838 860
Importantly, in a cohort of men with metastatic CRPC
undergoing ADT treatment with Enz (ADT-Enz) a longitu-
dinal analysis of CTCs from the same 10 patients before and
after ADT-Enz demonstrated an increase of
Malat1
and AR-
v7 expression in eight of 10 patients evaluated
( Fig. 1 H,
Supplementary Table 3), consistent with our in vitro study.
Furthermore, analysis of the The Cancer Genome Atlas
database revealed PCa patients harboring higher expression
of
Malat1
have poor overall survival (Supplementary
Fig. 1 F).
Together, results from multiple EnzR-PCa cells
( Fig. 1A–
G) and human clinical data
( Fig. 1 H), all suggest that
Malat1
may play essential roles in the development of EnzR-CRPC.
3.2.
Enz-induced Malat1 expression via suppressing AR
function
We further dissected mechanism(s) by which
Malat1
expression was enhanced during or after development of
Enz resistance. We found that transient treatment with Enz
increased
Malat1
expression in the presence of 1 nM
dihydrotestosterone (DHT), the androgen concentration
found in both CRPC C4-2 and and VCaP cells
( Fig. 2 A,
supplementary Fig. 2B, respectively), and knocking down
AR with short hairpin RNA (shRNA;
Fig. 2B), led to increased
Malat1
expression at 1 nM DHT in C4-2 cells
( Fig. 2B). In
contrast, adding functional AR-complementary DNA or
treating with DHT (both 1 nM DHT and 10 nM DHT)
resulted in suppressing the
Malat1
expression in the EnzR
cells (Supplementary Figs. 2C and 2D).
3.3.
Enz induced Malat1 expression via altering AR binding to
Malat1 promoter
Further mechanism dissection found AR could bind directly
to the predicted AR response elements (AREs) located on the
proximal promoter of the
Malat1
gene, and adding DHT
increased AR binding to these AREs
( Fig. 2 C) which might
then lead to suppress
Malat1
expression (Supplementary
Fig. 2C). In contrast, adding Enz might suppress the AR
binding to these AREs
( Fig. 2 D), which might then lead to
increase the
Malat1
expression
( Fig. 2 A).
Sequence analysis found a H3K4me3 enrichment region
(see the University of California Santa Cruz database;
Fig. 2 E, indicating the promoter activity) in the 2 kb
upstream of
Malat1
gene locus. The results from manipu-
lating AR activity with DHT or Enz revealed that adding DHT
could suppress H3K4me3 levels in
Malat1
promoter in the
C4-2 cells
( Fig. 2E lower panel), and transient or long-term
Enz (R1 cells) treatment might increase the
Malat1
promoter activity (Fig. 2F and G, respectively).
3.4.
Malat1 is indispensable for AR-v7 production and may
function via interacting with SF2 to splice the AR transcript
Next, to examine if Enz-enhanced
Malat1
expression is
required for AR-v7 expression, we knocked-down
Malat1
with two different siRNAs
( Fig. 3A) and results revealed that
suppressing
Malat1
by
Malat1
-siRNAs led to decrease AR-v7
expression in the EnzR C4-2 cells
( Fig. 3A) and VCaP cells
( Fig. 3 B, Supplementary Fig. 3A), suggesting that
Malat1
is
indispensable for Enz-induced AR-v7 expression.
Early studies suggested
Malat1
might function through
binding to (SF2, also called ARSF1) to increase the RNA
splicing capacity for the targets
[7,9] .We examined the
possible complex formation of
Malat1-
SF2 on the
AR
transcript by immunoprecipitating SF2 complex followed
by analysis of associated RNAs including
Malat1
and
AR
transcript. The results revealed that at the 1 nM DHT
castration condition, there is a significant increase in
binding of
Malat1
and
AR
transcript to SF2 in EnzR cells
compared with parental Enz-sensitive cells, supporting the
hypothesis that
Malat1
can enhance splicing of AR-v7
through direct modulation of SF2 binding and activity
( Fig. 3 C). Furthermore, we found that in EnzR cells there is a
higher level of SF2 activity determined by elevated
phosphorylation level, indicating enhanced splicing activity
of SF2 likely as a result of increased
Malat1
expression
( Fig. 3D). Importantly, knocking down of SF2 in VCaP cells
( Fig. 3E) and EnzR1 cells
( Fig. 3F) also led to attenuate AR-v7
production, suggesting that SF2 is indeed involved in the
splicing of AR-v7 as reported earlier
[10].
3.5.
Malat1-enhanced AR-v7 expression contributes to the
development of Enz resistance
To confirm our hypothesis that Enz could function through
induction of the
Malat1
/AR-v7 axis to accelerate develop-
ment of Enz-resistance, we examined the effects of Enz on
the growth of C4-2 cells stably transfected with AR-v7, and
MTT proliferation assay results revealed that Enz had a
reduced effect on cell growth suppression
( Fig. 4A), and
transducing AR-v7-shRNA could resensitize EnzR1 cells to
respond to Enz treatment
( Fig. 4 B).
We then applied another approach using the TALE-based
Malat1
induction to prove that increased
Malat1
(therefore
increased AR-v7) in C4-2 cells also resulted in a reduced
effect of Enz to suppress the cell growth
( Fig. 4C), and
adding
Malat1
-siRNA in EnzR cells can overcome Enz
resistance
( Fig. 4 D). Importantly, we found the contribution
of
Malat1
to Enz resistance can be blocked by AR-v7-shRNA
or SF2-shRNA
( Fig. 4E and F). Furthermore, reciprocal
expression of AR-v7 could reverse
Malat1
-siRNA
-
mediated
effects on both Enz sensitivity and cell growth (Supple-
mentary Figs. 4A and 4B).
Together, results from
Fig. 4 A–F and Supplementary Fig.
4A-B strengthen the notion that the
Malat1
/SF2/AR-v7 axis
is critically involved in Enz-resistance development.
3.6.
New therapies to suppress EnzR-PCa in in vitro cell lines
and in vivo mouse models
To prove the key roles of our newly identified Enz/
Malat1
/
AR-v7 axis in the development of Enz-resistance in the in-
vivo mouse model and seek new therapeutic approaches to
better suppress EnzR-PCa, we used our newly developed
EnzR cell lines to test the efficacy of targeting
Malat1
and
AR-v7 with
Malat1
-siRNA or AR-v7 degradation enhancer
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 8 3 5 – 8 4 4
838