Acute Myeloid Leukemia SUPPLEMENTARY APPENDIX Targeted therapy for a subset of acute myeloid leukemias that lack expression of aldehyde dehydrogenase 1A1
Maura Gasparetto, 1 Shanshan Pei, 1 Mohammad Minhajuddin, 1 Nabilah Khan, 1 Daniel A. Pollyea, 1 Jason R. Myers, 2 John M. Ashton, 2 Michael W. Becker, 3 Vasilis Vasiliou, 4 Keith R. Humphries, 5 Craig T. Jordan 1 and Clayton A. Smith 1
1Division of Hematology, University of Colorado, Aurora, CO, USA; 2Genomics Research Center, University of Rochester, NY, USA; 3De - partment of Medicine, University of Rochester Medical Center, NY, USA; 4Department of Environmental Health Sciences, Yale Univer - sity, New Haven, CT, USA and 5Terry Fox Laboratory, British Columbia Cancer Agency, Vancouver, BC, Canada
©2017 Ferrata Storti Foundation. This is an open-access paper. doi:10.3324/haematol. 2016.159053
Received: November 2, 2016. Accepted: March 8, 2017. Pre-published: March 9, 2017. Correspondence: [email protected] Supplemental Figure 1
A. Primary AMLs 10$
1$
0.1$
Rela%ve'Expression' 0.01$
0.001$
Figure 1A. qPCR for ALDH isoforms on 7 primary AML samples.
B. ALDH1A3 ALDH2 4.0 ALDH1A3 12 Normal ALDH2 Normal 4.0 12 Normal p<0.005 AML Normal 3.5 p<0.05 AML 10 p<0.005 AML
3.5 p<0.05 AML 10
3.0 8
3.0 8 Log2 Expression 2.5 Log2 Expression 6 Log2 Expression 2.5 Log2 Expression 6
2.0 4
2.0 4 BM (n=10) PB (n=10) BM (n=10) PB (n=10) AML (n=26) AML (n=26)
BM CD34+ (n=8) BM CD34+ (n=8) PB CD34+ (n=10) BM (n=10) PB (n=10) PB CD34+ (n=10) BM (n=10) PB (n=10) AML (n=26) AML (n=26)
BM CD34+ (n=8)PB CD34+ (n=10) BM CD34+ (n=8)PB CD34+ (n=10) ALDH3B1 ALDH8A1 10 3.0 ALDH3B1 ALDH8A1 AML 10 AML 3.0 p<0.005 9 Normal AML Normal p<0.005 AML 2.8 p<0.005 9 p<0.005 Normal 8 Normal 2.8 2.6 8 7 2.6 7 Log2 Expression Log2 Expression 2.4 6 Log2 Expression Log2 Expression 2.4 6 5 2.2
5 2.2 BM (n=10) PB (n=10) BM (n=10) PB (n=10) AML (n=26) AML (n=26)
BM CD34+ (n=8) BM CD34+ (n=8) PB CD34+ (n=10) BM (n=10) PB (n=10) PB CD34+ (n=10) BM (n=10) PB (n=10) AML (n=26) AML (n=26)
BM CD34+ (n=8)PB CD34+ (n=10) BM CD34+ (n=8)PB CD34+ (n=10)
Figure 1B. Differen al expression between normal CD34+ cells and AML of several ALDH isoforms. Data is from the GSE9476 data set. Supplemental Figure 1, cont
C. Correla3on% ALDH% coefficient% -0.10 -0.07 -0.03 0.00 0.03 0.07 0.10 -0.10 -0.07 -0.03 0.00 0.03 0.07 0.10 family% p%value% (r)%to% FPKM% genes% RPKM ALDH1A1% ALDH1A1% N/A% N/A% ALDH3B1% A0.241% 0.001% ALDH3B2% A0.177% 0.018% ALDH9A1% A0.126% 0.092% ALDH16A1% A0.100% 0.185% ALDH3A2% A0.093% 0.216% ALDH2% A0.026% 0.728% ALDH4A1% A0.018% 0.810% ALDH7A1% A0.012% 0.874% ALDH3A1% A0.006% 0.938% ALDH1B1% 0.048% 0.524% ALDH1A2% 0.051% 0.494% ALDH1L2% 0.061% 0.414% ALDH6A1% 0.177% 0.018% ALDH18A1% 0.191% 0.011% ALDH1L1% 0.199% 0.008% ALDH5A1% 0.202% 0.007% Favorable%(n=33)% Intermediate%(n=101)% Poor%(n=42)% N.D.%(n=3)% Favorable%(n=33)% Intermediate%(n=101)% Poor%(n=42)% N.D.%(n=3)% ALDH8A1% 0.271% 0.000% ALDH1A3% 0.536% <%0.0001% CytoAgene3c%Risks%
Figure 1C. Heat map representa on of expression of ALDH isoforms within AMLs expressing low to absent levels of ALDH1A1. Data is from the TCGA AML data set.
D.
10000
1000
100 RPKM
10
1
ALDH2 ALDH1A1ALDH1A2ALDH1A3ALDH1B1ALDH1L2 ALDH2-RALDH3A1ALDH3A2ALDH3B1ALDH4A1ALDH5A1ALDH6A1ALDH7A1ALDH8A1ALDH9A1 ALDH1A1-RALDH1A2-RALDH1A3-RALDH1B1-RALDH1L2-R ALDH3A1-RALDH3A2-RALDH3B1-RALDH4A1-RALDH5A1-RALDH6A1-RALDH7A1-RALDH8A1-RALDH9A1-RALDH16A1ALDH18A1 ALDH16A1-RALDH18A1-R
Figure 1D. Diagnosis and relapse (-R) AML subclones (LSC and Non-LSC) from 6 pa ents were analyzed for differen al expression of ALDH isoforms. Supplemental Figure 1, cont
E. Intermediate Risk Group (n=101) 1.0 1.0 Low (n=44) 0.9 ALDH1A1 Low (n=44) 0.9 ALDH1A1 0.8 High (n=57) 0.8 High (n=57) 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3
0.2 Overall Survival 0.2 Event Free Survival 0.1 Log-rank p=0.2138 0.1 Log-rank p=0.1545 0.0 0.0 0 25 50 75 100 125 0 25 50 75 100 125 months months
Intermediate Risk Group, <65y/o (n=49)
1.0 1.0 Low (n=23) 0.9 0.9 Low (n=23) ALDH1A1 High (n=26) ALDH1A1 0.8 0.8 High (n=26) 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3
0.2 Overall Survival 0.2 Event Free Survival 0.1 Log-rank p=0.1094 0.1 Log-rank p=0.1179 0.0 0.0 0 25 50 75 100 125 0 25 50 75 100 125 months months
Figure 1E. EFS and OS from all Intermediate risk group pa ents (n=101) and pa ents that are < 65 years old and were treated with standard chemotherapy regimens (n=49). A cut-off value of 1.5 RPKM was used to segregate pa ents into ALDH1A1-low (RPKM<1.5) and ALDH1A1-high (RPKM>1.5) groups. Data is from the TCGA AML data set. Supplemental Figure 2 B. A.
100K 150K 200K 250K FSC-A 100 50K
SSC-A 20 40 60 80 Figure 2A. Figure 2B. 0 0
10 2
10 10 Aldefluor
0
3
Aldefluor 10 10
1 4
Aldefluor
Aldefluor
10 10
2
5
Kasumi-1 MOLM-13 10 10 6
3
flow flow 100
FSC-A100K 150K 200K 250K
SSC-A 20 40 60 80 50K 0
0 10
2
cytometric cytometric 10 Aldefluor+DEAB 10
0
Aldefluor+DEAB 3
10 10
1
4
10 10
2 analysis of MOLM-13 cells and RNA- analysis of Kasumi-1 cells and
5
10 10
3 6
Rela-ve"Expression" 0.4" 1.2" FPKM% 0.8" 1.6" 0" 2" ALDH1A1" 0" 1" 2" 3" 4" 5" 6" ALDH1A1"
ALDH1A2" ALDH1A2" ALDH1A3" ALDH1A3" ALDH1B1"
ALDH1L1" ALDH1B1"
ALDH1L2" ALDH2" ALDH2" ALDH3A1" ALDH3A1"
ALDH3A2" qPCR ALDH3B1"
ALDH3B1" ALDH4A1" for ALDH isoforms.
Seq ALDH3B2" ALDH5A1" ALDH4A1" for ALDH isoforms. ALDH6A1" ALDH5A1"
ALDH6A1" ALDH7A1"
ALDH7A1" ALDH8A1" ALDH8A1" ALDH9A1" ALDH9A1" ALDH16A1" ALDH16A1"
ALDH18A1" ALDH18A1"
Supplemental Figure 2, cont
C. Kasumi-1 pLVX-Vector pLVX-ALDH1A1
5 105 10
10 4
3 10 4 10 Aldefluor Aldefluor 10 2 10 1 10 3 10 0 0 10 3 10 4 10 5 10 6 0 102 103 104 105 mCherry mCherry pLVX-Vector pLVX-ALDH1A1 ALDH1A1 10 1M 1M
8 800K 800K
600K 6 96% 600K 91%
SSC-A SSC-A 4 400K 400K
200K 200K 2 Rela ve Expression 0 0 0 3 4 5 6 0 10 10 10 10 0 103 104 105 106 mCherry mCherry
D. E. MOLM-13 100
pLVX-Vector pLVX-ALDH1A1 ALDH1A1 * ALDH1A1- 3 250K 250K 80 ALDH1A1+ 2.5 200K 200K 60 2 * 150K 150K 1.5 40 * * FSC-A 100K FSC-A 100K % Viability 1 50K 50K 20 Rela ve Expression 0.5 0 0 0 2 3 4 5 6 0 10 10 10 10 10 102 103 104 105 106 M] M] M] M] Aldefluor Aldefluor µ µ µ µ
Untreated ATO[1 4HC[6 4HC[12.5 M]+4HC[6 µ
ATO[1 F. Kasumi-3 2.5
2 100 100
80 80 1.5 82.5% 1.5%
60 60 1 SSC-A SSC-A 40 40
Rela ve expression 0.5 20 20
0 0 0 10 0 10 1 10 2 10 3 10 0 10 1 10 2 10 3 Aldefluor Aldefluor+DEAB Figure 2C. Restora on of ALDH1A1 in Kasumi-1 cells detected by qPCR and Aldefluor. Flow cytometric analysis of ALDH1A1+ and vector control cells showing infec on efficiency by m-Cherry and Aldefluor and post-sort enrichment phenotype.
Figure 2D. Restora on of ALDH1A1 in MOLM-13 cells detected by qPCR and Aldefluor. Flow cytometric analysis of ALDH1A1+ and vector control cells showing infec on efficiency by Aldefluor.
Figure 2E. Restora on of ALDH1A1 expression in MOLM-13 par ally blocked the effects of 4HC alone and in combina on with ATO depending on their concentra ons.
Figure 2F. ALDH1A1 ac vity in Kasumi-3 cell detected by qPCR (compared to CD34+HSCs) and Aldefluor.
Supplemental Figure 3
A. 10" AML[Aldefluor>10%]
- 100 Aldefluor 8" Aldefluor+ 80 Unsorted
6" 60 FPKM% 4" 40
% of Max 2" 20 0" 0 0 10 2 10 3 10 4 10 5 Aldefluor ALDH2" ALDH1A1"ALDH1A2"ALDH1A3"ALDH1B1"ALDH1L1"ALDH1L2" ALDH3A1"ALDH3A2"ALDH3B1"ALDH3B2"ALDH4A1"ALDH5A1"ALDH6A1"ALDH7A1"ALDH8A1"ALDH9A1" ALDH16A1"ALDH18A1" B. AML051804% 35" AML[Aldefluor>10%]
AML051804" - 30" 100 Aldefluor Aldefluor+ 25" 80 Unsorted
20" 60 FPKM% 15"
10" 40 % of Max
5" 20
0" 0 0 10 2 10 3 10 4 10 5 ALDH2" Aldefluor ALDH1A1"ALDH1A2"ALDH1A3"ALDH1B1"ALDH1L1"ALDH1L2" ALDH3A1"ALDH3A2"ALDH3B1"ALDH3B2"ALDH4A1"ALDH5A1"ALDH6A1"ALDH7A1"ALDH8A1"ALDH9A1" ALDH16A1"ALDH18A1" C. 10" AML[Aldefluor>10%]
- 100 Aldefluor 8" Aldefluor+ 80 Unsorted
6" 60
FPKM% 4" 40 % of Max 2" 20
0" 0 0 10 2 10 3 10 4 10 5 Aldefluor ALDH2" ALDH1A1"ALDH1A2"ALDH1A3"ALDH1B1"ALDH1L1"ALDH1L2" ALDH3A1"ALDH3A2"ALDH3B1"ALDH3B2"ALDH4A1"ALDH5A1"ALDH6A1"ALDH7A1"ALDH8A1"ALDH9A1" ALDH16A1"ALDH18A1"
Figure 3A-C. RNA-Seq analysis for ALDH isoforms from 3 AMLs [Aldefluor>10%] and Aldefluor measurements by flow cytometry. Primary AML samples containing both Aldefluor+ and Aldefluor- frac ons were sorted into the respec ve subsets for in vitro sensi vity studies (3 independent experiments). Supplemental Figure 4
A. AML [Aldefluor>0.1-1.5%]
5 10 10 5 10 5
4 4 4
10 10 10
3 3 10 10 103 CD123 CD71 HLA-DR
2 10 102 0 0 0
0 102 103 104 105 0 102 103 104 105 0 102 103 104 105 Aldefluor Aldefluor Aldefluor
5 10 5 10 5 10
10 4 10 4 Blasts 10 4
Blasts
Blasts 3 3 10 103 10 CD99 CD38 CD71
2 2 10 10 2 10 0 0 0 0 10 2 10 3 10 4 10 5 0 10 2 10 3 10 4 10 5 0 10 3 10 4 10 5 CD34 CD7 HLA-DR
Figure 4A. Flow cytometric analysis of an AML [Aldefluor>0.1-1.5%] with few ALDH+ cells (green) that overlay on the blast gates.
B. AML [Aldefluor>0.1-1.5%] 1"
0.8"
0.6"
0.4"
Rela%ve'Expression' 0.2"
0"
ALDH2" ALDH1A1"ALDH1B1" ALDH3A1"ALDH3B1"ALDH4A1"ALDH5A1"ALDH6A1"ALDH7A1"ALDH9A1" ALDH16A1"ALDH18A1" CD34+UCB 1.4"
1.2"
1"
0.8"
0.6"
0.4"
Rela%ve'Expression' 0.2"
0"
ALDH2" ALDH1A1"ALDH1B1" ALDH3A1"ALDH3B1"ALDH4A1"ALDH5A1"ALDH6A1"ALDH7A1"ALDH9A1" ALDH16A1"ALDH18A1"
Figure 4B. qPCR for ALDH isoforms from an AML [Aldefluor>0.1-1.5%] and from a CD34+UCB normal control. Supplemental Methods
Analysis of public databases For analysis of cytogene c risk groups, overall survival (OS) and event free survival (EFS), 165 AML specimens from The Cancer Genome Atlas (TCGA) with complete RNAseq, cytogene c risk, and survival outcome were u lized. In other analyses where cytogene c risk and survival outcome data were not needed, 179 AML specimens with complete RNA-Seq data were analyzed (TCGA).
Flow cytometry analysis Flow cytometric analysis was performed on a FACSCalibur device, FACSAria or an LSRII (Becton Dickinson (BD); San Jose, CA) and data was analyzed using FlowJo So ware (TreeStar; Palo Alto, CA). Cell sor ng was performed on FACSAria device equipped with 405nm violet, 488 nm argon and 633 HeNe lasers (BD).
Cytotoxicity studies Primary AML, Kasumi-1, Kasumi-3 and MOLM-13 cells were incubated overnight at 37°C with 1-5µM ATO and 6-25µM 4HC, or an equivalent volume of vehicle (EtOH or DMSO). Higher concentra ons of 4HNE (20-40µM) were used for short-term incuba on experiments where indicated in the figure legends. Viability was measured with the LIVE/DEAD® Fixable Near-IR Dead Cell kit (Invitrogen; Carlsbad, CA) for all experiments requiring fixa on/permeabiliza on. For ROS or apoptosis detec on, cells were incubated with surface an bodies for 30 min before adding 5µM MitoSOX, CellROX (Invitrogen, Molecular Probes) or 5µl AnnexinV (BD) for an addi onal 30 minutes at 37°C. Cells were resuspended in serum-free media or AnnexinV buffer, containing the cell viability dye 4’,6-diamidino-2-phenylindole (DAPI) at 1μg/ml.
PCR and Quan ta ve Real Time PCR (qPCR) Total mRNA was isolated with a PicoPure RNA isola on Kit (Thermo Fisher Scien fic; Waltham MA), mRNA samples were reverse transcribed using Transcriptor First Strand cDNA Synthesis Kit (Roche Diagnos cs USA) and quan ta ve real- me PCR (qPCR) was performed with LightCycler480 or 96 systems (Roche) using a LightCycler 480 SYBR Green I Master Mix or a Probes Master reac on mix for detec on of ALDH isoform RNAs.
Xenotransplant studies To test the drug sensi vity of the ALDH1A1- AML cell line MOLM-13 in vivo, 12 NSGS mice were treated intraperitoneally (IP) with busulfan (30mg/kg) then injected intravenously (IV) with 7x106 AML cells. Five days post injec on, 6 control mice were treated IP with saline while 6 mice were treated IP with 5µg/g ATO daily for 4 days together with 150µg/g Cy at day 1 and day 4. The mice were sacrificed by CO2 asphyxia on a er 6 more days and analyzed for MOLM-13 content by flow cytometry. To test primary AMLs in the xenotransplant model, first, three different primary ALDH1A1- AMLs were cultured in IMDM+Ins-Trans-Se+4%FBS+Pen/Strep (STEMCELL Technologies; Vancouver, BC, Canada) at a final concentra on of 1x106 cells/ml and treated with vehicle control or 4HC (30µM/ml) plus ATO (5µM/ml) for 24 hours. 5x106 treated cells were then transplanted per recipient into 8-10 week old NSGS mice; 12-13 weeks later, femora were flushed and human AML engra ment was analyzed by flow cytometry. To assess the sensi vity of primary AML cells in vivo to Cy and ATO, 5x106 ALDH1A1- or ALDH1A1+ AML cells were injected intravenously into 10 busulfan- condi oned NSGS mice. A er marrow engra ment was confirmed in 2 representa ve mice, Cy and ATO were administered as described above for the MOLM-13 studies. Fourteen or nineteen days later, marrow was harvested and human AML levels determined by flow cytometry. For studies of sensi vity of normal human HSCs and progenitors to 4HC and ATO, CD34+UCB cells were treated in vitro and transplanted as described above with human engra ment measured in marrow 13 weeks later using flow cytometry.
Study Approval NSGS mice used for xenotransplanta on studies were maintained in the University of Colorado in compliance with the University of Colorado Ins tu onal Animal Care and Use Commi ee and Na onal Ins tutes of Health Guidelines. Human bone marrow and peripheral blood samples were obtained from pa ents diagnosed with AML at the University of Rochester and the University of Colorado a er obtaining IRB approved informed consent. Cells typically had a post-thaw viability of >70%.