Skip Navigation



International Immunology Advance Access published online on July 2, 2007
International Immunology, doi:10.1093/intimm/dxm053
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
19/7/847    most recent
dxm053v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Hernandez, J. D.
Right arrow Articles by Baum, L. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hernandez, J. D.
Right arrow Articles by Baum, L. G.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Japanese Society for Immunology. 2007. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

T-cell activation results in microheterogeneous changes in glycosylation of CD45

Joseph D. Hernandez1, Jeffrey Klein1, Stephen J. Van Dyken3, Jamey D. Marth3 and Linda G. Baum1,2

1 Department of Pathology and Laboratory Medicine
2 Jonsson Comprehensive Cancer Center, UCLA School of Medicine, 10833 LeConte Avenue, Los Angeles, CA 90095, USA
3 Department of Cellular and Molecular Medicine, Howard Hughes Medical Institute, University of California, San Diego, La Jolla, CA 92093, USA

Correspondence to: Correspondence to: L. G. Baum; E-mail: lbaum{at}mednet.ucla.edu


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 References
 
During T-cell development and activation, dramatic changes occur in glycan structures that decorate cell-surface glycoproteins. These changes have been considered to be general cellular events that affect many glycans on many glycoproteins. For example, loss of sialic acid from core 1 O-glycans on T-cell surface glycoproteins CD45, CD43 and CD8, detected with peanut agglutinin (PNA), is a hallmark of immature thymocytes and activated peripheral T cells. Loss of cell-surface sialic acid during T-cell activation has been proposed to enhance TCR reactivity with antigen. However, CD4 T-cell activation also results in increased binding of the CZ-1 antibody that recognizes a sialic acid-containing epitope on CD45RB. This indicates that increased sialylation of the CZ-1 epitope occurs during CD4 T cell activation, and that loss of cell surface sialic acid during T-cell activation is a selective event rather than affecting all cell surface glycans. As specific glycans on specific glycoprotein backbones control critical events in T-cell maturation and survival, understanding mechanisms of selective glycoprotein glycosylation is important for regulating T-cell development and function. We define the sialylated O-glycan epitope recognized by CZ-1, and find that, paradoxically, CZ-1 and PNA binding are simultaneously increased on activated CD4+ T cells, demonstrating site-specific changes in CD45 sialylation. Moreover, we identify ST3Gal I as the sialyltransferase responsible for creating the CZ-1 epitope. Thus, changes in glycan structure during T-cell activation are microheterogeneous and unique to individual glycans on specific glycoproteins, implying that these glycans have precise functions in T-cell biology.

Keywords: glycosyltransferase, O-glycan, sialic acid


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 References
 
Glycosylation of cell-surface glycoproteins and glycolipids regulates multiple processes critical for T-cell function, including thymic development, trafficking, antigen recognition, activation and survival (1–11). Glycosylation is a complex process that creates an array of glycan structures on the surface of T cells (12), and distinct glycan structures regulate specific T-cell functions. For example, during thymocyte development, O-linked glycosylation of Notch by the Fringe glycosyltransferase controls T-cell lineage commitment (13). In addition, selective trafficking of T-cell subsets to tissues is controlled by sialylation and fucosylation of mucin type O-glycans (2, 10). Also, the threshold of T-cell response to antigen and cytokine production is controlled by expression of tetrantennary N-glycans synthesized by N-glucosaminyltransferase V (4, 8).

T-cell development in the thymus and activation in the periphery are accompanied by changes in N-glycan structures (4, 14), O-glycan structures (9, 15–17) and terminal addition of sialic acid to N- and O-glycans (9, 11, 18–23). Many of these changes in glycosylation were thought to be global as the changes were typically detected with plant lectins that recognize the same glycan structure on multiple polypeptide backbones (19, 24, 25). It is well known that T-cell activation in vitro and in vivo is accompanied by loss of sialic acid from core 1 O-glycans (SA{alpha}2,3Galß1,3GalNAc-Ser/Thr) to expose asialo core 1 O-glycans (Galß1,3GalNAc-Ser/Thr) (9, 19, 21); exposure of asialo core 1 O-glycans on activated T cells was detected with the plant lectin peanut agglutinin (PNA) that recognizes Galß1,3GalNAc sequences on several glycoproteins including CD8, CD43 and CD45 (26). Intriguingly, it was recently shown that increased PNA binding to activated T cells is primarily due to increased synthesis of CD45-bearing asialo core 1 O-glycans, suggesting that de novo synthesis of hyposialylated CD45, and not de-sialylation of existing glycoproteins, is responsible for increased binding of PNA on activated T cells (27). Reduced sialylation of CD45 on activated T cells may be important for promoting antigen recognition, TCR signaling and cytokine production (28, 29).

Changes in cellular glycosylation have also been detected using mAbs that recognize glycosylation-dependent differentiation markers on lymphocytes. CT-1, B220, 1B11, CLA (HECA-452), 6C10 and SLeX (CSLEX-1) are examples of glycosylation-dependent epitopes recognized by mAbs (30–33). Unlike plant lectins that can recognize a glycan ligand on multiple polypeptide backbones, some of these mAbs recognize glycosylation-dependent epitopes on specific polypeptides, such as B220 and 6C10 that recognize glycan epitopes only on CD45 and Thy-1, respectively, while others recognize glycosylation-dependent epitopes on several polypeptides, such as CT-1, 1B11 and SLeX. Because the precise structure of many of these epitopes is not well understood, it remains unclear whether these mAbs detect changes in glycosylation at all potential sites on the polypeptide backbone or changes at specific sites on specific polypeptides.

Welsh et al. described a sialic acid-dependent epitope on CD45RB that is up-regulated during CD4 T-cell activation, recognized by the CZ-1 mAb (34–36). This was surprising, given that increased PNA binding to activated T cells results from loss of sialic acid on core 1 O-glycans on CD45, as described above (21, 27). We have determined that the CZ-1 epitope requires one or more sialylated core 1 O-glycans (SA{alpha}2,3Galß1,3GalNAc-Ser/Thr) on CD45RB, and that expression of this sialic acid-dependent epitope paradoxically increases while overall O-linked sialylation of core 1 O-glycans decreases during CD4 T-cell activation. These results demonstrate that changes in glycosylation are not global, but are instead microheterogeneous, and are specific for both individual glycan chains and the glycoproteins that display these. These results indicate that unique glycan structures define precise points in T-cell activation.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 References
 
Cell lines and reagents
Murine BW5147, CD45– and phytohemagglutinin (PHA)R cell lines were maintained as previously described (37). CD45– cells are a CD45-deficient derivative of the BW5147 T-cell line. CD45R0, CD45RC and CD45RBC cells were produced by transfecting CD45– cells with plasmids encoding CD45 lacking the alternatively spliced exons, containing the C exon or containing the B and C exons, respectively (38). CD45R0, CD45RC and CD45RBC cells (the kind gift of Kim Bottomly, Yale University) were maintained as described for CD45– cells with the addition of 0.5 mg ml–1 hygromycin. PHAR cells were derived by selection of BW5147 cells with the lectin PHA and express increased levels of core 2 O-glycans on CD45 compared with the parental line (37, 39, 40). EL-4 cells were maintained in DMEM with 10% FBS, 2 mM L-glutamine, 1 mM MEM sodium-pyruvate, 100 U ml–1 penicillin and 0.1 mg ml–1 streptomycin. To modify cell-surface glycans, cells were grown in the presence of 2 mM benzyl-{alpha}-GalNAc to inhibit O-glycan elongation, 2 mM 1-deoxymannojirimycin (DMNJ) to inhibit N-glycan elongation, or 1% dimethylsulfoxide alone (vehicle control), for 2 days prior to analysis.

The 1B11 antibody that recognizes core 2 O-glycan-modified murine CD43, anti-CD45RB-FITC, anti-CD4-R-phycoerythrin (clone H129.19), rat IgG2a-FITC and rat IgM were obtained from BD Biosciences, San Diego, CA, USA. µ-Chain-specific goat anti-rat-IgM F(ab')2 conjugated to FITC or APC was obtained from Jackson Immunoresearch, West Grove, PA, USA. FITC-conjugated PNA and Sambucus nigra agglutinin (SNA) were obtained from Vector Laboratories, Burlingname, CA, USA. The CZ-1 antibody was the kind gift of Ray Welsh (University of Massachusetts Medical Center, Worcester, MA, USA).

Flow cytometry
Cells and/or splenocytes were washed with PBS/1%BSA. Splenocytes were blocked with 10 µg mouse IgG prior to staining. Cells were stained with antibodies or lectins at the concentrations recommended by the manufacturer with isotype-matched antibodies or FITC–BSA as respective controls. Following staining, cells were washed with PBS and resuspended in PBS with (cell lines) or without (splenocytes) 7-aminoactinomycinD (7-AAD, Molecular Probes, Eugene, OR, USA). Cells were analyzed on a FACScan or FACSCalibur flow cytometer (BD Biosciences). Analysis of cell lines was restricted to viable cells based on forward versus side-scatter profiles and 7-AAD negativity. Analysis of splenocytes was restricted to viable CD4 cells based on forward versus side-scatter profiles and CD4 expression.

Sialidase treatment of cells
To remove {alpha}2,3-linked sialic acid from cell-surface glycoproteins, cells were washed with HBSS and resuspended at 1 x 106 ml–1 in HBSS. Twenty-five units {alpha}2,3-specific sialidase L from Macrobdella leech (V Labs, Covington, LA, USA) or buffer alone was added to 106 cells for 2 h at 37°C. Cells were washed with HBSS prior to flow cytometry analysis. The efficacy of enzyme treatment was assessed by flow cytometric analysis with PNA and SNA to detect exposure of asialo core 1 O-glycans and {alpha}2,6-linked sialic acid, respectively.

Splenocyte stimulation
To analyze wild-type splenocytes, single-cell suspensions were made from C57Bl/6 mice. To assess the roles of individual sialyltransferases, single-cell suspensions of splenocytes were isolated from sialyltransferase-deficient mice or age-matched wild-type controls; the ST3Gal I–/–, ST3Gal II–/– and ST3Gal IV–/– mice have been previously described (9, 41, 42). Cells were resuspended in RPMI with 5%FBS, 2 mM Glutamax (Invitrogen, Carlsbad, CA, USA) and 10 mM HEPES and stimulated with 1 µg ml–1 anti-CD3 (clone 145-2C11, BD Biosciences) and 100 u ml–1 IL-2 (Biosource, Camarillo, CA, USA) for the indicated times prior to flow cytometry analysis.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 References
 
CZ-1 recognizes an {alpha}2,3 sialylated core 1 O-glycan-dependent epitope on CD45RB
CD45 is a large transmembrane tyrosine phosphatase with numerous N- and O-glycans [Fig. 1, for review see ref. (29)]. The membrane proximal extracellular portion of CD45 is composed of several fibronectin type III domains that are heavily N-glycosylated. The membrane distal portion contains three domains encoded by the alternatively spliced A, B and C exons. All three of these domains are abundantly O-glycosylated; the A and C domain also bear N-glycans, but the B domain contains no N-glycosylation sites.


Figure 1
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. Structure and glycosylation of CD45. (A) CD45 is a highly glycosylated transmembrane protein tyrosine phosphatase. Multiple isoforms of CD45 exist as a result of differential utilization of the A, B and C exons, indicated by the letters A, B and C. The membrane proximal region contains multiple N-glycans, while the membrane distal region contain multiple N-glycans and abundant O-glycans. CZ-1 recognizes a sialic acid-dependent epitope on the B domain of CD45. (B) Sialic acid containing glycan structures on CD45 include both N- and O-glycans. Core 1 O-glycans can be modified by both {alpha}2,3 and {alpha}2,6 sialic acid, while core 2 O-glycans can be modified by {alpha}2,3 sialic acid at the termini of both branches. The termini of N-glycans can be modified by either {alpha}2,3 or {alpha}2,6 sialic acid.

 
Both O- and N-glycans can be modified by addition of various sugars (Fig. 1B). O-glycans can be short core 1 structures or branched core 2 structures, and core 1 and core 2 O-glycans can be terminated with sialic acid residues. N-glycans are often highly branched structures that can also be terminated with sialic acid residues. Sialic acid residues decorating the termini of N- and O-glycans can be added in {alpha}2,3 or {alpha}2,6 linkages by a large family of sialyltransferase enzymes (43). As expression and activity of different sialyltransferases are modulated during T-cell development and activation (2, 9, 12 ,22, 23, 27), glycan structures are dynamic and reflect the functional state of the cell.

As CZ-1 binding was reported to identify a unique subset of activated CD4 cells, we wished to characterize the glycan epitope recognized by CZ-1. T-cell activation results in increased expression of core 2 O-glycans (44), suggesting that increased CZ-1 binding may be related to increased expression of core 2 O-glycans on these cells. We screened a panel of murine T-cell lines for expression of CD45RB, core 2 O-glycans and CZ-1 (Fig. 2A). The BW5147 T-cell line is CD4–CD8–, expresses CD45RB, and only expresses core 1 O-glycans (40; Fig. 1). We observed robust CZ-1 binding to BW5147 cells, indicating that core 2 O-glycans are not required for CZ-1 recognition. Similarly, we observed binding of CZ-1 to EL-4 T cells that, like BW5147 cells, are also CD4– CD8– and do not express core 2 O-glycans. The PHAR cell line is a lectin-selected derivative of BW5147 that expresses a specific glycosyltransferase that creates core 2 O-glycans on cell-surface glycoproteins (40). As PHAR cells are also CZ-1+, this demonstrates that core 2 O-glycan branching does not block CZ-1 binding. We confirmed that CD45 expression is required for CZ-1 binding using the CD45– derivative of BW5147 (Fig. 2B). We observed no CZ-1 binding to CD45– cells. CD45– cells transfected with cDNA-encoding CD45R0 or CD45RC likewise did not bind CZ-1. In contrast, cells transfected with CD45RBC bound both anti-CD45RB and CZ-1, confirming previous studies demonstrating that the CZ-1 epitope is CD45RB dependent (38).


Figure 2
View larger version (26K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. CZ-1 recognizes a core 2 O-glycan-independent epitope on CD45RB. (A) A panel of T-cell lines that either express or lack core 2 O-glycans, as detected with the 1B11 antibody, were screened for expression of CD45RB and the CZ-1 epitope. Core 2 O-glycan-deficient BW5147 (BW) and EL-4 cells expressed CD45RB (left panels, shaded histogram) and bound CZ-1 (right panels, shaded histogram). In addition, CZ-1 also bound to the core 2 O-glycan-expressing derivative of BW, PHAR, demonstrating that CZ-1 epitope expression is independent of O-glycan branching. Binding of respective isotype control antibodies are shown with dashed lines. (B) CZ-1 epitope expression is restricted to CD45RB-positive cells. CZ-1 did not bind to a CD45-deficient derivative of BW (CD45–). Similarly, CZ-1 and anti-CD45RB did not bind to cells that express CD45R0 or CD45RC. CD45 expression was verified in these cell lines by detection with a pan-specific CD45 antibody. In contrast, CD45–cells expressing CD45 that contains the B and C exons bound both anti-CD45RB and CZ-1, confirming that the CZ-1 epitope is restricted to CD45RB-positive cells. Staining is representative of at least two experiments for each cell line.

 
Welsh et al. originally observed that CZ-1 binding was also sialic acid dependent because treatment with a non-specific neuraminidase eliminated CZ-1 reactivity (34). N- and O-glycans on CD45 can be modified with {alpha}2,3 and/or {alpha}2,6-linked sialic acid (Fig. 1). To determine if CZ-1 binding is dependent on {alpha}2,3-linked or {alpha}2,6-linked sialic acid, EL-4 cells were treated with an {alpha}2,3-specific neuramindase from leech (Fig. 3A). Removal of {alpha}2,3-linked sialic acid from core 1 O-glycans was confirmed by increased binding of the lectin PNA to neuramindase-treated cells, and the {alpha}2,3-specific neuraminidase did not affect levels of {alpha}2,6-linked sialic acid on the cells as assessed by binding of the lectin SNA (data not shown). As shown in Fig. 3(A), CZ-1 binding to EL-4 cells was dramatically reduced following treatment with the {alpha}2,3-specific neuraminidase, demonstrating that {alpha}2,3-linked sialic acid is required for CZ-1 binding.


Figure 3
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. CZ-1 recognizes {alpha}2,3 sialylated core 1 O-glycans. (A) EL-4 cells were either treated with a neuraminidase (Nase) specific for {alpha}2,3 sialic acid from Macrobdella leech or mock treated and analyzed by flow cytometry for expression of CZ-1 (shaded histogram). Staining with isotype control antibody is depicted by a dashed histogram. Treatment with {alpha}2,3 neuraminidase dramatically reduced CZ-1 binding, demonstrating that the antibody is specific for {alpha}2,3 sialic acid. Similar results were obtained with BW5147 (BW) cells (data not shown). (B) EL-4 cells were grown in the presence of vehicle alone (mock), benzyl-{alpha}-GalNAc to inhibit O-glycan elongation, or DMNJ to inhibit N-glycan elongation and analyzed for expression of CZ-1 and CD45RB by flow cytometry (shaded histogram). Staining with isotype control antibodies is depicted with a dashed line. Inhibition of O-glycan, but not N-glycan, elongation dramatically reduced CZ-1 binding, demonstrating that CZ-1 is an O-glycan-dependent epitope. Expression of CD45RB was not affected by treatment with inhibitors. Similar results were obtained with BW and PHARcells.

 
Both N- and O-glycans on CD45 can bear glycans terminated with {alpha}2,3-linked sialic acid. Based on the structure of CD45 (Fig. 1), we thought it unlikely that CZ-1 recognizes an epitope containing sialic acid on an N-glycan as exon B of CD45 does not possess any sites for N-glycosylation. Thus, we predicted that CZ-1 would recognize an {alpha}2,3-linked sialic acid-dependent epitope on O-glycans; however, given the large size of N-glycans, it was formally possible that CZ-1 recognizes an epitope on the B domain that is partly determined by an N-glycan on another portion of CD45. To determine the requirement for O- versus N-glycans in creating the CZ-1 epitope, EL-4 cells were treated with benzyl-{alpha}-GalNAc to inhibit O-glycan elongation or DMNJ to inhibit N-glycan elongation (Fig. 3B). Treatment with both inhibitors had no effect on expression of CD45RB. However, cells treated with benzyl-{alpha}-GalNAc showed a dramatic decrease in CZ-1 epitope expression. In contrast, inhibition of N-glycan elongation did not decrease CZ-1 binding; in fact, DMNJ treatment actually resulted in increased CZ-1 binding. Thus, the CZ-1 epitope requires O-glycans on CD45. Not only are N-glycans dispensable for the CZ-1 epitope, the increase in CZ-1 binding that we observed to DMNJ-treated cells suggested bulky N-glycans could reduce accessibility of the CZ-1 epitope. Identical results were observed with PHAR cells treated with benzyl-{alpha}-GalNAc and DMNJ (data not shown). Taken together, the data in Figs 2 and 3 demonstrate that CZ-1 recognizes O-glycans-bearing Galß1,3GalNAc sequences terminated with {alpha}2,3-linked sialic acid on CD45.

CZ-1 expression is up-regulated during T-cell activation
CZ-1 binding was previously shown to increase during T-cell activation in vivo and in vitro (34–36). However, it was not clear whether up-regulation of the CZ-1 epitope during T-cell activation is due to increased expression of the CD45RB polypeptide or to changes in CD45 glycosylation. Importantly, it is well established that T-cell activation is accompanied by general loss of {alpha}2,3-linked sialic acid on O-glycans on CD45, detected by PNA binding to T cells and by structural characterization of isolated O-glycans (27). Thus, increased CZ-1 binding to {alpha}2,3-linked sialic acid on CD45 O-glycans during T-cell activation would indicate that sialic acid residues are added to specific O-glycans on CD45 expressed on activated T cells, while other O-glycans on CD45 lack sialic acid, implying very precise substrate specificity and selective patterns of glycosylation during T-cell activation.

To address this question, murine splenocytes were activated with anti-CD3 and IL-2 and analyzed for expression of CD4, CD8, CZ-1 and CD45RB. As shown in Fig. 4(A), CZ-1 binding to CD4+ cells increased with time following anti-CD3 stimulation. Simultaneously, PNA binding to these cells also increased, confirming the overall loss of sialic acid from core 1 O-glycans following T-cell activation. The pattern of CD45RB expression did not significantly change over the 72 h following T-cell activation. Thus, the increase in CZ-1 binding to CD4 T cells after activation did not result from increased CD45RB expression but rather from altered glycosylation of CD45RB.


Figure 4
View larger version (45K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Expression of the {alpha}2,3 sialylated CZ-1 epitope is paradoxically increased on activated T cells. Pooled splenocytes from two mice were activated with anti-CD3 and IL-2 and binding of PNA, CZ-1 and CD45RB were analyzed on CD4+cells at various time points up to 72 h (A). Activation resulted in increased binding of PNA, demonstrating a loss of {alpha}2,3 sialic acid from core 1 O-glycans on CD45. Paradoxically, expression of the CZ-1 epitope remained constant or increased over this same time period. CD45RB expression remained relatively constant throughout this time course. Control staining of viable lymphocytes, gated by forward versus side scatter, for each reagent on 72 h samples is shown. (B) Dual color analysis of CZ-1 and PNA binding to CD4+cells demonstrates that activated cells are both CZ-1 and PNA bright. The percentages of cells that are both CZ-1 and PNA bright are indicated by the boxed gate. Similarly, dual color analysis reveals that CZ-1 bright cells are CD45RB positive.

 
As mentioned above, increased PNA binding on activated CD4 and CD8 T cells (19, 21) results primarily from increased exposure of asialo core 1 O-glycans on CD45 (27), yet CZ-1 binding increases on activated CD4 cells. To ask if simultaneous loss of sialic acid from some CD45 glycans and increased CZ-1 binding occurs on activated CD4 cells, we directly compared PNA and CZ-1 binding to activated CD4 T cells. Splenocytes were activated with anti-CD3 and IL-2, and CD4 cells were analyzed at various time points for PNA and CZ-1 binding (Fig. 4B). CZ-1 epitope expression was highest on a sub-population of CD4+ cells that were also CD45RB bright (Fig. 4B). The mean fluorescence intensity of PNA binding increased from 22 on unstimulated CD4 splenocytes to 115 after 72 h of stimulation. Although core 1 O-glycan sialylation generally decreased upon T-cell activation, as detected by PNA binding, expression of the sialylated CZ-1 epitope increased. By 72 h after activation, we detected a distinct CD4 population that was both PNA+ and CZ-1+ that comprised 32% of the total CD4 T cells (Fig. 4B). Thus, these T cells displayed CD45 that had lost sialic acid from many core 1 O-glycans but had increased sialylation of the core 1 O-glycan epitope recognized by CZ-1. These results directly demonstrated that changes in glycan structure upon T-cell activation are microheterogeneous and may be specific to individual glycans on a single glycoprotein.

The CZ-1 epitope is synthesized by ST3Gal I
Several sialyltransferases can synthesize the {alpha}2,3 sialic acid linkage in the CZ-1 epitope, including ST3Gal I, ST3Gal II and ST3Gal IV (27, 41, 45). In contrast, ST3Gal III does not utilize core 1 O-glycans as a substrate and therefore was not considered as a candidate (45). Decreased {alpha}2,3 sialylation of Galß1,3GalNAc sequences on CD45 following T-cell activation is primarily due to decreased ST3Gal I activity (27). Increased expression of the CZ-1 epitope on activated CD4 T cells may therefore result from sialylation by ST3Gal II or ST3Gal IV following T-cell activation or alternatively may result from preferential sialylation by ST3Gal I despite overall decreased activity.

To identify the sialyltransferases responsible for creating the CZ-1 epitope and distinguish between these two mechanisms, we assessed CZ-1 binding to CD4 T cells from mice lacking ST3Gal I, ST3Gal II or ST3Gal IV (Fig. 5) (9, 41, 42). As expected, CZ-1 expression was low on resting CD4 T cells from wild type and all sialyltransferase-null mice. After stimulation of splenocytes with anti-CD3 and IL-2 for 72 h, CZ-1 binding to CD4 T cells from ST3Gal II–/– and ST3Gal IV–/– mice was dramatically increased, comparable to the increase seen for CD4 cells from age-matched wild-type controls. In contrast, there was minimal increase in CZ-1 binding to activated CD4 T cells from ST3Gal I–/– mice. Thus, the ST3Gal I enzyme is required for the high level of expression of the CZ-1 epitope on CD45RB during CD4 T-cell activation despite the overall decrease in ST3Gal I activity in these cells. Moreover, other sialyltransferases that add {alpha}2,3-linked sialic acid to O-glycans appear to only minimally contribute to synthesis of the CZ-1 epitope. This observation demonstrates preferential acceptor substrate specificity of ST3Gal I for this unique O-glycan on CD45.


Figure 5
View larger version (35K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. The ST3Gal I sialyltransferase is required for synthesis of the CZ-1 epitope. Splenocytes from ST3Gal I–/–, ST3Gal II–/–, ST3Gal IV–/–or age-matched wild-type controls were stimulated with anti-CD3 and IL-2 for 72 h. CZ-1 expression was analyzed on CD4+cells prior to and following stimulation. Control staining of viable lymphocytes, gated by forward versus side scatter, for activated samples is shown. CZ-1 expression increased on ST3Gal II–/–and ST3Gal IV–/–CD4+cells following stimulation, but not ST3Gal I–/–CD4+cells, indicating that ST3Gal I is required for creating the CZ-1 epitope. Results are representative for two individual knockout and control mice assayed for each genotype.

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
References
 
Profound changes in lymphocyte glycan structures were first appreciated using plant lectins over four decades ago. Plant lectins have been used to distinguish immature and mature thymocyte subsets (46, 47) and naive, effector and memory subsets of peripheral T cells (21, 48), as well as to isolate hematopoietic and neural stem cells (49, 50). Changes in glycan structure are determined, at least in part, by the pattern of expression of specific glycosyltransferases and glycosidases in distinct lymphocyte sub-populations (2, 9, 12, 21, 22, 51). For example, expression of the ST3Gal I sialyltransferase in medullary, but not cortical, thymocytes adds {alpha}2,3-linked sialic acid to Galß1,3GalNAc-Ser/Thr sequences, blocking PNA binding to medullary thymocytes (9, 22). Similarly, expression of the core 2 N-acetylglucosaminyltransferase I in cortical, but not medullary, thymocytes creates core 2 O-glycans on cortical thymocytes recognized by the 1B11 mAb (52).

Until recently, changes in glycan structure that were observed using plant lectins such as PNA were assumed to affect all cell-surface glycans as cells express many glycoproteins that are potential substrates for glycosyltransferases and glycosidases. However, it is becoming increasingly clear that changes in glycan structure are not universal to all glycoproteins on a cell. For example, binding of SNA, that recognizes SA{alpha}2,6Gal sequences, to medullary thymocytes appears to be due to ST6Gal I modification of a single T-cell glycoprotein, CD45 (1, 18). On medullary thymocytes, SNA binds the CD45RA isoform. However, while both medullary and cortical thymocytes express similar levels of ST6Gal I mRNA and protein, medullary thymocytes express the CD45RA isoform, while cortical thymocytes express the CD45R0 isoform, implying that CD45RA on medullary thymocytes is a preferred acceptor substrate for ST6Gal I (18). These observations indicate that determination of glycosyltransferase RNA and protein levels may not be sufficient to predict the repertoire of glycan structures that will be made by a cell, and that expression of glycan structures may also depend on expression of specific glycoprotein acceptor substrates.

Recent data have shown that T-cell activation results in the loss of {alpha}2,3 sialic acid specifically from O-glycans on CD45 (27). In contrast, we have observed increased {alpha}2,3 sialylation of one or more O-glycans on the CD45RB domain detected by the CZ-1 mAb. These observations suggest that changes in cellular glycosylation are not only heterogenous among different glycoproteins but are also microheterogeneous among individual glycans on a single glycoprotein. Microheterogeneous changes in glycan structures may be achieved through several mechanisms. Glycosyltransferases may exhibit hierarchal or absolute acceptor substrate specificity for individual glycan chains on a single glycoprotein. Hierarchal acceptor substrate specificity could determine glycan structure in instances where glycosyltransferase activity is limiting. For example, relatively low expression and activity of ST3Gal I in activated T cells (27) may result in preferred substrate sites on specific O-glycans, such as the CZ-1 epitopes, being sialylated by this enzyme, while less optimal substrate sites would remain non-sialylated and would bind PNA. Although multiple glycosyltransferases may synthesize the same glycan structures, the enzymes have clear glycoprotein substrate preferences as we observed a dramatic loss of CZ-1 epitope expression on CD4 T cells from ST3Gal I–/– mice but not on CD4 T cells from ST3Gal II–/– or ST3Gal IV–/– mice (Fig. 5). To fully understand the mechanisms underlying glycan microheterogeneity on T-cell subsets will require an understanding of factors determining glycosyltransferase substrate specificity as well as a comprehensive survey of glycosyltransferase expression patterns in various cell types.

Although glycan structures have been considered to have a composite effect on glycoprotein function, recent evidence demonstrates that glycoprotein function may be profoundly affected by the structure of a single glycan chain. For example, multiple O-glycans on P-selectin glycoprotein ligand-1 (PSGL-1) are fucosylated and therefore potential ligands for P-selectin (53). However, elimination of a single membrane distal O-glycosylation site on PSGL-1 profoundly inhibited P-selectin binding and P-selectin-mediated rolling adhesion (54, 55). Similarly, addition of a single sialic acid on an O-glycan in the stalk region of CD8ß does not affect protein structure but regulates CD8 interaction with MHC class I (3, 7, 56, 57). By affecting MHC class I interactions with CD8, a single sialic acid may thus significantly affect TCR cell signaling involved in thymocyte positive and negative selection.

The effects of distinct and specific glycan addition to CD45 are not yet known. However, CD45 tyrosine phosphatase activity is known to be regulated by O-glycosylation and sialylation (28), so that specific glycans may regulate the ability of CD45 to participate in T-cell activation via phosphatase-mediated activation of CD45-associated kinases. Similarly, CD45 localizes to specific membrane domains during TCR signaling, and specific glycans may regulate CD45 re-distribution to these membrane domains (29, 58).

As it becomes increasingly clear that unique glycan structures at individual glycosylation sites can profoundly affect glycoprotein function, it is critical that we understand the factors that control microheterogeneity of glycosylation. The biological effects of another type of post-translational modification, phosphorylation, are widely accepted to be substrate and site specific, and studies defining roles for protein phosphorylation in cellular activation have utilized ‘phospho-specific’ antibodies that recognize phosphorylation at specific sites on unique proteins. Similar strategies may be used with ‘glycan-specific’ antibodies that will allow us to understand the functions of specific glycans at specific sites on unique proteins during T-cell development and T-cell effector function in the periphery.


   Acknowledgements
 
We thank Mark Boton for assistance with the mouse experiments. This work was supported by GM63281 (to L.G.B) and HL57345 (to J.D.M.) from the National Institutes of Health. J.D.M. is an investigator for the Howard Hughes Medical Institute. J.D.H. was supported in part by National Institutes of Health Research Service Award Pre-doctoral Fellowships T32 GM08042, CA009120, AI52031, by a University of California Dissertation Year Fellowship and by the University of California Los Angeles Aesculapians. The Jonsson Comprehensive Cancer Center flow cytometry facility was supported by grants CA 16042 and AI28697 from the National Institutes of Health.


   Abbreviations
 
DMNJ, deoxymannojirimycin
PHA, phytohemagglutinin
PNA, peanut agglutinin
PSGL-1, P-selectin glycoprotein-1
SNA, Sambucus nigra agglutinin

   Notes
 
Transmitting editor: T. F. Tedder

Received 4 January 2007, accepted 10 April 2007.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 

  1. Amano M, Galvan M, He J, Baum LG. The ST6Gal I sialyltransferase selectively modifies N-glycans on CD45 to negatively regulate galectin-1-induced CD45 clustering, phosphatase modulation, and T cell death. J. Biol. Chem. (2003) 278:7469.[Abstract/Free Full Text]
  2. Blander JM, Visintin I, Janeway CA Jr, Medzhitov R. Alpha(1,3)-fucosyltransferase VII and alpha(2,3)-sialyltransferase IV are up-regulated in activated CD4 T cells and maintained after their differentiation into Th1 and migration into inflammatory sites. J. Immunol. (1999) 163:3746.[Abstract/Free Full Text]
  3. Daniels MA, Devine L, Miller JD, et al. CD8 binding to MHC class I molecules is influenced by T cell maturation and glycosylation. Immunity (2001) 15:1051.[CrossRef][Web of Science][Medline]
  4. Demetriou M, Granovsky M, Quaggin S, Dennis JW. Negative regulation of T-cell activation and autoimmunity by Mgat5 N-glycosylation. Nature (2001) 409:733.[CrossRef][Medline]
  5. Lowe JB. Glycosylation, immunity, and autoimmunity. Cell (2001) 104:809.[CrossRef][Web of Science][Medline]
  6. Mekala DJ, Geiger TL. Functional segregation of the TCR and antigen-MHC complexes on the surface of CTL. J. Immunol. (2003) 171:4089.[Abstract/Free Full Text]
  7. Moody AM, Chui D, Reche PA, Priatel JJ, Marth JD, Reinherz EL. Developmentally regulated glycosylation of the CD8alphabeta coreceptor stalk modulates ligand binding. Cell (2001) 107:501.[CrossRef][Web of Science][Medline]
  8. Morgan R, Gao G, Pawling J, Dennis JW, Demetriou M, Li B. N-acetylglucosaminyltransferase V (Mgat5)-mediated N-glycosylation negatively regulates Th1 cytokine production by T cells. J. Immunol. (2004) 173:7200.[Abstract/Free Full Text]
  9. Priatel JJ, Chui D, Hiraoka N, et al. The ST3Gal-I sialyltransferase controls CD8+ T lymphocyte homeostasis by modulating O-glycan biosynthesis. Immunity (2000) 12:273.[CrossRef][Web of Science][Medline]
  10. Smithson G, Rogers CE, Smith PL, et al. Fuc-TVII is required for T helper 1 and T cytotoxic 1 lymphocyte selectin ligand expression and recruitment in inflammation, and together with Fuc-TIV regulates naive T cell trafficking to lymph nodes. J. Exp. Med. (2001) 194:601.[Abstract/Free Full Text]
  11. Starr TK, Daniels MA, Lucido MM, Jameson SC, Hogquist KA. Thymocyte sensitivity and supramolecular activation cluster formation are developmentally regulated: a partial role for sialylation. J. Immunol. (2003) 171:4512.[Abstract/Free Full Text]
  12. Comelli EM, Sutton-Smith M, Yan Q, et al. Activation of murine CD4+ and CD8+ T lymphocytes leads to dramatic remodeling of N-linked glycans. J. Immunol. (2006) 177:2431.[Abstract/Free Full Text]
  13. Koch U, Lacombe TA, Holland D, et al. Subversion of the T/B lineage decision in the thymus by lunatic fringe-mediated inhibition of Notch-1. Immunity (2001) 15:225.[CrossRef][Web of Science][Medline]
  14. Lemaire S, Derappe C, Michalski JC, Aubery M, Neel D. Expression of beta 1-6-branched N-linked oligosaccharides is associated with activation in human T4 and T8 cell populations. J. Biol. Chem. (1994) 269:8069.[Abstract/Free Full Text]
  15. Piller F, Piller V, Fox RI, Fukuda M. Human T-lymphocyte activation is associated with changes in O-glycan biosynthesis. J. Biol. Chem. (1988) 263:15146.[Abstract/Free Full Text]
  16. Harrington LE, Galvan M, Baum LG, Altman JD, Ahmed R. Differentiating between memory and effector CD8 T cells by altered expression of cell surface O-glycans. J. Exp. Med. (2000) 191:1241.[Abstract/Free Full Text]
  17. Baum LG, Pang M, Perillo NL, et al. Human thymic epithelial cells express an endogenous lectin, galectin-1, which binds to core 2 O-glycans on thymocytes and T lymphoblastoid cells. J. Exp. Med. (1995) 181:877.[Abstract/Free Full Text]
  18. Baum LG, Derbin K, Perillo NL, Wu T, Pang M, Uittenbogaart C. Characterization of terminal sialic acid linkages on human thymocytes. Correlation between lectin-binding phenotype and sialyltransferase expression. J. Biol. Chem. (1996) 271:10793.[Abstract/Free Full Text]
  19. Chervenak R, Cohen JJ. Peanut lectin binding as a marker for activated T-lineage lymphocytes. Thymus (1982) 4:61.[Web of Science][Medline]
  20. Daniels MA, Hogquist KA, Jameson SC. Sweet ‘n’ sour: the impact of differential glycosylation on T cell responses. Nat. Immunol. (2002) 3:903.[CrossRef][Web of Science][Medline]
  21. Galvan M, Murali-Krishna K, Ming LL, Baum L, Ahmed R. Alterations in cell surface carbohydrates on T cells from virally infected mice can distinguish effector/memory CD8+ T cells from naive cells. J. Immunol. (1998) 161:641.[Abstract/Free Full Text]
  22. Gillespie W, Paulson JC, Kelm S, Pang M, Baum LG. Regulation of alpha 2,3-sialyltransferase expression correlates with conversion of peanut agglutinin (PNA)+ to PNA- phenotype in developing thymocytes. J. Biol. Chem. (1993) 268:3801.[Abstract/Free Full Text]
  23. Kaufmann M, Blaser C, Takashima S, Schwartz-Albiez R, Tsuji S, Pircher H. Identification of an alpha2,6-sialyltransferase induced early after lymphocyte activation. Int. Immunol. (1999) 11:731.[Abstract/Free Full Text]
  24. Sharon N. Lectin receptors as lymphocyte surface markers. Adv. Immunol. (1983) 34:213.[Web of Science][Medline]
  25. Sharon N, Lis H. History of lectins: from hemagglutinins to biological recognition molecules. Glycobiology (2004) 14:53R.[Abstract/Free Full Text]
  26. Wu W, Harley PH, Punt JA, Sharrow SO, Kearse KP. Identification of CD8 as a peanut agglutinin (PNA) receptor molecule on immature thymocytes. J. Exp. Med. (1996) 184:759.[Abstract/Free Full Text]
  27. Amado M, Yan Q, Comelli EM, Collins BE, Paulson JC. Peanut agglutinin high phenotype of activated CD8+ T cells results from de novo synthesis of CD45 glycans. J. Biol. Chem. (2004) 279:36689.[Abstract/Free Full Text]
  28. Xu Z, Weiss A. Negative regulation of CD45 by differential homodimerization of the alternatively spliced isoforms. Nat. Immunol. (2002) 3:764.[CrossRef][Web of Science][Medline]
  29. Hermiston ML, Xu Z, Weiss A. CD45: a critical regulator of signaling thresholds in immune cells. Annu. Rev. Immunol. (2003) 21:107.[CrossRef][Web of Science][Medline]
  30. Smith PL, Lowe JB. Molecular cloning of a murine N-acetylgalactosamine transferase cDNA that determines expression of the T lymphocyte-specific CT oligosaccharide differentiation antigen. J. Biol. Chem. (1994) 269:15162.[Abstract/Free Full Text]
  31. Ellies LG, Tsuboi S, Petryniak B, Lowe JB, Fukuda M, Marth JD. Core 2 oligosaccharide biosynthesis distinguishes between selectin ligands essential for leukocyte homing and inflammation. Immunity (1998) 9:881.[CrossRef][Web of Science][Medline]
  32. Fuhlbrigge RC, Kieffer JD, Armerding D, Kupper TS. Cutaneous lymphocyte antigen is a specialized form of PSGL-1 expressed on skin-homing T cells. Nature (1997) 389:978.[CrossRef][Medline]
  33. Brennan PJ, Saouaf SJ, Van Dyken S, et al. Sialylation regulates peripheral tolerance in CD4+ T cells. Int. Immunol. (2006) 18:627.[Abstract/Free Full Text]
  34. Brutkiewicz RR, O'Donnell CL, Maciaszek JW, Welsh RM, Vargas-Cortes M. The monoclonal antibody CZ-1 identifies a mouse CD45-associated epitope expressed on interleukin-2-responsive cells. Eur. J. Immunol. (1993) 23:2427.[Web of Science][Medline]
  35. Varga SM, Welsh RM. The CD45RB-associated epitope defined by monoclonal antibody CZ-1 is an activation and memory marker for mouse CD4 T cells. Cell Immunol. (1996) 167:56.[CrossRef][Web of Science][Medline]
  36. Vargas-Cortes M, O'Donnell CL, Appel MC, Maciaszek JW, Yurkunas KS, Welsh RM. A lymphocyte differentiation and activation antigen, CZ-1, that distinguishes between CD8+ and unstimulated CD4+ T lymphocytes. Eur. J. Immunol. (1992) 22:1043.[Web of Science][Medline]
  37. Nguyen JT, Evans DP, Galvan M, et al. CD45 modulates galectin-1-induced T cell death: regulation by expression of core 2 O-glycans. J. Immunol. (2001) 167:5697.[Abstract/Free Full Text]
  38. Novak TJ, Farber D, Leitenberg D, Hong SC, Johnson P, Bottomly K. Isoforms of the transmembrane tyrosine phosphatase CD45 differentially affect T cell recognition. Immunity (1994) 1:109.[CrossRef][Web of Science][Medline]
  39. Cummings RD, Trowbridge IS, Kornfeld S. A mouse lymphoma cell line resistant to the leukoagglutinating lectin from Phaseolus vulgaris is deficient in UDP-GlcNAc: alpha-D-mannoside beta 1,6 N-acetylglucosaminyltransferase. J. Biol. Chem. (1982) 257:13421.[Free Full Text]
  40. Galvan M, Tsuboi S, Fukuda M, Baum LG. Expression of a specific glycosyltransferase enzyme regulates T cell death mediated by galectin-1. J. Biol. Chem. (2000) 275:16730.[Abstract/Free Full Text]
  41. Ellies LG, Sperandio M, Underhill GH, et al. Sialyltransferase specificity in selectin ligand formation. Blood (2002) 100:3618.[Abstract/Free Full Text]
  42. Ellies LG, Ditto D, Levy GG, et al. Sialyltransferase ST3Gal-IV operates as a dominant modifier of hemostasis by concealing asialoglycoprotein receptor ligands. Proc. Natl Acad. Sci. USA (2002) 99:10042.[Abstract/Free Full Text]
  43. Harduin-Lepers A, Vallejo-Ruiz V, Krzewinski-Recchi MA, Samyn-Petit B, Julien S, Delannoy P. The human sialyltransferase family. Biochimie (2001) 83:727.[Medline]
  44. Jones AT, Federsppiel B, Ellies LG, et al. Characterization of the activation-associated isoform of CD43 on murine T lymphocytes. J. Immunol. (1994) 153:3426.[Abstract]
  45. Kono M, Ohyama Y, Lee YC, Hamamoto T, Kojima N, Tsuji S. Mouse beta-galactoside alpha 2,3-sialyltransferases: comparison of in vitro substrate specificities and tissue specific expression. Glycobiology (1997) 7:469.[Abstract/Free Full Text]
  46. Reisner Y, Linker-Israeli M, Sharon N. Separation of mouse thymocytes into two subpopulations by the use of peanut agglutinin. Cell Immunol. (1976) 25:129.[CrossRef][Web of Science][Medline]
  47. Rose ML, Malchiodi F. Binding of peanut lectin to thymic cortex and germinal centres of lymphoid tissue. Immunology (1981) 42:583.[Web of Science][Medline]
  48. Reisner Y, Pahwa S, Chiao JW, Sharon N, Evans RL, Good RA. Separation of antibody helper and antibody suppressor human T cells by using soybean agglutinin. Proc. Natl Acad. Sci. USA (1980) 77:6778.[Abstract/Free Full Text]
  49. Rietze RL, Valcanis H, Brooker GF, Thomas T, Voss AK, Bartlett PF. Purification of a pluripotent neural stem cell from the adult mouse brain. Nature (2001) 412:736.[CrossRef][Medline]
  50. Salner AL, Obbagy JE, Hellman S. Differing stem cell self-renewal of lectin-separated murine bone marrow fractions. J. Natl Cancer Inst. (1982) 68:639.[Web of Science][Medline]
  51. Van Dyken SJ, Green RS, Marth JD. Structural and mechanistic features of protein O-glycosylation linked to CD8+ T Cell apoptosis. Mol. Cell Biol. (2007) 27:1096.[Abstract/Free Full Text]
  52. Ellies LG, Tao W, Fellinger W, Teh HS, Ziltener HJ. The CD43 130-kD peripheral T-cell activation antigen is downregulated in thymic positive selection. Blood (1996) 88:1725.[Abstract/Free Full Text]
  53. Wilkins PP, McEver RP, Cummings RD. Structures of the O-glycans on P-selectin glycoprotein ligand-1 from HL-60 cells. J. Biol. Chem. (1996) 271:18732.[Abstract/Free Full Text]
  54. Xia L, Ramachandran V, McDaniel JM, Nguyen KN, Cummings RD, McEver RP. N-terminal residues in murine P-selectin glycoprotein ligand-1 required for binding to murine P-selectin. Blood (2003) 101:552.[Abstract/Free Full Text]
  55. Liu W, Ramachandran V, Kang J, Kishimoto TK, Cummings RD, McEver RP. Identification of N-terminal residues on P-selectin glycoprotein ligand-1 required for binding to P-selectin. J. Biol. Chem. (1998) 273:7078.[Abstract/Free Full Text]
  56. Moody AM, North SJ, Reinhold B, et al. Sialic acid capping of CD8beta core 1-O-glycans controls thymocyte-major histocompatibility complex class I interaction. J. Biol. Chem. (2003) 278:7240.[Abstract/Free Full Text]
  57. Merry AH, Gilbert RJ, Shore DA, et al. O-glycan sialylation and the structure of the stalk-like region of the T cell co-receptor CD8. J. Biol. Chem. (2003) 278:27119.[Abstract/Free Full Text]
  58. Johnson KG, Bromley SK, Dustin ML, Thomas ML. A supramolecular basis for CD45 tyrosine phosphatase regulation in sustained T cell activation. Proc. Natl Acad. Sci. USA (2000) 97:10138.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
L. A. Earl, S. Bi, and L. G. Baum
N- and O-Glycans Modulate Galectin-1 Binding, CD45 Signaling, and T Cell Death
J. Biol. Chem., January 22, 2010; 285(4): 2232 - 2244.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
S. M. Jackson, N. Harp, D. Patel, J. Wulf, E. D. Spaeth, U. K. Dike, J. A. James, and J. D. Capra
Key developmental transitions in human germinal center B cells are revealed by differential CD45RB expression
Blood, April 23, 2009; 113(17): 3999 - 4007.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
19/7/847    most recent
dxm053v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Hernandez, J. D.
Right arrow Articles by Baum, L. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hernandez, J. D.
Right arrow Articles by Baum, L. G.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?