any case their presence does not detract from the differences between the leukemic and remission samples of this patient. Miller et al. and Donti et al. suggest that the conclusions of our paper imply an understanding of the pathogenesis of APL or a direct association between rearranged MPO and critical DNA sequences involved in the t(15;17); this is not the case. Although questioned, it should be noted that only "one explanation for the presence of these novel bands in two of the M3 (APL) patients examined is that the breakpoint on chromosome 17 of the t(15;17) occurs within the MPO sequences. Alternatively, M3-associated rearrangements of the MPO gene such as submicroscopic deletions or inversions that are unrelated to the translocation breakpoint may have occurred" (1).We were not aware ofthe publications by Chang et al. in Leukemia (2) In the course of our previous studies of proteoglycans and other glycoconjugates in this neuronal cell line (2-4), we obtained data that raise several questions concerning this conclusion.1) Schubert et al. state that they purified the secreted form of a neuronal HSPG from PC12 cells and that they used this purified HSPG to obtain tryptic peptides that were fractionated by high-performance liquid chromatography. A major peptide labeled with 35SO4 (designated peptide 51) had an NH2-terminal amino acid sequence that was almost identical to the deduced sequence for residues 18 through 44 of human ABPP. The authors conclude that this peptide must be derived from an HSPG core protein, as HSPGs are the only proteoglycans produced by PC12 cells, and that all tryptic peptides labeled with 35s04 should therefore be derived from this proteoglycan.However, we previously reported (2) that chondroitin sulfate accounts for approximately 80% of the glycosaminoglycans secreted (in the form of proteoglycans) by the original line of PC12 cells (5), and we have recently found (4, 6) that chondroitin sulfate proteoglycans represent a smaller but still highly significant proportion (approximately 35%) ofthe proteoglycans secreted by the PC12 cells studied in two other laboratories (1, 7). The paper (7) cited as evidence that only one proteoglycan has been detected in PC12 cells reports an investigation specifically of HSPGs identified with monoclonal antibodies and does not address the question of whether other proteoglycans may also be present. Moreover, the reference cited by Schubert et al. in support of the statement that HSPGs are the major class of proteoglycans in nervous tissue clearly demonstrates that chondroitin sulfate, and not heparan sulfate, is the predominant sulfated glycosaminoglycan at all ages in brain (8).Because we have found [on the basis of gel filtration and SDS-PAGE (polyacrylamide gel electrophoresis)] that the molecular size of the HSPG secreted by PC12 cells (110 to 135 kD) is the same as that recently reported for the ABPP (9), we do not think it is surprising that these would elute together on Sepharose CL-4B. Therefore, in the absence of electrophoretic and fluo...