Professional Documents
Culture Documents
Abstract
Tumor recurrence in prostate cancer has been attributed to the presence of CD44-expressing tumor-
initiating cells. In this study, we report that miR-708 is a key negative regulator of this CD44þ subpopulation
of prostate cancer cells, with important implications for diagnosis and prognosis of this disease. miR-708 was
underexpressed in CD44þ cells from prostate cancer xenografts. Reconstitution of miR-708 in prostate cancer
cell lines or CD44þ prostate cancer cells led to decreased tumorigenicity in vitro. Intratumoral delivery of
synthetic miR-708 oligonucleotides triggered regression of established tumors in a murine xenograft model of
Materials and Methods after transfection, cells were counted and placed on control
Cell lines and cell culture inserts or Matrigel inserts at 1 105 cells/mL in serum-free
Nonmalignant epithelial prostate cell lines (RWPE-1 and medium and were allowed to migrate for 20 hours at 37 C. Cells
PWR-1E) and prostate carcinoma cell lines (LNCaP, Du145, were removed from the top of the inserts and cells that
PC3, VCaP, MDA-PCa-2b) were obtained from the American migrated/invaded though the polycarbonate basement mem-
Type Culture Collection (ATCC) and cultured under recom- brane were fixed, stained, and quantified at optical density of
mended conditions as described previously (ref. 13; also 560 nm after extraction. For clonogenicity assay, cells were
described in Supplementary Methods). These human-derived counted, seeded at low density (1,000 cells/plate) and allowed
cell lines were authenticated by DNA short tandem repeat to grow until visible colonies appeared. Then, cells were stained
analysis by ATCC. The experiments with cell lines were con- with Giemsa, and colonies were counted.
ducted within 6 months of their procurement/resuscitation.
Apoptosis assays
FACS analysis for apoptosis was done 72 hours after trans-
Prostate cancer xenografts
fection, using Annexin V-FITC/7-AAD Kit (Beckman Coulter,
Xenograft human prostate tumors LAPC-4 and LAPC-9 were
Inc.), according to the manufacturer's protocol. Stained cells
courtesy of R. Reiter (14, 15), PC3 xenograft tumors were
were immediately analyzed with a flow cytometer (Cell Lab
A CD44 3′ UTR
1 3066
Site 1 Site 2
B 1.2
Figure 1. CD44 is a functional target
of miR-708 in prostate cancer. A,
Relative miR-708 expression
1
schematic representation of CD44
0.8 30 -UTR showing the relative
P = 0.0098 positions of putative miR-708
0.6 CD44–
target sites. B, relative miR-708
P = 0.0009 CD44+
P = 0.0162 levels in purified CD44 and
0.4 CD44þ subpopulations of prostate
cancer xenografts (PC3 and
0.2 P = 0.0016
P = 0.0016 P = 0.0007 LAPC9). Data were normalized to
miR-CON
miR-CON
RT-PCR (left). Immunoblots for
Relative miR-708 expression
miR-708
miR-708
16,000
250 endogenous CD44 protein in
Mock
Mock
14,000
transfected PC3/LNCaP cells
200 12,000
(right). Glyceraldehyde-3-
10,000
150 phosphate dehydrogenase
8,000 CD44
100
(GAPDH) was used a loading
6,000
control. D, CD44 30 -UTR construct
50 4,000
GAPDH or the control construct was
2,000
0 cotransfected into PC3/LNCaP
0
Mock miR-CON miR-708 Mock miR-CON miR-708 cells along with miR-708/miR-
CON/mock-treated cells and
D PC3 LNCaP assayed for luciferase activity.
Control CD44 3′UTR Firefly luciferase values were
1.4 1.4 normalized to Renilla luciferase
activity and plotted as relative
1.2 1.2
Relative luciferase activity
luciferase activity
1 1 ( , P < 0.05).
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0 0
Mock miR-CON miR-708 Mock miR-CON miR-708
Student t test was used for comparisons between groups. algorithms that predict the mRNA targets of a miRNA, miRAN-
Results were considered statistically significant at P 0.05. DA (21) and TargetScan (22). From these analyses, miR-708
Full Methods are available in the Supplementary Materials stood out for the presence of 2 binding sites, suggesting
and Methods. cooperative binding and interaction (Fig. 1A). To test the
potential regulation of CD44 by miR-708, we purified CD44þ
and CD44 subpopulations of prostate cancer cells from 2
Results prostate cancer xenograft mouse models (PC3 and LAPC9)
CD44 is a functional target of miR-708 in prostate cancer followed by miR-708 expression profiling (Fig. 1B). miR-708
Because CD44 is an important marker that determines expression was significantly downregulated in CD44þ versus
tumorigenic behavior of prostate cancer cells, it is essential CD44 cells suggesting that miR-708 regulates the CD44þ
to elucidate regulatory mechanisms that converge on this subpopulation in prostate cancer. To validate this further, we
molecule. miRNAs constrain gene expression by binding to overexpressed miR-708 in prostate cancer cell lines (PC3 and
the 30 -UTRs of cognate mRNA targets (20). To explore the LNCaP). Transient transfection of miR-708 precursor led to
potential regulation of this TIC marker by miRNAs, we used 2 overexpression of miR-708 as determined by real-time PCR
(RT-PCR; Fig. 1C, left). In both the cell lines, CD44 protein was miR-708 expression is widely attenuated in prostate
specifically downregulated in miR-708–expressing cells (Fig. cancer
1C, right) suggesting that miR-708 regulates CD44. To validate To evaluate the role of miR-708 in prostate cancer, miR-708
CD44 as a miR-708 target, we conducted luciferase reporter expression was assayed in human prostate cell lines, which
assays with CD44 30 -UTR in miR-708/miR-CON/mock-trans- included normal epithelial prostate cell lines (RWPE-1 and
fected PC3/LNCaP cells (Fig. 1D). We observed a significant PWR-1E) and prostate carcinoma cell lines (LNCaP, LAPC4,
decrease in luciferase activity upon miR-708 transfection, Du145, PC3, VCaP, MDA-PCa-2b, LAPC9). Relative miR-708
suggesting that miR-708 represses CD44 directly. Collectively, expression was specifically attenuated in all the examined
these data indicate that CD44 is a direct target of miR-708 in prostate carcinoma cell lines compared with normal epithelial
prostate cancer. prostate cell lines (Fig. 2A). To examine the clinical relevance of
A C P = 0.0002
1.2
Relative miR-708 expression
1.6
1
P = 0.0623
1.2
0.6
Figure 2. miR-708 expression is 1
0.4
widely attenuated in prostate cancer.
0.8
A, qRT-PCR analysis of relative miR- 0.2 P = 0.0003
P = 0.0001 P = 0.0001 P = 0.0002 P = 0.0001 P = 0.0001 P = 0.0005
708 expression levels in normal 0
0.6
epithelial prostate cell lines (RWPE-1 0.4
and PWR-1E) and prostate
100
708 expression with 70-79 25/134 (19) 18/25 (72) 4/25 (16) 3/25 (12)
80–89 1/134 (1) 1/1 (100) - -
clinicopathologic characteristics of
Gleason score
patients with prostate cancer. E, 4-6 46/129 (36) 29/46 (63) 7/46 (15) 10/46 (22)
ROC curve analysis showing the 7 53/129 (41) 39/53 (74) 9/53 (17) 5/53 (9) 0.0098*
ability of miR-708 expression to 8-10 30/129 (23) 27/30 (90) 2/30 (7) 1/30 (3)
Pathologic stage
discriminate between malignant and pT2 71/125 (57) 51/71 (72) 8/71 (11) 12/71 (17)
100-Specificity
nonmalignant tissue samples. pT3 52/125 (42) 43/52 (83) 5/52 (10) 4/52 (8)
0.1095 F Remission
F, Kaplan–Meier survival curves for pT4 2/125 (2) 2/2 (100) - -
patients with prostate cancer, Biochemical recurrence
HR: 6
stratified on the basis of miR-708 Yes 22/92 (24) 18/22 (82) 3/22 (14) 1/22 (5) 0.0138*
Survival probability (%)
Time (mo)
this finding, we extended our analysis to human clinical receiver operating characteristic (ROC) analyses on the cohort
prostate samples (N ¼ 136; Fig. 2B and C, Supplementary Fig. of clinical samples (Fig. 2E). ROC analyses showed that miR-
S1). Clinicopathologic characteristics of the patients are sum- 708 expression can be a single significant parameter to dis-
marized in Supplementary Table S1. We examined the expres- criminate between normal and tumor tissues with an area
sion levels of miR-708 in laser capture microdissected (LCM) under the ROC curve (AUC) of 0.937 [95% confidence interval
prostate cancer tissues (n ¼ 96) and matched adjacent normal (CI), 0.901–0.963; P < 0.0001]. Furthermore, Kaplan–Meier
regions by RT-PCR (Fig. 2B). While the expression of miR-708 survival analysis for patients with prostate cancer, stratified
was unaltered in 13 of 96 cases (13%) and higher in 16 of 96 on the basis of miR-708 levels, showed that survival was
cases (17%), a major fraction of tissue samples (67 of 96, 70%) significantly reduced in patients with low miR-708 expression
showed lower miR-708 levels relative to matched normal (Fig. 2F). The HR between the cases with low or high miR-708
tissues. The differences were statistically significant with the expression was 6 with a 95% CI from 2.2 to 16.4 and with an
Wilcoxon signed rank test (P < 0.0001). We extended our associated P value of 0.0223. This analysis suggests that miR-
analysis of clinical specimens by assessing miR-708 levels in 708 has significant potential to be used as a diagnostic and
40 additional cases of prostate adenocarcinoma, 8 of which had prognostic marker for prostate cancer.
matched normal adjacent tissue by in situ hybridization anal-
ysis and again observed attenuated expression of miR-708 in miR-708 reexpression suppresses tumorigenicity in vitro
A B PC3 LNCaP
miR-CON miR-708 miR-CON miR-708
PC3 LNCaP
140
Mock
Relative cell viability (%)
120
120 miR-CON
miR-708 100
100
80 80
300 P = 0.0368 P = 0.0025
60
C miR-CON
PC3
miR-708 miR-CON
LNCaP
miR-708
0
Mock miR-CON miR-708
0
Mock miR-CON miR-708
D PC3 LNCaP
miR-CON miR-708 miR-CON miR-708
1.2
0.8
1 0.7 P = 0.0039 P = 0.0183
1.2
P = 0.0099 P = 0.0046
Invasion (OD 560 nm)
0.6 0.6
0.8
0.5 1
0.5
0.6 0.4
0.8
0.4 0.3 0.4
0.2 0.6
0.2 0.3
0.1
0 0.4
0 0.2
Mock miR-CON miR-708 Mock miR-CON miR-708
0.2 0.1
E miR-CON miR-708 0
Mock miR-CON miR-708
0
Mock miR-CON miR-708
% Apoptotic 5±2 % Apoptotic 23 ± 2
% Early apoptotic 4±1 % Early apoptotic 8±3
% Viables 90 ± 3 % Viables 66 ± 5
% Dead 2±1 % Dead 3±1
G
Mock miR-CON miR-708
P = 0.0170
P = 0.0028 P = 0.0140
Apoptosis (relative values)
4
P = 0.0009
3.5
3
2.5
2
F miR-CON miR-708
1.5
1
% Apoptotic 4±2 % Apoptotic 13 ± 2 0.5
% Early apoptotic 4±1 % Early apoptotic 16 ± 4 0
% Viables 91 ± 4 % Viables 68 ± 2 PC3 LNCaP
% Dead 2±1 % Dead 3±1
Total cells Total cells
Figure 3. Restoration of miR-708 expression suppresses tumorigenicity in vitro. To evaluate the role of miR-708 in prostate cancer, miR-708 precursor was
reintroduced in PC3/LNCaP cell lines by transient transfections followed by functional assays after 72 hours. Cell viability assay (A), colony formation assay (B),
Transwell migration assay (C), and invasion assay (D) with PC3/LNCaP cells mock-transfected/transfected with miR-CON or miR-708. E, apoptosis assay in
PC3 and LNCaP (F) cells. Seventy-two hours after transfection, cells were stained with Annexin V-FITC/7-AAD and immediately analyzed with a flow
cytometer. G, representation of relative apoptotic fractions (early þ late apoptotic) in each group for PC3/LNCaP cells. Apoptosis induced by miR-CON/miR-
708 was normalized to mock control ( , P < 0.05).
A B
miR-CON miR-708 miR-CON miR-708
1,200
1,000
Tumor volume (mm3)
800
600
400
200
C P = 0.0001 D
2,000 P = 0.0003
Relative miR-708 expression
1,800
P = 0.0006
Control
1,600
miR-708
P = 0.0002
1,400
T1 T2 T3 T4 T5
1,200
1,000
800 CD44
P = 0.0008
600
400
200
P = 0.1000 GAPDH
0
Sample C1 C2 T1 T2 T3 T4 T5
miR-CON miR-708
Figure 4. Intratumoral delivery of miR-708 leads to regression of tumors in a prostate cancer xenograft mouse model and attenuates CD44 expression. A, PC3
cells were subcutaneously injected into nude mice to form solid, palpable tumors (day 30), following which synthetic miR-708/control miR were intratumorally
delivered every 3 days for 4 weeks. Tumor volumes following miR-708 administration were significantly reduced ( , P < 0.05). B, representative
images of mice from the 2 groups on day 58 are shown. C, relative miR-708 expression in prostate cancer xenografts as assessed by RT-PCR. D, CD44 protein
levels in prostate cancer xenografts as assessed by immunoblot analysis.
models. To validate this, we extracted protein from these assay showed that apoptotic cell fractions were significantly
xenografts and gauged whether intratumoral administration increased upon CD44 knockdown compared with control, an
of miR-708 altered CD44 expression (Fig. 4D). Indeed, CD44 effect similar to that observed upon miR-708 reintroduction in
protein levels were repressed in miR-708 injected xenografts PC3 cells (Fig. 5E and F). These results suggest that CD44 is an
compared with the control tumors. This finding reinforces the important determinant of the tumorigenic properties of pros-
idea that CD44 is a direct target of miR-708. tate cancer cells and that CD44 inhibition partially pheno-
copies the antitumorigenic effects of miR-708 in prostate
CD44 knockdown partially phenocopies miR-708 cancer. Furthermore, in view of our present results showing
reexpression in prostate cancer cells that CD44þ cells from prostate cancer xenografts presented
To determine whether CD44 is a functionally relevant target lower miR-708 expression, we also examined whether CD44
of miR-708 in prostate cancer, we inhibited CD44 expression knockdown in PC3 cells altered miR-708 expression (Fig. 5G).
using siRNA to see whether CD44 knockdown functionally Interestingly, siRNA-mediated CD44 knockdown led to aug-
mimics the effects of miR-708 overexpression in prostate mentation of miR-708 expression. This observation reinforces
cancer. We treated PC3 cells with CD44 siRNA followed by in the regulatory interplay between miR-708 and CD44.
vitro functional assays. We initially tested 3 sets of siRNA
against CD44 to achieve efficient knockdown, as assessed by miR-708 inhibits tumor-initiating capacity of prostate
RT-PCR and immunoblot analysis (Fig. 5A). si1 produced the cancer cells in vitro
most efficient knockdown and was used in subsequent experi- Because CD44 has been identified as a major molecular
ments. siRNA inhibition of CD44 led to decreased cell viability, determinant of prostate cancer TICs (7–12), and in view of our
migration, and invasion of PC3 cells (Fig. 5B–D). Apoptosis present results suggesting that miR-708 regulates CD44, we
A B
NS siRNA
siRNA
CD44
NS siRNA CD44 siRNA
120
1
60
0.8 GAPDH 40
0.6 20
0
0.4
Days 0 1 2 3
0.2
F
1.2 1.2 25
1 1
P = 0.0051
Relative migration
20
P = 0.011
Relative invasion
0 0
5
NS siRNA CD44 siRNA NS siRNA CD44 siRNA
0
E NS siRNA CD44 siRNA
NS siRNA CD44 siRNA
% Apoptotic 3±1
G
% Apoptotic 14 ± 1 P = 0.0009
% Early apoptotic 4±1 45
% Early apoptotic 5±1
Relative miR-708 expression
% Viables 93 ± 2 % Viables 81 ± 2 40
% Dead 0.2 ± 0.1 % Dead 0.3 ± 0.1 35
Total cells Total cells
30
25
20
15
10
5
0
NS siRNA CD44 siRNA
Figure 5. CD44 knockdown partially phenocopies miR-708 reexpression in prostate cancer cells. PC3 cells were transfected with CD44 siRNA/nonspecific
(NS) control siRNA for 72 hours, followed by various assays. A, relative CD44 mRNA expression and protein expression after siRNA transfections as
assessed by RT-PCR and immunoblotting, respectively. Cell viability assay (B), Transwell migration assay (C), invasion assay (D), and apoptosis assay (E) in
PC3 cells after NS siRNA (left) or CD44 siRNA (right) treatment. F, representation of average apoptotic fractions in each group ( , P < 0.05). G, relative miR-708
levels after siRNA transfections as assessed by RT-PCR. Data were normalized to RNU48 control and are represented as mean SEM.
2,500
2,000
1,500
1,000
500 D
miR-CON miR-708
0
miR-CON miR-708
C 35
Figure 6. miR-708 inhibits tumor-
initiating capacity of prostate
18 Colonies (average no.) 30 þ
cancer cells in vitro. CD44 and
16 CD44 subpopulations were
Spheres (average no.)
25
P = 0.0024
0.2 G 30 P = 0.0424
0
Anti- Anti- 25
Spheres (average no.)
miR-CON miR-708
20
15
10
0
Anti- Anti-
miR-CON miR-708
asked whether miR-708 has the potential to alter the tumor- miR-708 on TICs, we conducted sphere formation and clono-
initiating capacity of prostate cancer cells. To this end, we genicity assays, which have been widely used to measure the
purified CD44þ and CD44 cells from PC3 xenografts, manip- activity of TICs (10–12, 16). miR-708 reconstitution significant-
ulated miR-708 in these subpopulations followed by in vitro ly inhibited sphere formation (Fig. 6C) and clonogenic poten-
assays for determining their tumor-initiating capacity (Fig. 6). tial (Fig. 6D) of the CD44þ subpopulation. We also inhibited
CD44þ prostate cancer cells are more tumorigenic than the miR-708 in the purified CD44 population from PC3 xenografts
isogenic CD44 prostate cancer cells (10, 12). CD44þ cells were (Fig. 6E–G) using anti-miR-708 oligos and examined its effect
transfected with control miR or miR-708 precursor (Fig. 6A–D). on the tumorigenic potential of CD44 prostate cancer cells.
We validated overexpression of miR-708 in CD44þ cells by qRT- Treatment with anti-miR-708 led to miR-708 inhibition com-
PCR (Fig. 6A). As shown in Fig. 6B, miR-708 overexpression led pared with control oligo (anti-miR-CON) as assessed by qRT-
to growth inhibition of CD44þ cells. To evaluate the effect of PCR (Fig. 6E). Inhibition of miR-708 led to increased growth
(Fig. 6F) and sphere formation ability (Fig. 6G) in the CD44 ical specimens. Taken together, these findings suggest that this
subpopulation. Taken together, these observations suggest widely observed miR alteration has significant potential as a
that miR-708 negatively regulates the tumor-initiating capacity disease biomarker for diagnosis and prognosis in prostate
of prostate cancer. These results suggest that the observed cancer. Furthermore, our in vitro and in vivo data suggest that
inhibitory effects of miR-708 on the tumorigenicity of bulk reconstitution of miR-708 in prostate cancer suppresses
prostate cancer cells in vitro and in vivo might be due to its tumorigenicity, confirming the tumor-suppressive role of
effects on TICs in prostate cancer. miR-708 in prostate cancer. Moreover, intratumoral delivery
of synthetic miR-708 oligonucleotides was sufficient to trigger
AKT2 is also a functional target of miR-708 in prostate in vivo regression of established tumors in a murine xenograft
cancer model of human prostate cancer, supporting the therapeutic
Using miRNA target prediction algorithms miRANDA (21) potential of this novel miRNA in prostate cancer.
and TargetScan (22), we found that AKT2 is another putative The existence of tumor-initiating subpopulations within the
miR-708 target as it possesses 3 potential miR-708–binding bulk of tumor has been described in murine and human
sites within its 30 -UTR (Fig. 7A). miR-708 reconstitution in PC3/ prostate cancer. These cells are endowed with high tumori-
LNCaP cells led to reduced endogenous protein levels of AKT2 genic capacity and drive tumor growth, metastasis, cause
(Fig. 7B). Luciferase reporter assays with control/AKT2 30 -UTR treatment resistance, and recurrence (23, 24). Several studies
miR-CON
miR-CON
1 3555
miR-708
miR-708
Site 1 Site 2 Site 3
Mock
Mock
Site 1 969: 5′… CCUCUGAAAAGGGAGGCUCCUA G … 3′ AKT2 3′ UTR
:
3′ GGGUCGAUCUAACA UUCGAGGAA 5′ hsa-miR-708
AKT2
Site 2 1602: 5 ...AGACCCCUGGGCCCCAGCUCCUC... 3′ AKT2 3′ UTR
Control
Site 3 2043: 5 ...UUGCUGUCACCUGGGAGCUCCUG...3′ AKT2 3′ UTR D
: miR-708
3′ GGGUCGAUCUAACA UUCGAGGAA 5′ hsa-miR-708
C PC3 LNCaP
T1 T2 T3 T4 T5
1
GAPDH
0.8
0 0 40
Mock miR-CON miR-708 Mock miR-CON miR-708
NS siRNA
20
siRNA
AKT2
0
E Days 0 1 2 3
1 2 3 G NS siRNA AKT2 siRNA
H NS siRNA AKT2 siRNA
1.2 AKT2
Relative AKT2 expression
0.8
GAPDH
1.2
0.6 1.2
Relative migration
1
Relative invasion
1
0.4
0.8
0.8
0.2 P = 0.0001 P = 0.0314
0.6
0.6
0.4
0
0.4
NS siRNA 1 2 3
0.2
AKT2 siRNA 0.2
0
NS siRNA AKT2 siRNA 0
NS siRNA AKT2 siRNA
I NS siRNA
% Apoptotic 3±1
AKT2 siRNA
% Apoptotic 9±1
% Early apoptotic
% Viables
3±1
93 ± 2
% Early apoptotic 5±1 J
% Viables 86 ± 1
% Dead 0.2 ± 0.1 P = 0.0081
% Dead 1 ± 0.4 16
12
% Apoptosis
10
0
NS siRNA AKT2 siRNA
Figure 7. AKT2 is also a functional target of miR-708 in prostate cancer. A, schematic representation of the AKT2 30 -UTR showing the relative positions of 3
putative miR-708 target sites. B, immunoblots for endogenous AKT2 protein in PC3/LNCaP cells transfected as indicated. GAPDH was used a loading control.
C, luciferase activity assay with the AKT2 30 UTR construct/control construct contransfected with miR-CON/miR-708. Firefly luciferase values were
normalized to Renilla luciferase activity and plotted as relative luciferase activity ( , P < 0.05). D, AKT2 expression in harvested tumors from mouse prostate
cancer xenografts intratumorally injected with either miR-CON or miR-708 as assessed by immunoblot analysis (T1–T5 are individual tumors). E, PC3 cells
were transfected with AKT2/NS siRNA for 72 hours. Relative AKT2 mRNA expression and protein expression after siRNA transfections as assessed
by RT-PCR and immunoblotting, respectively. Cell viability assay (F), Transwell migration assay (G), invasion assay (H), and apoptosis assay (I) in PC3 cells
after NS siRNA (left) or AKT2 siRNA (right) treatment. J, representation of average apoptotic fractions in each group ( , P < 0.05).
prostate cancer. miR-34a directly represses CD44 thereby miR-708 expression leads to prostate cancer initiation, pro-
inhibiting prostate cancer stem cells and metastasis gression, and development by regulating expression of CD44 as
(17, 24, 26). Here, we report that CD44 is a direct and functional well as AKT2, that in turn regulates proliferative, invasive, and
target of miR-708 in prostate cancer. migratory abilities of prostate cancer TICs. In view of the
miRNAs suppress gene expression posttranscriptionally via widespread attenuation of miR-708 observed in our patient
sequence-specific interactions with the 30 -UTRs of cognate cohort, its antitumorigenic effects on bulk or CD44þ subpop-
mRNA targets (20). An individual miRNA is capable of regu- ulation of prostate cancer cells and in vivo inhibition of
lating dozens of distinct mRNAs (32, 33), and it is known that prostate tumor xenografts, we envision that restoring miR-
pleiotropic suppression of multiple downstream effectors may 708 may provide a novel therapeutic modality for prostate
underlie the phenotypic changes observed upon perturbing the cancer.
levels of certain miRNAs (13, 34–37). We found that miR-708
targets AKT2 as well. Luciferase reporter assays showed that Disclosure of Potential Conflicts of Interest
miR-708 represses its dependent genes (CD44 and AKT2) by No potential conflicts of interest were disclosed.
direct binding to the complementary sites within their 30 -
UTRs. The serine/threonine kinase AKT2/PKBb is a core Authors' Contributions
Conception and design: S. Saini, R. Dahiya
member of the phosphoinositide 3-kinase (PI3K)/AKT signal-
References
1. Shen MM, Abate-Shen C. Molecular genetics of prostate cancer: new 8. Collins AT, Berry PA, Hyde C, Stower MJ, Maitland NJ. Prospective
prospects for old challenges. Genes Dev 2010;24:1967–2000. identification of tumorigenic prostate cancer stem cells. Cancer Res
2. Brooks DD, Wolf A, Smith RA, Dash C, Guessous I. Prostate cancer 2005;65:10946–51.
screening 2010: updated recommendations from the American Can- 9. Li H, Chen X, Calhoun-Davis T, Claypool K, Tang DG. PC3 human
cer Society. J Natl Med Assoc 2010;102:423–9. prostate carcinoma cell holoclones contain self-renewing tumor-initi-
3. Wolf AM, Wender RC, Etzioni RB, Thompson IM, D'Amico AV, Volk ating cells. Cancer Res 2008;68:1820–5.
RJ, et al. American Cancer Society guideline for the early detection 10. Patrawala L, Calhoun T, Schneider-Broussard R, Li H, Bhatia B, Tang
of prostate cancer: update 2010. CA Cancer J Clin 2010;60: S, et al. Highly purified CD44 þprostate cancer cells from xenograft
70–98. human tumors are enriched in tumorigenic and metastatic progenitor
4. Grubb RL III, Kibel AS. Prostate cancer: screening, diagnosis and cells. Oncogene 2006;25:1696–708.
management in 2007. Mol Med 2007;104:408–13. 11. Patrawala L, Calhoun T, Schneider-Broussard R, Zhou J, Claypool K,
5. Kasper S, Cookson MS. Mechanisms leading to the development of Tang DG. Side population is enriched in tumorigenic, stem-like cancer
hormone-resistant prostate cancer. Urol Clin North Am 2006;33: cells, whereas ABCG2þ and ABCG2- cancer cells are similarly tumor-
201–10. igenic. Cancer Res 2005;65:6207–19.
6. Tannock IF, de Wit R, Berry WR, Horti J, Pluzanska A, Chi KN, et al. 12. Patrawala L, Calhoun-Davis T, Schneider-Broussard R, Tang DG.
Docetaxel plus prednisone or mitoxantrone plus prednisone for Hierarchical organization of prostate cancer cells in xenograft tumors:
advanced prostate cancer. N Engl J Med 2004;351:1502–12. the CD44þalpha2beta1þ cell population is enriched in tumor-initiating
7. Tang DG, Patrawala L, Calhoun T, Bhatia B, Choy G, Schneider- cells. Cancer Res 2007;67:6796–805.
Broussard R, et al. Prostate cancer stem/progenitor cells: identifi- 13. Saini S, Majid S, Yamamura S, Tabatabai L, Suh SO, Shahryari V, et al.
cation, characterization, and implications. Mol Carcinog 2007; Regulatory role of mir-203 in prostate cancer progression and metas-
46:1–14. tasis. Clin Cancer Res 2011;17:5287–98.
14. Craft N, Chhor C, Tran C, Belldegrun A, DeKernion J, Witte ON, et al. 29. Garzon R, Marcucci G, Croce CM. Targeting microRNAs in cancer:
Evidence for clonal outgrowth of androgen-independent prostate rationale, strategies and challenges. Nat Rev Drug Discov 2010;9:
cancer cells from androgen-dependent tumors through a two-step 775–89.
process. Cancer Res 1999;59:5030–6. 30. Ivanovska I, Cleary MA. Combinatorial microRNAs: working together
15. Klein KA, Reiter RE, Redula J, Moradi H, Zhu XL, Brothman AR, et al. to make a difference. Cell Cycle 2008;7:3137–42.
Progression of metastatic human prostate cancer to androgen inde- 31. Yang K, Handorean AM, Iczkowski KA. MicroRNAs 373 and 520c are
pendence in immunodeficient SCID mice. Nat Med 1997;3:402–8. downregulated in prostate cancer, suppress CD44 translation and
16. Li H, Jiang M, Honorio S, Patrawala L, Jeter CR, Calhoun-Davis T, et al. enhance invasion of prostate cancer cells in vitro. Int J Clin Exp Pathol
Methodologies in assaying prostate cancer stem cells. Methods Mol 2009;2:361–9.
Biol 2009;568:85–138. 32. Baek D, Villen J, Shin C, Camargo FD, Gygi SP, Bartel DP. The impact
17. Liu C, Kelnar K, Liu B, Chen X, Calhoun-Davis T, Li H, et al. The of microRNAs on protein output. Nature 2008;455:64–71.
microRNA miR-34a inhibits prostate cancer stem cells and metastasis 33. Selbach M, Schwanhausser B, Thierfelder N, Fang Z, Khanin R,
by directly repressing CD44. Nat Med 2011;17:211–5. Rajewsky N. Widespread changes in protein synthesis induced by
18. Majid S, Saini S, Dar AA, Hirata H, Shahryari V, Tanaka Y, et al. microRNAs. Nature 2008;455:58–63.
MicroRNA-205 inhibits Src-mediated oncogenic pathways in renal 34. Saini S, Arora S, Majid S, Shahryari V, Chen Y, Deng G, et al. Curcumin
cancer. Cancer Res 2011;71:2611–21. modulates microRNA-203-mediated regulation of the Src-Akt axis in
19. Trang P, Medina PP, Wiggins JF, Ruffino L, Kelnar K, Omotola M, et al. bladder cancer. Cancer Prev Res (Phila) 2011;4:1698–709.
Regression of murine lung tumors by the let-7 microRNA. Oncogene 35. Valastyan S, Benaich N, Chang A, Reinhardt F, Weinberg RA. Con-
2010;29:1580–7. comitant suppression of three target genes can explain the impact of a