IOVS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

(Investigative Ophthalmology and Visual Science. 2006;47:4316-4322.)
© 2006 by The Association for Research in Vision and Ophthalmology, Inc.
DOI:  10.1167/iovs.05-1415
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hopkinson, A.
Right arrow Articles by Dua, H. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hopkinson, A.
Right arrow Articles by Dua, H. S.

Amniotic Membrane for Ocular Surface Reconstruction: Donor Variations and the Effect of Handling on TGF-ß Content

Andrew Hopkinson,1 Richard S. McIntosh,1 Patrick J. Tighe,1 David K. James,2 and Harminder S. Dua1

1From the Division of Ophthalmology and Visual Sciences and 2Foetomaternal Medicine, University Hospital Nottingham, Queen’s Medical Centre, Nottingham, United Kingdom.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
PURPOSE. Amniotic membrane (AM) transplantation is an accepted procedure in ocular surgery. However, little is known of the interdonor and intradonor variability within the membrane. In addition, the effects of the methods of processing, storage, and preoperative preparation on the membrane are not fully elucidated. The purpose of this study was to use TGF-ß as an example to investigate interdonor and intradonor variability and to determine the effect of "handling " on TGF-ß1 within fresh, processed and stored, and transplantation-ready AM (TRAM).

METHODS. Seventeen human AMs, both fresh and handled, were analyzed for TGF-ß1 by real-time polymerase chain reaction, immunohistochemistry, SDS-PAGE, and Western blotting.

RESULTS. TGF-ß1 was the highest normalized expressed isoform of TGF-ß in all samples, but it varied between membranes of different donors and at different sites within the same membrane. The highest concentration was noted in the spongy layer. Removal of the spongy layer successfully removed the bulk of TGF-ß1 from TRAM. Latency-associated protein (LAP) and a latent TGF-ß–binding protein (LTBP) were also detected.

CONCLUSION. TGF-ß1 is present in various regulatory forms in the AM. A degree of intermembrane and intramembrane variation is modified by handling. Unless a standardized protocol is adopted that delivers a membrane with consistent constituents, clinical outcomes may vary and comparisons may be invalid.

The amniotic membrane (AM) has acquired a unique place in ophthalmic surgery and is now used in many aspects of ocular surface reconstruction.1 Although not fully elucidated, a number of different mechanisms of action have been attributed to explain the beneficial effects of the membrane. Some of these are inferred from the composition of the membrane. By virtue of the abundance of growth factors contained within the membrane, it is believed that the membrane promotes epithelialization of denuded areas of the ocular surface.2 Another important role accorded to the membrane is to inhibit inflammation and scarring.3 Many of these constituents are likely to be affected by the methods used for handling the membrane. In this study the term handling was used to describe collectively all procedures carried out on the membrane before surgery. Handling thus included processing (procurement, separation of amnion from placenta and chorion, removal of spongy layer, and washing), storage or preservation, and preparation (thawing of stored membrane and washing immediately before transplantation, yielding a piece referred to as transplantation-ready amniotic membrane (TRAM). Generally, two different protocols for the processing and storing (preservation) of AM for clinical use have been established.4 5 Preoperative preparation of the membrane has not been standardized.

Despite its widespread use, the characteristics of the membrane that is procured, supplied, and applied to patients is far from standardized.6 The effect of handling on the membrane is unknown. In addition, little is known of the interdonor and intradonor variability within the membrane. Some of the variability in clinical outcomes that has been reported after the use of the membrane could well be related to such differences.1 6

One of the major reported properties attributed to AM is its ability to facilitate wound healing,7 8 9 which is associated with the action of TGF-ß. Three highly conserved isoforms of mammalian TGF-ß (TGF-ß1, TGF-ß2, and TGF-ß3)10 are encoded by distinct genes.11 TGF-ß is known to regulate the proliferation and differentiation of cells, inflammation, wound healing, scarring, angiogenesis, and extracellular matrix (ECM) remodeling in a variety of tissues and organs and during embryonic development. Almost all cells in the body produce TGF-ß and have receptors for it. Biologic activity of TGF-ß is regulated by a latency-associated protein (LAP) and a latent TGF-ß–binding protein (LTBP). Most secreted TGF-ß is latent and requires release from LAP and LTBP to effect its biologic action.

The purpose of this study was to assess variation among AM and to determine the effects of handing on AM. Because TGF-ß is a major modulator of wound healing implicated in the mechanism of action of the membrane and because of its abundance in soluble and insoluble forms, we selected this molecule as a candidate to determine the relative levels of TGF-ß within fresh and handled AM as an illustration of the differences between donors and the effect of handling.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Amniotic Membrane Procurement
Seventeen human AMs were collected from consenting patients undergoing elective cesarean section near term and delivering healthy infants. Patients with gestational problems were not included. Local ethics committee approval was obtained, and the study complied with the tenets of the Declaration of Helsinki. Membranes were preserved and handled according to a published methodology for AM transplantation (AMT).5 They were stored at –80°C for a minimum of 6 months to cover the window period of HIV infection before clinical use. Tissue has been stored for as long as 2 years before use. Sections of each fresh AM were also retained for protein and immunohistochemical analysis.

Preparation of AM
Membrane was prepared in accordance with a previously published procedure5 that was modified to allow for complete removal of the spongy layer (boundary layer between the fibroblast layer of the AM stroma and the cellular layer of the chorion12 ) and any contamination of it with donor blood (patent application submitted; briefly, the membrane was allowed to soak in balanced salt solution with gentle rocking for 30 minutes, which allowed the spongy layer to swell considerably and facilitated its removal as a single sheet). Cleaned membranes were washed before storage, as previously described.5 After washing, 1- or 2-inch–square pieces of AM were placed in 2 mL PBS with dimethyl sulfoxide (DMSO) and frozen at –80°C. Stored AM segments were thawed and thoroughly cleaned by three washes in 5 mL saline containing protease inhibitors (complete protease inhibitor tablets; Roche, Lewes, UK) for 10 minutes each wash. Storage medium and washes were concentrated and retained for protein analysis, as described previously.13

Isolation of RNA and cDNA Synthesis
Total RNA was isolated from AM samples (placental amnion [pa], apical amnion [aa], near apical [na], and unspecified reflectum region [am]), chorion segments, and placenta using a commercial kit (RNeasy kit and Qiashredder columns; Qiagen, Crawley, West Sussex, UK) according to the manufacturer’s protocol. Total RNA was quantitated spectrophotometrically (ND1000; Nanodrop Technologies, Wilmington, DE), and cDNA was synthesized from 2 µg RNA. First-strand cDNA was prepared (Ready-to-Go dT-primed cDNA kits; Amersham Biosciences, Chalfont St Giles, Bucks, UK), according to the manufacturer’s instructions.

Real-Time PCR
For selected primers that had previously been validated using conventional PCR (data not shown), real-time PCR analysis was performed to allow calculation of the relative abundance of TGF-ß1, TGF-ß2, and TGF-ß3 transcripts, in AM, chorion, and placenta of eight fresh fetal membranes, using previously published methods.14 Identical placenta and peripheral blood mononuclear cell (PBMC) samples were used as positive controls in all assays to compensate for interassay variation. Primers were: TGF-ß1 (product size, 196 bp) forward primer, 5'- CCAACTATTGCTTCAGCTCCAC-3'; TGF-ß1 reverse primer, 5'-TTATGCTGGTTGTACAGGGCC-3'; TGF-ß2 (product size, 233 bp) forward primer, 5'-CTGGAGCATGCCCGTATTTATG-3'; TGF-ß2 reverse primer, 5'-TTTGGTCTTGCCACTTTTCCAAG-3'; TGF-ß3 (product size, 211 bp) forward primer, 5'-CCAATTACTGCTTCCGCAACTT-3'; TGF-ß3 reverse primer, 5'-GCAGATGCTTCAGGGTTCAGA-3'.

One-Dimensional SDS-PAGE
Proteins were extracted from samples, as described previously.13 One-dimensional SDS-PAGE was performed by separating solubilized protein, under denaturing and reducing conditions, with 12-well Bis-Tris gels (NuPAGE Novex 4% to 12%; Invitrogen, Paisley, UK), according to manufacturer’s protocol. Protein visualization was performed using Coomassie blue staining (Simply Blue safe stain; Invitrogen).

Western Blot
Western blotting of gels was carried out according previously published methodologies. Polyvinylidene difluoride (PVDF) membranes were blocked for 1 hour at room temperature using Tris-buffered saline, pH 8.0 (Sigma, Poole, UK), 0.05% (vol/vol) Tween 20 (Promega, Southampton, UK), and 1% (wt/vol) nonfat milk powder (TBSTM), followed by immunodetection of human TGF-ß1 using monoclonal mouse anti–human TGF-ß1 (MCA797, clone TB21; Serotec, Oxford, UK) primary antibody (titrated concentration of 1:20); human epithelial growth factor (EGF) using monoclonal mouse anti–human EGF (MAB236, clone 10825; R&D Systems, Oxon, UK); or human hepatocyte growth factor (HGF) using affinity-purified polyclonal goat anti–human HGF (AF-294-NA; R&D Systems). Primary antibody was detected using alkaline-phosphatase–conjugated goat anti–mouse IgG (H+L) or rabbit anti–goat IgG (H+L), preadsorbed to bovine, horse, and human antibodies (Pierce, Cheshire, UK). Blots were developed with premixed alkaline phosphatase chromogen kit (BCIP/NBT; Sigma). TGF-ß1 purified from platelets and recombinant human (rh) EGF (both R&D Systems) were used as positive controls for antibody reactivity.

Immunohistochemistry
Six-micrometer sections were prepared from optimal cutting temperature (OCT; Raymond Lamb Ltd., East Sussex, UK)–embedded fresh AM and TRAM and were stained using a three-step alkaline phosphatase–anti-alkaline phosphatase (APAAP) protocol. Primary antibodies used were mouse monoclonal anti–human TGF-ß1 (clone TB21; Serotec); anti–human LAP (TGF-ß1) antibody (clone 27235.1; R&D Systems); and anti–human latent TGF-ß binding protein 1 antibody (LTBP-1, clone 35409; R&D Systems). A titration assay established the optimum antibody dilutions of 1:20 for TGF-ß1 and 1:100 for LAP and LTBP-1 when APAAP methods were used. Mouse primary antibodies were detected with rabbit anti–mouse antibody (Z0259; DAKO, Bucks, UK) at a dilution of 1:40, followed by alkaline phosphatase and mouse monoclonal APAAP (DAKO) according to the manufacturer’s recommendations. Antibody binding was detected with Fast Red (Sigma). Sections of OCT-embedded placenta were used as positive controls. For each staining run and each antibody, appropriate positive controls and negative controls (in which nonimmune immunoglobulin was substituted for the primary antibody) were performed to ensure quality control.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Real-Time qPCR Analysis
After normalization, TGF-ß1 was the highest expressed isoform in fresh AM (Fig. 1) , typically double that of TGF-ß2 and TGF-ß3. Some variation in TGF-ß1 expression between different regions within a membrane was observed. These differences were not consistent among different donors. TGF-ß2 was typically the lowest expressed isoform in AM; however, expression was increased in placental AM regions and in chorion samples. Normalized expression levels suggest TGF-ß1 and TGF-ß2 as lower in AM than in chorion (Fig. 2) . However, TGF-ß3 expression in AM and chorion was similar (Fig. 2) .


Figure 1
View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 1. TGF-ß isoform expression in various areas of the fetal membrane. Expression of TGF-ß isoforms was assessed using real-time PCR on cDNA derived from chorion (Ch) and amniotic membrane at the apical amniotic membrane site (AA), near apical site (NA), placental amnion site (PA), and amniotic membrane reflectum region (AM). Sample values are adjusted to compensate for interassay variation using identical positive control samples.

 

Figure 2
View larger version (7K):
[in this window]
[in a new window]
  		FIGURE 2. Normalized relative transcript abundance of TGF-ß isoform expression in AM and chorion. Expression of TGF-ß isoforms -ß1, -ß2, and -ß3 was assessed using real-time PCR on cDNA derived from chorion and amniotic membrane. Data were normalized against corresponding relative transcript abundance of TGF-ß isoform expression in PBMC/placenta. Horizontal line: mean expression level for positively amplified samples for each sample type.

 
Immunodetection of TGFß1 with Western Blot
Western blot analysis of protein extracted from fresh AM, TRAM, and spongy layer removed during processing confirmed the presence of TGF-ß1 protein in AM (Figs. 3A 3B) . TGF-ß1 protein was detected in all fresh AM samples (Fig. 3B , lanes 1–4), however, at varying intensities between membranes. In contrast, TGF-ß1 was detected in only one of the corresponding TRAM samples, suggesting TGF-ß1 elution from AM occurred during handling. The degree of elution also varied between membranes. The most intense staining was observed in the spongy layer removed by our modified handling technique, indicating a more thorough removal of the spongy layer and of the TGF-ß1 contained within.


Figure 3
View larger version (136K):
[in this window]
[in a new window]
  		FIGURE 3. Immunodetection of TGF-ß1 protein expression in AM 35 µg protein extracted from fresh AM (lanes 1–4), processed AM (lanes 5–8), and spongy layer (lanes 9–11) were separated on denaturing PAGE mini-gels under reducing conditions, Coomassie stained (A) or Western blotted to PVDF, and detected with anti–TGF-ß1 antibody (B). Four AMs were used, respectively, in lanes 1, 5, and 9, lanes 2, 6, and 10, lanes 3, 7, and 11, and lanes 4 and 8. Results of 1 of 5 representative experiments are shown.

 
Immunodetection of TGF-ß1 had demonstrated that the most intense staining was present in samples of spongy layer removed according to the modified handling technique and chorion. Staining in these samples varied considerably between membranes (Figs. 4A 4B) . On Western blot, at a protein load of 20 µg per lane, no TGF-ß1 was detectable in extracts obtained from membranes processed by the modified technique but were easily detected in extracts of membranes obtained by the conventional technique (Figs. 4A 4B) . This indicated a significant reduction in TGF-ß1 levels in samples handled according to the modified technique as opposed to the conventional technique. Immunodetection of HGF and EGF was also carried out as comparative controls to assess effects of handling on other factors reported in AM. Staining for HGF was variable between samples; HGF was not detected in all AM or derived samples (Fig. 4D) , whereas staining patterns for HGF in positive membranes was similar to that of TGF-ß1 (Fig. 4C) . EGF was not detected in any fresh or handled AM or chorion-derived samples (data not shown). However, staining for recombinant rhEGF control was detected as low as 2.5 ng per lane (equivalent to 2.5 ng in 20,000-ng protein load per lane).


Figure 4
View larger version (26K):
[in this window]
[in a new window]
  		FIGURE 4. Levels of TGF-ß1 (A, B) and HGF (C, D) protein expression in AM, removed during handling. Total protein was assayed, and equal amounts of 20 µg were loaded in each lane. Proteins were separated on denaturing PAGE mini-gels under reducing conditions, Western blotted to PVDF, and detected with anti–TGF-ß1 and anti–HGF antibodies. Spongy layer (lane 1), chorion (lane 2), fresh AM (lane 3), processed AM (lane 4), 10x storage medium (lane 5), 10x wash 1 (lane 6), 10x wash 2 (lane 7), and 10x wash 3 (lane 8). Results of 1 of 15 representative experiments (A, B) are shown.

 
No TGF-ß1 was detected in the concentrated washes retained during handling of fresh membranes before preservation/storage (data not shown). However, intense staining was detected in the storage medium of these membranes (for example, Fig. 4 , lane 5), suggesting release of TGF-ß1 during preservation. Subsequent sequential washing of preserved membranes resulted in the further elution of TGF-ß1, which decreased with each sequential wash, until staining was no longer detected in the washes (Fig. 4) . This typically occurred within three 10-mL saline washes of 5 minutes per wash.

Variation in TGF-ß1 content of fresh AM determined the relative amount released during handling and the amount of processing required until TGF-ß1 was no longer detectable (Figs. 4A 4B) . Nevertheless, with sufficient washing, TGF-ß1 was typically removed below a detectable level (with a 20-µg protein load; Fig. 4 ). However, TGF-ß1 remained detectable in subsequent washes in 3 of 17 AM samples (Fig. 5 , lane 7). In these cases, elution appeared much slower with consistent staining intensities in the storage medium and each sequential wash (Fig. 5 , lanes 4–7); TGF-ß1 was often detected even in the fifth wash (data not shown).


Figure 5
View larger version (48K):
[in this window]
[in a new window]
  		FIGURE 5. Levels of TGF-ß1 protein expression in AM not removed during processing. Total protein was assayed and equal amounts of 20 µg were loaded in each lane. Proteins were separated on denaturing PAGE mini-gels under reducing conditions, Western blotted to nitrocellulose, and reacted with anti–TGF-ß1 antibody. Spongy layer (lane 1), fresh AM (lane 2), processed AM (lane 3), 10x storage medium (lane 4), 10x wash 1 (lane 5), 10x wash 2 (lane 6), and 10x wash 3 (lane 7). Results of 1 of 7 representative experiments are shown.

 
Effect of Protease Inhibitors on TGF-ß1 Release
AM segments from the same amnion were prepared as if for surgery with and without protease inhibitors (Fig. 6) . When protease inhibitors were included in the wash medium, TGF-ß1 release was reduced and was not typically detectable in the eluent after the third wash. Conversely, increased staining was observed in the corresponding TRAM. In contrast, omitting protease inhibitors resulted in increased TGF-ß1 elution (Fig. 6) , which continued beyond the standard number of washes used during clinical preparation.


Figure 6
View larger version (27K):
[in this window]
[in a new window]
  		FIGURE 6. Effects of protease inhibitors. TGF-ß1 protein eluted from AM during processing after preservation. Total protein was assayed, and equal amounts of 20 µg were loaded in each lane. Proteins were separated on denaturing PAGE mini-gels under reducing conditions, Western blotted to nitrocellulose, and reacted with anti–TGF-ß1 antibody. Corresponding membrane segments were processed with protease inhibitors present (lanes 1–4) and absent (lanes 5–10) from the wash medium. 10x storage medium (lanes 1, 5), 10x wash 1 (lanes 2, 6), 10x wash 2 (lanes 3, 7), 10x wash 3 (lanes 4, 8), 10x wash 4 (lane 9), and 10x wash 5 (lane 10). Results of 1 of 4 representative experiments are shown.

 
Immunohistochemical Analysis of Amniotic Membrane
The partial release of TGF-ß1 during processing suggested at least two types of TGF-ß1 exist in AM. To investigate this further, in situ localization of TGF-ß1 and their associated proteins and the relative effects of handling were examined. Immunohistochemical analysis for TGF-ß1, ß1-LAP, and LTBP1 expression in eight AMs before and after handling was carried out. Control staining with nonspecific IgG was negative (data not shown).

Two distinct types of AM, different in gross morphology, were generally observed. The first was supported by a "thin" ECM sparsely populated by fibroblasts, which were preferentially concentrated along the amnion/spongy layer interface. The ECM of the second type was considerably thicker ("thick") and was populated by fibroblasts throughout its thickness. Five of 17 AM samples tested had thick ECM.

Staining for TGF-ß1
Staining for TGF-ß1 throughout AM varied depending on gross membrane morphology. Typically, fresh, thick ECM membranes stained for TGF-ß1 throughout the entire membrane, including the bone marrow (BM) and amniotic epithelial cells (AECs) (Fig. 7A) . Other distinct staining patterns were also observed in the ECM as a line along the BM of the AECs, around fibroblasts, and of varying intensity in the spongy layer. In fresh, thin ECM membranes, less general staining for TGF-ß1 occurred; however, staining was localized to the spongy layer and in the ECM around fibroblasts and BM (Fig. 7B) .


Figure 7
View larger version (76K):
[in this window]
[in a new window]
  		FIGURE 7. Immunohistochemistry for TGF-ß1 (A–D), ß1-LAP (E–H), and LTBP-1 (I–L) in amniotic membrane obtained fresh (A, B, E, F, I, J) and corresponding preserved processed membrane (C, D, G, H, K, L), from two different membranes, (A, C, E, G, I, K), and (B, D, F, H, J, L). Original magnification, x400. Positive staining is shown as red. Results of 1 of 7 representative experiments are shown.

 
Handling the membrane according to our modified technique reduced nonlocalized general staining, particularly in the thick membranes. In addition, the intensely stained spongy layer was also consistently removed. Localized staining for TGF-ß1 after handling was similar for all membranes, particularly in the ECM around fibroblasts and in the BM (Fig. 7D) . After handling, localization of staining was in the pattern described, but the stain intensity varied among all membranes (Figs. 7A 7B 7C 7D) . Punctate staining for TGF-ß1 was observed in the AECs of two fresh AMs, and this staining was not reduced by handling (data not shown).

Staining for LAP
Staining for LAP was similar across all membranes, irrespective of thickness (Figs. 7E 7F 7G 7H) . Staining in fresh AM was faint and colocalized in the AECs with TGF-ß1 as general cytoplasmic and punctate staining (Fig. 7E) . However, intense punctate staining was observed in 2 of 17 AMs assessed (Fig. 7F) .

Staining for LTBP
LTBP-1 is the binding protein for TGF-ß1. Staining for LTBP-1 was intense in all fresh samples (Fig. 7I 7J 7K 7L) and was specifically localized to the BM of AECs, to distinct regions in ECM surrounding fibroblasts, and to fibers interspersed in the ECM between fibroblasts (particularly in thick membranes; Fig. 7K ). In addition, staining for LTBP-1 was also intense at the interface with the spongy layer. In four cases, this staining pattern extended into the spongy layer itself. Typically, in any one membrane, staining for LTBP-1 was colocalized with that for TGF-ß1 and was not reduced by handling (Figs. 7D 7L) , suggesting that LTBP-1, and by inference at least some TGF-ß1, was ECM bound. In fresh, thick ECM membranes in which staining for TGF-ß1 was present throughout the entire membrane, handling resulted in elution of all TGF-ß1 except around fibroblasts. In these membranes, LTBP staining was still present in the BM, but colocalization could not be seen because of the elution of TGF-ß1 (Figs. 7C 7K) .


   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
This study confirms the expression of TGF-ß isoforms in AM at the gene level.2 The results show that at the protein and the gene levels, TGF-ß1 is the highest expressed isoform and that protein expression is lower in AM than in chorion. In addition, the results indicate variation in TGF-ß1 gene expression among different membranes (donors). Western blot analysis and immunohistochemical examination support these findings, particularly when taking account of differences in membrane thickness. Clinically, such variation between membranes is not differentiated before surgery; therefore, their effect on clinical efficacy is unknown.

TGF-ß1 protein was detected throughout the fetal membrane. An important observation was that maximal presence of TGF-ß1 occurred in the acellular spongy layer, suggesting that the spongy layer may act as a depot for chorion-derived TGF-ß1 and other factors such as HGF. Proteoglycans such as lumican,15 decorin,16 and mimecan (Hopkinson A, unpublished observations, 2005) have been demonstrated in the membrane, but their regional distribution is not fully characterized. These proteoglycans could determine the distribution of the bound form of TGF-ß1.

TGF-ß isoforms are expressed in a tissue-specific and a developmentally regulated fashion. The TGF-ß1 isoform is a key mediator involved in the scarring process and is typically secreted as a biologically inactive precursor cytokine in small (LAP-TGF-ß1) or large (LAP-TGFß1-LTBP) latent (L-TGF-ß) complexes.11 17 18 L-TGF-ß1 consists of a biologically active mature TGF-ß dimmer,11 19 complexed to TGF-ß LAP.20 LAP confers latency in such a way that dissociation or extensive modification of the structural conformation is required for TGF-ß1 to elicit its biologic activity.21 Latency is critical in the regulation of TGF-ß1 activity because increased TGF-ß1 expression does not necessarily correlate with increased TGF-ß1 activity.22 Most small complexes are not secreted but are retained in the Golgi apparatus23 and the cytoplasm.24

The large L-TGF-ß complex, which comprises the small L-TGF-ß1 (LAP-TGF-ß) complex and the LTBP,20 can also be secreted.11 LTBP-1 cannot interact with active free TGF-ß1 directly but specifically binds the LAP-TGF-ß1 complex through the LAP. LTBP-1 plays a central role in the regulation and secretion of TGF-ß1 as a large L-TGF-ß1 complex.23 25 The principal function of LTBP-1 is to bind small L-TGF-ß1, covalently targeting it to the ECM.26 27 28 ECM-associated LTBP-1 serves to regulate TGF-ß1 activity, localizing TGF-ß1 in a concentrated "ready-to-go" depot for activation.11 The presence of these molecules in the AM supports the notion that TGF-ß1 in AM is present in a potentially active form and can influence the wound healing response.

Handling compromises cellular viability,29 facilitating the elution of soluble cellular proteins.13 Therefore, reduced immunoreactivity for TGF-ß1 in TRAM suggests that cytoplasmic L-TGF-ß124 and HGF are liberated. Similarly, reduced staining throughout the ECM and spongy layer suggests soluble TGF-ß1 was also present in these structures. This may explain why TGF-ß1 elution occurs initially in the storage medium on thawing and then in a sequential manner until soluble TGF-ß1 is no longer detected in the washes. However, the amount and time required for soluble TGF-ß1 elution to occur varied unpredictably, suggesting a source of TGF-ß1 in addition to AECs. This may explain why HGF is also eluted after processing below a detectable level in TRAM. At this point, the preoperative preparation of the membrane was standardized for all membranes to produce comparative data. In doing so, washing was considerably more extensive than clinical procedures currently used. Despite this, TGF-ß1 elution and HGF varied, with TGF-ß1 often continuing beyond 20 minutes of extensive washing with agitation.

Conventional clinical preparation of AM does not specifically deal with the spongy layer other than to remove any parts of it that are stained with donor blood, resulting in partial removal of the spongy layer. In this study, we demonstrated that ineffective removal of spongy layer, as described, during handling prolonged the release of soluble TGF-ß1 from AM. Furthermore, even after washing, the membrane retained considerable amounts of TGF-ß1 when remnants of spongy layer were retained. Hence, any residual spongy layer could also act as a TGF-ß1 reservoir, resulting in increased amounts transplanted to the eye.

A new protocol to standardize removal of the spongy layer and the amount of TGF-ß1 left in TRAM has been developed (see Materials and Methods) that essentially eliminates most spongy layer–related detectable TGF-ß1 in AM. That substantial amounts of TGF-ß1 are contained in the spongy layer and inconsistent removal of the spongy layer can result in significant variation between membranes with regard to their TGF-ß1 content is amply supported by the evidence presented. Nevertheless, despite complete removal of the spongy layer and though HGF was eluted below a detectable level, variation in TGF-ß1 could still be detected in TRAM, emphasizing that variable amounts of this growth factor remained bound to ECM or was retained intracellularly (detected as intracellular punctate staining). The punctate staining of TGF-ß1 in TRAM is indicative of the presence of AEC residue containing Golgi-associated TGF-ß1 on such membranes.23 In contrast to the conclusions of Koizumi et al.,2 who reported that TGF-ß1 levels were similar in AM with and without AECs, that EGF levels were similar to TGF-ß1 levels in AM with cells but were dramatically reduced after handling, and that HGF levels were more than 40 times greater than TGF-ß1 and EGF levels in AM with cells but were dramatically reduced after epithelial removal, our observations indicated that TGF-ß1 existed in at least two regulatory states in AM, that HGF was detected at much lower levels than TGF-ß1 but was eluted during handling, and that EGF was not present at detectable levels in AM. Our results also highlighted the variability of AM among donors.

The observation that LTBP-1 is localized in fibers laterally throughout the ECM of AM agrees with reports that, in vivo, LTBP-1 plays a structural role in the ECM.30 However, a recent report showing reduced expression of LTBP-1 in the fetus, where TGF-ß1 is not expressed,28 also supports a function in AM related more to TGF-ß sequestration than to a structural role. Colocalization of LTBP-1 with TGF-ß1 in fresh AM and, more important, in TRAM, as demonstrated in this study, indicated that TGF-ß1 was present in inactive and ECM (i.e., LTBP-1)–bound forms. Colocalization of TGF-ß1 and LTBP-1 in the ECM in the basal aspect of AECs and most surrounding (fibroblast) cells suggested that AM cells secreted LTBP-1 as part of the large L-TGF-ß1 complex or as LTBP-1 alone, both of which are rapidly sequestered to the ECM immediately after secretion.25 27 28 31 LTBP-1 could then serve to "mop" up any free small L-TGF-ß1 as an inactive reservoir.

Staining of ECM-localized TGF-ß1, particularly at the BM, varied between membranes. Some membranes showed the pattern described, but most membranes showed colocalization of TGF-ß1 and LTBP-1 at the BM. LTBP-1 apparently colocalizes more with the inactive form of TGF-ß1. Activation of large matrix-associated L-TGF-ß1 requires release from the ECM by proteolysis.11 17 18 20 32 33 34 35 Reduced elution of TGF-ß1 in the presence of protease inhibitors suggested that proteases may be involved in the activation and release of TGF-ß1 from the ECM of AM during handling. Release and elution of TGF-ß1 during handling would explain why such colocalization could not be demonstrated in all membranes, as illustrated in Figures 7C and 7K . In addition, active proteolysis would be expected to have an effect on the membrane after transplantation, potentially releasing any residual stored TGF-ß1 close to the ocular surface.

In summary, using the important molecule TGF-ß1 as a prototype and HGF as a comparable control, we have demonstrated that handling procedures can substantially alter the nature and possibly the efficacy of the TRAM. Attempts to standardize the membrane would have to include aspects of handling and take into consideration interdonor variations. This study highlights the fact that other key factors critical in achieving the desired clinical effects may be inadvertently lost or retained, depending on the exact nature of the handling procedures used, necessitating further study into the effects of handling on AM proteins.


   Footnotes
 
Supported by the Royal Blind Asylum and School/Scottish National Institution for the War Blinded, and Royal College of Surgeons Edinburgh, UK, and in part by the Eye Research Institute, Wills Eye Hospital, Philadelphia, Pennsylvania.

Submitted for publication November 2, 2005; revised January 20 and April 23, 2006; accepted July 26, 2006.

Disclosure: A. Hopkinson, P; R.S. McIntosh, P; P.J. Tighe, P; D.K. James, P; H.S. Dua, P

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Corresponding author: Andrew Hopkinson, Division of Ophthalmology and Visual Sciences, B Floor, Eye and E.N.T. Centre, University Hospital Nottingham, Queen’s Medical Centre, NG7 2UH Nottingham, UK; andy.hopkinson{at}nottingham.ac.uk.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Dua HS, Gomes JA, King AJ, Maharajan VS. The amniotic membrane in ophthalmology. Surv Ophthalmol. 2004;49:51–77.[CrossRef][ISI][Medline][Order article via Infotrieve]
  2. Koizumi NJ, Inatomi TJ, Sotozono CJ, Fullwood NJ, Quantock AJ, Kinoshita S. Growth factor mRNA and protein in preserved human amniotic membrane. Curr Eye Res. 2000;20:173–177.[CrossRef][ISI][Medline][Order article via Infotrieve]
  3. Tseng SC, Li DQ, Ma X. Suppression of transforming growth factor-beta isoforms, TGF-beta receptor type II, and myofibroblast differentiation in cultured human corneal and limbal fibroblasts by amniotic membrane matrix. J Cell Physiol. 1999;179:325–335.[CrossRef][ISI][Medline][Order article via Infotrieve]
  4. Tseng SC, Prabhasawat P, Barton K, Gray T, Meller D. Amniotic membrane transplantation with or without limbal allografts for corneal surface reconstruction in patients with limbal stem cell deficiency. Arch Ophthalmol. 1998;116:431–441.[Abstract/Free Full Text]
  5. Tsubota K, Shimazaki J. Surgical treatment of children blinded by Stevens-Johnson syndrome. Am J Ophthalmol. 1999;128:573–581.[CrossRef][ISI][Medline][Order article via Infotrieve]
  6. Dua HS, Maharajan VS, Hopkinson A. Controversies and limitations of amniotic membrane in ophthalmic surgery. Reinhard T Larkin F eds. Corneal and External Eye Disease (Essentials in Ophthalmology). 2005;21–33. Springer Berlin.
  7. Border WA, Noble NA. Transforming growth factor beta in tissue fibrosis. N Engl J Med. 1994;331:1286–1292.[Free Full Text]
  8. Branton MH, Kopp JB. TGF-beta and fibrosis. Microbes Infect. 1999;1:1349–1365.[CrossRef][ISI][Medline][Order article via Infotrieve]
  9. Mangasser-Stephan K, Gressner AM. Molecular and functional aspects of latent transforming growth factor-beta binding protein: just a masking protein?. Cell Tissue Res. 1999;297:363–370.[CrossRef][ISI][Medline][Order article via Infotrieve]
  10. Lee SB, Li DQ, Tan DT, Meller DC, Tseng SC. Suppression of TGF-beta signaling in both normal conjunctival fibroblasts and pterygial body fibroblasts by amniotic membrane. Curr Eye Res. 2000;20:325–334.[CrossRef][ISI][Medline][Order article via Infotrieve]
  11. Khalil N. TGF-beta: from latent to active. Microbes Infect. 1999;1:1255–1263.[CrossRef][ISI][Medline][Order article via Infotrieve]
  12. Bourne GL. The microscopic anatomy of the human amnion and chorion. Am J Obstet Gynecol. 1960;79:1070–1073.[ISI][Medline][Order article via Infotrieve]
  13. Hopkinson A, McIntosh RS, Layfield R, Keyte J, Dua HS, Tighe PJ. Optimised two-dimensional electrophoresis procedures for the protein characterisation of structural tissues. Proteomics. 2005;5:1967–1979.[CrossRef][ISI][Medline][Order article via Infotrieve]
  14. McIntosh RS, Cade JE, Al-Abed M, et al. The spectrum of antimicrobial peptide expression at the ocular surface. Invest Ophthalmol Vis Sci. 2005;46:1379–1385.[Abstract/Free Full Text]
  15. Yeh LK, Chen WL, Li W, et al. Soluble lumican glycoprotein purified from human amniotic membrane promotes corneal epithelial wound healing. Invest Ophthalmol Vis Sci. 2005;46:479–486.[Abstract/Free Full Text]
  16. Wu WX, Zhang Q, Unno N, Derks JB, Nathanielsz PW. Characterization of decorin mRNA in pregnant intrauterine tissues of the ewe and regulation by steroids. Am J Physiol Cell Physiol. 2000;278:C199–C206.[Abstract/Free Full Text]
  17. Gleizes PE, Munger JS, Nunes I, et al. TGF-beta latency: biological significance and mechanisms of activation. Stem Cells. 1997;15:190–197.[Abstract/Free Full Text]
  18. Koli K, Saharinen J, Hyytiainen M, Penttinen C, Keski-Oja J. Latency, activation, and binding proteins of TGF-beta. Microsc Res Technique. 2001;52:354–362.[CrossRef][ISI][Medline][Order article via Infotrieve]
  19. Cheifetz S, Bassols A, Stanley K, Ohta M, Greenberger J, Massague J. Heterodimeric transforming growth factor beta: biological properties and interaction with three types of cell surface receptors. J Biol Chem. 1988;263:10783–10789.[Abstract/Free Full Text]
  20. Saharinen J, Hyytiainen M, Taipale J, Keski-Oja J. Latent transforming growth factor-beta binding proteins (LTBPs)—structural extracellular matrix proteins for targeting TGF- beta action. Cytokine Growth Factor Rev. 1999;10:99–117.[CrossRef][ISI][Medline][Order article via Infotrieve]
  21. McMahon GA, Dignam JD, Gentry LE. Structural characterization of the latent complex between transforming growth factor beta 1 and beta 1-latency-associated peptide. Biochem J. 1996;313(pt 1)343–351.[Medline][Order article via Infotrieve]
  22. Sporn MB. TGF-beta: 20 years and counting. Microbes Infect. 1999;1:1251–1253.[CrossRef][ISI][Medline][Order article via Infotrieve]
  23. Miyazono K, Thyberg J, Heldin CH. 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]
  24. Mizoi T, Ohtani H, Miyazono K, Miyazawa M, Matsuno S, Nagura H. Immunoelectron microscopic localization of transforming growth factor beta 1 and latent transforming growth factor beta 1 binding protein in human gastrointestinal carcinomas: qualitative difference between cancer cells and stromal cells. Cancer Res. 1993;53:183–190.[Abstract/Free Full Text]
  25. Miyazono K, Olofsson A, Colosetti P, Heldin CH. 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.[ISI][Medline][Order article via Infotrieve]
  26. Olofsson A, Miyazono K, Kanzaki T, Colosetti P, Engstrom U, Heldin CH. Transforming growth factor-beta 1, -beta 2, and -beta 3 secreted by a human glioblastoma cell line: identification of small and different forms of large latent complexes. J Biol Chem. 1992;267:19482–19488.[Abstract/Free Full Text]
  27. Taipale J, Miyazono K, Heldin CH, 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(1–2)171–181.[Abstract/Free Full Text]
  28. 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]
  29. Corton M, Villuendas G, Botella JI, San Millan JL, Escobar-Morreale HF, Peral B. Improved resolution of the human adipose tissue proteome at alkaline and wide range pH by the addition of hydroxyethyl disulfide. Proteomics. 2004;4:438–441.[CrossRef][ISI][Medline][Order article via Infotrieve]
  30. Isogai Z, Ono RN, Ushiro S, et al. Latent transforming growth factor beta-binding protein 1 interacts with fibrillin and is a microfibril-associated protein. J Biol Chem. 2003;278:2750–2757.[Abstract/Free Full Text]
  31. Taipale J, Lohi J, Saarinen J, Kovanen PT, Keski-Oja J. Human mast cell chymase and leukocyte elastase release latent transforming growth factor-beta 1 from the extracellular matrix of cultured human epithelial and endothelial cells. J Biol Chem. 1995;270:4689–4696.[Abstract/Free Full Text]
  32. Abe M, Oda N, Sato Y. Cell-associated activation of latent transforming growth factor-beta by calpain. J Cell Physiol. 1998;174:186–193.[CrossRef][ISI][Medline][Order article via Infotrieve]
  33. Blakytny R, Ludlow A, Martin GE, et al. Latent TGF-beta1 activation by platelets. J Cell Physiol. 2004;199:67–76.[CrossRef][ISI][Medline][Order article via Infotrieve]
  34. Lyons RM, Gentry LE, Purchio AF, Moses HL. Mechanism of activation of latent recombinant transforming growth factor beta 1 by plasmin. J Cell Biol. 1990;110:1361–1367.[Abstract/Free Full Text]
  35. Lyons RM, Keski-Oja J, Moses HL. Proteolytic activation of latent transforming growth factor-beta from fibroblast-conditioned medium. J Cell Biol. 1988;106:1659–1665.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 IOVSHome page
A. Abedin, I. Mohammed, A. Hopkinson, and H. S. Dua
A Novel Antimicrobial Peptide on the Ocular Surface Shows Decreased Expression in Inflammation and Infection
Invest. Ophthalmol. Vis. Sci., January 1, 2008; 49(1): 28 - 33.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hopkinson, A.
Right arrow Articles by Dua, H. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hopkinson, A.
Right arrow Articles by Dua, H. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS