The isolated oligomerisation domain of TSP-5/COMP has been demonstrated to self-assemble into a five-stranded alpha-helix. To establish the oligomerisation status of D-TSP experimentally and in view of the very high predicted molecular mass of a pentamer of the full-length polypeptide (591,240 Da), the equivalent region of D-TSP was expressed in His-tagged form in Escherichia coli and analysed by Tricine-PAGE under reducing or non-reducing conditions. The reduced protein resolved at the calculated molecular mass, 7.9 kDa, whereas the non-reduced material resolved at an apparent molecular mass of 38 kDa, which corresponded closely to the predicted molecular mass of a pentamer (39.5 kDa) (Adams, 2003).
The properties of full-length recombinant D-TSP were examined in comparison to other ECM molecules. Secreted, V5His6-tagged D-TSP resolved as two bands with apparent molecular masses of 135 kDa and 150 kDa under reducing conditions. The lower band may result from proteolytic susceptibility, as has been noted for mammalian TSPs. Under non-reducing conditions, D-TSP resolved as a single band of apparent molecular mass 600 kDa that was clearly larger than the TSP-5/COMP pentamer of 500 kDa (Adams, 2003).
To characterise native D-TSP, polyclonal antisera were raised to a C-terminal epitope. The affinity-purified immunoglobulins specifically detected proteins of approximately 125,000 Da in 0-16 hour embryos or soft tissue extracts of adult fly bodies under reducing conditions. A presumed proteolytic fragment of around 60,000 Da was also prominent in the fly body. Both bands were competed by the immunising peptide. The secretion of D-TSP by several D. melanogaster cell lines was investigated and it was found that the protein was low or undetectable in the medium of Kc-167 cells and S2 cells, but was readily detectable in the medium of clone 8 imaginal disc cells. To determine whether D-TSP was deposited into ECM, clone 8 cells were grown to high density and extracts made under conditions that preserve assembled, insoluble ECM while removing the cell layer. D-TSP was not extracted by 20 mM NH4OH, but was solubilised by 2 M or 8 M urea. These results demonstrate that endogenous secreted D-TSP becomes associated with ECM in cultured cells. With regard to the production of the 60,000 Da band detected in adult flies, it was noted that minor bands of similar molecular mass were detected in the long-term cell cultures (Adams, 2003).
To localise endogenous matrix-associated D-TSP, non-permeabilised clone 8 cells were stained with the affinity-purified IgG to D-TSP. D-TSP was specifically detected in a non-fibrillar, pericellular location around the cells. The intensity of staining was non-uniform, with clusters of cells showing bright staining and others little or no staining. Staining was blocked in the presence of the immunizing peptide. Pre-treatment of the cultures with 2 M urea abolished the staining completely (Adams, 2003).
Although the amino-terminal domain of insect TSPs did not correspond to the laminin-G-like fold predicted for TSPs 1-4, the presence of canonical heparin-binding motifs in the N-terminal domains of Drosophila and Anopheles TSPs suggested that insect TSPs might also have heparin-binding activity. To test this, conditioned medium from cells expressing recombinant D-TSP was precleared on glutathione-Sepharose beads, then incubated with heparin-Sepharose and washed in buffers with increasing concentration of NaCl. D-TSP bound specifically to heparin-Sepharose and binding activity was detectable even after elution with 0.6 M NaCl. According to normalization by scanning densitometry, 65% was retained in the presence of 0.6 M NaCl and 34% was retained in the presence of 1 M NaCl. To relate the activity to the amino-terminal domain, the same experiment was carried out with a truncated protein (designated P341), consisting of the amino-terminal domain, coiled-coil domain and first EGF-like domain of D-TSP with a His-tag was prepared. This protein resolved as a monomer of 43 kDa under reducing conditions. P341 bound specifically to heparin-Sepharose and the interaction was stable to elution by 1 M NaCl. According to normalisation by scanning densitometry, 75% of the bound material was resistant to elution with 0.6 M NaCl and 45% was retained in the presence of 1 M NaCl. Endogenous D-TSP secreted by clone 8 cells also bound to heparin effectively. These results identify D-TSP, and its amino-terminal domain, as a strong heparin-binding protein (Adams, 2003).
To obtain information on the expression of Drosophila TSP in relation to cell differentiation, the spatial expression of the D-TSP gene during embryogenesis was examined. Transcripts were first detected immediately after egg deposition, suggesting a maternal contribution. During the blastoderm stage, the message disappeared gradually and no message was observed during gastrulation. During germband extension, transcripts appeared again in a patch at the anterior dorsal side. This patch persisted throughout late germband extension and a repetitive pattern became apparent in the ectoderm and mesoderm. This pattern did not change during germband retraction, and at the retracted stage strong transcription was observed in a segmental pattern that corresponded to the future apodeme cells and a subset of myoblasts. Anteriorly, RNA expression was observed in the dorsal pharygneal myoblasts. At stage 15, strong expression persisted in most apodeme cells and in a subset of muscle cells. In embryos close to hatching, most cells of apodemal origin (i.e. intersegmental apodemes and intrasegmental apodemes 1 and 2) showed transcription of D-TSP. Interestingly, transcription of D-TSP was readily detectable in imaginal discs, in particular in the wing disc, where strongest expression was found in the presumptive wing notum and at low levels on the presumptive wing itself. Transcripts were also detected in several areas of the leg disc (Adams, 2003).
Previous studies showed that tsp mRNA is detected in segmental stripes that correspond to the muscle attachment sites from stage 11 of embryonic development (Adams, 2003). An antibody was raised to Tsp and used to correlate the distribution of Tsp protein with the process of somatic muscle development. The specificity of the antibody is demonstrated by its lack of reactivity with embryos homozygous for Df(2L)BSC9, which removes the tsp gene. At stage 12-13, prior to the formation of muscle-tendon junctions, Tsp protein is detected as scattered dots around the Stripe-expressing tendon cells. These dots appear to be external to the tendon cells, as judged by their distance from the tendon cell. At stage 16, following the establishment of the myotendinous junction, Tsp becomes highly concentrated at the muscle-tendon junction sites, as deduced from its localization between the muscle edges and the tendon cell nucleus. The change in Tsp distribution is consistent with a dynamic process, where following the formation of the myotendinous junction, Tsp associates preferentially with this site. To investigate further the relationship between Tsp distribution and muscle-tendon adherens junction formation, the distribution of Tsp was examined in myospheroid (mysXG43) mutant embryos lacking functional integrins. In mys mutant embryos, the muscles migrate normally towards their attachment sites and the muscle pattern appears normal at stage 14-15 of embryonic development. However, at stage 16, when the muscles initiate their contraction, they pull away from the attachment sites and become rounded. Tsp distribution in mysxg43 embryos at stage 16 is not as concentrated as in wild-type embryos and often appears as scattered dots, similarly to stage 12-13 wild-type embryos. Nevertheless, higher accumulation of Tsp is still detected at the muscle edges, suggesting partial integrin-independent association with the muscle cells (Subramanian, 2007).
Tsp was initially recovered in a microarray screen for genes that are downstream of Stripe by comparing the gene expression profile of embryos overexpressing Stripe in the ectoderm with that of wild-type embryos. Further analysis showed that in stripe mutant embryos Tsp protein is still detected, possibly because of earlier Stripe-independent transcriptional input. Importantly, overexpression of Stripe using the engrailed-gal4 driver leads to a significant induction of Tsp expression, confirming the microarray results and the ability of Stripe to induce Tsp expression. Because Stripe expression is greatly upregulated following muscle-tendon interaction, it is assumed that Stripe-dependent Tsp induction is linked to muscle-tendon interaction. It is concluded that Tsp distribution is dynamic and correlates with the biogenesis of adherens junction formation (Subramanian, 2007).
The genomic organization of the tsp gene suggests that the gene produces four splice variants, of which only TspA includes the conserved CTD and the Tsp type-3 repeats. A mutation in the tsp locus (tsp8R) was induced by imprecise excision of an EP element inserted within the tsp coding region, 847 nucleotides 5' to the stop codon of TspA. Although flies homozygous for precise excisions of this EP element are viable, the imprecise excision leads to embryonic lethality. The imprecise excision was mapped using PCR with primers flanking the insertion site and it was found that it removes 3457 nucleotides within the tsp gene. The deletion results in the putative production of a truncated protein, which lacks the conserved CTD, the Tsp type-3 repeats and four out of six EGF repeats. The gene on the 3' flank of the tsp gene (CG11327) is not affected. Staining of the tsp8R homozygous mutants with the anti-Tsp antibody (raised against the N-terminal domain) revealed that the truncated Tsp protein is not detected in these embryos (Subramanian, 2007).
To assess the contribution of Tsp to the assembly of somatic musculature, the phenotype of tsp mutant embryos was analyzed. The initial myotube fusion and the migration of muscles towards tendon cells appear normal in the tsp mutant embryos. However, a large proportion of the somatic muscles of stage 16 tsp8R mutant embryos are rounded. In addition, the muscles do not extend between their corresponding Stripe-expressing tendon cells, as in wild-type embryos. This phenotype is detected in all the muscle types at stage 16, although some variability exists between the distinct muscles, and the phenotype is more severe in embryos at late stage 16. No significant difference was detected between the phenotype of direct muscle-tendon junctions, e.g. the lateral transverse muscles, and indirect junctions, e.g. ventral-lateral muscles. The residual association of the muscles with tendons in tsp mutants may reflect the redundant function of laminin (Wing blister), which may contribute, although to a lesser extent, to muscle-tendon interaction. In addition, maternal tsp, which may still be present at this stage, may contribute to the residual muscle-tendon association. The rounded muscle phenotype is reminiscent of the mys mutant embryos, suggesting that in tsp mutant embryos the association of somatic muscles with tendon cells may be abrogated. The overall pattern of tendon cells was slightly aberrant, as deduced from the Stripe expression pattern, presumably reflecting the aberrant somatic muscle pattern. Essentially, a similar phenotype was observed in embryos trans-heterozygous for tsp8R and Df(2L)BSC9, which uncovers the entire tsp gene, suggesting that tsp8R represents a severe mutant allele of tsp. Importantly, the muscle phenotype of the tsp8R mutant embryos is rescued by overexpressing TspA in tendon cells using the stripe-gal4 driver, but not following overexpression of TspA in the muscles using the mef-2-gal4 driver. This suggests that the tendon-specific expression of Tsp is essential for its function (Subramanian, 2007).
To test whether the somatic muscles in the tsp mutant embryos are capable of forming integrin-mediated adherens junctions, the embryos were stained for integrin ßPS. The typical integrin-positive bands were still detected in each segment of the tsp mutant embryo, corresponding to the ends of the ventral and dorsal longitudinal muscles. However, ectopic integrin staining was detected at various locations, which appeared to correspond to regions of muscle-muscle interactions. This phenotype was detected in all tsp mutant embryos, in at least one segment. Importantly, the edges of each of the lateral transverse muscles, which normally interact with a single tendon cell, exhibited a large reduction in integrin staining except when the lateral transverse muscle was associated with a neighboring muscle cell. The positive staining of integrin at sites overlapping muscle-muscle interactions, as well as the rounding up of some of the muscles, raised the possibility that the somatic muscles of tsp mutant embryos bind primarily with neighboring muscle cells and not with tendon cells, and the relative 'normal' staining of ßPS integrin in the tsp mutant embryos represents sites of muscle-muscle-dependent adherens junctions. Indeed, a lateral view of tsp mutant embryos shows that in some cases ßPS observed at the ends of the muscles is not coupled to Stripe-expressing tendon cells, in contrast to wild-type embryos (Subramanian, 2007).
The muscle-muscle junctions detected by the staining for ßPS may utilize the Tiggrin ligand to assemble the αPS2ßPS integrins on both sides of the hemiadherens junction. Staining with anti-Tiggrin revealed that in tsp mutant embryos Tiggrin accumulation is not observed as stripes but rather in dots, and often Tiggrin-positive dots are detected in ectopic sites. The large Tiggrin dots observed between the longitudinal muscles are consistent with defects in the muscle-tendon interaction, since the tendon cells are arranged as a line of cells at this region, leading to the subsequent line of Tiggrin (and integrin) staining (Subramanian, 2007).
It is concluded that the somatic muscles in tsp mutant embryos fail to form junctions with tendon cells, but are still capable of forming integrin-mediated junctions with neighboring muscles, presumably using Tiggrin as an ECM ligand for the muscle-specific integrin (Subramanian, 2007).
The abnormal pattern of the somatic muscles in the tsp mutant embryo raised the possibility that the muscle-tendon integrin-mediated adhesion is defective in the mutant embryos. A hallmark of appropriate integrin-mediated adhesion is the accumulation of Talin at the cytoplasmic face of the hemiadherens junction, where it binds directly to the integrin cytoplasmic domain, modulating its ligand affinity and recruiting actin microfilaments to this site. A significant reduction of accumulated Talin levels in tsp mutant embryos is observed. Whereas Talin is still detected at the sites of muscle-muscle junctions, it was entirely missing at sites where individual muscles would normally form junctions with single tendon cells, in particular at the junction sites formed between the lateral transverse muscles and their corresponding tendon cells. The lack of Talin at these sites corresponds with the lack of ßPS-integrin staining and is consistent with the loss of appropriate myotendinous junction. Thus, in the absence of functional Tsp, individual myotubes fail to form integrin-mediated adherens junction with tendon cells (Subramanian, 2007).
The results so far are consistent with a model where the tendon-dependent Tsp promotes adhesion of the muscles by binding to the αPS2ßPS integrin receptors. To test this model directly, the CTD was cloned into a Histidine-tag expression vector. In addition, a mutated CTD (CTD*) where the KGD site was mutated into LGE was similarly produced. Both CTDs were produced in bacteria, purified on Ni-NTA agarose beads, concentrated, and diluted in PBS containing 0.3 mM CaCl2. The purified proteins were used at ~40 µg/ml to coat tissue culture plates. S2 cells expressing either αPS2m8ßPS integrin, a mutated αPS2m8 or αPS1ßPS receptors were plated on these cultured dishes and the percentage of spreaded cells was determined (Subramanian, 2007).
Tsp CTD induces a significant elevation in the number of αPS2m8ßPS cells spreading relative to a control where no ligand was added. Importantly, the mutated CTD (CTD-LGE) did not induce cell spreading and was similar to the no-ligand control. Similarly, cells expressing a mutated αPS2m8 did not induce cell spreading on the CTD and behaved like the control cells where no ligand was added. Cells expressing αPS1ßPS did not show a specific elevation in the number of cells spreading on the CTD. The relatively low percentage of spread cells on the Tsp-CTD may reflect a partial reconstitution of the Tsp-CTD produced and purified from bacteria. Large amounts of the Tsp N-terminal domain could not be produced, presumably because of its instability (Subramanian, 2007).
These experiments are consistent with a direct binding of αPS2ßPS integrin receptors with the KGD site that is included in the CTD of Tsp (Subramanian, 2007).
During Drosophila embryogenesis, the attachment of somatic muscles to epidermal tendon cells requires heterodimeric PS-integrin proteins (α- and β-subunits). The α-subunits are expressed complementarily, either tendon cell- or muscle-specific, whereas the β-integrin subunit is expressed in both tissues. Mutations of β-integrin cause a severe muscle detachment phenotype, whereas α-subunit mutations have weaker or only larval muscle detachment phenotypes. Furthermore, mutations of extracellular matrix (ECM) proteins known to act as integrin binding partners have comparatively weak effects only, suggesting the presence of additional integrin binding ECM proteins required for proper muscle attachment. This study reports that mutations in the Drosophila gene thrombospondin (tsp) cause embryonic muscle detachment. tsp is specifically expressed in both developing and mature epidermal tendon cells. Its initial expression in segment border cells, the tendon precursors, is under the control of hedgehog-dependent signaling, whereas tsp expression in differentiated tendon cells depends on the transcription factor encoded by stripe. In the absence of tsp activity, no aspect of muscle pattern is affected, nor is formation of the initial contact between muscle and tendon cells or muscle-to-muscle attachments. However, when muscle contractions occur during late embryogenesis, muscles detach from the tendon cells. The Tsp protein is localized to the tendon cell ECM where muscles attach. Genetic interaction studies indicate that Tsp specifically interacts with the αPS2 integrin and that this interaction is needed to withstand the forces of muscle contractions at the tendon cells (Chanana, 2007).
Attachment of muscles to tendon cells critically depends on integrin activity. The integrin-like muscle detachment phenotype of the tsp mutants as well as the co-localization of Tsp and integrin proteins strongly suggest that Tsp plays a decisive role in the integrin-mediated cell adhesion process of muscles and tendon cells. In fact, Tsp was shown to encode a pentameric glycosylated protein that is part of the ECM (Adams, 2003). Thus, it could indeed function as a direct binding partner of the PS-integrins (Chanana, 2007).
In order to establish such a functional link between tsp and PS-integrins by genetic means, double mutant embryos were generated that carry either a strong loss of function allele for the αPS1 subunit of integrin due to the mewM6 mutation or a hypomorphic allele for the αPS2 subunit due to the ifB2 mutation in combination with only one wild-type copy of tsp (tspΔ6 and tspΔ79) (Chanana, 2007).
In comparison with wild-type embryos mewM6 mutants bearing two copies of tsp develop a normal muscle pattern with muscle detachment in only few segments. mewM6 mutant embryos that have only one wild-type copy of tsp develop a weak muscle detachment phenotype affecting a small number of longitudinal muscles. In contrast, ifB2 mutants bearing two copies of tsp developed a mild detachment phenotype in several segments, which was strongly enhanced, both with respect to the extent of detachment and penetrance in embryos with only one remaining wild-type copy of tsp. In these embryos, longitudinal, ventral as well as dorsal muscles were found to be detached, a phenomenon not observed in tsp mutant embryos. Furthermore, in mewM6 mutants lacking both wild-type copies of tsp, only a mild enhancement of the tsp mutant phenotype was observed. In contrast, ifB2 lacking both wild-type copies of tsp develop dramatic muscle pattern defects beyond an additive effect of the two individual mutant phenotypes. The enhancement of the muscle detachment phenotype of if mutants by the removal of one copy of tsp and the dramatic enhancement that affects even the set of muscles that are not affected in each of the single if or tsp mutants establishes an essential role of Tsp in the αPS2-dependent muscle attachment process. In contrast, the weak effects of the reduction of the tsp dose in hemizygous mewM6 mutants makes a prominent role of Tsp in αPS1-dependent muscle attachment rather unlikely (Chanana, 2007).
Vertebrate Tsp is a glycosylated protein that forms oligomers and is capable of interacting with both calcium and heparin (Adams, 2001). Furthermore, it has been shown to directly interact with the extracellular part of integrin proteins (Lawler, 1988). This interaction depends on a highly conserved RGB motif, which is characteristic of integrin binding proteins of the ECM (Chanana, 2007).Vertebrate genomes code for up to five Tsps (Adams, 2001), whereas the Drosophila genome contains only a single tsp-coding sequence (Adams, 2003), which, however, encodes two Tsp variants which differ in their carboxyterminal regions. Previous biochemical studies on Drosophila Tsp showed that the protein is secreted and able to form a pentameric structure as suggested by the molecular weight of the secreted native complex (Adams, 2003). At the sequence level, the conserved Drosophila Tsp contains all functionally characterized domains including the critical RGD motif required for integrin binding. In contrast to vertebrate Tsp, the RGD motif in Drosophila Tsp is positioned in the aminoterminal region, close to two BBXB sequence motifs known to bind to heparin, instead of the third Tsp/COMP domain (Adams, 2003). In addition, a KGD motif is observed in the third Tsp/COMP domain of Drosophila Tsp which was shown to serve also as an interaction motif for integrins (Ruoslahti, 1996). Drosophila Tsp contains therefore two RGD/KGD motifs that would allow direct binding of PS2, the integrin heterodimer that was previously found to associate with Tig, an interaction that was shown to be dependent on the presence of the RGD motif. Furthermore, an RGD motif is also required to mediate the interaction of PS2 with the laminin α-chain Wb (Chanana, 2007).
Drosophila tsp is expressed in all ectodermal tendon precursor cells, strongly enriched in those positioned at the segment border of the embryo. Furthermore, tsp is expressed in all differentiated tendon cells after muscle contact. Therefore, tsp is expressed in all cells that have previously been identified by the expression of stripe. stripe encodes an EGR-type Zn-finger transcription factor that is required for tendon cell differentiation. Like stripe, the initial expression of tsp is controlled by Hedgehog signaling at the segment borders and requires stripe activity only during the later stages when the tendon cells are already differentiated. These results suggest that the genes stripe and tsp are activated in parallel by Hh-dependent Ci activity, and that stripe activity maintains the expression of tsp during the later stages when Ci activity has ceased (Chanana, 2007).
Tsp is secreted from epidermal tendon cells and accumulates at the tendon cell matrix, a specific ECM to which the muscles attach. The functional characterisation of the newly generated tsp alleles, which fail to express detectable amounts of tsp transcript, showed that Tsp is necessary for the proper anchoring of muscles at the tendons cells. As observed with mutants affecting the β subunit and the αPS2 subunit of integrin, mys and if, respectively, the muscles were found to detach from their epidermal attachment sites once muscle contraction occurs. Thus, tsp activity is not required for any aspect of muscle pattern formation and/or muscle guidance as well as proper adherence to tendon cells but plays an essential role in maintaining the interconnection between muscles and tendons cells once contraction occurs. Although the muscle detachments are less pronounced than in mys or if mutants, the detachment phenotype of tsp mutants is by far stronger than the corresponding phenotypes that are caused by the loss of other ECM proteins, such as Tig, Wb and LanA, known to be integrin interaction partners. The strong and specific enhancement of the detachment phenotype of mutants that carry a weak if allele, in response to the loss of one or both tsp wild-type alleles strongly suggests that Tsp functions as an essential ECM binding partner of αPS2 encoded by if. The specificity of the genetic interaction shown in this study is consistent with the finding that binding to αPS2 requires an RGD motif as has been found in Tig and Wb as well. Mutations of either tig or wb cause weak muscle detachment phenotypes as observed with tsp mutants, suggesting a redundant αPS2 integrin interaction system in which ECM binding is provided by different partners and that each of them is required for the proper anchoring of the muscles. This conclusion is consistent with the finding that tig, wb and tsp mutants display a weaker phenotype than the if loss of function mutants. Based on the specific expression of tsp in both tendon cell precursors and differentiated tendon cells, which differs from the multiple expression sites of tig and wb, together with the strong enhancement of the muscle detachment phenotype in if and tsp double mutants, it appears likely that tsp is the crucial interaction partner of the αPS2 integrin subunit to provide proper anchoring of muscles to tendon cells. This proposal, and the relative contribution of each of the by now three different αPS2 integrin subunit binding proteins, can be tested once double and triple mutant combinations for all the genes involved and biochemical test systems become available (Chanana, 2007).
Reference names in red indicate recommended papers.
Search PubMed for articles about Drosophila Thrombospondin
Adams, J. C. (2001). Thrombospondins: multifunctional regulators of cell interactions. Annu. Rev. Cell Dev. Biol. 17: 25-51. Medline abstract: 11687483
Adams, J. C., et al. (2003). Characterisation of Drosophila thrombospondin defines an early origin of pentameric thrombospondins. J. Mol. Biol. 328(2): 479-94. Medline abstract: 12691755
Adams, J. C. and Lawler, J. (2004). The thrombospondins. Int. J. Biochem. Cell Biol. 36: 961-968. Medline abstract: 15094109
Adolph, K. W. (2001). A thrombospondin homologue in Drosophila melanogaster: cDNA and protein structure. Gene 269(1-2): 177-84. Medline abstract: 11376949
Bornstein, P. (2001). Thrombospondins as matricellular modulators of cell function. J. Clin. Invest. 107(8): 929-34. Medline abstract: 11306593
Chanana, B., Graf, R., Koledachkina, T., Pflanz, R. and Vorbruggen, G. (2007). αPS2 integrin-mediated muscle attachment in Drosophila requires the ECM protein Thrombospondin. Mech Dev. 124(6): 463-75. Medline abstract: 17482800
Christopherson, K. S., Ullian, E. M., Stokes, C. C., Mullowney, C. E., Hell, J. W., Agah, A., Lawler, J., Mosher, D. F., Bornstein, P. and Barres, B. A. (2005). Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell 120: 421-433. Medline abstract: 15707899
Graner, M. W., Bunch, T. A., Baumgartner, S., Kerschen, A. and Brower, D. L. (1998). Splice variants of the Drosophila PS2 integrins differentially interact with RGD-containing fragments of the extracellular proteins tiggrin, ten-m, and D-laminin 2. J. Biol. Chem. 273: 18235-18241. Medline abstract: 9660786
Lawler, J., Weinstein, R. and Hynes, R. O. (1988). Cell attachment to thrombospondin: the role of ARG-GLY-ASP, calcium, and integrin receptors, J. Cell. Biol. 107: 2351-2361. Medline abstract: 2848850
Lawler, J. (2000). The functions of thrombospondin-1 and-2. Curr. Opin. Cell Biol. 12: 634-640. Medline abstract: 10978901
Martin, D., Zusman, S., Li, X., Williams, E. L., Khare, N., DaRocha, S., Chiquet-Ehrismann, R. and Baumgartner, S. (1999). wing blister, a new Drosophila laminin alpha chain required for cell adhesion and migration during embryonic and imaginal development. J. Cell Biol. 145: 191-201. Medline abstract: 10189378
Murphy-Ullrich, J. E. (2001). The de-adhesive activity of matricellular proteins: is intermediate cell adhesion an adaptive state? J. Clin. Invest. 107(7): 785-790. Medline abstract: 11285293
Ruoslahti, E. (2006). RGD and other recognition sequences for integrins, Annu. Rev. Cell. Dev. Biol. 12: 697-715. Medline abstract: 8970741
Scarborough, R. M., Naughton, M. A., Teng, W., Rose, J. W., Phillips, D. R., Nannizzi, L., Arfsten, A., Campbell, A. M. and Charo, I. F. (1993). Design of potent and specific integrin antagonists. Peptide antagonists with high specificity for glycoprotein IIb-IIIa. J. Biol. Chem. 268: 1066-1073. Medline abstract: 8419315
Subramanian, A., Wayburn, B., Bunch, T. and Volk, T. (2007). Thrombospondin-mediated adhesion is essential for the formation of the myotendinous junction in Drosophila. Development 134(7): 1269-78. Medline abstract: 17314133
date revised: 10 August 2007
Home page: The Interactive Fly © 2006 Thomas Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.