The ultrastructural composition of basement membranes in vivo

Authors
Citation
N. Miosge, The ultrastructural composition of basement membranes in vivo, HIST HISTOP, 16(4), 2001, pp. 1239-1248
Citations number
82
Categorie Soggetti
Medical Research Diagnosis & Treatment
Journal title
HISTOLOGY AND HISTOPATHOLOGY
ISSN journal
02133911 → ACNP
Volume
16
Issue
4
Year of publication
2001
Pages
1239 - 1248
Database
ISI
SICI code
0213-3911(200110)16:4<1239:TUCOBM>2.0.ZU;2-A
Abstract
The ultrastructure of basement membranes has a homogeneous appearance. The enormous cell biological importance of basement membranes and their compone nts for cell proliferation, migration and differentiation implies that thei r composition is more complex than their structure suggests. To elucidate t he molecular composition of basement membranes in vivo, we optimised immuno gold histochemistry to allow the determination of the molecular arrangement of matrix molecules. Basically, we apply a mild fixation and embed the tissues in the hydrophili c LR-Gold((R)). This preserves the basement membrane with a quality similar to freeze substitution. The application of two antibodies directed toward the C- and N-terminal ends of a molecule and coupled to gold particles of d ifferent sizes allows determination of the orientation of a molecule within the basement membrane. We were able to demonstrate that the molecular orie ntation of the laminin-1 molecule changes in the basement membrane accordin g to cell biological needs. We also showed that ultrastructurally identical basement membranes like the ones of the proximal and distal tubules of the kidney have a differing molecular arrangement. Integrin alpha7 influences the molecular composition of the basement membranes at the myotendinous jun ction. With the help of double labelling at the ultrastructural level we co uld show that nidogen-1 is co-localised with laminin-1 and only found in fu lly developed, mature basement membranes. In general, laminin-1, nidogen-1 and collagen type IV are localised over the entire width of basement membra nes, with laminin-1 and nidogen-1 co-localised, in accordance with the curr ent basement membrane models. Incidentally, our investigations warn us, that not every matrix protein fou nd at the light microscopic level as a linear staining pattern underneath a n epithelium (basement membrane zone) is a real basement membrane component when investigated at the ultrastructural level. Instead, one and the same molecule, e.g. endostatin, can be a basement membrane component in one orga n and a matrix molecule in another.