Skip Navigation

Cardiovascular Research 2002 53(4):971-983; doi:10.1016/S0008-6363(01)00512-0
© 2002 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Sinha, S.
Right arrow Articles by Kielty, C. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sinha, S.
Right arrow Articles by Kielty, C. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2002, European Society of Cardiology

Expression of latent TGF-beta binding proteins and association with TGF-beta1 and fibrillin-1 following arterial injury

Sanjay Sinhaa,b,1, Anthony M. Heagertyb, C.Adrian Shuttlewortha and Cay M. Kieltya,b,*

aWellcome Trust Centre for Cell-Matrix Research, 2.205 Stopford Building, University of Manchester, Manchester M13 9PT, UK
bDepartment of Medicine, University of Manchester, Oxford Road, Manchester M13 9PT, UK

* Tel.: +44-161-275-5739; fax: +44-161-275-5082 cay.kielty{at}man.ac.uk

Received 28 May 2001; accepted 15 October 2001


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Objectives: Transforming growth factor-β (TGF-β), a potent regulator of wound healing and scar formation, is thought to have a key role in the response to arterial injury. Latent TGF-β binding proteins (LTBPs), members of the fibrillin superfamily, govern TGF-β1 release, targeting and activation in vitro and also play a role as structural components of fibrillin-rich microfibrils. Despite the potential of LTBPs to modulate the response to arterial injury through either or both of these mechanisms, as yet their expression and function in the injured vasculature remain poorly defined. Methods: In this study, a porcine model of coronary angioplasty was used to investigate LTBP-1 and LTBP-2 synthesis and their association with TGF-β1 and fibrillin-1. Results: After angioplasty, increased LTBP-1 and LTBP-2 immunostaining was detected in a similar distribution to increased TGF-β1 expression in the neointima and in the neoadventitia. Overnight organ cultures revealed the formation of large latent TGF-β1 complexes containing LTBP-1. Increased LTBP-1 proteolysis after arterial injury correlated with increased active and latent TGF-β levels. LTBP-2 synthesis increased in response to arterial injury but was neither present in large latent complexes nor proteolytically processed. LTBP-1 and LTBP-2 both co-localised to fibrillin-rich fibrillar structures in the neointima and adventitia. Conclusions: These data suggest that LTBP-1 may have a TGF-β1 binding role in the arterial response to injury, and that LTBP-1 and LTBP-2 may have a structural role in association with microfibrils within the developing neointimal lesion. LTBP-1 proteolysis is potentially an important regulatory step for TGF-β activation in the vasculature and inhibition of proteolysis could represent a novel therapeutic modality for controlling the arterial injury response.

KEYWORDS Growth factors; Extracellular matrix; Angioplasty; Coronary circulation; Smooth muscle


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
Latent transforming growth factor (TGF)-β binding proteins (LTBPs) comprise a family of extracellular glycoproteins that are thought to have a dual role as TGF-β binding proteins and as structural components of extracellular matrix microfibrils (reviewed in Ref. [1]). To date, four isoforms have been identified although most studies have focussed on LTBP-1. LTBPs bind the small latent TGF-β complex, consisting of TGF-β and the latency associated peptide (LAP), through a covalent disulfide bond to form a large latent complex [2,3]. Efficient release of TGF-β from cells appears to be dependent upon large latent complex formation [4,5] while LTBP-1 may additionally be required for growth factor targeting and activation at the cell surface [6,7]. LTBPs are also thought to be important structural components of microfibrils [8,9] and as such may have an independent role in extracellular matrix remodelling in addition to their growth factor related functions.

There is abundant evidence that TGF-β1 has a key role in the vascular response to injury, in part by regulating the repair process [10–13]. However, despite the potential for LTBPs to modulate TGF-β function, as yet their presence and expression in the vasculature remains poorly defined. Furthermore, consistent with their structural role LTBPs in other tissues have been found to co-localise with fibrillin-1 [14–17], a closely related extracellular protein that is an integral component of microfibrils. However, it is unknown whether LTBPs and fibrillin-1 co-localise in the vasculature and whether LTBPs have a structural role in the vascular injury response. Since fibrillin-1 containing microfibrils appear to mediate SMC attachment to elastic fibres in the arterial wall [18,19], then microfibril assembly is likely to be an important component of arterial wound healing following injury. If LTBPs also have a structural role in the vascular injury response as vital microfibrillar components, then they might be expected to co-localise with fibrillin-1 in the injured vessel. However, no such data are presently available and indeed the effects of arterial injury on fibrillin-1 expression and localisation are not defined.

We therefore examined expression and synthesis of LTBP-1 and LTBP-2 in relation to TGF-β1 following arterial injury in a porcine coronary angioplasty model and investigated whether large latent complexes were formed in vivo. We also studied the effects of vascular injury on fibrillin-1 and determined whether this microfibrillar component co-localised with LTBP-1 or LTBP-2. The results presented in this paper constitute the first detailed investigation into LTBP-1 and LTBP-2 involvement in the arterial response to injury.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
2.1. Animal species and coronary angioplasty
Large white pigs (18–25 kg) were used for these studies under authorisation by the Home Office (UK) under the Animals (Scientific Procedures) Act 1986. All procedures conformed with the Guide for the care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85–23, revised 1996). Oversized balloon angioplasty was performed on the left anterior descending and/or circumflex arteries under general anaesthetic (2–2.5% halothane) using fluoroscopic guidance. The animals were sacrificed at varying times post-procedure (see below) and the vessels cut into 3 mm segments, embedded in OCT embedding medium and frozen in liquid nitrogen. Uninjured vessels were also harvested as controls.

2.2. Immunohistochemistry and in situ hybridisation
Only vessels demonstrating significant balloon injury (fracture of the internal elastic lamina (IEL) and dissection through at least part of the medial layer) were included in the immunohistochemical study (total n=19: 5 at 3 days, 3 at 1 week, 5 at 2 weeks, 3 at 4 weeks and 3 at 12 weeks) along with 4 uninjured control vessels. Acetone-fixed, hydrogen peroxide pretreated, frozen sections were blocked using 3% normal goat or horse serum and 3% bovine serum albumin. The following primary antibodies were applied overnight: anti-LTBP-1, 1:200–400 (Ab39 [20], gift from Dr K Miyazono, Tokyo, Japan), anti-LTBP-2, 1:400 (to the amino terminus, exons 1–10 of bovine LTBP-2 [21], gift from Dr R Mecham, St Louis, USA), anti-TGF-β1, 1:25–50 (sc-146, Santa Cruz, USA), anti-TGF-β1 type II receptor, 1:100 (R&D, Abingdon, UK), anti-fibrillin-1, 1:400 (a combination of 11C1.3, Neomarkers, USA, MAB2499 & MAB2502, Chemicon, USA). Equivalent concentrations of normal rabbit serum, rabbit and mouse immunoglobulins were used as controls. Peroxidase/diaminobenzidine staining was performed and sections were counterstained with haematoxylin. LTBP and fibrillin-1 co-localisation was investigated by double labelling indirect immunofluorescence with an anti-rabbit-fluorescein isothiocyanate conjugate (1:100; Sigma, Poole, UK) and an anti-mouse-Cy3 conjugate (1:200, Sigma) for detection. In situ hybridisation was performed using digoxygenin-labelled (Roche, Germany) sense and anti-sense riboprobes (400 ng/ml) to porcine TGF-β1 and visualised with an anti-digoxygenin antibody (Roche) conjugated to alkaline phosphatase.

2.3. Immunoprecipitation
A novel ex vivo organ culture model was established to investigate LTBP synthesis in the vasculature. Coronary arteries at 1 week post-injury and uninjured controls were dissected out of the heart immediately post-mortem, chopped into 1 mm3 pieces and cultured overnight under serum-free conditions in cysteine-free medium supplemented with 35S-cysteine. Low passage porcine vascular SMC, cultured from the aortic medial layer by enzyme dispersion, were also labelled with 35S-cysteine in a similar manner. Samples consisted of conditioned medium from both organ and cell cultures. Protease inhibitors were added to all samples and fibronectin was removed using gelatin-sepharose 4B. Anti-fibronectin antibodies and protein A sepharose were used to preclear the samples and these precipitates were run alongside the LTBP immunoprecipitates to identify residual contaminating fibronectin bands. The following antibodies were used for immunoprecipitation: anti-LTBP1 (Ab 39), anti-LTBP2 (amino terminal) [21], anti-LTBP2 (L9F [22], gift from Dr R.P. Mecham, St Louis, USA) and anti-LAP-β1 (LT-1 [4], gift from Dr C. Heldin, Uppsala, Sweden). All immunoprecipitates were run on SDS–polyacrylamide gels and placed against film after drying. Band intensities were quantified using two dimensional densitometry on the Aida 2.0 image analysis program. A two tailed student's t-test was used to compare the results of injured versus uninjured vessels.

2.4. TGF-β measurement
Active TGF-β in organ culture medium was assayed using mink lung epithelial cells (MLEC) that had been stably transfected with the plasminogen activator inhibitor-1 (PAI-1) promoter coupled to a luciferase reporter gene (cells donated by Dr D.B. Rifkin, New York, USA) as described by Abe et al. [23]. Total TGF-β was estimated by heat treating an aliquot of organ culture medium at 80°C for 10 min, prior to the assay. Two tailed student's t-tests were carried out to assess the significance of any differences between injured and uninjured vessels.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
3.1. Immunohistochemistry
3.1.1. Increased expression of LTBP-1 and LTBP-2 following vascular injury
In order to examine the temporal and spatial distribution of LTBP-1 and LTBP-2 after arterial injury, immunohistochemistry was performed on a porcine model of coronary artery injury. Both isoforms were detected in the internal elastic lamina (IEL) and external elastic lamina (EEL) of the uninjured vessel (Fig. 1A and E), structures rich in elastic fibres. Immunostaining at 1 and 2 weeks post-injury demonstrated the presence of both isoforms in the neointima and in the neoadventitia (Fig. 1B, C, F and G). However, LTBP-2 seemed to be more widespread than LTBP-1. The extent of immunostaining for LTBP-1 had diminished by 4 weeks and beyond, with little difference between distinct parts of the vessel wall and was similar in appearance to the uninjured vessel (Fig. 1D). LTBP-2 immunostaining also decreased but low levels of staining were still detectable at 12 weeks widely throughout the vessel wall (Fig. 1H). In summary, both LTBP isoforms were associated with elastic tissues and found in the regions of the vessel wall that exhibited an injury response. However, subtle differences in their expression patterns were noted with respect to extent and timecourse of staining.


Figure 1
View larger version (114K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 LTBP-1 and LTBP-2 immunolocalisation. LTBP-1 detected in the IEL and the EEL of the uninjured vessel (A). At 1 or 2 weeks (B, C) after injury, LTBP-1 was visible in neoadventitia and neointima. At 4 weeks and 12 weeks (D), immunostaining had decreased to the level of the uninjured vessel. The boxed area in B is shown at higher magnification in C. LTBP-2 seen in the IEL and EEL of normal vessels (E). LTBP-2 was detected at 1 week and at 2 weeks (F, G) in the neointima and neoadventitia. The box in panel F is shown at higher power in G. Immunostaining was significantly reduced by 12 weeks (H). Normal rabbit serum control (I). NIH=neointimal hyperplasia, L=lumen, NA=neoadventitia, m=media. (Magnification: A,C,G x100; B,D–F,H,I x25).

 
3.1.2. TGF-β1 and the type II receptor are upregulated and co-distributed with LTBPs
If LTBPs were involved in regulating TGF-β1 availability in the vasculature, then it might be predicted that LTBPs and TGF-β1 would have a similar pattern of expression following arterial injury. TGF-β1 was initially localised to the neoadventitia at 3 days and was found in both neoadventitia and neointima at 1 and 2 weeks (Fig. 2A and B). Virtually no staining for this growth factor was detected after this time or in normal uninjured vessels. Expression of TGF-β1 mRNA was also examined using in situ hybridisation. Little or no expression was detected in uninjured vessels, although there was upregulation, initially within the neoadventitia, at 3 days post-injury (Fig. 2D). By 1 week, the transcript was found in both the neoadventitia and the neointima (Fig. 2E). Expression of mRNA decreased rapidly beyond 2 weeks and no expression was seen at 4 weeks or later (Fig. 2G). No significant signal was seen using the corresponding sense strand as control (Fig. 2H–K).


Figure 2
View larger version (138K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 TGF-β1 and type II receptor immunolocalisation and TGF-β1 in situ hybridisation. TGF-β1 protein localized to the neoadventitia (A) at 3 days. Increased neointimal and neoadventitial immunostaining were detected at 2 weeks (B and C). The box in panel B is shown at higher magnification in C. TGF-β1 was not detected beyond 2 weeks (data not shown). Neoadventitial cells expressing TGF-β1 mRNA also appeared at 3 days (D). Subsequent mRNA expression was seen in the neointima and neoadventitia at 1 week (E), in the sub-luminal neointima and adventitia at 2 weeks (F) and by endothelial cells only at 12 weeks (G). Panels H–K represent sense controls for D–G respectively. Sections DandH incubated with alkaline phophatase substrate for 20 min only due to intensity of signal, E–G and I–K incubated for 6 h. Low levels of the type II receptor detected in normal uninjured media (L). At 1 week post-injury, neoadventitial and neointimal cells expressed the receptor (M), but by 2 weeks, most of the expression was confined to the neointima (N). Normal mouse serum control at 2 weeks (O). NIH=neointimal hyperplasia, L=lumen, m=media, NA=neoadventitia, AS=antisense. (Magnification: A,C x100; B,L–O x25, D–K x50).

 
In order to define regions of the vessel wall where TGF-β signalling may be occurring following injury, immunohistochemistry was performed for the TGF-β type II receptor. After injury, both neointimal and neoadventitial cells stained for the type II receptor at 1 week while the receptor was detected predominantly in the neointima by 2 weeks (Fig. 2M and N). This increase in type II receptor staining was transient and no significant staining was seen in any part of the vessel wall by 4 weeks, similar to the uninjured state (Fig. 2L).

These results confirm that following arterial injury, TGF-β1 and its type II receptor are expressed in a similar distribution to the putative extracellular binding proteins, LTBP-1 and LTBP-2.

3.1.3. Fibrillin-1 co-localises with LTBP-1 and LTBP-2 in the injured vasculature
LTBPs have been found to co-localise with fibrillin-1, a major component of elastic tissues, in skin, skeleton and kidney [14–17] although such a relationship has not yet been examined in the normal or injured vasculature. Evidence of such co-localisation during the injury response would suggest that LTBPs may interact with fibrillin-1 as integral components of new microfibrils formed during remodelling of the extracellular matrix following injury.

The distribution of fibrillin-1 was assessed initially in normal and injured vessels. Widespread fibrillin-1 staining was seen in vascular connective tissues. In uninjured vessels, there was immunostaining of the IEL, EEL and elastic tissue within the medial layer (Fig. 3A). At 1 and 2 weeks post-injury, fibrillin-1 was detected in the neoadventitia and in the neointima (Fig. 3B and C). Cellular staining intensity decreased by 4 weeks and at this point and at 12 weeks, the staining pattern exhibited a predominantly extracellular distribution (Fig. 3E and F).


Figure 3
View larger version (133K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Fibrillin-1 immunolocalisation and co-localisation with LTBP-1 and LTBP-2. Fibrillin-1 was detected in the IEL, the EEL, elastic fibres within the media and in the loose connective tissue surrounding the normal vessel (A). After injury, fibrillin-1 was found in the neointima and neoadventitia at 1 and 2 weeks (B and C). Fibrillin-1 persisted in these regions at 12 weeks in an extracellular distribution (E and F). Boxed areas in panels B and E displayed at higher magnification in C and F. Mouse immunoglobulin control (D) at 2 weeks. Double-labelling immunofluorescent studies demonstrated co-localisation of LTBP-1 and fibrillin-1 in uninjured vessels (G) and 2 weeks post-injury (H). Co-localisation was seen as yellow, prominently in the IEL and to a lesser extent in the EEL in the uninjured vessel (G). In the injured vessel, co-localisation was found to fibrillar structures in the neointima and fractured IEL (H). Co-localisation of LTBP-2 and fibrillin-1 in uninjured vessels (I) and 1 week post-injury (J). Co-localisation is seen as yellow, prominently in the IEL and to a lesser extent in the EEL in uninjured vessels (I). In the injured vessel, co-localisation was found to fibrillar structures in the neointima and IEL (J). Note lack of co-localisation of fibrillin-1 with either LTBP-1 or LTBP-2 in the normal or injured media. Smaller inset panels show corresponding immunostaining for LTBP-1 or LTBP-2 alone in green and for fibrillin-1 alone in red. NIH=neointimal hyperplasia, L=lumen, NA=neoadventitia, m=media. (Magnification: A,C,F,G–J x100, B,D,E x25).

 
Co-localisation of fibrillin-1 with LTBP-1 or LTBP-2 to extracellular fibrillar structures was demonstrated in both normal and injured vessels. Co-localisation was prominent in the uninjured IEL (Fig. 3G and I) and in the neointima in injured vessels and less apparent in the media (Fig. 3H and J). Controls were performed using normal rabbit serum or mouse immunoglobulins (as appropriate) in place of each one of the primary antibodies in turn to exclude secondary antibody cross-reactivity (data not shown). In addition, substitution of both primary antibodies simultaneously with normal rabbit serum and mouse immunoglobulins revealed no discernible image, thus confirming the absence of significant autofluorescence or background at the exposure settings used for photographing the images of the primaries.

In summary, fibrillin-1 appeared to be a significant extracellular component of the arterial injury response. Furthermore, the co-localisation between LTBPs and fibrillin-1 indicates that LTBPs may also have a structural role, in addition to any possible growth factor binding role, during the remodelling of the extracellular matrix in response to vascular injury.

3.2. Immunoprecipitation studies
3.2.1. Smooth muscle cell (SMC) synthesise LTBP-1 and LTBP-2
SMC have previously been found to synthesise LTBP-1 in culture [7] although their ability to synthesise LTBP-2 has not yet been investigated. In order to confirm that porcine SMC were able to synthesise both LTBP-1 and LTBP-2, immunoprecipitation studies using radiolabelled porcine SMC cultures were carried out.

Immunoprecipitation studies using Ab39 revealed two LTBP-1 bands in the conditioned medium at 190 kDa and 200 kDa (increasing to 210 and 220 kDa on reduction) (Fig. 4, lanes 2 and 4). Two LTBP-2 species were also immunoprecipitated from the conditioned medium at 205 kDa and 215 kDa (increasing to 230 kDa and 240 kDa on reduction, Fig. 4, lanes 6 and 8). In addition, similar bands were found using an alternative anti-LTBP-2 antibody (L9F, data not shown). Two distinct bands were detected for both LTBP-1 and LTBP-2 which may represent alternative splicing, variable post-translational modifications or proteolytic cleavage of these molecules whilst in culture overnight.


Figure 4
View larger version (60K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 LTBP-1 and LTBP-2 immunoprecipitation from SMC cultures. LTBP-1 immunoprecipitates from SMC conditioned medium were run non-reduced (lane 2) or reduced (lane 4) on an 8% polyacrylamide gel. Twin LTBP-1 bands (marked) were detected in the medium at 190 kDa and 200 kDa (lane 2). On reduction the LTBP-1 bands migrated at 210 kDa and 220 kDa (lane 4). Precleared fibronectin (FN) precipitate was also run alongside (lanes 1 and 3) to allow identification of contaminating fibronectin bands (marked *). LTBP-2 immunoprecipitates from SMC conditioned medium were also run non-reduced (lane 6) and reduced (lane 8) on an 8% polyacrylamide gel. Two LTBP-2 bands (marked, at 205 kDa and 215 kDa) were seen in the conditioned medium (lane 6). LTBP-2 also migrated more slowly under reducing conditions (lane 8). Precleared fibronectin (FN) precipitate was again run on the same gel (lanes 5 and 7) to allow identification of contaminating fibronectin bands (marked *).

 
Non-specific association of fibronectin to the anti-LTBP antibodies was addressed by preclearing the supernatant with gelatin-sepharose before immunoprecipitation. However, residual fibronectin bands were still detected as monomers at 250 kDa and as probable high molecular weight dimers in the stacking gel (non-reduced), that also migrated at 250 kDa on reduction. Thus, in all immunoprecipitations, fibronectin bands (marked * in Fig. 4) were distinguished from LTBPs by carrying out parallel anti-fibronectin immunoprecipitations from the supernatant samples. The anti-fibronectin precipitates (Fig. 4, lanes 1, 3, 5, 7) were run on the same gel as the anti-LTBP immunoprecipitates (lanes 2, 4, 6, 8) in order to distinguish LTBPs from fibronectin cross-contamination.

In SMC cultures, no significant evidence of a higher molecular weight large latent complex was detected, although this may have been masked by the fibronectin bands. Nor was there any evidence that LAP or TGF-β species were released under reducing conditions since no bands representing either the LAP monomer (37.5 kDa) or the TGF-β monomer (12.5 kDa) were detected. Moreover, none of the LTBP bands were seen to migrate faster upon reduction, which would have been expected had there been loss of the small latent complex (LAP-TGF-β dimer, approx 100 kDa). These findings suggest that in these cell cultures most of the LTBPs released are not complexed to LAP/TGF-β. In summary, porcine SMC were shown to synthesise both LTBP-1 and LTBP-2 although large latent complexes were not readily detected in these immunoprecipitation studies.

3.2.2. LTBP-1 is synthesised by the coronary vasculature and may show increased proteolysis following injury
To investigate LTBP synthesis and large latent complex formation after arterial injury, coronary arteries were obtained 7 days post-angioplasty. These and uninjured control vessels were radiolabelled in overnight organ cultures and immunoprecipitations carried out using the conditioned culture medium.

Both injured and uninjured vessels were found to synthesise LTBP-1, detected in the organ culture medium as a high molecular weight, diffuse band at approximately 210 kDa and also as three smaller well defined bands at 175 kDa, 160 kDa and 150 kDa, when reduced (Fig. 5). LTBP-1 is susceptible to proteolytic cleavage [24] and tryptic digestion of recombinant protein in previous studies [4] produced a very similar pattern to that seen in these organ cultures, with increasing trypsin incubation time leading to a greater number of smaller fragments. Since a variety of proteases are upregulated in response to injury [25,26], it is likely that the banding patterns seen here are due to proteolytic degradation of the full length LTBP-1 molecule.


Figure 5
View larger version (68K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 LTBP-1 immunoprecipitation from injured and uninjured vessel organ culture medium. Both uninjured (lanes 1 and 2) and injured (lanes 4 and 5) vessels were found to synthesise full length LTBP-1 (L) and at least three distinct faster migrating bands (L-{Delta}1, L-{Delta}2 and L-{Delta}3) which were consistent with proteolytically processed forms of LTBP-1. Densitometric analysis revealed no overall difference in band intensities between injured and uninjured vessels. However, there was a significant increase in the intensity of the smallest species (L-{Delta}3), a reduction in intensity of the larger species (L-{Delta}1) and a small but nonsignificant reduction in intensity of the full length species (L) in injured vessels compared to uninjured. Densitometry was performed by measuring band intensities as demonstrated by box a and box b. For each lane, an appropriately sized box c was used to measure the background which was subtracted from the band intensity measurement. Band intensities are given as mean±S.E.M. Control immunoprecipitations were carried out with normal rabbit serum (lane 3) and anti-fibronectin antibodies (lane 6). Schematic diagarams of full length and proteolytically processed forms of LTBP-1 adjacent to the bands on the gel are for general illustrative purposes only and should not be taken as representing precise cleavage sites or events. (6% polyacrylamide gel, reduced). FN=fibronectin.

 
Using densitometry, only a small non-significant increase in the total amount of radio-labelled LTBP-1 was detected, produced by the injured compared to the uninjured vessel organ cultures. However, there was a clear distinction in the banding patterns obtained (Fig. 5), such that in uninjured vessels, the higher molecular weight bands (L and L-{Delta}1) were predominant, while in injured vessels, there was a preponderance of the lower molecular weight bands (L-{Delta}2 and L-{Delta}3). These findings indicated increased proteolysis in injured compared to uninjured vessels.

3.2.3. LTBP-1 forms large latent complexes in the vasculature
To investigate whether large latent complexes were produced by the injured or uninjured vessels, immunoprecipitations were performed on conditioned medium using an antibody to LAP (LT-1). These precipitates were run in parallel with LTBP-1 immunoprecipitates, obtained using Ab39, and analysed in both reduced and non-reduced states (Fig. 6). If coronary vessels do synthesise large latent complexes consisting of LTBP-1 bound via a disulphide bond to LAP/TGF-β, then immunoprecipitation using LT-1 would be expected to extract a high molecular weight complex in the unreduced state and release LTBP-1 on reduction. As predicted, when anti-LAP precipitates were run in the reduced state, bands that co-migrated with truncated and full length LTBP-1 were detected (Fig. 6, lanes 11 and 12, marked with), highlighting that large latent complexes were indeed synthesised by the coronary vessels in organ culture. However, a discrete high molecular weight large latent complex was difficult to identify in the non-reduced state (Fig. 6, lanes 5 and 6), possibly due to the presence of residual fibronectin bands or other high molecular weight contaminants. Moreover, the bands isolated by the anti-LAP antibody were faint in comparison to the LTBP-1 bands precipitated by Ab39 which indicates that in the vasculature, only a proportion of the LTBP-1 is released as part of a large latent complex.


Figure 6
View larger version (56K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 LTBP-1 and LAP immunoprecipitation from organ cultures. LTBP-1 immunoprecipitation from organ culture media revealed a diffuse high molecular weight band at approximately 200 kDa (marked]) and several lower molecular weight bands (lanes 2–4). Immunoprecipitation using anti-LAP antibodies, on reduction, revealed a band that co-migrated with the smallest LTBP-1 fragment and a fainter band that co-migrated with full length LTBP-1 (both marked as in lanes 11 and 12). Endogenous immunoglobulin contamination (marked with) in addition to fibronectin (FN, marked *) was detected on preclearance with anti-fibronectin antibodies (lanes 1 and 7). (4–18% polyacrylamide gel). Lane 1: Injured medium/anti-fibronectin preclearance. Lanes 2 and 3: Injured media/anti-LTBP-1. Lane 4: Uninjured medium/anti-LTBP-1. Lane 5: Injured medium/anti-LAP. Lane 6: Uninjured medium/anti-LAP. Lanes 7–12: As lanes 1–6, but reduced before electrophoresis. Schematic diagarams of full length and proteolytically processed forms of LTBP-1 adjacent to the bands on the gel are for general illustrative purposes only and should not be taken as representing precise cleavage sites or events.

 
3.2.4. LTBP-2 synthesis in increased following arterial injury
Coronary arteries also synthesised LTBP-2, detected in the organ culture conditioned medium as a single species running at 215 kDa (Fig. 7). No smaller bands were seen to dissociate on reduction and bands from anti-LAP immunoprecipitation did not co-migrate with the LTBP-2 bands (data not shown). Thus, we did not find any evidence in this study to suggest that LTBP-2 formed a large latent complex with LAP/TGF-β in the vasculature. However, In contrast to LTBP-1, overall synthesis of LTBP-2 was significantly upregulated as a result of vessel injury, consistent with the immunohistochemical findings.


Figure 7
View larger version (67K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 LTBP-2 immunoprecipitation from injured and uninjured vessel organ culture medium. LTBP-2 production increased in injured vessel organ culture supernatants (lanes 4–6) compared to uninjured (lanes 1–3). (4–18% polyacrylamide gel, non-reduced). Densitometry was performed by calculating intensity in box a minus box b (background) for each lane. Mean band intensities are±S.E.M.

 
3.3. Increased TGF-β levels in injured vessels
LTBP proteolysis is a key step in the release of matrix associated TGF-β1 [7,24]. Since, there appeared to be increased LTBP-1 proteolysis following vessel injury in our model, we assessed TGF-β levels in the organ culture conditioned medium as a first step in investigating a potential role for LTBP-1 in governing TGF-β availability in the injured vasculature. Active and total TGF-β were measured in five injured and four uninjured vessel organ culture media (Fig. 8). The results indicate that both active and total TGF-β were present at significantly higher concentrations in the injured vessel organ cultures than in uninjured cultures.


Figure 8
View larger version (55K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 8 Active and total TGF-β in organ cultures. Both active and total TGF-β were significantly increased in the injured vessel organ cultures compared to uninjured vessels. Values in the Table 1 are mean values±S.E.M. and error bars on the chart represent S.E.M.

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
The distribution and expression of LTBPs, extracellular matrix proteins with the potential to regulate TGF-β1 function and to act as key structural elements, in the arterial response to injury remain poorly defined. LTBP-1 mRNA expression has been detected by northern blotting of extracts from ‘heart’ but it was unclear whether expression was vascular in origin or not [27]. LTBP-1 protein was detected in the arterial wall during a study of rheumatoid synovium [28] and recently, Kanzaki and colleagues [29] reported increased immunostaining for LTBP-1 in injured rat carotid neointima. No comparable data are available for the other LTBP isoforms although LTBP-2 mRNA was found in developing vessel walls in the mouse embryo [30] and the protein has been localised to elastin associated microfibrils in the vasculature [22]. Furthermore, the biochemical composition of LTBP-containing macromolecules in the vasculature and whether large latent TGF-β complexes occur in normal or diseased vessels have not been investigated. This paper characterises the spatial and temporal distribution of LTBP-1 and LTBP-2 in response to arterial injury and shows, for the first time in vascular tissues, that LTBP-1 but not LTBP-2, forms large latent TGF-β complexes and has increased susceptibility to proteolysis following injury.


Table 1
Uninjured Injured P
Active TGF-β (pg/ml) 563±30 746±59 0.04
Total TGF-β (pg/ml) 1400±178 4740±1213 0.047
Vessel weights (mg) 243±33 342±82 NS (0.34)
Number of vessels 4 5

Increased TGF-β1 expression and signalling are major determinants of the arterial injury response. Studies using co-cultures of endothelial cells and SMC or pericytes have shown that LTBP-1, released as part of a large latent complex, has a major role in the targeting and activation of the growth factor [7]. Initially, the latent complex is bound to the extracellular matrix through interactions with LTBP-1. Proteolytic cleavage of LTBP-1 subsequently allows the release of a truncated form of the large latent complex. Finally, LTBP-1 may also have a role in the activation of TGF-β1 at the cell surface. Despite many new insights into the role of LTBP-1 in targeting and activation of TGF-β1 in cell culture studies, there are virtually no insights into the events regulating TGF-β1 targeting and availability in vivo. The co-distribution and upregulation of TGF-β1, its type II receptor and LTBP isoforms in the neointima and neoadventitia demonstrated in this study suggest that the cell culture derived model [6,7], whereby LTBP-1 is involved in targeting and activating TGF-β1, may also be applicable to growth factor targeting and activation in the injured coronary artery.

A role for LTBP-1 in TGF-β1 targeting and activation is further supported by two findings: firstly, that LTBP-1 proteolysis is increased in coronary arteries post-injury (Fig. 5), most likely as a result of the well documented upregulation of a variety of proteases following arterial injury [25,26]. Secondly, immunoprecipitation using an antibody to LAP releases full length and truncated LTBP-1 under reducing conditions (Fig. 6, lanes 11 and 12) confirming that large latent complexes that include LTBP-1 bound to LAP via a disulfide bond are indeed synthesised and present in the vasculature. In cell culture, covalent binding of the large latent complex to the extracellular matrix with subsequent proteolytic release appears to be a prerequisite for growth factor activation [6,31]. Thus, it appears that LTBP-1 proteolysis has a similar role in regulating matrix bound TGF-β release and activation in the vasculature. Our findings that raised total and active TGF-β levels were associated with increased proteolysis in injured vessels, compared to non-injured coronary arteries, strongly supports this concept. Taken together these data indicate that LTBP-1 plays an important role in the targeting and release of TGF-β1 in the response to arterial injury and is thus a novel regulator of the injury response.

TGF-β1 may be released as a large latent complex from a variety of cell types following vascular injury including platelets, macrophages/monocytes, adventitial myofibroblasts as well as SMC, many of which have also been shown to release LTBP-1 [2,3,7,32]. While the relative contributions of different cell types is likely to depend upon the vessel type and extent of injury, our studies show that in the coronary balloon injury model in the pig, both adventitial myofibroblasts and neointimal SMC contribute significantly to growth factor release, as judged by in situ hybridisation (Fig. 2) and also stain for LTBPs by immunohistochemistry. The spatiotemporal distribution of TGF-β1 and LTBP expression in arterial injury models in which there is significantly less injury, such as endothlial denudation models, may be different and remains to be determined. However, during balloon angioplasty in patients, the extent of vessel injury is typically severe with dissection through the media [33]. Thus, our findings may have greater relevence to clinical angioplasty than data from models with lesser degrees of injury.

The immunoprecipitation studies demonstrated for the first time that SMC synthesise LTBP-2 in culture in addition to LTBP-1 [7]. These results corroborate the immunohistochemical findings and suggest that at least some of the increased staining was specifically due to increased synthesis by neointimal SMC. However, in contrast to LTBP-1, no LTBP-2 was co-precipitated by the anti-LAP antibody and we were unable to detect evidence for significant LTBP-2 proteolysis in the coronary artery organ cultures. These data suggest that LTBP-2 may not have a significant TGF-β modulating role in the vasculature. As such, the conclusions are consistent with recent studies by Saharinen and Keski-Oja [34] that challenge previous findings that LTBP-2 was able to bind the small latent TGF-β complex.

Earlier studies have established that LTBP-1 localises to fibrillin-rich microfibrils in a variety of cell and organ cultures [8,9,35] and in the skin [14,15] and skeleton [16], while LTBP-2 has been found to be associated with microfibrils around elastic fibres in bovine nuchal ligament [22]. Given their close structural homology, it is possible that LTBPs associate with fibrillins as integral components of the microfibril, although the precise molecular interactions are as yet unclear. Recently, studies based on mice homozygous for a targeted hypomorphic allele of fibrillin-1 have shown that fibrillin-1 abnormalities result in abnormal extracellular matrix turnover and phenotypic modulation of SMC [36] contributing to the vascular abnormalities found in Marfan syndrome, a disease associated with mutations in fibrillin-1. Thus microfibril structure and/or function appear to be important regulators of contractile vascular SMC phenotype and ECM turnover. These are two key processes that determine the arterial injury response and are thought to be dysregulated in restenosis. The co-localisation of LTBP-1 and LTBP-2 with fibrillin-1 in a fibrillar distribution in the injured vessel, seen in this study, strongly suggests that these LTBP isoforms interact with newly deposited microfibrils within the neointima and neoadventitia. Hence, LTBP-1 and LTBP-2 also have the potential to influence the arterial injury response as structural components of the microfibril, independent of a growth factor binding role.

This study has investigated LTBP expression within the vasculature after balloon injury. Our findings are consistent with a growth factor binding role for LTBP-1 in the arterial response to injury and a structural role for LTBP-1 and LTBP-2 as microfibrillar components within the developing neointimal lesion. The increased proteolysis of LTBP-1 following angioplasty may represent an important regulatory step in TGF-β activation in the injured vessel. Studies aimed at further characterising and then inhibiting LTBP-1 proteolytic release from extracellular matrix post-angioplasty are now warranted and may lead to novel therapeutic strategies for controlling the TGF-β driven injury response.

Time for primary review 38 days.


   Acknowledgements
 
We are grateful to the Medical Research Council for supporting SS (Clinical Training Fellow) and CMK (Senior Research Fellow). We thank Dr Ian S.D. Roberts, Department of Cellular Pathology, John Radcliffe Hospital, Oxford for advice on immunohistochemical staining.


   Notes
 
1 Current address: University of Virginia, Molecular Physiology and Biological Physics, PO Box 800736, Charlottesville, VA 22908-0736, USA. Back


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. Sinha S., Nevett C., Shuttleworth C.A., Kielty C.M. Cellular and extracellular biology of the latent transforming growth factor-beta binding proteins. Matrix Biol. (1998) 17:529–545.[CrossRef][Web of Science][Medline]
  2. Miyazono K., Hellman U., Wernstedt C., Heldin C.H. Latent high molecular weight complex of transforming growth factor beta 1. Purification from human platelets and structural characterization. J. Biol. Chem. (1988) 263:6407–6415.[Abstract/Free Full Text]
  3. Wakefield L.M., Smith D.M., Flanders K.C., Sporn M.B. Latent transforming growth factor-beta from human platelets. A high molecular weight complex containing precursor sequences. J. Biol. Chem. (1988) 263:7646–7654.[Abstract/Free Full Text]
  4. Miyazono K., Olofsson A., Colosetti P., Heldin C.H. A role of the latent TGF-beta 1-binding protein in the assembly and secretion of TGF-beta 1. EMBO J. (1991) 10:1091–1101.[Web of Science][Medline]
  5. Miyazono K., Thyberg J., Heldin C.H. Retention of the transforming growth factor-beta 1 precursor in the Golgi complex in a latent endoglycosidase H-sensitive form. J. Biol. Chem. (1992) 267:5668–5675.[Abstract/Free Full Text]
  6. Kojima S., Nara K., Rifkin D.B. Requirement for transglutaminase in the activation of latent transforming growth factor-beta in bovine endothelial cells. J. Cell. Biol. (1993) 121:439–448.[Abstract/Free Full Text]
  7. Flaumenhaft R., Abe M., Sato Y., Miyazono K., Harpel J., Heldin C.H., Rifkin D.B. Role of the latent TGF-beta binding protein in the activation of latent TGF-beta by co-cultures of endothelial and smooth muscle cells. J. Cell. Biol. (1993) 120:995–1002.[Abstract/Free Full Text]
  8. Taipale J., Saharinen J., Hedman K., Keski-Oja J. Latent transforming growth factor-beta 1 and its binding protein are components of extracellular matrix microfibrils. J. Histochem. Cytochem. (1996) 44:875–889.[Abstract]
  9. Dallas S.L., Miyazono K., Skerry T.M., Mundy G.R., Bonewald L. Dual role for the latent transforming growth factor-beta binding protein in storage of latent TGF-beta in the extracellular matrix and as a structural matrix protein. J. Cell. Biol. (1995) 131:539–549.[Abstract/Free Full Text]
  10. Majesky M.W., Lindner V., Twardzik D.R., Schwartz S.M., Reidy M.A. Production of transforming growth factor beta 1 during repair of arterial injury. J. Clin. Invest. (1991) 88:904–910.[Web of Science][Medline]
  11. Nikol S., Isner J.M., Pickering J.G., Kearney M., Leclerc G., Weir L. Expression of transforming growth factor-beta 1 is increased in human vascular restenosis lesions. J. Clin. Invest. (1992) 90:1582–1592.[Web of Science][Medline]
  12. Nabel E.G., Shum L., Pompili V.J., Yang Z.Y., San H., Shu H.B., Liptay S., Gold L., Gordon D., Derynck R. Direct transfer of transforming growth factor beta 1 gene into arteries stimulates fibrocellular hyperplasia. Proc. Natl. Acad. Sci. USA (1993) 90:10759–10763.[Abstract/Free Full Text]
  13. Wolf Y.G., Rasmussen L.M., Ruoslahti E. Antibodies against transforming growth factor-beta 1 suppress intimal hyperplasia in a rat model. J. Clin. Invest. (1994) 93:1172–1178.[Web of Science][Medline]
  14. Karonen T., Jeskanen L., Keski-Oja J. Transforming growth factor beta 1 and its latent form binding protein-1 associate with elastic fibres in human dermis: accumulation in actinic damage and absence in anetoderma. Br. J. Dermatol. (1997) 137:51–58.[CrossRef][Web of Science][Medline]
  15. Raghunath M., Unsold C., Kubitscheck U., Bruckner-Tuderman L., Peters R., Meuli M. The cutaneous microfibrillar apparatus contains latent transforming growth factor-beta binding protein-1 (LTBP-1) and is a repository for latent TGF-beta1. J. Invest. Dermatol. (1998) 111:559–564.[CrossRef][Web of Science][Medline]
  16. Dallas S.L., Keene D.R., Bruder S.P., Saharinen J., Sakai L.Y., Mundy G.R., Bonewald L.F. Role of the latent transforming growth factor beta binding protein 1 in fibrillin-containing microfibrils in bone cells in vitro and in vivo. J. Bone Miner Res. (2000) 15:68–81.[CrossRef][Web of Science][Medline]
  17. Sterzel R.B., Hartner A., Schlotzer-Schrehardt U., Voit S., Hausknecht B., Doliana R., Colombatti A., Gibson M.A., Braghetta P., Bressan G.M. Elastic fiber proteins in the glomerular mesangium in vivo and in cell culture. Kidney Int. (2000) 58:1588–1602.[CrossRef][Web of Science][Medline]
  18. Davis E.C. Smooth muscle cell to elastic lamina connections in developing mouse aorta: role in aortic medial organization. Lab. Invest. (1993) 68:89–99.[Web of Science][Medline]
  19. Davis E. Immunolocalization of microfibril and microfibril-associated proteins in the subendothelial matrix of the developing mouse aorta. J. Cell. Sci. (1994) 107:727–736.[Abstract]
  20. Kanzaki T., Olofsson A., Moren A., Wernstedt C., Hellman U., Miyazono K., Claesson-Welsh L., Heldin C.H. TGF-beta 1 binding protein: a component of the large latent complex of TGF-beta 1 with multiple repeat sequences. Cell (1990) 61:1051–1061.[CrossRef][Web of Science][Medline]
  21. Bashir M.M., Han M.D., Abrams W.R., Tucker T., Ma R.I., Gibson M., Ritty T., Mecham R., Rosenbloom J. Analysis of the human gene encoding latent transforming growth factor- beta-binding protein-2. Int. J. Biochem. Cell Biol. (1996) 28:531–542.[CrossRef][Web of Science][Medline]
  22. Gibson M.A., Hatzinikolas G., Davis E.C., Baker E., Sutherland G.R., Mecham R.P. Bovine latent transforming growth factor beta 1-binding protein 2: molecular cloning, identification of tissue isoforms, and immunolocalization to elastin-associated microfibrils. Mol. Cell. Biol. (1995) 15:6932–6942.[Abstract]
  23. Abe M., Harpel J.G., Metz C.N., Nunes I., Loskutoff D.J., Rifkin D.B. An assay for transforming growth factor-beta using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct. Anal. Biochem. (1994) 216:276–284.[CrossRef][Web of Science][Medline]
  24. Taipale J., Miyazono K., Heldin C.H., Keski-Oja J. Latent transforming growth factor-beta 1 associates to fibroblast extracellular matrix via latent TGF-beta binding protein. J. Cell. Biol. (1994) 124:171–181.[Abstract/Free Full Text]
  25. Galis Z.S., Muszynski M., Sukhova G.K., Simon-Morrissey E., Unemori E.N., Lark M.W., Amento E., Libby P. Cytokine-stimulated human vascular smooth muscle cells synthesize a complement of enzymes required for extracellular matrix digestion. Circ. Res. (1994) 75:181–189.[Abstract/Free Full Text]
  26. Aoyagi M., Yamamoto M., Azuma H., Nagashima G., Niimi Y., Tamaki M., Hirakawa K., Yamamoto K. Immunolocalization of matrix metalloproteinases in rabbit carotid arteries after balloon denudation. Histochem. Cell. Biol. (1998) 109:97–102.[CrossRef][Web of Science][Medline]
  27. Tsuji T., Okada F., Yamaguchi K., Nakamura T. Molecular cloning of the large subunit of transforming growth factor type beta masking protein and expression of the mRNA in various rat tissues. Proc. Natl. Acad. Sci. USA (1990) 87:8835–8839.[Abstract/Free Full Text]
  28. Taketazu F., Kato M., Gobl A., Ichijo H., ten Dijke P., Itoh J., Kyogoku M., Ronnelid J., Miyazono K., Heldin C.H. Enhanced expression of transforming growth factor-beta s and transforming growth factor-beta type II receptor in the synovial tissues of patients with rheumatoid arthritis. Lab. Invest. (1994) 70:620–630.[Web of Science][Medline]
  29. Kanzaki T., Shiina R., Saito Y., Oohashi H., Morisaki N. Role of latent TGF-beta 1 binding protein in vascular remodeling. Biochem. Biophys. Res. Commun. (1998) 246:26–30.[CrossRef][Web of Science][Medline]
  30. Fang J., Li X., Smiley E., Francke U., Mecham R.P., Bonadio J. Mouse latent TGF-beta binding protein-2: molecular cloning and developmental expression. Biochim. Biophys. Acta (1997) 1354:219–230.[Medline]
  31. Nunes I., Gleizes P.E., Metz C.N., Rifkin D.B. Latent transforming growth factor-beta binding protein domains involved in activation and transglutaminase-dependent cross-linking of latent transforming growth factor-beta. J. Cell. Biol. (1997) 136:1151–1163.[Abstract/Free Full Text]
  32. Nunes I., Shapiro R.L., Rifkin D.B. Characterization of latent TGF-beta activation by murine peritoneal macrophages. J. Immunol. (1995) 155:1450–1459.[Abstract]
  33. Losordo D.W., Rosenfield K., Pieczek A., Baker K., Harding M., Isner J.M. How does angioplasty work? Serial analysis of human iliac arteries using intravascular ultrasound. Circulation (1992) 86:1845–1858.[Abstract/Free Full Text]
  34. Saharinen J., Keski-Oja J. Specific sequence motif of 8-Cys repeats of TGF-beta binding proteins LTBPs, creates a hydrophobic interaction surface for binding of small latent TGF-beta. Mol. Biol. Cell (2000) 11:2691–2704.[Abstract/Free Full Text]
  35. Nakajima Y., Miyazono K., Kato M., Takase M., Yamagishi T., Nakamura H. Extracellular fibrillar structure of latent TGF beta binding protein-1: role in TGF beta-dependent endothelial-mesenchymal transformation during endocardial cushion tissue formation in mouse embryonic heart. J. Cell. Biol. (1997) 136:193–204.[Abstract/Free Full Text]
  36. Bunton T.E., Biery N.J., Myers L., Gayraud B., Ramirez F., Dietz H.C. Phenotypic alteration of vascular smooth muscle cells precedes elastolysis in a mouse model of marfan syndrome. Circ. Res. (2001) 88:37–43.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Sinha, S.
Right arrow Articles by Kielty, C. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sinha, S.
Right arrow Articles by Kielty, C. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?