| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online September 4, 2003 as doi:10.1096/fj.03-0246fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Division of Surgical Research, Children’s Hospital, and Department of Ophthalmology, Harvard Medical School, Boston, Massachusetts, USA
2Correspondence: Department of Surgery, Children’s Hospital, and Department of Ophthalmology Harvard Medical School, 300 Longwood Ave., Boston, MA 02115, USA. E-mail: robert.damato{at}tch.harvard.edu
SPECIFIC AIMS
Our aim was to determine the inheritance pattern of VEGF responsiveness in mice and identify chromosomal regions associated with VEGF responsiveness.
PRINCIPAL FINDINGS
1. Variability in VEGF responsiveness among a wide variety of inbred mouse strains correlates to bFGF responsiveness
Previous studies have shown a substantial difference in angiogenic responsiveness to bFGF among various inbred mouse strains. To confirm that this variability extends to VEGF responsiveness, we performed the corneal neovascularization assay on 12 different inbred mouse strains using VEGF to stimulate vessel growth. As with bFGF, we observed dramatic variations in responsiveness to an angiogenic stimulus among various mouse strains (Fig. 1
), with the vessel area in response to a 180 ng VEGF pellet ranging from 0.4 to 2.2 mm2. We also observed that VEGF responsiveness correlated generally with bFGF responsiveness (Fig. 1)
, with 
60% of the variance in the bFGF responsiveness explicable by VEGF responsiveness. This is consistent with the observation that about half of the bFGF response in the corneal neovascularization assay can be blocked by VEGF antagonists and therefore relies on the VEGF pathway (R. D’Amato, unpublished results).
|
2. Variability in VEGF responsiveness among BXD Strains (which derive from just two parental strains) is nearly as great as among all available inbred mouse strains and is controlled by multiple genes that act in a semidominant manner
To map QTL associated with VEGF responsiveness, we used recombinant inbred (RI) strains. We mapped angiogenic responsiveness genes using a RI strain set derived from the C57BL/6J strain. C57BL/6J’s angiogenic responsiveness sits near the mode of the strains tested (Fig. 1)
, and its genome is being sequenced by the public genome project. Thus, C57BL/6J provide a good baseline for both genetic mapping and analysis of sequence variation. Since the BXD series is the largest set of RIs derived from the C57BL/6J mouse strain, we performed the corneal neovascularization assay on all that were commercially available.
Phenotyping of F1 progeny of C57BL/6J and DBA/2J matings revealed a mean of 1.1 ± 0.22 mm2, approximately halfway between the means of the parental strains (0.8±0.19 and 1.4±0.27 for C57BL/6J and DBA/2J, respectively). Phenotyping the BXD strains, which represent a homozygous recombination of the parental strains, resulted in a broad range of values from 0.3 (BXD-33) to 2.2 (BXD-16), significant deviations from the parental strain values in both directions. The mean (and mode) of the vessel areas for the 30 strains was 1.0 ± 0.34—slightly (not significantly) below that of the F1 animals. The broad unimodal nature of the RI data distribution combined with the intermediate nature of both the F1 value and the RI mean value suggests the existence of multiple QTL governing the responsiveness to VEGF.
3. Composite interval mapping and multiple interval mapping link VEGF responsiveness to AngVq1 and AngVq2, indicating they can explain a significant portion (75% in the MIM model) of the BXD strains’ variance in VEGF responsiveness
Simple interval mapping (SIM, Fig. 2
) revealed several regions of near-significant linkage with the vessel area, but no areas of significant linkage. Given the possibility of multiple-linked QTL suggested by marker association and the inappropriateness of simple interval mapping in such situations, we went on to perform composite interval mapping (CIM, Fig. 2
) with significance levels determined by permutation. This reduced the number of linked areas to two, and these areas exhibited highly significant linkage (P<0.001). We have designated these QTL as AngVq1 (on chromosome 10 near D10Mit20) and AngVq2 (on chromosome 2 near D2Mit6) for angiogenesis due to VEGF.
|
Multiple interval mapping (MIM) analysis was then performed. This revealed several areas of linkage, but only regions on chromosome 2 between D2Mit80 and D2Mit82 and on chromosome 10 between D10Mit194 and D10Mit42 were predicted to explain >20% of the variance in the BXD strains. These regions correspond to those predicted in composite interval mapping to be linked to VEGF responsiveness, thus confirming this linkage. In addition to location and effect, MIM allows the prediction of epistasis among QTL in the model. These tests were performed and no epistatic interactions were revealed. The final model R2 value was 0.92, indicating a good fit between the model and the data.
The positions of the QTL were transferred to a physical map using marker positions annotated in the UCSC (http://genome.ucsc.edu, Feb 2002 freeze) and the MGSCv3 data sets. Since the order of some markers is discordant between the linkage and the sequencing maps, the above analyses were repeated. Both marker map data sets produced linkage to the same areas on chromosomes 2 and 10. MIM with this map also confirmed those areas previously identified. We conclude therefore that there are polymorphisms between C57BL/6J and DBA/2J mice on chromosome 2 between 11.7 and 26.3 Mbp and on chromosome 10 between 46.7 and 82.8 Mbp that modify angiogenic responsiveness to VEGF. Marker genotypes in these regions are in good agreement with this conclusion.
CONCLUSIONS AND SIGNIFICANCE
In recent years it has become evident that the host contributes significantly to the progression of a tumor by providing the tumor’s blood supply. Factors that induce an angiogenic response have been identified and their regulation in the context of angiogenic systems is being carefully studied. However, the possibility of dramatic differences in host response to a given angiogenic stimulus have not been widely explored. Our experiments represent the first attempt at a genome-wide scan to identify polymorphisms that affect angiogenic responsiveness.
The location of the genetic regions we have identified suggest several candidate genes. Candidate genes known to be located near the AngVq1 peak include several extracellular matrix associated components such as proline 4-hydroxylase (P4ha1), several collagens (Col13a1, Col6a1, Col6a2, and Col18a1, the endostatin precursor), a matrix metalloproteinase (Mmp11), and the extracellular matrix receptor component integrin ß2 (Itgb2). All these may modify endothelial invasion and the associated matrix remodeling during angiogenesis. There are also several angiogenesis-related signaling molecules in the region including: the thromboxane A2 receptor (Tbxa2r), ephrin A2 (Efna2), macrophage migration inhibitory factor (Mif), and the adenosine A2a receptor (Adora2a). A few candidate genes in the AngVq2 region include prostaglandin D2 synthase (Ptgds), a cluster of interleukin 1-related molecules (Il1f9, Il1f10, and Il1rn, shown to inhibit corneal angiogenesis in rats), and endothelial differentiation-related factor 1 (Edf1, a calcium-calmodulin binding transcription factor that may regulate endothelial cell differentiation).
The locations of AngVq1 and AngVq1 suggest additional characteristics that may be affected by angiogenic responsiveness. For example, the region associated with AngVq2 appears to overlap with several loci whose effects might be explained by alterations in VEGF responsiveness. These include Kwq6 (Kidney weight), Igf1q1 (Igf1 serum concentration), and the Sluc16 (lung cancer susceptibility). Another lung cancer susceptibility locus, Sluc22, may overlap one of the minor peaks in CIM mapping and correspond to a minor MIM identified loci on chromosome 10.
The region associated with AngVq1 also overlaps loci whose effects might be explained by alterations in VEGF responsiveness. These include Pas11 (modifier of K-ras (pulmonary adenoma sensitivity-1) -1), Bsc1 (Brain size control), and Igfbp3q2 (Igfbp3 serum concentration). Also colocated with AngVq1 is Lifespan2, a QTL for overall animal lifespan. In the report identifying this QTL, the genotype was also associated with increased survival in mice that eventually die of cancer. This suggests that multiple late-life diseases, including cancer, may share pathophysiologic mechanisms affected by the same genomic regions. Since angiogenesis has been shown to be important in other pathogenic processes that may affect lifespan, including atherosclerosis and arthritis, angiogenesis would make a good candidate for the shared pathophysiologic mechanism influencing variation in murine lifespan.
Besides tumor growth, angiogenesis is involved in some normal and pathogenic processes, including atherosclerosis, arthritis, macular degeneration, diabetic retinopathy, psoriasis, endometriosis, wound healing, organ development and regeneration, fat growth, hair growth, and reproduction. We expect that the genes that regulate angiogenesis will have both prognostic and therapeutic value in many of these processes.
While the corneal neovascularization assay is indicative of angiogenic response in the eye and may not entirely predict the angiogenic response in other tissues, we believe that most of the polymorphisms we identified will also regulate angiogenesis systemically. Indeed, a good correlation exists between the corneal angiogenic response and the endothelial migratory ability of cells derived from aortic ring sections of inbred strains, as reported by us earlier. Thus, characterization of novel angiogenesis-regulating genes detected by the QTL mapping of the corneal angiogenic phenotype may broadly identify additional therapeutic targets for antiangiogenic agents. Because some drugs currently in development act on the VEGF pathway, an understanding of the genetics of VEGF response may identify loci that predict response to VEGF-directed antiangiogenic therapy. Finally, these experiments reveal the important effect of the host genetics on angiogenesis and therefore on tumor growth and other angiogenesis-associated pathologic processes. The identification of 
high angiogenic
individuals who are at risk for diseases dependent on angiogenesis will allow for increased monitoring of disease development and progression leading to improved prevention and better clinical management. The possibility that specific and novel antiangiogenic therapy can be developed and implemented at early stages of disease development may render affected individuals low angiogenic and increase survival.
|
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.03-0246fje; doi: 10.1096/fj.03-0246fje 
This article has been cited by other articles:
|
|
 |
|
  T. Udagawa, A. E. Birsner, M. Wood, and R. J. D'Amato Chronic Suppression of Angiogenesis following Radiation Exposure Is Independent of Hematopoietic Reconstitution Cancer Res., March 1, 2007; 67(5): 2040 - 2045. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Becker, R. D. Wright, R. Satchi-Fainaro, T. Funakoshi, J. Folkman, A. L. Kung, and R. J. D'Amato A Novel Noninvasive Model of Endometriosis for Monitoring the Efficacy of Antiangiogenic Therapy Am. J. Pathol., June 1, 2006; 168(6): 2074 - 2084. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Wood, T. Udagawa, Y. Hida, and R. J. D'Amato X-Linked Dominant Growth Suppression of Transplanted Tumors in C57BL/6J-scid Mice Cancer Res., July 1, 2005; 65(13): 5690 - 5695. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Beaudry, J. Force, G. N. Naumov, A. Wang, C. H. Baker, A. Ryan, S. Soker, B. E. Johnson, J. Folkman, and J. V. Heymach Differential Effects of Vascular Endothelial Growth Factor Receptor-2 Inhibitor ZD6474 on Circulating Endothelial Progenitors and Mature Circulating Endothelial Cells: Implications for Use as a Surrogate Marker of Antiangiogenic Activity Clin. Cancer Res., May 1, 2005; 11(9): 3514 - 3522. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Ragolia, T. Palaia, T. B. Koutrouby, and J. K. Maesaka Inhibition of cell cycle progression and migration of vascular smooth muscle cells by prostaglandin D2 synthase: resistance in diabetic Goto-Kakizaki rats Am J Physiol Cell Physiol, November 1, 2004; 287(5): C1273 - C1281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. ROGERS, R. M. ROHAN, A. E. BIRSNER, and R. J. D'AMATO Genetic loci that control the angiogenic response to basic fibroblast growth factor FASEB J, July 1, 2004; 18(10): 1050 - 1059. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
