Inhibiting autophagosome turnover
targets therapy resistant leukemia
cells through metabolic disruption
Kaitlyn Dykstra, PhD
Department of Medicine – Leukemia
Lab of Eunice Wang, MD
April 3, 2018
Hematopoiesis
Hematopoietic Stem cell
Myeloid
Progenitor
Lymphoid
Progenitor
Plasma Cell
Megakaryocitic
precursor
RBC precursor
Granulocytic
precursor
T cell /NK
precursor
B cell
precursor
T cell
NK cell
Platelets
Macrophage
RBC
Neutrophils
Slide courtesy
Monica Guzman,
Weill Cornell
Pluripotent Stem cell
Myeloid
Progenitor
Lymphoid
Progenitor
Neutrophils
Plasma Cell
Megakaryocitic
precursor
RBC precursor
Granulocytic
precursor
T cell /NK
precursor
B cell
precursor
T cell
NK cell
Platelets
Macrophage
RBC
Slide courtesy
Monica Guzman,
Weill Cornell
Acute
Myeloid
Leukemia
(AML)
Leukemia
AML Survival Rates Remain Low
Leukemia Subtype 5-year survival
rate
Acute lymphocytic
(ALL)
79.0%
Chronic lymphocytic
(CLL)
68.2%
Acute myeloid
(AML)
26.9%
Chronic myeloid
(CML)
66.9%
SEER Cancer Statistics Review (CSR), 1975-2014 American Cancer Society
CML 14%
Other 9%
ALL 10%
CLL 35%
AML 32%
Estimated New Cases (2018)
60,300 total
Why is AML difficult to cure? • Heterogeneous disease
– 23 different mutated genes
• Compare to CML – Almost all cases are
caused by “Philadelphia chromosome” translocation
– Can be treated with tyrosine kinase inhibitors
The Cancer Genome Atlas Research Network, NEJM, 2013
Why is AML difficult to cure? • AML has an older
patient population – Less tolerant to
chemotherapy
– Less tolerant to bone marrow transplant
• Compared to ALL – Most patients are
pediatric
Sallan, SE. Hematology Am Soc Hematol Educ Program, 2006.
Outcomes are particularly poor in
older adults
Juliusson et al., Blood 119:3890-3899,, 2012 Juliusson et al., Blood 113(18): 4179-4187, 2009
<60
>60
Standard 7+3 induction
chemotherapy
– Treatment has not
changed since the
1970s
– Despite most patients
achieving remission,
relapse inevitably
occurs
Standard of Care for AML
Daunorubicin
or Idarubicin
Cytarabine
(AraC) +
7 days + 3 days
Relapse occurs due to minimal residual
disease including leukemia stem cells in
the bone marrow
Leukemia stem cell (LSC)
AML Blast Minimal residual disease (MRD)
• No signs or symptoms of
disease (remission)
• Disease below the threshold
of diagnostic detection by
microscopy or flow cytometry
How can we eradicate MRD?
Hypoxia in the bone marrow microenvironment
contributes to chemoresistance in AML
Oxygen Level
Cellularity
Portwood et al., Clinical Cancer Research, 2013
O2 levels
• 6% in normal tissues
• 1-3% in bone marrow
Hypoxia contributes to chemoresistance in
AML cell lines
* p<0.05 ** p<0.005 *** p<0.0005 n=3 Live Cells: Annexin V negative/7-AAD negative
How does hypoxia lead to
chemoresistance?
• AraC targets proliferating cells
• Hypoxia reduces cell proliferation
• Is reduced proliferation in hypoxia sufficient to explain AraC resistance?
Hypoxia
Quiescence Metabolic
homeostasis
Therapy resistance
Survival Quiescence
Hypoxia does not reduce the anti-proliferative
or DNA damaging effects of AraC
U n t A ra C U n t A ra C
0
2 5
5 0
% p
-H2
.AX
po
sit
ive
ce
lls
H E L -L u c
n s
2 1 % O 2 1 % O 2
U n t A ra C U n t A ra C
0
5 0 0 0 0 0
1 0 0 0 0 0 0
1 5 0 0 0 0 0
2 0 0 0 0 0 0
2 5 0 0 0 0 0
Liv
e c
ell
s/m
L
H E L -L u c
n s
2 1 % O 2 1 % O 2
Cells/ml (Trypan Blue) DNA Damage (p-H2.AX)
How does hypoxia lead to
chemoresistance? • Hypoxia has previously
been shown to upregulate anti-apoptotic proteins (e.g. Bcl-2) and downregulated pro-apoptotic proteins (e.g. Bax)
• Is reduced chemosensitivity due to changes in expression of these proteins?
Hypoxia
Quiescence Metabolic
homeostasis
Therapy resistance
Survival Quiescence Survival
Expression of anti-apoptotic Bcl-2 and pro-
apoptotic Bax are not affected by hypoxia
Bcl2
Actin
Normoxia Hypoxia Normoxia Hypoxia
Bax
Actin
How does hypoxia lead to
chemoresistance?
• Other survival
pathways altered
under hypoxia
– Autophagy?
Hypoxia
Quiescence Metabolic
homeostasis
Therapy resistance
Survival Survival
Autophagy is a prosurvival process that
is upregulated under hypoxic conditions
Normoxia Hypoxia
LC3-I
LC3-II
p62
actin
- + - + - +
HEL-Luc HL60 MOLM13
Hypoxia
DA
PI/
Cyto
ID
Autophagic flux is upregulated in AML cell
lines after prolonged exposure to hypoxia
2 1 % O 2 1 % O 2
0
5
1 0
1 5
2 0
2 5
3 0
3 5
%C
yto
ID P
os
itiv
e
Ce
lls
*
H E L -L u c H L 6 0 M O L M 1 3
0 .0
0 .2
0 .4
0 .6
Ra
tio
L
C3
BII
/I
N o rm o x ia
H y p o x ia
H E L -L u c H L 6 0 M O L M 1 3
0 .0 0
0 .2 5
0 .5 0
0 .7 5
1 .0 0
p6
2 N
orm
ali
ze
d
to N
orm
ox
ia
N o rm o x ia
H y p o x ia
* p<0.05 n=3
Does autophagy drive chemoresistance?
U n t A ra C S p a u 1A ra C
+ S p a u 1
U n t A ra C S p a u 1A ra C
+ S p a u 1
0
2 5
5 0
7 5
1 0 0
%L
ive
Ce
lls
ns ns*
2 1 % O 2 1 % O 2
M O L M 1 3
Autophagosome formation does not
drive chemoresistance under
hypoxia
U n t A ra C M R T A ra C
+ M R T
U n t A ra C M R T A ra C
+ M R T
0
2 5
5 0
7 5
1 0 0
%L
ive
Ce
lls
ns
*
*
2 1 % O 2 1 % O 2
M O L M 1 3
Cyto-ID
MRT68921 (ULK1 inhibitor)
Spautin-1 (Class II PI3K inhibitor)
* p<0.05 n=3
Autophagosome formation does not
drive chemoresistance under
hypoxia
N.S. LC3B-
LC3-I
LC3-II
actin
N.S. Atg7-
Atg7
actin
U n t A ra C U n t A ra C
0
2 5
5 0
7 5
1 0 0
%L
ive
C
ell
s
n o n -s i le n c in g s h R N A
A tg 7 s h R N A
L C 3 B s h R N A
2 1 % O 2 1 % O 2
M O L M 1 3
n s
n s
U n t A ra C U n t A ra C
0
2 5
5 0
7 5
1 0 0
%L
ive
C
ell
s
n o n -s i le n c in g s h R N A
A tg 7 s h R N A
L C 3 B s h R N A
2 1 % O 2 1 % O 2
M O L M 1 3
n s
BafA1 and Chloroquine inhibit autophagosome
turnover through distinct but related mechanisms
Bafilomycin A1 Chloroquine
Vacuolar-type H+-ATPase
inhibitor
• Increases lysosomal pH
Lysosomotropic agent
• Trapped in lysosomes
• Increases lysosomal
pH
Increased lysosomal pH reduces fusion competence.
U n t A ra C B a fA 1 A ra C
+ B a fA 1
U n t A ra C B a fA 1 A ra C
+ B a fA 1
0
2 5
5 0
7 5
1 0 0
%L
ive
Ce
lls
ns
*
*
2 1 % O 2 1 % O 2
M O L M 1 3
Inhibiting autophagosome turnover overcomes
hypoxia induced chemoresistance
U n t A ra C C Q A ra C
+ C Q
U n t A ra C C Q A ra C
+ C Q
0
2 5
5 0
7 5
1 0 0
%L
ive
Ce
lls
ns
*
*
2 1 % O 2 1 % O 2
M O L M 1 3
Bafilomycin A1 Chloroquine
* p<0.05 n=3
Relapse occurs due to minimal residual
disease (MRD) and leukemia stem cells
(LSCs) in the bone marrow
Leukemia stem cell (LSC)
AML Blast
Can late stage autophagy inhibitors
target leukemia stem cells?
LSC Characteristics
• CD34+/CD38-/Lin-
– not exclusively
• Preferentially localized in
hypoxic bone marrow
• Quiescent
• Increased autophagy
CB CD34+ BM CD34+
Figure 2: Basal level of LC3 II in normal CB CD34+ cells and primary AML samples. Representative single confocal sections of
cells expressing LC3 II (green), LAMP2 (red) and DAPI (blue). A:CD34+CB; B: normal BM;
C: CD34+CD38- cell; D:AML12; E:AML18; F: AML7.
A B C
D E F
Figure 2: Basal level of LC3 II in normal CB CD34+ cells and primary AML samples. Representative single confocal sections of
cells expressing LC3 II (green), LAMP2 (red) and DAPI (blue). A:CD34+CB; B: normal BM;
C: CD34+CD38- cell; D:AML12; E:AML18; F: AML7.
A B C
D E F AML (CD34+CD38-)
Figure 2: Basal level of LC3 II in normal CB CD34+ cells and primary AML samples. Representative single confocal sections of
cells expressing LC3 II (green), LAMP2 (red) and DAPI (blue). A:CD34+CB; B: normal BM;
C: CD34+CD38- cell; D:AML12; E:AML18; F: AML7.
A B C
D E F
Neng Yang, Guzman Lab LC3: green / Lysosome: red
Functional definition for LSC
Guzman Lab
Leukemia stem cell
Leukemia progenitor
Colony forming assays (CFU)
Functional definition for LSC
Guzman Lab
Leukemia stem cell
Leukemia progenitor
Xenotransplantation in sub-lethally irradiated immunodeficient mice
Late stage autophagy inhibitors inhibit
primary AML colony formation
1st
PTL AraC CLQ NH4CL Baf A1 3-MA0
20
40
60
80
100
120
FA1441
FG11452
AML032508
GJ1545
AML022806
AML072110
CP1215
FL1473
CF
U re
lativ
e to
UT
2nd
PTL AraC CLQ NH4CL Baf A1 3-MA0
20
40
60
80
100
FA1441
FG11452
AML032508
GJ1545
AML022806
AML072110
CP1215
FL1473
CF
U re
lative
to
UT
Neng Yang, Guzman Lab
Bafilomycin A1 selectively eradicates
leukemia stem cells
Neng Yang, Guzman Lab
4 weeks
AML
Primary AML
BafA1
(1mg/kg; 2X/week)
4 weeks
Why do late stage autophagy
inhibitors target LSCs?
• Inhibiting OXPHOS has
been demonstrated to
target leukemia stem cells
and therapy resistant cells
• Do late stage autophagy
inhibitors modulate
OXPHOS?
Viale et al., Cancer Research, 2015
Basal respiration
Maximal Respiration
Normoxia Untreated
Hypoxia Untreated
Normoxia 5nM BafA1
Hypoxia 5nM BafA1
Late stage autophagy inhibitors block OXPHOS
under hypoxia in AML cells
* p<0.05 ** p<0.005 *** p<0.0005 n=3
Late stage autophagy inhibitors block OXPHOS
under hypoxia in AML cells
Normoxia Untreated
Hypoxia Untreated
Normoxia 50µM CQ
Hypoxia 50µM CQ * p<0.05 ** p<0.005 *** p<0.0005 n=3
Basal respiration
Maximal Respiration
Late stage autophagy inhibitors block OXPHOS
in BafA1 sensitive primary AML samples
Normoxia Untreated
Hypoxia Untreated
Normoxia 25nM BafA1
Hypoxia 25nM BafA1
Patient Sample 1 Patient Sample 2 Patient Sample 3
Late stage autophagy inhibitors block OXPHOS
in BafA1 sensitive primary AML samples
Normoxia Untreated
Hypoxia Untreated
Normoxia 25nM BafA1
Hypoxia 25nM BafA1
Patient Sample 1
0
25
50
75
100
Unt 25nM Baf Unt 25nM Baf
Normoxia Hypoxia
%A
po
pto
tic
Patient Sample 1
n=1
Late stage autophagy inhibitors block OXPHOS
in BafA1 sensitive primary AML samples
Normoxia Untreated
Hypoxia Untreated
Normoxia 25nM BafA1
Hypoxia 25nM BafA1
Patient Sample 3
0
25
50
75
100
Unt 25nM Baf Unt 25nM Baf
Normoxia Hypoxia
%A
po
pto
tic
Patient Sample 3
n=1
Early stage autophagy inhibitors do not
modulate OXPHOS
* p<0.05 ** p<0.005 *** p<0.0005 n=3
OXPHOS inhibition by late stage autophagy
inhibitors is not due to decreases in
mitochondrial mass
MitoTracker MitoTracker
Unstained
Hypoxia Untreated
Hypoxia 5nM BafA1
Hypoxia 25µM CQ
Late stage autophagy inhibitors induce
mitochondrial ROS production
0
1
2
3
4
5
6
7
8
9
10
Untreated 25nMBafA1
50nMBafA1
50uM CQ 100uM CQ Untreated 25nMBafA1
50nMBafA1
50uM CQ 100uM CQ
Normoxia Hypoxia
Rel
ativ
e M
ito
SO
X M
FI
4h
8h
24h
n=2
BafA1 reduces glycolytic function in hypoxia
Normoxia Untreated
Hypoxia Untreated
Normoxia 5nM BafA1
Hypoxia 5nM BafA1
Glycolysis
Glycolytic Capacity
* p<0.05 ** p<0.005 *** p<0.0005 n=3
Chloroquine does not upregulated glycolytic
function in hypoxia
Glycolysis
Glycolytic Capacity
Normoxia Untreated
Hypoxia Untreated
Normoxia 50uM CQ
Hypoxia 50uM CQ * p<0.05 ** p<0.005 *** p<0.0005 n=3
Conclusions
• Hypoxia reduces chemosensitivity in
AML cell lines and increases
autophagic flux.
• Inhibiting the late stages of autophagy
overcomes both hypoxia induced
chemoresistance and eradicates LSCs
• Late stage autophagy inhibitors disrupt
both mitochondrial respiration and
glycolysis under hypoxia
Hypoxia
Glycolysis
Chemoresistance
Late Stage AI
OXPHOS
Leukemia Stem
Cell Maintenance
Future directions • Test BafA1 and chloroquine in
combination with AraC and in an MRD
model of AML
• Elucidate the connection between
blocking autophagosome
turnover/lysosomal disruption and
metabolism – Mitochondrial ROS
– Ca2+
• Identify new inhibitors with better
specificity/distribution – Bone marrow targeted nanoparticles
Acknowledgements Wang Lab
Eunice Wang, MD
Matthew Johnson
Scott Portwood, MS, MBA
Amanda Przespolewski, DO
Dirkje Hanekamp, MS
Megan Johnson
Weill Cornell
Monica Guzman, PhD
Neng Yang, PhD
NCI Cancer Center Support Grant
5P30 CA001656
- Flow and Image Cytometry
Shared Resource
- Immune Analysis Facility
Roswell Park Alliance Foundation
Jacquie Hirsch Leukemia Research Fund