B-cell chronic lymphocytic leukemia (B-CLL) shows an incredibly high heterogeneity in the clinical course, spanning from rapidly aggressive to completely indolent behaviour. In order to identify the correspondent gene expression variability, we investigated 29 cases of untreated B-CLL using microarray. Firstly, two robust CLL clusters were identified applying multiple unsupervised clustering algorithms. Their separation was mainly determined by the differential expression of several genes included into two gene groups designated OxPhos and Lyn clusters. OxPhos gene cluster, previously identified in a subset of diffuse large B-cell lymphomas, comprises genes coding for some respiratory chain enzymes, ribosomal proteins and translation factors. Moreover, increased levels of genes involved in the regulation of apoptosis and in the proteasome-ubiquitin complex and the down-regulation of LYN gene, member of B-cell receptor pathway, characterized the OxPhos CLL subset. These B-CLL biological clusters did not reveal any preferential distribution of Ig mutated or Ig unmutated CLL prognostic groups. Furthermore, we applied another unsupervised algorithm (Subtractive Unsupervised Analysis) after the exclusion of genes characterizing the previously identified CLL subsets. At this point, we could identify two new CLL clusters, showing a clear association to the Ig mutational status. In addition to the ZAP-70 and LPL genes, the patients with unmutated Ig expressed higher level of some interesting genes involved in B cell activation, cell cycle regulation, apoptosis resistance and angiogenesis. In conclusion, we showed that B-CLL is characterized by an intrinsic heterogeneity in gene expression pattern, which overcomes the influence of the immunoglobulin mutational status on B-cell chronic lymphocytic leukemia profiles. Keywords: chronic lymphocytic leukemia, gene expressione profile, immunuglobulin mutational status, clinical heterogeneity
Overall design
Twenty-nine B-CLL patients, who have attended the haematology clinic of Modena Hospital, were enrolled in this study. Total RNA from mononuclear cells was isolated using RNeasy Mini Kit (QIAGEN, Valencia, CA, USA) according to the manufacturer’s instructions For all 29 specimens, fluorescently-labelled cRNA was generated by in vitro transcription using Low RNA Input Fluorescent Linear Amplification Kit (Agilent Technologies, Palo alto, CA, USA) according to the manufacturer’s instructions. Amplified cRNA of each patient was labelled with cyanine 5-CTP (Perkin-Elmer, NEN Life Science, Boston, MA, USA) in each experiment. Mix of equal amount of total RNAs from our cohort was used as reference control in all microarray gene-profiling experiments. Amplified reference cRNA was labelled with cyanine 3-CTP (Perkin-Elmer, NEN Life Science, Boston, MA, USA) in each experiment. Moreover, cRNA products were purified using RNeasy columns (Qiagen). 1 ug of Cy5-labelled cRNA was mixed with the same amount of Cy3-labelled reference cRNA and then mixed cRNAs were fragmented to an average size of ~50-100 nt by incubation at 60°C for 30 min using in situ Hybridization kit-plus (Agilent). Samples were hybridized on Agilent Human 1A Oligo Microarray, ink-jet printed microarray, comprising 20,173 (60-mer) experimentally validated oligonucleotide probes (features).33 After hybridization for 17 h at 60°C, slides were washed according to Agilent SSC protocol instructions and then scanned using a confocal laser scanner (Agilent Technologies). Fluorescence data were analysed with Feature Extraction Software v.7.5 (Agilent Technologies). Log10 ratio of the dye-normalized Cy3 and Cy5 channel signals were calculated, then a p-value and a final error were obtained to establish the significance of each feature.
Supplement 1A CLL: List of 164 genes up-regulated in Cluster II OxPhos relative to Cluster I Lyn. These genes were selected using supervised analysis ( error-weighted one-way ANOVA). The analysis were performed on gene expression data of 27 CLLs divided into the two main Biological Clusters (I and II), previously identified through unsupervised clustering analyses. header descriptions
Supplement 1B CLL: List of 68 genes down-regulated in Cluster II relative to Cluster I. These genes were selected using supervised analysis (error-weighted one-way ANOVA). The analysis were performed on gene expression data of 27 CLLs divided into the two main Biological Clusters, previously identified through unsupervised clustering analyses. header descriptions
Sequence Name(s)
Sequence Description
Data table
Sequence Name(s)
Sequence Description
NM_014948
Homo sapiens likely ortholog of mouse ubiquitin conjugating enzyme 7 interacting protein 5 (UBCE7IP5), transcript variant 1, mRNA
AK026695
Homo sapiens cDNA: FLJ23042 fis, clone LNG02323
AK026323
Homo sapiens cDNA: FLJ22670 fis, clone HSI08684
A_23_P212180
Unknown
NM_017535
Homo sapiens hypothetical protein DKFZp566H0824 (DKFZp566H0824), mRNA
GP2
Homo sapiens glycoprotein 2 (zymogen granule membrane) (GP2), mRNA
BC014951
Homo sapiens chromosome 20 open reading frame 107, mRNA (cDNA clone IMAGE:4849719)
PDZK7
Homo sapiens PDZ domain containing 7 (PDZK7), mRNA
AK000707
Homo sapiens cDNA FLJ20700 fis, clone KAIA2250
A_23_P68137
Unknown
THC1858960
Unknown
ANKK1
Homo sapiens ankyrin repeat and kinase domain containing 1 (ANKK1), mRNA
BC044941
Homo sapiens, Similar to apoptosis related protein, clone IMAGE:5213487, mRNA
NM_032753
Homo sapiens hypothetical protein MGC15631 (MGC15631), mRNA
NM_153338
Homo sapiens hypothetical protein FLJ90165 (FLJ90165), mRNA
AL832180
Homo sapiens mRNA; cDNA DKFZp686P1116 (from clone DKFZp686P1116)
AK098134
Homo sapiens cDNA FLJ40815 fis, clone TRACH2010600
AF116688
Homo sapiens PRO2133 mRNA, complete cds
NM_032391
Homo sapiens small nuclear protein PRAC (PRAC), mRNA
BC034271
Homo sapiens Fanconi anemia, complementation group C, mRNA (cDNA clone IMAGE:4777682), with apparent retained intron
Total number of rows: 68
Table truncated, full table size 4 Kbytes.
Supplement 2: List of genes included in the 78-genes classifier. The classifier was constructed using Bayesian algorithm and it comprises 78 genes that best distinguish Cluster II OxPhos CLL samples from Cluster I Lyn CLL samples. The Confidence/Order column indicates the order of genes from the more significant one to the less one. The Likelihood/T-Score column indicates Fisher Test results (t). header descriptions
Homo sapiens ribosomal protein S15a (RPS15A), mRNA
10
1.3811
SNRPD2
Homo sapiens small nuclear ribonucleoprotein D2 polypeptide 16.5kDa (SNRPD2), transcript variant 2, mRNA
11
1.3687
RPS9
Homo sapiens ribosomal protein S9 (RPS9), mRNA
12
1.3550
CD48
Homo sapiens CD48 antigen (B-cell membrane protein) (CD48), mRNA
13
1.3441
NM_006830
Homo sapiens ubiquinol-cytochrome c reductase (6.4kD) subunit (UQCR), mRNA
14
1.3205
TMEM14B
Homo sapiens transmembrane protein 14B (TMEM14B), mRNA
15
1.3178
PFDN5
Homo sapiens prefoldin 5 (PFDN5), transcript variant 1, mRNA
16
1.3156
LAMR1
Homo sapiens laminin receptor 1 (ribosomal protein SA, 67kDa) (LAMR1), mRNA
17
1.3106
TPT1
Homo sapiens tumor protein, translationally-controlled 1 (TPT1), mRNA
18
1.3104
MYL6
Homo sapiens myosin, light polypeptide 6, alkali, smooth muscle and non-muscle (MYL6), transcript variant 2, mRNA
19
1.3037
Total number of rows: 78
Table truncated, full table size 6 Kbytes.
Supplement 3 CLL. Statistically different genes between Ig mutated and unmutated CLL subsets. These genes were selected using supervised analysis (error-weighted one-way ANOVA). The analysis were performed on gene expression data of 29 CLLs divided into the two groups by virtue of immunoglobulin mutational status, previously identified through direct sequencing. The cut off was 2%. header descriptions
Sequence Name(s)
Expression
Sequence Description
Factor 1 P-Value
Data table
Sequence Name(s)
Expression
Sequence Description
Factor 1 P-Value
HEMGN
up
Homo sapiens hemogen (HEMGN), transcript variant 1, mRNA
1.11E-15
GPR86
down
Homo sapiens G protein-coupled receptor 86 (GPR86), transcript variant 1, mRNA
1.90E-14
AK056531
down
Homo sapiens cDNA FLJ31969 fis, clone NT2RP7008013, highly similar to Mus musculus mRNA for Ten-m4
6.58E-12
A_23_P124442
up
Unknown
3.12E-09
IL10
down
Homo sapiens interleukin 10 (IL10), mRNA
5.77E-09
NPTX1
up
Homo sapiens neuronal pentraxin I (NPTX1), mRNA
2.18E-07
LPL
up
Homo sapiens lipoprotein lipase (LPL), mRNA
3.13E-07
TEAD3
up
Homo sapiens TEA domain family member 3 (TEAD3), mRNA
4.70E-07
PBP
up
Homo sapiens prostatic binding protein (PBP), mRNA
5.96E-07
ANGPT2
up
Homo sapiens angiopoietin 2 (ANGPT2), mRNA
3.02E-06
WSB2
up
Homo sapiens WD repeat and SOCS box-containing 2 (WSB2), mRNA
3.96E-06
CHST6
up
Homo sapiens carbohydrate (N-acetylglucosamine 6-O) sulfotransferase 6 (CHST6), mRNA
4.32E-06
NM_024745
down
Homo sapiens likely ortholog of mouse Shc SH2-domain binding protein 1 (SHCBP1), mRNA
6.35E-06
ESRRBL1
up
Homo sapiens estrogen-related receptor beta like 1 (ESRRBL1), mRNA [NM_018010]
1.00E-05
ZNF288
down
Homo sapiens zinc finger protein 288 (ZNF288), mRNA
4.00E-05
DFNA5
up
Homo sapiens deafness, autosomal dominant 5 (DFNA5), mRNA
5.00E-05
ZAP70
up
Homo sapiens zeta-chain (TCR) associated protein kinase 70kDa (ZAP70), transcript variant 1, mRNA
5.00E-05
ADAM29
down
Homo sapiens a disintegrin and metalloproteinase domain 29 (ADAM29), transcript variant 3, mRNA
5.00E-05
FCGRT
down
Homo sapiens Fc fragment of IgG, receptor, transporter, alpha (FCGRT), mRNA
7.00E-05
NM_024795
down
Homo sapiens hypothetical protein FLJ22800 (FLJ22800), mRNA
Less...
More...