Published online before print March 3, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Pathology and Infectious Diseases, Royal Veterinary College, Hatfield, United Kingdom;
Institute for Animal Health, Compton, Berkshire, United Kingdom;
Roslin Institute, Midlothian, United Kingdom; and
Centre for Tropical Veterinary Medicine, The Royal (Dick) School of Veterinary Medicine, Easter Bush Veterinary Centre, Roslin, United Kingdom
1Correspondence: Department of Pathology and Infectious Diseases, Royal Veterinary College, Hawkshead Lane, Hatfield, AL9 7TA, UK. E-mail: dwerling{at}rvc.ac.uk
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: dendritic cells • C-type lectin • gp120 • Mycobacterium bovis BCG
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
DC-SIGN, a type II transmembrane protein, is highly expressed on the surface of DC [11 ]. Although only DC were originally found to express DC-SIGN, subsequent studies detected expression on other cells such as decidual macrophages, alveolar macrophages, and B cells [12 13 14 ]. In vitro, DC-SIGN expression is found on immature and mature monocyte-derived DC (MoDC). In addition to its role as a cell adhesion molecule, DC-SIGN also functions as an antigen receptor and has been shown to specifically bind Mycobacterium tuberculosis, envelope proteins of hepatitis C, and Ebola virus, as well as the HIV-1 envelope gp120 [5 6 7 8 9 10 ]. DC-SIGN is the major receptor for gp120 on MoDC, recognizing the high mannose carbohydrate structure on the molecule [9 , 15 , 16 ]. After binding to DC-SIGN, gp120 can be internalized into lysosomes. This is facilitated by two internalization motifs: a dileucine (LL) motif and a tyrosine-based motif (YXXL) in the cytoplasmic domain of DC-SIGN [17 ].
DC-SIGN-related molecules have been identified in other species, including mouse [18 ], macaque and chimpanzee [19 ]. In cattle, only a few CLR have been identified at the mRNA or genome level. For some, such as the bovine mannose receptor (MR), their functional relevance has been confirmed [20 ]. Recently, DEC-205, dectin-1 and dectin-2 have been cloned and functionally characterized [21 22 23 ].
In contrast, no data are available regarding the expression of a bovine ortholog of DC-SIGN. We provide evidence, at the gene and protein level, for the existence of a bovine ortholog to human DC-SIGN expressed on bovine MoDC, which seems to exist in cattle as only one gene, that binds and internalizes HIV-1 gp120 in addition to Mycobacterium bovis bacillus Calmette-Guerin (BCG). Thus, it is likely that this molecule may contribute to the infection of bovine (DC) cells with M. bovis, similar to that described for human DC-SIGN.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Prediction of gene structures, organization, and protein domains
Predictions of gene structures and organization were based on Basic Local Alignment Search Tool (BLAST) hits of genome sequences using peptide or DNA sequences to interrogate the University of California Santa Cruz (UCSC; Santa Cruz, CA, USA) genome browser (http://genome.ucsc.edu/). Protein domains within peptide sequences were predicted by Simple Modular Architecture Research Tool (SMART; Version 5; http://smart.embl-heidelberg.de/) [24
]. A multiple sequence alignment of peptide sequences was made using ClustalX (Version 1.83; http://bips.u-strasbg.fr/fr/Documentation/ClustalX/) [25
] and default parameters. Multiple sequence alignments were displayed using Jalview (Version 2.3; http://www.jalview.org/) [26
].
Calculation of reconciled species and gene trees
Gene trees were determined by the neighbor-joining method [27
]. Trees were displayed using TreeView (Version 1.6.6) and GeneTree (Version 1.3.0; http://taxonomy.zoology.gla.ac.uk/software/index.html) [28
].
Cell preparation and generation of DC
Human and bovine blood was collected from healthy donors in accordance with home office license and ethical committee approval. Blood was collected into sterilized glass bottles containing citrate buffer, and buffy coat was prepared as described recently [29
]. Human and bovine PBMC were separated by density gradient centrifugation (Histopaque 1.077 g ml–1 and 1.083 g ml–1, respectively; Sigma, Poole, Dorset, UK) as described [20
]. Monocytes, B cells, and CD4+ T cells were isolated using the MidiMACS system (Miltenyi Biotec, Ltd., Bisley, Surrey, UK) using anti-CD14 microbeads (Miltenyi Biotec, Ltd.) or mAb specific for CD21 [CC21, Royal Veterinary College (RVC), University of London, UK] or CD4 (CC30, RVC), respectively, as described [20
]. Viability was assessed by Trypan blue exclusion (Sigma) and was greater than 95%. The purity of the cells was evaluated by flow cytometry using a FACSAria (BD Biosciences, Oxford, UK) and shown to be >98%.
Bovine and human MoDC were generated as described using 1 ml recombinant bovine (rbo)IL-4 (based on the dose inducing a half-maximal response in a B cell proliferation assay) and 333 µl rboGM-CSF [20 ] or 50 ng ml–1 recombinant human (rhu)IL-4 and 50 ng ml–1 rhuGM-CSF (both supplied by Biosource, Nivelles, Belgium). Medium was changed every 3 days, and cells were harvested after 6–7 days in culture. All incubations were at 37°C in 5% CO2/air unless otherwise stated.
RNA extraction and cDNA synthesis
Total RNA was prepared from bovine MoDC, B cells, and CD4+ T cells following the Qiagen RNeasy mini kit protocol (Qiagen, Crawley, UK). The quality and quantity of the total RNA were determined using a spectrophotometer. RNA (500 ng) was used in the subsequent cDNA reactions using Superscript II RT (Invitrogen, Paisley, UK) and following the manufacturer’s protocol. cDNA was diluted 1:2.5 prior to use.
PCR and sequencing
Primers (MWG Biotech, Ebersberg, Germany) were designed to the Btau_CD209L sequence to amplify the full-length coding sequence (CDS) as well as an internal fragment (INT). The primers CDS-forward (FOR; 5'-GCATGGCAGAGATGTATGAACACAAGG-3') and CDS-reverse (REV; 5'-TCAGAGGACTGGGCAGGGAA-3') amplified the predicted CDS of 778 bp. Primers INT-FOR (5'-GGAGCCAGATGACTCTGAGG-3') and INT-REV (5'-CCCAATTCTGAGGACAAGGA-3') amplified an INT of 383 bp, representing bases 24–406 of the CDS. These primers where then used in a standard PCR at an annealing temperature of 58°C using cDNA from bovine MoDC, B cells, and CD4+ T cells as template DNA (5 µl). The amplified products were visualized on a 1% agarose gel. Cleaned products were then used in sequencing reactions.
Antibodies and flow cytometry
Human and bovine monocytes and MoDC were analyzed for the expression of the cell surface molecules. The sources of antibodies, their isotypes, and methods for flow cytometry have been described in detail [30
]. The following mouse anti-bovine mAb were used: CD14 (CCG33, IgG1), CD40 (IL-A158, IgG1), MHC II-DQ (CC158, IgG2a), MHC II-DR (CC108, IgG1), CD80 (N32/52-3, IgG1), and CD86 (IL-A190, IgG1). For human cells, the following directly labeled mouse anti-human mAb were used: CD14-FITC, CD40-FITC, HLA class II-DR, -DP, and -DQ (all IgG2a), CD80-FITC, and CD86-FITC (both IgG1; all mAb from Serotec Ltd., Oxford, UK). To analyze surface expression of DC-SIGN in both species, a polyclonal anti-human DC-SIGN antibody (pAb; C20, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) was used. Where necessary, bound mAb was detected using FITC-labeled, isotype-specific secondary reagents (Southern Biotechnology, Birmingham, AL, USA). Immunofluorescent staining of 10,000-gated, live cells was analyzed using FlowJo software, Version 7.1 (Tree Star, Inc., Ashland, OR, USA).
M. bovis BCG expressing GFP (BCG-GFP)
BCG-GFP was cultured using methods described previously [31
]. Mycobacterial stocks were frozen and stored at –80°C. Stocks were enumerated by serial dilution and spotting onto 7H10 plates.
Western blot analysis for DC-SIGN
Human and bovine MoDC, T cells, B cells, and monocytes produced as described were pelleted by centrifugation at 300 g for 10 min at 4°C. Thereafter, cells were washed twice with ice-cold PBS and lysed in 50 µl M-PERTM lysis buffer (Pierce Biotechnology, Inc., Rockford, IL, USA) per 106 cells for 20 min at room temperature. After centrifugation at 16,000 g for 15 min at 4°C, protein extracts were collected and frozen at –20°C. Protein concentrations in cell lysates were measured using the Coomassie Plus BradfordTM assay kit (Pierce Biotechnology, Inc.) according to the manufacturer’s instructions. Laemmli buffer (5x) was added, and each sample was boiled for 3 min prior to storage at –20°C. For Western blot analysis, 24 µg sample in 20 µl was loaded per well, and proteins were separated by SDS-PAGE. After electrophoresis, 100 V for 90 min, proteins were transferred to a 0.2-µm nitrocellulose membrane (Amersham Biosciences, Germany), which was subsequently blocked with PBS containing 0.05% Tween 20 and 5% milk powder (PBS-TM) for 90 min at room temperature, followed by incubation with mouse anti-human DC-SIGN mAb (diluted 1:1000, clone DCN46, BD Biosciences) or mouse anti-human β-actin (diluted 1:2000, Sigma) overnight at 4°C. Excess antibody was removed by extensive washing in PBS-TM. The membrane was then incubated with HRP-conjugated anti-mouse IgG (Amersham Biosciences) at a 1:2000 dilution in PBS-TM for 90 min at room temperature. The membranes were washed extensively with PBS-Tween 20, and bands were visualized using the ECL system (Amersham Biosciences) and a Curix 60 processor (Agfa-Gevaert N.V., Mortsel, Belgium).
Binding assay for DC-SIGN ligands
Bovine and human MoDC were harvested and adjusted to a concentration of 3–4 x 105 ml–1. This cell suspension (100 µl) was added to each well of a 96-U-well microtiter plate, which was kept on ice. Binding of FITC-labeled OVA (FITC-OVA) was analyzed as described recently [20
]. Briefly, 4 µl FITC-OVA (125 µg ml–1; Molecular Probes, Inc., Eugene, OR, USA) was added to each well and incubated for 30 min on ice or at 37°C. To assess binding of FITC-labeled gp120 (FITC-gp120), FITC-gp120 strain IIIB (ImmunoDiagnostics, Inc., Woburn, MA, USA), at a concentration of 4 µg ml–1, was added to each well and incubated for 5 min or 30 min on ice or 37°C. To assess the specificity of binding, cells were incubated for 30 min on ice with 40 ng ml–1anti-human CD3 pAb (Sigma), anti-human DC-SIGN pAb (Santa Cruz Biotechnology, Inc.), or anti-DC-SIGN mAb (BD Biosciences) prior to adding FITC-OVA or FITC-gp120. In a second set of experiments, bovine lactoferrin (bLF; Sigma) was used to inhibit the DC-SIGN-gp120 interaction in accordance with a recent publication [32
]. MoDC were preincubated with medium alone or 0.01, 0.1, 1, 10, or 100 µM bLF for 30 min at 37°C, followed by incubation with FITC-gp120 for 30 min at 37°C. After three washes with ice-cold PBS, cells were harvested and analyzed by flow cytometry using a FACSAria. The specificity of gp120 binding was also evaluated using purified bovine CD4+ T cells, and the interaction of gp120 with these cells was analyzed as described above.
As DC-SIGN was also described as a receptor for mycobacteria, MoDC were incubated with BCG-GFP at a multiplicity of infection (MOI) of 10 for 2 h at 37°C or on ice as described above. Cells were washed by low-speed centrifugation (300 g) with ice-cold PBS before determining the percentage of cells that bound BCG-GFP by flow cytometry. To assess the specificity of the reaction, DC were preincubated with bLF or pAb to DC-SIGN as described above.
Confocal microscopy
Bovine and human MoDC were harvested, adjusted to a concentration of 1 x 106 ml–1, and incubated at 37°C with 1 µg ml–1 FITC-gp120 for 10 min, 30 min, or 120 min. Cells were washed twice and fixed for 15 min with 2% paraformaldehyde at room temperature. For the analysis of intracellular expression, cells were incubated for 10 min in PBS/BSA/azide /0.02% saponin (Sigma). Cells were stained with anti-DC-SIGN pAb (C-20, Santa Cruz Biotechnology, Inc.), anti-bovine MHC II mAb (CC158), anti-early endosome antigen-1 (EEA-1; ab2900, Abcam, Cambridge, MA, USA), and Cy3-conjugated cholera toxin (CTX; Quadratech, UK) for 60 min at room temperature. After two washes, cells were incubated for 60 min at room temperature with Cy3-, Cy5-, tetramethyl rhodamine isothiocyanate-, or PE-labeled, isotype-matched secondary antibodies (Amersham-Pharmacia, Piscataway, NJ, USA; Molecular Probes; and Southern Biotechnology, respectively). Slides were mounted using Prolong-Antifade (Molecular Probes) and analyzed using an LSM510 META confocal microscope and LSM 2.5 image analysis software (Carl Zeiss, Inc., Thornwood, NY, USA).
Statistical analysis
Data are presented as mean ± SD, and a representative set of data of at least three repeats is shown. Data were analyzed using GraphPad Prism, Version 4.0 (GraphPad Software Inc., San Diego, CA, USA). Differences between treatment groups were determined by means of a one-way ANOVA, followed by Bonferroni t-test for significant effects. Probability levels used for statistical significance were 95% in all cases.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (22K): [in a new window] |
Figure 1. Neighbor-joining tree and protein domains. (A) Gene tree for CD209-like genes determined by a neighbor-joining method and displayed using GeneTree [27
, 28
]. Predicted gene duplications (•) and gene deletions (bold lines). (B) SMART analysis was used to identify protein domains within CD209 proteins [24
]. As part of the SMART package, several regions were predicted. Transmembrane segments, as predicted by TMHMM2, are indicated by a blue bar; coiled-coil regions as determined by Coils2 are indicated by green bars in the neck domains. Segments of low compositional complexity, as determined by SEG, are indicated by pink areas. Signal peptides as determined by SignalP are indicated by red squares. Intron positions are indicated by vertical lines showing phase and position in amino acids. Btau, Bos taurus; Hsap, Homo sapiens; Mdom, Mus domesticus; Mmus, Mus musculus. CLECT, C-type lectin.
|
 View larger version (22K): [in a new window] |
Figure 2. PCR and Western blot analysis of bovine CD209L expression. (A) cDNA generated from bovine MoDC, B cells, and CD4+ T cells was analyzed by PCR for the presence of bovine CD209L transcripts using full-length and internal primers as described. Specific amplicons were detected in cDNA generated from MoDC (results of two animals are shown) but not T cells or B cells (data not shown). (B) Cytoplasmic extracts generated from human and bovine cells were prepared as described, and Western blot analysis was performed using antibodies to DC-SIGN (clone DCN46; upper panel) and β-actin (lower panel) in the appropriate dilutions.
|
Cell surface expression of DC-SIGN on human and bovine monocytes and MoDC
As the Western blot results suggested the presence of a DC-SIGN-like molecule expressed by bovine MoDC, we assessed DC-SIGN expression on the surface of freshly isolated CD14+ monocytes and MoDC by FACS analysis. As shown in Figure 3
, human and bovine monocytes and MoDC expressed a similar receptor repertoire on their cell surface. Differences observed were that MoDC generated from the human or bovine monocytes (at Day 6 of culture) showed no expression of monocyte marker CD14 and higher expression of the molecules CD40, CD80, CD86, and MHC II. Although surface expression of DC-SIGN was not detected on human and bovine monocytes, strong expression was observed in human MoDC with weaker expression on bovine MoDC.
 View larger version (30K): [in a new window] |
Figure 3. Phenotypic analysis of surface antigens expressed on human and bovine MoDC. Freshly isolated CD14+ monocytes and MoDC were stained with indicated markers and analyzed by flow cytometry. Data are presented as histogram of 10,000 cells. Filled histograms represent antigen staining, and open histograms represent isotype-matched controls. Results are from one representative experiment of at least three repeats.
|
 View larger version (25K): [in a new window] |
Figure 4. Binding of FITC-OVA and FITC-gp120 to human and bovine MoDC, which were generated as described and incubated with FITC-OVA (left columns) or FITC-gp120 (right columns) for 30 min at 37°C in the absence or presence of anti-CD3 antibody or anti-DC-SIGN pAb (A). Open histograms represent untreated cells. Results are from one representative experiment of at least four repeats. Data are expressed as histogram analysis of 10,000 cells. (B) Specificity of gp120 binding to DC-SIGN was analyzed by incubating human and bovine MoDC for 30 min at 37°C in the absence or presence of anti-DC-SIGN pAb or mAb. MFI was converted into percent binding by the following formula: (MFI of gp120 binding in the presence of anti-DC-SIGN antibody) – (MFI background, no gp120)/(MFI of the gp120 binding in the absence of anti-DC-SIGN antibody) – (MFI background, no gp120). A histogram summary of average percentage of gp120 binding from at least four experiments is shown. (C) Influence of bLF on FITC-gp120 binding in human and bovine MoDC, which were preincubated with varying concentrations of bLF for 30 min at 37°C, followed by incubation with FITC-gp120 for 30 min at 37°C. A histogram summary of average percent of gp120 binding from three experiments is shown. *, P < 0.05; ***, P < 0.001.
|
 View larger version (13K): [in a new window] |
Figure 5. DC-SIGN antibody reduces binding of BCG-GFP to bovine MoDC, which were incubated for 2 h on ice or at 37°C with BCG-GFP at a MOI of 10 (A) or preincubated with pAb to DC-SIGN (B) or 10 µM bLF for 30 min before exposure to BCG-GFP for 2 h at 37°C (C). Filled histograms represent BCG-GFP-incubated cells; open histograms represent cells incubated on ice (A) or incubated with pAb to DC-SIGN (B) or bLF (C), respectively. Results are from one representative experiment of three repeats. Data are expressed as histogram analysis of 5000 cells.
|
 View larger version (18K): [in a new window] |
Figure 6. Triple-color confocal fluorescence microscopy performed on bovine MoDC, which were incubated with FITC-gp120 for 0 min (A), 10 min (B), 30 min (C), and 120 min (D) at 37°C. DC-SIGN is visualized with Cy5 (blue), gp120 visualized with FITC (green), and MHC II visualized with Cy3 (red).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Multiple CD209-like sequences were found in opossum (2), mouse (5), and human (2) based on sequence homology and annotation data about the UCSC genome browser (data not shown). Analysis of the cattle genome strongly supports the existence of a single CD209-like gene, as no other CD209-like sequences were detected. This was based on available bovine and nonbovine sequences (further details shown in Supplemental Figs. 1 and 2). Despite the fact that the cow genome is not yet fully annotated, we are convinced that this is not a result of gaps in the bovine genome assembly. This sequence is predicted to be a direct descendent of an ancestral CD209-like gene. In contrast, opossum, mouse and human genomes contain multiple CD209-like genes, the products of gene duplications/gene deletions. Interestingly, only the human genome sequence shows extensive internal repeats. Our analysis suggests that the opossum sequences are the result of recent gene duplication in this lineage and so, are strictly CD209-like. In contrast to other species, the mouse has five highly divergent CD209-like genes and is therefore unlikely to be the product of any recent gene duplications in the rodent lineage. The most likely explanation is that all but CD209b have been deleted during the past and are no longer present in the human lineage. CD209b is most like the two human genes CD209 and CLEC4M, which are potentially the product of recent gene duplications and are therefore paralogs and orthologs to the bovine CD209L molecule described here. The SMART analysis also shows that human CD209 proteins are different in that they both contain a repeat domain, which seems to be unique to the human/primate molecules. Interestingly, the predicted and confirmed bovine CD209L molecule seemed to show the highest similarity to the murine SIGNR7. This molecule has recently been described to be one functional homologue to the human DC-SIGN molecule with similar carbohydrate-binding characteristics [33 ]. The fact that murine SIGNR7 has been described to have a shorter neck region compared with human DC-SIGN fits to our observation that the bovine CD209L molecule possesses a shorter sequence as well as potentially a slightly lower molecular weight.
Phenotypic analysis of surface antigens expressed on bovine and human cell subsets revealed that monocytes express lower levels of the costimulatory molecules CD40, CD80, and CD86 when compared with MoDC from both species, supporting the established phenotype of human monocytes and MoDC [20 ]. In addition, staining with antibodies to DC-SIGN revealed that human and bovine MoDC expressed DC-SIGN, whereas no expression was observed on monocytes from either species. A stronger reactivity of the anti-DC-SIGN antibody was observed with human MoDC than with bovine MoDC. As an antibody generated against the human DC-SIGN molecule was used in this experiment, the less-intense staining observed on bovine MoDC could reflect the level of cross-reactivity of the antibody.
As the initial binding of HIV-1 is mediated by its gp120 envelope protein to DC-SIGN, further investigation was done to show the specific binding of FITC-gp120 by bovine MoDC using human MoDC as controls. Binding of FITC-gp120 to bovine MoDC was DC-SIGN-dependent and was blocked in the presence of antibodies to DC-SIGN or bLF, similar to that described for human MoDC [5 , 32 ]. Although our study does not discriminate between the blocking of the DC-SIGN-gp120 interaction being a result of bLF binding to DC-SIGN or gp120, a recent study performed in the human system suggests that bLF binds to DC-SIGN and not gp120 [32 ]. In contrast, uptake of FITC-OVA via clathrin-coated pits [20 ] was unaffected by antibody treatment. Blocking of gp120 binding was higher in the presence of a pAb compared with that observed with the mAb in human and bovine cells. This observation could be explained by the fact that the mAb has a higher specificity compared with the pAb, recognizing fewer epitopes. A further possibility is that gp120 may bind to receptors other than DC-SIGN expressed on the surface of MoDC, such as CD4 and MR [34 ], and some of these receptors are also present on bovine MoDC [20 ].
Having established that a functional DC-SIGN ortholog is expressed on bovine MoDC, we then evaluated its role in the uptake of M. bovis BCG. DC-SIGN plays an important role in the functional subversion of DC in mycobacterial infection [8 , 35 ]. In addition, given the observation that DC-SIGN is also expressed on alveolar macrophages [36 ] and the observed differences in the responses of bovine cells to different mycobacterial strains [37 ], it is important to understand the function of this receptor in the bovine system. Similar as with gp120, the binding of BCG-GFP to bovine MoDC was reduced in the presence of a pAb to DC-SIGN and to a lesser extent, in the presence of bLF. These data clearly indicate the importance of the bovine DC-SIGN ortholog in mycobacterial infection and may help to explain the production of immunosuppressive IL-10 by bovine alveolar macrophages and DC in response to M. bovis (Lise Schreuder and T. J. Coffey, manuscript in preparation), as well as in vaccinated animals [38 ]. In addition, the presence of a high-affinity receptor for mycobacterial antigens on DC, such as DC-SIGN, and its known ability to target bound antigens to endosomal/lysosomal compartments may mature these cells. This leads to their migration to the regional draining lymph nodes, where they are able to prime a mycobacterial-specific T cell response. It is worth noting that the role of DC-SIGN and the resulting chemokine response may differ between host as well as mycobacterial species [37 , 39 ].
To verify that the DC-SIGN-bound antigen is internalized and delivered into endosomal/lysosomal compartments, confocal microscopy studies were performed. gp120 was internalized by human and bovine MoDC as early as 10 min after incubation. Our data are in accordance with a recent study showing that DC-SIGN resides within lipid rafts in the cellular membrane but also outside these in the cytoplasm [40 ].
In summary, our data describe the presence of a bovine ortholog to human DC-SIGN expressed by bovine MoDC, able to bind M. bovis BCG. In addition, this molecule is able to bind and internalize HIV gp120 in a specific manner.
Received August 7, 2007; revised January 29, 2008; accepted January 29, 2008.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. S. Seo, J. Y. Park, W. C. Davis, L. K. Fox, M. A. McGuire, Y. H. Park, and G. A. Bohach Superantigen-mediated differentiation of bovine monocytes into dendritic cells J. Leukoc. Biol., April 1, 2009; 85(4): 606 - 616. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
