The bone microenvironment (e.g. glycosaminoglycans (GAGs), growth factors) plays a major role in bone resorption, especially in the formation of osteoclasts which differentiate from the hematopoietic lineage in the presence of RANKL. Previous studies revealed that GAGs may influence osteoclastogenesis, but data are very controversial, some studies showing an inhibitory effect of GAGs on osteoclastic differentiation whereas others demonstrated a stimulatory effect. To clarify their activities, we investigated the effect of 5 families of GAGs in three different models of human/mouse osteoclastogenesis. The present data revealed that heparin inhibited osteoclastogenesis in these three models, which was confirmed by a decrease in mRNA expression of osteoclastic markers and by an inhibition of the bone resorption capacity. We also demonstrated in RAW 264.7 cells that other families of GAGs different from heparin inhibited RANKL-induced osteoclastogenesis, and that this inhibition was dependent on the length and the level of sulfation of GAGs. In the present work, heparin did not bind to RANKL and did not modulate RANKL signaling. Heparin acted at 2 distinct steps of osteoclastogenesis from human CD14(+) cells: first, heparin strongly decreased the adherence of osteoclast precursors, and secondly inhibited osteoclasts to spread and to be active. Furthermore, the second action of heparin was reversible as the removal of heparin at the end of the culture time allowed the condensed cells to spread out and showed the formation of morphological active osteoclasts. The present work clearly evidences that GAGs inhibit osteoclastogenesis in vitro and strengthens the therapeutic interest of defined GAGs in osteolytic diseases.