Page 55 - Read Online

Elton et al. Cancer Drug Resist 2020;3:161-70 I http://dx.doi.org/10.20517/cdr.2019.117 Page 165

A B C

Figure 2. A novel human TOP2α/90 isoform is overexpressed in acquired resistance to TOP2α-targeted drugs etoposide, mitoxantrone,
and amsacrine. A: TOP2α immunoassay utilizing K562 and K/VP.5 [35,36,55,56] cell lysates; B: TOP2α immunoassay utilizing HL-60 and HL-
60/MX2 [32,50,51] cell lysates; C: TOP2α immunoassay utilizing HL-60 and HL-60/AMSA [57] cell lysates. The immunoblots were probed
with an antibody specific for the N-terminal portion of TOP2α/170 (i.e., amino acids 14-27, denoted N-terminal Ab). (A) Image is from
Figure 2A published originally in the Journal of Pharmacology and Experimental Therapeutics; Kanagasabai et al. [35] , 2017. TOP2α:
topoisomerase IIα

TOP2α/90 (INTRON 19 RETENTION) AND CHEMORESISTANCE
Our laboratory has also investigated the molecular mechanisms which lead to decreased TOP2α/170
expression levels in acquired chemoresistance. Resistant human leukemia K562 cells were generated by
intermittent then continuous treatment with 0.5 µM etoposide followed by limiting dilution to isolate and
[55]
then characterize a clonal K/VP.5 cell line . Compared to parental K562 cells, the K/VP.5 subline was 30-
[56]
fold resistant to etoposide and cross-resistant to teniposide, mitoxantrone, doxorubicin, and amsacrine .
[56]
This multi-drug resistance was not mediated by overexpression of ABCB1 . K/VP.5 cells exhibited
reduced TOP2α/170 mRNA (by Northern blot analysis) with no change in transcription compared to K562
[55]
cells . In addition, using an antibody generated from the C-terminal 70 kDa of TOP2α, immunoassays of
cells lysates demonstrated reduced TOP2α/170 protein levels in K/VP.5 compared to K562 cells [55,56] .

Surprisingly, additional immunoblotting experiments using a N-terminal specific TOP2α/170 antibody
(generated against amino acids 14-27) revealed the presence of two major TOP2α proteins, the expected
wild-type TOP2α/170 isoform and a novel 90 kDa isoform, TOP2α/90 [Figure 2A] [35,36] . Compared to parental
K562 cells, the expression level of TOP2α/170 was attenuated as expected but TOP2α/90 was increased in K/
VP.5 cells [Figure 2A] [35,36] .Immunoassays utilizing cell lysates from two additional TOP2α-poison resistant cell
[32]
[57]
lines, HL-60/MX2 (mitoxantrone-resistant) and HL-60/AMSA (amsacrine-resistant) , also demonstrated
greater TOP2α/90 protein levels compared to parental HL-60 cells [Figure 2B and C].
Using 3’-rapid amplification of cDNA ends (3’-RACE), followed by PCR and sequencing, analyses revealed
that TOP2α/90 mRNA (2762 nt) shares the first 19 exons with the TOP2α/170 transcript. However, the
TOP2α/90 mRNA retains a processed intron 19 (380 nt) that harbors an in-frame stop codon, and two
[35]
consensus poly(A) sites [Figure 1A-iii] . TOP2α/90 mRNA lacks the published TOP2α/170 transcript
sequences from exon 20 to 35, and harbors a novel 3’-untranslated region (302 nt) [Figure 1A-iii] .
[35]
TOP2α/90 mRNA intron 19 retention was validated by mapping RNA-seq raw reads [Figure 1C].
The TOP2α/90 mRNA encodes a truncated TOP2α protein isoform of 786 aa with a calculated molecular
weight of 90,076 Da, which is approximately one half the size of the wild-type TOP2α/170 protein (i.e., 1531
aa, 174,385 Da) [Figure 1B] [35,36] . Although TOP2α/90 is identical to TOP2α/170 for the first 761 aa, this
protein is missing the C-terminal 770 aa present in TOP2α/170, which are replaced with 25 unique amino
acids encoded by the exon 19/intron 19 “read-through” [Figure 1B-iii]. As a result of intron 19 retention, the
truncated TOP2α/90 isoform does not harbor an active site tyrosine (Tyr805), which is present in the DNA
[1-4]
gate domain [Figure 1B] and is required for wild-type TOP2α/170 to generate double-strand DNA breaks .
Finally, TOP2α/90 is also missing two characterized dimerization domains (DD) (i.e., 1053-1069 aa and
1121-1143 aa) [41-44] and NLS 1454-1497 [46,47] present in wild-type TOP2α/170 [Figure 1B].