study demonstrates that the expression of ST6Gal.I and of 2,6-sialylated glycoconjugates can undergo dramatic variations as a consequence of neoplastic transformation of hepatocytes. The identification of the molecular events at the basis of such alterations will require further work.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Surgical specimens
Clinical information on the 21 patients who underwent surgery for HCC is reported in Table I. Informed consent was obtained from all patients. Samples from nonneoplastic and HCC tissues were removed at surgery. A portion of the tissue was formalin-fixed and paraffin-embedded for routine histopathological examination; other portions of the tissue were frozen in liquid nitrogen and stored at -80°C.
Sialyltransferase assay and blot analysis
Tissues were potter-homogenized in ice-cold distilled water; the protein concentration of the homogenate was measured by the Lowry method and adjusted to 10 mg/ml. ST6Gal.I activity was measured in whole homogenates in the range of linearity with respect to time and enzyme concentration as incorporation of [14C]-labeled sialic acid on asialotransferrin (prepared by mild acid hydrolysis of human transferrin) as detailed elsewere (Dall'Olio et al., 1996). For dot-blot analysis, 10 µg of the homogenates was applied to Hybond nitrocellulose membrane (Amersham, Little Chalfont, U.K.). Membranes were probed with 1 µg/ml digoxigenin-conjugated SNA (SNA-dig) (Boehringer) which was detected with horseradish peroxidase–labeled antidigoxigenin antibodies (Boehringer). The reaction was finally developed with SuperSignal West Pico Chemiluminescent Substrate (Pierce, Rockford, IL) and detected by autoradiography. Intensity of the dots was quantitated by the Kodak Digital Science 1D software. Details on dot-blot procedure have been reported previously (Dall'Olio et al., 2000). For lectin blot analysis, 100 µg of proteins from total tissue homogenates was electrophoresed under reducing conditions on a 10% polyacrylamide gel and electrotransferred to Hybond membrane. After a wash with phosphate buffered saline with Tween, the blot was processed as detailed for dot-blot.
RT-PCR analysis
Total RNA was extracted from normal and neoplastic tissues by RNAZolB B (Biotecx Laboratories, Houston, TX) method. Two micrograms of total RNA were reverse-transcribed using the TaKaRa RT-PCR kit Ver 2.1 (TaKaRa, Shuzo, Japan), using random 9-mers primers, according to the manufacturer's instructions. Two to five microliters of the cDNAs were amplified by a multiplex RT-PCR approach, which allows the simultaneous amplification of the YZ form, the H form, and the coding region of ST6Gal.I as well as of the ß-actin cDNA as an internal control. The PCR reaction contained in a final volume of 50 µl: 1x Taq polymerase buffer; 1.7 mM MgCl2; 0.2 mM each dATP, dCTP, dGTP, dTTP; 0.5 U InViTAQ DNA polymerase (Eppendorf, Milan, Italy); and the following concentrations of primers (the approximate position of the primers is indicated in Figure 1):
- 120 nM primer EIIL.2, forward primer for coding region of ST6Gal.I (5'-CTGCGTCCTGGTCTTTCTTC-3')
- 500 nM primer HepL.1, forward primer for H form (5'-GTCTCTTATTTTTTGCCTTTGCAG-3')
- 250 nM primer EZL.1, forward primer for YZ form (5'-GAGAAGTGGTGAATGTCATGGAG-3')
- 250 nM primer EVR.3, common reverse primer for ST6Gal.I (5'-TGGGTCCCATACAATTAGGAT-3')
- 40 nM each of the primers ACTL.3 (5' GGCATCGTGATGGACTCCG 3') and ACTR.3 (5' GCTGGAAGGTGGACAGCGA 3') for ß-actin
After a denaturation step for 1 min at 94°C, amplification was performed for 35 cycles of the following program: denaturing 94°C, 1 min; annealing 60°C, 1 min; elongation 72°C, 2 min. PCR products were analyzed on a 2% agarose gel stained with ethidium bromide. The intensity of the bands was quantified by the Kodak Digital Science 1D software and compared with that given by known amounts of YX and H cDNA standards, prepared as follows. The cDNA from HepG2 cells, a cell line known to express both the YZ and the H forms (Dall'Olio et al., 1999) was used as a source for the PCR amplification of the two forms. After gel isolation, the PCR products were cloned in pGEM-T easy vector (Promega, Madison, WI). The concentration of the purified plasmids was carefully determined by A260 optical density and brought to a concentration of standard cDNA of 25 fg/µl.
5'-RACE
The whole procedure was made according to the instructions of the GeneRacer kit (Invitrogen, Carlsbad, CA). Five micrograms of normal liver total RNA were reverse-transcribed using random primers, and the resulting cDNA was 5'-ligated with GeneRacer oligonucleotide (Invitrogen) and subjected to PCR amplification with GeneRacer primer (Invitrogen) and oligonucleotide EIIR.5 (5'-AATCAGACCCCATGGCCAATTTCC-3') as follows: preliminary denaturing step: 94°C, 1 min, then 37 cycles of the following program: denaturing 94°C, 1 min; annealing 62°C, 1 min; extension 72°C, 1 min. The resulting PCR product was gel isolated, cloned by using the TOPO TA cloning kit (Invitrogen), and sequenced. All sequence analysis were performed automatically using a Beckman-Coulter CEQ2000XL DNA analysis system.
SNA staining of histological sections
Sections were deparafinized in xylene, rehydrated in graded ethanol, and subjected to antigen retrieval (boiling in 10 mM citrate buffer, pH 6, for 15 min). Previous experiments had shown that this treatment maximizes SNA reactivity. SNA-dig staining and detection with alkaline phosphatase–conjugated antidigoxigenin antibodies (Boehringer) were as previously described (Dall'Olio and Trere, 1993).
In situ hybridization
Nonisotopic in situ hybridization was performed essentially as described previously (Fiorentino et al., 1999). Digoxigenin-labeled probes were prepared from the linearized pGEM-T easy vector containing the ST6Gal.I cDNA from HepG2 cells (see previous discussion). Sense and anti-sense RNA probes were generated with T3 and T7 RNA polymerase for 2 h at 37°C in 1x transcription buffer; 35 U RNAase inhibitor; 1 mM each ATP, CTP, GTP; and 1 mM of a mixture of unlabeled UTP and digoxigenin UTP (6.5:3.5 ratio) (Boehringer). The size of the probes was reduced to 50–100 nucleotides by means of alkaline hydrolysis. Before hybridization, tissue specimens were digested with 10 mg/ml proteinase K (Sigma, St. Louis, MO) 30 min at 37°C. Hybridization was performed at 46°C overnight with 10 pM of digoxigenin-labeled probe in 25 µl hybridization buffer (50% deionized formamide, 2x saline sodium citrate, 10% dextran sulfate, 1% sodium dodecyl sulfate). Posthybridization washes were at 60°C in 50% deionized formamide 2x saline sodium citrate for 30 min. Antidigoxigenin antibodies 1:500 diluted were applied overnight at 4°C. Detection was accomplished with nitro-blue tetrazolium/ 5-bromo-4-chloro-3-indolyl-phosphate for 3–6 h. Sections were then counterstained in methyl green. Control for specificity was performed with sense probe.
|
Acknowledgements |
|---|
This work was supported by grants from the University of Bologna to F.D. and W.F.G., Funds for Selected Research Topics to F.D., and from the MIUR to F.D. and W.F.G.
|
Footnotes |
|---|
1 To whom correspondence should be addressed; e-mail: dallolio{at}alma.unibo.it
|
Abbreviations |
|---|
HCC, hepatocellular carcinoma; RACE, rapid amplification of cDNA ends; RT-PCR, reverse transcriptase polymerase chain reaction; SNA, Sambucus nigra agglutinin; ST6Gal.I, ß-galactoside
2,6-sialyltransferase I; UTR, untranslated regions |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Aas-Eng, D.A., Asheim, H.C., Deggerdal, A., Smeland, E., and Funderud, S. (1995) Characterization of a promoter region supporting transcription of a novel human ß-galactoside
-2,6-sialyltransferase transcript in HepG2 cells. Biochim. Biophys. Acta, 1261, 166–169.[Medline]
Appenheimer, M.M., Huang, R.Y., Chandrasekaran, E.V., Dalziel, M., Hu, Y.P., Soloway, P.D., Wuensch, S.A., Matta, K.L., and Lau, J.T. (2003) Biologic contribution of P1 promoter-mediated expression of ST6Gal I sialyltransferase. Glycobiology, 13, 591–600.
Cao, Y., Merling, A., Crocker, P.R., Keller, R., and Schwartz-Albiez, R. (2002) Differential expression of beta-galactoside alpha2,6 sialyltransferase and sialoglycans in normal and cirrhotic liver and hepatocellular carcinoma. Lab Invest., 82, 1515–1524.[ISI][Medline]
Dall'Olio, F. (1996) Protein glycosylation in cancer biology: an overview. Clin. Molec. Pathol., 49, M123–M126.
Dall'Olio, F. (2000) The sialyl-2,6-lactosaminyl-structure: biosynthesis and functional role. Glycoconj. J., 17, 669–676.[CrossRef][Medline]
Dall'Olio, F. and Chiricolo, M. (2001) Sialyltransferases in cancer. Glycoconj. J., 18, 841–850.[CrossRef][Medline]
Dall'Olio, F. and Trere, D. (1993) Expression of 2,6-sialylated sugar chains in normal and neoplastic colon tissues. Detection by digoxigenin-conjugated Sambucus nigra agglutinin. Eur. J. Histochem., 37, 257–265.[ISI][Medline]
Dall'Olio, F., Malagolini, N., Di Stefano, G., Minni, F., Marrano, D., and Serafini-Cessi, F. (1989) Increased CMP-NeuAc:Galß1,4GlcNAc-R 2,6 sialyltransferase activity in human colorectal cancer tissues. Int. J. Cancer, 44, 434–439.[ISI][Medline]
Dall'Olio, F., Chiricolo, M., Lollini, P., and Lau, J.T. (1995) Human colon cancer cell lines permanently expressing 2,6-sialylated sugar chains by transfection with rat ß-galactoside 2,6 sialyltransferase cDNA. Biochem. Biophys. Res. Commun., 211, 554–561.[CrossRef][ISI][Medline]
Dall'Olio, F., Malagolini, N., Guerrini, S., Lau, J.T., and Serafini-Cessi, F. (1996) Differentiation-dependent expression of human ß-galactoside 2,6-sialyltransferase mRNA in colon carcinoma CaCo-2 cells. Glycoconj. J., 13, 115–121.[CrossRef][ISI][Medline]
Dall'Olio, F., Chiricolo, M., and Lau, J.T. (1999) Differential expression of the hepatic transcript of ß-galactoside 2,6-sialyltransferase in human colon cancer cell lines. Int. J. Cancer, 81, 243–247.[CrossRef][ISI][Medline]
Dall'Olio, F., Chiricolo, M., Ceccarelli, C., Minni, F., Marrano, D., and Santini, D. (2000) ß-Galactoside 2,6 sialyltransferase in human colon cancer: contribution of multiple transcripts to regulation of enzyme activity and reactivity with Sambucus nigra agglutinin. Int. J. Cancer, 88, 58–65.[CrossRef][ISI][Medline]
Dall'Olio, F., Chiricolo, M., Mariani, E., and Facchini, A. (2001) Biosynthesis of the cancer-related sialyl-2,6-lactosaminyl epitope in colon cancer cell lines expressing ß-galactoside 2,6-sialyltransferase under a constitutive promoter. Eur. J Biochem., 268, 5876–5884.[ISI][Medline]
Dalziel, M., Lemaire, S., Ewing, J., Kobayashi, L., and Lau, J.T. (1999) Hepatic acute phase induction of murine ß-galactoside 2,6 sialyltransferase (ST6Gal I) is IL-6 dependent and mediated by elevation of exon H-containing class of transcripts. Glycobiology, 9, 1003–1008.
Fiorentino, M., D'Errico, A., Altimari, A., Barozzi, C., and Grigioni, W.F. (1999) High levels of BCL-2 messenger RNA detected by in situ hybridization in human hepatocellular and cholangiocellular carcinomas. Diagn. Mol. Pathol., 8, 189–194.[CrossRef][ISI][Medline]
Grundmann, U., Nerlich, C., Rein, T., and Zettlmeissl, G. (1990) Complete cDNA sequence encoding human ß-galactoside -2,6-sialyltransferase. Nucleic Acids Res., 18, 667.
Harduin-Lepers, A., Vallejo-Ruiz, V., Krzewinski-Recchi, M., Samyn-Petit, B., Julien, S., and Delannoy, P. (2001) The human sialyltransferase family. Biochimie, 83, 727–737.[Medline]
Jain, N., Sudhakar, C., and Das, M.R. (1993) Regulation of expression of ß-galactoside 2,6-sialyltransferase in a rat tumor, Zajdela ascitic hepatoma. FEBS Lett., 317, 147–151.[CrossRef][ISI][Medline]
Kaneko, Y., Yamamoto, H., Colley, K.J., and Moskal, J.R. (1995) Expression of Gal ß1,4GlcNAc 2,6-sialyltransferase and 2,6-linked sialoglycoconjugates in normal human and rat tissues. J. Histochem. Cytochem., 43, 945–954.[Abstract]
Kaplan, H.A., Woloski, B.M., Hellman, M., and Jamieson, J.C. (1983) Studies on the effect of inflammation on rat liver and serum sialyltransferase. Evidence that inflammation causes release of Gal ß1,4GlcNAc 2,6 sialyltransferase from liver. J. Biol. Chem., 258, 11505–11509.
Kitagawa, H. and Paulson, J.C. (1994) Differential expression of five sialyltransferase genes in human tissues. J. Biol. Chem., 269, 17872–17878.
Krzewinski-Recchi, M.A., Julien, S., Juliant, S., Teintenier-Lelievre, M., Samyn-Petit, B., Montiel, M.D., Mir, A.M., Cerutti, M., Harduin-Lepers, A., and Delannoy, P. (2003) Identification and functional expression of a second human beta-galactoside alpha2,6-sialyltransferase, ST6Gal II. Eur. J Biochem., 270, 950–961.[ISI][Medline]
Lee, E.U., Roth, J., and Paulson, J.C. (1989) Alteration of terminal glycosylation sequences on N-linked oligosaccharides of Chinese hamster ovary cells by expression of ß-galactoside 2,6-sialyltransferase. J. Biol. Chem., 264, 13848–13855.
Lo, N.W. and Lau, J.T. (1996) Transcription of the ß-galactoside 2,6-sialyltransferase gene in B lymphocytes is directed by a separate and distinct promoter. Glycobiology, 6, 271–279.
Miyagi, T., Koseki, M., and Tsuiki, S. (1988) Comparative study of the levels of sialyltransferases responsible for the formation of sugar chains in glycoproteins and gangliosides in rat liver and hepatomas. Jpn. J. Cancer Res., 79, 742–749.[ISI][Medline]
Murayama, T., Zuber, C., Seelentag, W.K., Li, W.P., Kemmner, W., Heitz, P.U., and Roth, J. (1997) Colon carcinoma glycoproteins carrying 2,6-linked sialic acid reactive with Sambucus nigra agglutinin are not constitutively expressed in normal human colon mucosa and are distinct from sialyl-Tn antigen. Int. J. Cancer, 70, 575–581.[CrossRef][ISI][Medline]
Ozturk, M. (1999) Genetic aspects of hepatocellular carcinogenesis. Semin. Liver Dis., 19, 235–242.[ISI][Medline]
Paulson, J.C., Weinstein, J., and Schauer, A. (1989) Tissue-specific expression of sialyltransferases. J. Biol. Chem., 264, 10931–10934.
Pousset, D., Piller, V., Bureaud, N., Monsigny, M., and Piller, F. (1997) Increased 2,6 sialylation of N-glycans in a transgenic mouse model of hepatocellular carcinoma. Cancer Res., 57, 4249–4256.
Recchi, M.A., Hebbar, M., Hornez, L., Harduin-Lepers, A., Peyrat, J.P., and Delannoy, P. (1998) Multiplex reverse transcription polymerase chain reaction assessment of sialyltransferase expression in human breast cancer. Cancer Res., 58, 4066–4070.
Schauer, R. (2000) Achievements and challenges of sialic acid research. Glycoconj. J., 17, 485–499.[CrossRef][ISI][Medline]
Shah, S., Lance, P., Smith, T.J., Berenson, C.S., Cohen, S.A., Horvath, P.J., Lau, J.T., and Baumann, H. (1992) n-Butyrate reduces the expression of ß-galactoside 2,6-sialyltransferase in Hep G2 cells. J. Biol. Chem., 267, 10652–10658.
Shibuya, N., Goldstein, I.J., Broekaert, W.F., Nsimba-Lubaki, M., Peeters, B., and Peumans, W.J. (1987) The elderberry (Sambucus nigra L.) bark lectin recognizes the Neu5Ac(2-6)Gal/GalNAc sequence. J. Biol. Chem., 262, 1596–1601.
Stamenkovic, I., Asheim, H.C., Deggerdal, A., Blomhoff, H.K., Smeland, E.B., and Funderud, S. (1990) The B cell antigen CD75 is a cell surface sialytransferase. J. Exp. Med., 172, 641–643.
Svensson, E.C., Soreghan, B., and Paulson, J.C. (1990) Organization of the ß-galactoside 2,6-sialyltransferase gene. Evidence for the transcriptional regulation of terminal glycosylation. J. Biol. Chem., 265, 20863–20868.
Taatjes, D.J., Roth, J., Weinstein, J., and Paulson, J.C. (1988) Post-Golgi apparatus localization and regional expression of rat intestinal sialyltransferase detected by immunoelectron microscopy with polypeptide epitope-purified antibody. J. Biol. Chem., 263, 6302–6309.
Takashima, S., Tsuji, S., and Tsujimoto, M. (2002) Characterization of the second type of human ß-galactoside 2,6-sialyltransferase (ST6Gal II), which sialylates Galß1,4GlcNAc structures on oligosaccharides preferentially. Genomic analysis of human sialyltransferase genes. J. Biol. Chem., 277, 45719–45728.
Tsuji, S., Datta, A.K., and Paulson, J.C. (1996) Systematic nomenclature for sialyltransferases [letter]. Glycobiology, 6, v–vii.[Medline]
Varki, A. (1993) Biological roles of oligosaccharides: all of the theories are correct. Glycobiology, 3, 97–130.
Vertino-Bell, A., Ren, J., Black, J.D., and Lau, J.T. (1994) Developmental regulation of ß-galactoside 2,6-sialyltransferase in small intestine epithelium. Dev. Biol., 165, 126–136.[CrossRef][ISI][Medline]
Wang, X., O'Hanlon, T.P., Young, R.F., and Lau, J.T. (1990) Rat ß-galactoside 2,6-sialyltransferase genomic organization: alternate promoters direct the synthesis of liver and kidney transcripts. Glycobiology, 1, 25–31.[Medline]
Wang, X., Vertino, A., Eddy, R.L., Byers, M.G., Jani-Sait, S.N., Shows, T.B., and Lau, J.T. (1993) Chromosome mapping and organization of the human ß-galactoside 2,6-sialyltransferase gene. Differential and cell-type specific usage of upstream exon sequences in B-lymphoblastoid cells. J. Biol. Chem., 268, 4355–4361.
Wang, X.C., O'Hanlon, T.P., and Lau, J.T. (1989) Regulation of ß-galactoside 2,6-sialyltransferase gene expression by dexamethasone. J. Biol. Chem., 264, 1854–1859.
Weinstein, J., de Souza e Silva, U., and Paulson, J.C. (1982a) Purification of a Gal ß1,4GlcNAc 2,6 sialyltransferase and a Gal ß1,3(4)GlcNAc 2,3 sialyltransferase to homogeneity from rat liver. J. Biol. Chem., 257, 13835–13844.
Weinstein, J., de Souza e Silva, U., and Paulson, J.C. (1982b) Sialylation of glycoprotein oligosaccharides N-linked to asparagine. Enzymatic characterization of a Gal ß1,3(4)GlcNAc 2,3 sialyltransferase and a Gal ß1,4GlcNAc 2,6 sialyltransferase from rat liver. J. Biol. Chem., 257, 13845–13853.
Weinstein, J., Lee, E.U., McEntee, K., Lai, P.H., and Paulson, J.C. (1987) Primary structure of ß-galactoside 2,6-sialyltransferase. Conversion of membrane-bound enzyme to soluble forms by cleavage of the NH2-terminal signal anchor. J. Biol. Chem., 262, 17735–17743.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Nasirikenari, B. H. Segal, J. R. Ostberg, A. Urbasic, and J. T. Lau Altered granulopoietic profile and exaggerated acute neutrophilic inflammation in mice with targeted deficiency in the sialyltransferase ST6Gal I Blood, November 15, 2006; 108(10): 3397 - 3405. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||