Proteomic Analysis of Soluble Proteins Important in Pediatric Acute Lymphoblastic Leukemia.

Abstract 1448

Poster Board I-471

In acute leukemia (AL), some clinical symptoms may be caused by soluble factors secreted by AL cells into the bone marrow (BM) microenvironment. On the other hand, AL cells are often dependent on the microenviroment of the host, with most AL cells dying during first days after transferred to in vitro. It is not known whether soluble factors are responsible for some host-AL interactions since most of the attention is paid to the biology of the leukemic cells themselves rather than the soluble portion of BM microenvironment.

We aimed at identifying proteins in BM plasma of children with lymphoblastic AL (ALL), which may be responsible for ALL aggressiveness or for microenvironment-mediated survival of ALL cells. We used proteomic techniques to compare BM plasma at the diagnosis of ALL with non-malignant controls (BM plasma from patients with no signs of malignant disease with a BM check-up more than one year after BM transplantation or healthy donor blood plasma).

BM plasma and blood samples were analyzed by protein microarray and/or by two-dimensional electrophoresis (2D PAGE). Protein microarray enabled a simultaneous detection of 79 proteins per sample (mainly cytokines and chemokines). The aim of 2D PAGE was to compare the complex patterns of between ALL and controls and to search for differences among unknown or unexpected molecules.

We tested 79 soluble proteins in patients (n=15) and control samples (BM plasma, n=9 and blood plasma, n=5) by protein microarray. We detected 23 proteins with a significantly different concentration in patients (p<0.05). Of these, only 2 proteins TIMP-1 and LIF differed with a high level of significance after Bonferroni correction was applied (p<0.00064). The other differences need to be confirmed or excluded in a validation cohort. We observed no difference between control blood and control BM plasma. Both LIF and TIMP-1 are likely to be produced by non-leukemic cells given the low level of expression by leukemic cells in expression profiling. Causal relationship between LIF and TIMP-1 expression, as well as their importance for cell “stemness” which have been proven in other systems, need to be confirmed in leukemia.

Before 2D PAGE, BM plasma was immuno-depleted from 12 abundant proteins by affinity chromatography. This improved considerably the separation of proteins on 2D PAGE and increased the number of detectable minor plasma proteins with possible biologic importance. We identified 393 protein spots per gel. Of these, 16 proteins spots were significantly different between the BM plasma from ALL and controls.

The differently expressed spots were identified by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF). The MALDI-TOF identified the concentration of haptoglobin to be increased in ALL, whereas hemoglobin alpha, hemoglobin beta, catalase, retinol binding protein, peroxiredoxin, ferritin, and carbonic anhydrase were identified as more abundant in controls. At the time of abstract submission, we confirmed elevated concentration of haptoglobin in patients by immuno-turbidometry (8 TEL-AML1, 12 hyperdiploid, 20 non-hyperdiploid or fusion gene ) and compared with 10 control samples. Ferritin concentration, on the other hand, was higher in control samples, as confirmed by Enzyme-linked imunoassay (ELISA).

Presented data provide an insight into leukemic BM environment. The role of the observed differences in soluble factors in leukemia pathogenesis, diagnostic and therapeutic potentialis being investigated. Supported by : GAUK 7543/2007, IGA MZCR NR/9531-3