2017 Progress Report: Identifying In Utero Exposures that are Risk Factors for Childhood Leukemia

EPA Grant Number: R836159C002
Subproject: this is subproject number 002 , established and managed by the Center Director under grant R836159
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).

Center: Center for Integrative Research on Childhood Leukemia and the Environment - 2015
Center Director: Metayer, Catherine
Title: Identifying In Utero Exposures that are Risk Factors for Childhood Leukemia
Investigators: Rappaport, Stephen M. , Hubbard, Alan , Whitehead, Todd
Institution: University of California - Berkeley
EPA Project Officer: Louie, Nica
Project Period: September 1, 2015 through August 31, 2019
Project Period Covered by this Report: September 1, 2016 through August 31,2017
RFA: Children's Environmental Health and Disease Prevention Research Centers (2014) RFA Text |  Recipients Lists
Research Category: Health , Children's Health

Objective:

The overarching research theme of CIRCLE is to identify additional in utero chemical risk factors for childhood acute lymphoblastic leukemia (ALL) in an ethnically diverse population, and to understand how chemicals increase risk via immunological, genetic and epigenetic mechanisms. In this project, researchers will characterize the totality of endogenous and exogenous chemical exposures in maternal and neonatal biospecimens to identify novel in utero risk factors for childhood ALL by applying new methods for profiling small molecules and protein (Cys34) adducts, and by investigating likely sources of exposure. Researchers will also perform targeted analyses of small molecules that represent biomarkers of exposure to factors that have been previously associated with childhood ALL.

Aim 1: Finalize methods for profiling small molecules and Cys34 adducts in archive newborn specimens (ANBS).

Aim 2: Measure and annotate “omic” features in ANBS extracts from childhood ALL cases and matched controls.

Aim 3: Process and compare data from cases and controls to find discriminating “omic” features. 

Aim 4: Perform semi-targeted analyses of small molecules chosen a priori as possible biomarkers of childhood ALL. 

Aim 5: Measure and compare “omic” features between pairs of ANBS and maternal blood samples collected during pregnancy from 200 mothers of childhood ALL cases and 400 mothers of control children.   

Progress Summary:

Aim 1:  Researchers developed methods for profiling small molecules and Cys34 adducts of human serum albumin (HSA) from ANBS and then measured these chemicals via UPLC-HRMS. Each set of untargeted analyses is performed with a 4.7-mm punch from an ANBS. In Year 6, researchers validated the untargeted UPLC-HRMS analysis of small molecules with ANBS punches from 10 control children and adjacent filter paper from the same Guthrie cards. The method included measurement of potassium in the extracts to facilitate normalization of analyte levels for hematocrit content. After statistical filtering, several thousand small-molecule features were available for analysis. This UPLC-HRMS method was applied to ANBS from an additional 97 control children to evaluate the bioinformatic workflow developed in the laboratory (Edmands, et al., 2017). Of the roughly 1,000 prevalent small-molecule features that were tested, multivariate linear regression detected significant associations with ethnicity (three metabolites) and birth weight (15 metabolites) after adjusting for multiple testing (Petrick, et al., 2017). Because measurement of potassium for normalization of hematocrit proved to be very time consuming, measurements of hemoglobin (Hb) were validated with control ANBS as a simpler normalization method. 

Finally, the adductomics methodology was modified to simplify extraction of HSA from ANBS. The final method involves extraction of a 4.7-mm punch with 50 microliters of water, of which 3 microliters is removed for measurement of Hb. Then 50 microliters of methanol is added to precipitate Hb and some serum proteins while leaving HSA in solution at a purity of 70-80%. The adductomics methodology currently is being validated with ANBS from 50 pairs of control children from the CCLS, half of whom had mothers who actively smoked during pregnancy and half whose mothers were nonsmokers.  

Aim 2: Extracts from 4.7-mm punches from the first 100 pairs of ANBS from childhood ALL cases and matched controls were analyzed using the same UPLC-HRMS platform and bioinformatics/statistics methods described above, and using potassium to normalize for hematocrit.  Matching was based on gender, date of birth, and the child’s ethnicity. These analyses were extended to an additional 200 childhood leukemia cases and matched controls, using Hb to normalize for hematocrit. Researchers also have commenced adductomics analyses of the first 100 case/control pairs

Aim 3:  Several strategies were applied for statistical analysis of small-molecule clusters from 300 case/control pairs described above. Statistical analyses currently are being conducted using a combination of parametric statistics (paired Student’s t-statistic) and machine-learning algorithms (lasso and random forest) for selection of variables that discriminate between childhood leukemia cases and controls. Small-molecule features that are included in the list of discriminating variables are being targeted for annotation by matching of MS/MS spectra and comparisons with reference standards (Edmands, et al., 2017).   

Aim 4:  In parallel with Aim 2, researchers detected several targeted molecules that had been selected as putative causes of childhood leukemia (metabolites of benzene and biomarkers of coffee). These targeted molecules are being tested to determine whether they are present at higher levels in childhood leukemia cases.

Aim 5: This aim was not pursued.

Future Activities:

Aim 1: The adductomics workflow and bioinformatics will be finalized.  Statistical methods for selection of variables that discriminate between childhood ALL cases and controls will be further evaluated and refined with iterations involving the sets of case/control pairs that already have been analyzed for small molecules and adducts.

Aim 2: The final batch of 100 childhood ALL cases and controls will be analyzed for small molecules and the adductomics analyses of the current 300 case/control pairs will continue. MS/MS spectra will be used for annotations based on comparisons with in-house libraries and online databases. 

Aim 3: Statistical tests will be performed with batches on samples from Aim 2 to compare levels of small molecules and Cys34 adducts between childhood leukemia cases and controls. Tests will be performed batch-wise to validate results and with data combined from all batches to increase power.

Aim 4: Targeted small molecules representing possible childhood leukemia risk factors (metabolites of benzene and biomarkers of coffee) will be measured and statistical tests performed on each batch of samples.

Aim 5: Small-molecule omics will commence with samples of maternal blood collected from mothers of 200 childhood ALL case/control pairs.


Journal Articles on this Report : 2 Displayed | Download in RIS Format

Other subproject views: All 4 publications 2 publications in selected types All 2 journal articles
Other center views: All 25 publications 23 publications in selected types All 23 journal articles
Type Citation Sub Project Document Sources
Journal Article Edmands WMB, Petrick LM, Barupal DK, Scalbert A, Wilson MJ, Wickliffe JK, Rappaport SM. compMS2Miner:an automatable metabolite identification, visualization, and data-sharing R package for high-resolution LC-MS data sets. Analytical Chemistry 2017;89(7):3919-3928. R836159 (2017)
R836159C002 (2017)
  • Abstract from PubMed
  • Full-text: ACS Publications-Full Text HTML
    Exit
  • Abstract: ACS Publications-Abstract
    Exit
  • Other: ACS Publications-Full Text PDF
    Exit
  • Journal Article Petrick L, Edmands W, Schiffman C, Grigoryan H, Perttula K, Yano Y, Dudoit S, Whitehead T, Metayer C, Rappaport S. An untargeted metabolomics method for archived newborn dried blood spots in epidemiologic studies. Metabolomics 2017;13(3):27 (19 pp.). R836159 (2017)
    R836159 (2018)
    R836159C002 (2017)
  • Full-text from PubMed
  • Abstract from PubMed
  • Associated PubMed link
  • Full-text: ResearchGate-Abstract & Full Text-PDF
    Exit
  • Abstract: Springer-Abstract
    Exit
  • Relevant Websites:

    CIRCLE - The Center for Integrative Research on Childhood Leukemia and the Environment Exit

    Progress and Final Reports:

    Original Abstract
  • 2016 Progress Report
  • 2018

  • Main Center Abstract and Reports:

    R836159    Center for Integrative Research on Childhood Leukemia and the Environment - 2015

    Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
    R836159C001 In Utero Chemical Exposures, Immune Status, and Childhood Leukemia
    R836159C002 Identifying In Utero Exposures that are Risk Factors for Childhood Leukemia
    R836159C003 Prenatal Exposures, Constitutive Genetics, DNA Methylation & Childhood Leukemia