Novel variants in the PRDX6 Gene and the risk of Acute Lung Injury following major trauma

Article (PDF Available)inBMC Medical Genetics 12(1):77 · May 2011with22 Reads
DOI: 10.1186/1471-2350-12-77 · Source: PubMed
Abstract
Peroxiredoxin 6 (PRDX6) is involved in redox regulation of the cell and is thought to be protective against oxidant injury. Little is known about genetic variation within the PRDX6 gene and its association with acute lung injury (ALI). In this study we sequenced the PRDX6 gene to uncover common variants, and tested association with ALI following major trauma. To examine the extent of variation in the PRDX6 gene, we performed direct sequencing of the 5' UTR, exons, introns and the 3' UTR in 25 African American cases and controls and 23 European American cases and controls (selected from a cohort study of major trauma), which uncovered 80 SNPs. In silico modeling was performed using Patrocles and Transcriptional Element Search System (TESS). Thirty seven novel and tagging SNPs were tested for association with ALI compared with ICU at-risk controls who did not develop ALI in a cohort study of 259 African American and 254 European American subjects that had been admitted to the ICU with major trauma. Resequencing of critically ill subjects demonstrated 43 novel SNPs not previously reported. Coding regions demonstrated no detectable variation, indicating conservation of the protein. Block haplotype analyses reveal that recombination rates within the gene seem low in both Caucasians and African Americans. Several novel SNPs appeared to have the potential for functional consequence using in silico modeling. Chi2 analysis of ALI incidence and genotype showed no significant association between the SNPs in this study and ALI. Haplotype analysis did not reveal any association beyond single SNP analyses. This study revealed novel SNPs within the PRDX6 gene and its 5' and 3' flanking regions via direct sequencing. There was no association found between these SNPs and ALI, possibly due to a low sample size, which was limited to detection of relative risks of 1.93 and above. Future studies may focus on the role of PRDX6 genetic variation in other diseases, where oxidative stress is suspected.

Figures

RESEARCH ARTICLE Open Access
Novel variants in the PRDX6 Gene and the risk of
Acute Lung Injury following major trauma
Melanie Rushefski
2,3
, Richard Aplenc
2,3
, Nuala Meyer
1
, Mingyao Li
2
, Rui Feng
2
, Paul N Lanken
1
, Robert Gallop
2
,
Scarlett Bellamy
2
, A Russell Localio
2
, Sheldon I Feinstein
4
, Aron B Fisher
4
, Steven M Albelda
1
and
Jason D Christie
1,2*
Abstract
Background: Peroxiredoxin 6 (PRDX6) is involved in redox regulation of the cell and is thought to be protective
against oxidant injury. Little is known about genetic variation within the PRDX6 gene and its association with acute
lung injury (ALI). In this study we sequenced the PRDX6 gene to uncover common variants, and tested association
with ALI following major trauma.
Methods: To examine the extent of variation in the PRDX6 gene, we performed direct sequencing of the 5 UTR,
exons, introns and the 3 UTR in 25 African American cases and controls and 23 European American cases and
controls (selected from a cohort study of major trauma), which uncovered 80 SNPs. In silico modeling was performed
using Patrocles and Transcriptional Element Search System (TESS). Thirty seven novel and tagging SNPs were tested
for association with ALI compared with ICU at-risk controls who did not develop ALI in a cohort study of 259 African
American and 254 European American subjects that had been admitted to the ICU with major trauma.
Results: Resequencing of critically ill subjects demonstrated 43 novel SNPs not previously reported. Coding regions
demonstrated no detectable variation, indicating conservation of the protein. Block haplotype analyses reveal that
recombination rates within the gene seem low in both Caucasians and Africa n Americans. Several novel SNPs
appeared to have the potential for functional consequence using in silico modeling. Chi
2
analysis of ALI incidence
and genotype showed no significant association between the SNPs in this study and ALI. Haplotype analysis did
not reveal any association beyond single SNP analyses.
Conclusions: This study revealed novel SNPs within the PRDX6 gene and its 5 and 3 flanking regions via direct
sequencing. There was no association found between these SNPs and ALI, possibly due to a low sample size,
which was limited to detection of relative risks of 1.93 and above. Future studies may focus on the role of PRDX6
genetic variation in other diseases, where oxidative stress is suspected.
Keywords: Peroxiredoxin, Acute Lung Injury, Oxidant Stress, Genetic Polymorphisms
Background
Acute Res piratory Distress Syndrome (ARDS) and Acute
Lung Injury (ALI) affect 100,000-150,000 patients each
year in the United States alone [1,2]. ALI is an inflamma-
tory syndrome chara cterized by acute respiratory failure
due to non-cardiogenic pulmonary edema and hypoxe-
mia [3]. Oxi dant st ress caused by reactive oxygen species
(ROS ) is thought to be a major contributor to the patho-
genesis of ALI. ROS can be generated by inflammatory
cells or pulmonary endothelium and cause damage to
proteins, DNA, and lipids [4].
The risk of developing ALI/ARDS is not un iformly dis-
tributed in the critically ill population, suggesting a
genetic influence on outcomes [ 5]. Per oxiredoxins are a
superfamily of non-heme and non-selenium p eroxidases
that are widely distributed throughout all phyla [6]. The
Peroxiredoxin 6 gene (PRDX6) is located on chromosome
1q24 and is appro ximately 12 Kb in length, containing 5
exons. The Prdx6 protein encoded by PRDX6 is involved
* Correspondence: jchristi@mail.med.upenn.edu
1
Division of Pulmonary and Critical Care Medicine, Department of Medicine,
University of Pennsylvania School of Medicine, 3600 Spruce Street,
Philadelphia, 19104, USA
Full list of author information is available at the end of the article
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
© 2011 Rushefski et al; licensee BioM ed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution Lice nse (http://creativecommons.org/licenses/by/2.0), which perm its unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.
in redox regulation of the cell and has been shown in cell
and animal models to be protective against oxidative
injury [7]. Prdx6 also h as phospholipase A
2
activity and
has an important role in lung surfactant metabolism [7].
The protein product, Prdx6, has been shown to affect the
cellular level of H
2
O
2
produced in cells stimulated with
platelet-derived growth factor or tumor necrosis factor-
a, and modulating signaling induced by those li gands [8],
thus indicating that Prdx6 can have an effect on cytokine
levels and cell signaling cascades. Recent studies suggest
that Prdx6 is only active following heterodimerization
with glutathione-S-transferase pi, indicating that there i s
an important i nteraction between Prdx6 and GSTpi [6].
Despite these important functions, l ittle is known about
genetic variation within the PRDX6 gene [9] and its asso-
ciation with ALI.
In order to determine if variation within PRDX6 is
ass ociated with ALI risk in either the African American
(AA) or European American (EA) populations, we per-
formed dire ct seque ncing of the 5 UTR, ex ons, introns,
and the 3 untranslated region (UTR) in 48 subje cts (2 5
African Americans and 23 European Americans) and
identified 80 variants, many of which have not been pre-
viously reported. Eighteen of the eighty variants, a long
with 19 tagging SNPs selected using HapMap http://
hapmap.ncbi.nlm.nih.gov/, were tested for association
with ALI using a custom genotyping platform.
Methods
Patient population
Between 1999 and 2006, patients were enrolled in a major
trauma cohort study designed to study molecular risks for
acute lung injury [10-12]. Participants met the following
inclusion criteria: 1) admission to the intensive care unit
(ICU) as a result of acute trauma directly from the field or
via that hospitals Emergency Department; and 2) have an
Injury Severity Score (ISS) 16 as calculated on the basis
of information available during their first 24 hours of hos-
pitalization. The following demographic and clinical vari-
ables were collected upon admission to t he ICU: age,
gender, ISS, blunt mec hanism, and acute physiology and
chronic health evaluation (APACHE) (Table 1). Exclusion
criteria were death or discharge from the ICU within
24 hours of admi ssion, less than 13 years of age, curren t
or past evidence of congestive heart failure (CHF) or
recent acute myocardial infarction, severe chronic respira-
tory disease, morbid obesity, burns on over 30% of
the total body surface area, and lung or bone marrow
transplant [10].
The d efinition of ALI was in accordance with the
American European Consensus Conference (AECC) [3].
ALI and ARDS were defined as: acute onset; bila teral
pulmonary infiltrat es on chest X-ray consistent wit h
pulmonary edema; absence of evidence of left atrial
hypertension; a nd poor systemic oxygen ation, and a
ratio of arterial oxygen (PaO2) to the fraction of
inspired oxygen (FiO2) less than or equal to 300 for ALI
and 200 for A RDS [3]. All chest x-rays w ere reviewed
independently by 2 trained observers. In our population,
greater than 85% of subjects meeting criteria f or ALI
also met criteria for ARDS.
Clinical Data and Biosample Collection
Clinical data were collected by trained study nurses
using a standardized research case report form designed
for the trauma cohort study. Blood for analysis was
obtained from residual blood samples in tubes contain-
ing ethyledenediaminetetraacetic acid (EDTA) that had
been previously drawn for other clinical purposes. Study
personnel collected residual samples each day, centri-
fuged, and separated the buffy coat layers, which were
frozen at -80°C [10]. All clinical and biosample collec-
tion protocols were approved by the institutional review
board (IRB) at the University of Pennsylvania School o f
Medicine under a waiver of informed consent.
PRDX6 resequencing
Genomic DNA was extracted from whole blood using
Qiagen Qiamp DNA Blood Midi Kits (Qiagen USA) and
stored in the provided tris-EDTA buffer. DNA from 25
African American and 23 European American subjects
selected from the major trauma cohort, with ALI status
equally distributed within each group, were selected fo r
sequencing of PCR fragments, providing a power of 99%
to detect minor allele frequencies of at least 5% [13].
PCR primers for 4 Kb upstream of the ATG start site,
all exons, introns, and 4 Kb of the 3 UTR were
designed using PCRoverlap (ChildrensHospitalofPhi-
ladelphia (CHOP) bioinformatic s c ore) to generate
amplicons between 600 and 800 bp that overlapped by
Table 1 Clinical data for individuals enrolled in the study
by ancestry
ALI No ALI
African Americans (N = 285) (N = 71) (N = 187)
Age SD) 36.7 ± 17.5 31.9 ± 12.4
Gender (% of males ) 32 84
ISS SD) 25.3 ± 8.9 22.8 ± 6.3
Blunt Mechanism (%)
(% of Blunt Trauma)
25 59
APACHE SD) 45.4 ± 12.4 40.2 ± 13.5
European Americans (N = 269) (N = 86) (N = 168)
Age SD) 41.8 ± 19.9 39.3 ± 18.5
Gender (% of males ) 27 61
ISS SD) 26.9 ± 7.5 25.2 ± 7.2
Blunt Mechanism (%) 4 7
APACHE SD) 41.9 ± 14.7 35.7 ± 12.8
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 2 of 15
at least 100 bp (Additional file 1). Following primer
optimization, DNA was amplified and sequenced in the
forward and reverse direction using a 3730 automated
sequencer at the CHOP Nucleic Acid and Protein Core
facility (Philadelphia, PA). Sequencher 4.8 (Gene Codes
Ann Arbor, MI) was used to fa cilitate secondary peak
calls and to compare the s equence data to the NCBI
reference sequence.
SNP genotyping
Novel variants with a minor allele frequency (MAF) of
0.04orhigherandtaggingSNPs,fromHapMaphttp://
hapmap.ncbi.nlm.nih.gov/, were validated using the
SNPlex genotyping system (Applied Biosystems Inc.
Foster City, CA). Tagging SNPs were selected using
Taggers pairwise testing methods described by Bakker
and colleague s [14]. Genotyping novel variants not only
served to test for association, but allowed us to validate
those SNPs in a larger population. Tagging SNPs were
also selected to provide better coverage of the haplotype
structure of PRDX6. SNPlex utilizes an oligonucl eotide
ligation/PCR assay with universal ZipChute probe detec-
tion to perform genotyping of up to 48 SNPs in a single
reaction. ZipChute probes were custom designed and
detected by capillary elec trophoresis us ing the Applied
Biosystems 3130 Analyzer and genotype calls were
determined using Gene Mapper 4.0 (Applied Biosystems
Foster City, CA).
All genotyping was performed in the University of
Pennsylvanias Lab oratory for Molecular Epidemiology
(LME). Staff was blinded to the disease status and geno-
typing calls were performed in subsamples by plate.
Each plate contained six positive c ontrols to test for
concordance. Genotyping calls were performed automa-
tically using the algorithm described by Da La Vega and
colleagues [15].
In silico modeling of putative function in SNP sites
We sou ght to tes t inferred function in silico using tran-
scription fac tor binding and mRNA binding tools. TESS
is a web-based software tool for l ocating possible tran-
scription factor binding sites in DNA sequences using
weight matrix models. It can also be used for browsing
information about relevant transcription factors in the
TRANSFAC database [16]. All SNPs discovered within
the 5 UTR and the first intron wer e submitted to TESS
as 21 base pair long FASTA sequences with the refer-
ence allele of the SNP of interest in the 11
th
position. A
second search was performed using the alternative allele
in the 11
th
position. To eliminate any poor matches due
to background noise, transcription factors with log-like-
lihood scores (La) less than 12, were eliminated. TESS
results were compared with experimental transcription
factor binding site (TFBS) data registered i n the
University of California Santa Cruz (UCSC) Genome
Browser by the Encyclopedia of DNA Elements
(ENCODE) consortium [17]. The ENCODE data were
filtered by chromosome and position.
A search query was performed for potential miRNA
binding sites in the 3 UTR of PRDX6 using Patrocles
http://www.patrocles.org/. Patrocles is an online database
containing DNA sequence polymorphisms that are pre-
dicted to interrupt miRNA-mediated gene regulation
[18]. The search was performed using PRDX6 as a key
word in the target gene id field and miRNA target motifs
were defined by Xie et al. [19] and Lewis et al [20].
Statistical Analysis of ALI association
259 African American and 254 European American sub-
jectsenrolledinthemajortraumacohortwereusedto
test for association of novel variants and ta gging SNPs
with ALI. Association of each PRDX6 SNP with ALI
was determined separately for European Americans and
African Americans using an additive model Chi
2
test,
with a p-value < 0.0014 for African Americans consid-
ered significant. Dominant and re cessive inheritan ce
models were also tested using Chi
2
analysis. Multivari-
able analyses of potential confounding were performed
using logistic regression methods. Power was calculated
using the power for genetic associat ion analyses (PGA)
[21]. Using PGA, we estimated that a sample size of 250
subjects per race category would provide 80% power to
detect relative risks of 2.26 or greater for SNPs with a
prevalence of 0.05 or greater and 1.93 or greater for
SNPs with a prevalence of 0.10 or greater, assuming a
Bonferroni-corrected alpha = 0.0014 for African Ameri-
cans, and an incidence of ALI = 0.30 (Addi tional file 2).
These statistical analyses were perfor med using STATA
11 (STATA Data Corp, College Station, TX). Pairwise
linkage disequilibrium was evaluated using Haploview
http://www.broadinstitute.org/mpg/haploview. Geno-
types with a completion rate of 95% or greater were
considered for analysis in Haploview. LD was calculated
in terms of r
2
values and blocks were defined using the
default algorithm using the confidence intervals methods
of Gabriel and colleagues [22].
Haplotypes were inferred using the standard expecta-
tion maximization algorithm in Haploview [23,24] and
the following confidence interval (CI) criteria: CI
minima for strong LD: 0.7 - 0.98; u pper CI maximum
for strong LD: 0.98; fracti on of strong LD in informative
comparisons 0.95; and exclude markers with minor
allele frequency (MAF) < 0.05. Haploty pes were tested
for association with ALI first in a global association test,
which performed contingency testing using all haplo-
types of an LD block compared to no haplotypes, and
then as individual haplotypes versus ALI coded in an
additive fashion PLINK [25]. Haplotype multiple testing
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 3 of 15
was ad dressed by applying permu tation tests (10,000
permutations).
Results
Identification of novel polymorphisms in PRDX6
Direct sequencing of the PRDX6 gene in 48 subject s
revealed 80 genetic variants, none of which were in cod-
ing regions (31 in the 3 UTR, 22 in the 5 UTR and 27
intronic) (Additional file 3). The variants identified via
direct sequencing were compared with t hose registered
in the NCBI dbSNP database (Build 130) and 43 were
found to be novel SNPs (Table 2). Thirty seven were
matched with S NPs catalogued in dbSNP (Build 130)
and Genewindow http://genewindow.nci.nih.gov/Wel-
come based on chromosome position (Table 3). Twenty
five of the novel SNPs uncovered had a MAF > 0.04 and
were submitted to the NCBI to be catalogued and
assigned ss numbers in the submitter records section of
dbSNP (Table 2). Novel SNPs were a lso compared with
SNPs registered in the 1000 Genomes d atabase. Thirty
six out of thirty seven known SNPs overlapped with
SNPs registered in 1000 Genomes, but only sixteen out
of forty-three novel SNPs ide ntified via our sequencing
effort were also registered in 1000 Genomes (Table 4).
Several variants were only observed in one individual.
As a quality control m easure we present the confidence
scores for these genotypes in additional file 4. Confi-
dence scores are reported as the percentage of overlap
between heterozygote peaks. Previous studies indicated
that two t ranscription factor binding sites, the ARE1
(-357 to -349) [26] and GRE2 (-750 to -738) (A. Fisher,
unpublished observations), may play a role in the regula-
tion of PRDX6.WewereunabletosequencetheARE1
region and portions of the intronic regions due to the
GC rich content of the flanking sequence (Figure 1).
The GRE2 region was successfully sequenced, but
showed no variation.
In Silico function of novel SNPs in PRDX6
The TES S results showed severa l potential transcript ion
factor binding motifs in both the reference and alterna-
tive sequence. The reference and alternative sequences
were submitted as independent queries and transcrip-
tion factors were returned for 19 pos itions in the refer-
ence sequence and 21 positions in the alternative
sequence (Table 5). Twenty seven out of twenty nine
sequences submitted were shown to create, abolish, or
change a transcription factor binding site. Fourteen of
these SNPs wer e novel. Comparison of the transcription
factors returned from the TESS query with the data
from ENCODE showed that only 3 of these putative
transcription factor bindin g sites have been tested by
the ENCODE consortium, SP1, GATA-1, and c-Myc.
ENCODE data fo r SP1, GATA-1, and c-Myc reveale d
that there is no evidence of binding affinity with the
sequence results from the PRDX6 gene when filtered for
PRDX6 and A549 cells.
A Patrocles miRNA database search for PRDX6
revealed eight SNPs in the 3 UTR of PRDX6 as poten-
tial miRNA binding sites (Table 6). Of the eight SNPs
returned from the search query, three matched SNPs
from this study ( rs4611, rs36005931, and rs2000).
rs4611 and rs36005931 are located within octamers that
have been conserved among several species, but do not
correspond to a known miRNA. The G allele of rs2000
is part of an octamer capable of binding miR-942. A lit-
erature search for miR-942 returned only sequence data,
with no known function to date.
Association of PRDX6 with ALI
The trauma cohort described in Table 1 was genotyped
for 37 PRDX6 SNPs using SNPlex. All SNPs were tested
for Hardy-Weinberg equilibrium (Additional file 5). Chi
2
analysis of incidence of ALI compared to genotype
using an additive model showed no significant a ssocia-
tion between any of the SNPs in this study and ALI
(Table 7). Dominant and recessive models failed to
demonstrate an association betwe en our SNPs of inter-
est and ALI (Additional file 6). The genotype concor-
dance rate based on assay positive controls was 100%
and the frequency of missing genotypes is presented in
Table 7. Logistic regress ion analysis after adjustment for
age and ISS showe d no assoc iation between ALI and
our SNPs (Table 8).
Haplotype Analysis
Haplotype blocks were created for both African and
European Americans using 27 and 28 SNP markers,
respecti vely. Haplotype blocks were created for a region
spanning 100.6 kb of chromosome 1. For African Amer-
icans, there were 14 SNP markers with genotyping com-
pletion rate of less than 95% and were thus excluded
from the haplotype analysis. For European Americans,
there were 9 SNPs with a genotype completion rate of
less than 95% and were excluded from the haplotype
analysis.
In African Americans, the re were 2 blocks (blo ck1 =
rs491636 2 and rs10753081; block2 = PRDX6_171711459,
rs34619706, hCV9040425, rs35244306, rs9425725,
rs912767, rs2000, hCV1948447, rs6702828, rs670 2835,
rs7521536, rs7529377) (Figure 2). In European Americans,
there were 2 block s (block1 = rs4916362, rs10753081,
PRDX6_171711459, rs34 619706; block 2 = rs33942654,
PRDX6_171715019, hCV9040425, rs35244306, rs9425725,
rs912767, rs2000, hCV1948447, rs6702835, rs752 1536,
rs7529377) (Figure 3). The haplotype structure of PRDX6
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 4 of 15
Table 2 Novel SNPs discovered via direct sequencing
Novel SNPs Region Chr. Position SNP 5 Flanking Sequence 3 Flanking Sequence MAF in EA MAF in AA ss# (dbSNP)
PRDX6_171709541 5 UTR 171709541 G/C CTTCAAGGTTC ACCCTTATAGC 0.04 0.02 ss217326279
PRDX6_171709910 5 UTR 171709910 G/T ATGATCATTTTT GAAATATACAG 0.00 0.10 ss217326288
PRDX6_171710327 5 UTR 171710327 C/T ACCCTAGCCCC TGTGCTGGCA 0.04 0.00 ss217326273
PRDX6_171710490 5 UTR 171710490 C/T TGCACTGCGGA GCAGGGACCT 0.00 0.06 ss217326283
PRDX6_171710775 5 UTR 171710775 G/C CTTATGGCTGG GTGAGACATG 0.00 0.02
PRDX6_171710821 5 UTR 171710821 C/T ACTGCACTGAG TTGTGTAAAGT 0.00 0.10 ss217326292
PRDX6_171711029 5 UTR 171711029 C/T ACTCAGAGACC GGGTCCTCCG 0.00 0.02
PRDX6_171712345 5 UTR 171712345 A/T ATGGTTCATAA AGAAAGGGGA 0.87 0.64 ss217326296
PRDX6_171713694 intron 171713694 G/T TCACTTCCCCG AGTGCCCAGG 0.00 0.04 ss217326300
PRDX6_171713738 intron 171713738 G/T CCTCCGTTCTG TGCTCCCTGG 0.00 0.02
PRDX6_171713872 intron 171713872 C/T GCACAAAATGT TAAAACCACTA 0.00 0.12 ss217326323
PRDX6_171713919 intron 171713919 G/C AAAGACTTTTTG AGCCGCCTCC 0.02 0.00
PRDX6_171714107 intron 171714107 C/T CCAGGACACGT TCCCCAACTTT 0.00 0.04 ss217326304
PRDX6_171714984 intron 171714984 C/T GATCAAAAGTG TTATCAGGGAG 0.04 0.04 ss217326307
PRDX6_171715019 intron 171715019 A/G AGGAACACGGT TATCTGCATTT 0.00 0.10 ss217326318
PRDX6_171715596 intron 171715596 A/G GGGAGGGAAG TGAACTGGCTT 0.00 0.02
PRDX6_171716007 intron 171716007 A/G AAACCTTGGGA GTGGCAGCCG 0.00 0.04 ss217326311
PRDX6_171716032 intron 171716032 G/C TAAGTAGGAAG TGCCCTTGTCT 0.00 0.02
PRDX6_171716554 intron 171716554 A/C AGAAGCCAAGT AACTTTAATTTT 0.00 0.02
PRDX6_171716572 intron 171716572 A/T TCAACTTTAATT TAAATAGAAGA 0.00 0.02
PRDX6_171716582 intron 171716582 A/G TTACATATAAAT ATAGAAACCTA 0.00 0.04 ss217326315
PRDX6_171716584 intron 171716584 A/T ACATATAAATAG AGAAACCTATT 0.00 0.02
PRDX6_171716603 intron 171716603 A/G AACCTATTTATT ATTACATAATTT 0.00 0.02
PRDX6_171723151 Intron 171723151 C/T AAAGCTAGCAT TGGAGAAGAA 0.00 0.02
PRDX6_171723403 Intron 171723403 C/T CTTGATTAGTCT AGCACCTGTAG 0.00 0.02
PRDX6_171723889 3UTR 171723889 G/T AAAACTCAAAT GGATCTCTGCA 0.00 0.04
PRDX6_171723918 3UTR 171723918 A/G GCTTGTGACCA GTCATATTTGT 0.00 0.02
PRDX6_171724000 3UTR 171724000 G/C TAACTGTCCTAT TCCTCTCCTGT 0.00 0.04 ss217326227
PRDX6_171724128 3UTR 171724128 G/T TTTTTTTAATAT TGATCACAGAA 0.00 0.04 ss217326232
PRDX6_171724182 3UTR 171724182 A/G CATATTCTTTTA TCTTGATCACA 0.00 0.04 ss217326236
PRDX6_171724286 3UTR 171724286 A/T TTGCTATAAAAA TTTGTGATAAG 0.00 0.02
PRDX6_171724949 3UTR 171724949 C/T ACTCTACTAATA CAGGTTTAGAA 0.26 0.00 ss217326270
PRDX6_171725122 3UTR 171725122 G/T GGACCTGCTTC TTGTAGTTTGC 0.00 0.04 ss217326239
PRDX6_171725183 3UTR 171725183 C/T GGGATCATCGC GTCTCATAAGG 0.00 0.04 ss217326242
PRDX6_171725257 3UTR 171725257 A/T CCTCCCAAAGG CATCCAAATAC 0.00 0.04 ss217326247
PRDX6_171725485 3UTR 171725485 C/T CCTGCCTCAGC GAGCAGCTGG 0.00 0.06 ss217326250
PRDX6_171725681 3UTR 171725681 G/T ATATTTTTATTG TAGAATAATGT 0.00 0.02
PRDX6_171725826 3UTR 171725826 A/G TCTGGGGAATG TTTGAAAGAGA 0.04 0.00 ss217326254
PRDX6_171727146 3UTR 171727146 G/C CTGTGATTCCT TTGTGGTCTTG 0.02 0.00
PRDX6_171727831 3UTR 171727831 C/T ATGCATGGGAT ATTATCCTCTA 0.04 0.02 ss217326261
PRDX6_171728416 3UTR 171728416 G/T CCTCATTAGGG CTCTTAGCCCT 0.04 0.02 ss217326266
PRDX6_171728455 3UTR 171728455 G/C AATCGGGAGGC TGTTAACAGGT 0.00 0.04 ss217326257
PRDX6_171729049 3UTR 171729049 C/T GTTCTTAAACTA AATAGCATGAG 0.00 0.02
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 5 of 15
appears to be in low LD, in both subgroups. Haplotype
analyses of association with ALI did not reveal any signifi-
cant associations above and single SNP analyses (Table 9).
Discussion
Prdx6 is a member of the t hiol-specific antioxidant
protein family and in overexpressing cell and mouse
models has been shown to b e protective against oxi-
dant stress which null models show sensitivity to
oxidants [7,9,27]. Thus, PRDX6 is a suitable candidate
gene for ALI risk. The extent of genetic variation
within PRDX6 remains largely un known, therefore we
performed direct sequencing of the PRDX6 gene, and
identified novel variants for future study. We also
tested the newly discovered SNPs and tagging SNPs
for association with ALI using our trauma cohort, and
did not demonstrate an association with trauma-
related ALI.
Table 3 Known SNPs discovered via direct sequencing
Knowns SNPs Region Chr. Position SNP MAF in EA (discovery) MAF in AA (discovery) MAF in EA (dbSNP) MAF in AA (dbSNP)
rs13376447 5 UTR 171709896 A/G 0.00 0.06 0.01 0.09
rs10753081 5 UTR 171710154 C/T 0.39 0.30 0.73 0.21
rs34282688 5 UTR 171710819 C/T 0.07 0.00 NA NA
rs9425722 5 UTR 171711268 C/T 1.02 0.70 0.04 0.24
rs35152701 5 UTR 171711269 A/G 0.02 0.00 0.00 0.00
rs12739142 5 UTR 171711278 A/G 0.04 0.00 0.02 0.04
rs4354572 5 UTR 171711459 C/T 1.02 0.66 0.00 0.04
rs34619706 5 UTR 171711670 A/G 0.09 0.02 0.07 0.04
rs4382766 5 UTR 171711699 C/T 0.54 0.28 0.30 0.50
rs13376392 5 UTR 171711701 C/T 0.00 0.06 0.00 0.04
rs11576174 5 UTR 171712301 G/T 0.04 0.00 0.15 0.00
rs34977864 5 UTR 171712466 G/T 0.00 0.06 0.00 0.10
rs35441546 5 UTR 171712652 C/T 0.00 0.06 0.00 0.03
rs35133735 5 UTR 171712738 C/T 0.00 0.06 0.00 0.00
rs6671141 intron 171713557 G/T 0.26 0.26 0.24 0.30
rs35749242 intron 171714171 A/G 0.00 0.12 0.00 0.03
rs35918328 intron 171714199 A/G 0.00 0.02 0.00 0.10
rs35899698 intron 171714279 C/T 0.00 0.10 0.00 0.00
rs33942654 intron 171714884 A/G 0.26 0.18 0.23 0.33
rs35679908 intron 171715013 A/G 0.00 0.02 0.00 0.00
rs9425723 intron 171715118 A/G 0.33 0.62 0.23 0.63
rs9425724 intron 171715123 A/G 0.33 0.54 0.23 0.53
rs7540065 intron 171715659 A/G 0.33 0.56 0.23 0.53
rs35244306 intron 171716247 C/T 0.00 0.14 0.00 0.07
rs4611 3UTR 171723640 C/T 0.33 0.50 0.77 0.47
rs3833536 3UTR 171724172 C/- 0.00 0.00 0.05 0.04
rs7314 3UTR 171724222 A/G 0.20 0.50 0.23 0.52
rs36005931 3UTR 171724224 A/G 0.00 0.04 0.00 0.10
rs2000 3UTR 171724457 A/G 0.04 0.00 0.03 0.00
rs34129563 3UTR 171724720 G/C 0.11 0.00 0.02 0.03
rs9425727 3UTR 171725216 G/C 0.02 0.14 0.00 0.04
rs35358649 3UTR 171725429 C/T 0.02 0.00 0.00 0.00
rs35547740 3UTR 171725569 (-/T) 0.00 0.04 0.00 0.07
rs6702835 3UTR 171725723 A/G 0.33 0.48 0.23 0.54
rs60587131 3UTR 171726777 G/C 0.00 0.02 0.05 NA
rs57032935 3UTR 171726836 G/C 0.15 0.22 0.18 0.31
rs6664925 3UTR 171729033 A/G 0.00 0.18 0.00 0.16
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 6 of 15
Table 4 Comparison of PRDX6 discovery SNPs with 1000 Genomes database
PRDX6 SNPs 1000 Genomes Comparison PRDX6 SNPs 1000 Genomes Comparison
PRDX6_171709910 Matched PRDX6_171709541 Not Matched
PRDX6_171713872 Matched PRDX6_171710327 Not Matched
PRDX6_171714107 Matched PRDX6_171710490 Not Matched
PRDX6_171723151 Matched PRDX6_171710775 Not Matched
PRDX6_171723918 Matched PRDX6_171710821 Not Matched
PRDX6_171724000 Matched PRDX6_171711029 Not Matched
PRDX6_171724128 Matched PRDX6_171712345 Not Matched
PRDX6_171724182 Matched PRDX6_171713694 Not Matched
PRDX6_171725122 Matched PRDX6_171713738 Not Matched
PRDX6_171725183 Matched PRDX6_171713919 Not Matched
PRDX6_171725257 Matched PRDX6_171714984 Not Matched
PRDX6_171725826 Matched PRDX6_171715019 Not Matched
PRDX6_171727146 Matched PRDX6_171715596 Not Matched
PRDX6_171727831 Matched PRDX6_171716007 Not Matched
PRDX6_171728416 Matched PRDX6_171716032 Not Matched
PRDX6_171728455 Matched PRDX6_171716554 Not Matched
rs10753081 Matched PRDX6_171716572 Not Matched
rs11576174 Matched PRDX6_171716582 Not Matched
rs12739142 Matched PRDX6_171716584 Not Matched
rs13376392 Matched PRDX6_171716603 Not Matched
rs13376447 Matched PRDX6_171723403 Not Matched
rs2000 Matched PRDX6_171723889 Not Matched
rs33942654 Matched PRDX6_171724286 Not Matched
rs34129563 Matched PRDX6_171724949 Not Matched
rs34282688 Matched PRDX6_171725485 Not Matched
rs34619706 Matched PRDX6_171725681 Not Matched
rs34977864 Matched PRDX6_171729049 Not Matched
rs35133735 Matched rs3833536 Not Matched
rs35152701 Matched
rs35244306 Matched
rs35358649 Matched
rs35441546 Matched
rs35547740 Matched
rs35679908 Matched
rs35749242 Matched
rs35899698 Matched
rs35918328 Matched
rs36005931 Matched
rs4354572 Matched
rs4382766 Matched
rs4611 Matched
rs57032935 Matched
rs60587131 Matched
rs6664925 Matched
rs6671141 Matched
rs6702835 Matched
rs7314 Matched
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 7 of 15
We identified 43 novel variants among African Ameri-
can and European American subjects with either ALI or
control status. None of the 43 SNPs identified were in
coding regions which may indicate that the Prdx6 pro-
tein is highly conserved across phyla. Approximately 19
kb on chromosome 1 was sequenced in order to achieve
adequate coverage of the PRDX6 gene and f lanking 5
and 3 UTRs. Special attentionwasgiventotheGRE2
and ARE1 regions -749 to -737 and -357 to -349,
respectively. The ARE1 within the PRDX6 promoter was
showntoplayaroleinregulationoftranscriptionand
to be inducible under conditions of oxidative stress [26]
and the GRE2 may be capable of binding transcription
factors under oxidative stress conditions [28]. Due to
the GC rich content of the region surrounding the
ARE1, we were unable to optimize PCR reaction condi-
tions in a way to prime through the secondary structure.
The GRE2 region was sequenced, but no variation was
noted. The GC rich region wit hin the PRDX6 promot er
might war rant further investigation since methylation of
DNA cytosine residues are often found in the sequence
context CpG. Several new sequencing approaches are
emerging that target methylation sites using restriction
enzyme treatment followed by sequence by synthesis
[29].
In addition to comparing our results with NCBI s
dbSNP, we compared our novel and known SNPs with
the resequencing data registered in 1000 Genomes. The
1000 Genomes project aims to find most genetic var-
iants with frequencie s of at least 1%. Thus far three
sequencing projects contribute to the database, low cov-
erage sequencing of 179 individuals from 4 populations,
high coverage sequenci ng of 2 mother-father-child trios,
and exon targe ting sequencing of 697 individuals from 7
populations [30]. Although 1000 Genomes aims to iden-
tify over 95% of variat ion in any individual, 27 of our
novel S NPs and 1 previously recorded SNP are not pre-
sent in the database, signifying a need for resequencing
of extreme phenotypes, such as ALI cases.
Novel and previously recorded SNPs in the 5 UTR
and first intron of PRDX6 were submitted to TESS to
determine their likelihood of being in transcription fac-
tor binding sites. We found 19 motifs in the reference
sequences that are capable of binding known transcrip-
tion factors and 21 in the alternative sequence. A com-
parison between the results of the ref erence sequence
search and the alternative revealed that in most cases,
the SNP of interest changes the motif enough to cause a
different transcription factor to bind that site or can
cause a binding site to disappear and vice versa. After
comparison with the ENCODE data, we found that our
sequences have not yet been shown to b ind the three
overlapping transcription factors tested in ENCODE
experimentally.
Known SNPs validated in the sequencing effort were
compared using a Patrocles search query for miRNA
target sites within PRDX6 to de termine if any of our
SNPs were in putative target sites for miRNAs. Three of
the eight SNPs returned from the search corresponded
with our known SNPs. Only one of the three SNPs was
found to have a corresponding known miRNA (miR-
942). Some miRNAs are known to control the expres-
sion of genes at the posttranscriptional level [31]. How-
ever, very limited data are available on miR-942.
We performed an association study for ALI using
newly un covered SNPs and SNPs se lected from Hap-
map and NCBI s dbSNP and o bserved no significant
association between any of the SNPs in this study and
Table 4 Comparison of PRDX6 discovery SNPs with 1000 Genomes database (Continued)
rs7540065 Matched
rs9425722 Matched
rs9425723 Matched
rs9425724 Matched
rs9425727 Matched
Figure 1 PRDX6 gene diagram with novel SNP positions. A schematic of the PRDX6 gene is presented with vertical lines and arrows above
the gene indicating novel SNPs and regions of interest, respectively. The horizontal arrows below the schematic are representative of the
regions of PRDX6 that were successfully sequenced.
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 8 of 15
Table 5 Potential transcription factor binding sites within the PRDX6 gene
SNP Location Base Change Reference Sequence Alternative Sequence
PRDX6_171709541 5 UTR G/C n/a Sp1, CACCC binding factor, PuF
rs13376447 5 UTR A/G GATA-1 GATA-1
PRDX6_171709910 5 UTR G/T n/a IL-6 RE-BP
rs10753081 5 UTR C/T GATA-3 GATA-1
PRDX6_171710327 5 UTR C/T C/EBPbeta C/EBPbeta
PRDX6_171710775 5 UTR G/C c-Myb n/a
rs9425722 5 UTR C/T n/a c-Myc
rs35152701 5 UTR A/G n/a TMF, TBP, TFIID
rs12739142 5 UTR A/G IPF1, Isl-1, IPF1 IPF1
rs4354572 5 UTR C/T C/EBPbeta AFP1, ATBF1-B
rs11576174 5 UTR G/T H4TF-1 ETF, TMF, TFIID, TBP
PRDX6_171712345 5 UTR A/T n/a SRY, TCF-4E
PRDX6_171712345 5 UTR A/T n/a SRY
rs35441546 5 UTR C/T Sp1, ETF CACCC-binding factor, Sp1
rs35133735 5 UTR C/T n/a CACCC-binding factor, Sp1
rs6671141 intron G/T GR alpha, PR, PR A, GR beta n/a
PRDX6_171713738 intron G/T n/a TEF-1
PRDX6_171713872 intron C/T n/a H4TF-1
rs35918328 intron A/G IPF1 n/a
rs33942654 intron A/G c-Myb n/a
PRDX6_171714984 intron C/T TBP GATA-1
PRDX6_171715019 intron A/G GATA-3 GATA-1
rs9425723 intron A/G TBP, TFIID AP-1, c-Jun
rs9425724 intron A/G USF1 AP-1, AP-4, CCK-1a, c-Myc, CREB, Max1, USF1
PRDX6_171715596 intron A/G c-Myb, c-Ets-2 n/a
PRDX6_171716007 intron A/G n/a Sp1, C/EBPbeta, CACCC-binding factor
PRDX6_171716572 intron A/T TMF, TFIID, ETF, TBP n/a
PRDX6_171716582 intron A/G GATA-1, GATA-3 n/a
PRDX6_171716584 intron A/T GATA-1, GATA-3 SRY
*Transcription factor binding site modeling was performed using TESS.
Table 6 PRDX6 SNPs thought to be miRNA target sites
SNP ID Chromosome Position Base Change Ancestral Derived miRNA
rs4611 171723640 T/C TTGGTGCT
CTGGTGCT
rs15268 171723695 C/A AGCAATTA hsa-miR-302f
ATTACATA hsa-miR-380
rs3211528 171724201 G/A CTGGGGGA hsa-miR-361-3p
rs36005931 171724224 A/G GTGCCTTC
TGTGCCTT
rs35820016 171724277 T/A TTTTGCTA hsa-miR-548p
TTGCAATA
rs1804053 171724351 G/A ATGTAGCA hsa-miR-221
hsa-miR-222
rs11544001 171724433 A/G GTGCATGA
TGCATGAA
TGTGCATG
rs2000 171724457 G/A AGAGAAGA hsa-miR-942
*miRNA target site modeling was performed using Patrocles.
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 9 of 15
ALI. This lack of association may be due to several
causes. First, the detectable effect size is modest
because of sample size limitations. We genotyped 513
subjects to test for an association between our selected
SNPs and ALI, but this sample size was inadequate to
detect relative risks below 1.93 and 1.69 for alleles
with MAFs of 0.05 and 0.10, respectively. Second, our
analyses were limited to patients with severe trauma.
Thus,ourstudydidnotevaluateapossibleassociation
with other causes of ALI such as sepsis. Finally, it is
possible that PRDX6 genetic variation may n ot modify
the risk of ALI.
Table 7 Association of 37 PRDX6 genotypes and risk of ALI using and additive model in a population of African and
European Americans with major trauma
European Americans African Americans
SNP Source MAF
(ALI)
MAF
(non-ALI)
Missing Genotype
Frequency
p value MAF
(ALI)
MAF
(non-ALI)
Missing Genotype
Frequency
p value
hCV1948447 Tagging 0.22 0.20 0.016 0.538 0.06 0.05 0.019 0.507
hCV25599136 Tagging 0.00 0.00 0.161 NA 0.03 0.03 0.230 0.842
hCV25599144 Tagging 0.00 0.01 0.399 0.999 0.00 0.00 0.296 NA
hCV9040425 Tagging 0.27 0.25 0.021 0.739 0.38 0.35 0.019 0.725
hCV9040434 Tagging 0.25 0.24 0.407 0.838 0.31 0.29 0.307 0.537
Position23855054 Sequencing 0.00 0.00 0.012 NA 0.09 0.08 0.019 0.535
Position23855203 Sequencing 0.00 0.00 0.679 NA 0.10 0.17 0.662 0.268
Position23859396 Sequencing 0.00 0.00 0.012 NA 0.04 0.02 0.019 0.302
Position23863249 Sequencing 0.00 0.00 0.012 NA 0.03 0.03 0.023 0.577
PRDX6_171709910 Sequencing 0.03 0.01 0.436 0.192 0.04 0.06 0.494 0.505
PRDX6_171711459 Sequencing 0.00 0.00 0.021 NA 0.11 0.18 0.043 0.088
PRDX6_171713872 Sequencing 0.00 0.01 0.037 0.999 0.08 0.05 0.070 0.331
PRDX6_171715019 Sequencing 0.00 0.00 0.029 0.999 0.05 0.05 0.058 0.879
PRDX6_171724949 Sequencing 0.22 0.19 0.029 0.455 0.07 0.05 0.043 0.481
rs10753081 Sequencing 0.34 0.29 0.008 0.246 0.37 0.34 0.000 0.676
rs2000 Tagging 0.05 0.05 0.045 0.966 0.02 0.02 0.031 0.793
rs33942654 Sequencing 0.23 0.20 0.021 0.321 0.34 0.29 0.054 0.138
rs34129563 Sequencing 0.07 0.03 0.021 0.080 0.01 0.01 0.039 0.559
rs34619706 Sequencing 0.08 0.08 0.037 0.900 0.03 0.01 0.047 0.287
rs35244306 Sequencing 0.00 0.01 0.021 0.999 0.11 0.08 0.039 0.517
rs35749242 Sequencing 0.00 0.01 0.021 0.999 0.08 0.06 0.039 0.516
rs4354572 Sequencing 0.00 0.00 0.354 NA 0.11 0.17 0.358 0.251
rs4382766 Sequencing 0.33 0.29 0.025 0.312 0.35 0.34 0.047 0.830
rs4916362 Tagging 0.34 0.29 0.021 0.308 0.35 0.32 0.019 0.670
rs57032935 Sequencing 0.23 0.19 0.037 0.296 0.34 0.29 0.082 0.131
rs6671141 Sequencing 0.23 0.19 0.210 0.421 0.38 0.33 0.163 0.170
rs6699179 Tagging 0.00 0.00 0.214 NA 0.01 0.00 0.171 0.999
rs6702828 Tagging 0.00 0.00 0.012 NA 0.01 0.00 0.027 0.778
rs6702835 Sequencing 0.27 0.25 0.008 0.606 0.42 0.37 0.000 0.493
rs7314 Sequencing 0.27 0.25 0.045 0.666 0.34 0.30 0.078 0.571
rs7367963 Tagging 0.34 0.29 0.029 0.284 0.35 0.31 0.019 0.574
rs7521536 Tagging 0.24 0.20 0.025 0.360 0.32 0.29 0.043 0.320
rs7529377 Tagging 0.24 0.20 0.016 0.331 0.28 0.26 0.027 0.321
rs7540065 Sequencing 0.27 0.25 0.021 0.726 0.37 0.35 0.039 0.997
rs912767 Tagging 0.24 0.20 0.016 0.331 0.29 0.26 0.027 0.240
rs9425722 Sequencing 0.00 0.00 0.342 NA 0.08 0.17 0.335 0.048
rs9425725 Tagging 0.00 0.00 0.012 NA 0.13 0.20 0.027 0.091
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 10 of 15
The genotype data were u sed to construct hapl otype
blocks to bet ter assess the PRDX6 gene structure. Hap-
lotype anal ysis plays an important role in association
studies between genotype and phenotype, since SNPs
found to be in strong LD can capture most of the
genetic variation across fairly large regions [24]. The
haplotype blocks constructed from our g enotype data
did not show strong linkage disequilibrium using
confidence intervals, therefore tagging SNP strategies in
future studies should be approached with caution.
Our resequencing data did not show any variation in
the coding region of PRDX6. Had nonsynonymous
SNPs been discovered, it would have prompted us to
investigate whet her any of these SNPs had any effect on
protein structure, which could cause a loss of function
in Prdx6. Since we cannot make a connection between
Table 8 Multivariate analysis adjusted for age and injury severity score
African Americans European Americans
SNP ORtrend Ptrend ORdom Pdom ORrec Prec ORtrend Ptrend ORdom Pdom ORrec Prec
hCV1948447 1.33 0.507 1.19 0.705 3.87E+09 0.999 1.16 0.538 1.06 0.840 2.41 0.204
hCV25599136 1.15 0.842 1.15 0.842 NA NA NA NA NA NA NA NA
hCV25599144 NA NA NA NA NA NA 4.19E-10 0.999 4.19E-10 0.999 NA NA
hCV9040425 1.08 0.725 1.08 0.781 1.13 0.760 1.08 0.739 1.04 0.885 1.32 0.593
hCV9040434 1.17 0.537 1.30 0.442 1.04 0.947 1.06 0.838 0.94 0.871 2.05 0.365
Position23855054 1.25 0.535 1.39 0.398 2.07E-09 0.999 NA NA NA NA NA NA
Position23855203 0.56 0.268 0.60 0.374 1.52E-09 0.999 NA NA NA NA NA NA
Position23859396 1.73 0.302 2.17 0.190 1.70E-09 0.999 NA NA NA NA NA NA
Position23863249 1.36 0.577 1.61 0.431 1.70E-09 0.999 NA NA NA NA NA NA
PRDX6_171709910 0.66 0.505 0.72 0.636 1.74E-09 0.999 5.13 0.192 5.13 0.192 NA NA
PRDX6_171711459 0.58 0.088 0.61 0.146 2.02E-09 0.998 NA NA NA NA NA NA
PRDX6_171713872 1.49 0.331 1.69 0.229 1.06E-09 0.999 1.28E-09 0.999 1.28E-09 0.999 NA NA
PRDX6_171715019 0.93 0.879 1.01 0.989 1.07E-09 0.999 1.77E-09 0.999 1.77E-09 0.999 NA NA
PRDX6_171724949 1.35 0.481 1.21 0.681 4.18E+09 0.999 1.20 0.455 1.10 0.740 2.50 0.186
rs10753081 1.09 0.676 1.19 0.543 0.98 0.970 1.28 0.246 1.34 0.288 1.44 0.439
rs2000 1.21 0.793 1.21 0.793 NA NA 0.98 0.966 1.25 0.661 1.30E-09 0.999
rs33942654 1.39 0.138 1.78 0.057 1.05 0.923 1.27 0.321 1.14 0.640 2.96 0.103
rs34129563 1.73 0.559 1.73 0.559 NA NA 2.16 0.080 2.09 0.115 3.42E+09 0.999
rs34619706 2.13 0.287 2.13 0.287 NA NA 0.96 0.900 0.89 0.773 2.32 0.554
rs35244306 1.25 0.517 1.41 0.349 1.34E-09 0.999 1.29E-09 0.999 1.29E-09 0.999 NA NA
rs35749242 1.30 0.516 1.43 0.399 1.09E-09 0.999 1.29E-09 0.999 1.29E-09 0.999 NA NA
rs4354572 0.65 0.251 0.70 0.386 1.85E-09 0.999 NA NA NA NA NA NA
rs4382766 0.96 0.830 0.94 0.838 0.94 0.892 1.24 0.312 1.26 0.409 1.50 0.394
rs4916362 1.09 0.670 1.19 0.551 0.99 0.987 1.24 0.308 1.28 0.381 1.44 0.436
rs57032935 1.40 0.131 1.81 0.053 1.07 0.888 1.29 0.296 1.16 0.600 2.98 0.101
rs6671141 1.37 0.170 1.43 0.256 1.64 0.261 1.24 0.421 1.21 0.549 1.86 0.398
rs6699179 1.14E+09 0.999 1.14E+09 0.999 NA NA NA NA NA NA NA NA
rs6702828 1.59 0.778 1.59 0.778 NA NA NA NA NA NA NA NA
rs6702835 1.15 0.493 1.78 0.063 0.61 0.242 1.12 0.606 1.12 0.686 1.28 0.632
rs7314 1.13 0.571 1.26 0.437 1.02 0.963 1.10 0.666 1.07 0.820 1.38 0.542
rs7367963 1.12 0.574 1.13 0.676 1.26 0.589 1.26 0.284 1.30 0.347 1.45 0.432
rs7521536 1.24 0.320 1.47 0.191 1.01 0.977 1.24 0.360 1.11 0.705 2.95 0.104
rs7529377 1.25 0.321 1.31 0.360 1.40 0.506 1.26 0.331 1.14 0.651 2.93 0.106
rs7540065 1.00 0.997 1.03 0.922 0.95 0.896 1.08 0.726 1.04 0.883 1.35 0.570
rs912767 1.30 0.240 1.40 0.250 1.40 0.505 1.26 0.331 1.14 0.651 2.93 0.106
rs9425722 0.44 0.048 0.44 0.065 1.97E-09 0.999 NA NA NA NA NA NA
rs9425725 0.60 0.091 0.63 0.154 2.03E-09 0.998 NA NA NA NA NA NA
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 11 of 15
coding region SNPs and conformational changes in the
protein, we examined regulatory effects. We found sev-
eral promoter SNPs that change the sequence of poten-
tial TFBSs based on conservation data. We were unable
to confirm that these sequences were in fact TFBSs due
to the lack of available data. However if any of our pro-
moter SNPs showed a significant association with ALI
or another phenotype perhaps using a l arger sample
size, futu re studies using promoter constructs could
offer more information on upregulation of PRDX6. We
also found several SNP s in the 3UTR. It is possible that
one or more of these SNPs is responsible for changing
an miRNA binding site, thus repressing protein
translation.
Our study has several limitations. One potential lim-
itation of this study is the number of genotype call fail-
ures. Ten and nine markers for African Americans and
European Americans respectively were eliminated from
our analysis since they were under the 95% completion
rate cut-off. This high rate of genotype failure was due
to difficulties with c onsistent assay performance rather
than DNA quality. If these genotypes had been obtained,
itisapossiblethatanassociationmayhavebeen
observed. Also, we did not adjust our results for ances-
try informative markers (AIMs). Instead our population
was st ratified based on skin co lor, which may not be an
adequate proxy for population admixture effects.
Another possible limitation is a candidate gene approach
that focused on a single gene: PRDX6.ALIriskmaybe
considered a complex phe notype, and thus likely is not
fullyexplainedbyavariationinasinglegene[10].
Finally, we only tested for association in patients with
ALI from severe trauma. Thus, it is possible that PRDX6
may play a role in the initiation or severity of ALI after
other insults, including sepsis, or in determining recov-
ery from ALI.
PRDX6 has been shown to play a role not only in ALI,
but othe r diseases as well. A recent studied demon-
strated that PRDX6 promotes lung cancer metastasis
and invasion via phospholipase A
2
activity in mi ce [32].
Figure 2 Haplotype Structure of 17 PRDX6 SNPs in African Americans with r
2
Values (N = 259).
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 12 of 15
Figure 3 Haplotype Structure of 17 PRDX6 SNPs in European Americans with r
2
Values (N = 254).
Table 9 Haplotype Analysis among African and European Americans
African Americans European Americans
Block Haplotype Population Frequencies p value Block Haplotype Population Frequencies p value
Block 1 GT 0.654 0.719 Block 1 ACA 0.690 0.520
AC 0.330 0.833 GTA 0.230 0.479
GC 0.016 0.563 GTG 0.080 0.999
Block 2 TGTTACGTG 0.282 0.449 Block 2 GGACGTG 0.734 0.883
TATTGCACA 0.212 0.386 AAGTACA 0.202 0.634
CATCACATG 0.153 0.083 GAACATG 0.050 0.297
TACTACGTG 0.099 0.485
TGTTACATG 0.072 0.455
TATTACATG 0.055 0.935
TATTGTACA 0.051 0.894
TATTACACG 0.036 0.969
TATCACATG 0.028 0.953
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 13 of 15
Another publication reported that PRDX6 transfected
breast cancer cells metastasized more readily to the
lungs when compared with control cells [ 26]. It is possi-
ble that our novel SNPs may function in lung cancer as
well as ALI. The interaction between GSTpi and PRDX6
is another interesting subject for future studies. GSTpi
expression is elevated in tumors from a variety of can-
cers, including lung cancer, compared to normal tissue
[33]. Testing gene-gene interactions between PRDX6
and GSTpi would be an interesting future direction
both in ALI and other diseases such as cancer.
Conclusion
In conclusion, this study revealed novel SNPs within the
important anti-oxidant PRDX6 ge ne and its 5 and 3
flanking re gions via direct sequencing. Several of these
variants have putative function and m ay be useful for
future gene association studies. Although there was no
association discovered between our novel and tagging
SNPs with trauma-related ALI, future stu dies m ay focus
on the role of PRDX6 variation in other at risk groups, as
well as other diseases.
Additional material
Additional file 1: PCR primers and cycling conditions.
Additional file 2: Detectable relative risk vs. disease allele
frequency.
Additional file 3: Sequencing variants separated by race and case-
control status.
Additional file 4: Confidence score for sequencing genotypes with
only one variant.
Additional file 5: Hardy-Weinberg equilibrium values.
Additional file 6: Dominant and recessive models in African
Americans and European Americans .
Acknowledgements
This work was supported by HL60290, HL079063, The Doris Duke Charitable
Foundation
Author details
1
Division of Pulmonary and Critical Care Medicine, Department of Medicine,
University of Pennsylvania School of Medicine, 3600 Spruce Street,
Philadelphia, 19104, USA.
2
Department of Biostatistics and Epidemiology,
University of Pennsylvania School of Medicine, 423 Guardian Drive,
Philadelphia, 19104, USA.
3
Division of Oncology, Childrens Hospital of
Philadelphia, 34
th
Street and Civic Center Boulevard, Philadelphia, 19104,
USA.
4
Institute for Environmental Medicine, University of Pennsylvania, 3620
Hamilton Walk, Philadelphia, 19104, USA.
Authors contributions
MR carried out the sequencing and genotyping analysis and drafted the
manuscript. RA participated in the design of the study, supervision of
laboratory assays, and interpretation of data. NM participated in the design
of the study, interpretation of data, and manuscript editing. ML performed
statistical analyses. RF performed the statistical analyses. PNL participated in
the design of the study, collection of data, and manuscript editing. RG
performed statistical analyses. SB performed the statistical analyses. ARL
performed statistical analyses. SIF participated in the design of the study and
interpretation of data. ABF participated in the design of the study,
interpretation of data, and manuscript editing. SMA participated in the
design of the study, and interpretation of data. JDC participated in the
design of the study, data collection, interpretation of the data, manuscript
drafting and manuscript editing. All authors read and approved the final
manuscript
Competing interests
The authors declare that they have no competing interests.
Received: 15 July 2010 Accepted: 31 May 2011 Published: 31 May 2011
References
1. Neff M, Rubenfeld G: Clinical Epidemiology of Acute Lung Injury. Seminars
in Respiratory and Critical Care Medicine 2001, 22:237-246.
2. Rubenfeld GD: Epidemiology of acute lung injury. Critical Care Medicine
2003, 31:S276-84.
3. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K , Hudson L, Lamy M,
LegallJR,MorrisA,SpraggR:The A merican European consensus
conference on ARDS: definitions, mechanisms, relevant outcomes
and clinical trials coordination. Am J Resp Crit Care Med 1994,
149:818-824.
4. Gunther A, Walmrath D, Grimminger F, Seeger W: Pathophysiology of
Acute Lung Injury. Seminars in Respiratory and Critical Care Medicine 2001,
22:247-258.
5. Moss M, Mannino DM: Race and gender differences in acute respiratory
distress syndrome deaths in the United States: an analysis of multiple-
cause mortality data (1979-1996). Critical Care Medicine 2003,
30:1679-1685.
6. Manevich Y, Feinstein SI, Fisher AB: Activation of the antioxidant enzyme
1-CYS peroxiredoxin requires glutathionylation mediated by
heterodimerization with πGST. Proceedings of the National Academy of
Sciences of the United States of America 2004, 101:3780-3785.
7. Schremmer B, Manevich Y, Feinstein SI, Fisher AB: Peroxiredoxins in the
lung with emphasis on peroxiredoxin VI. Sub-cellular Biochemistry 2007,
44:317-344.
8. Rhee GS, Chae HZ, Kim K: Peroxiredoxins: A historical overview and
speculative preview of novel mechanisms and emerging concepts in
cell signaling. Free Radical Biology & Medicine 2005, 38:1543-1552.
9. Manevich Y, Sweitzer T, Pak JH, Feinstein SI, Muzykantov V, Fisher AB: 1-Cys
peroxiredoxin overexpression protects cells against phospholipid
peroxidation-mediated membrane damage. Proceedings of the National
Academy of Sciences of the United States of America 2002, 99:11599-604.
10. Christie JD, Ma SF, Aplenc R, Li M, Lanken PN, Shah CV, Fuchs B,
Albelda SM, Flores C, Garcia JGN: Variation in the myosin light chain
kinase gene is associated with development of acute lung injury after
major trauma. Critical Care Medicine 2008, 36:2794-2800.
11. Shah CV, Localio AR, Lanken PN, Kahn JM, Bellamy S, Gallop R, Finkel B,
Gracias VH, Fuchs BD, Christie JD: The impact of development of acute
lung injury on hospital mortality in critically ill trauma patients. Critical
Care Medicine 2008, 36:2309-2315.
12. Meyer NJ, Li M, Feng R, Bradfield J, Gallop R, Bellamy S, Fuchs BD,
Lanken PN, Albelda SM, Rushefski M, Aplenc R, Abramova H, Atochina
Vasserman EN, Beers MF, Calfee CS, Cohen MJ, Pittet JF, Christiani DC,
OKeefe GE, Ware LB, May AK, Wurfel MM, Hakonarson H, Christie JD:
ANGPT2 Genetic Variant is Associated with Trauma-Associated Acute
Lung Injury and Altered Plasma Angiopoietin-2 Isoform Ratio. American
Journal of Respiratory and Critical Care Medicine 2011.
13. Kruglyak L, Nickerson DA: Variation is the spice of life. Nature Genetics
2001, 27
:234-236.
14.
De Bakker PIW, Yelensky R, Peer I, Gabriel SB, Daly MJ, Altshuler D:
Efficiency and Power in Genetic Association Studies. Nature Genetics
2005, 37:1217-1223.
15. De La Vega FM, Lazaruk KD, Rhodes MD, Wenz MH: Assessment of two
flexible and compatible SNP genotyping platforms: Taqman SNP
genotyping assays and the SNPlex genotyping system. Mutation Research
2005, 573:111-135.
16. Schug J, Overton GC: TESS: Transcription Element Search Software on the
WWW. Technical Report CBIL-TR-1997-1001-v0.0. Computational Biology
and Informatics Laboratory School of Medicine University of Pennsylvania 1997
[http://www.cbil.upenn.edu/tess].
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 14 of 15
17. The ENCODE Project Consortium: Identification and analysis of functional
elements in 1% of the human genome by the ENCODE pilot project.
Nature 2007, 447:799-816.
18. Hiard S, Charlier C, Coppieters W, Georges M, Baurain D: Patrocles: a
database of polymorphic miRNA-mediated gene regulation in
vertebrates. Nucleic Acids Research 2010, D640-D651.
19. Xie X, Lu J, Kulbokas EJ, Golub TR, Mootha V, Lindblad-Toh K, Lander ES,
Kellis M: Systematic discovery of regulatory motifs in human promoters
and 3 UTRs by comparison of several mammals. Nature 2005,
434:338-345.
20. Lewis BP, Burge CB, Bartel DP: Conserved seed pairing, often flanked by
adenosines, indicates that thousands of human genes are microRNA
targets. Cell 2005, 120:15-20.
21. Menashe I, Rosenberg PS, Chen BE: PGA: Power calculator for case-control
genetic association analyses. BMC Genetics 2008, 9:36.
22. Gabriel SB, Schaffner SF, Nguyen H, Moore JM, Roy J, Blumenstiel B,
Higgins J, DeFelice M, Lochner A, Faggart M, Liu-Cordero SN, Rotimi C,
Adeyemo A, Cooper R, Ward R, Lander ES, Daly MJ, Altshuler D: The
structure of haplotype blocks in the human genome. Science 2002,
296:2225-2229.
23. Barrett JC, Fry B, Maller J, Daly MJ: Haploview: Analysis and visualization
of ld and haplotype maps. Bioinformatics 2005, 21:263-265.
24. Gabriel SB, Schaffner SF, Nguyen H, Moore JM, Roy J, Blumenstiel B,
Higgins J, DeFelice M, Lochner A, Faggart M, Liu-Cordero SN, Rotimi C,
Adeyemo A, Cooper R, Ward R, Lander ES, Daly MJ, Altshuler D: The
structure of haplotype blocks in the human genome. Science 2002,
296:2225-2229.
25. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, Maller J,
Sklar P, de Bakker PI, Daly MJ, Sham PC: Plink: A tool set for whole-
genome association and population-based linkage analyses. American
Journal of Human Genetics 2007, 81:559-575.
26. Chowdhury I, Mo Y, Gao L, Kazi A, Fisher AB, Feinstein SI: Oxidant Stress
Stimulates Expression of the Human Peroxiredoxin 6 (Prdx6) Gene by a
Transcriptional Mechanism Involving an Antioxidant Response Element.
Free Radical Biology and Medicine 2009, 46:146-153.
27. Wang Y, Feinstein SI, Fisher AB: H
2
O
2
-dependent Hyperoxidation of
Peroxiredoxin 6 (Prdx6) Plays a Role in Cellular Toxicity via Up-
regulation of iPLA2. Journal of Cellular Biochemistry 2008, 104:1274-1285.
28. Lukosz M, Jakob S, Büchner N, Zschauer T, Altschmied T, Haendeler J:
Antioxidants & Redox Signaling. 2010, 12:713-742.
29. Laird PW: Principles and Challenges of Genome-wide DNA Methylation
Analysis. Nature Reviews Genetics 2010, 11:191-203.
30. The 1000 Genomes Project Consortium: A map of human genome
variation from population-scale sequencing. Nature 2010, 467:1061-1073.
31. Lim LP, Glasner ME, Yekta S, Burge CB, Bartel DP: Vertebrate microRNA
Genes. Science 2003, 299:1540.
32. Ho JN, Lee SB, Lee SS, Yoon SH, Kang GY, Hwang SG, Um HD:
Phospholipase A2 activity of peroxiredoxin 6 promotes invasion and
metastasis of lung cancer cells. Molecular Cancer Therapuetics 2010,
9(4):825-32.
33. Moscow JA, Fairchild CR, Madden MJ, Ransom DT, Wieand HS, OBrien EE,
Poplack DG, Cossman J, Myers CE, Cowan KH: Expression of anionic
glutathione-S-transferase and P-glycoprotein in human tissues and
tumors. Cancer Research 1989, 49(6):1422-1428.
Pre-publication history
The pre-publication history for this paper can be accessed here:
http://www.biomedcentral.com/1471-2350/12/77/prepub
doi:10.1186/1471-2350-12-77
Cite this article as: Ru shefski et al.: Novel variants in the PRDX6 Gene
and the risk of Acute Lung Injury following major trauma. BMC Medical
Genetics 2011 12:77.
Submit your next manuscript to BioMed Central
and take full advantage of:
Convenient online submission
Thorough peer review
No space constraints or color figure charges
Immediate publication on acceptance
Inclusion in PubMed, CAS, Scopus and Google Scholar
Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submit
Rushefski et al. BMC Medical Genetics 2011, 12:77
http://www.biomedcentral.com/1471-2350/12/77
Page 15 of 15
  • [Show abstract] [Hide abstract] ABSTRACT: Peroxiredoxins, which reduced intracellular peroxides as a noval kind of antioxidant protein, were extensively expressed in various types of cancers and were thought as a biomarker of cancer cells. In this work, we performed genotyping analyses for tag SNP of Prdx 1, 2 and 6, and then evaluated the association with susceptibility and clinic stage of esophageal squamous cell carcinoma (ESCC) in a case-control study. The protein level of these Prdx isoforms in ESCC cancer samples was evaluated by Western blot. Then 356 ESCC cancer cases and 315 controls were genotyped by SNPshot assay. Differences in frequencies of the genotypes of the SNPs variant between the cases and controls were evaluated by using the chi-square test. Our result of Western blot confirmed the aberrant expression of Prdx 1, 2 and 6 in ESCC samples, which was coincident with other studies. After genotyping by SNPshot assay, the result showed that the allele and genotype frequencies did not differ between the patients and controls. And no association between the polymorphism and the progression of ESCC including tumor grade and stage was found. Our data suggested that polymorphisms of Prdx 1, 2 and 6 were not associated with esophageal cancer.
    Article · Jan 2012
  • [Show abstract] [Hide abstract] ABSTRACT: In this study, a genetically-diverse panel of 43 mouse strains was exposed to phosgene and genome-wide association mapping performed employing a high density SNP assembly. Transcriptomic analysis was also used to improve the genetic resolution in the identification of genetic determinants of phosgene-induced acute lung injury. We prioritized the identified genes based on whether the encoded protein was previously associated with lung injury or contained a nonsynonymous SNP within a functional domain. In addition, candidates were selected that contained a promoter SNP that could alter a putative transcription factor binding site and had variable expression by transcriptomic analyses. The latter 2 criteria also required that ≥10% of mice carried the minor allele and that this allele could account for ≥10% of the phenotypic difference noted between the strains at the phenotypic extremes. This integrative functional approach revealed 14 candidate genes that included Atp1a1, Alox5, Galnt11, Hrh1, Mbd4, Phactr2, Plxnd1, Ptprt, Reln, and Zfand4, which had significant SNP associations, and Itga9, Man1a2, Mapk14, and Vwf, which had suggestive SNP associations. Of the genes with significant SNP associations, Atp1a1, Alox5, Plxnd1, Ptprt, and Zfand4 are particularly noteworthy and could be associated with acute lung injury in several ways. Using a competitive electrophoretic mobility shift analysis, Atp1a1 promoter (rs215053185) oligonucleotide containing the minor G-allele formed a major distinct faster-migrating complex. In addition, a gene with a suggestive SNP association, Itga9, is linked to TGFB1-signaling, which previously has been associated with the susceptibility to acute lung injury in mice.
    Full-text · Article · Apr 2013
  • [Show abstract] [Hide abstract] ABSTRACT: Objective: Moderate-to-severe psoriasis is a chronic skin condition that often requires systemic therapy and biologics are the newest systemic treatment available. A favorable aspect of biologics is that they are thought to have minimal risks for drug-drug interactions compared to oral systemic medications such as cyclosporine and methotrexate. However, this assumption has not been recently or adequately reviewed. We reviewed the literature to identify possible drug-drug interactions with biologics and other medications. Methods: We searched PubMed for published case reports, clinical studies, reviews, and Food and Drug Administration (FDA) labels discussing possible drug-drug interactions with biologics for the treatment of moderate-to-severe psoriasis. Results: There were only a small number of published articles describing drug-drug interactions with biologics. Our review identified two case reports, five clinical studies, and three pharmacokinetics reviews. The majority of articles did not observe clinically relevant drug-drug interactions with biologics. FDA labels do suggest a possible relationship between tumor necrosis factor (TNF)-alpha inhibitors and cytochrome P450 (CYP) activity. Conclusion: The paucity of information regarding drug-drug interactions reaffirms the idea that biologics have limited susceptibility to drug-drug interactions compared to other oral medications. Further studies are needed to adequately assess drug-drug interactions with biologics.
    Article · Jul 2013
Show more