Fluorescent sensors based on bacterial fusion proteins.

Fluorescence proteins are widely used as markers for biomedical and technological purposes. Therefore, the aim of this project was to create a fluorescent sensor, based in the green and cyan fluorescent protein, using bacterial S-layers proteins as scaffold for the fluorescent tag. We report the cloning, expression and purification of three S-layer fluorescent proteins: SgsE-EGFP, SgsE-ECFP and SgsE-13aa-ECFP, this last containing a 13-amino acid rigid linker. The pH dependence of the fluorescence intensity of the S-layer fusion proteins, monitored by fluorescence spectroscopy, showed that the ECFP tag was more stable than EGFP. Furthermore, the fluorescent fusion proteins were reassembled on silica particles modified with cationic and anionic polyelectrolytes. Zeta potential measurements confirmed the particle coatings and indicated their colloidal stability. Flow cytometry and fluorescence microscopy showed that the fluorescence of the fusion proteins was pH dependent and sensitive to the underlying polyelectrolyte coating. This might suggest that the fluorescent tag is not completely exposed to the bulk media as an independent moiety. Finally, it was found out that viscosity enhanced the fluorescence intensity of the three fluorescent S-layer proteins.

Biomaterial and cellular properties as examined through atomic force microscopy, fluorescence optical microscopies and spectroscopic techniques.

Colloids, polymers, gels, and biological materials are widely used for numerous technological applications. The design, fabrication, and understanding of the physico-chemical properties of such (bio)materials, however, represent a challenge for scientists and technologists. This review is a concise update of the latest achievements in surface and bulk analytical techniques applied to biomaterials and soft matter systems. Specific emphasis is devoted to their structural, mechanical, surface, and chemical properties. The review introduces the basics of atomic force microscopy (AFM) and discusses its combination with optical microscopies and spectroscopic techniques. In particular, we report the AFM combination with fluorescence, confocal microscopy, stimulated emission depletion, Raman, and infrared spectroscopy. The review highlights the impact and the potential applications of the combination of such experimental techniques on current problems and questions related to material and polymer science, cell biology, biochemical engineering, and protein chemistry. This review is an accessible introduction to AFM for the newcomer and a source of new experimental information for the experienced researcher.

2D crystalline protein layers as immobilization matrices for the development of DNA microarrays.

There is a growing demand for functional layers for the immobilization of (bio)molecules on different kinds of substrates in the field of biosensors, microarrays, and lab-on-a-chip development. These functional coatings should have the ability to specifically bind (bio)molecules with a high binding efficiency, while showing low unspecific binding during the following assay. In this paper we present rSbpA surface layer proteins (S-layer proteins) as a versatile immobilization layer for the development of DNA microarrays. S-layer proteins show the ability to reassemble into two-dimensional arrays on solid surfaces and their functional groups, such as carboxylic groups, are repeated with the periodicity of the lattice, allowing for immobilization of other (bio)molecules. Different fluorescently labeled amino functionalized DNA oligomers were covalently linked to the S-layer matrices to allow the characterization of DNA binding on S-layers. Hybridization and dissociation of DNA-oligomers were studied on S-layer coated slides, revealing low levels of unspecific adsorption of DNA on S-layer based immobilization matrices. In the following the principle was transferred to a DNA microarray design showing successful spotting and hybridization on whole microarray slides. Besides common laser scanning for fluorescence detection, S-layer based microarrays were evaluated with a compact, low cost platform for direct fluorescence imaging based on surface plasmon enhanced fluorescence excitation. It could be shown that S-layer protein layers are promising as immobilization matrices for the development of biosensors and microarrays.

A fusion tag to fold on: the S-layer protein SgsE confers improved folding kinetics to translationally fused enhanced green fluorescent protein.

Genetic fusion of two proteins frequently induces beneficial effects to the proteins, such as increased solubility, besides the combination of two protein functions. Here, we study the effects of the bacterial surface layer protein SgsE from Geobacillus stearothermophilus NRS 2004/3a on the folding of a C-terminally fused enhanced green fluorescent protein (EGFP) moiety. Although GFPs are generally unable to adopt a functional confirmation in the bacterial periplasm of Escherichia coli cells, we observed periplasmic fluorescence from a chimera of a 150-amino-acid N-terminal truncation of SgsE and EGFP. Based on this finding, unfolding and refolding kinetics of different S-layer-EGFP chimeras, a maltose binding protein-EGFP chimera, and sole EGFP were monitored using green fluorescence as indicator for the folded protein state. Calculated apparent rate constants for unfolding and refolding indicated different folding pathways for EGFP depending on the fusion partner used, and a clearly stabilizing effect was observed for the SgsE_C fusion moiety. Thermal stability, as determined by differential scanning calorimetry, and unfolding equilibria were found to be independent of the fused partner. We conclude that the stabilizing effect SgsE_C exerts on EGFP is due to a reduction of degrees of freedom for folding of EGFP in the fused state.

Overexpression of G protein-coupled receptor 5D in the bone marrow is associated with poor prognosis in patients with multiple myeloma.

BACKGROUND: G protein-coupled receptor 5D (GPRC5D) is a novel surface receptor. As this new subtype of G protein-coupled receptors was discovered, little is known about the role of this gene. MATERIALS AND METHODS: In this retrospective study, we investigated GPRC5D mRNA expression by real-time polymerase chain reaction (RT-PCR) in bone marrow (BM) of 48 patients with multiple myeloma (MM). RESULTS: Highly variable levels of GPRC5D (median, 288; quartiles, 17-928) were detected in patients with MM, whereas only low expression was detected in normal tissues (median, 1; quartiles, 1-23). High mRNA expression of GPRC5D correlated positively with high plasma cell count in bone marrow (r = 0·64, P < 0·001), high ß(2) -microglobulin (r = 0·42, P = 0·003) and poor-risk cytogenetics: deletion 13q14 (rb-1), P = 0·003; and 14q32 translocation t(4;14)(p16;q32), P = 0·029. GPRC5D mRNA expression showed a significant correlation with overall survival (P = 0·031). The estimated overall survival of patients expressing GPRC5D above or below the median of 288 was 43·9% vs. 70·2% at 48 months. Here, we report, for the first time, the association of GPRC5D expression and cancer. CONCLUSIONS: Overexpression in poor-risk myeloma, low expression in normal tissues and cell surface expression identify GPRC5D as a potential novel cancer antigen. Our data demonstrate that GPRC5D is a prognostic factor in MM correlating with other major risk factors.

Fluorescence energy transfer in the bi-fluorescent S-layer tandem fusion protein ECFP-SgsE-YFP.

This work reports for the first time on the fabrication of a bi-functional S-layer tandem fusion protein which is able to self-assemble on solid supports without losing its functionality. Two variants of the green fluorescent protein (GFP) were genetically combined with a self-assembly system having the remarkable opportunity to interact with each other and act as functional nanopatterning biocoating. The S-layer protein SgsE of Geobacillus stearothermophilus NRS 2004/3a was fused with the cyan ECFP donor protein at the SgsE N-terminus and with the yellow YFP acceptor protein at the C-terminus. The fluorescence energy transfer was studied with spectrofluorimetry, confocal microscopy and flow cytometry, whilst protein self-assembly (on silicon dioxide particles) and structural investigations were carried out with atomic force microscopy (AFM). The fluorescence resonance energy transfer efficiency of reassembled SgsE tandem protein was 20.0 ± 6.1% which is almost the same transfer efficiency shown in solution (19.6 ± 0.1%). This work shows that bi-fluorescent S-layer fusion proteins self-assemble on silica particles retaining their fluorescent properties.

Cell surface display of chimeric glycoproteins via the S-layer of Paenibacillus alvei.

The Gram-positive, mesophilic bacterium Paenibacillus alvei CCM 2051(T) possesses a two-dimensional crystalline protein surface layer (S-layer) with oblique lattice symmetry composed of a single type of O-glycoprotein species. Herein, we describe a strategy for nanopatterned in vivo cell surface co-display of peptide and glycan epitopes based on this S-layer glycoprotein self-assembly system. The open reading frame of the corresponding structural gene spaA codes for a protein of 983 amino acids, including a signal peptide of 24 amino acids. The mature S-layer protein has a theoretical molecular mass of 105.95kDa and a calculated pI of 5.83. It contains three S-layer homology domains at the N-terminus that are involved in anchoring of the glycoprotein via a non-classical, pyruvylated secondary cell wall polymer to the peptidoglycan layer of the cell wall. For this polymer, several putative biosynthesis enzymes were identified upstream of the spaA gene. For in vivo cell surface display, the hexahistidine tag and the enhanced green fluorescent protein, respectively, were translationally fused to the C-terminus of SpaA. Immunoblot analysis, immunofluorescence staining, and fluorescence microscopy revealed that the fused epitopes were efficiently expressed and successfully displayed via the S-layer glycoprotein matrix on the surface of P. alvei CCM 2051(T) cells. In contrast, exclusively non-glycosylated chimeric SpaA proteins were displayed, when the S-layer of the glycosylation-deficient wsfP mutant was used as a display matrix.

Absorption, steady-state fluorescence, fluorescence lifetime, and 2D self-assembly properties of engineered fluorescent S-layer fusion proteins of Geobacillus stearothermophilus NRS 2004/3a.

S-layer fusion protein technology was used to design four different fluorescent fusion proteins with three different GFP mutants and the red fluorescent protein mRFP1. Their absorption spectra, steady-state fluorescence, and fluorescence lifetime were investigated as a function of pH. It was found that fluorescence intensities and lifetime of the GFP mutant S-layer fusion proteins decreased about 50% between pH 6 and pH 5. The spectral properties of the red S-layer fusion protein were minimally affected by pH variations. These results were compared with His-tagged reference fluorescent proteins, demonstrating that the S-layer protein did not change the general spectral properties of the whole fusion protein. In addition, the pK(a) values of the fluorescent S-layer fusion proteins were calculated. Finally, it was shown that the S-layer fusion proteins were able to self-assemble forming 2D nanostructures of oblique p2 symmetry with lattice parameters of about a = 11 nm, b = 14 nm, and gamma = 80 degrees . The fluorescence tag did not hinder the natural self-assembly process of the S-layer protein. The combination of the fluorescence properties and the self-assembly ability of the engineered fusion proteins make them a promising tool to generate biomimetic surfaces for future applications in nanobiotechnology at a wide range of pH.

Optical oxygen sensors based on Pt(II) porphyrin dye immobilized on S-layer protein matrices.

This paper describes the development of planar and fiber optic oxygen sensors utilizing surface layer (S-layer) proteins as immobilization matrix for oxygen sensitive dyes. S-layer proteins have the intrinsic capability to reassemble into two-dimensional arrays in suspension and at interfaces. Due to their crystalline character the distribution of functional groups, such as carboxylic groups, is repeated with the periodicity of the lattice and thus allows the reproducible and geometrically distinct binding of functional molecules. For the development of oxygen sensors an oxygen sensitive Pt(II) porphyrin dye was covalently bound to the S-layer matrix. Measurement of the oxygen concentration was performed by phase modulation fluorimetry. Setups comprising low cost optoelectronic components like LEDs and silicon photodiodes were constructed. For both sensor setups (planar and fiber optic) variations in the oxygen concentrations resulted in distinct and reproducible changes in luminescence lifetime and intensity. The luminescence quenching efficiency of these sensors was found to be 1.5-1.9 (expressed as the ratio of signal under nitrogen and air) which compares well to other sensor systems using luminophores embedded in polymer matrices. These results demonstrated the application potential of S-layers as immobilization matrices in the development of (bio-)sensors.

Genetic engineering of the S-layer protein SbpA of Lysinibacillus sphaericus CCM 2177 for the generation of functionalized nanoarrays.

The mesophilic organism Lysinibacillus sphaericus CCM 2177 produces the surface (S)-layer protein SbpA, which after secretion completely covers the cell surface with a crystalline array exhibiting square lattice symmetry. Because of its excellent in vitro recrystallization properties on solid supports, SbpA represents a suitable candidate for genetically engineering to create a versatile self-assembly system for the development of a molecular construction kit for nanobiotechnological applications. The first goal of this study was to investigate the surface location of 3 different C-terminal amino acid positions within the S-layer lattice formed by SbpA. Therefore, three derivatives of SbpA were constructed, in which 90, 173, or 200 C-terminal amino acids were deleted, and the sequence encoding the short affinity tag Strep-tag II as well as a single cysteine residue were fused to their C-terminal end. Recrystallization studies of the rSbpA/STII/Cys fusion proteins indicated that C-terminal truncation and functionalization of the S-layer protein did not interfere with the self-assembly capability. Fluorescent labeling demonstrated that the orientation of the crystalline rSbpA(31-1178)/STII/Cys lattice on solid supports was the same, like the orientation of wild-type S-layer protein SbpA on the bacterial cell. In soluble and recrystallized rSbpA/STII/Cys fusion proteins, Strep-tag II was used for prescreening of the surface accessibility, whereas the thiol group of the end-standing cysteine residue was exploited for site-directed chemical linkage of differently sized preactivated macromolecules via heterobifunctional cross-linkers. Finally, functionalized two-dimensional S-layer lattices formed by rSbpA(31-1178)/STII/Cys exhibiting highly accessible cysteine residues in a well-defined arrangement on the surface were utilized for the template-assisted patterning of gold nanoparticles.

Recombinant glycans on an S-layer self-assembly protein: a new dimension for nanopatterned biomaterials.

Crucial biological phenomena are mediated through carbohydrates that are displayed in a defined manner and interact with molecular scale precision. We lay the groundwork for the integration of recombinant carbohydrates into a "biomolecular construction kit" for the design of new biomaterials, by utilizing the self-assembly system of the crystalline cell surface (S)-layer protein SgsE of Geobacillus stearothermophilus NRS 2004/3a. SgsE is a naturally O-glycosylated protein, with intrinsic properties that allow it to function as a nanopatterned matrix for the periodic display of glycans. By using a combined carbohydrate/protein engineering approach, two types of S-layer neoglycoproteins are produced in Escherichia coli. Based on the identification of a suitable periplasmic targeting system for the SgsE self-assembly protein as a cellular prerequisite for protein glycosylation, and on engineering of one of the natural protein O-glycosylation sites into a target for N-glycosylation, the heptasaccharide from the AcrA protein of Campylobacter jejuni and the O7 polysaccharide of E. coli are co- or post-translationally transferred to the S-layer protein by the action of the oligosaccharyltransferase PglB. The degree of glycosylation of the S-layer neoglycoproteins after purification from the periplasmic fraction reaches completeness. Electron microscopy reveals that recombinant glycosylation is fully compatible with the S-layer protein self-assembly system. Tailor-made ("functional") nanopatterned, self-assembling neoglycoproteins may open up new strategies for influencing and controlling complex biological systems with potential applications in the areas of biomimetics, drug targeting, vaccine design, or diagnostics.

Myeloid leukemias express a broad spectrum of VEGF receptors including neuropilin-1 (NRP-1) and NRP-2.

Vascular endothelial growth factor (VEGF) is produced in neoplastic cells in various myeloid neoplasms and may act as an autocrine growth-regulator. We have examined the expression of five VEGF receptors (VEGR1/Flt-1, VEGFR2/KDR, Flt-4, neuropilin-1 = NRP-1, NRP-2) in leukemic cells obtained from patients with acute myeloid leukemia (n = 28), chronic myeloid leukemia (n = 14), chronic eosinophilic leukemia (n = 3), chronic myelomonocytic leukemia (n = 9), or mast cell leukemia/systemic mastocytosis (n = 3) as well as in respective cell lines. Expression of VEGFR mRNA was analyzed by RT-PCR, and expression of VEGFR protein by immunocytochemistry. In most patients, leukemic cells expressed NRP-1 mRNA and NRP-2 mRNA independent of the type of disease. By contrast, transcripts for Flt-1, KDR, and Flt-4 were expressed variably without a clear correlation to the type of leukemia. Expression of VEGF receptors was also demonstrable at the protein level in all cases tested. In conclusion, neoplastic cells in myeloid leukemias frequently express VEGFR including NRP-1 and NRP-2.

Overexpression of the paternally expressed gene 10 (PEG10) from the imprinted locus on chromosome 7q21 in high-risk B-cell chronic lymphocytic leukemia.

We report high expression of the maternally imprinted gene PEG10 in high-risk B-CLL defined by high LPL mRNA expression. Differential expression was initially identified by microarray analysis and confirmed by real time PCR in 42 B-CLL patients. mRNA expression ranged from 0.3- to 375.4-fold compared to normal peripheral blood mononuclear cells (PBMNC). Expression levels in CD19+ B-CLL cells were 100-fold higher than in B-cells from healthy donors. PEG10 expression levels in B-CLL patient samples remained stable over time even after chemotherapy. High PEG10 expression correlated with high LPL expression (p=0.001) and a positive Coombs' test (p=0.04). Interestingly, similar expression patterns were observed for the neighbouring imprinted gene sarcoglycan-epsilon (SGCE). Monoallelic expression and maintained imprinting of PEG10 were found by allele- or methylation-specific PCR. The intensity of intracellular staining of PEG10 protein corresponded to mRNA levels as confirmed by immunofluorescence staining. Short term knock-down of PEG10 in B-CLL cells and HepG2 cells was not associated with changes in cell survival but resulted in a significant change in the expression of 80 genes. However, long term inhibition of PEG10 led to induction of apoptosis in B-CLL cells. Our data indicate (i) a prognostic value of PEG10 in B-CLL patients; (ii) specific deregulation of the imprinted locus at 7q21 in high-risk B-CLL; (iii) a potential functional and biological role of PEG10 protein expression. Altogether, PEG10 represents a novel marker in B-CLL.

Upregulation of retinoic acid receptor-beta by the epidermal growth factor-receptor inhibitor PD153035 is not mediated by blockade of ErbB pathways.

Inhibiting epidermal growth factor-receptor (ErbB-1) represents a powerful anticancer strategy. Activation of retinoid pathways is also in development for cancer treatment. Retinoic acid receptor-beta-the tumor suppressor and main retinoid mediator--is silenced in many tumors. The ErbB-1 inhibitor PD153035 cooperates with retinoic acid during growth inhibition and induces retinoic acid receptor-beta suggesting that ErbB-1 controls retinoic acid receptor-beta. However, here we demonstrate that ErbB pathways are not involved in PD153035-mediated retinoic acid receptor-beta-upregulation. PD153035 inhibits ErbB-1-phosphorylation, whereas its derivative EBE-A22 is inactive. Yet both inhibit cell growth and upregulate retinoic acid receptor-beta in ErbB-1-overexpressing (MDA-MB-468), moderately expressing (OVCAR-3), ErbB-1-negative (MDA-MB-453) or ErbB-negative cells (CEM, Jurkat). Both bind DNA, whereas the closely related ErbB-1 inhibitors AG1478 and ZD1839, which are inactive on retinoic acid receptor-beta, do not significantly bind DNA. None of the other ErbB-1/ErbB-2 inhibitors tested (RG-14620, LFM-A12, AG879, AG825) affect retinoic acid receptor-beta. PD153035 decreases methylation of the retinoic acid receptor-beta2 promoter. In OVCAR-3, it stimulates dislodgement of histone deacetylase 1 from the promoter and acetylation of histones H3 and H4. Consequently, PD153035 facilitates recruitment of RNA polymerase II to the promoter and stimulates transcriptional activity. Moreover, PD153035 increases the retinoic acid receptor-beta mRNA half-life. No other retinoid receptor, nor estrogen receptor-alpha, nor RASSF1A is upregulated by PD153035. Thus PD153035 induces retinoic acid receptor-beta by ErbB-independent transcriptional and post-transcriptional mechanisms. This report highlights a triple action for an ErbB-1 inhibitor (ErbB-1 inhibition, DNA intercalation, retinoic acid receptor-beta-induction). Such multitargeting drugs bear great potential for cancer treatment.

Limited value of FLT3 mRNA expression in the bone marrow for prognosis and monitoring of patients with acute myeloid leukemia.

We studied wild-type FLT3 mRNA expression at diagnosis in bone marrow samples from 85 patients with acute myeloid leukemia (AML), 23 of whom were in complete remission, and determined its utility as a marker for minimal residual disease (MRD). We conclude that FLT3 expression is of limited value as a prognostic marker and for MRD monitoring.

An EGF receptor inhibitor induces RAR-beta expression in breast and ovarian cancer cells.

Inhibition of the epidermal growth factor (EGF)-receptor (EGFR) has become a promising anticancer treatment strategy. In addition, application of retinoids yields encouraging results for cancer prevention and therapy. Many tumors express no or low amounts of retinoic acid receptor-beta2 (RAR-beta2) due to epigenetic silencing via DNA hypermethylation. RAR-beta2 is the main mediator of the antiproliferative effect of retinoids. RAR-beta2 re-expression causes reversal of transformation, cell cycle arrest, and restoration of retinoid sensitivity. RAR-beta2 is thus a tumor suppressor. Western blotting, colorimetric in vitro cell proliferation assays, and reverse transcription-polymerase chain reaction demonstrated that the EGFR inhibitor PD153035 not only blocked activation of EGFR and inhibited cell growth, but also stimulated RAR-beta expression in MDA-MB-468 breast and OVCAR-3 ovarian carcinoma cells. Upregulation of RAR-beta by PD153035 was confirmed by real-time reverse transcription-polymerase chain reaction. In contrast, expression of other retinoid receptors and of estrogen receptor-alpha was not affected. PD153035-mediated re-induction of RAR-beta was associated with demethylation of the RAR-beta2 gene promoter P2 as demonstrated by methylation-specific polymerase chain reaction. These novel results on the ErbB/retinoid receptor cross-talk may be useful for designing future anticancer combination regimens.

Highly refractory acute myeloid leukemia.

In this study we evaluated 103 patients suffering from acute myeloid leukemia (AML) who did not respond to induction chemotherapy and defined a sub-group of patients with highly refractory disease characterized by a persistence of more than 1 G/L blast cells in the peripheral blood between days 12 and 16 of the first induction cycle. Only seven patients (one female, six males) met these criteria. Their median age was 65 years (range 41-82 years). Four had de novo AML and three secondary AML. Cytogenetic analysis was performed in six patients: complex aberrations were detected in four patients and, unexpectedly, normal karyotypes were found in the other two. Analysis of multidrug-resistance factors revealed high co-expression of P-glycoprotein (P-gp) and lung resistance protein (LRP) in all four patients with highly refractory disease tested a finding in only 6% of patients with refractory disease and 3% of patients who achieved complete remission (CR) of disease. Furthermore, patients with highly refractory AML had substantially higher leukocyte counts than patients with refractory AML or CR, although this was not significant statistically. Overall, patients with highly refractory AML are characterized by a high incidence of complex cytogenetic aberrations and marked expression of drug transporters.

Relation of in vitro growth characteristics to cytogenetics and treatment outcome in acute myeloid leukemia: prognostic significance in patients with a normal karyotype.

We analyzed in vitro growth characteristics of bone marrow mononuclear cells (BMMCs) from 322 patients with acute myeloid leukemia (AML) in relation to cytogenetic abnormalities. Median colony growth was low in each of the cytogenetic changes associated with a favorable outcome. Most karyotypic abnormalities in the intermediate prognosis group were associated with low growth potential, but 11 q23 abnormalities exhibited 8 times higher in vitro growth. Cytogenetic changes that included abn(3q) seemed to display the highest colony growth in the unfavorable prognosis group, whereas isolated -7 may have been associated with limited growth potential. In vitro growth behavior was predictive of neither rate of complete remission (CR) nor survival of AML patients within the 3 cytogenetic risk groups. In contrast, colony growth differed significantly in the subgroup of patients with a normal karyotype who achieved remission with induction treatment and those who had no remission (10 versus 81.5/10(5) BMMCs; P = .015). Significantly more patients with normal cytogenetics and colony growth below the 50th percentile went into CR than did patients with colony growth above the 50th percentile (82.8% versus 71.2%). Only 4 (6.8%) of the patients in the low growth group had no remission, compared with 12 (23.1%) of the patients with higher in vitro growth (P = .031, chi-square test). In conclusion, colony growth may prove useful as a prognostic factor for early treatment failure in AML patients with a normal karyotype.

Variable prognostic value of FLT3 internal tandem duplications in patients with de novo AML and a normal karyotype, t(15;17), t(8;21) or inv(16).

INTRODUCTION: Internal tandem duplication of the FLT3 gene (FLT3/ITD) has been linked to poor outcome in acute myeloid leukemia (AML). However, the prognostic value of FLT3/ITD in various cytogenetic risk groups is still a matter of debate. The aim of this study was to evaluate the prognostic significance in patients with de novo AML and a normal karyotype or a t(15;17), t(8;21) or inv(16) (good risk group). PATIENTS AND METHODS: Diagnostic samples of 100 patients were investigated by single-step PCR of exons 11 and 12 of the FLT3 gene in a single center retrospective analysis. Subgroups included 53 patients with normal karyotype, 21 patients with t(15;17), 9 patients with t(8;21) and 17 patients with inv(16). RESULTS: FLT3/ITD was found in 26 out of 100 patients: 30% of patients with a normal karyotype and 38% of t(15;17) patients tested positive. The complete remission (CR) rates for the ITD(+) or ITD(-) groups were 50 vs 76% in normal karyotypes, and 100 vs 53% in t(15;17) patients, while the relapse rates were 75 vs 25% in normal karyotypes and 50 vs 42% in t(15;17) patients. Overall survival (OS) and disease free survival (DFS) at 5 years were significantly different in patients with normal cytogenetics: ITD(+) vs ITD(-): OS 6 vs 28% (P<0.003); DFS 13 vs 41% (P<0.02) Interestingly, FLT3/ITD had no significant effect on the outcome of t(15;17) patients: ITD(+) vs ITD(-): OS 85 vs 53% (P=0.056), DFS 45 vs 60% (P=0.6) at 50 months. CONCLUSIONS: These data suggest a high prognostic value of FLT3/ITD in patients with normal cytogenetics. However, we find no evidence that FLT3/ITD is a predictive marker for patients with t(15;17).