Niche-modulated and niche-modulating genes in bone marrow cells

Abstract

Bone marrow (BM) cells depend on their niche for growth and survival. However, the genes modulated by niche stimuli have not been discriminated yet. For this purpose, we investigated BM aspirations from patients with various hematological malignancies. Each aspirate was fractionated, and the various samples were fixed at different time points and analyzed by microarray. Identification of niche-modulated genes relied on sustained change in expression following loss of niche regulation. Compared with the reference ('authentic') samples, which were fixed immediately following aspiration, the BM samples fixed after longer stay out-of-niche acquired numerous changes in gene-expression profile (GEP). The overall genes modulated included a common subset of functionally diverse genes displaying prompt and sustained 'switch' in expression irrespective of the tumor type. Interestingly, the 'switch' in GEP was reversible and turned 'off-and-on' again in culture conditions, resuming cell-cell-matrix contact versus respread into suspension, respectively. Moreover, the resuming of contact prolonged the survival of tumor cells out-of-niche, and the regression of the 'contactless switch' was followed by induction of a new set of genes, this time mainly encoding extracellular proteins including angiogenic factors and extracellular matrix proteins. Our data set, being unique in authentic expression design, uncovered niche-modulated and niche-modulating genes capable of controlling homing, expansion and angiogenesis.

Figure 1

Sustained ‘switch' in GEP following BM aspiration. BM samples from 10 patients with MM, 4 with AML and 1 case with DLBCL (BCL) were fractionated and fixed (either by flash freezing or using RNAlater) at different time points following aspiration. The samples were analyzed by microarray, and the common changes in gene expression (⩾2-folds) are displayed as fold changes relative to the signals recorded from the corresponding reference samples (fixed immediately following aspiration).

Figure 2

Induction of AREG. The signal intensities are depicted against the time to RNA fixation following BM aspirations obtained from patients with MM (cases A–J), leukemic phase of DLBCL (case B′) and AML (cases A′, C′–E′) or alternatively following mobilization from the bottom well into suspension of cultured BM cells from cases B and C (see details in the next section).

Figure 3

The ‘switch' in GEP is contact dependent. Cultured BM cells (CC) were flash frozen at different time points following their mobilization from the bottom well into suspension. Pleural fluid (PF) cell sediment containing pure MM cell population was split with one part left in the bottom tube and the second was transferred into another tube and maintained in suspension after redilution with pleural fluid. Both parts of the sediment were flash frozen at 270 min after the split. The frozen samples were analyzed by microarray, and the differences in signal intensities are displayed as fold changes relative to the GEPs recorded from the reference samples (fixed immediately following mobilization into suspension). Unannotated genes are denoted by their Affymetrix probe set ID number. The results showed similar changes in GEP following cell mobilization and maintenance in suspension as evolved following BM aspiration, indicating that the switch in GEP is related to the loss of cell–cell–matrix contact relationships.

Figure 4

‘Switch' in GEP during remission versus before treatment. A comparison between the top post-aspiration changes in GEP (⩾2-folds) evolved in the BM samples from case B during complete remission (CR) (right column) versus the changes recorded before treatment (MM). The results are displayed as fold changes relative to corresponding reference samples (fixed immediately following aspiration). The lower field demonstrates the genes modulated in the pretreatment samples only. The upper field demonstrates overlapping genes modulated in both disease stages. Unannotated genes are denoted by their Affymetrix probe set ID number.

Figure 5

Niche-modulating genes. The top induced genes in cultured BM samples from three patients with MM are listed . The changes were recorded after culture for 1, 7 and 29 days and are presented as fold changes (green) relative to the signal intensities recorded from corresponding fresh BM samples flash frozen immediately following aspiration. The genes induced during culture were totally different from the ‘contact-modulated genes' except for IL8, which was also induced following aspiration.