Glycoprotein (GP)IIb/IIIa is an integrin complex normally restricted in its expression to platelets and the megakaryocytes from which they are derived. This complex functions as a receptor for fibrinogen and other ligands and is involved in platelet aggregation. The receptor complex is expressed at high levels during final megakaryocyte differentiation. Further, while GPIIIa is expressed in other tissues as part of the vitronectin receptor, GPIIb is only expressed on maturing megakaryocytes and the platelets derived from them. Thus studies of the GPIIb gene may serve as a model of gene regulation during this process. Over the past several years, the genes for both GPIIb and IIIa have been cloned and analyzed. The GPIIb gene contains 30 exons over 18 kilobases (kb). The transcriptional start site has been determined and there does not appear to be a TATA-box immediately upstream of this site. Studies have been done to define regulatory elements upstream of the transcriptional start site. Most of these studies focused on the human promoter and on studies using megakaryocytic cell lines. These studies have defined several important tissue-specific promoter elements including a GATA454 site (454 basepairs upstream of the transcriptional start site that involves a GATA-binding consensus sequence), a GATA54 site and an Ets35 site (that involves an Ets-binding consensus sequence). Expression studies with megakaryocytic cell lines suggest that each of these sites effects expression approximately threefold. Further, an Ets510 site was also described that had a similar effect. While these studies were underway, we pursued studies of the rat 5'-flanking region using a rat primary marrow expression system. Qualitatively, our data support the human data; however, quantitatively, we found significant differences from the human studies done in cell lines. We found that the major tissue-specific promoter element was the GATA454 site. Mutations altering this site result in an approximately fiftyfold drop in expression. In comparison, eliminating the Ets510 site by truncation or point mutation had only a twofold effect on expression. Mutations at the Ets35 site did effect expression at a high level, decreasing expression approximately fifteenfold, while mutations at the GATA54 site effected expression by approximately ninefold. In addition, using 50 bp deletions, we have preliminarily defined two domains from -450 to -351 bp and -150 to -101 bp upstream of the transcriptional start site that effected expression. The former appears to contain a positive regulatory element, while the latter appears to be a silencer element.(ABSTRACT TRUNCATED AT 400 WORDS)