Complete Inhibition of Goiter in Mice Requires Combined Gene Therapy Modification of Angiopoietin, Vascular Endothelial Growth Factor, and F
http://www.100md.com
内分泌学杂志 2005年第7期
Division of Medical Sciences (J.D.R., M.A.B., S.E., J.C.W., M.C.E.) and Cancer Research U.K. Institute for Cancer Studies (V.M.), University of Birmingham, Birmingham B15 2TT, United Kingdom
Address all correspondence and requests for reprints to: Dr. M. C. Eggo, The Medical School, University of Birmingham, Birmingham B15 2TT, United Kingdom. E-mail: m.c.eggo@bham.ac.uk.
Abstract
In goiter, increased expression of growth factors and their receptors occurs. We have inhibited the action of some of these growth factors, alone and in combination, to determine which are important in goitrogenesis. Recombinant adenovirus vectors (RAds) expressing truncated, secreted forms of human Tie2 (RAd-sTie2) and vascular endothelial growth factor receptor 1 (RAd-sVEGFR1) or a truncated, dominant-negative fibroblast growth factor receptor 1 (RAdDN-FGFR1) were used. Goiters in mice were induced by feeding an iodide-deficient diet, containing methimazole and sodium perchlorate. RAds were administered to mice simultaneously with the goitrogenic regimen, which was continued for 14 d. RAd treatment did not significantly affect increases in TSH or reductions in thyroid hormone or thyroid hyperactivity seen in goitrogen-treated controls mice, suggesting no effect on pituitary or thyroid responses to hypothyroidism. In control goiters, a 4-fold increase in vascular volume accompanied a 2-fold increase in thyroid mass. Complete inhibition of these increases was found when animals were treated with the three RAds in combination. In thyroids from three RAd-treated animals, there was marked, significant inhibition of Tie2, FGFR1, VEGFR1, FGF-2, and VEGF expression, compared with control goiters. When used individually, RAdDN-FGFR1 partially prevented goiter and RAd-sVEGFR1 partially reduced vascular volume. Their effects were not additive. RAd-sTie2 did not reduce goiter mass or vascular volume when used alone but was essential for complete goiter inhibition. VEGF and VEGFR1 expression was reduced in these thyroids. Limitation of physiologic organ growth is complex, requiring inhibition of multiple, interdependent growth factor axes.
Introduction
IN IN VIVO models, dietary deprivation of iodide combined with a goitrogen such as methylmercaptoimidazole provokes a rapid goitrogenesis, with typical increase in thyroid mass in rodents of 2- to 5-fold after 2 wk (1, 2, 3). There is an early vascular response with an increase in capillary density and size. Subsequently thyroid follicular cell growth increases, reaching a maximum at 5 d and slowly declining thereafter, although remaining elevated (4). Elevations of TSH accompany the fall in thyroid hormone levels and are thought to be the initiating stimulus for the cascade of events leading to goitrogenesis. TSH is known to regulate the synthesis of several angiogenic growth factors (5), and we have sought ways to limit goiter by manipulating the activity of three of these factors.
The mRNAs of the receptors for vascular endothelial growth factor (VEGF) are increased in goiter induced in rats. The increases in VEGF receptor (VEGFR)1 and VEGFR2 are subsequent to the rise in plasma TSH and in parallel with thyroid capillary proliferation (6). VEGFRs are reported to have a distribution limited to endothelial cells, and inhibition of their signaling should limit endothelial cell proliferation and thus thyroid angiogenesis. In human thyroid follicular cells in vitro, the ligands for VEGFR, the VEGFs are increased by factors signaling, like TSH, through cAMP and protein kinase C (7). The fibroblast growth factors (FGFs) are also angiogenic growth factors, but their receptors are also expressed on thyroid follicular cells, cells of the mesenchyme, and pericytes surrounding the vasculature. FGF receptor (FGFR)1 activation increases thyroid follicular cell growth in rat thyroid cells in vitro (8), and FGF-1 is a potent goitrogen in rats, increasing thyroid weight 43% after 6 d of treatment (9). FGF-1 and -2 and FGFR1 are expressed by human follicular cells, and expression of all is increased in multinodular goiter in humans (10) and goiter induced in rats (11). Using a recombinant adenovirus vector (RAd) expressing a dominant-negative (DN) form of FGFR1 (RAdDN-FGFR1), we recently showed inhibition of goiter growth in mice (12), but whether the principal effect is on thyroid follicular cells or endothelial cells or both is not clear.
Angiopoietin (Ang)-1 and -2 are angiogenic growth factors important in the destabilization of existing vessels or maturation of the neovasculature. Ang-2 antagonizes Ang-1-mediated phosphorylation of its receptor Tie2 (13, 14, 15), but apparently effects are context and concentration dependent (16, 17). Angiopoietins are secreted by human thyroid follicular cells, and mRNA levels for Ang-1 are increased in iodide-deficient goiter induced in rats (18). Although originally thought to be restricted to endothelial cells, Tie2 mRNA and protein are expressed in thyroid follicular cells (both human and rat) as well as thyroid endothelial cells. Tie2 expression on follicular cells is increased by TSH through elevations in cAMP (18) and is elevated in goiter. The function of follicular cell expression in goitrogenesis is unknown as yet.
To limit the actions of VEGFs and angiopoietins, soluble forms of their receptors, expressed in E1/E3-deleted, replication-defective adenovirus vectors, have been used. AdExTek, which expresses murine soluble Tie2, is postulated to work by binding angiopoietins, thus reducing their bioavailability (19). This strategy is well established in adenovirus gene therapy, in which the adenovirus-mediated transcription of recombinant protein chiefly occurs in the liver (20, 21). A gene therapy approach to limit organ growth has been used by several laboratories to block the growth of introduced tumors (22, 23, 24), but our study differs because induced goiter is a physiologic, programmed growth of normal cells initiated by endogenous hormones and growth factors.
Materials and Methods
Virus construction
RAd-sTie2.
An adenovirus expressing the extracellular domain of human Tie2 and green fluorescent protein (GFP) was constructed using the AdEasy system (25). Briefly, human Tie2 extracellular domain (710 amino acids) was cloned from human umbilical vein endothelial cell cDNA (kind gift of Drs. P. W. Hewett and J. C. Murray, CRC Academic Department of Clinical Oncology, City Hospital, Nottingham, UK) and transferred into an adenoviral shuttle vector that harbors a cytomegalovirus promoter-driven GFP and a cytomegalovirus promoter-flanked multiple cloning site for the insertion of the gene of interest. This shuttle vector was linearized with PmeI and cotransformed into Escherichia coli strain BJ5183 with the adenoviral backbone, pAdEasy-1, which contains the genome of E1/E3 deleted Ad5 adenovirus (26). After transfection in 911 cells, the recombinant adenovirus, named RAd-sTie2, was plaque purified and grown up through three further rounds of amplification.
RAdExTek.
This adenovirus expresses the extracellular domain of murine Tie2, which is 89.6% homologous to human extracellular domain of Tie2 (19). The recombinant protein is the full length of the extracellular domain, except for 8 amino acids from the transmembrane domain.
RAdDN-FGFR1.
This adenovirus contains the extracellular and transmembrane domains of mouse FGFR1 and blocks FGFR signaling in thyroid follicular cells (12). In other studies RAd-expressing DNFGFR1 has been found to inhibit the action of FGFR2–4 as well as FGFR1 (27).
RAd-sVEGFR1.
This virus was a kind gift from Dr. R Mulligan (Children’s Hospital, Boston, MA). RAd-sVEGFR1 expresses the soluble extracellular domain of VEGFR1, which is also known as Flt-1 (23).
Control adenovirus.
E1/E3-deleted adenovirus expressing GFP (RAd-GFP) was used as a control adenovirus.
Purification of viruses.
For in vivo use RAds were banded by cesium chloride density gradient centrifugation. The RAds were desalted by dialysis in a Slidalyzer (Pierce, Rockford, IL) against buffer [PBS, 10 mM Na2HPO4, 2.7 mM KCl, 137 mM NaCl, 10 mM CaCl2, 0.5 mM MgCl2 and 10% glycerol (pH 7.4)] at 4 C. RAds for in vitro and in vivo experiments were produced in a single large batch, aliquotted, and frozen at –70 C. All experiments were conducted with the same batch of recombinant adenovirus. The particle count was determined by assaying DNA content of virus, which had been solubilized in sodium dodecyl sulfate (SDS) (28) and denatured by heating to 56 C for 30 min, using the Picogreen DNA assay kit (Molecular Probes, Eugene, OR). Infectivity was determined by plaque assay.
In vitro viral experiments and Western blotting
Human thyroid follicular cells were prepared as described (29) from consented surgical specimens in accordance with local ethical guidelines. Cells were grown until 50% confluent and infected for 90 min with RAd-sTie2 at a multiplicity of infection of 10 particles per cell in a small volume of serum-free medium. Cells were washed with Hank’s balanced salt solution, and the medium was replaced for 72 h before analysis of proteins. For Western blotting, 100 μl of the cell-conditioned medium was adjusted to 2% SDS and 2% 2-mercaptoethanol and boiled for 5 min. Secreted proteins were analyzed by electrophoresis on 10% SDS-polyacrylamide gels with 5% stacking gel as described previously (18). Primary antibodies to epitopes at the C and N terminus of Tie2 (Santa Cruz Biotechnologies, Santa Cruz, CA) were used with enhanced chemiluminescence for detection.
Animal experiments
Adult (8 wk old) BALB/C male inbred mice were used for all experiments in accordance with ethical guidelines. Goiter was induced by feeding a diet of low iodine chow (less than 0.05 ppm iodide; Lillico Biotechnology, Surrey, UK) mixed with an equal weight of a solution of 0.15% 2-mercapto-1-methylimidazole (MMI) in water. Drinking water contained 1% sodium perchlorate (NaClO4). This regimen was maintained for 14 d. All experimental groups consisted of four to eight mice yielding eight to 16 thyroid lobes for examination.
Adenovirus-treated mice
Mice were injected with 1010 particles of each banded adenovirus in 100 μl PBS via tail vein under isoflurane anesthesia, and on recovery from anesthesia, given the low-iodine diet (LID) with MMI and NaClO4 to induce goiter formation. Mice were weighed daily to monitor toxicity and their general condition observed twice daily. Fourteen days later, the mice were anesthetized with isoflurane. Some mice were injected iv with 0.3 mg griffonia (Bandeiraea) simplicifolia Lectin I conjugated to rhodamine (Vector Laboratories, Burlingame, CA), which circulated intravitally for 2 min as described (30, 31). This lectin binds specifically to endothelial cells in microvessels, primarily capillaries but also terminal arterioles and venules. At the end of this time, mice were killed by cervical dislocation and blood taken by cardiac puncture.
RIA of serum levels of total T3 and T4 and TSH
Whole blood was collected from the mice and allowed to clot. The serum was used in single assays to measure total T3, total T4, and TSH by RIA (ICN Pharmaceuticals, High Wycombe, UK). The rat TSH kit was used to measure mouse serum TSH values.
Vascular analysis
Unfixed thyroids from lectin-treated mice were examined by confocal microscopy on a MRC 600 confocal microscope (Bio-Rad Laboratories, Hercules, CA) after intravital endothelial staining with lectin conjugated to rhodamine as described above. The thyroid glands were mounted in a drop of Hank’s balanced salt solution on a coverslip and optically sectioned. Blood vessel structure could be seen up to 100 μm into the tissue, but the morphology was best seen in the most superficial 50 μm. For analysis, Z-series of images were collected at x200 magnification and with 10-μm steps. This allowed each section to be optically independent from surrounding sections. To calculate the vascular volume (Vv) of the thyroid specimens, standard stereological techniques were used (32). The images of the thyroid vasculature were projected on a stereological grid, and then grid intersections overlying the vasculature were point counted. The observer was blinded to the identity of the samples. The points overlying blood vessels divided by total points overlying the tissue determined the Vv. The first section (of the surface) was not representative and was discounted. Sections at 10, 20, and 30 μm deep into the thyroid in all samples were examined. Z-series were taken from at least three thyroid glands in each group, and for each gland three series of three sections were taken and used in calculations. Within each mouse thyroid, the data were consistent with no more than 10% error within the determinations. The Vv was calculated by multiplying the mass of the thyroid by the Vv fraction, assuming a specific gravity of 1 for thyroid tissue.
Immunostaining
Immunostaining of 7 μm, paraffin wax-embedded, formaldehyde-fixed thyroid sections was performed as described previously using the ABC technique (16). Photographs of the entire section were taken and quantitation of immunostaining for all sections was assessed by five blinded, independent observers using a 0–4 scoring system. This was an arbitrary scale, with 4 showing every cell with intense staining, 3 showing strong staining with the majority of the cells stained, 2 showing moderate staining with some cells stained, and 1 showing light staining in some. Comparisons were made only between sections stained in the same run with the appropriate controls of RAd-GFP-treated goiter animals, control goiter animals, and background controls in which the first antibody was omitted. Sections (n = 3) from thyroids from at least two animals were analyzed in this way. Antisera used were FGF-2 (Santa Cruz I47), FGFR1 (Santa Cruz H76), VEGF (Oncogene PC315; Oncogene, Boston, MA), VEGFR1 (Santa Cruz C17), and Tie2 (Santa Cruz H176).
Statistics
Factoral ANOVA with Student-Newman-Keuls post hoc test was used to quantify intergroup differences. For analysis of immunostaining data, Student’s t test was used.
Results
Characterization of RAd-sTie2
To establish that cells transduced with RAd-sTie2 could secrete the extracellular soluble portion of human Tie2, we used primary cultures of human thyroid follicular cells. A Western blot of conditioned medium collected 72 h post transduction with RAd-sTie2 was probed with antisera specific for either the C or N terminus of Tie2. A 110-kDa fragment of sTie2 was detected only by antiserum to the N terminus of Tie2 (Fig. 1), and there was no sTie2 detectable in the conditioned medium of control cells or in cells transduced with RAd-GFP virus. Also shown on this figure is the secretion of immunoreactive sTek from human thyroid cells transduced with RAdExTek virus. A protein of the same size as sTie2 was secreted from cells treated with this virus.
FIG. 1. Western blot of 50 μl of conditioned medium of human follicular cells analyzed by SDS-PAGE. and probed with antibody to N-term of Tie2 (Santa Cruz). C, Control cells; GFP, cells transduced with RAd-GFP for 72 h; Tie, cells infected with RAd-sTie2 for 72 h; and Tek, cells infected with RAdExTek for 72 h. The same blot was stripped and reprobed with antibody to C-term Tie2 (Santa Cruz) and no staining was detectable.
The circulating levels of human Tie2 and sFlt in the serum of male BALB/C mice 14 d post injection were determined using an ELISA specific for human Tie2 and murine sFlt as described by the manufacturer (R&D Systems, Minneapolis, MN). Circulating levels were 30.1 ± 0.6 ng/ml for Tie2 and 9.4 ng/ml sFlt. There was no detectable circulating human Tie2 in control mice. Kuo et al. (23) have shown that peak levels of adenovirus transgene expression occur within the first 3 d and that there is a progressive decline thereafter. These values are therefore likely to be an underestimate of the circulating levels throughout the experiment.
Analysis of lectin-labeled blood vessels in normal thyroid by confocal microscopy
A normal mouse was injected with rhodamine-labeled lectin under anesthesia and terminated 2 min later. The thyroid was removed and examined using the confocal microscope. The confocal image is shown in Fig. 2A, and the lectin-labeled capillaries are evident. Figure 2B is a phase-contrast image of the same area, and Fig. 2C is a merged image of the light micrograph in green, and the blood vessels in red showing that the capillaries pass around the follicles. The dark areas in Fig. 2A correspond to the follicles in Fig. 2B.
FIG. 2. Confocal micrographs of normal mouse thyroid gland from a mouse intravitally injected with rhodamine-labeled lectin taken at x200 magnification. A, Lectin-labeled capillaries. B, Phase-contrast micrograph of the same area of thyroid as A. C, Merged image with the light micrograph in green and the blood vessels in red showing that the capillaries pass around the follicles, and that the dark central area as in A correspond to follicles in B. Scale, Width = 660 μm.
Effect of RAdDN-FGFR1, RAd-sVEGFR1, and RAd-sTie2 on thyroid activity, mass, and vascular volume in mice on a goitrogenic regime
Circulating TSH increased from 1.1 ± 0.6 in control mice to 7.4 ± 0.1 ng/ml (P < 0.001) in goitrogen-treated mice and was not significantly changed by treatment with any of the RAds (DN-FGFR1, sVEGFR1, sTie2, GFP) alone or in combination. Total T3 fell from 2.93 ± 0.09 to 0.53 ± 0.41 nmol/liter (P < 0.0001) and total T4 fell from 60.8 ± 17.9 to 16.7 ± 2.9 nmol/liter (P < 0.001) at the end of 14 d treatment with LID and goitrogens. Again RAd treatment had no significant effect on these changes. The general well-being of the mice was not adversely affected by viruses when used alone or three viruses combined (3 x viral load), and although the weight loss was greater in this group, it was not statistically different from the other groups.
Figure 3A shows thyroid mass and Fig. 3B the vascular volume changes occurring after treatment with RAds alone and in combination. The goitrogenic regime induced a more than 2-fold, significant increase in thyroid mass and a 4-fold increase in vascular volume. Goiter mass in mice treated with goitrogen alone, although smaller, did not differ significantly from those treated with goitrogens and injected with RAd-GFP, confirming that adenovirus infection alone does not affect goiter induction. Data from the RAd-GFP-treated mice are shown on the figure. There was no significant difference between the effects of RAd-sTie2 and RAdExTek on goiter mass, and the results were combined for simplicity.
FIG. 3. A, Thyroid mass (milligrams) from control mice (cont) and mice treated with goitrogens and RAd expressing GFP (G), sVEGFR1 (V), DN-FGFR1 (F), sTie2(T), DN-FGFR1+sVEGFR1 (F+V), DN-FGFR1+sTie2 (F+T), and sVEGFR1+sTie2+DN-FGFR1 (V+F+T). *, P < 0.05, ***, P < 0.001, compared with control; ###, P < 0.001, compared with goitrogen + RAdGFP-treated (G) mice. The numbers in the open boxes show the number of thyroid lobes weighed (mean ± SEM). B, Vascular volume (microliters) from control mice (cont) and mice treated with goitrogens and Rad-expressing GFP (G), sVEGFR1 (V), DN-FGFR1 (F), sTie2(T), DN-FGFR1+sVEGFR1 (F+V), DN-FGFR1+sTie2 (F+T), and sVEGFR1+sTie2+DN-FGFR1 (V+F+T). **, P < 0.01, ***, P < 0.001, compared with control; ##, P < 0.01, ###, P < 0.001, compared with goitrogen + RAdGFP-treated (G) mice (n = 3; mean ± SEM).
The difference between the goiter mass in the RAdDN-FGFR1-treated group and the thyroids from the controls was not significant, suggesting that goiter development had been inhibited by this treatment. The excess thyroid mass induced with goitrogen treatment had been reduced by almost 60%. In this analysis this did not quite reach statistical significance, but in our earlier report (12), using a smaller database and comparisons on mice treated and terminated on the same date, P < 0.05, compared with the goiter animals. For the other treatment groups, RAd-sVEGFR1 also reduced goiter mass (43%), but the goiter mass in these mice was significantly different from control (P < 0.05) and was not significantly reduced, compared with goiter control. The decrease in vascular volume with this RAd (Fig. 3B) was significantly different at 50% of the control goiter (P < 0.001). Although RAdDN-FGFR1 treatment reduced the vascular volume, this was not a significant decrease, compared with goitrogen treatment and was significantly different from control. RAd-sVEGFR1 and RAdDN-FGFR1 in combination inhibited the goiter, compared with the control mice, but no more than either virus used alone. Their effects were not additive on goiter mass or vascular volume.
When all three adenoviruses were used in combination, there was significant and complete inhibition of goiter, measured as thyroid mass. There was no significant difference in the mass of these thyroids, compared with the control mice, and there was a significant difference, compared with goiter mice (P < 0.001) (Fig. 3A). There was significant and complete inhibition of vascular volume increases, as shown in Fig. 3B. The thyroids from these animals were, however, red, suggesting that they were hypervascular and confocal micrographs of the vasculature are consistent with this. Figure 4A shows the vasculature from control mice. Figure 4B shows the vasculature from mice treated with goitrogen without RAd treatment. There is a marked increase in vessel diameter as well as a marked increase in intensity of vascular staining consistent with angiogenesis evident in Fig. 4B. Figure 4C shows the thyroid vasculature from mice treated with RAd-sVEGFR1 and goitrogens, and Fig. 4D shows that from mice treated with the combination of three RAds and goitrogens. Treatment with either RAd-sVEGFR1 or the combination of the three RAds produced a vasculature showing an increase of vessel diameter comparable with normal goitrogenesis (Fig. 4B) but a reduction in the number of new vessels. This latter effect was even more marked when animals were treated with three RAds than with RAd-sVEGFR1.
FIG. 4. Confocal micrographs showing thyroid vasculature control (A), treated with goitrogen (B), goitrogen with RAd-sVEGFR1 (C), and goitrogen with RAd-sTie2, RAd-sVEGFR1, and RAdDN-FGFR1 (D). All images are x200 magnification. Scale, Width = 660 μm.
Expression of FGF-2 and Tie2 after goitrogen treatment and the effect of RAd-sTie2
Figure 5A shows the expression of Tie2 and FGF-2 in normal mouse thyroid sections. Expression of both antigens in the thyroid follicular cells of normal mice is evident. The vasculature also stained positively for Tie2 and FGF-2. After 14 d treatment with the goitrogenic regimen, there was a marked increase in expression of both Tie2 and FGF-2 as shown in the second column. Also evident is the follicular cell hyperplasia and the loss of follicles, which occurs with goitrogenesis. In the third column, thyroid sections from a mouse treated with RAd-sTie2 at the start of the goitrogenic regimen are shown. There is a clear reduction in Tie2 expression, compared with the section directly above. Reductions in FGF-2 expression are also seen in the goiters from mice treated with RAd-sTie2, although this effect is not marked. From this we conclude that when Tie2 signaling is blocked, induction of Tie2 is blocked, consistent with positive feedback of Tie-2 signaling on its own receptor. We cannot, however, preclude the possibility that this is an indirect effect. Even more marked reductions in the expression of Tie2 and FGF-2 were seen in mice treated with goitrogen and RAdDN-FGFR1 shown in the last column. This was particularly marked for Tie2, in which expression was reduced to levels less than control mice. We conclude that FGF signaling, which is increased in goitrogenesis, may mediate the expression of Tie2 in mouse thyroids. We also conclude that FGF signaling, directly or indirectly, may increase the expression of FGF-2 in vivo.
FIG. 5. A, Immunostaining of mouse thyroid sections for Tie2 (upper panel) and FGF-2 (lower panel) from control mice (first column), mice treated with goitrogen (G) alone (second column), mice treated with goitrogen and RAd-sTie2 (third column), and mice treated with goitrogen and RAdDN-FGFR1 (fourth column). Magnification, x100. B, Immunostaining of thyroid sections from mice treated with goitrogen (G) + RAd-GFP (A–E) and from thyroids from mice treated with G+ RAds expressing sTie2, DN-FGFR1, and sVEGFR1 in combination (A’-E’). Antibodies: A/A’, FGF-2; B/B’, FGFR1; C/C’, VEGF; D/D’, VEGFR1; E/E’, Tie2. Bar, 100 μm.
Effects of the combination of RAdDN-FGFR1, RAd-sTie2, and RAd-sVEGFR1 on expression of FGF-2, FGFR1, VEGF, VEGFR1, and Tie2 in thyroid tissue sections
The effects of the combination of three RAds on growth factor and receptor expression are shown in Fig. 5B in which representative sections from the thyroids of mice treated with goitrogen and RAd-GFP (upper panel) and those treated with goitrogen and the combination of the three RAds (lower panel) were stained for FGF-2 (A), FGFR1 (B), VEGF (C), VEGFR1 (D), and Tie2 (E). In animals treated with three viruses, there were marked reductions in immunostaining intensity for all the growth factors and their receptors.
To examine the interrelationships between the growth factors further, a more quantitative comparison of the effects of individual RAd treatment on the expression of the growth factors and their receptors was undertaken. Significance was determined from the mean scoring (0–4) of the five independent, blinded observers and compared with the goitrogen-treated, RAd-GFP treated animals.
The rank order for inhibition for animals treated with RAd-sTie2 alone was VEGF greater than VEGFR1 greater than FGFR1 equal to Tie2 greater than FGF-2. These effects were significant for VEGF and VEGFR1. From these data we conclude that Tie2 signaling is required for the induction and maintenance of VEGF expression in thyroid cells.
In thyroids from animals treated with RAdDN-FGFR1 or RAd-sVEGFR1, inhibition of growth factor and receptor expression was less and more variable. For both RAds there was prevention of the increase in Tie2 and FGFR1 expression with no effect on VEGFR1 expression, compared with control goiter. In goiters in RAd-sVEGFR1-treated animals, FGF-2 expression was unaffected. We conclude from this that VEGF expression is independent of VEGFR1 and FGFR1 signaling and that FGFR and Tie2 induction are regulated by VEGFR1 and FGFR1 signaling.
Analyses of the sections of the mice treated with all three viruses, examples of which are shown in Fig. 5B, showed significant inhibition of all the growth factors and their receptors. The order of inhibition was Tie2 greater than FGFR1 equal to VEGFR1 equal to FGF greater than VEGF.
Discussion
Treatment of mice with perchlorate, MMI, and a LID makes them hypothyroid and leads to goitrogenesis. The 4-fold increase in vascular volume in goiter (i.e. a 2-fold increase in vascular volume density) agrees with previous studies using paraffin-embedded sections or electron microscope images to analyze the vascular changes in goiter in rats (1, 3, 33, 34). The increase in vascular volume has been attributed to both increases in the size of existing blood vessels and capillary sprouting.
Our data show that this goiter can be completely inhibited by combined delivery of RAds inhibiting the actions of three angiogenic growth factor signaling pathways. The actions of the angiopoietins, the VEGFs, and FGF signaling have to be inhibited together to effect full inhibition of goiter induced in normal thyroid. We have previously shown that inhibition of FGF signaling with RAdDN-FGFR1 produces a partial reduction in the mass of the goiter produced in response to hypothyroidism (12). Although the vasculature was decreased with this treatment, the greater, significant effect was on goiter mass. From the immunostaining analyses, inhibition of FGF signaling was accompanied by an inhibition of FGF-2 expression and its receptor FGFR1 as well as a reduction in Tie2. There was no inhibition of VEGF, which increases endothelial cell proliferation through VEGFR2. This may explain why inhibition of FGF signaling did not produce effective inhibition of the goiter.
When the thyroid gland is stimulated with TSH, the morphology of the follicles and the follicular cells changes dramatically, the follicular lumens becoming insignificant and the cells hyperactive and columnar. Mass changes may not therefore directly relate to an increase in cell number. Whether follicular cell proliferation is prevented by RAdDN-FGFR1 and whether the mitogenic effects of TSH persist are not known. RAdDN-FGFR1 could potentially inhibit the growth of any FGFR-expressing cells in the thyroid including fibroblasts, endothelial cells, pericytes, and follicular cells, which may contribute to the reduction in goiter mass.
We found that RAd-sVEGFR1 can significantly but only partially limit the development of new vasculature to a goiter, reducing the number of new vessels formed during goitrogenesis but not affecting the increase in diameter of existing vessels. We found that RAds-VEGFR1 reduced the expression of FGFR1 and Tie2 in the goiters, which would also be antiangiogenic. A recent study in mice showed that 293 cells expressing sVEGFR1 could block the growth of a tumor derived from a follicular thyroid carcinoma cell line by 70% (34). The inhibition of physiologic goiter in our study was less dramatic, but the studies differ in that in our model the stimulus for angiogenesis is derived from normal thyroid cells and not those known to have genetic derangements.
RAd-sTie2 did not block the increase in vascularity that occurs during goitrogenesis in mice. Goiter mass in fact was higher in these animals, although this was not statistically significant. Tie2 is thought to be involved more in the restructuring and maturation of the vasculature than its proliferation (13). We found that mRNA for Ang-1 but not Ang-2 was elevated in an iodide-deficiency/goitrogen model of goiter in rats. The sequestration of Ang-1 should prevent the stabilization of the vasculature through phosphatidylinositol 3-kinase but, based on data from tumor models, whether this will have proangiogenic or antiangiogenic consequences is unpredictable (13, 35). The increase in VEGF and VEGFR1, induced in goiter, was significantly reduced with RAd-sTie2, which should limit angiogenesis. The FGF axis was less profoundly affected and whether this can compensate for a reduction in VEGF signaling is unknown. Inhibition of angiopoietin:Tie2 signaling on the endothelial cells and/or the thyroid follicular cells may also affect cell-cell adhesion. There are two studies (36, 37) suggesting Tie2-independent effects of the angiopoietins due to their interactions with integrins. Whether by blocking angiopoietin effects, cell adhesion is blocked in vivo is unknown, although knockout mice models lend support to this role (13). The role of Tie2 on the follicular cells is unknown and has not been explored in the knockout mice. To examine whether its actions could be important in vivo, we determined the effects of RAd-sTie2 on thyroid growth and function in normal 8-wk-old mice. Experiments were terminated 14 d after injection of RAd-sTie2. Although there was an increase in thyroid mass from 1.4 ± 0.4 to 1.9 ± 0.5 mg, this was not significant. There was no effect on thyroid function measured by T4 and TSH levels. Further studies at different time points are required to determine whether Tie2 can provide survival signals in thyroid follicular cells through Akt signaling or affect thyroid function.
Our finding that the combined inhibition of the three angiogenic pathways is required to limit goiter growth completely is likely related to the complete inhibition of the induction of the receptors and their ligands with this treatment. Because elevations in TSH are not blocked by combined RAd treatment, we conclude that the induction of the growth factor receptors and their ligands is not under direct TSH control but is regulated by other growth factors including those under examination in this study. It is interesting that the increase in Tie2 expression in goiter was limited by treatment with any of the three RAds and implies a more complex regulation of follicular Tie2 expression than our earlier in vitro work suggested. We found that TSH, through cAMP, was able to regulate follicular cell Tie2 expression in vitro (18); however, we also found that TSH regulates FGF-2 production in FRTL5 cells (12). Conceivably TSH effects on Tie2 may therefore be mediated through FGF-2 or another intermediary. We also conclude that FGF is, in part, responsible for the increase in FGFR1 synthesis because RAdDN-FGFR1 was able to inhibit FGFR1 expression. The ligands for VEGFRs are increased by both cAMP and growth factors (6, 7). Our data suggest that Tie2 also mediates the production of the VEGFs in the thyroid.
Our data demonstrate that growth of the thyroid vasculature is regulated by the production of local growth factors. We have examined parts of three of the potential angiogenic/goitrogenic pathways. This does not preclude the importance of other pathways or the possibility of redundancies. It must be remembered that we examined the expression of only one of the five VEGFRs, only one of the four VEGFs, only one of the four FGFRs, and only one of 23 of the FGFs. Furthermore, we have not examined the potential contribution of other growth factors of likely importance such as hepatocyte growth factor, IGFs, and TGFs (5, 38). We do show, however, that when goiter was effectively inhibited, all the antigens examined were reduced, suggesting cooperation between the pathways we used. We also found evidence that Tie2 signaling, probably in thyroid follicular cells, regulates the increases in follicular cells of VEGF as well as that of its own receptor and that of VEGFR1 and FGFR1. This strongly suggests a functional role on thyroid cells.
We show that physiologic thyroid growth can be modified by the use of RAds. There have been no studies to our knowledge of the effects of gene therapy on the growth of endogenous tissue driven by endogenous hormones and locally produced endogenous growth actors. Other gene therapy studies have shown the effects on the growth of introduced tumors in mice. Although gene therapy is unlikely to replace surgery as the primary modality of treatment of large multinodular goiters and differentiated thyroid cancers, its use in anaplastic carcinoma of the thyroid, which is invasive and rapidly fatal, may be more relevant.
Acknowledgments
We thank Dr. Jackson Kirkman-Brown for generous help with the confocal microscope.
References
Wollman S, Herveg J, Zeligs J, Ericson L 1978 Blood capillary enlargement during the development of thyroid hyperplasia in the rat. Endocrinology 103:2306–2312
Denef J, Haumont S, Cornette C, Beckers C 1981 Correlated functional and morphometric study of thyroid hyperplasia induced by iodine deficiency. Endocrinology 108:2352–2358
Many M, Denef J, Haumont S 1984 Precocity of the endothelial proliferation during a course of rapid goitrogenesis. Acta Endocrinol 105:487–491
Redmond O, Tuffery AR 1981 Thyroid proliferation, body weight, thyrotropin and thyroid hormones in chronic antithyroid (carbimazole) treatment in rats. J Anat 133:37–47
Eggo MC 2003 Angiogenesis in goitrogenesis. Curr Opin Endocrinol Diabetes 10:341–346
Viglietto G, Romano A, Manzo G, Chiappetta G, Paoletti I, Califano D, Galati MG, Mauriello V, Bruni P, Lao CT, Fusco A, Persico MG 1997 Upregulation of the angiogenic factors PlGf, VEGF and their receptors (Flt-1, Flk-1/KDR) by TSH in cultured thyrocytes and in the thyroid gland of thiouracil-fed rats suggest a TSH-dependent paracrine mechanism for goiter hypervascularization. Oncogene 15:2687–2698
Sato K, Yamazaki K, Shizume K, Kanaji Y, Obara T, Ohsumi K, Demura H, Yamaguchi S, Shibuya M 1995 Stimulation by thyroid-stimulating hormone and Graves’ immunoglobulin G of vascular endothelial growth factor mRNA expression in human thyroid follicles in vitro and Flt mRNA expression in the rat thyroid in vivo. J Clin Invest 96:1295–1302
Isozaki O, Emoto N, Tsushima T, Sato Y, Shizume K, Demura H, Akamizu T, Kohn LD 1992 Opposite regulation of deoxyribonucleic acid synthesis and iodide uptake in rat thyroid cells by basic fibroblast growth factor: correlation with opposite regulation of c-fos and thyrotropin receptor gene expression. Endocrinology 131:2723–2732
DeVito W, Chanoine J, Alex S, Fang S, Stone S, Huber A, Shalhoub V, Linn J, Stein G, Braverman L 1992 Effect of in vivo administration of recombinant acidic FGF on thyroid function in the rat: induction of colloid goiter. Endocrinology 131:729–735
Thompson SD, Franklyn JA, Watkinson JC, Verhaeg JM, Sheppard MC, Eggo MC 1998 Fibroblast growth factors 1 and 2 and fibroblast growth factor receptor 1 are elevated in thyroid hyperplasia. J Clin Endocrinol Metab 83:1336–1341
Patel VA, Hill DJ, Eggo MC, Sheppard MC, Becks GP, Logan A 1996 Changes in the immunohistochemical localisation of fibroblast growth factor-2, transforming growth factor-?1 and thrombospondin-1 are associated with early angiogenic events in the hyperplastic rat thyroid. J Endocrinol 148:485–499
Davies EL, Ramsden JD, Cocks H, Searle PF, Watkinson JC, Ueno H, Logan A, Franklyn JA, Mautner V, Eggo MC 2003 Adenoviral delivery of dominant negative fibroblast growth factor receptor 1 blocks FGF effects in thyroid cells and reduces goitrogenesis in mice. J Clin Endocrinol Metab 88:4472–4480
Thurston G 2003 Role of angiopoietins and Tie receptor tyrosine kinases in angiogenesis and lymphangiogenesis. Cell Tissue Res 314:61–68
Maisonpierre PC, Suri C, Jones PF, Bartunkova S, Wiegand SJ, Radziejewski C, Compton D, McClain J, Aldrich TH, Papadopoulos N, Daly TJ, Davis S, Sato TN, Yancopoulos GD 1997 Angiopoietin-2, a natural antagonist for Tie2 that disrupts in vivo angiogenesis. Science 277:55–60
Davis S, Papadopoulos N, Aldrich TH, Maisonpierre PC, Huang T, Kovac L, Xu A, Leidich R, Radziejewski C, Rafique A, Goldberg J, Jain V, Bailey K, Karow M, Fandl J, Samuelsson SJ, Ioffe E, Rudge JS, Daly TJ, Rad-ziejewski C, Yancopoulos GD 2003 Angiopoietins have distinct modular domains essential for receptor binding, dimerization and superclustering. Nat Struct Biol 10:38–44
Teichert-Kuliszewska K, Maisonpierre PC, Jones N, Campbell AI, Master Z, Bendeck MP, Alitalo K, Dumont DJ, Yancopoulos GD, Stewart DJ 2001 Biological action of angiopoietin-2 in a fibrin matrix model of angiogenesis is associated with activation of Tie2. Cardiovasc Res 49:659–670
Kim I, Kim HG, So JN, Kim JH, Kwak HJ, Koh GY 2000 Angiopoietin-1 regulates endothelial cell survival through the phosphatidylinositol 3'-kinase/Akt signal transduction pathway. Oncogene 19:4549–4552
Ramsden JD, Cocks HC, Shams M, Nijjar S, Watkinson JC, Sheppard MC, Ahmed AS, Eggo MC 2001 Tie-2 is expressed on thyroid follicular cells, is increased in goiter, and is regulated by thyrotropin through cyclic adenosine 3',5'-monophosphate. J Clin Endocrinol Metab 86:2709–2716
Lin P, Buxton JA, Acheson A, Radziejewski C, Maisonpierre PC, Yancopoulos GD, Channon KM, Hale LP, Dewhirst MW, George SE, Peters KG 1998 Antiangiogenic gene therapy targeting the endothelium-specific receptor tyrosine kinase Tie2. Proc Natl Acad Sci USA 95:8829–8834
Schneider M, French B 1993 The advent of adenovirus gene therapy for cardiovascular disease. Circulation 88:1937–1942[Free Full Text]
Carmeliet P, Collen D 1997 Molecular analysis of blood vessel formation and disease. Am J Physiol 273:H2091–H2104
Lin P, Polverini P, Dewhirst M, Shan S, Rao PS, Peters K 1997 Inhibition of tumor angiogenesis using a soluble receptor establishes a role for Tie2 in pathologic vascular growth. J Clin Invest 100:2072–2078
Kuo CJ, Farnebo F, Yu EY, Christofferson R, Swearingen RA, Carter R, von Recum HA, Yuan J, Kamihara J, Flynn E, D’Amato R, Folkman J, Mulligan RC 2001 Comparative evaluation of the antitumor activity of antiangiogenic proteins delivered by gene transfer. Proc Natl Acad Sci USA 98:4605–4610
Hangai M, Moon YS, Kitaya N, Chan CK, Wu DY, Peters KG, Ryan SJ, Hinton DR 2001 Systemically expressed soluble Tie2 Inhibits intraocular neovascularization. Hum Gene Ther 12:1311–1321
He TC, Zhou S, da Costa LT, Yu Y, Kinzler KW, Vogelstein B 1998 A simplified system for generating recombinant adenoviruses. Proc Natl Acad Sci USA 95:2509–2514
Ghosh-Choudhury G, Haj-Ahmad Y, Brinkley P, Rudy J, Graham FL 1986 Human adenovirus cloning vectors based on infectious bacterial plasmids. Gene 50:161–171
Ueno H, Gunn M, Dell K, Tseng Jr A, Williams L 1992 A truncated form of fibroblast growth factor receptor 1 inhibits signal transduction by multiple types of fibroblast growth factor receptor. J Biol Chem 267:1470–1476
Murakami P, McCaman MT 1999 Quantitation of adenovirus DNA and virus particles with the PicoGreen fluorescent dye. Anal Biochem 274:283–288
Eggo MC, King WJ, Black EG, Sheppard MC 1996 Functional human thyroid cells and their insulin-like growth factor binding proteins. J Clin Endocrinol Metab 81:3056–3062
Debbage PL, Griebel J, Ried M, Gneiting T, DeVries A, Hutzler P 1998 Lectin intravital perfusion studies in tumor-bearing mice: micrometer-resolution, wide-area mapping of microvascular labeling, distinguishing efficiently and inefficiently perfused microregions in the tumor. J Histochem Cytochem 46:627–639
Thurston G, McLean JW, Rizen M, Baluk P, Haskell A, Murphy TJ, Hanahan D, McDonald DM 1998 Cationic liposomes target angiogenic endothelial cells in tumors and chronic inflammation in mice. J Clin Invest 101:1401–1413
Howard C, Reed M 1998 Unbiased stereology three-dimensional measurement in microscopy. Oxford, UK: BIOS Scientific Publishers Ltd.
Smeds S, Wollman S 1983 3H-thymidine labeling of endothelial cells in thyroid arteries, veins and lymphatics during thyroid stimulation. Lab Invest 48:285–291
Ye C, Feng C, Wang S, Wang KZQ, Huang N, Liu X, Lin Y, Li M 2004 sFlt-1 gene therapy of follicular thyroid carcinoma. Endocrinology 145:817–822
Stoeltzing O, Ahmad SA, Liu W, McCarty MF, Wey JS, Parikh AA, Fan F, Reinmuth N, Kawaguchi M, Bucana CD, Ellis LM 2003 Angiopoietin-1 inhibits vascular permeability, angiogenesis and growth of hepatic colon cancer tumors. Cancer Res 63:3370–3377
Carlson TR, Feng Y, Maisonpierre PC, Mrksich M, Morla AO 2001 Direct cell adhesion to the angiopoietins mediated by integrins. J Biol Chem 276:26516–26525
Xu Y, Yu Q 2001 Angiopoietin-1, unlike angiopoietin-2, is incorporated into the extracellular matrix via its linker peptide region. J Biol Chem 276:34990–34998
Eggo MC, Quiney V, Campbell S 2003 Local factors regulating growth and of human thyroid cells in vitro and in vivo. Mol Cell Endocrinol 213:47–58(James D. Ramsden, Malcolm)
Address all correspondence and requests for reprints to: Dr. M. C. Eggo, The Medical School, University of Birmingham, Birmingham B15 2TT, United Kingdom. E-mail: m.c.eggo@bham.ac.uk.
Abstract
In goiter, increased expression of growth factors and their receptors occurs. We have inhibited the action of some of these growth factors, alone and in combination, to determine which are important in goitrogenesis. Recombinant adenovirus vectors (RAds) expressing truncated, secreted forms of human Tie2 (RAd-sTie2) and vascular endothelial growth factor receptor 1 (RAd-sVEGFR1) or a truncated, dominant-negative fibroblast growth factor receptor 1 (RAdDN-FGFR1) were used. Goiters in mice were induced by feeding an iodide-deficient diet, containing methimazole and sodium perchlorate. RAds were administered to mice simultaneously with the goitrogenic regimen, which was continued for 14 d. RAd treatment did not significantly affect increases in TSH or reductions in thyroid hormone or thyroid hyperactivity seen in goitrogen-treated controls mice, suggesting no effect on pituitary or thyroid responses to hypothyroidism. In control goiters, a 4-fold increase in vascular volume accompanied a 2-fold increase in thyroid mass. Complete inhibition of these increases was found when animals were treated with the three RAds in combination. In thyroids from three RAd-treated animals, there was marked, significant inhibition of Tie2, FGFR1, VEGFR1, FGF-2, and VEGF expression, compared with control goiters. When used individually, RAdDN-FGFR1 partially prevented goiter and RAd-sVEGFR1 partially reduced vascular volume. Their effects were not additive. RAd-sTie2 did not reduce goiter mass or vascular volume when used alone but was essential for complete goiter inhibition. VEGF and VEGFR1 expression was reduced in these thyroids. Limitation of physiologic organ growth is complex, requiring inhibition of multiple, interdependent growth factor axes.
Introduction
IN IN VIVO models, dietary deprivation of iodide combined with a goitrogen such as methylmercaptoimidazole provokes a rapid goitrogenesis, with typical increase in thyroid mass in rodents of 2- to 5-fold after 2 wk (1, 2, 3). There is an early vascular response with an increase in capillary density and size. Subsequently thyroid follicular cell growth increases, reaching a maximum at 5 d and slowly declining thereafter, although remaining elevated (4). Elevations of TSH accompany the fall in thyroid hormone levels and are thought to be the initiating stimulus for the cascade of events leading to goitrogenesis. TSH is known to regulate the synthesis of several angiogenic growth factors (5), and we have sought ways to limit goiter by manipulating the activity of three of these factors.
The mRNAs of the receptors for vascular endothelial growth factor (VEGF) are increased in goiter induced in rats. The increases in VEGF receptor (VEGFR)1 and VEGFR2 are subsequent to the rise in plasma TSH and in parallel with thyroid capillary proliferation (6). VEGFRs are reported to have a distribution limited to endothelial cells, and inhibition of their signaling should limit endothelial cell proliferation and thus thyroid angiogenesis. In human thyroid follicular cells in vitro, the ligands for VEGFR, the VEGFs are increased by factors signaling, like TSH, through cAMP and protein kinase C (7). The fibroblast growth factors (FGFs) are also angiogenic growth factors, but their receptors are also expressed on thyroid follicular cells, cells of the mesenchyme, and pericytes surrounding the vasculature. FGF receptor (FGFR)1 activation increases thyroid follicular cell growth in rat thyroid cells in vitro (8), and FGF-1 is a potent goitrogen in rats, increasing thyroid weight 43% after 6 d of treatment (9). FGF-1 and -2 and FGFR1 are expressed by human follicular cells, and expression of all is increased in multinodular goiter in humans (10) and goiter induced in rats (11). Using a recombinant adenovirus vector (RAd) expressing a dominant-negative (DN) form of FGFR1 (RAdDN-FGFR1), we recently showed inhibition of goiter growth in mice (12), but whether the principal effect is on thyroid follicular cells or endothelial cells or both is not clear.
Angiopoietin (Ang)-1 and -2 are angiogenic growth factors important in the destabilization of existing vessels or maturation of the neovasculature. Ang-2 antagonizes Ang-1-mediated phosphorylation of its receptor Tie2 (13, 14, 15), but apparently effects are context and concentration dependent (16, 17). Angiopoietins are secreted by human thyroid follicular cells, and mRNA levels for Ang-1 are increased in iodide-deficient goiter induced in rats (18). Although originally thought to be restricted to endothelial cells, Tie2 mRNA and protein are expressed in thyroid follicular cells (both human and rat) as well as thyroid endothelial cells. Tie2 expression on follicular cells is increased by TSH through elevations in cAMP (18) and is elevated in goiter. The function of follicular cell expression in goitrogenesis is unknown as yet.
To limit the actions of VEGFs and angiopoietins, soluble forms of their receptors, expressed in E1/E3-deleted, replication-defective adenovirus vectors, have been used. AdExTek, which expresses murine soluble Tie2, is postulated to work by binding angiopoietins, thus reducing their bioavailability (19). This strategy is well established in adenovirus gene therapy, in which the adenovirus-mediated transcription of recombinant protein chiefly occurs in the liver (20, 21). A gene therapy approach to limit organ growth has been used by several laboratories to block the growth of introduced tumors (22, 23, 24), but our study differs because induced goiter is a physiologic, programmed growth of normal cells initiated by endogenous hormones and growth factors.
Materials and Methods
Virus construction
RAd-sTie2.
An adenovirus expressing the extracellular domain of human Tie2 and green fluorescent protein (GFP) was constructed using the AdEasy system (25). Briefly, human Tie2 extracellular domain (710 amino acids) was cloned from human umbilical vein endothelial cell cDNA (kind gift of Drs. P. W. Hewett and J. C. Murray, CRC Academic Department of Clinical Oncology, City Hospital, Nottingham, UK) and transferred into an adenoviral shuttle vector that harbors a cytomegalovirus promoter-driven GFP and a cytomegalovirus promoter-flanked multiple cloning site for the insertion of the gene of interest. This shuttle vector was linearized with PmeI and cotransformed into Escherichia coli strain BJ5183 with the adenoviral backbone, pAdEasy-1, which contains the genome of E1/E3 deleted Ad5 adenovirus (26). After transfection in 911 cells, the recombinant adenovirus, named RAd-sTie2, was plaque purified and grown up through three further rounds of amplification.
RAdExTek.
This adenovirus expresses the extracellular domain of murine Tie2, which is 89.6% homologous to human extracellular domain of Tie2 (19). The recombinant protein is the full length of the extracellular domain, except for 8 amino acids from the transmembrane domain.
RAdDN-FGFR1.
This adenovirus contains the extracellular and transmembrane domains of mouse FGFR1 and blocks FGFR signaling in thyroid follicular cells (12). In other studies RAd-expressing DNFGFR1 has been found to inhibit the action of FGFR2–4 as well as FGFR1 (27).
RAd-sVEGFR1.
This virus was a kind gift from Dr. R Mulligan (Children’s Hospital, Boston, MA). RAd-sVEGFR1 expresses the soluble extracellular domain of VEGFR1, which is also known as Flt-1 (23).
Control adenovirus.
E1/E3-deleted adenovirus expressing GFP (RAd-GFP) was used as a control adenovirus.
Purification of viruses.
For in vivo use RAds were banded by cesium chloride density gradient centrifugation. The RAds were desalted by dialysis in a Slidalyzer (Pierce, Rockford, IL) against buffer [PBS, 10 mM Na2HPO4, 2.7 mM KCl, 137 mM NaCl, 10 mM CaCl2, 0.5 mM MgCl2 and 10% glycerol (pH 7.4)] at 4 C. RAds for in vitro and in vivo experiments were produced in a single large batch, aliquotted, and frozen at –70 C. All experiments were conducted with the same batch of recombinant adenovirus. The particle count was determined by assaying DNA content of virus, which had been solubilized in sodium dodecyl sulfate (SDS) (28) and denatured by heating to 56 C for 30 min, using the Picogreen DNA assay kit (Molecular Probes, Eugene, OR). Infectivity was determined by plaque assay.
In vitro viral experiments and Western blotting
Human thyroid follicular cells were prepared as described (29) from consented surgical specimens in accordance with local ethical guidelines. Cells were grown until 50% confluent and infected for 90 min with RAd-sTie2 at a multiplicity of infection of 10 particles per cell in a small volume of serum-free medium. Cells were washed with Hank’s balanced salt solution, and the medium was replaced for 72 h before analysis of proteins. For Western blotting, 100 μl of the cell-conditioned medium was adjusted to 2% SDS and 2% 2-mercaptoethanol and boiled for 5 min. Secreted proteins were analyzed by electrophoresis on 10% SDS-polyacrylamide gels with 5% stacking gel as described previously (18). Primary antibodies to epitopes at the C and N terminus of Tie2 (Santa Cruz Biotechnologies, Santa Cruz, CA) were used with enhanced chemiluminescence for detection.
Animal experiments
Adult (8 wk old) BALB/C male inbred mice were used for all experiments in accordance with ethical guidelines. Goiter was induced by feeding a diet of low iodine chow (less than 0.05 ppm iodide; Lillico Biotechnology, Surrey, UK) mixed with an equal weight of a solution of 0.15% 2-mercapto-1-methylimidazole (MMI) in water. Drinking water contained 1% sodium perchlorate (NaClO4). This regimen was maintained for 14 d. All experimental groups consisted of four to eight mice yielding eight to 16 thyroid lobes for examination.
Adenovirus-treated mice
Mice were injected with 1010 particles of each banded adenovirus in 100 μl PBS via tail vein under isoflurane anesthesia, and on recovery from anesthesia, given the low-iodine diet (LID) with MMI and NaClO4 to induce goiter formation. Mice were weighed daily to monitor toxicity and their general condition observed twice daily. Fourteen days later, the mice were anesthetized with isoflurane. Some mice were injected iv with 0.3 mg griffonia (Bandeiraea) simplicifolia Lectin I conjugated to rhodamine (Vector Laboratories, Burlingame, CA), which circulated intravitally for 2 min as described (30, 31). This lectin binds specifically to endothelial cells in microvessels, primarily capillaries but also terminal arterioles and venules. At the end of this time, mice were killed by cervical dislocation and blood taken by cardiac puncture.
RIA of serum levels of total T3 and T4 and TSH
Whole blood was collected from the mice and allowed to clot. The serum was used in single assays to measure total T3, total T4, and TSH by RIA (ICN Pharmaceuticals, High Wycombe, UK). The rat TSH kit was used to measure mouse serum TSH values.
Vascular analysis
Unfixed thyroids from lectin-treated mice were examined by confocal microscopy on a MRC 600 confocal microscope (Bio-Rad Laboratories, Hercules, CA) after intravital endothelial staining with lectin conjugated to rhodamine as described above. The thyroid glands were mounted in a drop of Hank’s balanced salt solution on a coverslip and optically sectioned. Blood vessel structure could be seen up to 100 μm into the tissue, but the morphology was best seen in the most superficial 50 μm. For analysis, Z-series of images were collected at x200 magnification and with 10-μm steps. This allowed each section to be optically independent from surrounding sections. To calculate the vascular volume (Vv) of the thyroid specimens, standard stereological techniques were used (32). The images of the thyroid vasculature were projected on a stereological grid, and then grid intersections overlying the vasculature were point counted. The observer was blinded to the identity of the samples. The points overlying blood vessels divided by total points overlying the tissue determined the Vv. The first section (of the surface) was not representative and was discounted. Sections at 10, 20, and 30 μm deep into the thyroid in all samples were examined. Z-series were taken from at least three thyroid glands in each group, and for each gland three series of three sections were taken and used in calculations. Within each mouse thyroid, the data were consistent with no more than 10% error within the determinations. The Vv was calculated by multiplying the mass of the thyroid by the Vv fraction, assuming a specific gravity of 1 for thyroid tissue.
Immunostaining
Immunostaining of 7 μm, paraffin wax-embedded, formaldehyde-fixed thyroid sections was performed as described previously using the ABC technique (16). Photographs of the entire section were taken and quantitation of immunostaining for all sections was assessed by five blinded, independent observers using a 0–4 scoring system. This was an arbitrary scale, with 4 showing every cell with intense staining, 3 showing strong staining with the majority of the cells stained, 2 showing moderate staining with some cells stained, and 1 showing light staining in some. Comparisons were made only between sections stained in the same run with the appropriate controls of RAd-GFP-treated goiter animals, control goiter animals, and background controls in which the first antibody was omitted. Sections (n = 3) from thyroids from at least two animals were analyzed in this way. Antisera used were FGF-2 (Santa Cruz I47), FGFR1 (Santa Cruz H76), VEGF (Oncogene PC315; Oncogene, Boston, MA), VEGFR1 (Santa Cruz C17), and Tie2 (Santa Cruz H176).
Statistics
Factoral ANOVA with Student-Newman-Keuls post hoc test was used to quantify intergroup differences. For analysis of immunostaining data, Student’s t test was used.
Results
Characterization of RAd-sTie2
To establish that cells transduced with RAd-sTie2 could secrete the extracellular soluble portion of human Tie2, we used primary cultures of human thyroid follicular cells. A Western blot of conditioned medium collected 72 h post transduction with RAd-sTie2 was probed with antisera specific for either the C or N terminus of Tie2. A 110-kDa fragment of sTie2 was detected only by antiserum to the N terminus of Tie2 (Fig. 1), and there was no sTie2 detectable in the conditioned medium of control cells or in cells transduced with RAd-GFP virus. Also shown on this figure is the secretion of immunoreactive sTek from human thyroid cells transduced with RAdExTek virus. A protein of the same size as sTie2 was secreted from cells treated with this virus.
FIG. 1. Western blot of 50 μl of conditioned medium of human follicular cells analyzed by SDS-PAGE. and probed with antibody to N-term of Tie2 (Santa Cruz). C, Control cells; GFP, cells transduced with RAd-GFP for 72 h; Tie, cells infected with RAd-sTie2 for 72 h; and Tek, cells infected with RAdExTek for 72 h. The same blot was stripped and reprobed with antibody to C-term Tie2 (Santa Cruz) and no staining was detectable.
The circulating levels of human Tie2 and sFlt in the serum of male BALB/C mice 14 d post injection were determined using an ELISA specific for human Tie2 and murine sFlt as described by the manufacturer (R&D Systems, Minneapolis, MN). Circulating levels were 30.1 ± 0.6 ng/ml for Tie2 and 9.4 ng/ml sFlt. There was no detectable circulating human Tie2 in control mice. Kuo et al. (23) have shown that peak levels of adenovirus transgene expression occur within the first 3 d and that there is a progressive decline thereafter. These values are therefore likely to be an underestimate of the circulating levels throughout the experiment.
Analysis of lectin-labeled blood vessels in normal thyroid by confocal microscopy
A normal mouse was injected with rhodamine-labeled lectin under anesthesia and terminated 2 min later. The thyroid was removed and examined using the confocal microscope. The confocal image is shown in Fig. 2A, and the lectin-labeled capillaries are evident. Figure 2B is a phase-contrast image of the same area, and Fig. 2C is a merged image of the light micrograph in green, and the blood vessels in red showing that the capillaries pass around the follicles. The dark areas in Fig. 2A correspond to the follicles in Fig. 2B.
FIG. 2. Confocal micrographs of normal mouse thyroid gland from a mouse intravitally injected with rhodamine-labeled lectin taken at x200 magnification. A, Lectin-labeled capillaries. B, Phase-contrast micrograph of the same area of thyroid as A. C, Merged image with the light micrograph in green and the blood vessels in red showing that the capillaries pass around the follicles, and that the dark central area as in A correspond to follicles in B. Scale, Width = 660 μm.
Effect of RAdDN-FGFR1, RAd-sVEGFR1, and RAd-sTie2 on thyroid activity, mass, and vascular volume in mice on a goitrogenic regime
Circulating TSH increased from 1.1 ± 0.6 in control mice to 7.4 ± 0.1 ng/ml (P < 0.001) in goitrogen-treated mice and was not significantly changed by treatment with any of the RAds (DN-FGFR1, sVEGFR1, sTie2, GFP) alone or in combination. Total T3 fell from 2.93 ± 0.09 to 0.53 ± 0.41 nmol/liter (P < 0.0001) and total T4 fell from 60.8 ± 17.9 to 16.7 ± 2.9 nmol/liter (P < 0.001) at the end of 14 d treatment with LID and goitrogens. Again RAd treatment had no significant effect on these changes. The general well-being of the mice was not adversely affected by viruses when used alone or three viruses combined (3 x viral load), and although the weight loss was greater in this group, it was not statistically different from the other groups.
Figure 3A shows thyroid mass and Fig. 3B the vascular volume changes occurring after treatment with RAds alone and in combination. The goitrogenic regime induced a more than 2-fold, significant increase in thyroid mass and a 4-fold increase in vascular volume. Goiter mass in mice treated with goitrogen alone, although smaller, did not differ significantly from those treated with goitrogens and injected with RAd-GFP, confirming that adenovirus infection alone does not affect goiter induction. Data from the RAd-GFP-treated mice are shown on the figure. There was no significant difference between the effects of RAd-sTie2 and RAdExTek on goiter mass, and the results were combined for simplicity.
FIG. 3. A, Thyroid mass (milligrams) from control mice (cont) and mice treated with goitrogens and RAd expressing GFP (G), sVEGFR1 (V), DN-FGFR1 (F), sTie2(T), DN-FGFR1+sVEGFR1 (F+V), DN-FGFR1+sTie2 (F+T), and sVEGFR1+sTie2+DN-FGFR1 (V+F+T). *, P < 0.05, ***, P < 0.001, compared with control; ###, P < 0.001, compared with goitrogen + RAdGFP-treated (G) mice. The numbers in the open boxes show the number of thyroid lobes weighed (mean ± SEM). B, Vascular volume (microliters) from control mice (cont) and mice treated with goitrogens and Rad-expressing GFP (G), sVEGFR1 (V), DN-FGFR1 (F), sTie2(T), DN-FGFR1+sVEGFR1 (F+V), DN-FGFR1+sTie2 (F+T), and sVEGFR1+sTie2+DN-FGFR1 (V+F+T). **, P < 0.01, ***, P < 0.001, compared with control; ##, P < 0.01, ###, P < 0.001, compared with goitrogen + RAdGFP-treated (G) mice (n = 3; mean ± SEM).
The difference between the goiter mass in the RAdDN-FGFR1-treated group and the thyroids from the controls was not significant, suggesting that goiter development had been inhibited by this treatment. The excess thyroid mass induced with goitrogen treatment had been reduced by almost 60%. In this analysis this did not quite reach statistical significance, but in our earlier report (12), using a smaller database and comparisons on mice treated and terminated on the same date, P < 0.05, compared with the goiter animals. For the other treatment groups, RAd-sVEGFR1 also reduced goiter mass (43%), but the goiter mass in these mice was significantly different from control (P < 0.05) and was not significantly reduced, compared with goiter control. The decrease in vascular volume with this RAd (Fig. 3B) was significantly different at 50% of the control goiter (P < 0.001). Although RAdDN-FGFR1 treatment reduced the vascular volume, this was not a significant decrease, compared with goitrogen treatment and was significantly different from control. RAd-sVEGFR1 and RAdDN-FGFR1 in combination inhibited the goiter, compared with the control mice, but no more than either virus used alone. Their effects were not additive on goiter mass or vascular volume.
When all three adenoviruses were used in combination, there was significant and complete inhibition of goiter, measured as thyroid mass. There was no significant difference in the mass of these thyroids, compared with the control mice, and there was a significant difference, compared with goiter mice (P < 0.001) (Fig. 3A). There was significant and complete inhibition of vascular volume increases, as shown in Fig. 3B. The thyroids from these animals were, however, red, suggesting that they were hypervascular and confocal micrographs of the vasculature are consistent with this. Figure 4A shows the vasculature from control mice. Figure 4B shows the vasculature from mice treated with goitrogen without RAd treatment. There is a marked increase in vessel diameter as well as a marked increase in intensity of vascular staining consistent with angiogenesis evident in Fig. 4B. Figure 4C shows the thyroid vasculature from mice treated with RAd-sVEGFR1 and goitrogens, and Fig. 4D shows that from mice treated with the combination of three RAds and goitrogens. Treatment with either RAd-sVEGFR1 or the combination of the three RAds produced a vasculature showing an increase of vessel diameter comparable with normal goitrogenesis (Fig. 4B) but a reduction in the number of new vessels. This latter effect was even more marked when animals were treated with three RAds than with RAd-sVEGFR1.
FIG. 4. Confocal micrographs showing thyroid vasculature control (A), treated with goitrogen (B), goitrogen with RAd-sVEGFR1 (C), and goitrogen with RAd-sTie2, RAd-sVEGFR1, and RAdDN-FGFR1 (D). All images are x200 magnification. Scale, Width = 660 μm.
Expression of FGF-2 and Tie2 after goitrogen treatment and the effect of RAd-sTie2
Figure 5A shows the expression of Tie2 and FGF-2 in normal mouse thyroid sections. Expression of both antigens in the thyroid follicular cells of normal mice is evident. The vasculature also stained positively for Tie2 and FGF-2. After 14 d treatment with the goitrogenic regimen, there was a marked increase in expression of both Tie2 and FGF-2 as shown in the second column. Also evident is the follicular cell hyperplasia and the loss of follicles, which occurs with goitrogenesis. In the third column, thyroid sections from a mouse treated with RAd-sTie2 at the start of the goitrogenic regimen are shown. There is a clear reduction in Tie2 expression, compared with the section directly above. Reductions in FGF-2 expression are also seen in the goiters from mice treated with RAd-sTie2, although this effect is not marked. From this we conclude that when Tie2 signaling is blocked, induction of Tie2 is blocked, consistent with positive feedback of Tie-2 signaling on its own receptor. We cannot, however, preclude the possibility that this is an indirect effect. Even more marked reductions in the expression of Tie2 and FGF-2 were seen in mice treated with goitrogen and RAdDN-FGFR1 shown in the last column. This was particularly marked for Tie2, in which expression was reduced to levels less than control mice. We conclude that FGF signaling, which is increased in goitrogenesis, may mediate the expression of Tie2 in mouse thyroids. We also conclude that FGF signaling, directly or indirectly, may increase the expression of FGF-2 in vivo.
FIG. 5. A, Immunostaining of mouse thyroid sections for Tie2 (upper panel) and FGF-2 (lower panel) from control mice (first column), mice treated with goitrogen (G) alone (second column), mice treated with goitrogen and RAd-sTie2 (third column), and mice treated with goitrogen and RAdDN-FGFR1 (fourth column). Magnification, x100. B, Immunostaining of thyroid sections from mice treated with goitrogen (G) + RAd-GFP (A–E) and from thyroids from mice treated with G+ RAds expressing sTie2, DN-FGFR1, and sVEGFR1 in combination (A’-E’). Antibodies: A/A’, FGF-2; B/B’, FGFR1; C/C’, VEGF; D/D’, VEGFR1; E/E’, Tie2. Bar, 100 μm.
Effects of the combination of RAdDN-FGFR1, RAd-sTie2, and RAd-sVEGFR1 on expression of FGF-2, FGFR1, VEGF, VEGFR1, and Tie2 in thyroid tissue sections
The effects of the combination of three RAds on growth factor and receptor expression are shown in Fig. 5B in which representative sections from the thyroids of mice treated with goitrogen and RAd-GFP (upper panel) and those treated with goitrogen and the combination of the three RAds (lower panel) were stained for FGF-2 (A), FGFR1 (B), VEGF (C), VEGFR1 (D), and Tie2 (E). In animals treated with three viruses, there were marked reductions in immunostaining intensity for all the growth factors and their receptors.
To examine the interrelationships between the growth factors further, a more quantitative comparison of the effects of individual RAd treatment on the expression of the growth factors and their receptors was undertaken. Significance was determined from the mean scoring (0–4) of the five independent, blinded observers and compared with the goitrogen-treated, RAd-GFP treated animals.
The rank order for inhibition for animals treated with RAd-sTie2 alone was VEGF greater than VEGFR1 greater than FGFR1 equal to Tie2 greater than FGF-2. These effects were significant for VEGF and VEGFR1. From these data we conclude that Tie2 signaling is required for the induction and maintenance of VEGF expression in thyroid cells.
In thyroids from animals treated with RAdDN-FGFR1 or RAd-sVEGFR1, inhibition of growth factor and receptor expression was less and more variable. For both RAds there was prevention of the increase in Tie2 and FGFR1 expression with no effect on VEGFR1 expression, compared with control goiter. In goiters in RAd-sVEGFR1-treated animals, FGF-2 expression was unaffected. We conclude from this that VEGF expression is independent of VEGFR1 and FGFR1 signaling and that FGFR and Tie2 induction are regulated by VEGFR1 and FGFR1 signaling.
Analyses of the sections of the mice treated with all three viruses, examples of which are shown in Fig. 5B, showed significant inhibition of all the growth factors and their receptors. The order of inhibition was Tie2 greater than FGFR1 equal to VEGFR1 equal to FGF greater than VEGF.
Discussion
Treatment of mice with perchlorate, MMI, and a LID makes them hypothyroid and leads to goitrogenesis. The 4-fold increase in vascular volume in goiter (i.e. a 2-fold increase in vascular volume density) agrees with previous studies using paraffin-embedded sections or electron microscope images to analyze the vascular changes in goiter in rats (1, 3, 33, 34). The increase in vascular volume has been attributed to both increases in the size of existing blood vessels and capillary sprouting.
Our data show that this goiter can be completely inhibited by combined delivery of RAds inhibiting the actions of three angiogenic growth factor signaling pathways. The actions of the angiopoietins, the VEGFs, and FGF signaling have to be inhibited together to effect full inhibition of goiter induced in normal thyroid. We have previously shown that inhibition of FGF signaling with RAdDN-FGFR1 produces a partial reduction in the mass of the goiter produced in response to hypothyroidism (12). Although the vasculature was decreased with this treatment, the greater, significant effect was on goiter mass. From the immunostaining analyses, inhibition of FGF signaling was accompanied by an inhibition of FGF-2 expression and its receptor FGFR1 as well as a reduction in Tie2. There was no inhibition of VEGF, which increases endothelial cell proliferation through VEGFR2. This may explain why inhibition of FGF signaling did not produce effective inhibition of the goiter.
When the thyroid gland is stimulated with TSH, the morphology of the follicles and the follicular cells changes dramatically, the follicular lumens becoming insignificant and the cells hyperactive and columnar. Mass changes may not therefore directly relate to an increase in cell number. Whether follicular cell proliferation is prevented by RAdDN-FGFR1 and whether the mitogenic effects of TSH persist are not known. RAdDN-FGFR1 could potentially inhibit the growth of any FGFR-expressing cells in the thyroid including fibroblasts, endothelial cells, pericytes, and follicular cells, which may contribute to the reduction in goiter mass.
We found that RAd-sVEGFR1 can significantly but only partially limit the development of new vasculature to a goiter, reducing the number of new vessels formed during goitrogenesis but not affecting the increase in diameter of existing vessels. We found that RAds-VEGFR1 reduced the expression of FGFR1 and Tie2 in the goiters, which would also be antiangiogenic. A recent study in mice showed that 293 cells expressing sVEGFR1 could block the growth of a tumor derived from a follicular thyroid carcinoma cell line by 70% (34). The inhibition of physiologic goiter in our study was less dramatic, but the studies differ in that in our model the stimulus for angiogenesis is derived from normal thyroid cells and not those known to have genetic derangements.
RAd-sTie2 did not block the increase in vascularity that occurs during goitrogenesis in mice. Goiter mass in fact was higher in these animals, although this was not statistically significant. Tie2 is thought to be involved more in the restructuring and maturation of the vasculature than its proliferation (13). We found that mRNA for Ang-1 but not Ang-2 was elevated in an iodide-deficiency/goitrogen model of goiter in rats. The sequestration of Ang-1 should prevent the stabilization of the vasculature through phosphatidylinositol 3-kinase but, based on data from tumor models, whether this will have proangiogenic or antiangiogenic consequences is unpredictable (13, 35). The increase in VEGF and VEGFR1, induced in goiter, was significantly reduced with RAd-sTie2, which should limit angiogenesis. The FGF axis was less profoundly affected and whether this can compensate for a reduction in VEGF signaling is unknown. Inhibition of angiopoietin:Tie2 signaling on the endothelial cells and/or the thyroid follicular cells may also affect cell-cell adhesion. There are two studies (36, 37) suggesting Tie2-independent effects of the angiopoietins due to their interactions with integrins. Whether by blocking angiopoietin effects, cell adhesion is blocked in vivo is unknown, although knockout mice models lend support to this role (13). The role of Tie2 on the follicular cells is unknown and has not been explored in the knockout mice. To examine whether its actions could be important in vivo, we determined the effects of RAd-sTie2 on thyroid growth and function in normal 8-wk-old mice. Experiments were terminated 14 d after injection of RAd-sTie2. Although there was an increase in thyroid mass from 1.4 ± 0.4 to 1.9 ± 0.5 mg, this was not significant. There was no effect on thyroid function measured by T4 and TSH levels. Further studies at different time points are required to determine whether Tie2 can provide survival signals in thyroid follicular cells through Akt signaling or affect thyroid function.
Our finding that the combined inhibition of the three angiogenic pathways is required to limit goiter growth completely is likely related to the complete inhibition of the induction of the receptors and their ligands with this treatment. Because elevations in TSH are not blocked by combined RAd treatment, we conclude that the induction of the growth factor receptors and their ligands is not under direct TSH control but is regulated by other growth factors including those under examination in this study. It is interesting that the increase in Tie2 expression in goiter was limited by treatment with any of the three RAds and implies a more complex regulation of follicular Tie2 expression than our earlier in vitro work suggested. We found that TSH, through cAMP, was able to regulate follicular cell Tie2 expression in vitro (18); however, we also found that TSH regulates FGF-2 production in FRTL5 cells (12). Conceivably TSH effects on Tie2 may therefore be mediated through FGF-2 or another intermediary. We also conclude that FGF is, in part, responsible for the increase in FGFR1 synthesis because RAdDN-FGFR1 was able to inhibit FGFR1 expression. The ligands for VEGFRs are increased by both cAMP and growth factors (6, 7). Our data suggest that Tie2 also mediates the production of the VEGFs in the thyroid.
Our data demonstrate that growth of the thyroid vasculature is regulated by the production of local growth factors. We have examined parts of three of the potential angiogenic/goitrogenic pathways. This does not preclude the importance of other pathways or the possibility of redundancies. It must be remembered that we examined the expression of only one of the five VEGFRs, only one of the four VEGFs, only one of the four FGFRs, and only one of 23 of the FGFs. Furthermore, we have not examined the potential contribution of other growth factors of likely importance such as hepatocyte growth factor, IGFs, and TGFs (5, 38). We do show, however, that when goiter was effectively inhibited, all the antigens examined were reduced, suggesting cooperation between the pathways we used. We also found evidence that Tie2 signaling, probably in thyroid follicular cells, regulates the increases in follicular cells of VEGF as well as that of its own receptor and that of VEGFR1 and FGFR1. This strongly suggests a functional role on thyroid cells.
We show that physiologic thyroid growth can be modified by the use of RAds. There have been no studies to our knowledge of the effects of gene therapy on the growth of endogenous tissue driven by endogenous hormones and locally produced endogenous growth actors. Other gene therapy studies have shown the effects on the growth of introduced tumors in mice. Although gene therapy is unlikely to replace surgery as the primary modality of treatment of large multinodular goiters and differentiated thyroid cancers, its use in anaplastic carcinoma of the thyroid, which is invasive and rapidly fatal, may be more relevant.
Acknowledgments
We thank Dr. Jackson Kirkman-Brown for generous help with the confocal microscope.
References
Wollman S, Herveg J, Zeligs J, Ericson L 1978 Blood capillary enlargement during the development of thyroid hyperplasia in the rat. Endocrinology 103:2306–2312
Denef J, Haumont S, Cornette C, Beckers C 1981 Correlated functional and morphometric study of thyroid hyperplasia induced by iodine deficiency. Endocrinology 108:2352–2358
Many M, Denef J, Haumont S 1984 Precocity of the endothelial proliferation during a course of rapid goitrogenesis. Acta Endocrinol 105:487–491
Redmond O, Tuffery AR 1981 Thyroid proliferation, body weight, thyrotropin and thyroid hormones in chronic antithyroid (carbimazole) treatment in rats. J Anat 133:37–47
Eggo MC 2003 Angiogenesis in goitrogenesis. Curr Opin Endocrinol Diabetes 10:341–346
Viglietto G, Romano A, Manzo G, Chiappetta G, Paoletti I, Califano D, Galati MG, Mauriello V, Bruni P, Lao CT, Fusco A, Persico MG 1997 Upregulation of the angiogenic factors PlGf, VEGF and their receptors (Flt-1, Flk-1/KDR) by TSH in cultured thyrocytes and in the thyroid gland of thiouracil-fed rats suggest a TSH-dependent paracrine mechanism for goiter hypervascularization. Oncogene 15:2687–2698
Sato K, Yamazaki K, Shizume K, Kanaji Y, Obara T, Ohsumi K, Demura H, Yamaguchi S, Shibuya M 1995 Stimulation by thyroid-stimulating hormone and Graves’ immunoglobulin G of vascular endothelial growth factor mRNA expression in human thyroid follicles in vitro and Flt mRNA expression in the rat thyroid in vivo. J Clin Invest 96:1295–1302
Isozaki O, Emoto N, Tsushima T, Sato Y, Shizume K, Demura H, Akamizu T, Kohn LD 1992 Opposite regulation of deoxyribonucleic acid synthesis and iodide uptake in rat thyroid cells by basic fibroblast growth factor: correlation with opposite regulation of c-fos and thyrotropin receptor gene expression. Endocrinology 131:2723–2732
DeVito W, Chanoine J, Alex S, Fang S, Stone S, Huber A, Shalhoub V, Linn J, Stein G, Braverman L 1992 Effect of in vivo administration of recombinant acidic FGF on thyroid function in the rat: induction of colloid goiter. Endocrinology 131:729–735
Thompson SD, Franklyn JA, Watkinson JC, Verhaeg JM, Sheppard MC, Eggo MC 1998 Fibroblast growth factors 1 and 2 and fibroblast growth factor receptor 1 are elevated in thyroid hyperplasia. J Clin Endocrinol Metab 83:1336–1341
Patel VA, Hill DJ, Eggo MC, Sheppard MC, Becks GP, Logan A 1996 Changes in the immunohistochemical localisation of fibroblast growth factor-2, transforming growth factor-?1 and thrombospondin-1 are associated with early angiogenic events in the hyperplastic rat thyroid. J Endocrinol 148:485–499
Davies EL, Ramsden JD, Cocks H, Searle PF, Watkinson JC, Ueno H, Logan A, Franklyn JA, Mautner V, Eggo MC 2003 Adenoviral delivery of dominant negative fibroblast growth factor receptor 1 blocks FGF effects in thyroid cells and reduces goitrogenesis in mice. J Clin Endocrinol Metab 88:4472–4480
Thurston G 2003 Role of angiopoietins and Tie receptor tyrosine kinases in angiogenesis and lymphangiogenesis. Cell Tissue Res 314:61–68
Maisonpierre PC, Suri C, Jones PF, Bartunkova S, Wiegand SJ, Radziejewski C, Compton D, McClain J, Aldrich TH, Papadopoulos N, Daly TJ, Davis S, Sato TN, Yancopoulos GD 1997 Angiopoietin-2, a natural antagonist for Tie2 that disrupts in vivo angiogenesis. Science 277:55–60
Davis S, Papadopoulos N, Aldrich TH, Maisonpierre PC, Huang T, Kovac L, Xu A, Leidich R, Radziejewski C, Rafique A, Goldberg J, Jain V, Bailey K, Karow M, Fandl J, Samuelsson SJ, Ioffe E, Rudge JS, Daly TJ, Rad-ziejewski C, Yancopoulos GD 2003 Angiopoietins have distinct modular domains essential for receptor binding, dimerization and superclustering. Nat Struct Biol 10:38–44
Teichert-Kuliszewska K, Maisonpierre PC, Jones N, Campbell AI, Master Z, Bendeck MP, Alitalo K, Dumont DJ, Yancopoulos GD, Stewart DJ 2001 Biological action of angiopoietin-2 in a fibrin matrix model of angiogenesis is associated with activation of Tie2. Cardiovasc Res 49:659–670
Kim I, Kim HG, So JN, Kim JH, Kwak HJ, Koh GY 2000 Angiopoietin-1 regulates endothelial cell survival through the phosphatidylinositol 3'-kinase/Akt signal transduction pathway. Oncogene 19:4549–4552
Ramsden JD, Cocks HC, Shams M, Nijjar S, Watkinson JC, Sheppard MC, Ahmed AS, Eggo MC 2001 Tie-2 is expressed on thyroid follicular cells, is increased in goiter, and is regulated by thyrotropin through cyclic adenosine 3',5'-monophosphate. J Clin Endocrinol Metab 86:2709–2716
Lin P, Buxton JA, Acheson A, Radziejewski C, Maisonpierre PC, Yancopoulos GD, Channon KM, Hale LP, Dewhirst MW, George SE, Peters KG 1998 Antiangiogenic gene therapy targeting the endothelium-specific receptor tyrosine kinase Tie2. Proc Natl Acad Sci USA 95:8829–8834
Schneider M, French B 1993 The advent of adenovirus gene therapy for cardiovascular disease. Circulation 88:1937–1942[Free Full Text]
Carmeliet P, Collen D 1997 Molecular analysis of blood vessel formation and disease. Am J Physiol 273:H2091–H2104
Lin P, Polverini P, Dewhirst M, Shan S, Rao PS, Peters K 1997 Inhibition of tumor angiogenesis using a soluble receptor establishes a role for Tie2 in pathologic vascular growth. J Clin Invest 100:2072–2078
Kuo CJ, Farnebo F, Yu EY, Christofferson R, Swearingen RA, Carter R, von Recum HA, Yuan J, Kamihara J, Flynn E, D’Amato R, Folkman J, Mulligan RC 2001 Comparative evaluation of the antitumor activity of antiangiogenic proteins delivered by gene transfer. Proc Natl Acad Sci USA 98:4605–4610
Hangai M, Moon YS, Kitaya N, Chan CK, Wu DY, Peters KG, Ryan SJ, Hinton DR 2001 Systemically expressed soluble Tie2 Inhibits intraocular neovascularization. Hum Gene Ther 12:1311–1321
He TC, Zhou S, da Costa LT, Yu Y, Kinzler KW, Vogelstein B 1998 A simplified system for generating recombinant adenoviruses. Proc Natl Acad Sci USA 95:2509–2514
Ghosh-Choudhury G, Haj-Ahmad Y, Brinkley P, Rudy J, Graham FL 1986 Human adenovirus cloning vectors based on infectious bacterial plasmids. Gene 50:161–171
Ueno H, Gunn M, Dell K, Tseng Jr A, Williams L 1992 A truncated form of fibroblast growth factor receptor 1 inhibits signal transduction by multiple types of fibroblast growth factor receptor. J Biol Chem 267:1470–1476
Murakami P, McCaman MT 1999 Quantitation of adenovirus DNA and virus particles with the PicoGreen fluorescent dye. Anal Biochem 274:283–288
Eggo MC, King WJ, Black EG, Sheppard MC 1996 Functional human thyroid cells and their insulin-like growth factor binding proteins. J Clin Endocrinol Metab 81:3056–3062
Debbage PL, Griebel J, Ried M, Gneiting T, DeVries A, Hutzler P 1998 Lectin intravital perfusion studies in tumor-bearing mice: micrometer-resolution, wide-area mapping of microvascular labeling, distinguishing efficiently and inefficiently perfused microregions in the tumor. J Histochem Cytochem 46:627–639
Thurston G, McLean JW, Rizen M, Baluk P, Haskell A, Murphy TJ, Hanahan D, McDonald DM 1998 Cationic liposomes target angiogenic endothelial cells in tumors and chronic inflammation in mice. J Clin Invest 101:1401–1413
Howard C, Reed M 1998 Unbiased stereology three-dimensional measurement in microscopy. Oxford, UK: BIOS Scientific Publishers Ltd.
Smeds S, Wollman S 1983 3H-thymidine labeling of endothelial cells in thyroid arteries, veins and lymphatics during thyroid stimulation. Lab Invest 48:285–291
Ye C, Feng C, Wang S, Wang KZQ, Huang N, Liu X, Lin Y, Li M 2004 sFlt-1 gene therapy of follicular thyroid carcinoma. Endocrinology 145:817–822
Stoeltzing O, Ahmad SA, Liu W, McCarty MF, Wey JS, Parikh AA, Fan F, Reinmuth N, Kawaguchi M, Bucana CD, Ellis LM 2003 Angiopoietin-1 inhibits vascular permeability, angiogenesis and growth of hepatic colon cancer tumors. Cancer Res 63:3370–3377
Carlson TR, Feng Y, Maisonpierre PC, Mrksich M, Morla AO 2001 Direct cell adhesion to the angiopoietins mediated by integrins. J Biol Chem 276:26516–26525
Xu Y, Yu Q 2001 Angiopoietin-1, unlike angiopoietin-2, is incorporated into the extracellular matrix via its linker peptide region. J Biol Chem 276:34990–34998
Eggo MC, Quiney V, Campbell S 2003 Local factors regulating growth and of human thyroid cells in vitro and in vivo. Mol Cell Endocrinol 213:47–58(James D. Ramsden, Malcolm)