Supplement
Fig. S1. p53 primer location and sequences
A. Forward and reverse primer locations for p53 (red), p53β/γ (purple), and Δ40p53 (black) and predicted PCR products
We used primers published in a previous study to search for alternatively spliced transcripts encoding Δ40p53 (1).The forward primer is in an intron II sequence of 118bp unique to Δ40p53. (Most Δ40p53 is generated by utilizing an alternative translation start site in the p53 mRNA.) For the nested PCR used to detect other low abundance p53 isoforms, we used the p53F/R primers (red) for the first round of amplification. These primers hybridize to common exons generated from the upstream promoter and can amplify p53, as well as p53β, p53γ, 40p53, and p53. Thep53β,γF and p53βR or p53γR (purple) primers used in the second round of amplification hybridize to unique sequences in 3' end of exon 9 that result from alternative splicing and will specifically amplify either p53β or γ transcripts, respectively. The second round of amplification also contains p53F/R primers at 1/10th the original concentration, which are carried over with the PCR products of the first round of amplification. We used the primer sequences and protocol for the nested PCR reactions developed by Bourdon and colleagues (2). Predicted product sizes are given for each amplicon. (B) Primer sequences for p53, p53β, and p53γ.
Fig. S2. Analysis of p53 isoforms by RT-PCR
A. 40p53-specific transcripts
Using the 40p53 primers shown in Fig. S1, we executed 192 unique PCR settings in effort to amplify the predicted 360bp PCR product in MCF7 cells, which are known to express Δ40p53. We did detect a correctly sized PCR product in MCF7 RNA in the presence of reverse transcriptase (+RT) that was not present in the absence of reverse transcriptase (-RT) or input cDNA (H2O), or in MCF7 DNA (genomic). However, the sequence of this amplicon did not match Δ40p53 or other transcripts of the p53 family, as shown in the bottom of this panel. We took a closer look at the sequence of this alternative transcript using NCBI Blast and found that, in addition to Δ40p53, the internal repeats found in intron II have homology to over 3000 targets. This can account for the numerous amplicons we saw when we used suboptimal PCR conditions, as well as for the non-p53 sequence in the 360bp amplicon detected in MCF7 cells, as shown here. Based on these results, we think that alternatively spliced transcripts play only a minor role, if any, in generating the Δ40p53 isoform seen in GBM.
B. Additional isoform-specific transcripts
In order to determine if other p53 isoforms in the same MW range as 40p53 (45-48kD by western blot analysis) were expressed at the mRNA level, we carried out RT-PCR to detect p53β and γ in glioblastoma xenograft (GBM XT) samples using previously established protocols for single and nested PCR (2). Using p53β primers in a single PCR reaction, we were able to detect very low levels of the predicted 1kb productin four GBM XT samples (GBM10, 12, 16 and 22)(B, top panel), which were not detected in PCR preparations without reverse transcriptase (data not shown). Likewise, we used PCR primers specific for the p53γ isoform, but did not detect any bands at the predicted 1kb length (B, third panel). To increase detection of these very low abundance transcripts, we switched to a nested PCR protocol with primers common to p53, p53β,p53γ(Fig. S1) in round one, followed by a second round of PCR with primers specific for either p53β or p53γ and 2/20l of the previous PCR reaction. In addition to the specific primers, this second reaction also included 0.1x p53 F/R primers carried over from the first reaction. The nested PCR for p53β revealed a 1kb (Amp2) product in GBM12 and an approximately 1.2kb (Amp1) product in other GBM XT samples except GBM8 (B, second panel). For p53γ, we found amplicons at approximately 1.2kb (Amp1) and 1kb (Amp3), in addition to a blurred band at approximately 850bp (B, fourth panel). We submitted the 1.2kb Amp1 and 1kb Amp2 and Amp3 PCR products for sequencing and found that the 1.2kb amplicon matched the sequence of full-length p53 (data not shown). We attribute this to reamplification of p53 with residual p53F/R primers and p53 PCR products carried over from the first reaction. The 1kb amplicons contained p53 sequences, but not those specifically found in p53β (top) or p53γ (bottom), which are highlighted in red in panel (C). Amp2 contained sequences from exon 10, but not the alternatively spliced exon 9, which are present in p53 and p53, respectively, as shown in panel (D)(region unique to p53β indicated in red). Thus, Amp2 cannot be p53. The prominent 1kb band seen in GBM12 is most likely due to a splice variant produced by the p53 mutation in this xenograft (see main text and Table I). The sequence of Amp3 is aligned with p53 and p53γ in panel (E) (region unique to p53γ indicated in red). Again, we found sequences from exon 10 (p53), but not the alternatively spliced exon 9 unique to p53, indicating that Amp3 cannot represent p53 transcripts. One possibility is that this PCR product corresponds to p53 transcripts present in GBM16 and at lower levels in GBM22 and 14. There is a faint band corresponding to a protein with the MW of p53, which is slightly smaller than 40p53, that can be seen in GBM22 on western blots with antibody DO1. There is currently no commercially available antibody that could confirm p53 protein in GBM.
Fig. S3. Additional bands detected by p53 western blot analysis
On full-sized blots, two higher molecular weight bands at approximately 70 and 60kD (Fig. S1 A-D, indicated by * and **) were detected at varying levels across the majority of samples in glioblastoma xenograft (GBM XT) (A), glioblastoma patient tissue (GBM PT) (B), cerebral cortex (CTX) (C), and gliosis (GLS) (D). We also identified a CM1-reactive protein migrating slightly above the 50kD marker (most prominently seen in gliosis samples, indicated by ***). This p53 species did not react with the N-terminus antibodies DO1 or DO7, indicating that it is not full-length p53, p53β or γ (data not shown). Using pAb421, we also detected a lighter ~35kD band in all of the GBM XT samples (main text Fig. 1A), which we postulated could be Δ133p53 based on the molecular weight and antibody binding pattern (detected by pAb421 but not DO1). However, we did not observe similar bands at ~35kD using CM1 (A). Arrow indicates a nonspecific band that was also present in p53-null cells (Fig. S3).
Fig. S4. Analysis of higher molecular weight bands detected by CM1
We aligned representative glioblastoma patient (GBM PT) and xenograft (GBM XT) samples with p53 null cells (p53-/- mouse neural stem cells, H1299 human lung carcinoma and LNZ308 human glioblastoma cells) to determine if the higher molecular weight bands detected by CM1 were p53 species (A). The 70 and 60kD bands (indicated by * and **) were found in all three p53-null samples, and the third band migrating slightly above 50kD (designated ***) was not. From these data, we determined that the 70 and 60kD bands were nonspecific and that the ~55kD band is a p53 species. No human p53 peptide sequences were recovered by mass spectrometric analysis of these higher molecular weight bands immunoprecipitated with 4 different p53 antibodies (data not shown) and they did not react with antibodies against SUMO-1, p63, or p73 (B). The band at approximately 70kD did react with an antibody against hsp70, but no hsp70 peptides were detected by mass spectrometry.
Fig. S5. Detection of 40kD molecular weight band in in p53-deficient cells
In order to determine if the band migrating at approximately 40kD detected by DO1 and CM1 (see main text, Fig. 2B) is a p53 species, we aligned p53-null cells (mouse neural stem cells, human lung carcinoma cells H1299, and human glioblastoma cells LNZ308) with representative samples from glioblastoma patient tissue (GBM PT), gliosis (GLS), and cerebral cortex (CTX) and examined the banding pattern using DO1. We found a band at ~40kD across all samples, both p53+/+ and p53-/-, and concluded that this band is not a p53-species.
Fig. S6. Maintenance of p53 expression in late xenograft passages
Western blot detection of p53 species in glioblastoma xenograft (GBM XT) samples at later passages ranging from 30 to 46. Nonspecific band at approximately 60kD is indicated by ** (see Fig. S3 for explanation). The CM1 polyclonal antibody was used in the analysis.
Fig. S7. Differential recognition of mouse p53 and Δ40p53 by 1C12 N-terminal p53 antibody
In order to confirm that the band migrating below the 50kD at approximately 45kD in mouse samples is Δ40p53, we probed mouse embryonic stem cells (ESC), neural stem cells (NSC), and cortex with the N-terminal p53 antibody, 1C12(A). We did not detect the band below 50kD designated as Δ40p53 and did observe a band at approximately 50kD in ESCs, which express both p53 and Δ40p53 (3). We also confirmed the presence of Δ40p53 in mouse SVZ cellsusing an additional p53 antibody, CM1 (B and C), which detected a band below the 50kD corresponding to the Δ40p53 band detected by pAb421 (main text, Fig. 4 and 5). Mouse SVZ cells were derived from embryonic day 14 (E14), 2.5- and 14-month old mice (2.5mo and 14mo) (B). Panel (C) indicates differentiated SVZ cells on day 0, 4, 5, 6, and 7 of the differentiation protocol described in the main text.
Fig. S8. Immunocytochemical and morphological analysis of human and mouse subventricular zone (SVZ) progenitor cells
Human (top panel) and mouse (bottom panel) SVZ cells were stained with the nuclear dye DAPI (left most column) and with combinations of various neural differentiation markers: β-tubulin or oligodendrocyte (O4) with glial fibrillary acidic protein (GFAP) as indicated. Nestin and GFAP double positive cells (nestin+GFAP+) are multipotent precursor cells (4-9) that have the potency to differentiate into astrocytes (GFAP single positive), oligodendrocytes (O4 single positive), or neurons (β-tubulin single positive) (10-12). Based on the number of positively stained cells, we found that the proportions of nestin+GFAP+ cells were 36% and 19% for human and mouse SVZ cells, respectively (main text, Table III). The morphology of nestin+GFAP+ cells varied within each species sample. In humans, the morphology and staining pattern could be categorized into two subpopulations: cells with small nuclei and cells with large nuclei. In the small nuclei population, most cells had small cytoplasmic bodies with processes that were greater than twice the length of the cell body. The number of processes varied approximately from 2 to 10 within this population. In contrast, the diameter of the large nuclei were approximately 2-3 times that of small nuclei and these cells appeared to have a large cytoplasmic component with ruffled borders reminiscent of astrocytes. The morphology of mouse SVZ cells ranged from small-bodied cells that had 2-3 processes to those that appeared stellate with >10 short processes.
89% of mouse and 39% of human SVZ cells stained for GFAP only (main text, Table III). In human SVZ cultures, these were generally cells with large nuclei, while mouse GFAP+ cells had small nuclei with less cytoplasm and processes. There were no cells that stained for nestin only in either mouse or human samples, nor were there any cells that were exclusively β-tubulin positive in either population. The oligodendrocyte staining pattern differed between the two species, where approximately 11% of the human NSC population stained O4 positive and none of the mouse SVZ cells were O4 positive.
Fig. S9. Mdm2 expression in glioblastoma xenograft (GBM XT) tissues
To determine if Mdm2 levels were altered in the presence of differential Δ40p53 isoform expression, we examined Mdm2 levels in GBM XT samples by western blot. Mouse embryonic stem cells (ESC) derived from p44 transgenic mice (p44) overexpressing Δ40p53 is used as a positive control (13) and wildtype (WT) counterpart as a negative control.
Supplemental References
1.Avery-Kiejda KA, Zhang XD, Adams LJ, Scott RJ, Vojtesek B, Lane DP, et al. Small molecular weight variants of p53 are expressed in human melanoma cells and are induced by the DNA-damaging agent cisplatin. Clin Cancer Res. 2008 Mar 15;14(6):1659-68.
2.Boldrup L, Bourdon JC, Coates PJ, Sjostrom B, Nylander K. Expression of p53 isoforms in squamous cell carcinoma of the head and neck. Eur J Cancer. [Research Support, Non-U.S. Gov't]. 2007 Feb;43(3):617-23.
3.Ungewitter E, Scrable H. Delta40p53 controls the switch from pluripotency to differentiation by regulating IGF signaling in ESCs. Genes Dev. 2010 Nov 1;24(21):2408-19.
4.Lendahl U, Zimmerman LB, McKay RD. CNS stem cells express a new class of intermediate filament protein. Cell. 1990 Feb 23;60(4):585-95.
5.Doetsch F, Caille I, Lim DA, Garcia-Verdugo JM, Alvarez-Buylla A. Subventricular zone astrocytes are neural stem cells in the adult mammalian brain. Cell. 1999 Jun 11;97(6):703-16.
6.Laywell ED, Kukekov VG, Steindler DA. Multipotent neurospheres can be derived from forebrain subependymal zone and spinal cord of adult mice after protracted postmortem intervals. Exp Neurol. 1999 Apr;156(2):430-3.
7.Imura T, Kornblum HI, Sofroniew MV. The predominant neural stem cell isolated from postnatal and adult forebrain but not early embryonic forebrain expresses GFAP. J Neurosci. 2003 Apr 1;23(7):2824-32.
8.Imura T, Nakano I, Kornblum HI, Sofroniew MV. Phenotypic and functional heterogeneity of GFAP-expressing cells in vitro: differential expression of LeX/CD15 by GFAP-expressing multipotent neural stem cells and non-neurogenic astrocytes. Glia. 2006 Feb;53(3):277-93.
9.Garcia AD, Doan NB, Imura T, Bush TG, Sofroniew MV. GFAP-expressing progenitors are the principal source of constitutive neurogenesis in adult mouse forebrain. Nat Neurosci. 2004 Nov;7(11):1233-41.
10.Reynolds BA, Tetzlaff W, Weiss S. A multipotent EGF-responsive striatal embryonic progenitor cell produces neurons and astrocytes. J Neurosci. 1992 Nov;12(11):4565-74.
11.Reynolds BA, Weiss S. Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science. 1992 Mar 27;255(5052):1707-10.
12.Reynolds BA, Weiss S. Clonal and population analyses demonstrate that an EGF-responsive mammalian embryonic CNS precursor is a stem cell. Dev Biol. 1996 Apr 10;175(1):1-13.
13.Maier B, Gluba W, Bernier B, Turner T, Mohammad K, Guise T, et al. Modulation of mammalian life span by the short isoform of p53. Genes Dev. 2004 Feb 1;18(3):306-19.