Marked sensitivity to mRNA targeting by LNA antisense in tumor cells without a delivery
system: Lessons learned
Yixian Zhang, Melissa Dumble, Yaming Wu, Zhengxing Qu, Victoria Shi, Jessica Kearney, Mary Mileski, Ying Gao, Jenny Pak, Steve Youngster, Lee M. Greenberger
Enzon Pharmaceuticals, Piscataway, New Jersey 08854
P H A R M A C E U T I C A L S
Abstract # 5647
yixian.zhang@enzon.com
LNA
LNALNA DNA
LNA
DNA L NA
DNA
LNA
LNA Gapmer
LNA Gapmer
LNA-ONs USED IN THE STUDIES
EZN-3920
EZN-3892
EZN-3889
EZN-4150
EZN-4176
EZN-2968
EZN-3046 scrambled control
Tar get
HER3
β-catenin
β-catenin
PI3KCA
Androgen Receptor
HIF-1α
None
%Growth
HLF
BEL-­‐7 404
HLE
SNU398
SNU739
SNU423
9810
7721
SNU449
NOZ
7402
JHH-­‐6
Hep3B
QCY7703
HUH6
SNU387
JHH2
SNU475
SNU368
SNU182
SNU886
JHH-­‐1
SNU878
HUH1
MHCC97-­‐H
PLC/5
HUH7
MHCC97-­‐L
QGY7701
SNU354
LIM050
LIF639
HCCLM3
JHH5
JHH7
SK-­‐HEP-­‐1
OCUG-­‐1
Hep G2
JHH4
10
8
0
Tumor Type
•Potency in target down-regulation varied among tumor samples
•Tumor type C demonstrated superior permissiveness to LNA-ON
Ct values
0
2
4
6
Days
8
10
12
100
1000
0
1200
40%
600
71%
400
Colo-205 (Colon)
200
0
5
10
Days
15
4
6
40
A549
8
Days20
EZN-­‐3892, 100 mg/kg, Q3D
(Lung)
10
12
14
120
ResistantNCI-­‐H1581 Parental
Colo205 Resistant
Resistant
Colo205 Parental
Colo205 Resistant
60
60
3
40
40
2
20
20
200
0
2
4
Saline
Saline
QDx1
QDx1
Q3Dx2 Q3Dx3
Q3Dx3 Q3Dx4
Q3Dx4
Q3Dx2
1
1
Tumor(singledose)
dose)
Tumor(single
120
0
EZN3920 ( µM)
EZN3046 ( µM)
100
80
4. Resistant
cells are cross resistant to
60
multiple LNA-ONs
60
40
20
40
20
0
NCI-­‐H1581 Parental NCI-­‐H1581 Resistant
EZN3920 ( µM)
EZN3046 ( µM)
6
8
10
12
TumorConc(single)µM
Conc(single)µM
Tumor
14
Tumor
Conc(multiple)µM
Tumor
Conc(multiple)µM
88
60
60
66
40
40
44
20
20
22
88
Time (h)
Time (h)
2424
4848
Pair 1
1528
2137
67
Chip Hybridizations to HG-U133 Plus 2.0
Pair 2
863
94
3
41
59
Pair 3
1044
Pair 2
879
142
3
34
Pair 3
996
Genes significantly differentially expressed in 3 isogenic pairs
RT-PCR, Immuno blot validation
1212
80
80
44
Pair 1
cDNA/IVT (in vitro transcription) amplification
1010
00
20
0
100
18 RNA samples [3 isogenic pairs (parental vs resistant), triplicate]
HCT116 (Colon)
100
100
00
40
20
Generation of 3 resistant cell lines
400
bearing mice were treated with 40 mg/kg EZN-­‐4176 at indicated dosing schedule.
Twenty four hours after the last dose, mice were sacrificed and tumors harvested. AR
mRNA level (grey bars) and EZN-­‐4176 concentration in the tumors (line) were
measured. (B) NSCLC HCC827 tumor-­‐bearing mice were treated with 30 mg/kg EZN-­‐
3920 either single dose or multiple doses (q3dx3). At indicated times after the last
dose, mice were sacrificed and analyzed for HER3 mRNA level and EZN-­‐3920
concentration. All data are mean + SEM (n = 5).
4
60
40
Fig. 8. Genomic analysis of the mechanism of resistance.
600
00
• mRNA antagonist concentration increases in tumor with repeated dosing
• Increasing concentration of mRNA antagonist in the tumor is associated with increased
down-regulation of target mRNA
Up-regulated Gene in
Resistant Lines
(based on 1.5 fold change)
Down-regulated Gene in
Resistant Lines
(based on 1.5 fold change)
Functional Assessment (effect RNAi on LNA target down-regulation)
Fig. 6. Intratumoral PK/PD correlation. (A) Castration-­‐resistant C4-­‐2b tumor-­‐
80
80
EZN3889 ( µM)
80
Genomic analysis of resistant cell lines
Intratumoral PK/PD relationship
5
100
80
NCI-­‐H1581 Resistant
EZN3889 ( µM)
20
10
2.
5
1.
3
0.
3
0.
1
0
Parental
Colo205 Parental
Parental
0
120
100
60
20uM
20
0
120
EZN3892 (µM)
40
EZN3892 20
(µM)
20uM
Days
120
120
40
60
20
10
2.
5
1.
3
0.
3
0.
1
0
20
10
2.
5
1.
3
0.
3
0.
1
0
5uM
60
80
20
10
2.
5
1.
3
0.
3
0.
1
0
Cell proliferation (%)
10uM
Genechip analysis, QC assessment
AR
ARmRNA
mRNA
μM
[EZN-4176],
[EZN-4176], μM
m
M
20
80
0
Tumor growth inhibition correlated with in vitro proliferation data and target mRNA down-regulation
in vivo. Similar data (not shown) were obtained with EZN-3920 LNA-ON
B
40
5uM 0
10uM
16
0
EZN-­‐3892, 50 mg/kg, Q3D
60
Colo205
Parental
Resistant
Colo205R
Colo205
Colo205R
100
20
10
2.
5
1.
3
0.
3
0.
1
0
40
80
2.5uM
2.5uM
100
120
20
10
2.
5
1.
3
0.
3
0.
1
0
20
100
UT
UT
80
2. Resistance is associated with reduced
KD of target
EZN-­‐3892, 25 mg/kg, Q3D
20
40
0
0
Control Oligo, 100 mg/kg, Q3D
800
0
0
2
60
20
Saline 1000
800
60
0
14
Saline
Control Oligo, 100 mg/kg, Q3Dx6
EZN-3892, 100 mg/kg, Q3Dx6
EZN-3892, 50 mg/kg, Q3Dx6
EZN-3892, 25 mg/kg, Q3Dx6
1200
200
80
100
80
120
20
10
2.
5
1.
3
0.
3
0.
1
0
NCI-H1581 (Lung)
300
60
resistant cells in response to EZN-3892
treatment
120
4
2
1
0.
5
0.
3
0.
1
0
400
100
80
3. No change in β-catenin or GAPDH mRNA in
20
10
2.
5
1.
3
0.
3
0.
1
0
73% TGI
800
R^2
0.969
0.601
0.969
0.452
120
120
EZN-3892 , 25 mg/kg, Q3D
400
D
15.27
122.47
19.674
124.145
Cell proliferation (%)
1200
C
0.945
0.664
0.573
0.5
Cell proliferation (%)
studies)
EZN-3892, 50 mg/kg, Q3D
500
B
2.649
1.849
2.795
1.991
1. Resistant cells do not demonstrate growth
inhibition compared to parental (two independent
EZN-3892, 100 mg/kg, Q3D
600
A
116.348
102.638
115.735
102.569
GAPDH
GAPDH
β-catenin
b-­‐catenin
20
10
2.
5
1.
3
Cell proliferation (%)
0.
3
0.
1
0
y = ( (A - D)/(1 + (x/C)^B ) ) + D:
3892 Colo TCF2 (3892 Colo 205 TCF 2: Concentrati...
3892 Colo 205 R2 (3892 Colo R2: Concentration vs ...
3889 Colo 205 TCF2 (3889 Colo 205 TCF2: Concent...
3889 Colo 205 R2 (3889 Colo 205 R2: Concentratio...
Control LNA, 100 mg/kg, Q3D
700
uM
[EZN-3892],
µM
Parental
100
4
2
1
0.
5
0.
3
0.
1
0
EZN-3892, 50 mg/kg, Q3Dx5
10
20
10
2.
5
1.
Cell proliferation (%)3
0.
3
0.
1
0
800
1
Resistant
20
10
2.
5
1.
3
0.
3
0.
1
0
0.1
Cell proliferation (%)
0
0.01
Parental
4
2
1
0.
5
0.
3
0.
1
0
%Growth
20
4
2
1
0.
5
0.
3
0.
1
0
EZN-3892, 100 mg/kg, Q3Dx5
1600
00
A A A A A A A A A A A A A B B B B B C C C C C C C C C C C C C C C C
40
20
10
2.
5
1.
3
0.
3
0.
1
0
900
100
100
2
60
Cell proliferation (%)
2000
EZN-4176-SCR, 100 mg/kg, Q3Dx5
120
120
4
80
Cell proliferation (%)
2400
AA
6
100
100
Saline
20
18
16
14
12
10
8
6
4
2
0
Resistant
% RQ (GAPDH/B-­‐Catenin
1000
Saline
Down-regulation of mRNA in
primary tumor cells
cells from 3 tumor types were prepared from cancer patients according to Precision
Therapeutics Inc’s procedure, plated, and treated with EZN-­‐3920 without lipofection at various
concentrations. Cells were harvested 2 days after treatment. qRT-­‐PCR assay was performed.
Pancreas Prostate Skin
120 simply dissolved in HCT116 tumor bearing mice were dosed IV with EZN-­‐3892 or control LNA saline with indicated doses and schedules. 1400
Fig. 3. Down-­‐regulation of mRNA in primary tumor cells by LNA-­‐ON. Primary tumor
Lung
Fig. 5. Effect of LNA-­‐ON on tumor growth in mice. NCI-­‐H1581, Colo-­‐205, A549, and 0.6
β-catenin mRNA down-regulation > 50% was observed in 69% of HCC cell
lines. Similar data (not shown) were obtained with EZN-2968 HIF-1α LNA-ON
Liver
•SKBR3, Colo-­‐205, NCI-­‐H747 and NCI-­‐H1581 cells were more sensitive to EZN-­‐3892 than others (EC50 ~ 250 nM-­‐8 µM) in growth assay
• β-­‐catenin mRNA was down-­‐regulated efficiently (EC50~ 200 nM to 3 µM) in these cell lines
•Colo-­‐205 and NCI-­‐H1581 were chosen for in vivo efficacy test because they grow well on the flank of mice; HCT116 and A549 were also chosen for in vivo efficacy test because potency in growth inhibition and target down-­‐regulation were poor (EC50 > 20 µM)
0
0
Glioma
Colo 205 Resistant @ 8uM EZN3892 R
140
120
0.8
0.4
Colon
% RQ (GAPDH/B-­‐Catenin
1
Breast
Mean Tumor Volume (mm3)
Knockdown of β-­‐catenin mRNA in 39 HCC cell lines treated with 10 uM of EZN-­‐3892 for 72 h
were tested for their responses to LNA-­‐ONs.
Mean Tumor Volume (mm3)
carcinoma cell (HCC) lines were treated with EZN-­‐3892 at 10 µM. Cells were harvested 3 days
after treatment. β-­‐catenin qRT-­‐PCR assay was performed.
Fig. 8. Characterization of type 2 resistant cell lines. Parental and resistant cells Mean Tumor Volume (mm3)
MDA-MB361
MDA-MB468
HCT116
U87MG
786-O
PC3
LNCaP
H1975
HCC827
SKOV3
MCF-7
Fig. 2. Effect of LNA-­‐ONs without lipofection on mRNA. Multiple human hepatocellular
mRNA
mRNA
mRNA
3rd generation LNA-­‐based antisense technology results in: • Improved Tm (+2 to +80 per LNA monomer)
• Excellent plasma stability (T1/2 = 15 hrs)
• Long tissue residence time (Liver T1/2 = weeks)
• Short sequence (14-­‐16-­‐mer)
• IV administration without delivery vehicle
Compounds
20
EZN-3920 Concentration (µM)
EZN-3920 Concentration (µM)
•
O
40
Cell line 2
Cell line 3
Cell line 4
Traffic ↓ Two types of resistant mechanisms were identified:
Type 1: Level of target protein reduced compared to parent cell line (cell line 1)
Type 2: Level of target protein remained the same; down-­‐regulation efficiency of mRNA by LNA-­‐ON reduced (cell lines 2-­‐3)
HER3/GAPDH (%RQ)
(%R
RQ)
HO
60
Cell line 1
Gene ID
Chronic exposure
to increasing
amount LNA-ONs
77days
daystreatment
treatment
AR
R mRNA,
%Con
ntrol
AR
mRNA, %Control
2’
RNase H
Target ↓
EZN-3892
(µM)
EZN-3892
(µM)
Down-regulation mRNA by LNA-ON
in hepatocellular cell lines
IC50s for HER3 mRNA down-­‐regulation, µM
O
5 mM
48 hous
0.2
RNase H
RNase H
H
RNase H
Fig. 7. Generation of LNA-­‐ON-­‐resistant cell lines. EZN-­‐3892 to cell cultures at doses up to 20 µM without lipofection for 7 days prior to MTT assay. 80
0
1.2
Generating cells resistant to LNA-ON
Figure 4. Inhibition of solid tumor cell lines growth by LNA-­‐ON without lipofection. Cells from various tumor histologies were treated by adding M
μM
[4176], m
μM
requires DNA:RNA hybrid
4’
100
Correlation between in vitro and in
vivo growth inhibition
Mean Tumor Volume (mm3)
WHY USE LNA TECHNOLOGY?
RNase H Cleavage of mRNA
HO
120
Relative m RNA level
LNA-ONs were either added to tissue culture media (i.e. no
transfection) in vitro or prepared in saline and given IV in vivo.
Endpoints were measured by qRT-PCR, MTT, Western Blot
analysis, and tumor size. Gene Expression Profiling was
performed by Asuragen, Inc. Concentration of LNA-ONs in tissues
was measured by LC/MS/MS. Primary tumors samples obtained
and treated by Precision Therapeutics Inc (PTI) and downregulation analysis carried out by Enzon Pharmaceuticals.
Base
2 days after treatment. PIK3CA qRT-­‐PCR assay was performed.
58% of the cell line show > 50% PIK3CA mRNA down-regulation without
lipofection. Similar data were obtained with EZN-3920 HER3 LNA-ON
(Zhang et al, 2011 Gene Therapy. 18:326)
METHODS
Locked Nucleic Acid Fig. 1. Effect of LNA-­‐ONs without lipofection on mRNA in human
cancer cell lines.
Cells were treated with EZN-­‐4150 at 5 µM. Cells were harvested
PIKCA mRNA knockdown in multiple cell lines
Colo205
Unlike siRNAs, single-stranded locked nucleic acid-based
antisense oligonucleotides (LNA-ONs) have shown the ability to
down regulate mRNAs in vitro and in vivo without any delivery
systems such as transfection reagents or liposomes. Hence, LNAONs may have significant advantages as a therapeutic compared
to siRNAs. Investigation of LNA-ONs that target HIF-1α, survivin,
or androgen receptor in cancer patients is ongoing. We assess
here 1) down-regulation efficiency using LNA-ONs against
PIK3CA, β-catenin, and HER3 in multiple cell lines and in cells
prepared directly from patient tumors, 2) the correlation of target
down-regulation and growth inhibition in vitro with tumor growth
inhibition in vivo, 3) the correlation of intratumoral LNA-ON
concentration with target down-regulation in vivo in xenograft
models, and 4) genes that are differentially expressed in LNAONs resistant cells as compared to permissive cells.
Down-regulation of mRNA by LNAON in cell lines
PIK3CA mRNA, %Control
INTRODUCTION
Targets assessment for pharmacological manipulation or patient selection
•Preliminary results suggest that genes are differentially expressed in resistant cells
•Validation of these differentially expressed genes and understanding of their roles in target
down-regulation may assist in selection of tumors and patients who may benefit the most
from LNA-ON therapy
CONCLUSIONS
• LNA-­‐based oligonucleotides (LNA-­‐ONs) result in down-­‐regulation of target in vitro •
•
•
•
and in vivo in many cancer cell lines including primary tumor cells without the need for transfection
In vitro activities as assessed by potency in growth inhibition and target down-­‐
regulation may predict in vivo anti-­‐tumor response
LNA-­‐ONs administered IV achieve tumor concentration required to effectively down regulate target in vitro (Zhang et al, 2011 Gene Therapy. 18:326)
Cells resistant to LNA-­‐ONs have been generated
Understanding the mechanism of resistance to LNA-­‐ON may assist in the selection of tumors and patients for therapy with LNA-­‐ON
LNA-ONs are being developed by Enzon under a license with Santaris Pharma A/S.
AACR, Chicago, IL 2012