Reply to MacLeod et al.: Multiple myeloma plasma cells have chameleon characteristics

MacLeod et al. question whether our lot of H929 multiple myeloma (MM) plasma cells were contaminated with K562 chronic myeloid leukemia (CML) cells (1). Before addressing this issue, we want to state that the central purpose of our paper was to demonstrate that hypothesis-directed proteomics using immunoprecipitation (IP)–liquid chromatography–tandem mass spectrometry (LC-MS/MS) was capable of detecting a cancer cell’s defective and activated signaling pathways. Using three IPs with antibodies against the endogenous bait proteins p85 (PI3K), Grb2 and pTyr peptides, we could successfully capture the BCR–ABL complex in our MM cells. We suggest that this proteomics strategy may be an effective method for helping to choose the appropriate single or combination of tyrosine kinase inhibitor drugs for a personalized medicine approach from patient tumor cells.

We believe that the specific lot of H929 cells used in our study was derived from MM plasma cells that acquired a BCR–ABL fusion. It appears that MacLeod et al. performed only cytogenetic tests (i.e., FISH and RT-PCR) to make a determination of contamination, and we agree that our cells possessed the e14a2 form of BCR–ABL fusion, as we reported using similar genetic tests (1, 2). Together with e13a2 fusion form, e14a2 and e13a2 represent the most common forms of BCR–ABL. The argument we presented for our H929 cells being plasma in origin came from histopathology data (immunoperoxidase/morphology) and not from genetic tests. We originally suspected CML contamination, but our data suggested otherwise. To summarize, our lot of H929 MM cells showed positive markers for plasma cells such as CD138 but were negative for standard markers of CML (K562) cells, including CD34, CD68, and myeloperoxidase. Additionally, inspection of the cells with Wright–Giemsa staining showed an abundant blue cytoplasm more deeply stained toward the periphery, prominent Golgi complex, eccentrically located nuclei, and a large and prominent central nucleolus, all features characteristic of plasma cells but not of CML cells (2).

It is important to note that BCR–ABL fusions have been detected from MM patient plasma cells in bone marrow, so our findings represent a rare but not impossible occurrence (3, 4). It is also known that some patients with MM have developed subsequent BCR–ABL–positive CML (5), and so a similar genetic situation may have occurred in our H929 cells. However, the histological data pointed to the BCR–ABL cells having been of plasma origin and not leukocyte. Because our H929 cells lacked interferon regulatory factor 4 expression, this may have contributed to the expression of BCR–ABL (6). Our lot of H929 cells may have been a special case because we tested other lots of H929 cells, and because they contained significantly lower levels of BCR–ABL by Western blotting, we did not follow up with cytogenetic tests (2). It is possible that a BCR–ABL fusion occurred somewhere in the cultured history of our H929 plasma cells that was similar to the fusion found in K562 cells, and our IP-LC-MS/MS technology showed that one could detect such cell abnormalities and use it to our advantage when testing patient cancers.