Skip Navigation


Human Molecular Genetics Advance Access originally published online on July 30, 2007
Human Molecular Genetics 2007 16(20):2423-2432; doi:10.1093/hmg/ddm199
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
16/20/2423    most recent
ddm199v1
Right arrow Alert me when this article is cited
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 ISI Web of Science
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
Google Scholar
Right arrow Articles by Marmorstein, L. Y.
Right arrow Articles by Marmorstein, A. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Marmorstein, L. Y.
Right arrow Articles by Marmorstein, A. D.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Formation and progression of sub-retinal pigment epithelium deposits in Efemp1 mutation knock-in mice: a model for the early pathogenic course of macular degeneration

Lihua Y. Marmorstein1,2,*, Precious J. McLaughlin1, Neal S. Peachey4,5,6, Takako Sasaki7 and Alan D. Marmorstein1,3

1 Department of Ophthalmology and Vision Science, 2 Department of Physiology, 3 Optical Sciences Center, University of Arizona, Tucson, AZ 85711, USA, 4 Research Service, Cleveland VA Medical Center, 5 Cole Eye Institute, Cleveland Clinic Foundation, 6 Department of Ophthalmology, Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, OH, USA and 7 Department of Biochemistry and Molecular Biology, Oregon Health and Science University, Portland, OR, USA

* To whom correspondence should be addressed at: Department of Ophthalmology and Vision Science, University of Arizona, 655 N Alvernon Way, Suite 108, Tucson, AZ 85711, USA. Tel: +1 5206260447; Fax: +1 5206260457; Email: Lmarmorstein{at}eyes.arizona.edu

Received June 14, 2007; Accepted July 17, 2007


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Malattia leventinese (ML) is a dominantly inherited macular degenerative disease characterized by the presence of sub-retinal pigment epithelium (RPE) deposits. With the exception of an earlier age of onset, ML patients exhibit symptoms and histopathology compatible with the diagnosis of age-related macular degeneration (AMD), the most common cause of incurable blindness. ML is caused by a mutation (R345W) in the gene EFEMP1 which encodes fibulin-3, a protein of unknown function. We generated a knock-in mouse carrying the disease-associated mutation in the murine Efemp1 gene. Small, isolated sub-RPE deposits developed as early as 4 months of age in both heterozygous and homozygous knock-in mice. Over time these deposits increased in size and number eventually becoming continuous sheets. In older mice membranous debris was observed within the deposits and within Bruch’s membrane, and was accompanied by general RPE and choroidal abnormalities including degeneration, vacuolation, loss or disruption of the RPE basal infoldings, choroidal atrophy, and focal thickening of and invasion of cellular processes into Bruch’s membrane. Fibulin-3 was found to accumulate in the sub-RPE deposits. Thus, the Efemp1 knock-in mice reconstitute the most important histopathologic symptoms of both ML and AMD. We conclude that these mice are a valuable tool for studying the primary pathogenic course of basal deposits associated with macular degeneration and for testing prevention and treatment strategies for this class of diseases.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Macular degeneration refers to a variety of conditions that are characterized by progressive loss of central vision associated with abnormalities of the retinal pigment epithelium (RPE) and Bruch’s membrane (1). These conditions include age-related macular degeneration (AMD) and a number of rarer, earlier-onset inherited macular disorders that have strong similarities to AMD (1). The inherited maculopathy most similar to AMD is malattia leventinese (ML), also known as Doyne’s honeycomb retinal dystrophy (2,3). ML is an autosomal dominantly inherited condition with full penetrance (2,3). It initially presents in early adulthood as small drusen in the macula and surrounding the optic nerve head. In some cases, these small drusen will form a radial pattern extending into the peripheral retina (2,3). Gradually, the drusen becomes bigger and more numerous, eventually merging to form a solid plaque. Histopathological studies of ML have established that the small radial drusen are continuous with or internal to the basement membrane of the RPE (4,5), and the large drusen are external to the basement membrane of the RPE and within Bruch’s membrane (6). Patients are usually asymptomatic until the age of 30–40 years. In the later stages at the age of 40–50 years, complications of geographic atrophy and choroidal neovascularization develop and result in central vision loss (2,3). There is no treatment currently available to halt the progression of the disease or to prevent vision loss.

ML is caused by a missense mutation (R345W) in the gene EFEMP1 (EGF-containing fibulin-like extracellular matrix protein 1, also know as FBLN3, FBNL, or S1-5) that encodes fibulin-3 (7). No other EFEMP1 mutations have been found in ML, or in any other retinal disorders. Fibulin-3 is one of six members of the fibulin family of extracellular matrix (ECM) proteins that are characterized by tandem arrays of calcium-binding (cb) EGF domains and a C-terminal fibulin-type module (8,9). All of the fibulins are expressed broadly throughout the body (9), but little is known about their functions. Fibulin-1 and 2 bind to various ECM ligands including basement membrane and elastic fiber components (9). Mice lacking fibulin-1 show defects in capillary endothelial cells (10). We have reported that targeted disruption of fibulin-4 (also known as EFEMP2 or MBP1) in mice abolishes elastogenesis and results in perinatal lethality due to severe vascular and lung defects, indicating an indispensable role of fibulin-4 in elastic fiber formation (11). Fibulin-5 (also known as DANCE or EVEC) deficient mice are also found to exhibit disrupted and disorganized elastic fibers (12,13). The biological importance of fibulins is highlighted by the findings that mutations of individual members have been associated with a growing list of human diseases. In addition to the mutation in EFEMP1 which causes ML, missense variations in fibulin-5 and 6 have been detected in AMD patients (14,15), mutations in fibulin-4 and fibulin-5 have been found in some cutis laxa patients (16–18), and fibulin-4 is a target of autoimmunity in osteoarthritis patients (19).

Of the six fibulins, fibulin-3 is the least studied. In normal tissues, fibulin-3 is highly expressed by epithelial and endothelial cells and is present in blood vessels of different sizes (20–23). It has been shown to inhibit angiogenesis both in vitro and in vivo (24). In the eye, fibulin-3 is expressed by RPE and cells in the outer and inner nuclear layers of the retina (7,25,26). The R345W mutation found in ML patients is located in the C-terminal most cbEGF domain (7). We have shown that mutant fibulin-3 containing the R345W mutation is misfolded and secreted less efficiently than the normal protein (26). In both ML and AMD donor eyes, fibulin-3 accumulates in basal deposits and between the RPE and drusen (26). In a two-hybrid screen, TIMP-3 (tissue inhibitor of metalloproteinase-3) was identified as a strong binding partner for fibulin-3 (27). Mutations in TIMP-3 cause Sorsby’s fundus dystrophy (SFD), another inherited macular degenerative disease with similarities to AMD (28). Like ML, an early feature of SFD is abnormal sub-RPE deposits (29). These data linking fibulin-3 to AMD, SFD and ML, provide the first connection between two different macular degeneration genes, and suggest that fibulin-3 is involved in a general pathogenic pathway leading to macular degeneration. However, the pathogenesis of AMD, ML and SFD are poorly understood. Because these diseases are difficult to study in humans due to the late age of onset, limited number of patients, and limited donor tissues, investigators have turned to animal models, primarily mice.

Though mice that generate basal deposits due to genetic manipulation, environmental manipulation, or both have been produced (30–33), none of the mouse models of a human inherited maculopathy exhibit large basal deposits and none of the genes which cause basal deposit or drusen formation in mice is known to cause macular degeneration in humans. Here we report on the phenotype of a knock-in mouse harboring the R345W mutation in the murine Efemp1 gene. We find that the Efemp1 knock-in mice closely recapitulate the histopathology observed in ML patients and in many individuals with AMD. These knock-in mice provide a simple model of a naturally occurring disease and offer unique opportunities to study the biochemical pathways underlying the development and progression of sub-RPE deposits and other RPE abnormalities involved in macular degeneration.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Generation of Efemp1 knock-in mice carrying the R345W mutation
In order to create a mouse model of ML, we chose to generate a knock-in mouse carrying the R345W mutation in the endogenous mouse Efemp1 loci (Fig. 1) rather than a transgenic mouse over-expressing mutant fibulin-3 from an exogenous promoter. This choice was made because in knock-in mice, the mutant alleles are targeted to replace rather than augment the endogenous wild-type alleles, heterozygotes carry one wild-type Efemp1 allele and one mutated allele harboring the R345W mutation, and homozygotes carry two mutated alleles without the presence of wild-type alleles.


Figure 1
View larger version (50K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Targeting strategy to generate Efemp1 knock-in mice containing the R345W mutation. (A) Schematic diagram of the wild-type locus, targeting vector and mutant loci before (‘+’) and after (mutant) the cre-mediated deletion of the floxed neor marker. The asterisk mark indicates the location of the R345W mutation. (B–D) Southern blot analysis of mouse tail DNA. Before cre-mediated deletion of the floxed neor marker, the 8.3-kb wild-type (+/+) and the 10.1-kb mutant (+/ki) (‘+’) fragments were detected by the 5' probe/AvrII + AhdI digestion. The 9.4-kb wild-type and the 7.5-kb mutant (‘+’) fragments were detected by the 3' probe/BamHI digestion (B). After cre-mediated deletion, both the wild-type and mutant bands are 8.3 kb detected by 5' probe/AvrII+AhdI digestion. In the case of BamHI digestion/3' probe hybridization, the mutant fragment (7.5 kb) remains to be different from the wild-type fragment (9.4 kb) due to the introduction of BamHI site within the loxP sequence (C and D). Control: DNA from a heterozygote prior to the marker deletion. (E) Nucleotide changes created the R345W mutation and eliminated the BsmI restriction site and PCR/BsmI analysis of tail genomic DNA. PCR with primers flanking the R345W mutation resulted in a 810-bp fragment. BsmI cut the PCR product amplified from the wild-type locus (+/+) to generate 430-bp and 380-bp fragments. The PCR product from the mutant locus (+/ki) was not cut by BsmI and remained to be 810 bp, confirming the incorporation of the mutation into the locus.

 
Consistent with the phenotype of ML patients, both heterozygous (Efemp1+/ki) and homozygous (Efemp1ki/ki) Efemp1 knock-in mice were born with an expected mendelian frequency, and showed no early postnatal lethality. Their gross appearance, lifespan and reproductivity are similar to those of wild-type littermates.

Sub-RPE deposits in Efemp1 knock-in mice
To determine whether Efemp1 knock-in mice exhibit abnormalities in the eye, we performed fundoscopy and histological analysis on mice aged 2–23 months. While no remarkable features were observed in the fundus, electron microscopy revealed the presence of sub-RPE deposits in both Efemp1+/ki and Efemp1ki/ki mice as young as 4 months (Fig. 2). At this age, the deposits are small (with a diameter of ~0.5 µm or less), isolated, and infrequent in occurrence (Fig. 2B and C). No deposits were found in wild-type littermates (Fig. 2A) at this age. The content of the deposits appeared to be homogenous and similar to basement membrane materials. For most part, the deposits were located between the plasma membrane and basement membrane of the RPE. Thus, they are basal laminar deposits (BLDs). Occasionally, deposits appeared within RPE cells, suggesting that the origin of the deposits is from the RPE rather than Bruch’s membrane. At 8 months of age, the deposits were larger, with a diameter up to 1 µm, and occurred more frequently than at 4 months. The deposits remained isolated and far apart from each other (Fig. 2E and F). By 12 months of age, the deposits became so numerous in both Efemp1+/ki and Efemp1ki/ki mice that they merged into continuous patches in many areas (Fig. 2H and I). The deposits tended to be heavier in areas under the central retina and the far periphery approaching the ciliary body. Overall, they covered >50% of the area under the retina at this age. The deposits continued to increase in size and frequency as the mice aged. At 18 months, continuous sheets of thick basal deposit (Fig. 2K and L) were observed covering >80% of the area under the retina. The deposits reached over 4 µm in height by 23 months of age (Fig. 2 N and O) when virtually the entire area under the retina was covered with deposits. Although, by this age, deposits were also found in the wild-type littermate controls (Fig. 2M), they were small and isolated, and more akin to those found in 4-month old Efemp1+/ki or Efemp1ki/ki mice. We did not find any deposits in wild-type littermates at 18 months of age or younger (Fig. 2A, D, G and J).


Figure 2
View larger version (171K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Development and progression of sub-RPE deposits in Efemp1 knock-in mice. Representative electron micrographs showing basal RPE and Bruch’s membrane of wild-type, Efemp1+/ki, and Efemp1ki/ki mice. The ages of the mice are indicated at the side panel. The genotypes of the mice are indicated across the top: +/+, wild-type; +/ki, Efemp1+/ki and ki/ki, Efemp1ki/ki. Arrows indicate isolate sub-RPE deposits. White dotted lines outline the inner borders of continuous sub-RPE deposits in older knock-in mice. Asterisks indicate membranous debris; BI, RPE basal infoldings and BrM, Bruch’s membrane. The magnification is the same for all panels. Scale bar: 500 nm.

 
At younger ages (12 months or younger), deposits appeared to be slightly larger in homozygous than heterozygous Efemp1 knock-in mice (Fig. 2H and I). However, this difference seemed to fade in older mice. Based on deposit size, we could not distinguish whether samples were from homozygous or heterozygous knock-in mice at 18 or 23 months of age. While there were variations in deposit sizes among individual mice and some homozygous mice had larger deposits than heterozygous mice, the converse was also true where deposits of some heterozygous mice were larger than in homozygous mice at the same age. In human patients, disease severity and deposit sizes also vary among individuals. Due to the rare occurrence of the disease, no statistics are available on whether patients carrying homozygous mutations have larger deposits than patients carrying only one mutation. Our observations in Efemp1 knock-in mice suggest that the phenotype associated with one mutant allele can be as severe as that with two mutant alleles.

Membranous debris in the sub-RPE deposits in Efemp1 knock-in mice
The sub-RPE deposits in Efemp1 knock-in mice were initially homogenous in appearance. However, by 12 months of age, electron microscopy revealed that membrane bound debris had begun to accumulate within the deposits (Fig. 2I). As the mice continued to age these membranous debris became very prominent and were interdigitated amongst the deposits in both Efemp1+/ki and Efemp1ki/ki mice (Fig. 2L, N and O; and Fig. 3A and B). Additionally, debris containing membrane coated vesicles appeared diffusely within the collagenous and elastic layers of Bruch’s membrane (Fig. 3C and Fig. 4C). Membranous debris accumulated in thin layers within Bruch’s membrane is referred to as basal linear deposit, and large quantities of debris built up focally in Bruch’s membrane form soft drusen (34). Although BLDs are regarded as the hallmark of macular degeneration in humans, membranous debris has been shown to correlate more closely with the severity of degeneration (34,35).


Figure 3
View larger version (169K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Membranous debris in the sub-RPE deposits of Efemp1 knock-in mice. (A) Electron micrograph showing extensive membranous debris interdigitating within the BLDs of a 23-month old Efemp1ki/ki mouse. The dotted line outlines the inner border of the deposits. (B) The higher magnification of box ‘b' area of (A). Arrow indicates membranous debris. (C) Debris containing membrane coated vesicles (arrow) within Bruch’s membrane between the basement membranes of the RPE and choroidal endothelial cells. Double-headed arrow indicates the width of Bruch’s membrane. BLD, basal laminar deposits. Scale bars: 5 µm (A) and 500 nm (B,C).

 


Figure 4
View larger version (132K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. RPE and Bruch’s membrane abnormalities in Efemp1 knock-in mice. (A) Unhealthy RPE cells with degenerating nucleus (arrowhead), vacuoles (V) containing membranous material, and misaligned and disorganized RPE basal infoldings (arrows) in a 23-month old Efemp1ki/ki mice. Sometimes large vacuoles separate the abnormal basal infoldings from the BLDs. (B) A necrotic cell in the RPE appears to be electron dense (arrowheads). (C) Thickened Bruch’s membrane (double-headed arrow) contains membrane bound vacuoles (concave arrow) and vesicles (arrows), and cellular processes (arrowheads). (D) An atrophic choroidal capillary with a degenerating endothelial cell (arrowhead) leaves multiple rings of basement membranes behind (arrow). BLD, basal laminar deposits; OS, photoreceptor outer segments and V, vacuoles. Scale bars: 2 µm (A,B) and 1 µm (C,D).

 
RPE and Bruch’s membrane abnormalities in Efemp1 knock-in mice
Up to 1 year of age, the only obvious defect we noticed other than BLDs was a loss of RPE basal infoldings near the deposits (Fig. 2H and I). In general, the RPE and Bruch’s membrane of young Efemp1 knock-in mice appeared to be indistinguishable from those of the wild-type littermates by either light or electron microscopy. However, in both 18 and 23-month old Efemp1+/ki and Efemp1ki/ki mice, we found that the appearance of the RPE had deteriorated (Fig. 4). The RPE began to lose its uniformity and became highly vacuolated (Fig. 4A). Occasionally large vacuoles separated the BLDs from the RPE (Fig. 4A). In some instances, RPE cells had become thin and spread out. We also observed abnormally appearing RPE to retract from Bruch’s membrane in some regions. While most of the RPE basal infoldings disappeared as the deposits accumulated, those remaining were disorganized and misaligned (Fig. 4A). We also observed RPE cell degeneration in focal retinal areas associated with BLDs (Fig. 4B). In ML patients, RPE pigmentary changes are often observed (2). We were not able to assess this feature in the knock-in mice because these mice have a mixed 129Sv/J and BALB/c genetic background that results in an albino or light gray (chinchilla) coat color and the RPE is minimally pigmented or non-pigmented.

We also observed significant changes in Bruch’s membrane in the older knock-in mice. At 18 and 23 months Efemp1+/ki and Efemp1ki/ki mice, focal thickening of the Bruch’s membrane often accompanied thick BLDs and membranous debris (Fig. 4C). Thickened Bruch’s membrane contained either amorphous materials or membrane bound vesicles or debris. Sometimes the vesicles were enclosed in large membrane bound vacuoles (Fig. 4C). There were frequent invasions of cellular processes into the thickened Bruch’s membrane (Fig. 4C). The origin of these processes was presumably choroidal endothelial cells or fibroblasts. Choriocapillaris degeneration was also evident in these mice. The degeneration left behind rings of endothelial basement membranes (Fig. 4D). As a result, the intercapillary pillars were widened.

Histology of the neurosensory retina in Efemp1 knock-in mice
At the ages we examined (up to 2 years old), neither Efemp1+/ki nor Efemp1ki/ki mice showed significant changes in the photoreceptors or other layers of the neurosensory retina (Fig. 5B) compared to their wild-type littermates (Fig. 5A). There was localized shortening of outer segments or thinning of the outer nuclear layer in some of the older knock-in mice; however, there was no gross retinal degeneration. Small vesicles were observed between the outer and inner nuclear layers in some areas of the knock-in retina (Fig. 5B), but we also observed them at a similar frequency in wild-type mice. We found that the outer segments were frequently detached from the RPE in 18 or 23-month old knock-in mice (Fig. 3A). Though retinal detachment can be an artifact of histological processing of tissue sections, these detachments were rarely observed in sections from the wild-type littermates, which were processed in parallel with knock-in sections. Moreover, the RPE microvilli were always exposed in the detached areas, suggesting a weakened RPE-retina interaction rather than a processing artifact. To determine whether this was associated with a gross loss of retinal function, electroretinograms (ERGs) were obtained from 18-month old mice. There were no significant differences in the amplitude or implicit time (not shown) of the a- and b- waves between the knock-in mice and wild-type littermates (Fig. 5C and D). This correlates with the relatively intact retina in the knock-in mice, and is consistent with the unremarkable ERG findings reported in ML patients (36).


Figure 5
View larger version (79K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. ERGs and morphology of the neurosensory retina of Efemp1 knock-in mice. (A) The retina from a 18-month old wild-type mouse. (B) The retina from an 18-month old Efemp1ki/ki mouse. The magnification is the same for (A) and (B). Size bar: 50 µm. (C, D) Amplitude of the ‘a-wave' and ‘b-wave' components of the dark-adapted ERG plotted as a function of stimulus intensity of 18-month old Efemp1ki/ki mice and their wild-type littermates. Symbols indicate the mean ± SEM of three mice.

 
Accumulation of mutant fibulin-3 in the sub-RPE deposits in Efemp1 knock-in mice
We have previously found that fibulin-3 accumulates within basal deposits in AMD donor eyes and within and between drusen and the RPE in both AMD and ML donor eyes (26). To determine whether fibulin-3 accumulates in the BLDs in Efemp1 knock-in mice, we stained frozen eye sections with an anti-mouse fibulin-3 antibody. Fibulin-3 staining in Bruch’s membrane was relatively weak in the eyes of the wild-type mice. A thin signal in Bruch’s membrane, staining slightly heavier than the surrounding tissues (RPE or choriocapillaris) (Fig. 6A), was observed when the eye was examined at high magnification (1000x). In a previous study we reported the absence of fibulin-3 in normal human Bruch’s membrane using a monoclonal antibody against human fibulin-3 sequences (26). Using different antibodies, we have since detected fibulin-3 in Bruch’s membrane of normal human donor eyes (our unpublished data). The antibody used in the present study was raised against mouse fibulin-3 and did not label Bruch’s membrane in Efemp1 knockout mice (our unpublished data). Thus, we believe that fibulin-3 is a normal component of Bruch’s membrane in both human and mouse and the particular antibody and staining method used previously (26) was not able to detect the low level of fibulin-3 in normal human Bruch’s membrane. In young mice when there were no prominent sub-RPE deposits in the knock-in mice, significant differences in fibulin-3 staining were not found between the eyes of wild-type and knock-in animals (data not shown). In older knock-in mice with substantial sub-RPE deposits, however, heavy immunostaining for fibulin-3 was observed (Fig. 6B and C). The staining in the deposits was non-uniform, exhibiting a granular or clumped appearance. The increased immunostaining signal appeared to be the result of fibulin-3 focal accumulation in the deposits, because the overall level of fibulin-3 protein was not increased in the knock-in mice in western blot analysis (data not shown). Although fibulin-3 is normally present in photoreceptor outer segments and other areas of the retina (25,26), we did not observe its accumulation in the retina. Wild-type mice at these ages did not demonstrate any accumulation of fibulin-3.


Figure 6
View larger version (29K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 6. Fibulin-3 accumulation in the sub-RPE deposits of Efemp1 knock-in mice. Frozen sections from 22-months old mice were stained with a polyclonal antibody against mouse-fibulin-3 (red). (A) Wild-type mouse control. Note that Bruch’s membrane (arrows) was stained stronger than the RPE or choroid. (B) Heterozygous Efemp1 knock-in mouse. Note the heavy fibulin-3 staining along Bruch’s membrane and the sub-RPE deposits (arrows). (C) Homozygous Efemp1 knock-in mouse. Again, note the heavy fibulin-3 staining in Bruch’s membrane and the sub-RPE deposits (arrows). The nuclei were stained with DAPI (blue). The magnification is the same for all the panels. Scale bar: 10 µm.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
In this study we generated and characterized a knock-in mouse strain carrying the R345W mutation in Efemp1 that causes ML. Our results demonstrate that the knock-in mouse closely recapitulates the early pathophysiology of human patients, including sub-RPE deposits, RPE and Bruch’s membrane defects, and choroidal abnormalities, and is therefore a valid model for ML. This animal model allows us to evaluate primary pathological changes that are nearly impossible to follow in human patients due to the rare nature of the disease and the scarcity of donor tissues obtained at early disease stages. In ML patients, the first symptoms observed clinically are drusen. In the knock-in mice we are able to identify BLDs as the earliest change through histological analysis. BLDs are not visible in a fundus exam and are not detectable by any other currently available clinical tests. Since it is not possible to perform histological analysis on young ML patients, the knock-in model provides a unique opportunity to dissect the early pathogenic events that occur in ML.

In older Efemp1 knock-in mice, membranous debris accumulates in BLDs and within Bruch’s membrane. Membranous debris is considered to be a critical determinant leading to advanced macular degeneration in humans and the quantity and site of membranous debris accumulation closely correlates with the degree of degeneration (34,35). Histological studies on AMD donor eyes have shown that membranous debris appears in BLDs, accumulates as rounded collections between the RPE plasma membrane and its basement membrane to form basal mounds, collects within Bruch’s membrane to form basal linear deposits, and builds up focally in Bruch’s membrane to form soft drusen (34,35,37). Basal linear deposits and soft drusen are two forms of the same lesion (34,35,37). Though BLDs and basal linear deposits are not visible in fundus exam, basal mounds can appear in fundoscopy as small drusen (35). There are few histopathologic studies on ML in humans. The small radial drusen found in younger ML patients are internal to the basement membrane of the RPE (4,5) and may therefore be basal mounds, and the large drusen in later stages of ML would be focal accumulation of membranous debris in Bruch’s membrane. Results from Efemp1 knock-in mice demonstrate that BLDs are the initial histopathologic manifestation of the disease, and indicate that membranous debris develops subsequently within the BLDs and in Bruch’s membrane as basal linear deposits. Within the range of ages that we studied (up to 2 years old) the Efemp1 knock-in mice have not developed drusen as seen in ML patients. This is consistent with the anatomical differences between mouse and human. The mouse retina more closely resembles the peripheral retina of humans and does not have a macula. In both ML and AMD patients, the large drusen are always found in the macula, while the peripheral retina is usually well preserved. Many unique features of the human macula may aggravate the insults and promote faster accumulation of membranous debris and the formation of drusen. For example, the macula has a higher photoreceptor to RPE ratio (38) and there is a lack of retinal blood vessels in its center, the fovea. These features likely increase the metabolic burden of central RPE cells supporting macular photoreceptors and the stress on a damaged RPE (38).

Why mutant fibulin-3 leads to the formation of BLDs and the subsequent pathologic events remain unknown. Consistent with our previous finding that fibulin-3 accumulates in basal deposits and between the RPE and drusen in ML or AMD donor eyes, we find fibulin-3 heavily accumulates in the BLDs in both heterozygous and homozygous Efemp1 knock-in mice. Previous study has shown that mutant fibulin-3 is misfolded and secreted inefficiently from cells overexpressing it and causes ER stress that may lead to RPE cell dysfunction (26,39). It is possible that the formation of BLDs is a reaction of the RPE to cellular stress or damage by the mutant fibulin-3. Consistent with this hypothesis, other studies have found that BLDs formed in response to increased stress or damage of RPE caused by lack of SOD1 (32), blue light exposure (40), mutant lysosomal enzymes (41), or laser photochemical injury (42). It is also possible that mutant fibulin-3 causes the formation of a structurally altered Bruch’s membrane that induces a common injury response from RPE cells by secreting excess basement membrane. For example, mice lacking collagen XVIII, another component of Bruch’s membrane, also develop BLDs (43). As BLD becomes continuous sheets and thickens, it progressively separates the RPE from its basement membrane and choroidal blood supply, causes decreased permeability of Bruch’s membrane, metabolic insufficiency, and results in subsequent pathologic events. Alternatively, fibulin-3 may function as a localization cue for other basement membrane components such as TIMP-3. Accumulation of fibulin-3 could nucleate the accumulation of other basement membrane components to result in the formation of BLDs. The Efemp1 knock-in mouse model will be an invaluable tool to further study the mechanism by which the R345W mutation causes macular degeneration.

Efemp1 knock-in mice do not appear to have other systemic defects. ML patients are also not known to have other consistent systemic symptoms. Since fibulin-3 is expressed throughout the body, it is puzzling that mutant fibulin-3 seems to only cause macular degeneration, but not other abnormalities. We do not believe that the limited phenotype in ML patients and Efemp1 knock-in mice reflects a different function of fibulin-3 in Bruch’s membrane from other basement membranes. Fibulin-3 is highly expressed by epithelial and endothelial cells, and it may have an important role in maintaining the normal function of adjacent epithelial and endothelial cells in all basement membranes. However, the requirements for proper interaction with the ECM might differ between the RPE and other epithelial cells. RPE cells do not normally undergo mitosis once differentiated, while most other epithelial cells can regenerate. Bruch’s membrane is strategically located between the RPE/photoreceptor complex and their blood supply in the choroid. The non-dividing property and high metabolic activity of the RPE throughout life may make RPE cells more susceptible to subtle changes in the ECM, and induce the formation of sub-RPE deposits over time.

The etiology of macular degeneration is multifactorial, with different genetic or environmental factors acting or interacting to produce clinical conditions in different patients. Even ML patients carrying the same mutation develop symptoms at different ages. A previous study has found that fibulin-3 and TIMP-3 physically interact with each other (27). It is possible that all or some of the macular degeneration genes act in different aspects of a common disease pathway leading to macular degeneration. Considering the many phenotypic similarities between ML and AMD, the identification of biological pathways involving fibulin-3 offers a significant starting point toward identifying the steps in the pathogenesis of these diseases. The Efemp1 knock-in mice will be particularly useful model in understanding how a wide variety of symptoms and causative factors are interrelated in macular degeneration.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Generation of Efemp1 knock-in mice containing the R345W mutation
A BAC clone containing the entire mouse genomic DNA for Efemp1 was isolated from a genomic 129/SvJ mouse library. A 10 kb fragment from intron IX to 3' end of the mouse Efemp1 gene was used as the homology sequence and was cloned into the vector PGKneolox2DTA (kindly provided by Dr Philippe Soriano, Fred Hutchinson Cancer Research Center, Seattle, Washington). The neor gene expression cassette floxed by two direct loxP sequences was inserted in inverse orientation with respect to the Efemp1 gene at a ClaI site so the floxed cassette was introduced into intronic sequence (Fig. 1). The R345W mutation was generated using the QuikChange site-directed mutagenesis method (Stratagene). A 41-mer primer and its complimentary anti-sense primer containing three nucleotide changes (A to G, C to T, and A to G) were used: 5'-GTGAGACCACCAATGAGTGCTGGGAAGATGAGATGTGCTGG-3'. One nucleotide change eliminated a BsmI restriction site without changing codons for amino acids. The eliminated BsmI site provides an easy means to distinguish the mutant and wild-type loci in the subsequent verification steps. The mutation site was verified by sequencing. The targeting construct was linearized and electroporated into 129 Sv/J ES cells (Cell and Molecular Technologies, Inc.). Homologous recombination between the wild-type locus and the targeting vector resulted in the incorporation of the mutation into the endogenous locus and the floxed neor cassette into intron XI. The mutation was verified by PCR/sequencing for the clones used for microinjection. Recombinant clones carrying the R345W mutation were injected into C57BL/6 blastocysts to generate chimeras. Germline-transmitting chimeras were crossed with transgenic BALB/c females expressing Cre-recombinase under the control of a human CMV promotor (Jackson Laboratory) to delete the floxed neor cassette. The F1 heterozygous offspring (Efemp1+/ki) determined by Southern blot and PCR/BsmI digestion analysis of genomic DNA were inbred to segregate the Cre transgene from the Efemp1 mutation and to yield Efemp1 mutants homozygous (Efemp1ki/ki) for the R345W mutation.

Southern blot analysis and PCR
Southern blot analysis and PCR were performed as previously described for identification of homologous recombinants and the mutation (11). DNA was digested with AvrII and AhdI and hybridized with a 5' external probe or digested with BamHI and hybridized with a 3' external probe. The external probe sequences are outside of the homologous arm regions. PCR with primers flanking the R345W mutation (5'-ATCTCAGTCAGAGAGTGACC-3' and 5'-TGTAGACCGAGGTGCTGATG-3') resulted in a 810 bp fragment. BsmI cut the PCR product amplified from the wild-type locus (R/R) to generate a 430 bp and a 380 bp fragment. The PCR product from the mutant locus was not cut by BsmI and remains 810 bp.

Fundoscopy and electroretinography
Fundus exams were performed using a Leica MZ7.5 stereomicroscope equipped with a CCD camera. Mice were examined for fundus abnormalities prior to sacrifice for histology experiments. ERGs were recorded from 18-month old mice as previously described (44).

Histology
Mice were killed by CO2 asphyxiation. The eyes are enucleated and fixed in half strength Karnovsky’s fixative (2.5% gluteraldehyde and 2% paraformaldehyde in 0.1M cacodylate buffer, pH 7.2) overnight and then post-fixed with 1% osmium tetroxide, stained in 2% tannic acid, dehydrated in a graded series of alcohols, and embedded in epoxy resin. Semithin sections (1 µm) were cut and stained with methylene blue. Thin sections were cut on a Reichert Ultracut microtome and stained with uranyl acetate and lead citrate. Samples were examined and photographed using a Philips CM-12 electron microscope equipped with an AMT CCD camera (Advanced Microscopy Techniques Corp., Danvers, MA, USA) and AMTV542 software.

Immunofluorescence
Mouse eyes were fixed in 4% paraformaldehyde in PBS for 1 h, after which the lens and anterior segment were removed. Posterior poles were returned to 4% paraformaldehyde overnight, rinsed in PBS, and then following cryoprotection in 30% sucrose in PBS, poles were embedded in OCT at –20°C. Immunofluorescence staining of 10 µm cryosections was performed as previously described using an affinity purified polyclonal anti-mouse fibulin-3 antibody (22). Cell nuclei were stained with 4',6-diamidino-2-phenylindole (DAPI). Stained sections were examined and photographed using a Nikon E600 microscope equipped with a CCD camera and ACT1 software.


   ACKNOWLEDGEMENTS
 
We thank Peggy McCuskey for assistance with electron microscopy; J. Brett Stanton for technical assistance; and Dr Philippe Soriano for providing the PGKneolox2DTA vector. This work was supported by NIH/NEI grants EY13847, EY13160, R24EY15638, VA Medical Research Service, Research to Prevent Blindness, and the Macular Vision Research Foundation.

Conflict of Interest statement. None declared.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Stone E.M., Sheffield V.C., Hageman G.S. Molecular genetics of age-related macular degeneration. Hum. Mol. Genet. (2001) 10:2285–2292.[Abstract/Free Full Text]

  2. Piguet B., Haimovici R., Bird A.C. Dominantly inherited drusen represent more than one disorder: a historical review. Eye (1995) 9(Pt 1):34–41.[Web of Science][Medline]

  3. Marmorstein L. Association of EFEMP1 with malattia leventinese and age-related macular degeneration: a mini-review. Ophthalmic Genet. (2004) 25:219–226.[CrossRef][Medline]

  4. Dusek J., Streicher T., Schmidt K. Hereditary drusen of Bruch’s membrane. II: Studies of semi-thin sections and electron microscopy results. Klin. Monatsbl. Augenheilkd. (1982) 181:79–83.[Medline]

  5. Streicher T., Schmidt K., Dusek J. Hereditary drusen of Bruch’s membrane. I. Clinical and light microscopical study. Klin. Monatsbl. Augenheilkd. (1982) 181:27–31.[Medline]

  6. Collins T. A pathological report on a case of Doyne’s choroiditis (‘honeycomb’ or ‘family choroiditis’). Ophthalmoscope (1913) 11:537–538.

  7. Stone E.M., Lotery A.J., Munier F.L., Heon E., Piguet B., Guymer R.H., Vandenburgh K., Cousin P., Nishimura D., Swiderski R.E., et al. A single EFEMP1 mutation associated with both Malattia Leventinese and Doyne honeycomb retinal dystrophy. Nat. Genet. (1999) 22:199–202.[CrossRef][Web of Science][Medline]

  8. Argraves W.S., Greene L.M., Cooley M.A., Gallagher W.M. Fibulins: physiological and disease perspectives. EMBO Rep. (2003) 4:1127–1131.[CrossRef][Web of Science][Medline]

  9. Timpl R., Sasaki T., Kostka G., Chu M.L. Fibulins: a versatile family of extracellular matrix proteins. Nat. Rev. Mol. Cell. Biol. (2003) 4:479–489.[CrossRef][Web of Science][Medline]

  10. Kostka G., Giltay R., Bloch W., Addicks K., Timpl R., Fassler R., Chu M.L. Perinatal lethality and endothelial cell abnormalities in several vessel compartments of fibulin-1-deficient mice. Mol. Cell. Biol. (2001) 21:7025–7034.[Abstract/Free Full Text]

  11. McLaughlin P.J., Chen Q., Horiguchi M., Starcher B.C., Stanton J.B., Broekelmann T.J., Marmorstein A.D., McKay B., Mecham R., Nakamura T., et al. Targeted disruption of fibulin-4 abolishes elastogenesis and causes perinatal lethality in mice. Mol. Cell. Biol. (2006) 26:1700–1709.[Abstract/Free Full Text]

  12. Nakamura T., Lozano P.R., Ikeda Y., Iwanaga Y., Hinek A., Minamisawa S., Cheng C.F., Kobuke K., Dalton N., Takada Y., et al. Fibulin-5/DANCE is essential for elastogenesis in vivo. Nature (2002) 415:171–175.[CrossRef][Medline]

  13. Yanagisawa H., Davis E.C., Starcher B.C., Ouchi T., Yanagisawa M., Richardson J.A., Olson E.N. Fibulin-5 is an elastin-binding protein essential for elastic fibre development in vivo. Nature (2002) 415:168–171.[CrossRef][Medline]

  14. Schultz D.W., Weleber R.G., Lawrence G., Barral S., Majewski J., Acott T.S., Klein M.L. HEMICENTIN-1 (FIBULIN-6) and the 1q31 AMD locus in the context of complex disease: review and perspective. Ophthalmic Genet. (2005) 26:101–105.[CrossRef][Medline]

  15. Stone E.M., Braun T.A., Russell S.R., Kuehn M.H., Lotery A.J., Moore P.A., Eastman C.G., Casavant T.L., Sheffield V.C. Missense variations in the fibulin 5 gene and age-related macular degeneration. N. Engl. J. Med. (2004) 351:346–353.[Abstract/Free Full Text]

  16. Hucthagowder V., Sausgruber N., Kim K.H., Angle B., Marmorstein L.Y., Urban Z. Fibulin-4: a novel gene for an autosomal recessive cutis laxa syndrome. Am. J. Hum. Genet. (2006) 78:1075–1080.[CrossRef][Web of Science][Medline]

  17. Loeys B., Van Maldergem L., Mortier G., Coucke P., Gerniers S., Naeyaert J.M., De Paepe A. Homozygosity for a missense mutation in fibulin-5 (FBLN5) results in a severe form of cutis laxa. Hum. Mol. Genet. (2002) 11:2113–2118.[Abstract/Free Full Text]

  18. Markova D., Zou Y., Ringpfeil F., Sasaki T., Kostka G., Timpl R., Uitto J., Chu M.L. Genetic heterogeneity of cutis laxa: a heterozygous tandem duplication within the fibulin-5 (FBLN5) gene. Am. J. Hum. Genet. (2003) 72:998–1004.[CrossRef][Web of Science][Medline]

  19. Xiang Y., Sekine T., Nakamura H., Imajoh-Ohmi S., Fukuda H., Yudoh K., Masuko-Hongo K., Nishioka K., Kato T. Fibulin-4 is a target of autoimmunity predominantly in patients with osteoarthritis. J. Immunol. (2006) 176:3196–3204.[Abstract/Free Full Text]

  20. Ehlermann J., Weber S., Pfisterer P., Schorle H. Cloning, expression and characterization of the murine Efemp1, a gene mutated in Doyne-Honeycomb retinal dystrophy. Gene. Expr. Patterns (2003) 3:441–447.[CrossRef][Medline]

  21. Giltay R., Timpl R., Kostka G. Sequence, recombinant expression and tissue localization of two novel extracellular matrix proteins, fibulin-3 and fibulin-4. Matrix. Biol. (1999) 18:469–480.[CrossRef][Web of Science][Medline]

  22. Kobayashi N., Kostka G., Garbe J.H., Keene D.R., Bachinger H.P., Hanisch F.G., Markova D., Tsuda T., Timpl R., Chu M.L., et al. A comparative analysis of the fibulin protein family. Biochemical characterization, binding interactions, and tissue localization. J. Biol. Chem. (2007) 282:11805–11816.[Abstract/Free Full Text]

  23. Lecka-Czernik B., Lumpkin C.K. Jr, Goldstein S. An overexpressed gene transcript in senescent and quiescent human fibroblasts encoding a novel protein in the epidermal growth factor-like repeat family stimulates DNA synthesis. Mol. Cell. Biol. (1995) 15:120–128.[Abstract]

  24. Albig A.R., Neil J.R., Schiemann W.P. Fibulins 3 and 5 antagonize tumor angiogenesis in vivo. Cancer Res. (2006) 66:2621–2629.[Abstract/Free Full Text]

  25. Blackburn J., Tarttelin E.E., Gregory-Evans C.Y., Moosajee M., Gregory-Evans K. Transcriptional regulation and expression of the dominant drusen gene FBLN3 (EFEMP1) in mammalian retina. Invest. Ophthalmol. Vis. Sci. (2003) 44:4613–4621.[Abstract/Free Full Text]

  26. Marmorstein L.Y., Munier F.L., Arsenijevic Y., Schorderet D.F., McLaughlin P.J., Chung D., Traboulsi E., Marmorstein A.D. Aberrant accumulation of EFEMP1 underlies drusen formation in Malattia Leventinese and age-related macular degeneration. Proc. Natl Acad. Sci. USA (2002) 99:13067–13072.[Abstract/Free Full Text]

  27. Klenotic P.A., Munier F.L., Marmorstein L.Y., Anand-Apte B. Tissue inhibitor of metalloproteinases-3 (TIMP-3) is a binding partner of epithelial growth factor-containing fibulin-like extracellular matrix protein 1 (EFEMP1). Implications for macular degenerations. J. Biol. Chem. (2004) 279:30469–30473.[Abstract/Free Full Text]

  28. Weber B.H., Vogt G., Pruett R.C., Stohr H., Felbor U. Mutations in the tissue inhibitor of metalloproteinases-3 (TIMP3) in patients with Sorsby’s fundus dystrophy. Nat. Genet. (1994) 8:352–356.[CrossRef][Web of Science][Medline]

  29. Capon M., Marshall J., Krafft J.I., Alexander R.A., Hiscott P.S., Bird A.C. Sorsby’s fundus dystrophy: a light and electron microscopic study. Ophthalmology (1989) 95:1769–1777.

  30. Edwards A.O., Malek G. Molecular genetics of AMD and current animal models. Angiogenesis (2007) 10:119–132.[CrossRef][Medline]

  31. Elizabeth Rakoczy P., Yu M.J., Nusinowitz S., Chang B., Heckenlively J.R. Mouse models of age-related macular degeneration. Exp. Eye Res. (2006) 82:741–752.[CrossRef][Web of Science][Medline]

  32. Imamura Y., Noda S., Hashizume K., Shinoda K., Yamaguchi M., Uchiyama S., Shimizu T., Mizushima Y., Shirasawa T., Tsubota K. Drusen, choroidal neovascularization, and retinal pigment epithelium dysfunction in SOD1-deficient mice: a model of age-related macular degeneration. Proc. Natl Acad. Sci. USA (2006) 103:11282–11287.[Abstract/Free Full Text]

  33. Marmorstein A.D., Marmorstein L.Y. The challenge of modeling macular degeneration in mice. Trends Genet. (2007) 23:225–231.[CrossRef][Web of Science][Medline]

  34. Sarks S.H., Sarks J.P. Age-related maculopathy: Non-neovascular age-related macular degeneration and the evolution of geographic atrophy. In: Retina—Ryan S.J., ed. (2001) 2, 3rd edn. Mosby: St. Louis. 1064.

  35. Sarks S., Cherepanoff S., Killingsworth M., Sarks J. Relationship of basal laminar deposit and membranous debris to the clinical presentation of early age-related macular degeneration. Invest. Ophthalmol. Vis. Sci. (2007) 48:968–977.[Abstract/Free Full Text]

  36. Haimovici R., Wroblewski J., Piguet B., Fitzke F.W., Holder G.E., Arden G.B., Bird A.C. Symptomatic abnormalities of dark adaptation in patients with EFEMP1 retinal dystrophy (Malattia Leventinese/Doyne honeycomb retinal dystrophy). Eye (2002) 16:7–15.[CrossRef][Web of Science][Medline]

  37. Curcio C.A., Millican C.L. Basal linear deposit and large drusen are specific for early age-related maculopathy. Arch. Ophthalmol. (1999) 117:329–339.[Abstract/Free Full Text]

  38. Dorey C.K., Wu G., Ebenstein D., Garsd A., Weiter J.J. Cell loss in the aging retina. Relationship to lipofuscin accumulation and macular degeneration. Invest. Ophthalmol. Vis. Sci. (1989) 30:1691–1699.[Abstract/Free Full Text]

  39. Roybal C.N., Marmorstein L.Y., Vander Jagt D.L., Abcouwer S.F. Aberrant accumulation of fibulin-3 in the endoplasmic reticulum leads to activation of the unfolded protein response and VEGF expression. Invest. Ophthalmol. Vis. Sci. (2005) 46:3973–3979.[Abstract/Free Full Text]

  40. Cousins S.W., Espinosa-Heidmann D.G., Alexandridou A., Sall J., Dubovy S., Csaky K. The role of aging, high fat diet and blue light exposure in an experimental mouse model for basal laminar deposit formation. Exp. Eye Res. (2002) 75:543–553.[CrossRef][Web of Science][Medline]

  41. Rakoczy P.E., Zhang D., Robertson T., Barnett N.L., Papadimitriou J., Constable I.J., Lai C.M. Progressive age-related changes similar to age-related macular degeneration in a transgenic mouse model. Am. J. Pathol. (2002) 161:1515–1524.[Abstract/Free Full Text]

  42. Dithmar S., Sharara N.A., Curcio C.A., Le N.A., Zhang Y., Brown S., Grossniklaus H.E. Murine high-fat diet and laser photochemical model of basal deposits in Bruch membrane. Arch. Ophthalmol. (2001) 119:1643–1649.[Abstract/Free Full Text]

  43. Marneros A.G., Keene D.R., Hansen U., Fukai N., Moulton K., Goletz P.L., Moiseyev G., Pawlyk B.S., Halfter W., Dong S., et al. Collagen XVIII/endostatin is essential for vision and retinal pigment epithelial function. EMBO J. (2004) 23:89–99.[CrossRef][Web of Science][Medline]

  44. Chang B., Heckenlively J.R., Bayley P.R., Brecha N.C., Davisson M.T., Hawes N.L., Hirano A.A., Hurd R.E., Ikeda A., Johnson B.A., et al. The nob2 mouse, a null mutation in Cacna1f: anatomical and functional abnormalities in the outer retina and their consequences on ganglion cell visual responses. Vis. Neurosci. (2006) 23:11–24.[Web of Science][Medline]


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


This article has been cited by other articles:


Home page
 Mol Cancer ResHome page
B. Hu, K. K. Thirtamara-Rajamani, H. Sim, and M. S. Viapiano
Fibulin-3 Is Uniquely Upregulated in Malignant Gliomas and Promotes Tumor Cell Motility and Invasion
Mol. Cancer Res., November 1, 2009; 7(11): 1756 - 1770.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
16/20/2423    most recent
ddm199v1
Right arrow Alert me when this article is cited
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 ISI Web of Science
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
Google Scholar
Right arrow Articles by Marmorstein, L. Y.
Right arrow Articles by Marmorstein, A. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Marmorstein, L. Y.
Right arrow Articles by Marmorstein, A. D.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?