Expression Level Of A Coagulation Factor Gene, F13A1 Defines Characteristic Subpopulations Of Childhood Acute Lymphoblastic Leukemia

Using multiparameter flow cytometry, Western blot, ELISA and laser scanning microscopy, leukemic B-cell progenitor lymphoblasts were identified as a novel expression site of coagulation factor XIII subunit A (FXIIIA) (Kiss et al Thromb Haemost 2006). Three-year overall survival (OS) of children with FXIIIA-positive acute lymphoblastic leukemia (ALL) was significantly higher (87%) than 3-yr OS of patients with FXIIIA-negative ALL (65%). Here we report on gene expression profiles (GEP) associated with FXIIIA-positive vs. FXIIIA–negative childhood (ch)ALL. Finnally, the expression levels of F13A1 gene in cytogenetic subgroups were investigated.

GEP of 11 FXIIIA-positive and 9 FXIIIA-negative samples of patients treated at the Department of Hematology-Oncology, University of Debrecen was investigated using HG-U 133 plus 2.0 Affymetrix Platform Array. Expression sequence tags (EST) characterized by expression levels with at least 2 logs difference among FXIIIA-positive vs. FXIIIA-negative samples were selected and validated with real-time quantitative PCR. Extending the analysis, we have searched the database containing GEPs of 317 children with ALL treated at the Division of Hemato-Oncology, Department of Women’s and Children’s Health, University of Padova, using R package and Partek Genomic Suite softwares.

FXIIIA-positive and FXIIIA-negative samples exhibited a characteristically different GEP. In the FXIIIA-positive samples one of the overexpressed genes was F13A1. FXIIIA-negative samples contained two characteristic groups of overexpressed genes. One of these consisted of genes participating in B-cell development: EBF1, IKZF1 and PAX5. The second set consisted of genes encoding for tyrosine kinases: JAK1, JAK3, NC07523, NC08002, NC06523, NC08345 and for serine-threonine kinase NC00027. In contrast, FXIIIA-positive samples contained only two overexpressed tyrosine kinases: C-KIT and JAK2.

A wide variation of F13A1 expression levels of the 317 Padova patients was observed. Since expression level of FXIIIA protein of these patients has not been determined, we have arbitrarily defined F13A1 gene expression levels below 10⁶ as “low” and gene expression levels exceeding 10⁹ as “high”. Considering these two groups, we have investigated the distribution of F13A1 expression among the known cytogenetic subgroups of ALL. Low F13A1 expression was prevalent among “B-other” samples, high F13A1 expression was associated with t(1;19).

The pattern of overexpressed ESTs, accumulation of low F13A1 expressing samples in the “B-other” cytogenetic group of ALL and the unfavorable disease outcome of FXIIIA-negative cases may suggest a possible overlap between FXIIIA-negative ALL and BCR-ABL1-like ALL identified recently in the “B-other” group. The nature and extent of this overlap will be investigated prospectively in the BFM ALL-IC 2009 clinical trial. In contrast, high expression levels of F13A1 accumulated preferentially within the t(1;19) genetic group, associated with a good prognosis. Detection of FXIIIA expression by flow cytometry may offer an easy and non-expensive tool for defining new prognostic subpopulations of ALL.