This Thesis consists of 4 experimental Chapters preceded by a General Introduction and followed by a brief section dealing with conclusions and future. In Chapter 1, the prognostic value of CD38 and the relationship between surface expression of this molecule and IgVH mutation are considered. It is shown that surface CD38 expression on CLL cells is highly predictive of nonhypermutated IgVH gene mutation status, while CD38 negativity lacks such predictive value, since IgVH mutated and uninutated cases occur with approximately equal frequency in this subgroup. It is also shown, in agreement with the literature, that CD38 expression is associated with shorter survival and male preponderance. Abundant data are available on the intracellular expression of CD38 in other cell types, but not in CLL. Therefore, Chapter 2 deals with intracellular CD38 in CLL cells and with the different molecular forms expressed in sCD38+ and sCD38- CLL clones. Using a number of different techniques, it is shown that CD38 is expressed intracellularly in all CLL cells irrespectively of the surface expression of the molecule. In keeping with these findings, CD38 mRNA was found in both sCD38+ and sCD38- CLL clones at comparably low levels. It is documented by Western blotting and immunoprecipitation that CLL cells express, in addition to the 45 kD CD38 monomer, other molecular forms of 27, 60 and 205 M Among these, the high molecular weight molecule probably represents a tetrameric form of CD38 and is described in CLL for the first time. A main difference found between the sCD38+ and sCD38- clones was the almost complete absence of the 45 kD monomeric form in the sCD38- clones. An unexpected and interesting finding was a surface immunoreactivity with the anti- CD38 antibody Ab-4 on CLL cells previously classified as IICD38 negative" with the HB-7 antibody. In Chapter 3, the topology of different molecular forms of CD38 is studied further and their enzymatic functions are examined. By subcellular fractionation it is demonstrated that the tetrameric form of CD38 is most abundant in the membrane fraction. Surface radioiodination of CLL cells indicated, for the first time that the main forms of CD38 on the surface of CLL cells are the 45 kD and the -205 kD molecules. The 205 kD molecule was demonstrable on both sCD38+ and sCD38- cells, while the 45 kD molecule was present only on sCD38+ cells. Surface forms of CD38 on CLL cells were characterised further by enzymatic assays performed before and after protease digestion of surface proteins. Results indicated that surface forms are the major sources of CD38 enzymatic activity in CLL cells. Total cyclase and hydrolase activities were also compared in sCD38+ and sCD38- CLL cells and it was shown that both cell types possess these enzymatic functions, although activities were lower in sCD38- cells. It seems likely that the 45 kD molecule (probably present as non-covalently linked dimers) is responsible for the greater enzymatic activity of sCD38+ clones. Enzymatic assays performed after recovering proteins from polyacrylamide gels indicated that eluates from the HMW (>116 kD) section of the gels containing the putative CD38 tetramers have both cyclase and hydrolase activities.Chapter 4 addresses the question of why CD38-positive CLL is characterised by progressiveness and poor outcome. Since CD38 has a welldocumented role in cell cycle regulation and cell proliferation in other cell types, it was hypothesized that a similar linkage might be also present in CLL, contributing to the the adverse clinical outcome. To test this hypothesis, CLL cells were co-stained for surface CD38 and nuclear Ki-67 following a saponinbased permeabilization procedure. CLL clones classified as CD38+ expressed significantly higher levels of Ki-67 than did sCD38- clones. Clones classified as Ki-67+ (>5%) expressed significantly higher levels of CD38 than did Ki-67- ones. Relating surface CD38% and nuclear Ki-67% revealed a linear correlation between these two parameters. Also, the Ki-67+ subpopulation within a given CLL clone expressed significantly higher CD38 values than did the Ki-67- subpopulation. Furthermore, larger CLL cells showed significantly higher Ki-67 values than small CLL lymphocytes. Finally, CD38 immunoreactivity in CLL lymph node sections was found to be strongest in the proliferation centres. These data therefore indicate, for the first time, that surface CD38 expression is a marker of cell cycling activity in CLL and help to explain the adverse prognosis of CD38+ CLL and CLUPL.