scholarly journals Copy number alterations in B-cell development genes, drug resistance, and clinical outcome in pediatric B-cell precursor acute lymphoblastic leukemia

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Elisabeth M. P. Steeghs ◽  
Judith M. Boer ◽  
Alex Q. Hoogkamer ◽  
Aurélie Boeree ◽  
Valerie de Haas ◽  
...  
Blood ◽  
2014 ◽  
Vol 124 (21) ◽  
pp. 3582-3582
Author(s):  
Jeyanthy Eswaran ◽  
Paul Sinclair ◽  
Sirintra Nakjang ◽  
Christine J. Harrison

Abstract Acute lymphoblastic leukemia (ALL) is characterized by the expansion of immature hematopoietic cells in the bone marrow and blood. It is the most common childhood malignancy and is one of the major causes of death in children. Treatment improvements have increased the cure rate to more than 80% for children and about 40% for adults. However, current chemotherapy causes acute and long-term toxicity. Hence, there is a compelling need to understand the development of ALL and identify key players that may be used in targeted therapy. The B-cell receptor (BCR) and its precursor, pre-BCR, control B cell development, which is arrested in ALL. This blockade occurs at the first quality control checkpoint of B-cell development; the pre BCR checkpoint. To enhance our understanding of B-cell precursor (BCP)-ALL specific membrane associated signaling, we have investigated the functions of the recently discovered Chemokine factor like Marvel like Trans Membrane proteins (CMTM) that interact with BCR in pre-BCR checkpoint. We have identified heterozygous focal deletions of CMTM family genes, specifically CMTM6, 7 and 8 in BCP-ALL, including the subtype with intrachromosomal amplification of chromosome 21 (iAMP21). Similar focal deletions were found in iAMP21 xenograft models after serial passage, indicating their possible link to survival advantage. Although CMTM family proteins were first described in 2003, little is known about their physiological functions. The loss of a small gene cluster at chromosome 3p22, including CMTM6 and 7 has been reported in several cancers, including esophageal squamous cell, nasopharyngeal and lung carcinomas, indicating their potential roles as tumor suppressor genes. Among the CMTM family members, CMTM3 and 7 were initially identified as interacting partners of B-cell receptors. To characterize the CMTM mediated macromolecular assemblage in BCP-ALL, immunoprecipitation (IP) studies were performed using CMTM7 antibody. Initially, the expression of CMTM7 and the sensitivity of the CMTM7 antibodies were tested using various BCP-ALL cell lines. Due to their positive expression levels, the pre-B697 and NALM6 cell lines were selected for the IP studies. When the CMTM7-mediated membrane protein complex was isolated using CMTM7 antibody, we determined that the well-established tumor suppressor, B-cell linker (BLNK), interacted with CMTM7 in pre-B697 and NALM6. CMTM7-interaction partners are being verified by mass spectrometry. Next, to identify possible physiological functions, we performed a phylogenetic analysis and discovered that the CMTM family genes were homologous to myelin and lymphocyte (MAL) proteins, tricellulins, plasmolipins and occludin families, which comprise the tetra-spanin trans-membrane domain known as MARVEL (MAL and related protein for vesicle trafficking and membrane linking). These proteins are associated with cell communication and intracellular transport. To investigate the molecular mechanism of action of CMTM6, 7 and 8 in BCP-ALL and B cell development, we cloned, expressed and purified all three members and preliminary functional studies in-vitro indicated that they formed oligomers. Taken together, these data identify a critical membrane regulator, CMTM7, which may function as a tumor suppressor, communicating signals from membrane to cytosolic components through BLNK signaling in BCP-ALL. Disclosures No relevant conflicts of interest to declare.


Blood ◽  
2011 ◽  
Vol 118 (3) ◽  
pp. 544-553 ◽  
Author(s):  
Julia Hauer ◽  
Charles Mullighan ◽  
Estelle Morillon ◽  
Gary Wang ◽  
Julie Bruneau ◽  
...  

Abstract In human B-acute lymphoblastic leukemia (B-ALL), RAG1-induced genomic alterations are important for disease progression. However, given that biallelic loss of the RAG1 locus is observed in a subset of cases, RAG1's role in the development of B-ALL remains unclear. We chose a p19Arf−/−Rag1−/− mouse model to confirm the previously published results concerning the contribution of CDKN2A (p19ARF /INK4a) and RAG1 copy number alterations in precursor B cells to the initiation and/or progression to B-acute lymphoblastic leukemia (B-ALL). In this murine model, we identified a new, Rag1-independent leukemia-initiating mechanism originating from a Sca1+CD19+ precursor cell population and showed that Notch1 expression accelerates the cells' self-renewal capacity in vitro. In human RAG1-deficient BM, a similar CD34+CD19+ population expressed p19ARF. These findings suggest that combined loss of p19Arf and Rag1 results in B-cell precursor leukemia in mice and may contribute to the progression of precursor B-ALL in humans.


Blood ◽  
2011 ◽  
Vol 118 (21) ◽  
pp. 3580-3580
Author(s):  
Ilaria Iacobucci ◽  
Heike Pfifer ◽  
Annalisa Lonetti ◽  
Cristina Papayannidis ◽  
Anna Ferrari ◽  
...  

Abstract Abstract 3580 Introduction: Although treatment with tyrosine kinase inhibitors (TKIs) has revolutionized the management of adult patients with BCR-ABL1 -positive acute lymphoblastic leukemia (ALL) and significantly improved response rates, relapse is still an expected and early event in the majority of them. It is usually attributed to the emergence of resistant clones with mutations in BCR-ABL1 kinase domain or to BCR-ABL1 -independent pathways but many questions remain unresolved about the genetic abnormalities responsible for relapse after TKI and chemotherapy-based regimens. Patients and methods: In an attempt to better understand the genetic mechanisms responsible for this phenomenon, we have analyzed matched diagnosis-relapse samples from 30 adult BCR-ABL1 -positive ALL patients using high resolution Affymetrix single nucleotide polymorphism (SNP) arrays (GeneChip® Human Mapping 250K NspI, n=15 pairs and Genome-Wide Human SNP 6.0, n= 15 pairs). Genetic differences were analyzed in terms of copy number changes and loss of heterozygosity (LOH) events. 20 patients were enrolled in clinical trials of GIMEMA AL Working Party and treated with imatinib alone or in combination with conventional chemotherapy (40%) or dasatinib as frontline therapy (60%). The median age at diagnosis was 54 years (range 23–74) and the median blast cell count was 97% (range 60–99). The median time to relapse was 27 months (range, 9–104). 10 patients were treated according to the GMALL trials, a high-dose chemotherapy based protocol in combination with imatinib. The median age at diagnosis was 65 years (range 19–79) and the median leucocyte count was 37300/μl (range 5000 – 220000/μl). The median time to relapse was 9.8 months (range, 3 – 25). Results: First, we compared diagnosis and relapse samples for the presence of macroscopic (> 1.5 MB) copy number alterations (CNA). Novel acquired macroscopic CNAs were detected in 7/20 (35%) TKI relapse cases and included losses of 3p12-p14, 5q34, 9q34, 10q24 and 12p13-p12 and gains of 1q, 9q34-q33 and 22q and in 4/10 (40%) chemotherapy-relapse cases and included losses of 9p21 and 12q21–22 and gains of all chromosome 8 or part of it in 2 patients. Since no common patterns of acquired alterations were observed, it is likely that relapse may be due to a more generalized genetic instability rather than to specific mechanisms. Moreover, chemotherapy did not select resistant clones with higher number of alterations. 8/20 (40%) TKI resistant cases and 4/10 chemotherapy resistant patients harbored the same CNAs present in the matched diagnosis sample (losses of 9p21 in 7 cases, 7p and 22q11 in single cases and gains of chromosomes 1q, 4, 8q, 17q and 21), indicating a common clonal origin. In contrast, in 5/20 (25%) TKI resistant cases and 4/10 (40%) chemotherapy resistant patients macroscopic CNAs present at diagnosis were lost at relapse (losses of chromosomes 7, 11q, 14q, 15q, 16q and 19p and gains of 5q, 8q, 9q34 and 22q11). Thereafter, we compared diagnosis and relapse samples for microscopic CNAs (< 1.5 MB). The alteration most frequently acquired at relapse was loss of the tumor suppressor CDKN2A (53% vs 33 % of diagnosis). Other common acquired CNAs at relapse included gains of ABC transporter genes, such as ABCC1, ABCC6 (1q41) and BCL8 (15q11); losses affected EBF1 (5q33) and IGLL3 (22q11) genes involved in B-cell development, BTG1 (12q21) involved in cell cycle regulation and CHEK2 (22q12) involved in DNA repair. Interestingly, for all relapse cases analysis of IKZF1 deletions, identified in 80% of patients, demonstrated a clonal relationship between diagnostic and relapse samples, suggesting that this alteration is not acquired with relapse but it is maintained with fidelity from diagnosis working as a marker of disease. The majority (92%) of relapse samples harbored at least some of the CNAs present in the matched diagnosis sample, indicating a common clonal origin. Conclusions: Genomic copy number changes evolving from diagnosis to relapse have been identified demonstrating that a diversity of alterations contributes to relapse and with the most common alterations targeting key regulators of tumor suppression, cell cycle control, and lymphoid/B cell development. Supported by European LeukemiaNet, AIL, AIRC, Fondazione Del Monte di Bologna e Ravenna, FIRB 2006, PRIN 2009, Ateneo RFO grants, PIO program, Programma di Ricerca Regione – Università 2007 – 2009. Disclosures: Soverini: Novartis: Consultancy; ARIAD: Consultancy; Bristol-Myers Squibb: Consultancy. Baccarani:Pfizer Oncology: Consultancy; Novartis: Consultancy; BMS: Consultancy; Ariad: Consultancy; Novartis: Research Funding; Pfizer Oncology: Honoraria; Novartis: Honoraria; BMS: Honoraria; Ariad: Honoraria; Novartis: Membership on an entity's Board of Directors or advisory committees; BMS: Membership on an entity's Board of Directors or advisory committees; Ariad: Membership on an entity's Board of Directors or advisory committees. Ottmann:Novartis Corporation: Consultancy, Honoraria, Research Funding, Speakers Bureau. Martinelli:Novartis: Consultancy, Honoraria; BMS: Consultancy, Honoraria; Pfizer: Consultancy.


Blood ◽  
2012 ◽  
Vol 120 (21) ◽  
pp. 1466-1466
Author(s):  
Christopher D Chien ◽  
Elizabeth D Hicks ◽  
Paul P Su ◽  
Haiying Qin ◽  
Terry J Fry

Abstract Abstract 1466 Pediatric acute lymphoblastic leukemia (ALL) is the most common childhood malignancy. Although cure rates for this disease are approximately 90%, ALL remains one of the leading causes cancer-related deaths in children. Thus, new treatments are needed for those patients that do not respond to or recur following standard chemotherapy. Understanding the mechanisms underlying resistance of pediatric ALL to therapy offers one approach to improving outcomes. Recent studies have demonstrated the importance of communication between cancer cells and their microenvironment and how this contributes to the progression and therapeutic resistance but this has not been well studied in the context of ALL. Since the bone marrow is presumed to be the site of initiation of B precursor ALL we set out in our study to determine how ALL cells utilize the bone marrow milieu in a syngeneic transplantable model of preB cell ALL in immunocompetent mice. In this model, intravenously injected preB ALL develops first in the bone marrow, followed by infiltration into the spleen, lymph node, and liver. Using flow cytometry to detect the CD45.2 isoform following injection into B6CD45.1+ congenic recipients, leukemic cells can be identified in the bone marrow as early as 5 days after IV injection with a sensitivity of 0.01%-0.1%. The pre-B ALL line is B220+/CD19+/CD43+/BP1+/IL-7Ralpha (CD127)+/CD25-/Surface IgM-/cytoplasmic IgM+ consistent with a pre-pro B cell phenotype. We find that increasing amounts of leukemic infiltration in the bone marrow leads to an accumulation of non-malignant developing B cells at stages immediately prior to the pre-pro B cell (CD43+BP1-CD25-) and a reduction in non-malignant developing pre B cells at the developmental stage just after to the pre-pro B cell stage (CD43+BP1+CD25+). These data potentially suggest occupancy of normal B cell developmental niches by leukemia resulting in block in normal B cell development. Further supporting this hypothesis, we find significant reduction in early progression of ALL in aged (10–12 month old) mice known to have a deficiency in B cell developmental niches. We next explored whether specific factors that support normal B cell development can contribute to progression of precursor B cell leukemia. The normal B cell niche has only recently been characterized and the specific contribution of this niche to early ALL progression has not been extensively studied. Using a candidate approach, we examined the role of specific cytokines such as Interleukin-7 (IL-7) and thymic stromal lymphopoietin (TSLP) in early ALL progression. Our preB ALL line expresses high levels of IL-7Ralpha and low but detectable levels of TLSPR. In the presence of IL-7 (0.1 ng/ml) and TSLP (50 ng/ml) phosphSTAT5 is detectable indicating that these receptors are functional but that supraphysiologic levels of TSLP are required. Consistent with the importance of IL-7 in leukemia progression, preliminary data demonstrates reduced lethality of pr-B cell ALL in IL-7 deficient mice. Overexpression of TSLP receptor (TSLPR) has been associated with high rates of relapse and poor overall survival in precursor B cell ALL. We are currently generating a TSLPR overepressing preBALL line to determine the effect on early ALL progression and are using GFP-expressing preB ALL cells to identify the initial location of preB ALL occupancy in the bone marrow. In conclusion, or model of early ALL progression provides insight into the role of the bone marrow microenvironment in early ALL progression and provides an opportunity to examine how these microenvironmental factors contribute to therapeutic resistance. Given recent advances in immunotherapy for hematologic malignancies, the ability to study this in an immunocompetent host will be critical. Disclosures: No relevant conflicts of interest to declare.


Blood ◽  
2011 ◽  
Vol 118 (11) ◽  
pp. 3080-3087 ◽  
Author(s):  
Jinghui Zhang ◽  
Charles G. Mullighan ◽  
Richard C. Harvey ◽  
Gang Wu ◽  
Xiang Chen ◽  
...  

Abstract We sequenced 120 candidate genes in 187 high-risk childhood B-precursor acute lymphoblastic leukemias, the largest pediatric cancer genome sequencing effort reported to date. Integrated analysis of 179 validated somatic sequence mutations with genome-wide copy number alterations and gene expression profiles revealed a high frequency of recurrent somatic alterations in key signaling pathways, including B-cell development/differentiation (68% of cases), the TP53/RB tumor suppressor pathway (54%), Ras signaling (50%), and Janus kinases (11%). Recurrent mutations were also found in ETV6 (6 cases), TBL1XR1 (3), CREBBP (3), MUC4 (2), ASMTL (2), and ADARB2 (2). The frequency of mutations within the 4 major pathways varied markedly across genetic subtypes. Among 23 leukemias expressing a BCR-ABL1-like gene expression profile, 96% had somatic alterations in B-cell development/differentiation, 57% in JAK, and 52% in both pathways, whereas only 9% had Ras pathway mutations. In contrast, 21 cases defined by a distinct gene expression profile coupled with focal ERG deletion rarely had B-cell development/differentiation or JAK kinase alterations but had a high frequency (62%) of Ras signaling pathway mutations. These data extend the range of genes that are recurrently mutated in high-risk childhood B-precursor acute lymphoblastic leukemia and highlight important new therapeutic targets for selected patient subsets.


Blood ◽  
2014 ◽  
Vol 124 (21) ◽  
pp. 287-287
Author(s):  
Stefan Koehrer ◽  
Ondrej Havranek ◽  
R Eric Davis ◽  
Felix Seyfried ◽  
Greg Coffey ◽  
...  

Abstract A pivotal step during B-cell development is the expression of the precursor B-cell receptor (pre-BCR) by pre-B lymphocytes (cyto-Igµ+, surface-IgM-). The pre-BCR represents an immature form of the BCR and consists of two immunoglobulin heavy chains (IgH), two surrogate light chains (SLC) and the signal transducing adapter proteins Igα and Igβ. A functional pre-BCR drives proliferation of pre-B-cells, ensuring their further differentiation into mature B-cells. By immunophenotype, ~20% of B-cell acute lymphoblastic leukemia (B-ALL) cases originate from the pre-B-cell stage (pre-B-ALL) of lymphocyte development and might therefore also express the pre-BCR. In view of the importance of pre-BCR signaling for normal pre-B-cell development, we hypothesize that it is exploited by pre-B-ALL for malignant growth and proliferation. A hallmark of active pre-BCR signaling is the continuous internalization of pre-BCRs, resulting in low pre-BCR surface expression. Using this phenotype of active pre-BCR signaling (low pre-BCR expression and high phosphorylation of the pre-BCR associated kinases LYN and SYK), we identified pre-BCR+ ALL cell lines (RCH-ACV, SMS-SB and Nalm6) and xenograft expanded patient samples. To study the role of the pre-BCR in these cells, we rendered RCH-ACV and SMS-SB pre-BCR null by using CRISPR/CAS9 gene editing with guide RNAs specific for the hypervariable region (recombined V, D, and J segments) of their expressed IgH allele. As identified by flow cytometry for the pre-BCR, deficient RCH-ACV and SMS-SB cells exhibited reduced viability and impaired proliferation when compared to their pre-BCR+ controls (Figure 1). Pre-BCR- cells showed reduced baseline phosphorylation of CD19, VAV1 and AKT. Interestingly, BTK and ERK phosphorylation were not affected. These results provide evidence for the dependency of pre-BCR+ ALL on pre-BCR signaling and suggest selective involvement of the PI3K-AKT pathway. We also investigated the effects of pharmacological pre-BCR inhibition by treating pre-BCR+ and pre-BCR- ALL cell lines and xenograft expanded primary patient samples with PRT318, a small-molecule inhibitor of spleen tyrosine kinase (SYK). In pre-BCR+ ALL PRT318 blocked cell proliferation and selectively inhibited AKT phosphorylation, thus mimicking the effects of IgH knockout. Pre-BCR- ALL cells were resistant to PRT318. Key effectors of the pre-BCR during normal B-cell development are FOXO transcription factors. In line with this, we found reduced FOXO1 phosphorylation and increased FOXO1 total protein levels after IgH knockout as well as after treatment with PRT318. This was accompanied by an increase in the FOXO1 transcriptional targets p27 and BLNK, suggesting increased FOXO1 transcriptional activity in response to the inhibition of pre-BCR signaling. To study the contribution of FOXO1 to the effects of IgH knockout and SYK inhibition more thoroughly, we expressed constitutively active FOXO1 (FOXO1-3A) in the pre-BCR+ ALL cell line RCH-ACV and consequently assessed its effects on cell proliferation and protein expression. Similar to IgH knockout and PRT318, FOXO1-3A reduced cell proliferation and increased p27 and BLNK protein levels, confirming FOXO1 as an important downstream target of pre-BCR signaling in B-ALL. To identify additional effectors of the pre-BCR in B-ALL we performed gene expression profiling (GEP) to compare pre-BCR+ and pre-BCR- cells of RCH-ACV and SMS-SB. Gene set enrichment analysis (GSEA) showed that IgH knockout resulted in significant enrichment for gene sets associated with down-modulation of MYC activity. This was confirmed by Western blot analysis of MYC total protein levels, and consistent with the finding of reduced MYC protein in PRT318-treated and FOXO1-3A-expressing pre-BCR+ cells, all indicating that pre-BCR signaling modulates MYC activity through a mechanism involving SYK and FOXO1. In conclusion, we provide evidence for the dependence of certain B-ALL subgroups on pre-BCR signaling. According to our data this is mainly due to pre-BCR-induced inactivation of FOXO1 and the subsequent deregulation of MYC. Importantly, pharmacological inhibition of pre-BCR signaling with the SYK inhibitor PRT318 completely reversed these effects, therefore providing a rationale for the use of SYK inhibitors in pre-BCR+ subgroups of B-ALL. Figure 1: Figure 1 Figure 1. Disclosures Coffey: Portola Pharmaceuticals: Employment, Equity Ownership.


2016 ◽  
Vol 104 (3) ◽  
pp. 368-377 ◽  
Author(s):  
Nuket Yurur Kutlay ◽  
Esra Pekpak ◽  
Sule Altıner ◽  
Talia Ileri ◽  
Arzu Nedime Vicdan ◽  
...  

Blood ◽  
2014 ◽  
Vol 124 (21) ◽  
pp. 790-790 ◽  
Author(s):  
Maike Buchner ◽  
Eugene Park ◽  
Lars Klemm ◽  
Huimin Geng ◽  
Dragana Kopanja ◽  
...  

Abstract Background & Significance: Pre-B acute lymphoblastic leukemia (ALL) emerges in virtually all cases from B cell precursors that are arrested at the pre-B cell receptor checkpoint. In a gene expression survey of early B cell development, we found specific upregulation of FOXM1 at this developmental stage. FOXM1 belongs to the forkhead box transcription factor family and is a key regulator of cell growth by promoting cell cycle progression and has been implicated in progression of solid tumors. Therefore, we characterized the function and regulation of FOXM1 in normal B cell development as well as in pre-B ALL. Results: First, we verified the upregulation of FOXM1 during B cell development by qRT-PCR on sorted human and murine B cell progenitor populations. Then, we crossed Mb1-Cre tg mice to a Foxm1 conditional knockout mouse model (Foxm1fl/fl) and analyzed the early B cell populations according to the Hardy fractions. Despite the observed high expression of Foxm1 mRNA in fraction C’ and D, Foxm1 deletion did not alter B cell development. In order to investigate a potential role of FOXM1 in transformed B cells, we compared FOXM1 protein levels in patient-derived pre-B ALL samples with healthy B cells and B cell precursors and found 10-60-fold higher expression in the transformed B cell progenitors. To evaluate a potential predictive value of FOXM1 levels in patient-derived ALL samples, we measured FOXM1 mRNA levels at the time of diagnosis which strikingly correlate with risk stratification of ALL (intermediate-risk ALL n=31 vs. high risk ALL n=21; P=7.3e-5; BFM-REZ 2002). To further study the function and regulation of FOXM1, we cultured murine B cell precursors in the presence of IL7 and induced transformation with a retroviral BCR-ABL1 expression vector. BCR-ABL1 expression increased levels of FOXM1 compared to the normal IL7-dependent pre-B cells. Short-term inhibition of BCR-ABL1 did not affected protein levels of FOXM1. However, after 4 days of tyrosine kinase inhibition (TKI) treatment, FOXM1 protein levels were significantly downregulated in a dose-dependent manner. BCL2 overexpression prevented apoptosis induction by TKI but FOXM1 downregulation was retained. In addition, we found evidence that inactivation of FOXO factors by the PI3K/AKT pathway contributes to high FOXM1 expression in Ph+ALL. Overexpression of a constitutively active form of Akt to prevent activation of FOXO factors in the presence of TKI abrogated FOXM1 downregulation. Similarly, BCR-ABL1+ ALL derived from FOXO3a knockout mice prevented TKI-mediated FOXM1 reduction. Overexpression of a constitutively active form of FOXO3a but not FOXO1 significantly reduces levels of FOXM1 expression. In line with this, we found a significant inverse correlation of FOXM1 with FOXO3A mRNA levels in Ph+ ALL patients from the ECOG E2993 trial. However, the requirement of long-term treatment indicates, that, in addition to FOXO3a activation, epigenetic regulation of the FOXM1 promoter downstream of BCR-ABL1 is required. Consistent with this finding, the FOXM1 promoter region was found to be de-methylated in a large fraction of ALL. In order to further study FOXM1 function, we transduced pre-B cells derived from Foxm1fl/fl mice with BCR-ABL1 and with an inducible ERT2-Cre vector. Deletion of Foxm1 in BCR-ABL1-driven leukemia decreases cell viability, colony formation, and proliferative capacity in vitro as well as leukemia formation in vivo. FOXM1-deleted ALL cells revealed a strikingly higher sensitivity towards TKI-treatment compared to the control cells in Imatinib dose-response curves (IC50 EV:420 nM vs IC50 Cre-ERT2: 160 nM) as well as annexin V staining. We identified the ROS scavenger Catalase as a critical target of FOXM1 in mediating this drug resistance. As potential therapeutic agents to target FOXM1, we evaluated the effects of a previously described ARF peptide and the natural occurring antibiotic Thiostrepton. Both bind FOXM1 and inhibit its function as shown by reduced mRNA expression of FOXM1 target genes (Cyclin B1, PLK1, AURKB) and induced apoptosis in ALL and prolonged survival of patient-derived ALL transplant recipient mice. Conclusion: We have identified a critical function of the transcription factor FOXM1 in mediating proliferation and drug-resistance in B cell lineage ALL, but not in normal B cell progenitors and validated FOXM1 as a therapeutic target in a large fraction of drug-resistant B cell lineage ALL. Disclosures No relevant conflicts of interest to declare.


Sign in / Sign up

Export Citation Format

Share Document