HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mwangi, W. Articles by Palmer, G. H. Search for Related Content PubMed PubMed Citation Articles by Mwangi, W. Articles by Palmer, G. H.

Identification of Fetal Liver Tyrosine Kinase 3 (Flt3) Ligand Domain Required for Receptor Binding and Function Using Naturally Occurring Ligand Isoforms^{1 ,2}

Department of Veterinary Microbiology and Pathology, Washington State University, Pullman, WA 99164

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
We used a comparative approach to identify the fetal liver tyrosine kinase 3 (flt3) ligand structure required for binding and function. Two conserved bovine flt3 ligand isoforms, which differ in a defined region within the extracellular domain, were identified and shown to be uniformly transcribed in individuals with diverse MHC haplotypes. Notably, at the amino acid level, the extracellular domain of the bovine flt3 ligand isoform 1 is 81 and 72% identical with the extracellular domains of the human and murine flt3 ligands, respectively, whereas isoform-2 has a deletion within this domain. Bovine flt3 ligand isoform 1, but not 2, bound the human flt3 receptor and stimulated murine pro B cells transfected with the murine flt3 receptor. This retention of binding and function allowed definition of key residues by identifying sequences conserved among species. We have shown that a highly conserved, 18 aa sequence within the flt3 ligand extracellular domain is required for flt3 receptor binding and function. However, a peptide representing this sequence is insufficient for receptor binding as demonstrated by its failure to inhibit the bovine flt3 ligand isoform 1 binding to the human flt3 receptor. The requirement for flanking structure was confirmed by testing bovine flt3 ligand isoform 1 constructs truncated at specific residues outside the 18 aa sequence. Overall, the flt3 ligand structure required for function is markedly similar to that of the related hemopoietic growth factors, CSF-1 and steel factor. This definition of the required flt3 ligand structure will facilitate development of agonists to enhance dendritic cell recruitment for vaccines and immunotherapy.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
The survival, proliferation, and differentiation of hemopoietic cells is regulated, in part, by ligands binding tyrosine kinase receptors (1, 2). Fetal liver tyrosine kinase 3 (flt3)⁴ ligand binds the flt3 receptor, resulting in proliferation of murine hemopoietic progenitor cells and human CD34⁺ cord blood cells and bone marrow cells enriched for hemopoietic stem and progenitor cells (3, 4, 5, 6, 7, 8). Importantly, triggering flt3 stimulates, both in vitro and in vivo, generation of large numbers of both myeloid and lymphoid dendritic cells (DC) (9, 10, 11). This potent effect of flt3 ligand on DC generation is attributed to its ability to selectively target and expand the CD34⁺ progenitor cells (9). As DC are rare, specialized professional APCs and mature DC are efficient inducers of T cell activation (12), developing natural or synthetic agonists for the flt3 receptor would provide an adjuvant targeted at enhancing early steps in T cell priming. Accordingly, recombinant flt3 ligand has been shown to induce significant increases in DC in lymphoid tissues and enhance the sensitivity of Ag-specific B and T cell responses to a soluble protein, influence the class of Ab produced, and enable productive immune responses to otherwise tolerogenic protocols (13). These properties of flt3 ligand make it a promising agent for use in DC-based immunotherapy and as a vaccine adjuvant.

The flt3 ligand is a type 1 transmembrane protein that undergoes proteolytic cleavage to generate a soluble form (6, 7, 8). Regulation of mRNA splicing is also likely to control the generation of membrane bound vs soluble forms of flt3 ligand (14). Both the membrane-bound and the soluble forms bind the flt3 receptor and are biologically active, confirming that the predicted extracellular domains are functional (6, 7, 8, 14). Studies on the structurally related hemopoietic growth factors, CSF-1 and steel factor (SLF) (1, 2), have demonstrated that the conformation of the extracellular domain is critical to maintaining function. Flt3 ligand, CSF-1, and SLF are all type 1 transmembrane proteins characterized by short cytoplasmic domains and four conserved cysteines in the extracellular domain (1, 2, 6, 7, 8). These four conserved cysteines form two disulfide bonds in CSF-1 and SLF, the first pairing with the third, and the second pairing with the fourth (15, 16, 17).

Although the extracellular domain of flt3 ligand has been shown to bind the receptor, the residues within this domain that are required for binding and function have not been defined. Several flt3 ligand isoforms have been identified in human and murine T cell clones (7, 8, 14, 18). The extracellular domains of the three biologically active human flt3 ligand isoforms, 8, 9, and 22-1, are identical as are the extracellular domains of the three biologically active murine flt3 ligand isoforms, 6C, 5H, and E6 (also designated as T110, T118, and T169) (6, 7, 8, 14, 18). The differences between the biologically active human flt3 ligand isoforms occur downstream of the fourth conserved cysteine; this is also true for the biologically active murine flt3 ligand isoforms (6, 7, 8, 14, 18). Consequently, the extracellular domain structure required for receptor binding cannot be inferred from these human and murine flt3 ligand isoforms. We have identified two conserved flt3 ligand isoforms, uniformly expressed in cattle with diverse MHC haplotypes, which contain the four conserved cysteines but differ in a defined region within the extracellular domain. In this report, we use these naturally occurring isoforms to identify residues within the extracellular domain that are essential for flt3 receptor binding and biological function.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cells

Bovine PBMC were isolated by density gradient centrifugation using Lymphoprep (Nycomed Pharma AS, Oslo, Norway). B lymphocytes were isolated from PBMC by positive selection using dynabeads (Dynal, Lake Success, NY) as previously described (19). Peripheral blood monocyte-derived macrophages were generated as previously described (20). BL-3 (CRL 8037; American Type Culture Collection, Manassas, VA) is a bovine leukemia B cell line. Ba/F3 is a murine pro B cell line and Baflt is the murine pro B cell line permanently transfected with a construct expressing the murine flt3 receptor (7). The clone 42.2F7 is a Babesia bovis 42-kDa merozoite surface Ag-1-specific bovine CD4⁺ T cell clone (21) and G1.1H5 is a Fasciola hepatica-specific bovine CD4⁺ T cell clone (22). Bone marrow, cerebrum, kidney, liver, lung, lymph node, pancreas, spleen, and thymus tissue samples were collected postmortem from a Holstein calf. The tissue samples were finely minced and passed through an 100-µm nylon cell strainer (Falcon 2360; Becton Dickinson, Franklin Lakes, NJ) to generate a single cell suspension in PBS.

Cloning of cDNAs encoding two bovine flt3 ligand isoforms

Two bovine flt3 ligand cDNA fragments were first amplified using RT-PCR. Total RNA was isolated from bovine PBMC using TRIzol (Life Technologies, Gaithersburg, MD), and poly(A)⁺ mRNA was isolated from total RNA using Oligotex mRNA Mini kit (Qiagen, Valencia, CA). A minimum of 2 µg poly(A)⁺ mRNA (8) was reverse transcribed using the Thermoscript RT-PCR System (Life Technologies). Neospora caninum poly(A)⁺ mRNA was used as a negative control. A forward primer (F1; 5'-AGGCATGAGGGCCCCCGGC-3') was designed from a region upstream of the human flt3 ligand start codon that is 95% identical with nucleotides upstream of the murine flt3 ligand start codon (6, 8). A reverse primer (R1; 5'-CCTCCGCCGCGTCCTCTCCCA-3') was designed from a region upstream of the human flt3 ligand stop codon that is 89% identical with nucleotides upstream of the murine flt3 ligand stop codon (6, 8). Using these primers, two bovine flt3 ligand cDNA fragments spanning the signal sequence, extracellular, transmembrane, and part of the cytoplasmic domains were amplified by PCR using the high fidelity Pwo polymerase (Roche, Indianapolis, IN). The two bovine flt3 ligand cDNA fragments were cloned into PCR-Blunt vector (Invitrogen, Carlsbad, CA) and positive clones were sequenced using an ABI PRISM (Applied Biosystems, Foster City, CA) automated fluorescence DNA sequencer.

Cloning of the full-length bovine flt3 ligand cDNAs

To obtain full-length bovine flt3 ligand cDNAs, we used rapid amplification of cDNA ends (RACE) (23) to isolate and clone the unknown 3' end. A total of 2 µg poly(A)⁺ mRNA was reverse transcribed using the 5'/3'-RACE Kit (Roche) with the oligo dT anchor primer (5'-GACCACGCGTATCGATGTCGACTTTTTTTTTTTTTTTTG-3'). The cDNA obtained was then subjected to 3'-RACE PCR amplification using a bovine flt3 ligand-specific forward primer (F2; 5'-ATGACAGTGCTGGCGCCAGCCTGG-3') designed from the partial bovine flt3 ligand cDNA sequences obtained above, and the PCR anchor reverse primer (5'-GACCACGCGTATCGATGTCGAC-3'). The amplified cDNAs were analyzed by agarose gel (1.5%) electrophoresis and cloned into the PCR-Blunt vector. Nine clones were then sequenced from both directions using M13 forward (–20) and M13 reverse primers (Invitrogen).

Bovine flt3 ligand isoforms 1 (BFLT3L1) and 2 (BFLT3L2) expression in calves

To determine whether the identified bovine flt3 ligand cDNAs represented conserved isoforms in cattle, flt3 ligand cDNAs were RT-PCR amplified from calves with different MHC class I and class II haplotypes (24). The F2 forward primer and the PCR anchor reverse primer were used to amplify full-length cDNAs as described above. The PBMC used in the RT-PCR were isolated as described above from calves 96B6, 96B9, C97, and G3, which represent three different breeds. The amplified cDNAs were cloned into the PCR-Blunt vector and sequenced from both directions using the M13 forward (–20) and M13 reverse primers.

Cell and tissue distribution

Using the F2 and R1 primers, the bovine flt3 ligand cDNAs were RT-PCR amplified from the following cells and tissues: peripheral blood-derived B lymphocytes (isolated as previously described, Ref. 19); BL-3 cells (bovine leukemia B cell line, CRL 8037; American Type Culture Collection); bone marrow; CD4⁺ T cells (Babesia bovis 42-kDa merozoite surface Ag-1-specific bovine T cell clone 42.2F7, Ref. 21 and Fasciola hepatica-specific bovine T cell clone G1.1H5, Ref. 22); cerebrum; kidney; liver; lung; lymph node; peripheral blood monocyte-derived macrophages (isolated as previously described; Ref. 20); macrophages stimulated with LPS for 6 h; pancreas; unstimulated PBMC; PBMC stimulated with Con A for 6 h; B cell and macrophage-depleted PBMC (B cells were depleted by positive selection, and macrophages were depleted by adherence to polystyrene petri dishes); spleen; and thymus. The amplified cDNAs were analyzed by agarose gel (1.5%) electrophoresis. The cDNA from normal B-lymphocytes, CD4⁺ T cells, cerebrum, and macrophages were cloned into the PCR-Blunt vector and two clones from each sample were sequenced from both directions using the M13 forward and reverse primers.

Sequence analysis

The cDNA and deduced amino acid sequence analysis was performed using the GCG software package (Genetics Computer Group, University of Wisconsin, Madison, WI). The predicted signal peptide cleavage site was identified using the algorithm of Henrik (25), and residues within the transmembrane domain were identified using the algorithm of Hofmann (26). Multiple sequence alignment was aided by the ClustalW alignment editor (27), and sequence comparison was completed by using the BLAST programs (28). Hydropathic profiles of the deduced bovine flt3 ligand amino acid sequences were predicted by using the algorithm of Hofmann (26). The method of Hopp and Woods (29) was used to select an immunogenic peptide for generating peptide antiserum.

Peptide antiserum

A synthetic peptide CKTVAGSEMEKLLEDVNTE was used to induce bovine flt3 ligand peptide antiserum in BALB/c mice. This peptide was designed from the hydrophilic region on the extracellular domain of bovine flt3 ligand and is present in both isoforms. The peptide (2 mg) was coupled to a carrier protein (2 mg) using Imject Maleimide Activated Immunogen Conjugation kit with mariculture KLH (Pierce, Rockford, IL) according to the manufacturer’s instructions. A total of 150 µg of the conjugated protein was emulsified with 1 volume of Freund’s complete adjuvant and used to immunize three BALB/c mice by s.c. injection. Booster injections, using Freund’s incomplete adjuvant, were given at 2-wk intervals. Sera were obtained just before immunization and then 1 wk after each boost.

Expression and receptor binding assay

The open reading frames encoding the full-length bovine flt3 ligand isoforms 1 and 2 were PCR amplified using a bovine flt3 ligand-specific forward primer (F3; 5'-ATAGATATCATGACAGTGCTGGCGCCAGCCTGG-3') and a bovine flt3 ligand-specific reverse primer (R2; 5'-ATAGGATCCTTATATACAATTTCCTGGGGACAAGGGC-3'). The open reading frames encoding the extracellular domains of isoforms 1 and 2 were also PCR amplified using the forward primer F3 and a reverse primer (R3; 5'-AGGATCCTAAGGGGACTGAGGGCCCGGCAGGGA-3') that was designed from nucleotides encoding the last eight residues of the predicted extracellular domain. The forward primer F3 in combination with two reverse primers, R4 (5'-TAGGATCCTACTGTAGTTCCAGGCACCGGG-3') and R5 (5'-TAGGATCCTAACACTGTAGTTCCAGGCACCG-3'), were used to amplify two other fragments from the extracellular domain of BFLT3L1. The first fragment spans from the start codon to the nucleotides encoding the glutamine residue upstream of the last conserved cysteine (residues 1–158), whereas the second fragment spans from the start codon to the nucleotides encoding the last conserved cysteine (residues 1–159). The PCR products were then subcloned into the VR-1055 eukaryotic expression vector (Vical, San Diego, CA) and sequenced from both directions using VR-1055-specific primers. Large scale plasmid DNA was prepared using a Plasmid Maxiprep kit (Qiagen) and sterilized using 100% ethanol. A total of 20 µg DNA were transfected into one 100-mm plate of COS-7L cells (Life Technologies) using 30 µl LipofectAMINE reagent and 20 µl PLUS reagent (Life Technologies) according to the manufacturer’s instructions. Mock and VR-1055-transfected COS-7L cells were included as negative controls. One day after transfection, each plate was split to generate duplicate plates.

Three days posttransfection, the COS-7L cell monolayers were fixed using a 1:1 mixture of acetone and cold 100% ethanol for 10 min. The monolayers were then rinsed with PBS and blocked with PBS containing 5% bovine serum (blocking buffer). One set of the COS-7L cell monolayers was probed with 5 µg/ml human flt3-Fc chimera (R&D Systems, Minneapolis, MN) in blocking buffer, whereas the second set was probed with the murine anti-bovine flt3 ligand peptide antiserum (1:200) in blocking buffer to control for protein expression. Following washes in blocking buffer, monolayers probed with the human flt3-Fc chimera were incubated with a 1:2500 dilution of alkaline phosphatase-conjugated caprine anti-human Ab (Tropix, Bedford, MA) in blocking buffer, while the monolayers probed with the murine anti-bovine flt3 ligand peptide antiserum were incubated with a 1:2500 dilution of alkaline phosphatase-conjugated caprine anti-murine Ab (Tropix). Following washes in blocking buffer, the alkaline phosphatase activity was detected using Fast Red TR/Naphthol AS-MX substrate (Sigma, St. Louis, MO). Stained cells were visualized and photographed using an inverted phase contrast microscope model CK-2 (Olympus Optical, Tokyo, Japan).

Competing receptor binding

A synthetic 26-mer peptide (SCAFQPLPSCLRFVQANISHLLQDTH), which includes the underlined 18-mer peptide present in BFLT3L1 (residues 111–136) but absent in isoform 2, was used to determine whether it could compete with the full-length BFLT3L1 for the flt3 receptor. One 100-mm plate of COS-7L cells was transfected with the construct expressing full-length BFLT3L1 as above and 1 day posttransfection, the cells were diluted and replated into six-well plates. Three days posttransfection, the COS-7L cell monolayers were processed as above, and the receptor binding assay was conducted in the presence of the 26-mer peptide. Cells were incubated with a mixture containing 5 µg/ml human flt3-Fc chimera and 10-fold dilutions of the 26-mer peptide ranging in concentration from 5 µg/ml to 5 mg/ml. Additional cells were incubated with the murine anti-bovine flt3 ligand peptide antiserum diluted 1:200 to verify protein expression. Binding of the human flt3-Fc chimera and the BFLT3L1 protein expression were detected as above.

Western blot

Supernatant from COS-7L cells expressing the extracellular domain of the isoform 1 was concentrated 60-fold by using Centriprep centrifugal filter devices with a 10-kDa molecular mass cut off (Millipore, Bedford, MA). Supernatants from the COS-7L cells, COS-7L cells transfected with the empty vector, and COS-7L cells expressing the extracellular domain of isoform 2 were concentrated 120-fold because isoform 2 protein was not detectable when the supernatant was concentrated 60-fold. Eighty microliters of each concentrated supernatant was mixed with 20 µl 5x SDS-PAGE protein loading buffer and boiled for 5 min. Twenty microliters of the protein samples were resolved on a 14% polyacrylamide gel and transferred onto NitroBind nitrocellulose membrane (Micron Separations, Westborough, MA). The blot was probed with a 1:7000 dilution of the murine anti-bovine flt3 ligand peptide sera, and a replica blot was probed with a 1:7000 dilution of a negative control peptide antiserum (murine anti-Babesia bigemina Rhoptry-associated protein (RAP)-1c peptide antiserum). A caprine anti-murine alkaline phosphatase-conjugated secondary Ab (Tropix) was used at 1:15,000. The resolved proteins were detected using the Western-Star chemiluminescent detection system (Tropix) according to the manufacturer’s instructions.

Bioassay

Biological activity of recombinant soluble bovine flt3 ligand isoforms was assayed on the murine pro B cell line (Baflt) permanently transfected with a construct expressing the murine flt3 receptor (7). The murine pro B cell line (Ba/F3) was used as a negative control. Ba/F3 or Baflt cells (1 x 10⁴ cells per well) were incubated in triplicate with COS-7L supernatants diluted 1:10, 1:80, 1:800, 1:8,000, and 1:80,000 for 48 h at 37°C with 5% CO₂ in a humidified chamber. The cells were radiolabeled with 0.25 µCi/ml [³H]thymidine for the last 4 h and harvested. The mean [³H]thymidine incorporation in cpm was plotted against COS-7L supernatant dilutions.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cloning of the bovine flt3 ligand cDNAs

Two bovine flt3 ligand cDNA fragments were first amplified using RT-PCR from bovine PBMC because flt3 ligand message was previously demonstrated to be abundant in human PBMC by Northern blot analysis (7, 8). PCR amplification with primers F1 and R1 yielded two DNA fragments of 570 and 620 bp. The primers were designed based on two conserved regions of human (8) and murine (6) flt3 ligand sequences. These fragments were ligated into PCR-Blunt vector, and four clones were sequenced from both directions. The sequence from clone Bflt3L6 was used to search GenBank using BLAST programs, and 83% homology to the human and 81% homology to the murine flt3 ligand complete cDNAs was obtained. The insert of clone Bflt3L6 started from an ATG that corresponded to the start codon of human and murine flt3 ligand and ended in the cytoplasmic region that corresponded to nucleotides 90 bp upstream of the human and 66 bp upstream of the murine flt3 ligand stop codons (6, 8).

3'-RACE was performed using the bovine flt3 ligand-specific primer F2 and the PCR anchor primer that generated two DNA fragments (Fig. 1). These fragments were ligated into PCR-Blunt vector, and 12 clones were sequenced from both directions. The inserts in all of the clones started with an ATG and ended with a 16-bp-long poly(A) tail. Inserts in some clones were 857 bp long, whereas the rest were 911 bp long. Two clones, C97.4 and B6.8, were chosen to represent the 857-bp and the 911-bp clones, respectively. Clone B6.8 was designated BFLT3L1, whereas clone C97.4 was designated BFLT3L2 (Fig. 2). Multiple sequence alignment showed that all of the 857-bp clones had a 54-bp (base pair 345–399) deletion within the extracellular domain relative to the longer isoform, but maintained the same reading frame (Fig. 2).

View larger version (18K): [in this window] [in a new window]   FIGURE 1. Detection of two flt3 ligand isoform transcripts by RT-PCR. Lane 1, 100-bp ladder. Lane 2, Negative control (Neospora caninum poly(A)+ mRNA). Lane 3, Bovine flt3 ligand isoforms of 857 and 911 bp.

View larger version (40K): [in this window] [in a new window]   FIGURE 2. Bovine flt3 ligand (isoforms BFLT3L1 and BFLT3L2) nucleotide sequence alignment and deduced amino acid sequences. BFLT3L2 nucleotides identical with BFLT3L1 nucleotides are indicated by a dot. A dash in the BFLT3L2 sequence indicates a missing nucleotide as compared with BFLT3L1. The predicted signal sequence cleavage site is indicated by an arrow, and the predicted transmembrane region is underlined. Two potential N-linked glycosylation sites are overlined with a thin line, and the amino acids in bold are residues present in BFLT3L1 but missing in BFLT3L2. The nucleotides overlined with a bold line is a sequence similar to a consensus poly(A) addition signal (TATAAA as compared with the consensus AATAAA).

Analysis of the BFLT3L1 and BFLT3L2 cDNA sequences showed open reading frames of 876 and 822 bp, respectively. The two open reading frames have 35 bp of 3' noncoding sequences (Fig. 2). Fifteen base pairs upstream of the poly(A) tail was a sequence similar to a consensus polyadenylation signal (TATAAA as compared with the consensus AATAAA). This sequence was exactly the same as the polyadenylation signal in human flt3 ligand and occurred at exactly the same position (8). The BFLT3L1 and BFLT3L2 open reading frames encode transmembrane proteins of 292 and 274 aa, respectively (Fig. 2). The isoform BFLT3L2 lacks 18 aa (residues 116–133) within the extracellular domain, but the rest of the protein is completely identical with the isoform BFLT3L1 (Fig. 2). Both isoforms have a signal peptide of 27 aa and in the isoform BFLT3L1, the signal peptide is followed by an extracellular domain of 159 aa, whereas the extracellular domain of the isoform BFLT3L2 has 141 aa. Both isoforms have a 24-aa transmembrane domain and an 82-aa cytoplasmic domain. The extracellular domain of the isoform BFLT3L1 has two predicted N-linked glycosylation sites, whereas the extracellular domain of the isoform BFLT3L2 has one site. These predicted N-linked glycosylation sites are at the same positions as observed in the human and murine flt3 ligands (6, 8, 14). The second predicted N-linked glycosylation site absent in isoform BFLT3L2 is contained within the 18-aa deletion (Fig. 2).

Alignment of the bovine, human, and murine flt3 ligand sequences showed that the four conserved cysteines present in the extracellular domains of the human and murine flt3 ligands are also conserved in the two bovine flt3 ligand isoforms (Fig. 3). The cytoplasmic domains of both the bovine flt3 ligand isoforms are 52 aa longer than the human and 61 aa longer than the murine cytoplasmic domains. In their extracellular domains, the isoform BFLT3L1 is 81 and 72% identical with the human and murine flt3- ligands, respectively. The extracellular domain of isoform BFLT3L2 is 61 and 56% identical with the extracellular domains of the human and murine flt3 ligands, respectively (Fig. 3).

View larger version (61K): [in this window] [in a new window]   FIGURE 3. Bovine, human, and murine flt3 ligand amino acid sequence alignment. Completely conserved residues are indicated by a black background; conservative changes are indicated by a gray background. A white background indicates positions of residues that vary among species. Missing residues are indicated by a dash. The four cysteines that are conserved between flt3 ligand, SLF, and CSF-1 are denoted by an asterisk.

To determine whether the identified bovine flt3 ligand cDNAs represented conserved, predominant isoforms in cattle, flt3 ligand cDNAs were amplified by RT-PCR from four calves with different MHC haplotypes. The F2 forward primer and the PCR anchor reverse primer were used to amplify full-length bovine flt3 ligand cDNAs. The amplified cDNAs were cloned into the PCR-Blunt vector and sequenced from both directions. The cDNA sequences obtained showed that both the BFLT3L1 and BFLT3L2 isoforms were expressed in all of the four calves (Fig. 4).

View larger version (11K): [in this window] [in a new window]   FIGURE 4. Analysis of bovine flt3 ligand cDNA sequences (only base pairs 333–413 are shown) cloned from PBMC of four genetically different calves. Sequences from eight clones are shown; clones B6FLT3L8 and BFLT3L10 were obtained from calf 96B6, clones B9FLT3L17 and B9FLT3L18 were obtained from calf 96B9, clones C97FLT3L2 and C97FLT3L4 were obtained from calf C97, and clones G3FLT3L6 and G3FLT3L7 were obtained from calf G3. A dot indicates a nucleotide identical with that shown in the B6FLT3L10 cDNA sequence, and a dash indicates a missing nucleotide.

Transcripts of bovine flt3 ligand isoforms in bovine cells and tissues were examined by RT-PCR using the F2 and R1 primers. Two transcripts of 570 and 620 bp were detected in mRNA from all the cells and tissues sampled (Fig. 5). In most tissues, isoform 1 transcripts appear to be expressed at higher levels than isoform 2. However, this observation is based on semiquantitative analysis and will require confirmation. Transcripts from B and T lymphocytes, cerebrum, and macrophages were completely sequenced. Multiple sequence alignment of these cDNA sequences showed that the two bovine flt3 ligand isoforms were expressed in these cells (Fig. 6).

View larger version (31K): [in this window] [in a new window]   FIGURE 5. Transcription of flt3 ligand isoforms in bovine cells and tissues. Lanes 1 and 9, 100-bp ladder. Lane 2, Normal B cells. Lane 3, BL-3 cells (bovine leukemia B cell line). Lane 4, Cerebrum. Lane 5, Bone marrow. Lane 6, CD4+ T cells (B. bovis 42-kDa merozoite surface Ag-1-specific bovine T cell clone 42.2F7). Lane 7, CD4+ T cells (F. hepatica-specific bovine T cell clone G1.1H5). Lane 8, Liver. Lane 10, Lymph node. Lane 11, Macrophages. Lane 12, Macrophages stimulated with LPS for 6 h. Lane 13, Pancreas. Lane 14, Unstimulated PBMC. Lane 15, PBMC stimulated with Con A for 6 h. Lane 16, B cell and macrophage-depleted PBMC. Lane 17, Spleen. Lane 18, Thymus.

View larger version (9K): [in this window] [in a new window]   FIGURE 6. Detection of transcripts for both flt3 ligand isoforms in B and T lymphocytes, and macrophages (only bases 333–413 are shown). Sequences from six clones are shown; clones BCFLT3L1 and BCFLT3L2 were obtained from B cells, clones CDFLT3L1 and CDFLT3L2 were obtained from CD4+ T cells (B. bovis 42-kDa merozoite surface Ag-1-specific T cell clone 42.2F7), and clones MCFLT3L1 and MCFLT3L2 were obtained from macrophages. A dot indicates a nucleotide identical with that shown in the BCFLT3L1 cDNA sequence, and a dash indicates a missing nucleotide.

Hydropathic profiles of the bovine flt3 ligand isoforms are very similar, but the profile of the extracellular domain of isoform 1 is slightly different from that of isoform 2 (Fig. 7). This difference within the extracellular domains could have an effect on receptor binding because the flt3 ligand residues involved in receptor binding are located in the extracellular domain (7, 8, 14). To determine whether the bovine flt3 ligand isoforms could bind the flt3 receptor, the full-length open reading frames and the extracellular domains (Fig. 2) were expressed in COS-7L cells. Binding of the soluble human flt3-Fc chimera to the flt3 ligand was assessed by in situ immunocytochemistry, which showed that the BFLT3L1, but not isoform 2, bound to the human flt3 receptor (Fig. 8). Protein expression by the COS-7L transfectants was assessed by in situ immunocytochemistry using the murine anti-bovine flt3 ligand peptide antiserum as described in Materials and Methods. The assays showed that the COS-7L cells transfected with the constructs encoding full-length bovine flt3 ligand isoforms and the COS-7L cells transfected with the constructs encoding the extracellular domains expressed the flt3 ligand protein (Fig. 9), but only the isoform 1 bound to the human flt3 receptor (Fig. 8).

View larger version (24K): [in this window] [in a new window]   FIGURE 7. Hydropathic profiles of bovine flt3 ligand isoforms. A, flt3 ligand isoform 1 (BFLT3L1); B, isoform 2 (BFLT3L2). SP, Signal peptide; ED, extracellular domain; TM, transmembrane domain; CD, cytoplasmic domain. The filled rectangle in A indicates the 18-mer insertion in BFLT3L1. The vertical dotted lines indicate domain boundaries.

View larger version (57K): [in this window] [in a new window]   FIGURE 8. Binding of BFLT3L1, but not isoform 2, to the human flt3 receptor. Monolayers of COS-7L cells transfected with DNA constructs encoding bovine flt3 ligand isoforms were assayed for human flt3-fc binding, as described in Materials and Methods. A, COS-7L cells. B, COS-7L cells transfected with empty vector. C, COS-7L cells transfected with a DNA construct encoding full-length flt3 ligand isoform-1. D, COS-7L cells transfected with a DNA construct encoding the extracellular domain of isoform 1. E, COS-7L cells transfected with a DNA construct encoding full-length isoform 2. F, COS-7L cells transfected with a DNA construct encoding the extracellular domain of isoform 2.

View larger version (55K): [in this window] [in a new window]   FIGURE 9. Expression of bovine flt3 ligand isoforms in COS-7L cells. Monolayers of COS-7L cells transfected with DNA constructs encoding bovine flt3 ligand isoforms were assayed for flt3 ligand expression using a mouse antibovine flt3 ligand peptide sera, by in situ immunocytochemistry, as described in Materials and Methods. A, COS-7L cells. B, COS-7L cells transfected with empty vector. C, COS-7L cells transfected with a DNA construct encoding full-length flt3 ligand isoform 1. D, COS-7L cells transfected with a DNA construct encoding the extracellular domain of isoform 1. E, COS-7L cells transfected with a DNA construct encoding full-length isoform 2. F, COS-7L cells transfected with a DNA construct encoding the extracellular domain of isoform 2.

Competing receptor binding

A synthetic 26-mer peptide (SCAFQPLPSCLRFVQANISHLLQDTH), which includes the underlined 18-mer peptide present in BFLT3L1 (residues 111–136) but absent in isoform 2 (Fig. 2), was used to determine whether it could compete with the BFLT3L1 for the flt3 receptor. Binding of the soluble human flt3-Fc chimera to BFLT3L1 was assayed in the presence of the 26-mer peptide as described in Materials and Methods. The bound human flt3-Fc chimera was detected by in situ immunocytochemistry, demonstrating that the 26-mer peptide could not competitively inhibit binding of the flt3-fc chimera to the BFLT3L1 (data not shown).

Biological activity of the bovine flt3 ligand isoforms

Supernatants from COS-7L cells expressing the extracellular domains of the bovine flt3 ligand isoforms 1 and 2 were tested for their capacity to stimulate murine pro-B cells (Baflt cells) expressing the murine flt3 receptor (7). The supernatant from COS-7L cells expressing the extracellular domain of BFLT3L1, but not isoform 2, stimulated the proliferation of Baflt cells (stimulation index = 5, Fig. 10). Supernatants from COS-7L cells, and COS-7L cells transfected with the empty vector, did not stimulate Baflt cells (Fig. 10). None of the supernatants stimulated proliferation of non-flt3 receptor-transfected murine pro B-cells (data not shown).

View larger version (19K): [in this window] [in a new window]   FIGURE 10. Analysis of the biological activity of soluble bovine flt3 ligand isoforms by proliferation assay using Baflt cells (murine pro-B cell line transfected with murine flt3 receptor). Baflt cells (1 x 104 cells per well) were incubated in the presence of the diluted COS-7L cell medium and incubated for 48 h at 37°C with 5% CO2 in a humidified chamber. The cells were pulsed with 0.25 µCi/ml [3H]thymidine in the last 4 h and harvested. , Vector-transfected COS-7L cell supernatant; , supernatant from COS-7L cells expressing the extracellular domain of flt3 ligand isoform 1; , supernatant from COS-7L cells expressing the extracellular domain of flt3 ligand isoform 2. Each point represents the mean number of [3H]thymidine incorporation in cpm from three wells ± SD.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The requirement for flanking structure was confirmed by showing that a polypeptide containing amino acids 1–159 of bovine isoform 1 bound the receptor, whereas deletion of the fourth conserved cysteine, residue 159, abrogated receptor binding. Within their extracellular domains, the bovine, human, and murine flt3 ligands each have four conserved cysteines, which are also conserved in the closely related hemopoietic growth factors, CSF-1 and SLF (1, 2, 6, 7). These four conserved cysteines form two disulfide bonds in CSF-1 and SLF, the first pairing with the third, and the second pairing with the fourth (15, 16, 17). These conserved cysteines are thought to be similarly involved in the folding of the extracellular domain of flt3 ligand because the overall structure of flt3 ligand is similar to that of the four-helix bundle proteins CSF-1 and SLF (1, 2, 6, 7, 17). Thus the deletion of the fourth conserved cysteine in the extracellular domain of BFLT3L1 and abrogation of receptor binding is consistent with the required maintenance of this structure for flt3 ligand function.

The BFLT3L2 is a novel isoform because it contains the four conserved cysteines present in biologically active BFLT3L1, human isoforms 8, 9, and 22-1 (8), and murine isoforms 6C, 5H, and E6 (6, 14) (also designated T110, T118, and T169; Refs. 7, 18) but does not bind to the flt3 receptor or stimulate proliferation of Baflt cells. Unique to the BFLT3L2 is an 18-aa deletion in the extracellular domain upstream of the fourth conserved cysteine that does not disrupt the reading frame. The 18-aa deletion provided definitive evidence for the requirement of this sequence in flt3 receptor binding and function. In comparison, the extracellular domains of the three biologically active human flt3 ligand isoforms, 8, 9, and 22-1, are identical, as are the extracellular domains of the three biologically active murine flt3 ligand isoforms, 6C, 5H, and E6 (6, 7, 8, 14, 18). In all cases, the sequence changes in the biologically active human and murine flt3 ligand isoforms occur downstream of the fourth conserved cysteine (6, 7, 8, 14, 18). The human flt3 ligand isoforms 14 and 24 are unlikely to encode a biologically active protein as a result of frame shifts that markedly alter the extracellular domains (8). The human flt3 ligand isoform 14 has a 17-bp insertion within the signal sequence that disrupts the entire reading frame of the extracellular domain, whereas isoform 24 has a 139-bp deletion (nucleotides 426–564) in the extracellular domain, which results in deletion of amino acid residues 115–160 and changes the reading frame (8). The murine flt3 ligand isoform E616 has a 16-bp deletion in the extracellular domain that results in deletion of amino acid residues 159–164 and changes the reading frame (14). This isoform is biologically inactive, assumed to be due to the lack of the fourth conserved cysteine, residue 161 (14). We have confirmed this assumption by demonstrating that the fourth conserved cysteine in the extracellular domain of flt3 ligand is required for flt3 receptor binding.

The cytoplasmic domains of the human and murine flt3 ligands are short (6, 7, 8, 14, 18). Surprisingly, the cytoplasmic domains of both the bovine flt3 ligand isoforms are 52 amino acids longer than the human and 61 amino acids longer than the murine cytoplasmic domains. This is unusual because ligands for type III tyrosine kinase receptors are characterized by short cytoplasmic tails (1, 2, 6, 7, 8). Whether there is a unique cellular signaling pattern induced by bovine isoform 1 as compared with the flt3 ligand in humans and mice is unknown.

Although the overall tissue pattern of flt3 ligand expression appears similar among species, we detected transcripts of bovine flt3 ligand isoforms in the cerebrum. This is notable because transcripts of human and murine flt3 ligands were not detected in brain analyzed by both Northern blot and RT-PCR (7, 8, 18). Sequences of the bovine flt3 ligand transcripts from the cerebrum were identical with those shown in Fig. 2. Importantly, similar levels of transcripts of both flt3 ligand isoforms were present in most cells and tissues examined, as well as in cattle with different MHC haplotypes, suggesting that both isoforms are constitutively expressed.

The inability of BFLT3L2 to bind the flt3 receptor, the broad tissue distribution, and presence in individuals with different MHC haplotypes suggest that the BFLT3L2 has a function not requiring flt3 binding. In contrast, the BFLT3L1, human isoforms 8, 9, and 22-1 (8), and murine isoforms 6C, 5H, and E6 (6, 14) bind the flt3 receptor, hence they are true ligands for flt3. The inability of BFLT3L2 to bind the flt3 receptor also suggests that it does not act as a competitive antagonist. Splicing to switch from BFLT3L1 to isoform 2 or formation of mixed heterodimers to regulate isoform 1 function is a possible role for isoform 2 function; however, we found no evidence of differential transcription of isoforms 1 and 2 as would be expected if this was a regulatory mechanism. The most attractive hypothesis is that isoform 2 binds a different receptor and induces functions that may be related to or distinct from functions induced by triggering the flt3 receptor. This hypothesis is currently being tested.

	Acknowledgments

 
We thank Dr. Chuck Hannum (DNAX Research Institute, Palo Alto, CA) for generously providing Ba/F3 and Baflt cells; Dr. Harris A. Lewin (University of Illinois, Urbana, IL) for identifying MHC haplotypes; Dr. Travis C. McGuire for critically reviewing the manuscript; and Beverly Hunter and Kim Kegerreis for excellent technical assistance.

	Footnotes

 
¹ This work was supported by U.S. Department of Agriculture-National Research Initiative Competitive Grants Program Grant 99-35204-8274.

² The sequence data have been submitted to the National Center for Biotechnology Information nucleotide sequence databases under the accession numbers AF282985 and AF282986.

³ Address correspondence and reprint requests to Waithaka Mwangi, Department of Veterinary Microbiology and Pathology, Washington State University, Pullman, WA 99164-7040.

⁴ Abbreviations used in this paper: flt3, fetal liver tyrosine kinase 3; DC, dendritic cell(s); SLF, steel factor; RACE, rapid amplification of cDNA ends; BFLT3L1, bovine flt3 ligand isoform 1; BFLT3L2, bovine flt3 ligand isoform 2.

Received for publication July 6, 2000. Accepted for publication September 25, 2000.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Stanley, E. R., L. J. Guilbert, R. J. Tushinski, S. H. Bartelmez. 1983. CSF-1: a mononuclear phagocyte lineage specific hematopoietic growth factor. J. Cell. Biochem. 21:151.[Medline]
2. Williams, D. E., P. de Vries, A. E. Namen, M. B. Widmer, S. D. Lyman. 1992. The steel factor. Dev. Biol. 151:368.[Medline]
3. Rosnet, O., C. Schiff, M.-J. Pebusque, S. Marchetto, C. Tonnelle, T. Toiron, D. Birnbaum. 1993. Human flt3/flk2 gene: cDNA cloning and expression in hematopoietic cells. Blood 82:1110.[Abstract/Free Full Text]
4. Rosnet, O., S. Marchetto, O. de Lapeyriere, D. Birnbaum. 1991. Murine flt3, a gene encoding a novel tyrosine kinase receptor of the PDGFR/CSF1R family. Oncogene 6:1641.[Medline]
5. Mathews, W., C. T. Jordan, G. W. Wiegand, D. Pardoll, I. R. Lemischka. 1991. A receptor tyrosine kinase specific to hematopoietic stem and progenitor cell-enriched populations. Cell 65:1143.[Medline]
6. Lyman, S. D., L. James., T. VandenBos, P. de Vries, K. Brasel, B. Gliniak, L. T. Hollingsworth, K. S. Picha, H. J. McKenna, R. R. Splett, et al 1993. Molecular cloning of a ligand for the flt3/flk-2 tyrosine kinase receptor: a proliferative factor for primitive hematopoietic cells. Cell 75:1157.[Medline]
7. Hannum, C., J. Culpepper, D. Campbell, T. McClanahan, S. Zurawski, J. F. Bazan, R. Kastelein, S. Hudak, J. Wagener, J. Mattson, et al 1994. Ligand for flt3/flk2 receptor tyrosine kinase regulates growth of hematopoietic stem cells and is encoded by variant RNAs. Nature 368:643.[Medline]
8. Lyman, S. D., L. James, L. Johnson, K. Brasel, P. de Vries, S. S. Escobar, H. Downey, R. R. Splett, M. P. Beckmann, H. J. McKenna. 1994. Cloning of the human homologue of the murine flt3 ligand: a growth factor for early hematopoietic progenitor cells. Blood 83:2795.[Abstract/Free Full Text]
9. Siena, S., M. DiNicola, M. Bregni, R. Mortarini, A. Anichini, L. Lombardi, F. Ravagnani, G. Parmiani., A. M. Gianni. 1995. Massive ex vivo generation of functional dendritic cells from mobilized CD34⁺ blood progenitors for anti-cancer therapy. Exp. Haematol. 23:1463.[Medline]
10. Maraskovsky, E., K. Brasel, M. Teepe, E. R. Roux, S. D. Lyman, K. Shortman, H. J. McKenna. 1996. Dramatic increase in the number of functionally mature dendritic cells in the flt3 ligand-treated mice: multiple dendritic cell subpopulations identified. J. Exp. Med. 184:1953.[Abstract/Free Full Text]
11. Lebsack, M. E., E. Maraskovsky, E. Roux, M. Teepe, D. Hirschstein, J. Hoek, H. J. McKenna, C. Maliszewski, D. Caron. 1998. Increased circulatory dendritic cells in healthy human volunteers following administration of flt3 ligand alone or in combination with GM-CSF or G-CSF. Blood 92:507a.
12. Steinman, R. M.. 1991. The dendritic cell system and its role in immunogenicity. Annu. Rev. Immunol 9:271.[Medline]
13. Pulendran, B., J. L. Smith, M. Jenkins, M. Schoenborn, E. Maraskovsky, C. R. Maliszewski. 1998. Prevention of peripheral tolerance by a dendritic cell growth factor: flt3 ligand as an adjuvant. J. Exp. Med. 188:2075.[Abstract/Free Full Text]
14. Lyman, S. D., L. James, S. S. Escobar, H. Downey, P. de Vries, K. Brasel, K. Stocking, M. P. Beckmann, N. G. Copeland, L. S. Cleveland, et al 1995. Identification of soluble and membrane-bound isoforms of the murine flt3 ligand generated by alternative spicing of mRNAs. Oncogene 10:149.[Medline]
15. Bazan, J. F.. 1991. Genetic and structural homology of stem cell factor and macrophage colony-stimulating factors. Cell 65:9.[Medline]
16. Lu, H. S., C. L. Clogston, J. Wypych, P. R. Fausset, S. Lauren, E. A. Mendiaz, K. M. Zsebo, K. E. Langley. 1991. Amino acid sequence and post-translational modification of stem cell factor isolated from buffalo rat liver cell-conditioned medium. J. Biol. Chem. 266:8102.[Abstract/Free Full Text]
17. Pandit, J., A. Bohm, J. Jancarik, R. Halenbeck, K. Koths, S. H. Kim. 1992. Three-dimensional structure of dimeric human recombinant macrophage colony-stimulating factor. Science 258:1358.[Abstract/Free Full Text]
18. McClanahan, T., J. Culpepper, D. Campbell, J. Wagner, K. Franz-Bacon, J. Mattson, S. Tsai, J. Luh, M. J. Guimaraes, M.-G. Mattei, et al 1996. Biochemical and genetic characterization of multiple splice variants of the flt3 ligand. Blood 9:3371.
19. Brown, W. C., D. M. Estes, S. E. Chantler, K. A. Kegerres, C. E. Suarez. 1998. DNA and a CpG oligonucleotide derived from Babesia bovis are mitogenic for bovine B cells. Infect. Immun. 66:5423.[Abstract/Free Full Text]
20. Shoda, L. K. M., D. S. Zarlenga, A. Hirumi, W. C. Brown. 1999. Cloning of a cDNA encoding bovine interleukin-18 and analysis of IL-18 expression in macrophages and its IFN--inducing activity. J. Interferon Cytokine Res. 19:1169.[Medline]
21. Brown, W. C., G. H. Palmer, S. A. T .F. McElwin, S. A. Hines, D. A. E. Dobbelaere. 1993. Babesia bovis: characterization of the T helper cell response against he 42-kDa merozoite surface antigen (MSA-1) in cattle. Exp. Parasitol. 77:97.[Medline]
22. Brown, W. C., W. C. Davis, D. A. E. Dobbelaere, A. C. Rice-Ficht. 1994. CD4⁺ T-cell clones obtained from cattle chronically infected with Fasciola hepatica and specific for adult worm antigen express both unrestricted and Th2 cytokine profiles. Infect. Immun. 62:818.[Abstract/Free Full Text]
23. Frohman, M. A., M. K. Dush, G. R. Martin. 1988. Rapid production of full-length cDNAs from rare transcripts: amplification using a single gene-specific oligonucleotide primer. Proc. Natl. Acad. Sci. USA 85:8998.[Abstract/Free Full Text]
24. Lewin, H. A., G. C. Russell, E. J. Glass. 1999. Comparative organization and function of the major histocompatibility complex of domesticated cattle. Immunol. Rev. 167:145.[Medline]
25. Henrik, N., J. Engelbrecht, S. Brunak, G. von Heijne. 1997. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng. 10:1.[Abstract/Free Full Text]
26. Hofmann, K., W. Stoffel. 1993. TMbase: a database of membrane spanning proteins segments. Biol. Chem. 347:166.
27. Thompson, J. D., D. G. Higgins, T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 2:4673.
28. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403.[Medline]
29. Hopp, T. P., K. R. Woods. 1981. Prediction of protein antigenic determinants from amino acid sequences. Proc. Natl. Acad. Sci. USA 78:3824.[Abstract/Free Full Text]
30. Williamson, E., G. M. Westrich, J. L. Viney. 1999. Modulating dendritic cells to optimize mucosal immunization protocols. J. Immunol. 163:3668.[Abstract/Free Full Text]
31. Chen, K., S. Braun, S. Lyman, Y. Fan, C. M. Traycoff, E. A. Wiebke, J. Gaddy, G. Sledge, H. E. Broxmeyer, K. Cornetta. 1997. Antitumor activity and immunotherapy properties of flt3-ligand in a murine breast cancer model. Cancer Res. 57:3511.[Abstract/Free Full Text]

This article has been cited by other articles:

				Y. Zhuang, W. Mwangi, W. C. Brown, W. C. Davis, J. C. Hope, and G. H. Palmer Characterization of a Phenotypically Unique Population of CD13+ Dendritic Cells Resident in the Spleen. Clin. Vaccine Immunol., September 1, 2006; 13(9): 1064 - 1069. [Abstract] [Full Text] [PDF]

				W. Mwangi, W. C. Brown, G. A. Splitter, Y. Zhuang, K. Kegerreis, and G. H. Palmer Enhancement of antigen acquisition by dendritic cells and MHC class II-restricted epitope presentation to CD4+ T cells using VP22 DNA vaccine vectors that promote intercellular spreading following initial transfection J. Leukoc. Biol., August 1, 2005; 78(2): 401 - 411. [Abstract] [Full Text] [PDF]

				W. Mwangi, W. C. Brown, H. A. Lewin, C. J. Howard, J. C. Hope, T. V. Baszler, P. Caplazi, J. Abbott, and G. H. Palmer DNA-Encoded Fetal Liver Tyrosine Kinase 3 Ligand and Granulocyte Macrophage-Colony-Stimulating Factor Increase Dendritic Cell Recruitment to the Inoculation Site and Enhance Antigen-Specific CD4+ T Cell Responses Induced by DNA Vaccination of Outbred Animals J. Immunol., October 1, 2002; 169(7): 3837 - 3846. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mwangi, W. Articles by Palmer, G. H. Search for Related Content PubMed PubMed Citation Articles by Mwangi, W. Articles by Palmer, G. H.