Cancer Research The Future of Cancer Research: Science and Patient Impact 09 AM Call for Abstracts
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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 Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jeffers, M.
Right arrow Articles by Lichenstein, H. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jeffers, M.
Right arrow Articles by Lichenstein, H. S.
[Cancer Research 61, 3131-3138, April 1, 2001]
© 2001 American Association for Cancer Research

Molecular Biology and Genetics

Identification of a Novel Human Fibroblast Growth Factor and Characterization of Its Role in Oncogenesis

Michael Jeffers1, Richard Shimkets, Sudhirdas Prayaga, Ferenc Boldog, Meijia Yang, Catherine Burgess, Elma Fernandes, Beth Rittman, Juliette Shimkets, William J. LaRochelle and Henri S. Lichenstein

CuraGen Corporation, New Haven, Connecticut 06511


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The fibroblast growth factor (FGF) family of signaling molecules has been implicated in normal developmental and physiological processes, as well as in human malignancy. Using a homology-based genomic DNA mining process, we identified a human gene encoding a novel member of the FGF family, that we designate FGF-20. The FGF-20 cDNA was isolated, and its sequence confirmed the gene prediction. FGF-20 is expressed in normal brain, particularly the cerebellum, and in some cancer cell lines. Recombinant FGF-20 protein induces DNA synthesis in a variety of cell types and is recognized by multiple FGF receptors. Ectopic expression of FGF-20 in NIH 3T3 cells renders the cells transformed in vitro and tumorigenic in nude mice. These results underscore the utility of mining genomic DNA databases and reveal FGF-20 to be a novel oncogene that may play a role in human cancer.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Genomic mining is the process by which genes are predicted from genomic DNA sequences. This process involves identifying a genomic region of interest based on homology to a known gene, followed by prediction of exon/intron boundaries. Once a gene prediction is made, the corresponding cDNA is obtained to confirm the expression of the gene.

To identify protein therapeutic drugs and/or monoclonal antibody targets, we are systematically mining human genomic databases and functionally annotating genes that are relevant to disease. In this regard, a gene family of particular interest is the FGF2 group of cytokines. The FGF family consists of at least 21 members that regulate diverse cellular functions such as growth, survival, apoptosis, motility, and differentiation (1) . These molecules transduce signals intracellularly via high-affinity interactions with a number of cell surface tyrosine kinase FGFRs (2 , 3) . FGFs also bind, albeit with low affinity, to HSPGs present on most cell surfaces and ECMs. Interactions between FGFs and HSPGs serve to stabilize FGF-FGFR interactions and to sequester FGF and protect it from degradation (1) . Due to its growth-promoting capabilities, one member of the FGF family, FGF-7, is currently in clinical trials for the treatment of chemotherapy-induced mucositis (4) .

In addition to participating in normal growth and development, FGFs have also been implicated in the generation of pathological states, including cancer (5) . FGFs may contribute to malignancy by directly enhancing the growth of tumor cells. For example, autocrine growth stimulation through the coexpression of FGF and FGFR in the same cell leads to cellular transformation (6) . Likewise, the constitutive activation of FGFR via mutation or rearrangement leads to uncontrolled proliferation (7 , 8) . Finally, because some FGFs are angiogenic (9) , they may contribute to the tumorigenic process by facilitating the development of the blood supply needed to sustain tumor growth. Not surprisingly, at least one FGF is currently under investigation as a potential target for cancer therapy (10) .

Through a homology-based genomic mining process, we identified a novel human FGF and subsequently isolated its corresponding cDNA. We demonstrate that the protein product of this gene has growth-stimulatory and oncogenic properties. Furthermore, overexpression of the FGF mRNA was noted in specific cancer cell lines. These observations suggest that this FGF may represent a target for the treatment of human malignancy.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Identification of the FGF-20 Gene
The FGF-20 gene was identified after a TBLASTN (11) search of GenBank genomic DNA sequences with Xenopus FGF-20 (Ref. 12 ; GenBank accession number AB012615) as query. This search identified a locus (GenBank accession number AB020858) of high homology on chromosome 8. Intron/exon boundaries were deduced using standard consensus splicing parameters (13) , together with homologies derived from known FGFs. The FGF-20 initiation codon localizes to bp 16214 of locus AB020858, and the remaining 3' portion of this exon continues to bp 15930. The 5'-UTR of FGF-20 was extended an additional 606 bp upstream of the initiation codon using public ESTs (GenBank accession numbers AA232729, AA236522, AI272876, and AI272878). The remaining structure of the FGF-20 gene as it relates to locus AB020858 is as follows: (a) intron 1 (bp 15929 to 9942); (b) exon 2 (bp 9941 to 9838); (c) intron 2 (bp 9837 to 7500); (d) exon 3 [bp 7499-? (the composition of the 3'-UTR has not yet been determined)].

Identification of the FGF-20 cDNA
Oligonucleotide primers were designed to PCR amplify the complete predicted FGF-20 ORF. The primers were as follows: (a) forward primer, 5'-CTCGTCAGATCTCCACCATGGCTCCCTTAGCCGAAGTC-3'; and (b) reverse primer, 5'-CTCGTCCTCGAGAGTGTACATCAGTAGGTCCTTG-3'. The forward primer includes a BglII restriction site and a Kozak (14) consensus sequence (CCACC). The reverse primer contains a XhoI restriction site. Both primers contain a 5' clamp (CTCGTC). PCR reactions contained 5 ng of human prostate cDNA template, 1 µM each primer, 5 µM deoxynucleotide triphosphate (Clontech, Palo Alto, CA), and 1 µl of 50x Advantage-HF2 polymerase (Clontech) in a 50-µl volume. A touchdown PCR protocol was used, with an initial annealing temperature of 70°C that was decreased by 1°C/cycle until 60°C, at which point an additional 25 cycles were performed. A single PCR product of the appropriate size (~640 bp) was isolated from an agarose gel and subcloned into the pCR2.1 vector (Invitrogen, Carlsbad, CA) to generate the pCR2.1/FGF-20 construct.

Generation of FGF-20 Expression Constructs
pFGF-20.
A BglII-XhoI fragment containing the FGF-20 sequence was isolated from pCR2.1/FGF-20 and subcloned into BamHI/XhoI-digested pcDNA3.1 (Invitrogen).

pIg{kappa}-FGF-20.
We first modified the pCEP4 vector (Invitrogen) to contain an immunoglobulin {kappa} signal sequence followed by multiple cloning sites, a V5 epitope tag, and a polyhistidine tag. To this end, the V5 and polyhistidine regions were PCR amplified from pcDNA3.1 (Invitrogen) using the following primers: (a) forward primer, 5'-CTCGTCCTCGAGGGTAAGCCTATCCCTAAC-3'); and (b) reverse primer, 5'-CTCGTCGGGCCCCTGATCAGCGGGTTTAAAC-3'. The PCR product was digested with XhoI-ApaI and subcloned into XhoI-ApaI-digested pSecTag2B vector (Invitrogen), and the resulting plasmid was digested with PmeI-NheI. A DNA fragment containing the immunoglobulin {kappa} signal sequence and V5/polyhistidine tags was then subcloned into BamHI-Klenow and NheI-treated pCEP4 vector. The FGF-20 cDNA was isolated from pCR2.1/FGF-20 by PCR using primers 5'-AGATCATGGCTCCCTTAGCCGAAGTC-3' (forward primer) and 5'- CTCGTCCTCGAGAGTGTACATCAGTAGGTCCTTG-3' (reverse primer) and subcloned into the BamHI-XhoI-digested modified pCEP4 vector described above.

Real-Time Quantitative PCR Expression Analysis
RNA samples comprising normal human tissues were obtained from commercial sources (Clontech; Invitrogen; and Research Genetics, Huntsville, AL), and those comprising tumor cell lines were derived from cultured cells. Real-time quantitative PCR (15) was performed on an ABI Prism 7700 Sequence Detection System (PE Applied Biosystems, Foster City, CA) using TaqMan reagents (PE Applied Biosystems). RNAs were normalized using human ß-actin and glyceraldehyde-3-phosphate dehydrogenase TaqMan probes according to the manufacturer’s instructions. Equal quantities of normalized RNA were used as template in PCR reactions with FGF-20-specific reagents to obtain threshold cycle (CT) values. For graphic representation, CT numbers were converted to percentage expression relative to the sample exhibiting the highest level of expression. The following FGF-20-specific primers and probe, each possessing a minimum of three mismatches with corresponding regions of the highly homologous human FGF-9 and FGF-16 genes, were used: (a) forward primer, 5'-GGACCACAGCCTCTTCGGTA-3'; reverse primer, 5'-TGTCCACACCTCTAATACTGACCAG-3'; and (c) TaqMan probe, 5'-FAM-CCCACTGCCACACTGATGAATTCCAA-TAMRA-3'.

Cell Lines
Mammalian cell lines were obtained from the following sources: (a) NIH 3T3, CCD-1070sk, CCD-1106 KERTr, and 786-O were obtained from American Type Culture Collection (Manassas, VA); and (b) 293-EBNA was obtained from Invitrogen.

Generation of Recombinant FGF-20
293-EBNA cells were transfected using LipofectAMINE 2000 according to the manufacturer’s protocol (Life Technologies, Inc., Gaithersburg, MD). Cells were supplemented with 10% fetal bovine serum (Life Technologies, Inc.) 5 h after transfection. To generate protein for BrdUrd and growth assays, cells were washed and fed with DMEM (Life Technologies, Inc.) 18 h after transfection. After 48 h, the media were discarded, and the cell monolayer was incubated with 100 µM suramin (Sigma, St. Louis, MO) in 0.5 ml of DMEM for 30 min at 4°C. The suramin-containing conditioned media were then removed, clarified by centrifugation (5 min; 2000 x g), and subjected to TALON metal affinity chromatography according to the manufacturer’s instructions (Clontech). FGF-20 protein concentrations were estimated by Western analysis using a standard curve generated with a V5-tagged protein of known concentration. To generate control protein, 293-EBNA cells were transfected with pCEP4 plasmid (Invitrogen) and subjected to the purification procedure outlined above.

Western Analysis
NIH 3T3 cells were transfected using LipofectAMINE Plus according to the manufacturer’s protocol (Life Technologies, Inc.). For Western analysis, conditioned media were harvested 48 h after transfection, clarified by centrifugation (5 min; 2000 x g), and mixed with an equal volume of 2x gel-loading buffer. Cell monolayers were dissolved in 1x gel-loading buffer.

Samples were then boiled for 10 min, resolved on 4–20% gradient polyacrylamide gels (Novex, Dan Diego, CA) under reducing conditions, and transferred to nitrocellulose filters (Novex). Western analysis was performed according to standard procedures (16) using horseradish peroxidase-conjugated anti-V5 antibody (Invitrogen) and the enhanced chemiluminescence detection system (Amersham Pharmacia Biotech, Piscataway, NJ).

BrdUrd Assays
Cells were cultured in 96-well plates to ~100% confluence, washed and fed with DMEM (NIH 3T3, CCD-1070sk, and 786-O) or keratinocyte-serum-free media without supplements [Life Technologies, Inc. (CCD-1106 KERTr)], and incubated for 24 (NIH 3T3) or 48 h (CCD-1070sk, 786-O, and CCD-1106 KERTr). Recombinant FGF-20 or control protein was then added to the cells for 18 h.

To analyze the effect of soluble FGFRs on FGF-20 activity, recombinant FGF-20, aFGF, or PDGF-BB (final concentrations of 10, 5, and 3 ng/ml, respectively) was mixed with soluble receptors (final concentrations of 0.2, 1, and 5 µg/ml) and incubated for 30 min at 37°C before addition to serum-starved NIH 3T3 cells. Factor concentrations represent the amount of ligand needed to generate a half-maximal BrdUrd response in NIH 3T3 cells. Soluble FGFRs were Fc chimeras obtained from R&D Systems (Minneapolis, MN). The BrdUrd assay was performed according to the manufacturer’s specifications (Roche Molecular Biochemicals, Indianapolis, IN) using a 4 h BrdUrd incorporation time.

Focus Formation Assay
NIH 3T3 cells were transfected using LipofectAMINE Plus according to the manufacturer’s protocol (Life Technologies, Inc.). Cells were supplemented with 10% CS (Life Technologies, Inc.) 5 h after transfection. Two days after transfection, cells were transferred to 90-mm dishes and cultured for 2 weeks in DMEM + 5% CS. The cells were then stained with a 0.2% crystal violet/70% ethanol solution and photographed. Each 90-mm dish represents half of the cells from a 35-mm dish that had been transfected with 1.5 µg of plasmid DNA.

In Vivo Nude Mouse Tumor Assay
NIH 3T3 cells were transfected with LipofectAMINE Plus according to the manufacturer’s protocol (Life Technologies, Inc.). Cells were supplemented with 10% CS (Life Technologies, Inc.) 5 h after transfection. Two days after transfection, pFGF-20-transfected cells were split into DMEM/10% CS supplemented with 600 µg/ml Geneticin (Life Technologies, Inc.), and pIg{kappa}-FGF-20-transfected cells were split into DMEM/10% CS supplemented with 500 µg/ml hygromycin B (Life Technologies, Inc.). To generate control cells, NIH 3T3 cells were transfected with the appropriate empty vectors and selected as described above. After 2 weeks of culture, pools of transfected cells were trypsinized, neutralized with DMEM/10% CS, washed with PBS, and counted. One million cells in PBS were injected into the lateral subcutis of female athymic nude mice. Tumors were measured with calipers every 3–4 days.

Radiation Hybrid Mapping
Mapping was performed as described previously (17) . Logarithm of odds (LOD) scores were >3.0.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Identification of the Human FGF-20 Gene and Isolation of Its Corresponding cDNA.
The process of genomic mining was used to identify a gene predicted to encode a novel human FGF family member. This gene is comprised of 3 exons and 2 introns (Fig. 1)Citation . The DNA sequence predicts an ORF of 211 amino acids, with an in-frame stop codon 117 bp upstream of the initiator methionine. PROSITE (18) computer analysis predicts a characteristic FGF signature motif located between amino acids 125 and 148. The DNA segment used to mine the gene was annotated as mapping to chromosome 8p21.3-p22. We confirmed and further refined this location by radiation hybrid analysis, placing the gene 1.61 cR from marker WI-5104 on chromosome 8 (see "Materials and Methods").



View larger version (46K):
[in this window]
[in a new window]
  		Fig. 1. Nucleotide and deduced amino acid sequence of FGF-20. The initiation and stop codons are in bold and boxed, and an in-frame stop codon residing in the 5'-UTR is underlined. A FGF signature motif, G-x-[LI]-x-[STAGP]-x(6,7)-[DE]-C-x-[FLM]-x-E-x(6)-Y, identified by a PROSITE (18) search is double-underlined, and intron/exon boundaries are depicted with arrows. Introns 1 and 2 are 5988- and 2338-bp long, respectively. The 5'-UTR sequence was derived from public ESTs and is not shown in its entirety.

 
To verify that the ORF identified by genomic mining was correct, we used PCR to obtain a cDNA corresponding to the predicted genomic clone. The nucleotide sequence of this cDNA precisely matched that of the predicted gene. The protein encoded by the cDNA is most closely related to Xenopus FGF-20 (XFGF-20), human FGF-9, and human FGF-16 (80%, 70% and 64% amino acid identity, respectively; Fig. 2Citation ). Based on the strong homology with XFGF-20, the gene we identified may represent its human orthologue. Therefore, we provisionally name the gene FGF-20.



View larger version (50K):
[in this window]
[in a new window]
  		Fig. 2. Homology of FGF-20 with other FGF family members. FGF-20 was aligned with human FGF-9, human FGF-16, and Xenopus FGF-20 (GenBank accession numbers D14838, AB009391, and AB012615, respectively) by ClustalW analysis (32) . The internal hydrophobic domain involved in FGF-9 secretion (22) spans amino acids 95–120 of the FGF-9 sequence.

 
FGF-20 mRNA Expression in Normal Human Tissues and Tumor Cell Lines.
We examined the mRNA expression profile of FGF-20 using real-time quantitative PCR (15) . The primer/probe set used was designed to be FGF-20 specific and, as such, should not detect other known FGF family members. The results from a representative experiment are shown in Fig. 3Citation ; similar results were obtained with an independent primer/probe set (data not shown). In 39 human tissues examined, FGF-20 was found to be most highly expressed in the brain, particularly in the cerebellum. Other tissues expressed much lower levels of FGF-20. Of the 54 human tumor cell lines examined, FGF-20 was found to be most highly expressed in a lung carcinoma (LX-1), a colon carcinoma (SW-480), and a gastric carcinoma (NCI-N87). It is noteworthy that the level of FGF-20 expression in the LX-1 lung carcinoma was higher than that of any normal tissue.



View larger version (33K):
[in this window]
[in a new window]
  		Fig. 3. Expression analysis of FGF-20 using real-time quantitative PCR. Real-time quantitative PCR analysis was performed using FGF-20-specific TaqMan reagents on normalized RNA derived from normal human tissue samples (top panel) or tumor cell lines (bottom panel). The particular tumor cell lines used are indicated and grouped according to tumor type (BR, breast; OV, ovary; CNS, central nervous system; CO, colon; GA, gastric; RE, renal; LI, liver; LU, lung; PA, pancreas; PR, prostate; ME, melanoma). Expression is plotted as a percentage of the sample exhibiting the highest level of expression.

 
Expression of Recombinant FGF-20 Protein.
Although it possesses a moderately hydrophobic NH2 terminus (data not shown), FGF-20 lacks a classical NH2-terminal signal sequence as predicted by PSORT (19) and SIGNALP (20) computer algorithms. The two FGF family members to which FGF-20 is most similar, FGF-9 and FGF-16, also lack classical NH2-terminal signal sequences but are nonetheless secreted (6 , 21, 22, 23) . To determine whether FGF-20 is also secreted, the cDNA encoding the full-length FGF-20 protein was subcloned into a mammalian expression vector that we designate pFGF-20. This construct incorporates an epitope tag (V5) and a polyhistidine tag into the COOH terminus of the protein to aid in its identification and purification, respectively, and should generate a protein of Mr ~27,000. After transient transfection into NIH 3T3 cells and Western blotting with an anti-V5 antibody, a doublet of Mr ~23,000 and Mr ~27,000 was identified in conditioned media from cells transfected with pFGF-20 (Fig. 4Citation , Lane 6). As expected, conditioned media from cells transfected with control vector did not react with the antibody (Fig. 4Citation , Lane 2). These results indicate that FGF-20 is secreted, although it lacks a classical signal peptide.



View larger version (41K):
[in this window]
[in a new window]
  		Fig. 4. Western analysis of FGF-20. Samples from NIH 3T3 cells transiently transfected with the indicated construct were examined by Western analysis under reducing conditions using anti-V5 antibody. CM, conditioned media; P, cell pellet. Molecular weight markers are indicated on the right.

 
In addition to its secretion into conditioned media, we also found FGF-20 associated with the cell pellet/ECM (Fig. 4Citation , Lane 5). Because FGFs are known to bind to HSPGs present on the surface of cells and in the ECM, we investigated the possibility that FGF-20 was being sequestered in this manner. To this end, we treated FGF-20-transfected cells with suramin, a compound known to disrupt low-affinity interactions between growth factors and HSPGs (24) . After suramin treatment, we found that a significant quantity of FGF-20 could in fact be released from the cell surface/ECM, indicating that HSPGs are likely to play a role in sequestering this protein (data not shown).

With the goal of enhancing protein secretion, we generated a construct (pIg{kappa}-FGF-20) in which the FGF-20 cDNA was fused in frame with a cleavable NH2-terminal secretory signal sequence derived from the immunoglobulin {kappa} gene. The resulting protein also contained COOH-terminal V5 and polyhistidine tags as described above for pFGF-20. After transfection into NIH 3T3 cells, a doublet of Mr ~22,000 and Mr ~28,000 was found in the cell pellet and in the conditioned medium (Fig. 4Citation , Lanes 3 and 4, respectively). We found that whereas the pIg{kappa}-FGF-20 construct enhanced overall FGF-20 protein production relative to the pFGF-20 construct, the proportion of secreted to total protein was approximately the same for each construct. This indicates that native FGF-20 protein is efficiently secreted.

Induction of DNA Synthesis and Growth by Recombinant FGF-20 Protein.
To obtain partially purified protein for biological assays, 293 cells were transiently transfected with pIg{kappa}-FGF-20 and treated with suramin to release sequestered protein. This material was then enriched by TALON affinity chromatography, taking advantage of the COOH-terminal polyhistidine tag. Recombinant FGF-20 was tested for its ability to induce DNA synthesis in a BrdUrd incorporation assay. We found that FGF-20 induced DNA synthesis in NIH 3T3 mouse fibroblasts and CCD-1070sk normal human skin fibroblasts (Fig. 5A)Citation , CCD-1106 KERTr human keratinocytes and 786-O human renal carcinoma cells (Fig. 5B)Citation , and MG-63 human osteosarcoma cells and human breast epithelia cells (data not shown). DNA synthesis was generally induced at a half-maximal concentration of ~10 ng/ml. In contrast, protein purified from cells transfected with control vector did not induce DNA synthesis in any of these cell types (data not shown).



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 5. Biological activity of recombinant FGF-20: effects on DNA synthesis and cell growth. A and B, BrdUrd incorporation assay. NIH 3T3 mouse fibroblasts (red circles) and CCD-1070sk human fibroblasts [green squares; (A)] and CCD-1106 KERTr human keratinocytes (red circles) and 786-O human renal carcinoma cells [green squares (B)] were serum starved, incubated with partially purified FGF-20 for 18 h, and analyzed by a BrdUrd incorporation assay. Data points represent the average of triplicate wells. Background incorporation from cells incubated in the absence of FGF-20 was subtracted from all samples. C, growth assay. NIH 3T3 cells were incubated with serum-free media supplemented with the indicated factor and counted after 48 h. FGF-20 was used at 150 ng/ml. Data points represent the average of duplicate wells.

 
To determine whether recombinant FGF-20 sustained cell growth, NIH 3T3 cells were cultured with FGF-20 or control protein for 48 h and then counted (Fig. 5C)Citation . We found that FGF-20 induced a ~3-fold increase in cell number relative to control protein in this assay. In addition to reaching a higher cell density, NIH 3T3 cells cultured in the presence of FGF-20 exhibited a disorganized pattern of growth, indicating a loss of contact inhibition, and individual cells were found to be spindly and refractile (data not shown). These results show that FGF-20 acts as a growth factor and suggest that recombinant FGF-20 may mediate the morphological transformation of NIH 3T3 cells (see below).

Ectopic Expression of FGF-20 in NIH 3T3 Cells Induces Their Morphological Transformation in Vitro and Tumorigenicity in Nude Mice.
To assess the effect of ectopic FGF-20 expression on cell growth in culture, NIH 3T3 cells were transfected with FGF-20 expression plasmids (pFGF-20 and pIg{kappa}-FGF-20) or control vector. We found that cells transfected with either FGF-20 expression vector generated foci of morphologically transformed cells ~2 weeks after transfection, whereas cells transfected with control vector retained their normal morphology (Fig. 6)Citation . The pIg{kappa}-FGF-20 construct proved to be significantly more efficient at foci formation than the pFGF-20 construct.



View larger version (47K):
[in this window]
[in a new window]
  		Fig. 6. In vitro foci formation. NIH 3T3 cells transfected with the indicated constructs were cultured for 2 weeks in DMEM/5% CS, stained, and photographed. The foci generated by the pIg{kappa}-FGF-20 construct are numerous, but they are small due to overcrowding.

 
To determine whether ectopic FGF-20 expression induced the tumorigenicity of NIH 3T3 cells, stable transfectants were generated and injected s.c. into nude mice. By 11 days, all of the animals injected with pFGF-20- or pIg{kappa}-FGF-20-transfected cells possessed rapidly growing tumors, whereas none of the animals injected with control cells had tumors (Fig. 7)Citation . Tumors generated from pIg{kappa}-FGF-20-transfected cells formed with a decreased latency of several days relative to tumors generated from pFGF-20-transfected cells.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 7. In vivo tumor formation. NIH 3T3 cells stably transfected with the indicated constructs were injected into the subcutis of athymic nude mice, and mice were examined for tumor formation over a 2-week period. Tumor volume was obtained by measuring the tumors with calipers and using the following formula: . A minimum of four animals were used for each data point. All animals injected with pFGF-20- or pIg{kappa}-FGF-20-transfected cells developed rapidly growing tumors by 1–2 weeks. No animals injected with control cells that had been transfected with the appropriate empty vectors developed tumors by 3 weeks, at which time the experiment was terminated.

 
Receptor Binding Specificity of FGF-20.
To determine the receptor binding specificity of FGF-20, we examined the effect of soluble FGFRs on the induction of DNA synthesis in NIH 3T3 cells by recombinant FGF-20 (Fig. 8)Citation . Soluble receptors for FGFR1ß(IIIc), FGFR2{alpha}(IIIb), FGFR2ß(IIIb), FGFR2{alpha}(IIIc), FGFR3{alpha}(IIIc), and FGFR4 were used. We found that soluble forms of each of these FGFRs were able to specifically inhibit the biological activity of FGF-20. Complete or nearly complete inhibition was obtained with soluble FGFR2{alpha}(IIIb), FGFR2ß(IIIb), FGFR2{alpha}(IIIc), and FGFR3{alpha}(IIIc), whereas partial inhibition was achieved with soluble FGFR1ß(IIIc) and FGFR4. None of the soluble receptor reagents interfered with the induction of DNA synthesis by PDGF-BB (Fig. 8)Citation , thereby demonstrating their specificity. The integrity of each soluble receptor reagent was demonstrated by showing their ability to inhibit the induction of DNA synthesis by aFGF, a factor known to interact with all of the FGFRs under analysis (Fig. 8)Citation .



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 8. Receptor binding specificity of FGF-20. NIH 3T3 cells were serum-starved, incubated with the indicated factor (green square, PDGF-BB; blue triangle, aFGF; red circle, FGF-20) either alone or together with the indicated soluble FGFR, and analyzed by a BrdUrd incorporation assay. Data points represent the average obtained from triplicate wells and are represented as the percentage of BrdUrd incorporation relative to cells receiving factor alone.

 

   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
FGF-20 represents a novel human gene whose existence was predicted from homology-based mining of genomic DNA. FGF-20 localizes to chromosome 8 and is the first FGF family member identified to date on this chromosome. Like most FGFs (1) , the genomic structure of FGF-20 reveals its protein coding region to be composed of three exons. We verified the existence of FGF-20 by cDNA cloning and found that the encoded protein is most homologous (80% amino acid identity) to XFGF-20. Other human/Xenopus orthologues within the FGF family have been reported to possess identities as high as 92% [e.g., FGF-9 (25) ] and as low as 71% [e.g., FGF-3 (26) ]. These results suggest that XFGF-20 and the gene we identified may represent orthologues.

A search of public databases reveals only four ESTs corresponding to FGF-20, all of which are found within the genes’ first exon. Two of these ESTs were obtained from a colon carcinoma, consistent with mRNA expression analysis showing FGF-20 to be highly expressed in the SW-480 colon carcinoma cell line. The other ESTs were from mixed organs, some of which were not examined in the expression analysis. The poor coverage of FGF-20 by public ESTs may indicate that FGF-20 represents a rarely transcribed gene. These results confirm the utility of genomic mining for identifying such genes.

It is likely that the cDNA we identified encodes full-length FGF-20 for several reasons (see Fig. 1Citation ). First, the earliest methionine in the FGF-20 protein resides within a context predicted to be competent for initiating translation (14) . Second, an in-frame stop codon is predicted to reside 117 nucleotides upstream of this methionine. Finally, the first four amino acids of FGF-20 are identical to those of FGF-9 and XFGF-20 (see Fig. 2Citation ).

FGF-20 has a moderately hydrophobic NH2 terminus that is not predicted to function as a classical signal sequence. However, many FGFs, including FGF-9 (6 , 22 , 23) and FGF-16 (21) , lack classical NH2-terminal signal sequences and are nonetheless secreted. The observation that native FGF-20 is detected in the conditioned medium of cells after transfection suggests that FGF-20 is also secreted. In the case of FGF-9, two domains have been found to play a role in its secretion: (a) a weakly hydrophobic NH2-terminal region (23) ; and (b) a strongly hydrophobic central region (22) . Interestingly, FGF-20 also possesses a strongly hydrophobic internal domain that shares 92% amino acid identity with the region shown to be important for FGF-9 secretion (Fig. 2)Citation . Thus, it is possible that FGF-20 and FGF-9 share a common secretory mechanism.

Recombinant FGF-20 protein stimulates DNA synthesis and cell proliferation, effects that are likely mediated via high-affinity binding of FGF-20 to a cell surface receptor(s) and modulated via low-affinity interactions with HSPGs. Suramin extraction data suggest that FGF-20 binds to HSPGs present on the cell surface/ECM, and the results obtained with soluble FGFRs demonstrate that FGF-20 interacts with FGFRs encoded by each of the four known FGFR genes (Fig. 8)Citation . Thus, FGF-20 is a promiscuous ligand for FGFRs.

In a survey of normal human tissues, we found FGF-20 RNA expressed predominantly in the cerebellum (Fig. 3)Citation . The expression of FGF-20 in the cerebellum was confirmed in an independent experiment using real-time quantitative PCR on cDNAs derived from various regions of the brain (data not shown). The tissue-specific expression of FGF-20 implies that it may play an important role in the brain. Studies in rats indicate that FGF-9 is also expressed in a cerebellum-specific fashion at the protein level (27) and in the brain and kidney at the RNA level (28) . Whereas it is possible that the reagents used in the FGF-9 studies cross-react with FGF-20, it is likewise possible that these two highly related genes have retained similar cerebellum-specific transcriptional regulatory elements subsequent to diverging from a common ancestral gene.

In addition to normal cerebellum, we found FGF-20 expression in several human tumor cell lines including carcinomas of the lung, stomach, and colon (Fig. 3)Citation . Additional analysis has shown FGF-20 to be expressed in 5 of 15 colon and 2 of 8 gastric cancer cell lines (data not shown). The lack of FGF-20 expression in normal lung, stomach, and colon and its presence in tumor lines from these tissues indicate that these cancer cell lines exhibit an inappropriate expression of FGF-20. It is interesting to note that the chromosomal region to which FGF-20 maps is commonly altered in colorectal, lung, and gastric carcinomas (29 , 30) . Whereas the role that FGF-20 expression plays in these human malignancies remains to be investigated, it is possible that the establishment of a FGF-20-driven autocrine growth loop in these cells contributes to their initial tumorigenic conversion and/or their subsequent expansion. The possibility that FGF-20 secretion by tumor cells stimulates their in vivo growth via paracrine effects on stromal cells also deserves consideration.

Ectopic expression of FGF-20 in NIH 3T3 cells induces their in vitro transformation and in vivo tumorigenicity, effects that are mediated by both native FGF-20 (construct pFGF-20) and FGF-20 expressed with a heterologous immunoglobulin {kappa} signal sequence at its NH2 terminus (construct pIg{kappa}-FGF-20). However, it should be noted that pIg{kappa}-FGF-20 is more oncogenically active than pFGF-20, as evidenced by its greater in vitro transforming ability (Fig. 6)Citation and in vivo tumorigenicity (Fig. 7)Citation . The superior oncogenicity of pIg{kappa}-FGF-20 relative to pFGF-20 is likely due to the fact that in NIH 3T3 cells, the former construct generates significantly more FGF-20 protein (both secreted and cell-associated) than the latter construct (Fig. 4)Citation .

Like FGF-20, other FGFs have been shown to transform cells after ectopic expression, and in some cases the blockade of FGF signaling has been shown to suppress cell transformation (6 , 31) . We are currently generating the necessary reagents to further validate the role of FGF-20 in human malignancy and are exploring the possibility of harnessing its growth-promoting properties for therapeutic purposes.


   ACKNOWLEDGMENTS
 
We thank Alison Bendele for performing the nude mouse tumor assays; Steven Colman and Kevin Thompson for performing the radiation hybrid mapping; Amanda Mezick, Nancy Twomlow, and Liang-Xian Cao for technical assistance; and William McDonald, Rajeev Chillakuru, and Stacey Minskoff for helpful discussions.


   FOOTNOTES
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 To whom requests for reprints should be addressed. Present address: CuraGen Corporation, 322 East Main Street, Branford, CT 06405. Phone: (203) 871-4356; Fax: (203) 315-3301; E-mail: mjeffers{at}curagen.com Back

2 The abbreviations used are: FGF, fibroblast growth factor; FGFR, FGF receptor; ECM, extracellular matrix; HSPG, heparan sulfate proteoglycan; ORF, open reading frame; BrdUrd, bromodeoxyuridine; UTR, untranslated region; EST, expressed sequence tag; aFGF, acidic FGF; PDGF, platelet-derived growth factor; CS, calf serum. Back

Received 9/19/00. Accepted 1/26/01.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Szebenyi G., Fallon J. F. Fibroblast growth factors as multifunctional signaling factors. Int. Rev. Cytol., 185: 45-106, 1999.[Medline]
  2. Klint P., Claesson-Welsh L. Signal transduction by fibroblast growth factor receptors. Front. Biosci., 4: 165-177, 1999.
  3. Xu X., Weinstein M., Li C., Deng C. Fibroblast growth factor receptors (FGFRs) and their roles in limb development. Cell Tissue Res., 296: 33-43, 1999.[Medline]
  4. Danilenko D. M. Preclinical and early clinical development of keratinocyte growth factor, an epithelial-specific tissue growth factor. Toxicol. Pathol., 27: 64-71, 1999.[Abstract/Free Full Text]
  5. Basilico C., Moscatelli D. The FGF family of growth factors and oncogenes. Adv. Cancer Res., 59: 115-165, 1992.[Medline]
  6. Matsumoto-Yoshitomi S., Habashita J., Nomura C., Kuroshima K., Kurokawa T. Autocrine transformation by fibroblast growth factor 9 (FGF-9) and its possible participation in human oncogenesis. Int. J. Cancer, 71: 442-450, 1997.[Medline]
  7. Lorenzi M., Horii Y., Yamanaka R., Sakaguchi K., Miki T. FRAG1, a gene that potently activates fibroblast growth factor receptor by C-terminal fusion through chromosomal rearrangement. Proc. Natl. Acad. Sci. USA, 93: 8956-8961, 1996.[Abstract/Free Full Text]
  8. Li Y., Mangasarian K., Mansukhani A., Basilico C. Activation of FGF receptors by mutations in the transmenbrane domain. Oncogene, 14: 1397-1406, 1997.[Medline]
  9. Gerwins P., Skoldenberg E., Claesson-Welsh L. Function of fibroblast growth factors and vascular endothelial growth factors and their receptors in angiogenesis. Crit. Rev. Oncol. Hematol., 34: 185-194, 2000.[Medline]
  10. Gasparini G. The rationale and future potential of angiogenesis inhibitors in neoplasia. Drugs, 58: 17-38, 1999.
  11. Altschul S. F., Gish W., Miller W., Myers E. W., Lipman D. J. Basic local alignment search tool. J. Mol. Biol., 215: 403-410, 1990.[Medline]
  12. Koga C., Adati N., Nakata K., Mikoshiba K., Furuhata Y., Sata S., Tei H., Sakati Y., Kurokawa T., Shiokawa K., Yokoyama K. K. Characterization of a novel member of the FGF family, XFGF-20, in Xenopus laevis. Biochem. Biophys. Res. Commun., 261: 756-765, 1999.[Medline]
  13. Mount S. M. A catalogue of splice junction sequences. Nucleic Acids Res., 10: 459-472, 1982.[Abstract/Free Full Text]
  14. Kozak M. Interpreting cDNA sequences: some insights from studies on translation. Mamm. Genome, 7: 563-574, 1996.[Medline]
  15. Heid C. A., Stevens J., Livak K. J., Williams P. M. Real time quantitative PCR. Genome Res., 6: 986-994, 1996.[Abstract/Free Full Text]
  16. Sambrook J., Fritsch E. F., Maniatis T. Molecular Cloning: A Laboratory Manual 2nd ed. Cold Spring Harbor Laboratory Press Cold Spring Harbor, NY 1989.
  17. Steen R. G., Kwitek-Black A. E., Glenn C., Gullings-Handley J., Van Etten W., Atkinson O. S., Appel D., Twigger S., Muir M., Mull T., Granados M., Kissebah M., Russo K., Crane R., Popp M., Peden M., Matise T., Brown D. M., Lu J., Kingsmore S., Tonellato P. J., Rozen S., Slonim D., Young P., Knoblauch M., Provoost A., Ganten D., Colman S. D., Rothberg J., Lander E. S., Jacob H. J. A high-density integrated genetic linkage and radiation hybrid map of the laboratory rat. Genome Res., 9: 793-798, 1999.[Free Full Text]
  18. Bucher P., Bairoch A. A generalized profile syntax for biomolecular sequence motifs and its function in automatic sequence interpretation. Ismb., 2: 53-61, 1994.
  19. Nakai K., Kanehisa M. A knowledge base for predicting protein localization sites in eukaryotic cells. Genomics, 14: 897-911, 1992.[Medline]
  20. Nielsen H., Engelbrecht J., Brunak S., von Heijne G. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng., 10: 1-6, 1997.[Abstract/Free Full Text]
  21. Miyake A., Konishi M., Martin F. H., Hernday N. A., Ozaki K., Yamamoto S., Mikami T., Arakawa T., Itoh N. Structure and expression of a novel member, FGF-16, of the fibroblast growth factor family. Biochem. Biophys. Res. Commun., 243: 148-152, 1998.[Medline]
  22. Miyakawa K., Hatsuzawa K., Kurokawa T., Asada M., Kuroiwa T., Inamura T. A hydrophobic region locating at the center of fibroblast growth factor-9 is crucial for its secretion. J. Biol. Chem., 274: 29352-29357, 1999.[Abstract/Free Full Text]
  23. Revest J-M., DeMoerlooze L., Dickson C. Fibroblast growth factor 9 secretion is mediated by a non-cleaved amino-terminal signal sequence. J. Biol. Chem., 275: 8083-8090, 2000.[Abstract/Free Full Text]
  24. La Rocca R. V., Stein C. A., Myers C. E. Suramin: prototype of a new generation of antitumor compounds. Cancer Cells, 2: 106-115, 1990.[Medline]
  25. Song J., Slack J. M. W. XFGF-9: a new fibroblast growth factor from Xenopus embryos. Dev. Dyn., 206: 427-436, 1996.[Medline]
  26. Kiefer P., Mathieu M., Clode M. J., Peters G., Dickson C. FGF3 from Xenopus laevis. EMBO J., 12: 4159-4168, 1993.[Medline]
  27. Matsumoto-Yoshitomi S., Kuroshima K., Nomura C., Habashita J., Kurokawa T. Construction of a sensitive enzyme immunoassay for human fibroblast growth factor 9. Hybridoma, 15: 299-305, 1996.[Medline]
  28. Miyamoto M., Naruo K-I., Seko C., Matsumoto S., Kondo T., Kurokawa T. Molecular cloning of a novel cytokine cDNA encoding the ninth member of the fibroblast growth factor family, which has a unique secretion property. Mol. Cell. Biol., 13: 4251-4259, 1993.[Abstract/Free Full Text]
  29. Emi M., Fujiwara Y., Nakajima T., Tsuchiya E., Tsuda H., Hirohashi S., Maeda Y., Tsuruta K., Miyaki M., Nakamura Y. Frequent loss of heterozygosity for loci on chromosome 8p in hepatocellular carcinoma, colorectal cancer, and lung cancer. Cancer Res., 52: 5368-5372, 1992.[Abstract/Free Full Text]
  30. Baffa R., Santoro R., Bullrich F., Mandes B., Ishii H., Croce C. M. Definition and refinement of chromosome 8p regions of loss of heterozygosity in gastric cancer. Clin. Cancer Res., 6: 1372-1377, 2000.[Abstract/Free Full Text]
  31. Li Y., Basilico C., Mansukhani A. Cell transformation by fibroblast growth factors can be suppressed by truncated fibroblast growth factor receptors. Mol. Cell. Biol., 14: 7660-7669, 1994.[Abstract/Free Full Text]
  32. Thompson J. D., Higgins D. G., Gibson T. J. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res., 22: 4673-4680, 1994.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
K. Miyakawa and T. Imamura
Secretion of FGF-16 Requires an Uncleaved Bipartite Signal Sequence
J. Biol. Chem., September 12, 2003; 278(37): 35718 - 35724.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
E. Alvarez, E. G. Fey, P. Valax, Z. Yim, J. D. Peterson, M. Mesri, M. Jeffers, M. Dindinger, N. Twomlow, A. Ghatpande, et al.
Preclinical Characterization of CG53135 (FGF-20) in Radiation and Concomitant Chemotherapy/Radiation-induced Oral Mucositis
Clin. Cancer Res., August 1, 2003; 9(9): 3454 - 3461.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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 Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jeffers, M.
Right arrow Articles by Lichenstein, H. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jeffers, M.
Right arrow Articles by Lichenstein, H. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online