The Link domain [1] is a hyaluronan(HA)-binding region found in proteins of vertebrates that are involved in the assembly of extracellular matrix, cell adhesion, and migration. It is about 100 amino acids in length. The structure has been shown [2] to consist of two α helices and two antiparallel β sheets arranged around a large hydrophobic core similar to that of C-type lectin <See PDOC00537>. As shown in the schematic representation this domain contains four conserved cysteines involved in two disulfide bonds.

                              +----------+
                              |          |
                  ************|***********
              xxxxCxxxxxxxxxxxCxxxxxxxxxxCxxxxxxxxxxxxxxxCxxxxx
                  |                                      |
                  +--------------------------------------+

'C': conserved cysteine involved in a disulfide bond.
'*': position of the pattern.

The Link domain has also been termed HABM [1] (HA binding module) and PTR [3] (proteoglycan tandem repeat). Proteins with such a domain are listed below:

• The cartilage link protein (LP), a proteoglycan that togethers with HA and aggrecan forms multimolecular aggregates. It consists of an Ig-like V-type domain and two copies of the Link domain.
• The proteoglycans aggrecan, brevican, neurocan and versican, which are expressed in the CNS. These proteins are composed of an Ig-like V-type region, two or four (only in aggrecan) Link domains, up to two EGF-like repeats, a variable length domain containing the site of attachments of the sugars, followed, in the C-terminal by a C-type lectin and a Sushi domain.
• CD44 antigen. The main cell surface receptor for HA. CD44 is known by many different names and also exists in many different forms due to extensive alternative splicing of its 19 exons. It contains a single N-terminal Link domain, which has been shown to be involved in HA-binding [4].
• Tumor necrosis factor-inducible protein TSG-6. It is possibly involved in cell-cell and cell-matrix interactions during inflammation and tumorgenesis. It contains a Link domain and a CUB domain.

We developed a signature pattern for this domain, which is located in the middle of the sequence and covers the first three conserved cysteines. We also developed a profile, which covers the entire Link domain.

PROSITE is copyrighted by the SIB Swiss Institute of Bioinformatics and distributed under the Creative Commons Attribution-NonCommercial-NoDerivatives (CC BY-NC-ND 4.0) License, see https://prosite.expasy.org/prosite_license.html --------------------------------------------------------------------------------.