The S130K fibroblast growth factor–1 mutant induces heparin-independent proliferation and is resistant to thrombin degradation in fibrin glue☆☆☆★★★
Article Outline
Abstract
Objective: Site-directed mutagenesis is an important technique that can alter cytokine function, thereby eliciting desired responses. S130K is a mutation of fibroblast growth factor–1 (FGF-1), with lysine replacing serine in the heparin-binding site. We measured molecular stability and mitogenic activity of FGF-1 and S130K, both in the media and when suspended in fibrin glue (FG), on smooth muscle cells (SMCs) and endothelial cells (ECs) to determine if the mutation altered the function and potential clinical applicability. Methods: EC and SMC proliferation of soluble FGF-1 or S130K at 0, 0.1, 1, 10, or 100 ng/mL with heparin at 0, 5, 50, or 500 units (U)/mL was measured on growth-arrested cells in serum-free media. EC and SMC proliferation assays with cells on FG containing either FGF-1 or S130K at 0, 1, 10, 100, or 1000 ng/mL in combination with heparin at 0, 5, 50 or 500 U/mL were also performed during the exponential growth phase. Molecular degradation by thrombin was measured by sodium dodecylsulfate-polyacrylamide gel electrophoresis. Results: S130K induces greater EC and SMC proliferation in the absence of heparin than FGF-1 does (P <.0001 for both the 10 and 100 ng/mL doses). S130K is also significantly more potent than FGF-1 in the presence of heparin. Heparin in the media enhances cytokine-induced SMC and EC proliferation at doses of 5 U/mL, but inhibits SMC proliferation at concentrations of 500 U/mL. For the FG data, unlike FGF-1, S130K induces EC and SMC proliferation in the absence of heparin. The addition of 5 U/mL of heparin enhances the proliferation induced by S130K. For ECs, as the heparin dose increases to 50 U/mL, proliferation decreases, as compared with the 5 U/mL concentration when either FGF-1 or S130K in the FG was compared at concentrations of 10, 100, and 1000 ng/mL (P <.01). S130K is more potent in FG than is FGF-1 both with and without heparin and exhibits maximal EC and SMC proliferation at 10 ng/mL, whereas FGF-1 activity is maximal at 100 ng/mL. Gel electrophoresis demonstrated that S130K was relatively more resistant to thrombin degradation than FGF-1. Conclusions: Site-directed mutagenesis changed the potency and the heparin dependency on cellular proliferation of FGF-1 in vitro. These techniques should allow the delivery of mutant growth factors to areas of vascular intervention to induce specific, desired responses. We believe that these studies will enhance our knowledge of the function of various regions of the FGF-1 molecule, allowing us to more precisely design increasingly more useful FGF-1 mutants. (J Vasc Surg 2000;31:382-90.)
Site-directed mutagenesis is an important technique that allows specific amino acids to be altered in a molecule. This provides opportunities to study structure/function relationships and the relative contribution of specific regions to the overall function of the molecule. In addition, changing specific amino acids may allow the design of cytokines with more useful functions than the wild-type molecule. Workers in our laboratory have extensively studied the effects of fibroblast growth factor–1 (FGF-1), also known as acidic FGF, on smooth muscle cell (SMC) and endothelial cell (EC) proliferation. We have used fibrin glue (FG) as an extended delivery system for FGF-1 on vascular grafts and angioplasty sites. Cytokines and heparin can be mixed with fibrinogen before the addition of thrombin. Thrombin cleaves the fibrinogen into fibrin, and a matrix is created that contains the cytokines and heparin. The body's fibrinolytic system slowly degrades the matrix, presumably releasing the cytokines. FGF-1 was chosen for its strong chemoattractant and mitogenic activity for ECs, whereas in the absence of heparin, FGF-1 induces minimal SMC proliferation. With the addition of small amounts of heparin, SMC proliferation in response to FGF-1 dramatically increases. However, the addition of heparin is necessary in FG to protect FGF-1 from thrombin cleavage.
The synergism of FGF-1 and heparin is variable on different cell lines. In the absence of heparin, FGF-1 has about 1% or less of the proliferative activity on human umbilical vein endothelial cells as it does in the presence of heparin.1 Other cell lines, such as bovine aortic EC,2 mouse lung capillary EC,2 and BALB/c 3T3 fibroblasts3 exhibit FGF-1–induced mitogenic activity in the absence of heparin. In addition, FGF-1 stimulates bovine aortic ECs, but the mitogenic effect is augmented by the addition of heparin.4
The interactions of heparin and FGF-1 are complex and at this time, poorly understood. Heparin, or heparin-like molecules, facilitate the binding of FGF-1 to the transmembrane FGF receptor. The FGF receptors are protein tyrosine kinases that are activated by oligomerization.5 FGF-1 molecules complex with the sugars on the polysaccharide chains of heparin. Currently, heparin is believed to form a biologically active dimer of FGF-1 that modulates the signaling through the FGF receptors.6
Our group has previously reported an FG system containing FGF-1 and heparin used on grafts implanted into the aortoiliac and thoracoabdominal aortic positions in dogs.7, 8 FG containing FGF-1 and heparin placed onto expanded polytetrafluoroethylene grafts demonstrated complete endothelial coverage at 4 weeks, which was never seen in the control grafts. However, the inner capsule thickness was also greater in the grafts treated with FGF-1 as compared with the controls. FGF-1 stimulates both EC and SMC proliferation and thereby may cause intimal hyperplasia. An optimized FG would theoretically maximize EC proliferation while minimizing SMC proliferation that may lead to intimal hyperplasia. Concern over intimal hyperplasia in the canine grafts receiving FGF-1 and FG led us to speculate that the functions of the FGF-1 molecule could be changed by site-directed mutagenesis to induce either more EC compared with SMC proliferation, or more angiogenic activity when compared with wild-type FGF-1. Molecular stability may also be enhanced, thus allowing the use of the altered molecule within FG in the absence of heparin. S130K is a mutant of FGF-1 that has a lysine at position 130 instead of a serine. Residues 122-137 of human FGF-1 constitute a major heparin-binding domain in the protein.9 By changing the heparin-binding characteristics of FGF-1, we hoped to alter the functions of the molecule. The purpose of this paper is to directly compare the activity of FGF-1 versus S130K on EC and SMC proliferation with the cytokines contained within the media and within FG, and the stability of the proteins to thrombin-induced degradation.
Methods
Animal care complied with The Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, Commission on Life Sciences, National Research Council, 1996) and the “Principles of Laboratory Animal Care” (National Institutes of Health publication no. 85-23, revised 1985).
Construction of mutant FGF-1 prokaryotic expression plasmids
Full-length (β) FGF-1 complementary DNA was a gift of Dr Mike Jaye (RhonePoulenc Rorer Central Research). The pMJ17 plasmid containing FGF-1 was digested with Eco RI and Hind III to generate a 1000–base pair fragment with the Eco RI site at the 5' end. β FGF-1 was amplified from the complementary DNA with the following primers: 5' β AATTCCATATGGCTGAAGGGAAATCACC and 3' UTR GATCAGATCTAAGTTGCTTACAAATTCAGGCTC.
The mutant was then generated with the following two consecutive polymerase chain reactions (PCRs). The first PCR incorporates the mutagenic primer CTCAAGAAGAATGGGAAGTGCAAACGC
GGTCC, 3' UTR primer and β template (from above) and is amplified with Vent DNA Polymerase (New England Biolabs, Beverly, Mass). The mutant fragment is then gel purified and used in the second PCR. The second PCR incorporates the 5' β primer, the 3' mutant fragment, and the β template and is amplified with Taq Polymerase (Boehringer Mannheim Corporation, Indianapolis, Ind). The PCR product from the second reaction is then ligated into pBluescript and transformed into XL-1 Blue competent cells. Plasmid DNA purified from individual colonies were manually sequenced by the dideoxy chain termination method using a Sequenase 2.0 kit (US Biochemicals Corporation, Cleveland, Ohio) and [35S]dATP (1000 Ci/mmol, Amersham Corporation, Arlington Heights, Ill) to verify mutation. The mutant was then digested out of pBluescript with Nde I and Bgl II and directionally ligated into the Pet3C expression vector.
Production of wild-type and mutant recombinant FGF-1 and subsequent purification were performed as described.9
Endothelial and SMC harvest
Adult, mongrel dogs were anesthetized with thiopental sodium (Abbott Laboratories, Morris Plains, NJ), intubated, and ventilated. Anesthesia was maintained with nitrous oxide and halothane. Bilateral neck incisions were made. For ECs, bilateral external jugular veins were removed, inverted, and processed per our previously reported protocol.11 Briefly, the veins were sequentially placed into 0.05% trypsin/ethylenediamine tetraacetic acid (EDTA) 0.53 mmol/L (Gibco, Grand Island, NY) and collagenase 100 units (U)/mL (Gibco) at 37°C for 10 minutes each. After the veins were discarded, the trypsin and collagenase solutions were centrifuged at 1000 revolutions per minute (rpm) for 10 minutes. The supernatants were discarded, and the cells were suspended in 5 mL of EC growth media consisting of M-199 (Gibco) supplemented with 10% fetal bovine serum (FBS) (Hyclone, Logan, Utah), FGF-1 5 ng/mL (American Red Cross, Rockville, Md), bovine lung heparin 5 U/mL (Pharmacia & Upjohn, Kalamazoo, Mich), 100 U/mL penicillin, and 100 μg streptomycin (Gibco). The ECs were plated onto a fibronectin (American Red Cross)–coated T-25 culture flask (2.5 μg/cm2) and incubated at 37°C in a 5% humidified carbon dioxide chamber. The EC growth media was changed every 2 to 3 days, and confluent cells were passaged using trypsin-EDTA. EC identity was confirmed using dual staining with Factor VIII (Dako Corp, Carpenteria, Calif) and α-actin (Sigma Chemical Company, St Louis, Mo) antibody immunofluorescent staining. Only EC cultures exhibiting 95% or more positive Factor VIII staining and 5% or less positive α-actin staining were used for the proliferation assays. ECs were used within passages 1-4.
SMCs were obtained from canine carotid arteries using a previously published9 explant technique. Briefly, the carotid arteries were opened longitudinally, and the intima and adventitia were removed by scraping and dissecting with a scalpel. The medial layer was minced and placed into SMC growth media consisting of Dulbecco modified Eagle medium (DMEM) supplemented with 10% FBS (Hyclone), 10 mmol/L L-nonessential amino acids, 100 U/mL penicillin, 100 μg/mL streptomycin, 100 mmol/L L-sodium pyruvate, and 50 μg/mL gentamicin (all from Gibco unless otherwise stated). Primary SMC migrating from the explants were used in all experiments. SMC identity was confirmed using immunofluorescent staining with α-actin antibody (Sigma Chemical Company) and only cultures exhibiting 95% or more positive staining were used in the proliferation assays.
EC and SMC proliferation assays
SMCs, 10,000 cells per well, were plated into 96-well polystyrene plastic plates (Becton Dickinson, Lincoln Park, NJ) in 200 μL of SMC growth media and allowed to proliferate for 3 days until the wells were approximately 80% confluent. The media was removed, the cells were washed with phosphate-buffered saline (PBS), and 200 μL of serum-free quiescent media consisting of DMEM:F12 (Gibco), penicillin/streptomycin (Gibco), insulin 1 μmol/L, L-ascorbic acid 0.2 mmol/L, and transferrin 5 μg/mL (all from Sigma Chemical Company) were placed on each well. After 24 hours, 50 μL of FBS (positive control), PBS (negative control), or test solutions were placed on the wells. After 48 hours of the cells being in quiescent media (24 hours after the addition of cytokines), 1 μCi per well of 3H-thymidine (NEN Life Science Products, Boston, Mass) was placed on the wells, and the plates were processed 24 hours later. The media was removed, and the wells were washed with 0.9% saline to process. The cells were fixed in 100% methanol (Fischer Scientific, Fair Lawn, NJ) for 10 minutes and lysed with distilled water. The DNA was precipitated with 5% trichloroacetic acid (Sigma Chemical Company). The cell lysates were washed with distilled water, and the DNA was solubilized by adding 100 μL of 0.3 mol/L sodium hydroxide (Sigma Chemical Company). This solution was placed into 10 mL of scintillation fluid that contained 20 μL of acetic acid (Fisher Scientific) to avoid opacification of the scintillation fluid with an accompanying dampening of the counts per minute (CPM).
Test solutions (50 μL) were placed on the cells and contained various amounts of FGF-1, S130K, and heparin, so that the final concentration in the well was the stated value. S130K or FGF-1 at 0, 0.1, 10, or 100 ng/mL with heparin 0, 5, 50, or 500 U/mL was placed on the SMCs. Each plate contained a column of wells that received 50 μL of FBS (positive control). The data were normalized by dividing the CPM for each well by the average of the positive control wells on that plate and multiplying by 100% (percent positive control). The variability among the positive control wells was typically less than 10%. There were six replicates per plate for each treatment, and the experiments were performed in duplicate (n = 12 for each observation). The data are expressed as mean percent positive control ± standard deviation.
EC assays were performed in an identical fashion to the SMC assays except that the plates were coated with fibronectin (American Red Cross) 2.5 μg/cm2. The growth media contained M199, 10% FBS, and penicillin/streptomycin. The quiescent media contained Iscove's Modified Dulbecco's Medium (Gibco), penicillin/streptomycin, L-glutamine 2 mmol/L (Gibco), insulin-transferrin-selenium 1x (Gibco), bovine serum albumin 5% (Sigma), and 2-mercaptoethanol 1x (Gibco). ECs were plated at 10,000 cells per well and maintained in growth media for 3 days. Quiescent media was placed on the cells, and 1 day later test solutions were added to the wells. 3H-thymidine was added 24 hours after the test solutions, and the plates were processed in an identical fashion as in the SMC assays.
FG plate preparation
FG was made by reconstituting lyophilized human fibrinogen (American Red Cross) prepared from donor-pooled plasma with 0.9% saline. For modified FG, heparin and FGF-1 or S130K were also added to obtain a solution with a final concentration of fibrinogen 32 mg/mL, FGF-1 or S130K 0, 1, 10, 100, or 1000 ng/mL, and heparin 0, 5, 50, or 500 U/mL. Lyophilized human thrombin (American Red Cross) was reconstituted with 0.9% saline and added to all solutions to make a final concentration of 0.32 U/mL. A solution containing fibrinogen and thrombin only was made and used as the negative control. Before polymerization, the solutions were pipetted in 30 μL aliquots into 96-well polystyrene plastic plates (Becton Dickinson) and allowed to completely polymerize, about 1 hour, before solutions were placed in the wells that contained FG.
EC and SMC proliferation assays on FG
Media was placed on the FG for 24 hours and removed before plating cells to minimize the effect of soluble FGF-1 and heparin known to be released from the FG by diffusion. This release is virtually completed by the first 24 hours.12 Confluent cells were trypsinized and plated on FG at 5000 cells per well in 200 μL of media. ECs were plated in media, which was devoid of FGF-1 and heparin (M199, 10% FBS, penicillin, and streptomycin). SMCs were plated in SMC growth media (DMEM, 10% FBS, L-sodium pyruvate, L-nonessential amino acids, penicillin, streptomycin, and gentamicin).
After plating, the cells were allowed to grow for 48 hours, then 3H-thymidine 1 μCi per well was added, and the plates were processed 24 hours later. The plates were processed by removing the media and washing the wells with 0.9% saline. The cells were fixed in 100% methanol for 10 minutes and lysed with distilled water. The DNA was precipitated with 5% trichloroacetic acid. The cell lysates were washed with distilled water, and the FG was solubilized by adding 100 μL of 0.3 mol/L sodium hydroxide and heating at 60°C for 35 minutes. This solution was placed into 10 mL of scintillation fluid that contained 20 μL of acetic acid to avoid opacification of the scintillation fluid with an accompanying dampening of the CPM.
During preliminary experiments, we found that the FG itself has a very high background that was unaffected by the composition of the FG and exhibited very little variance. Therefore, every plate contained wells with FG but no cells, and the average of these wells was subtracted from the CPM of the test wells to eliminate the high FG background. Cells plated on FG with no additives were also placed on each plate as a negative control, and the data were normalized by dividing the CPM of each well by the average of the negative control wells and multiplied by 100%. % of negative control =
After the normalization of data for each experiment, the observations from the two experiments were combined to calculate the mean of the percent negative control ± standard deviation. All proliferation experiments were performed in duplicate, and each treatment had replicates of n = 5 in each experiment (n = 10 for the combined data).
Molecular degradation by thrombin
Purified, recombinant FGF-1 and S130K (20 μg) were resuspended in 0.05 mol/L HEPES, 0.25 mol/L sodium chloride, 5 mmol/L calcium chloride, pH 7.5 (600 μL) in the presence of 6 U of human thrombin. The samples were incubated at 37°C for the indicated times. Aliquots (100 μL) were mixed with 20 μL of 6X Laemmlis' sample buffer. The samples were subjected to polyacrylamide gel electrophoresis in the presence of sodium dodecylsulfate (SDS/PAGE) with 15% polyacrylamide gels using reducing conditions. The gels were stained with Coomassie brilliant blue to visualize intact FGF-1 and thrombin-derived fragments.
Statistics
Analysis of variance with Fischer's Least Square Differences was used as an adjunct test, and an unpaired Student t test was performed. The results for these tests were reported as significant at the P level <.05.
Results
EC and SMC proliferation assays with soluble cytokines
With the cytokines soluble in the media, similar results were obtained for both EC and SMC assays. Briefly, S130K induced heparin-independent proliferation in both cell lines, whereas in the absence of heparin, FGF-1 induced only moderate proliferation on SMCs and no significant proliferation on ECs. With the addition of heparin 5 U/mL, the proliferation induced by FGF-1 was dramatically increased on both cell lines, and S130K mitogenic activity was augmented, especially on the ECs. As the heparin dose increased, proliferation decreased for both cell lines and with both S130K and FGF-1. Overall, S130K was more potent, inducing greater proliferation at lower concentrations as compared with FGF-1. For ECs, maximal proliferation occurred for both cytokines at concentrations of 1 ng/mL in the presence of heparin at 5 or 50 U/mL (Table I).
Table I. Effect of soluble S130K and FGF-1 with heparin on EC proliferation
| Heparin U/mL | 0 | 5 | 50 | 500 |
|---|---|---|---|---|
| No cytokine | 13 ± 3* | 14 ± 2 | 13 ± 2 | 3 ± 1* |
| S130K 0.1 ng/mL | 19 ± 7 | 55 ± 5†‡ | 51 ± 7† | 8 ± 4 |
 1 | 20 ± 6 | 77 ± 7†‡ | 78 ± 5† | 34 ± 3† |
| 10 | 22 ± 7† | 77 ± 5† | 80 ± 5† | 58 ± 5† |
| 100 | 34 ± 8† | 73 ± 3† | 76 ± 3† | 61 ± 3† |
| FGF-1 0.1 ng/mL | 8 ± 1 | 9 ± 7 | 14 ± 3 | 8 ± 6 |
| 1 | 10 ± 2 | 69 ± 5† | 67 ± 3† | 37 ± 9† |
| 10 | 10 ± 1 | 65 ± 4† | 72 ± 5† | 52 ± 8† |
| 100 | 15 ± 1 | 59 ± 5† | 64 ± 6† | 51 ± 9† |
| *P <.001 for PBS versus heparin 500 U/mL. †P <.001 versus same heparin dose with no cytokine. ‡P <.001 for S130K 0.1 ng/mL versus S130K 1 ng/mL, both with heparin 5 U/mL. | ||||
Table II. Effect of soluble S130K and FGF-1 with heparin on SMC proliferation
| Heparin U/mL | 0 | 5 | 50 | 500 |
|---|---|---|---|---|
| No cytokine | 6 ± 1 | 4 ± 1 | 2 ± 0 | 2 ± 0 |
| S130K 0.1 ng/mL | 7 ± 1 | 5 ± 1 | 2 ± 1 | 2 ± 0 |
| 1 | 10 ± 1 | 168 ± 23*† | 103 ± 13* | 10 ± 2 |
| 10 | 111 ± 18* | 158 ± 14* | 143 ± 21* | 69 ± 10* |
| 100 | 174 ± 26*‡ | 174 ± 18* | 131 ± 21* | 72 ± 12* |
| FGF-1 0.1 ng/mL | 6 ± 1 | 3 ± 1 | 2 ± 0 | 2 ± 0 |
| 1 | 6 ± 1 | 31 ± 5*† | 8 ± 2 | 2 ± 0 |
| 10 | 23 ± 3* | 146 ± 18* | 120 ± 16* | 66 ± 12* |
| 100 | 85 ± 15*‡ | 132 ± 13* | 113 ± 8* | 63 ± 11* |
| *P <.001 versus same heparin dose with no cytokine. †P <.001 for S130K 1 ng/mL versus FGF-1 1 ng/mL, both with heparin 5 U/mL. ‡P <.0001 for S130K 100 ng/mL versus FGF-1 100 ng/mL, both in the absence of heparin. | ||||
EC and SMC proliferation assays on FG
Once again, the results for both EC and SMC proliferation on FG were similar. Briefly, in the absence of heparin, S130K induced significant proliferation at the higher concentrations, whereas FGF-1 within FG was not significantly different from the negative controls. Thrombin in the FG degrades FGF-1, and this finding suggests that the S130K mutation may protect from thrombin degradation. The addition of heparin 5 U/mL dramatically increases the proliferation induced by both cytokines with S130K being more potent with maximal proliferation occurring at the 10 ng/mL concentration as compared with the 100 ng/mL concentration of FGF-1 for both cell lines (Tables III and IV).
Table III. Effect of S130K and FGF-1 with heparin contained within FG on EC proliferation
| Heparin U/mL | 0 | 5 | 50 |
|---|---|---|---|
| No cytokine | 100 ± 12 | 155 ± 30 | 139 ± 60 |
| S130K 1 ng/mL | 126 ± 17 | 181 ± 38 | 141 ± 44 |
| 10 | 125 ± 12 | 374 ± 100*† | 267 ± 100* |
| 100 | 149 ± 29 | 395 ± 112* | 304 ± 121* |
| 1000 | 241 ± 38* | 382 ± 125* | 287 ± 133* |
| FGF-1 1 ng/mL | 124 ± 14 | 168 ± 40 | 145 ± 56 |
| 10 | 119 ± 15 | 249 ± 62*† | 161 ± 64 |
| 100 | 123 ± 23 | 378 ± 112* | 248 ± 104* |
| 1000 | 120 ± 22 | 375 ± 103* | 261 ± 131* |
| *P <.01 versus same heparin dose with no cytokine. †P <.01 for S130K 10 ng/mL versus FGF-1 10 ng/mL, both with heparin 5 U/mL. | |||
Table IV. Effect of S130K and FGF-1 with heparin contained within FG on SMC proliferation
| Heparin U/mL | 0 | 5 | 50 | 500 |
|---|---|---|---|---|
| No cytokine | 100 ± 11 | 122 ± 22 | 97 ± 15 | 85 ± 18 |
| S130K 1 ng/mL | 116 ± 13 | 196 ± 13* | 139 ± 8* | 136 ± 30* |
| 10 | 122 ± 14 | 331 ± 38*† | 280 ± 29* | 312 ± 48* |
| 100 | 138 ± 15* | 328 ± 29* | 289 ± 29* | 335 ± 60* |
| 1000 | 226 ± 20* | 316 ± 29* | 292 ± 41* | 328 ± 51* |
| FGF-1 1 ng/mL | 123 ± 12 | 131 ± 21 | 107 ± 20 | 96 ± 28 |
| 10 | 118 ± 25 | 182 ± 13*† | 180 ± 19* | 182 ± 52* |
| 100 | 122 ± 14 | 319 ± 30* | 303 ± 36* | 331 ± 59* |
| 1000 | 127 ± 13 | 321 ± 34* | 302 ± 32* | 323 ± 58* |
| *P <.01 versus same heparin dose with no cytokine. †P <.001 for S130K 10 ng/mL versus FGF-1 10 ng/mL, both with heparin 5 U/mL. | ||||
Molecular degradation by thrombin
Incubation of FGF-1 and S130K with thrombin and subsequent analysis with SDS/PAGE revealed that thrombin degraded both FGF-1 and S130K. However, there was a relative retardation and diminution of thrombin-induced degradation with the S130K, suggesting that the mutation may somehow protect the molecule (Fig 1).
Fig. 1.
Molecular degradation of FGF-1 and S130K by thrombin. FGF-1 and S130K were incubated with thrombin 6 U for the specified times (x-axis) and then analyzed by SDS/PAGE. Molecular markers (y-axis) are shown on the left for both gels. FGF-1 and S130K are 17 kd molecules. Both molecules are susceptible to thrombin degradation, but S130K is relatively more resistant than FGF-1, as demonstrated by the delayed appearance and diminished intensity of low molecular weight fragments, suggesting that the mutation offers some protection from thrombin degradation.
Discussion
Site-directed mutagenesis is an important technique that can alter a molecule by making single or multiple amino acid substitutions. This allows for the study of the functions of different regions in a molecule and, hopefully, the ability to design a more useful molecule than the parent compound. Our group has been interested in FGF-1 because of its relatively strong EC mitogenic and chemoattractant activity. Unfortunately, FGF-1 also stimulates SMC proliferation, and this may explain the increased cellular proliferation and capsular thickness observed in our previously reported canine studies.7, 8 In this paper, we report our results with S130K, a mutant of FGF-1. Residues 122-137 of human FGF-1 constitute a major heparin-binding domain in the FGF-1 protein.9 Multiple FGF-1 mutants have been generated in the laboratory of W. H. Burgess. These studies have demonstrated that mutations of the lysines in this region (residues 126, 127, 132) result in a reduction in the mitogenic potential of the mutant FGF-1.
Substituting a glutamic acid or glycine for lysine 132 results in as much as a 1000-fold reduction in mitogenic activity when compared with the wild-type protein on various cell lines including NIH 3T3 fibroblasts, mouse keratinocytes (BALB-MK), and human ECs. Interestingly, the mutant retains its ability to bind to high-affinity cell receptors and to induce all of the early events associated with FGF-1 stimulation.14, 15 The mutant is also as potent as wild-type FGF-1 in inducing differentiation of Xenopus species embryos.16 These results establish the feasibility that site-directed mutagenesis of the protein can dissect the functions associated with FGF-1.
We reported previously that amino acids 122-137 of FGF-1 comprise a major heparin-binding domain of the protein.9 It is known that the heparin dependence of FGF-1 mitogenic activity is reduced when cysteine 131 is mutated to a serine residue.17 We have shown that mutation of lysine 132 to a glutamic acid residue reduces the affinity of FGF-1 for heparin and its mitogenic activity.14 It is unknown whether this modification affects cysteine-mediated instability of the protein. More recently we have begun to mutate nonbasic amino acids in the 122-137 region of FGF-1 to lysine residues in search of gain of function mutations. The serine 130 to lysine mutation described here is the first to exhibit altered function.
Armed with the knowledge of the importance of the lysine residues in the heparin-binding region of FGF-1, we generated S130K, which substitutes a lysine at the serine 130 position. The S130K mutant exhibits a threefold increase in mitogenic potency for BALB-MK cells and a threefold reduction in the mitogenic activity on NIH 3T3 cells as compared with wild-type FGF-1. The BALB-MK cell line demonstrates a greater heparin dependency for wild-type FGF-1 than the NIH 3T3 cell line. This suggests that heparin-independent mitogenic activity of the mutants may be predicted according to their ability to stimulate the BALB-MK cell line (W. H. Burgess, oral communication, 1998). This appears to be the case with S130K, in that it induces heparin-independent proliferation on ECs and SMCs. These qualities are likely to apply in a positive way to in vivo models of vascular healing.
With the above information, we decided that S130K would be an interesting molecule to test on EC and SMC proliferation studies and that it would be interesting to study its stimulatory properties while contained within FG. Soluble S130K and FGF-1 were tested on ECs and SMCs, and similar results were obtained for both cell lines. S130K induced heparin-independent proliferation in both cell lines, whereas FGF-1 induced moderate proliferation on SMCs and no significant proliferation on ECs in the absence of heparin. With the addition of heparin 5 U/mL, the proliferation induced by FGF-1 was dramatically increased on both cell lines, and S130K mitogenic activity was augmented, especially on the ECs. As the heparin dose increased, proliferation decreased for both cell lines and with both S130K and FGF-1. Both molecules appear to induce EC and SMC proliferation in the presence of heparin, but the dose-response curve of S130K suggests that is it more potent, inducing greater proliferation at lower concentrations as compared with FGF-1.
When the cytokines were contained within FG, once again, the results for both EC and SMC proliferation on FG were similar. In the absence of heparin, S130K induced significant proliferation at the higher concentrations, whereas FGF-1 at any concentration within FG was not significantly different from the negative controls. Thrombin in the FG degrades FGF-1, and this finding suggests that the S130K mutation may protect from thrombin degradation. Incubation of FGF-1 and S130K with thrombin and subsequent analysis by SDS/PAGE revealed that both molecules are susceptible to thrombin degradation, but S130K is relatively more resistant to thrombin degradation compared with FGF-1. The addition of heparin 5 U/mL dramatically increases the proliferation induced by both cytokines with S130K being more potent, with maximal proliferation occurring at a 10–ng/mL concentration and at the 100–ng/mL concentration of FGF-1 for both cell lines. The major difference between ECs and SMCs lies in their response to heparin in the FG. For ECs, as the heparin dose increased to 50 U/mL, overall proliferation decreased for both cytokines. At heparin 500 U/mL, the ECs died. For SMCs, heparin concentrations as high as 500 U/mL did not significantly affect proliferation as compared with the heparin–5 U/mL concentration. Earlier reports from our laboratory11 and another laboratory18 have reported that heparin 500 U/mL in FG inhibits SMC proliferation. The differences in these studies and the current study can be explained by the release of cytokines and heparin from the FG within the first 24 hours. These earlier studies plated cells on FG immediately after the FG polymerized. In this study, media was placed on the FG for 24 hours and then removed before plating cells in new media. Using radiolabeled FGF-1 and heparin, we have shown that 70% ± 1% of the FGF-1 and 59% ± 2% of the heparin in the FG was released into the overlying media in the first 24 hours with minimal release occurring thereafter (66% ± 1% and 67% ± 1% at 96 hours for FGF-1 and heparin, respectively). The cell type or absence of cells did not affect release, but there was five times more FGF-1 and four times more heparin in the media at 72 hours for the immediate versus delayed plating because of diffusive release in the first 24 hours.12
Thrombin, which is present in FG, stimulates SMC proliferation. Thus, FG itself can induce SMC proliferation. Previous studies from our laboratory, however, show that soluble thrombin concentrations of 10 U/mL or more are needed to induce SMC proliferation.19 Our FG formulation contains thrombin concentrations of 0.32 U/mL, and therefore, should not induce SMC mitogenesis.
Ultimately, the goal of this research is to find an FG mixture that will enhance endothelialization of prosthetic grafts and angioplasty sites without causing intimal hyperplasia from SMC proliferation. Although S130K is an interesting molecule with many potential uses, it does not appear to be a good candidate for our canine in vivo studies because of its proliferative effect on SMCs. Using site-directed mutagenesis to alter cytokine function is an exciting avenue of research. With S130K, we are currently exploring its angiogenic potential, and we will also continue to screen new FGF-1 mutants. These techniques should allow the delivery of mutant growth factors to areas of vascular intervention to induce specific, desired responses. We believe that these studies will enhance our knowledge of the function of various regions of the FGF-1 molecule, allowing us to more precisely design increasingly more useful FGF-1 mutants.
References
- . Human brain-derived acidic and basic fibroblast growth factors, amino terminal sequences and specific mitogenic activities. Biochem Biophys Res Comm. 1986;135:541–548
- Pure brain-derived acidic fibroblast growth factor is a potent angiogenic vascular endothelial cell mitogen with sequence homology to interleukin 1. Proc Natl Acad Sci U S A. 1985;82:6409–6413
- . Purification and characterization of acidic fibroblast growth factor from bovine brain. Proc Natl Acad Sci U S A. 1984;81:357–361
- A unique family of endothelial cell polypeptide mitogens: the antigenic and receptor cross-reactivity of bovine endothelial cell growth factor, brain-derived acidic fibroblast growth factor, and eye-derived growth factor-II. J Cell Biol. 1985;101:1623–1626
- . Signal transduction by allosteric receptor oligomerization. Trends Biochem Sci. 1988;13:443–447
- Structure of a heparin-linked biologically active dimer of fibroblast growth factor. Nature. 1998;393:812–817
- FGF-1 affixation stimulates ePTFE endothelialization without intimal hyperplasia. J Surg Res. 1994;57:596–612
- Enhanced endothelialization of expanded polytetrafluoroethylene grafts by fibroblast growth factor type 1 pretreatment. Surgery. 1992;112:244–255
- . Analysis of putative heparin-binding domains of FGF-1 using site-directed mutagenesis and peptide analogues. J Biol Chem. 1995;270:25805–25811
- Fibrin glue containing fibroblast growth factor type-1 and heparin with autologous endothelial cells reduces intimal hyperplasia in a canine carotid artery balloon injury model. J Vasc Surg. 1997;25:840–849
- . Selective stimulation of endothelial cell proliferation with inhibition of smooth muscle cell proliferation by plus heparin delivered from fibrin glue suspensions. Surgery. 1995;118:280–287
- . Modulation of vascular cell growth kinetics by local cytokine delivery from fibrin glue. J Vasc Surg. 1999;29:852–861
- . Heparin inhibits thrombin-induced mitogen-activated protein kinase signaling in arterial smooth muscle cells. J Vasc Surg. 1998;27:512–520
- . Possible dissociation of the heparin-binding and mitogenic activities of heparin-binding (acidic fibroblast) growth factor-1 from its receptor-binding activities by site-directed mutagenesis of a single lysine residue. J Cell Biol. 1990;111:2129–2138
- . Structure-function studies of heparin-binding (acidic fibroblast) growth factor-1 using site-directed mutagenesis. J Cell Biochem. 1991;45:131–138
- . Structure-function studies of FGF-1: dissociation and partial reconstitution of certain of its biological activities. Mol Reprod Dev. 1994;39:5660
- Conversion of cysteine to serine residues alters the activity, stability, and heparin dependence of acidic fibroblast growth factor. J Biol Chem. 1991;266:5842–5846
- . Vascular endothelial growth factor and heparin in a biologic glue promotes human aortic endothelial cell proliferation with aortic smooth muscle cell inhibition. Surgery. 1996;120:433–490
- . Smooth muscle cell inhibition with endothelial cell stimulation by growth regulators within a fibrin glue delivery vehicle. Surg Forum. 1994;45:355–358
☆ Competition of interest: nil.
☆☆ Supported by grants from the NIH (R01 HL41272) and the Department of Veterans' Affairs.
★ Reprint requests: Howard P. Greisler, MD, Loyola University Medical Center, Department of Surgery, 2160 South First Avenue, Maywood, IL 60153.
★★ 0741-5214/2000/$12.00 + 0 24/6/104567
PII: S0741-5214(00)90168-X
doi:10.1016/S0741-5214(00)90168-X
© 2000 Society for Vascular Surgery and International Society for Cardiovascular Surgery, North American Chapter. Published by Elsevier Inc. All rights reserved.
