Supporting Information

Files in this Data Supplement:

Fig. 6. Nucleotide sequence of human BIC cDNA and associated sequences. The 5' end of BIC RNA corresponds to position 1 (ref. 1 and W.T., unpublished data), and locations and sizes of introns present in the primary transcript are indicated, as is the region corresponding to mature-length miR-155 (dotted line beginning at position 308). Filled arrows A, B, and C denote the oligonucleotides used in RT-PCR assays, with primer pairs A/C and B/C detecting spliced BIC RNA and unspliced pre-BIC RNA (Figs. 1A and 2A), respectively. The shaded nucleotides represent sequences detected by Invader reactions (positions 196–237 and 1019–1070 for BIC and 308–329 for miR-155). Two polyadenylation signals that generate two isoforms of BIC RNA (1) are indicated in bold italics. Open arrows denote the ends of the BIC exon 3 sequence cloned into pcDNA3.BIC and subsequently expressed to generate miR-155 in transfected HEK-293T cells (Fig. 1C). At the bottom are shown the sequence and predicted secondary structure of pre-miR-155. A mature 22-nt miR-155 sequence is in bold.

Fig. 7. Quantification of miR-155 compared with U6 RNA in a fixed amount (40 ng) of total cellular RNA. Biplex formats (1, 2) of Invader assay were used for detection of miR-155 and U6 RNA (see Table 1 for assay oligonucleotides). Data are plotted by using net signal values (background subtracted by using the tRNA signal). Samples include a subset of the samples analyzed in Fig. 3A plus HEK-293T (293T –/+ transfection with pcDNABIC) and two clinical DLBCL samples (sample D6 from Fig. 3C and sample D21, which is not a de novo case). Note that miR-155 levels in 293+BIC were not corrected for the ≈30% transfection efficiency. Error bars represent one standard deviation. Significant variations in expression of U6 RNA (>5 fold) were observed in the different cell types, likely because of differences in the activity of RNA polymerase III in transformed cells (3, 4). Thus, to normalize micro RNA expression, we assumed that each cell contained 20 pg of total RNA; we recognize that cell size and physiology can affect this number.

1. Eis, P. S. (2002) Nat. Biotechnol. 20, 307.

2. Eis, P. S., Olson, M. C., Takova, T., Curtis, M. L., Olson, S. M., Vener, T. I., Ip, H. S., Vedvik, K. L., Bartholomay, C. T., Allawi, H. T., et al. (2001) Nat. Biotechnol. 19, 673–676.

3. Hirsch, H. A., Jawdekar, G. W., Lee, K. A., Gu, L. & Henry, R. W. (2004) Mol. Cell. Biol. 24, 5989–5999.

4. White, R. J. (2004) Oncogene 23, 3208–3216.

Fig. 8. Structure and specificity of the miR-155 Invader assay. (A) The Invader oligo and probe contained 2'-O-methylated hairpins (blue sequence) at their 5' and 3' ends, respectively (1). Coaxial stacking with both ends of miR-155 (red sequence) stabilizes the interaction. The overlap structure (green box) formed upon binding of the Invader oligo and probe to miR-155 enables cleavage (vertical arrow) of the probe’s noncomplementary 5' arm, which is detected in a subsequent FRET-based Invader reaction (2, 3). (B) Disruption of the structure shown in A by an internal hairpin in BIC RNA, which encodes the ≈60-nt pre-miR-155 (Fig. 6). (C) Specificity of the miR-155 assay for mature miR-155 over its precursors. A 1,488-nt in vitro BIC transcript was detected only when it was present in 100- to 1,000-fold higher levels than those required for detection of miR-155.

1. Allawi, H. T., Dahlberg, J. E., Olson, S., Lund, E., Olson, M., Ma, W. P., Takova, T., Neri, B. P. & Lyamichev, V. I. (2004) RNA 10, 1153–1161.

2. Eis, P. S. (2002) Nat. Biotechnol. 20, 307.

3. Eis, P. S., Olson, M. C., Takova, T., Curtis, M. L., Olson, S. M., Vener, T. I., Ip, H. S., Vedvik, K. L., Bartholomay, C. T., Allawi, H. T., et al. (2001) Nat. Biotechnol. 19, 673–676.

Fig. 9. Possible targets of miR-155 in conserved sequences in the 3' UTR regions of mRNAs of humans, mice, and chickens. Numbers next to the mRNA sequences indicate their locations in the 3' UTRs. Note that species-specific differences in the sequences of these RNAs do not appear to disrupt the interactions. (A) PU.1 mRNAs. The interaction of miR-155 with human PU.1 mRNA was proposed in table S2 of John et al. (1). The mRNA of pig PU.1 has a comparable target in its 3' UTR, but the sequence of the pig miR-155 has not yet been reported, so a potential interaction is not shown here. Contrary to a prediction made in ref. 1, we were unable to find any target for miR-155 in the 3' UTR of the mRNA for GAB1. (B) C/EBP/b mRNA, which, like PU.1 mRNA, encodes a developmentally regulated transcription factor (2). The interaction between miR-155 and a sequence in the 3' UTR of human C-CEB/b mRNA is as proposed by Lewis et al. (3).

1. John, B., Enright, A. J., Aravin, A., Tuschl, T., Sander, C. & Marks, D. S. (2004) PLoS Biol 2, e363.

2. Xie, H.,Ye, M., Feng, R. & Graf, T. (2004) Cell 117, 663–676.

3. Lewis, B. P., Burge, C. B. & Bartel, D. P. (2005) Cell 120, 15–20.

Fig. 10. Summary of quantification of miR-15a, miR-16, and let-7a in the immortalized lymphoma cells lines that were analyzed for BIC RNA and miR-155 in Fig. 3A (DLBCL, diffuse large B cell lymphoma; HL, Hodgkin lymphoma). Note that all of these cultured cells, particularly the Hodgkin cell lines, had lower levels of miR-15a, miR-16, and let-7a RNA than did normal circulating B cells (Fig. 4 and Table 2).