Chest ACCP Member Benefits
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
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 PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sheppard, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sheppard, D.
(Chest. 2001;120:S49-S53.)
© 2001 American College of Chest Physicians

Integrin-Mediated Activation of Transforming Growth Factor-ß1 in Pulmonary Fibrosis*

Dean Sheppard, MD

* From the Lung Biology Center, Center for Occupational and Environmental Health, Cardiovascular Research Institute, Department of Medicine, University of California, San Francisco, San Francisco, CA.

Correspondence to: Dean Sheppard, Lung Biology Center, UCSF Box 0854, San Francisco, CA 94143

Abstract

The integrin {alpha}vß6 is restricted to epithelial cells and is dramatically induced in response to injury and inflammation. Mice expressing a null mutation of this integrin develop exaggerated inflammation of the lungs and skin, but are dramatically protected from bleomycin-induced pulmonary fibrosis. This phenotype led to the identification of a unique role for this integrin in binding to and activating latent extracellular complexes of the anti-inflammatory, profibrotic cytokine, transforming growth factor-ß1. This integrin-mediated activation is tightly spatially restricted and appears to require direct presentation of the activated cytokine to receptors on adjacent cells. The process also requires distinct regions of the ß6-subunit cytoplasmic domain and an intact actin cytoskeleton, suggesting the existence of additional cellular mechanisms to regulate this process. If this mechanism is found to be as important in humans as it is in mice, the integrin and as yet to be identified pathways for cellular regulation of this process could be exciting new targets for intervention in fibrotic diseases of the lung and other epithelial organs.

Key Words: {alpha}vß6 • integrin • pulmonary fibrosis • transforming growth factor-ß activation

Pulmonary epithelial cells express at least eight distinct integrin heterodimers.1 Two of these integrins, {alpha}3ß1 and {alpha}6ß4, recognize the epithelial basement membrane protein, laminin 5, as a ligand. Knockout mice lacking either of these integrins have defects in epithelial integrity,2 3 4 5 suggesting that these integrins may function as adhesion receptors and play important roles in the maintenance of epithelial integrity. However, the other six integrins expressed on lung epithelial cells recognize ligands that are not normally present in the epithelium or its basement membrane. In contrast, many of the ligands for these integrins, including osteopontin, fibronectin, and tenascin-C, are components of the provisional matrix that is produced in response to injury and inflammation.1 These observations suggest that many of the integrins expressed on lung epithelial cells may be present to regulate responses of these cells to tissue injury and inflammation, and could thus participate in the development of inflammatory lung diseases.

My laboratory has had a special interest in the integrin, {alpha}vß6, the only integrin that is restricted in its expression to epithelial cells. This integrin, initially identified from primary cultures of guinea pig airway epithelial cells,6 is expressed at only low levels in healthy airway and alveolar epithelial cells of adults, but is rapidly induced at both sites in response to a variety of insults.7 8 9 10 The initial ligands identified for this integrin, fibronectin,11 tenascin-C,12 13 and vitronectin,14 all bind the integrin through a linear tripeptide sequence, arginine-glycine-aspartic acid (RGD), which is also the sequence recognized by several other integrins, including all of the integrins that share the {alpha}v subunit.15

Initial studies of the biological role of the {alpha}vß6 integrin depended on examining the behavioral effects of heterologously expressing this integrin in cells that did not normally express it. These studies demonstrated two unique effects of this integrin: enhanced proliferation in three-dimensional culture16 and induction of expression of the matrix metalloproteinase-9.17 Interestingly, both of these effects appeared to depend on the presence of a unique 11 amino-acid sequence at the carboxyl terminus of the ß6 subunit.16 Cells transfected to express a mutant version of the integrin containing a ß6 subunit lacking the last 11 amino acids continued to bind and spread on {alpha}vß6 ligands, but were incapable of proliferating in either three-dimensional collagen gels or in vivo in nude mice. These data further supported potential roles for this integrin in dynamic responses to injury such as epithelial proliferation or migration through extracellular matricies.

To obtain more direct information about the role of {alpha}vß6 in vivo, we inactivated the ß6 subunit in embryonic stem cells and generated lines of mice homozygous for this null mutation. Because the ß6 subunit is only present in a single integrin heterdimer, these mice are effectively {alpha}vß6 knockouts. ß6 knockout mice had a completely unexpected phenotype, demonstrating functionally significant inflammation in the lungs and skin.18 Inflammation in the skin was present principally at sites of low-grade trauma, was characterized morphologically by large numbers of macrophages infiltrating the dermis, and resulted in destruction of hair follicles and inflammatory baldness. This response appears to reflect an interaction between this genetic alteration and an environmental insult (in this case, low-grade trauma), since the inflammation and baldness are most prominent in the region where mothers lift their infants with their teeth, and resolved after weaning. However, inflammatory baldness generally persists over the inner thighs, an area subjected to low-grade friction, even in adults.

In addition to inflammation in the skin, ß6 knockout mice develop progressive and lifelong inflammation in the lungs.18 This effect also likely requires an environmental insult since it is more prominent in mice kept in unventilated isolator cages than in those kept in ventilated cages. In the lung, there is an increase in several types of inflammatory cells, including macrophages, lymphocytes, neutrophils, and eosinophils. The lymphocytes are mostly "activated" as determined by expression of the activation marker, CD25 (the interleukin-2–receptor {alpha} subunit). Because this phenotype was unexpected, and not explained by any known biological effect of {alpha}vß6, it was important to determine whether this effect was actually due to inactivation of the ß6 subunit gene, rather than some unanticipated effect of unknown adjacent genes on the same chromosome. To address this issue, we generated a line of mice that overexpressed the human ß6 subunit under the control of the human surfactant protein-C promoter that drives expression in alveolar type-II cells and a subset of bronchiolar epithelial cells. We then backcrossed these mice to ß6 knockout mice and examined the effects of this "rescue" transgene on lung inflammation, as assessed by BAL.19 Surprisingly, expression of the ß6 subunit in this subset of epithelial cells in the lung periphery was sufficient to prevent the dramatic increases in macrophages, lymphocytes, eosinophils, and neutrophils in the airspaces of ß6 knockout mice. These findings definitively identified the ß6 gene itself as the critical cause of the exaggerated lung inflammation in ß6 knockout mice.

Because ß6 knockout mice clearly manifest increased sensitivity to inflammatory stimuli, we naively reasoned that these mice would be a good model of enhanced susceptibility to lung damage that results from inflammatory insults. To test this idea, we first examined whether these mice would have increased susceptibility to bleomycin-induced pulmonary fibrosis,20 a response that has been thought to be a late consequence of bleomycin-induced inflammation. We treated either wild-type or otherwise genetically identical ß6 knockout mice with a single dose of intratracheal bleomycin and examined the degree of fibrosis by assessing lung morphology and measuring lung hydroxyproline content 15, 30, and 60 days later.21 As expected, the degree of bleomycin-induced inflammation was greater in ß6 knockout mice at every time point examined. However, to our great surprise, ß6 knockout mice did not develop exaggerated fibrosis. In fact bleomycin-induced fibrosis was nearly completely absent in ß6 knockout mice at every time point, whereas wild-type mice developed progressively more severe fibrosis throughout the time period examined.

This pattern of enhanced inflammation but dramatically reduced fibrosis strongly suggested that the cytokine transforming growth factor (TGF)-ß1 might be downstream in a pathway involving the integrin {alpha}vß6. The reasons to suspect a role for TGF-ß1 include the dramatic tissue inflammation seen in TGF-ß1 knockout mice22 and the well-established central role of TGF-ß in tissue inflammation at multiple sites, including the lung. Indeed, there is considerable evidence that antagonists of TGF-ß can prevent bleomycin-induced pulmonary fibrosis. However, despite multiple efforts, we were never able to identify any difference in expression of TGF-ß messenger RNA or protein between wild-type or ß6 knockout mice either at baseline or at any time after treatment with bleomycin. An alternative explanation for how TGF-ß might act downstream of {alpha}vß6 was suggested by work done by John Munger, a pulmonary physician scientist at New York University. Dr. Munger showed that two other integrins related to {alpha}vß6, {alpha}vß1, and {alpha}vß5, bound to the tripeptide sequence RGD that is present in the N-terminal portion of the TGF-ß1 gene product called the latency-associated protein (LAP).23 We had previously shown that {alpha}vß6 recognizes this same sequence in each of its previously identified ligands (fibronectin, vitronectin, and tenascin-C). Importantly, it has been known for some time that the TGF-ß gene product is processed in the secretory apparatus through cleavage by the endoprotease, furin, and is assembled prior to secretion into a double homodimer composed of two copies of LAP and two copies of mature TGF-ß. This complex, called the small latent complex, is functionally inactive and is unable to bind to TGF-ß receptors or activate known biological effects of mature TGF-ß.24 Although these latent complexes can be easily activated in vitro by denaturing conditions or proteases, the precise mechanisms controlling activation have been poorly understood. Dr. Munger hypothesized that integrin binding to the RGD site(s) in LAP might induce activation of these latent complexes. However, he was unable to demonstrate such activation mediated by {alpha}vß1 or {alpha}vß5.

After seeing Dr. Munger’s data at the annual meeting of the American Thoracic Society, we established a collaboration between our laboratories that allowed us together to demonstrate that TGF-ß1 LAP was by far the most effective ligand for {alpha}vß6 identified to date.21 We therefore examined the ability of {alpha}vß6 to activate latent complexes, using a co-culture bioassay system combining ß6-expressing cells and mink lung epithlelial cells stably transfected with a highly TGF-ß–sensitive reporter system composed of a portion of the plasminogen activator inhibitor-1 promoter driving expression of firefly luciferase.25 These studies clearly demonstrated that four different cell lines and two types of primary epithelial cells expressing {alpha}vß6 all induced TGF-ß activity, an effect that could be equally well inhibited by antibodies to active TGF-ß1 or {alpha}vß6. Thus, TGF-ß1 does appear to be directly downstream of {alpha}vß6 as a result of extracellular integrin-mediated activation of latent complexes.

Given the potent ability of active TGF-ß to induce tissue fibrosis, it would be attractive to posit a mechanism by which TGF-ß activity could be spatially restricted, especially in organs with critically important fine structure, such as the gas exchange regions of the lung. Since integrins are spatially restricted to discrete regions of the plasma membrane, integrin-mediated activation could provide such a mechanism, provided that activated complexes did not dissociate and allow free diffusion of active TGF-ß away from the cell surface. To test this hypothesis, we performed co-culture bioassays in dishes containing microporous inserts, and either cultured {alpha}vß6-expressing cells and reporter cells on the same side of the filter or on opposite sides, to prevent test and reporter cells from touching each other. Active TGF-ß could freely diffuse across these filters, but TGF-ß that remained attached to the integrin-expressing cells could not. For each cell line tested, integrin-induced TGF-ß activity was dramatically reduced when cells were plated on opposite sides of the filter, strongly suggesting that this process does provide a mechanism for spatially restricted activation of TGF-ß (Fig 1) .21



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Regulation of {alpha}vß6 expression as a mechanism of regulating TGF-ß activation. TGF-ß itself is one of the most potent inducers of {alpha}vß6 expression; however, the mechanism(s) responsible for activation by tissue injury and inflammation in vivo remain to be determined.

 
We next sought to determine whether binding of latent complexes to {alpha}vß6 was sufficient to induce activation or whether this process could be subject to regulation by the integrin-expressing cell. Such an additional level of regulation would be attractive as a potential mechanism for fine-tuning both the spatial and temporal pattern of activation. To address this question, we took advantage of cell lines we had previously generated that express mutant forms of the ß6 subunit containing a variety of truncations of the cytoplasmic domain.16 26 27 With this approach, we were able to identify two mutants that allowed us to separate binding of TGF-ß1 LAP from activation. Both the full-length integrin and a mutant lacking the C-terminal 11 amino acids bound LAP and fully activated latent complexes. In contrast, a more extreme truncation, lacking the last 18 amino acids, bound LAP perfectly normally but did not induce activation. Thus, integrin binding is not sufficient to activate latent TGF-ß, suggesting that cells could potentially regulate this effect. Experiments utilizing the actin-depolymerizing agent, cytochalasin D, which completely abolished TGF-ß activation without affecting the response of the reporter cells to exogenous active TGF-ß, strongly suggest that the cytoskeleton participates in this activation process (Fig 2) .



View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Tight spatial and temporal regulation of integrin-mediated TGF-ß activation is probably regulated by a signaling pathway that involves close association of {alpha}vß6 with the actin cytoskeleton. The extracellular and cytoplasmic proteins that regulate this pathway remain to be determined. Spatial regulation is enhanced by the restriction of this process to the paracrine-activation pathway and the requirement for direct cell-cell contact.

 
The data described above clearly identify a pathway by which the integrin, {alpha}vß6, could activate latent TGF-ß and thereby negatively regulate lung inflammation and enhance bleomycin-induced pulmonary fibrosis. However, all of the direct evidence in support of this hypothesis comes from in vitro experiments, and these results cannot, by themselves, establish whether such a mechanism is the sole or even principal explanation for the phenotype of ß6 knockout mice. To examine this issue in a relatively unbiased fashion, we have utilized mu6500 (Affymetrix; Santa Clara, CA) gene chips to examine the global pattern of gene expression in wild-type and ß6 knockout 129 strain mice, at baseline and at various times after treatment with bleomycin. In parallel, in collaboration with John Allard and Renu Heller at Roche Bioscience (Palo Alto, CA), we also examined the global pattern of gene expression in another strain of mice, wild-type C57bl/6 mice.28 With this approach, we were able to identify a cluster of 66 genes that appeared to be specifically involved in the fibrotic response to bleomycin, since these genes were expressed at similar levels at baseline in all strains of mice, but were induced at considerably higher levels in the two strains of mice that developed fibrosis than in the ß6 knockout mice, which did not. Interestingly, a substantial fraction of the genes on the chip known to be induced by TGF-ß were included in this group. Furthermore, when we compared the level of expression of approximately 100 genes that we could identify as known TGF-ß–responsive genes between bleomycin-treated wild-type and ß6 knockout mice and between bleomycin-treated and vehicle-treated C57bl/6 mice, nearly all of these genes were expressed at higher levels in bleomycin-treated wild-type animals. Certainly, these descriptive data do not prove that TGF-ß activation is responsible for the contribution of {alpha}vß6 to in vivo pulmonary fibrosis, but they do at least provide further support for this hypothesis.

Thus, we have taken advantage of fortuitous observations in genetically manipulated mice to identify a mechanism by which an epithelially restricted integrin can activate latent TGF-ß1 that is stored in the extracellular space. This process has the potential for tight spatial and temporal regulation and is thus an attractive mechanism to restrict this potent profibrotic cytokine to the sites where it is most needed, thereby limiting the undesirable consequences of free diffusion. If this mechanism can be shown to function at other sites and in humans as well as mice, {alpha}vß6 itself, and the as yet to be identified cellular regulators of this process could be attractive new targets for treatment.

Footnotes

Abbreviations: LAP = latency-associated protein; RGD = arginine- glycine-aspartic acid; TGF = transforming growth factor

References

  1. Sheppard, D (1996) Epithelial integrins. Bioessays 18,655-660[CrossRef][ISI][Medline]
  2. van der Neut, R, Krimpenfort, P, Calafat, J, et al (1996) Epithelial detachment due to absence of hemidesmosomes in integrin ß4 null mice. Nat Genet 13,366-369[CrossRef][ISI][Medline]
  3. Georges-Labouesse, E, Messaddeq, N, Yehia, G, et al (1996) Absence of integrin {alpha}6 leads to epidermolysis bullosa and neonatal death in mice. Nat Genet 13,370-373[CrossRef][ISI][Medline]
  4. DiPersio, CM, Hodivala-Dilke, KM, Jaenisch, R, et al (1997) {alpha}3ß1 Integrin is required for normal development of the epidermal basement membrane. J Cell Biol 137,729-742[Abstract/Free Full Text]
  5. Dowling, J, Yu, Q-C, Fuchs, E (1996) ß4 Integrin is required for hemidesmosome formation, cell adhesion and cell survival. J Cell Biol 134,559-572[Abstract/Free Full Text]
  6. Sheppard, D, Rozzo, C, Starr, L, et al (1990) Complete amino acid sequence of a novel integrin ß subunit (ß6) identified in epithelial cells using the polymerase chain reaction. J Biol Chem 265,11502-11507[Abstract/Free Full Text]
  7. Breuss, JM, Gillett, N, Lu, L, et al (1993) Restricted distribution of integrin ß6 mRNA in primate epithelial tissues. J Histochem Cytochem 41,1521-1527[Abstract]
  8. Breuss, JM, Gallo, J, DeLisser, HM, et al (1995) Expression of the ß6 integrin in development, neoplasia, and tissue repair suggests a role in epithelial remodeling. J Cell Sci 108,2241-2251[Abstract]
  9. Haapasalmi, K, Zhang, K, Tonnesen, M, et al (1996) Keratinocytes in human wounds express alpha v beta 6 integrin. J Invest Dermatol 106,42-48[CrossRef][ISI][Medline]
  10. Weinacker, A, Ferrando, R, Elliott, M, et al (1995) Distribution of integrins {alpha}vß6 and {alpha}9ß1 and their known ligands, fibronectin and tenascin, in human airways. Am J Respir Cell Mol Biol 12,547-556[Abstract]
  11. Busk, M, Pytela, R, Sheppard, D (1992) Characterization of the integrin {alpha}vß6 as a fibronectin-binding protein. J Biol Chem 267,5790-5796[Abstract/Free Full Text]
  12. Prieto, AL, Edelman, GM, Crossin, KL (1993) Multiple integrins mediate cell attachment to cytotactin/tenascin. Proc Natl Acad Sci U S A 90,10154-10158[Abstract/Free Full Text]
  13. Yokosaki, Y, Monis, H, Chen, J, et al (1996) Differential effects of the integrins {alpha}9ß1, {alpha}vß3, and {alpha}vß6 on cell proliferative responses to tenascin: roles of the ß subunit extracellular and cytoplasmic domains. J Biol Chem 271,24144-24150[Abstract/Free Full Text]
  14. Huang, XZ, Wu, JF, Spong, S, et al (1998) The integrin {alpha}vß6 is critical for keratinocyte migration on both its known ligand, fibronectin, and on vitronectin. J Cell Sci 111,2189-2195[Abstract]
  15. Hynes, RO (1992) Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69,11-25[CrossRef][ISI][Medline]
  16. Agrez, M, Chen, A, Cone, R, et al (1994) The {alpha}vß6 integrin promotes proliferation of colon carcinoma cells through a unique region of the ß6 cytoplasmic domain. J Cell Biol 127,547-556[Abstract/Free Full Text]
  17. Niu, J, Gu, X, Turton, J, et al (1998) Integrin mediated signalling of gelatinase B secretion in colon carcinoma cells. Biochem Biophys Res Comm 249,287-291[CrossRef][ISI][Medline]
  18. Huang, XZ, Wu, JF, Cass, D, et al (1996) Inactivation of the integrin ß6 subunit gene reveals a role of epithelial integrins in regulating inflammation in the lung and skin. J Cell Biol 133,921-928[Abstract/Free Full Text]
  19. Huang, XZ, Wu, JF, Zhu, W, et al (1998) Expression of the human integrin ß6 subunit in alveolar type II cells and bronchiolar epithelial cells reverses lung inflammation in ß6 knockout mice. Am J Respir Cell Mol Biol 19,636-642[Abstract/Free Full Text]
  20. Jones, AW, Reeve, NL (1978) Ultrastructural study of bleomycin-induced pulmonary changes in mice. J Pathol 124,227-233[CrossRef][ISI][Medline]
  21. Munger, JS, Huang, XZ, Kawakatsu, H, et al (1999) The integrin {alpha}vß6 binds and activates latent TGF-ß1: a mechanism for regulating pulmonary inflammation and fibrosis. Cell 96,319-328[CrossRef][ISI][Medline]
  22. Shull, MM, Ormsby, I, Kier, AB, et al (1992) Targeted disruption of the mouse transforming growth factor-ß1 gene results in multifocal inflammatory disease. Nature 359,693-699[CrossRef][Medline]
  23. Munger, JS, Harpel, JG, Giancotti, FG, et al (1998) Interactions between growth factors and integrins: latent forms of transforming growth factor-ß are ligands for the integrin {alpha}vß1. Mol Biol Cell 9,2627-2638[Abstract/Free Full Text]
  24. Munger, JS, Harpel, JG, Gleizes, P, et al (1997) Latent transforming growth factor-ß: structural features and mechanisms of activation. Kidney Int 51,1376-1382[ISI][Medline]
  25. Abe, M, Harpel, JG, Metz, CN, et al (1994) An assay for transforming growth factor-ß using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct. Anal Biochem 216,276-284[CrossRef][ISI][Medline]
  26. Cone, RI, Weinacker, A, Chen, A, et al (1994) Effects of ß subunit cytoplasmic domain deletions on the recruitment of the integrin {alpha}vß6 to focal contacts. Cell Adhes Commun 2,101-113[ISI][Medline]
  27. Dixit, RB, Chen, A, Chen, J, et al (1996) Identification of a sequence within the integrin ß6 subunit cytoplasmic domain that is required to support the specific effect of {alpha}vß6 on proliferation in three-dimensional culture. J Biol Chem 271,25976-25980[Abstract/Free Full Text]
  28. Kaminiski, N, Allard, J, Pittet, J-F, et al (2000) Global analysis of gene expression in pulmonary fibrosis reveals distinct programs regulating lung inflammation and remodeling. Proc Nat Acad Sci U S A 97,1778-1783[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
E. Addis-Lieser, J. Kohl, and M. G. Chiaramonte
Opposing Regulatory Roles of Complement Factor 5 in the Development of Bleomycin-Induced Pulmonary Fibrosis
J. Immunol., August 1, 2005; 175(3): 1894 - 1902.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Lung Cell. Mol. Physiol.Home page
M. Ulanova, L. Puttagunta, M. Marcet-Palacios, M. Duszyk, U. Steinhoff, F. Duta, M.-K. Kim, Z. K. Indik, A. D. Schreiber, and A. D. Befus
Syk tyrosine kinase participates in {beta}1-integrin signaling and inflammatory responses in airway epithelial cells
Am J Physiol Lung Cell Mol Physiol, March 1, 2005; 288(3): L497 - L507.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
M. Sun, M. A. Opavsky, D. J. Stewart, M. Rabinovitch, F. Dawood, W.-H. Wen, and P. P. Liu
Temporal Response and Localization of Integrins {beta}1 and {beta}3 in the Heart After Myocardial Infarction: Regulation by Cytokines
Circulation, February 25, 2003; 107(7): 1046 - 1052.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
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 PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sheppard, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sheppard, D.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS