Content uploaded by Lorena Madrigal
Author content
All content in this area was uploaded by Lorena Madrigal on May 17, 2021
Content may be subject to copyright.
Fetal Hemoglobin Modulators May Be
Associated With Symptomology of Football
Players with Sickle Cell Trait
Carroll Flansburg, MA,MPH, Christina M. Balentine, BS, Ryan W. Grieger, MS, Justin Lund, MA,
Michelle Ciambella, BS, Deandre White, BS, Eric Coris, MD, Eduardo Gonzalez, MD,
Anne C. Stone, PhD, and Lorena Madrigal, PhD
Objectives: This study investigates whether genetic modifiers previ-
ously shown to influence adult fetal hemoglobin (HbF) levels and
glucose-6-phosphate dehydrogenase deficiency were associated with
variable symptomology in a small sample of collegiate football players
with sickle cell trait.
Methods: Survey data on self-assessed symptoms and genotype data
from five single nucleotide polymorphisms (SNPs) related to HbF
production and two SNPs that cause glucose-6-phosphate dehydro-
genase def iciency were collected from current and former college
football players.
Results: In this sample, SNPs found within the β-globin gene cluster
were found to be associated with a previous diagnosis of exertional sick-
ling and experience of extreme heat during and after training.
rs10189857 in the BCL11A gene was associated with body mass index
and weight and with experiencing extreme thirst during and after train-
ing. No significant correlations were found between the other SNPs and
symptoms within this sample.
Conclusions: These findings show that genetic variation known to
affect sickle cell disease symptomology may partly explain why some
football players with sickle cell trait experience adverse clinical out-
comes during periods of extreme physical exertion and others do not.
Key Words: adult HbF production, athletes with SCT, developmental
activation of hemoglobin genes, ECAST, sickle cell trait (SCT)
Sickle cell disease (SCD) results from the inheritance of two
β-globin alleles (HbS) with an amino acid substitution at
the sixth position, where valine replaces glutamic acid.
1
Clinicians
have noted that patients with SCD differ in their disease expres-
sion,
2
mostly because of patients’levels of fetal hemoglobin
Key Points
•Exercise collapse associated with sickle cell trait (SCT) remains
largely unexplained.
•This study assessed whether genetic modifiers previously shown
to affect adult fetal hemoglobin levels and clinical variance
among patients with sickle cell disease also are associated with
variable symptomology in a small sample of collegiate football
players with SCT.
•We genotyped collegiate football players with SCT for single
nucleotide polymorphisms (SNPs) previously shown to affect
levels of fetal hemoglobin and asked the athletes to complete a
survey about the presence of symptoms associated with exercise
collapse associated with SCT, and to compare themselves with
their peers without SCT.
•We found statistically significant associations between SNPs and
symptoms, and between one SNP and greater body weight and
body mass index.
•We have demonstrated that SNPs previously shown to affect clinical
variation in patients with sickle cell disease also are significantly
associated with clinical variation in football players with SCT.
From the Geisinger Health System, Danville, Pennsylvania, the School of Human
Evolution and Social Change, Arizona State University, Tempe, Arizona, the
Sch ool of Life Sciences, Arizona State University, Tempe, Arizona, the Department
of Anthropology, University of South Florida, Tampa, the Morsani College of Med-
icine, University of South Florida, Tampa, and the Center for Evolution and Medi-
cine and the School of Life Sciences, Arizona State University, Tempe, Arizona
Correspondence to Dr Lorena Madrigal, Department of Anthropology, University
of South Florida, 4202 E. Fowler Ave, Tampa, FL 33620. E-mail:
madrigal@usf.edu. To purchase a single copy of this article, visit sma.org/smj-
home. To purchase larger reprint quantities, please contact
Reprintsolutions@wolterskluwer.com.
Carroll Flansburg and Christina M. Balentine: These authors contributed equally.
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text, and links to the digital files are provided in the
HTML text of this article on the journal’s Web site (http://sma.org/smj-home).
This projectwas funded by an NCAA graduate studentresearch grant, awarded to
C.F. while at the University of South Florida.
C.M.B. has received compensation from Arizona State University; the
Environmental Science Institute; the University of Texas at Austin for the
Scientist in Residence program and the Ecology, Evolution, and Behavior
program; the Arizona Science Center; the National Science Foundation
Graduate Research Fellowship Program; the Society for the Study of
Evolution; and Sigma Xi. E.C. has received compensation from NCAA and
the US Department of Justice. A.C.S. has received compensation from
NCAA, the National Institutes of Justice, and the National Science
Foundation. L.M. has received compensation from NCAA. The remaining
authors did not report any financial relationships or conflicts of interest.
Accepted January 9, 20 19.
Copyright © 2019 by The Southern Medical Association
0038-4348/0–2000/112-289
DOI: 10.14423/SMJ.0000000000000976
Original Article
Southern Medical Journal •Volume 112, Number 5, May 2019 289
No copyeditor name!
Copyright © 2019 The Southern Medical Association. Unauthorized reproduction of this article is prohibited.
(HbF), such that increasing levels are associated with better
patient outcomes.
3,4
Hemoglobin is one of the best-studied proteins in humans
because of its role in distributing oxygen to the organs and other
peripheral tissues and because of its clinical importance as the
cause of inherited anemias, including SCD. Hemoglobin consists
of four polypeptide chains, two αand two βchains (α
2
β
2
). The
α-andβ-globin genes that produce these protein subunits are
found in clusters on chromosomes 16 and 11, respectively. The
expression of the three different α-like and five β-like globin
genes involved in the production of hemoglobin is a classic exam-
ple of cell and developmental-stage specific regulation.
5–7
In
brief, during the embryonic stage of development, three types of
embryonic hemoglobin (HbE) are produced through the activity
of the HBZ (ζ), HBA, and HBA2 (α)α-like globin genes, and
the HBE1 (ε), HBG1, and HBG2 (γ)β-like globin genes (as
the tetramers ζ
2
ε
2
,α
2
ε
2
,andα
2
γ
2
). During the first 3 months
of gestation, production of HbE declines as HbF becomes pre-
dominant. HbF consists of two αand two γchains (α
2
γ
2
). In
addition, during the fetal stage, the HBB (β)β-like globin gene
is expressed at a low level and increases as birth approaches. After
birth, a minor adult β-like gene, HBD (δ)alsoisexpressed.Dur-
ing the first year of life, the adult forms of hemoglobin (α
2
β
2
,
α
2
δ
2
) become predominant. Most babies are born producing both
fetal and adult hemoglobin; however, levels of adult hemoglobin
2(α
2
δ
2
) remain low because of past changes in regulatory ele-
ments in the promoter of HBD.
8
In adulthood, most people pro-
duce >95% of adult hemoglobin 1 (α
2
β
2
). Although HbF
typically declines after birth, there is variation in the levels of
HbF in adults. It has been shown that greater levels of HbF in
adults ameliorate symptoms of hemoglobinopathies.
6,9–12
Adults with HbF levels greater than ~5% have hereditary
persistence of fetal hemoglobin (or HPFH) as a result of variation
in the regulation of HBG.
13–15
Important modifiers of HbF pro-
duction discovered thus far include the transcriptional repressors
produced by the ZBTB7A and BCL11A genes and regulatory
changes found in such regions as the HBS1L-MYB intergenic
interval and in the promoter region of the HBG genes.
14,16–20
Researchers have quantified the variation in levels of HbF as a
result of these single nucleotide polymorphisms (SNPs) to be
up to 15% to 20%.
5,18
As such, although SCD is the result of
a single Mendelian mutation, its severity and clinical manifesta-
tions are the result of polygenic interactions, including those
that affect HbF, which contradict a simple Mendelian view.
21
Although the diverse clinical manifestations of SCD have
been established and explained for decades,
22
sickle cell trait
(SCT)—the heterozygous condition—was previously believed
to be uniformly benign. Most textbooks note that individuals
with SCTare symptom free except under extreme conditions.
23–25
Such conditions include altitude-hypoxic environments
26,27
and
periods of extreme exertion such as military
28–31
or sports train-
ing.
32
Following the work by Kark and colleagues during the
1980s, the US military engaged in an intervention study from
1982 to 1991 to determine whether better prevention of heat
illness would reduce the deaths of recruits and reduce excess
SCT deaths.
33
This study, which has not been published, showed
that the intervention reduced or eliminated the excess mortality
for recruits with SCT.
28
Because of this reduction in mortality,
the US Army dropped SCT screening upon recruitment; however,
as noted by Ferster and Eichner in 2012, exercise-related deaths
continued.
28
A more recent study demonstrated that although
US Army soldiers with SCT did not have higher mortality, they
did have a higher risk of exertional rhabdomyolysis, which is
the severe breakdown of skeletal muscle tissue precipitated
by strenuous physical exertion.
34
It is unfortunate that the increased risk of death associated
with SCT in sports has not decreased.
35
The US Registry of Sud-
den Death in Athletes collected data for 31 years, 1980–2011. It
notes that 23 deaths occurred in athletes with SCT, all of whom
were African American and who died in remarkably similar cir-
cumstances involving noninstantaneous collapse with rapid dete-
rioration following vigorous physical exertion at the early phase
of training and conditioning. A majority of the athletes with
SCT (19/23) were football players.
32
Notably, National Collegiate
Athletic Association (NCAA) Division I football players with
SCT had a risk of death that was 37 times higher than players
who did not have SCT.
36
These alarming statistics led a 2012
expert panel of military and civilian experts in SCT and sports
medicine to introduce the term “exercise collapse associated
with SCT,”or ECAST, to refer to the exercise-related complica-
tions experienced by the carriers of SCT.
37
Most football players with SCT do not have clinical com-
plaints or die; in fact, some have successful careers on profes-
sional teams. The question then becomes, what causes an
individual with SCT to be at an increased risk of ECAST,
whereas another exhibits no symptoms?
32
Most of the literature
addresses this differential risk in terms of environmental factors
such as hypoxia, extreme heat and humidity, dehydration,
asthma, fatigue, lack of sleep, poor conditioning, and high exer-
cise intensity.
37,38
To our knowledge, there has been no previous
work testing the hypothesis that genetic modifiers of HbF pro-
duction may at least in part explain why some but not all athletes
with SCT experience exercise-induced complications.
The purpose of our study was to assess whether genetic
modifiers previously shown to affect adult HbF levels and clinical
variance among patients with SCD
9,17,39,40
also are associated
with variable symptomology in a small sample of collegiate foot-
ball players with SCT. We restricted our study to football players
because of their increased risk of death
32,36
and to maintain
homogeneity in our sample. We focused on symptoms associ-
ated with exercise-induced complications in football players
with SCT. We hypothesized that a higher risk of experiencing
at least some of these symptoms could be caused by the presence
or absence of SNPs associated with HbF production.
Methods
The protocol for this study was reviewed and approved by the
University of South Florida medical institutional review board
Flansburg et al •Fetal Hemoglobin Modulators and SCT in Football Players
290 © 2019 The Southern Medical Association
Copyright © 2019 The Southern Medical Association. Unauthorized reproduction of this article is prohibited.
and reviewed at the Arizona State University Office of Research
Integrity and Assurance on a yearly basis. The criteria for inclu-
sion were being male, being at least 18 years of age, having had
confirmation of SCT status with a blood test, and having played
or currently playing American football. When the participants
signed the consent form, they agreed to a statement to the effect
that completion and submission of the survey and cheek swab
constituted consent for the information to be used in a research
paper in which their anonymity was ensured.
Study Design
We worked with current football players diagnosed as hav-
ing SCT from blood tests and whose symptoms were self-
assessed. The clinicians who designed and reviewed the survey
were blinded to the genetic data of the participants, the geneti-
cists were blinded to the clinical data of the participants, and
the statistical analysts were blinded to both.
Study Sample
We embarked on an intensive recruitment campaign of
NCAA Division I and II football teams from 2012 through
2016. Throughout 4 years of recruitment, the only successful
strategy for obtaining participants was via university coaching
and medical staff, who proved invaluable in encouraging current
and even former team players with SCT to participate in our
study. We decided to close recruitment when we obtained 31
samples from male football players (current and former), and
we then proceeded with the data analysis. Because of missing
data, our final sample size was 29.
Data Collection
Participants received a brief written survey, which asked
them to check whether they experienced the presence or a higher
frequency of symptoms than did their non-SCT teammates.
These symptoms included drenching sweats, extreme thirst during
and after training, hard muscles, whole-body muscle cramps,
hematuria, and low-back pain. All of the symptoms were there-
fore self-assessed. The only symptom that was not self-assessed
(although it was self-reported) was “Have you ever been told you
suffered from exertional sickling?”The survey can be found in the
Supplemental Digital Content (http://links.lww.com/SMJ/A148).
Because the most important modifiers of HbF production
are the intergenic region HBS1L-MYB, the BCL11A gene,
and the promoter region of the γgene itself, we genotyped five
SNPs found in these three regions. There are two SNPs at the pro-
moter region of the γgene within the β-globin gene cluster
(rs7482144 and rs10128556), two at the BCL11A region
(rs10189857 and rs4671393), and one at the HBS1L-MYB
intergenic region (rs9402686).
18,41,42
To refer to the two forms
of the SNP as ancestral or derived, we follow the assignment
given by the SNP Database (https://www.ncbi.nlm.nih.gov/snp).
The SNP Database is an active archive maintained by the
National Center of Biotechnology Information, which determines
which nucleotide has been changed (derived) from its original
(ancestral) form. Also included in the study were two types of
glucose-6-phosphate-dehydrogenase deficiency (G6PDD) because
of the possibility of oxidative stress in red blood cells adding
to the risk of exertional sickling.
DNA was extracted from participants’cheek swabs following
the standard phenol-chloroform extraction protocol, and each sample
was genotyped at the seven SNPs noted above using polymerase
chain reaction with restriction digest and quantitative polymerase
chain reaction-based genotyping methods. The five HbF SNPs are
autosomal, so homozygous or heterozygous results were expected
and observed in this sample. Conversely, the G6PD gene is found
on the X chromosome; thus, hemizygous results (which should look
the same as homozygosity) were expected in this all-male cohort.
One individual, however, exhibited heterozygosity for both of the
G6PDD SNPs, so standard methods were used to rule out possible
contamination of the sample by female laboratory researchers.
Although this anomaly was likely not due to contamination dur-
ing DNA extraction and genotyping, this individual was never-
theless removed from the analyses of G6PDD to maintain
homogeneity in the sample. A detailed description of the DNA
extraction and genotyping methods used can be found in the Sup-
plemental Digital Content (http://links.lww.com/SMJ/A148).
Analytic Methods
We determined whether the frequency of the symptoms
from the survey was associated with the frequency of any of
the SNPs using the Fisher exact test. We used SAS version 9.4
(SAS Institute, Cary, NC) for our statistical analyses.
43
Results
Gene Frequencies
Table 1 shows the allelefrequencies offive SNPs associated
with HbF production and two SNPs associated with G6PDD for
the entire sample (n = 29 for the HbF SNPs and n = 28 for the
G6PD deficiency SNPs, because one male participant appeared
to be heterozygous at this X-linked gene and was therefore
excluded from the analysis). The frequencies of both β-globin
gene cluster SNPs are identical as a result of their tight linkage;
in other words, both SNPs are so close to each other in the gene
that their frequency in this sample is identical.
All of the SNPs, except rs10189857 in the BCL11A gene,
were in Hardy-Weinberg equilibrium, which means that the
observed and expected frequencies are not significantly differ-
ent. For rs10189857, however, the number of heterozygotes
(AG = 16) exceeds that of the ancestral-type homozygotes
(AA = 13), whereas no derived GG homozygotes are observed
(P< 0.05). Previous studies have shown that the derived allele
of the rs10189857 in the BCL11A SNP increases HbF in adults.
Original Article
Southern Medical Journal •Volume 112, Number 5, May 2019 291
Copyright © 2019 The Southern Medical Association. Unauthorized reproduction of this article is prohibited.
The Homozygotes and Heterozygotes for Rs10189857
SNP at the BCL11A Differ in their Mean Body Mass
Index (BMI) and Weight
We hypothesized that because heterozygotes for the
rs10189857 SNP at the BCL11A produce more HbF, athletes
would be able to train more and achieve greater musculature.
To test this hypothesis, we divided participants into genotypic
groups (AA and AG) for the rs10189857 SNP at the BCL11A
and computed their mean height, weight, and BMI. According
to our expectations, the heterozygotes were 10 k heavier
(P= 0.04) and almost 2 BMI units larger than the homozygotes
(P= 0.04), whereas their heights were virtually identical. A one-
way test is appropriate in this case because our hypothesis was
that the heterozygotes would achieve greater weight and BMI
but not an equal or lesser weight or BMI with a median two-
sample test (Table 2).
Association of Alleles with Symptoms
Three symptoms were significantly associated with the
SNPs analyzed in this study. In the following, we present the
odds ratios (ORs) according to how the odds of each symptom
change from one SNP state to the other:
•A previous diagnosis of exertional sickling was associated with both
SNPs in the β-globin gene cluster, where the derived alleles are pro-
tective of a diagnosis of exertional sickling (OR
ancestral-derived
0.68,
95% confidence interval [CI] 0.51–0.9071). In our sample, the
two individuals who had been diagnosed with exertional sickling
were homozygote for the ancestral allele. Being heterozygote for
the derived allele, which has been shown to result in higher levels
of HbF, is protective of a previous diagnosis of exertional sickling.
•Experiencing extreme thirst during and after training was significantly
associated with the rs10189857 allele in BCL11A. Of the 16 heterozy-
gotes for the SNP, 10 reported the symptom (OR
ancestral-derived
9.17, 95%
CI 1.5–56.3). Being heterozygote for the derived allele, which has been
shown to result in higher HbF levels, is associated with significant odds
of extreme thirst during and after training.
•Experiencing extreme heat during and after training was significantly
associated with the derived allele of rs7482144 at the β-globin gene
cluster. Of the 4 individuals who reported the symptom, 3 were
heterozygotes for this SNP (OR
ancestral-derived
15.75, 95% CI
1.28–192.46). Being heterozygote for the derived allele, which
has been shown to result in higher HbF levels, is associated with
a significant odds of extreme heat during and after training.
Table 1. Allele frequencies for the entire sample
SNP rs7482144 rs10128556 rs10189857 rs4671393 rs9402686 rs1050829 rs1050828
Location β-Globin gene cluster β-Globin gene cluster BCL11A BCL11A HBS1L-MYB G6PD
a
G6PD
a
Alleles G = ancestral;
A=derivedallele
C = ancestral;
T = derived allele
A = ancestral;
G=derivedallele
G = ancestral;
A=derivedallele
G = ancestral;
A = derived allele
A = ancestral;
G = derived allele
G = ancestral;
A = derived allele
Allele frequencies
(N = 29)
P= 0.88, q = 0.12;
GG = 22, GA = 7
P= 0.88, q = 0.12;
GG = 22, GA = 7
P= 0.72, q = 0.28;
AA=13,AG=16
P= 0.81, q = 0.19;
GG = 19, GA = 9, AA =1
P= 0.91, q = 0.09;
GG = 23, GA = 6
P= 0.75, q = 0.25;
A=21,G=7
P= 0.82, q = 0.18;
G=23,A=5
Hardy-Weinberg equilibrium χ
2
=0.55,ns χ
2
= 0.55, ns χ
2
=4.21;P<0.05 χ
2
=0,ns χ
2
=0.39,ns
ns, not statistically significant; SNP, single nucleotide polymorphism.
a
n=28(29–1, where the possible XXY participant was excluded).
Table 2. Comparison of the anthropometrics of the
homozygotes and heterozygotes of the rs10189857 SNP
at the BCL11A
Variable Homozygotes Heterozygotes P(one-way)
Weight, k 86.7 96.2 0.044
BMI 25.625 27.84 0.044
Height, m 1.84 1.84 ns
Sample size 13 16
BMI, body mass index; ns, not statistically significant; SNP, single-nucleotide
polymorphism.
Flansburg et al •Fetal Hemoglobin Modulators and SCT in Football Players
292 © 2019 The Southern Medical Association
Copyright © 2019 The Southern Medical Association. Unauthorized reproduction of this article is prohibited.
Lack of Association Between G6PDD
SNPs and Symptoms
No significant association was found between the two
G6PDD SNPs and any of the self-reported symptoms.
Discussion
This study is the first of which we are aware in which HbF mod-
ulators have been examined in SCT football players. Previous
studies in cohorts of individuals with SCD have shown that
higher HbF levels may be associated with milder symptomology.
This line of research has established the importance of SNPs at
the intergenic region HBS1L-MYB, BCL11A, and the β-globin
gene cluster in modulating the levels of HbF, and therefore the
expression of SCD.
13,24,44,45
The exact clinical effects of increased
HbF levels in individuals with SCT are unknown; however, in
discussing the results of this study, we operate within the assump-
tion that increased HbF levels in individuals with SCT
32
also are
clinically advantageous. In our sample, being a derived hetero-
zygote at both SNPs in the β-globin gene cluster was signifi-
cantly protective of a diagnosis of exertional sickling.
In the same manner, the rs10189857 at the BCL11A also is
an important modulator of HbF levels in thalassemic patients.
46
In our sample, we found that being heterozygote for this SNP is
associated with significantly higher BMI and weight (an obvi-
ous advantage in football). This result helps us understand
why our sample is in Hardy-Weinberg disequilibrium for this
SNP: In a highly competitive environment for this particular
sport, individuals with SCT with higher levels of HbF will have
better physiological means to succeed.
The questionof why the heterozygotes ofthe rs10189857 at
the BCL11A SNP also report extreme thirst during and after
training more frequently than do ancestral homozygotes is impor-
tant. We propose the following to explain these results: (1) Het-
erozygotes have greater exertional capability/limits because of
their comparatively greater access to oxygen, which may result
in greater conditioning with corresponding relatively higher
weight and BMI. (2) Heterozygotes require additional exertional
levels because they need to displace additional body mass in
comparison to a lighter-weight individual performing a training
task of the same intensity. (3) Heterozygotes have comparatively
greater heat production and require greater water consumption
because of this relatively higher exertion. A similar mechanism
probably explains our result that experiencing extreme heat
during and after training was significantly associated with the
derived allele of rs7482144 at the β-globin gene cluster. This
derived allele also has been reported to increase the production
of HbF. It is likely that individuals who carry this SNP push
themselves more because of their higher levels of HbF, and as
a result, they report higher levels of extreme heat. These results
suggest that these specific SNPs may be associated with these
symptoms as a by-product of the other phenotypic manifesta-
tions of the SNPs.
All of the participants were sampled from NCAA Division I
and Division II teams; therefore, these college athletes underwent
extremely high levels of physical endurance. It is difficult to imag-
ine that the homozygotes for the rs10189857 at the BCL11A SNP
were randomly sampled from colleges that favor a small body size,
whereas the heterozygotes for the rs10189857 at the BCL11A SNP
were sampled from colleges that favor a large body size.
It is important to note that we applaud and support the
efforts of medical staff who insist that players be hydrated and
train slowly at the start of the season. We are complementing
their work by pointing out that genetic variation also can play
a role in the clinical manifestation of SCT during extreme phys-
ical exercise.
A potential weakness in our study is that we relied on the
subjective reporting of symptoms and diagnosis by subjects.
Our study also suffers from the shortcoming of having a small
sample size (N = 29). Our small sample size stems from the
tremendous difficulty in enrolling participants. After 4 years
of aggressive recruitment campaign, we decided that it was
time to publish our data, because a sample size of 30 is con-
sidered the point at which samples follow a normal distribu-
tion.
47
Moreover, as has been argued by others,
48
the usual
expectation that every single paper conform to high sample sizes
is unwarranted and detrimental to the development of innovative
research avenues.
This study is the first of its kind to show that genetic mod-
ifiers related to variable HbF levels in adults do in fact influence
the clinical manifestation of SCT in athletes. This is one of many
calls for the recognition of SCT having actual clinical manifesta-
tions for some individuals. Future research with larger sample
sizes, including nonathletes and individuals from numerous
demographic populations, is required to support our results.
It is time that the medical sciences acknowledge the variation
within the clinical manifestations of SCT.
Conclusions
We have demonstrated that SNPs previously shown to affect
clinical variation in patients with SCD also are significantly
associated with clinical variation in football players with SCT.
Acknowledgments
We thank the trainers, coaches, and football players who
have supported our efforts by participating and by encouraging
others to participate in this project.
We dedicate this paper to the memory of the athletes with
sickle cell trait who have died.
References
1. Ingram VM. Gene mutations in human haemoglobin: the chemical
difference between normal and sickle cel l haemoglobin. Nature 1957;180:
326–328.
2. Bauer DE, Kamran SC, Lessard S, et al. An erythroid enhancer of BCL11A
subject to genetic variation determines fetal hemoglobin level. Science
2013;342: 253–257.
Original Article
Southern Medical Journal •Volume 112, Number 5, May 2019 293
Copyright © 2019 The Southern Medical Association. Unauthorized reproduction of this article is prohibited.
3. Meier ER, Fasano RM, Levett PR. A systematic review of the literature for
severity predictors in children with sickle cell anemia. Blood Cells Mol Dis
2017;65:8 6–94.
4. Serjeant GR. Fetal hemoglobin in homozygous sickle-cell disease. Clin
Haematol 1975;4:109–122.
5. Hardison RC, Blobel GA. Genetics. GWAS to therapy by genome edits?
Science 2013;342:206–207.
6. Smith EC, Orkin SH. Hemoglobin genetics: recent contributions of GWAS
and gene editing. Hum Mol Genet 2016;25:R99–R105.
7. Steinberg MH. Predicting clinical severity in sickle cell anaemia. Br J
Haematol 2005;129:465–481.
8. Tang DC, Ebb D, Hardison RC, et al. Restoration of the CCAAT box or
insertion of the CACCC motif activates [corrected] delta-globin gene
expression. Blood 1997;90:421–427.
9. Akinsheye I, Alsultan A, Solovieff N, et al. Fetal hemoglobin in sickle cell
anemia. Blood 2011;118:19–27.
10. Basak A, Hancarova M, Ulirsch JC, et al. BCL11A deletions result in fetal
hemoglobin persistence and neurodevelopmental alterations. J Clin Invest
2015;125:2363–2368.
11. Funnell AP, Prontera P, Ottaviani V, et al. 2p15-p16.1 microdeletions
encompassing and proximal to BCL11A are associated with elevated HbF
in addition to neurologic impairment. Blood 2015;126:89–93.
12. Uda M, Galanello R, Sanna S, et al. Genome-wide association study
shows BCL11A associated with persistent fetal hemoglobin and
amelioration of the phenotype of beta-thalassemia. Proc Natl Acad Sci
USA2008;105:1620–1625.
13. Antoniani C, Romano O, Miccio A. Concise review: epigenetic regulation of
hematopoiesis: biological insights and therapeutic applications. Stem Cells
Transl Med 2017;6:2106–2114.
14. Martyn GE, Wienert B, Yang L, et al. Natural regulatory mutations elevate
the fetal globin gene via disruption of BCL11A or ZBTB7A binding. Nat
Genet 2018;50:498–503.
15. Thein SL, Menzel S, Lathrop M, et al. Control of fetal hemoglobin: new
insights emerging from genomics and clinical implications. Hum Mol Genet
2009;18:R216–R223.
16. Bhatnagar P, Purvis S, Barron-Casella E, et al. Genome-wide association
study identif ies genetic variants influencing F-cell levels in sickle-cell patients.
JHumGenet2011;56:316–323.
17. Galarneau G, Palmer CD, Sankaran VG, et al. Fine-mapping at three loci
known to affect fetal hemoglobin levels explains additional genetic variation.
Nat Genet 2010;42:1049–1051.
18. Lettre G, Sankaran VG, Bezerra MA, et al. DNA polymorphisms at the
BCL11A, HBS1L-MYB, and beta-globin loci associate with fetal hemoglobin
levels and pain crises in sickle cell disease. Proc Natl Acad Sci U S A
2008;105:11869–11874.
19. Masuda T, Wang X, Maeda M, et al. Transcription factors LRF and BCL11A
independe ntly repress expressi on of fetal hemoglobin. Science 2016;351:
285–289.
20. Thein SL, Menzel S, Peng X, et al. Intergenic variants of HBS1L-MYB are
responsible for a major quantitative trait locus on chromosome 6q23
influencing fetal hemoglobin levels in adults. Proc Natl Acad Sci U S A
2007;104:11346–11351.
21. Bauer DE, Orkin SH. Hemoglobin switching's surprise: the versatile
transcription factor BCL11A is a master repressor of fetal hemoglobin.
Curr Opin Genet Dev 2015;33:62–70.
22. Kulozik AE, Kar BC, Satapathy RK, et al. Fetal hemoglobin levels and beta
(s) globin haplotypes in an Indian populations with sickle cell disease. Blood
1987;69:1742–1746.
23. Connes P, Reid H, Hardy-Dessources MD, et al. Physiological responses of
sickle cell trait carriers during exercise. Sports Med 2008;38:931–946.
24. National Heart, Lung, and Blood Institute. Questions and answers about
sickle cell trait. https://www.nhlbi.nih.gov/news/2010/questions-and-
answers-about-sickle-cell-trait. Published September 22, 2010. Accessed
February 26, 2019.
25. Tsaras G, Owusu-Ansah A, Boateng FO, et al. Complications associated
with sickle cell trait: a brief narrative review. Am J Med 2009;122:507–512.
26. Hayashi TY, Matsuda I, Hagiwara K, et al. Massive splenic infarction and
splenic venous thrombosis observed in a patient with acute splenic syndrome
of sickle cell traits on contrast-enhanced thin-slice computed tomography.
Abdom Radiol (NY) 2016;41:1718–1721.
27. Gupta M, Lehl SS, Singh K, et al. Acute splenic infarction in a hiker with
previously unrecognised sickle cell trait. BMJ Case Rep 2013;2013:
bcr2013008931.
28. Ferster K, Eichner ER. Exertional sickling deaths in Army recruits with
sickle cell trait. Mil Med 2012;177:56–59.
29. Murray MJ, Evans P. Sudden exertional death in a soldier with sickle cell
trait. Mil Med 1996;161:303–305.
30. Way A, Ganesan S, McErlain M. Multiple limb compartment syndromes in a
recruit with sickle cell trait. J R Army Med Corps 2011;157:182–183.
31. Weisman IM, Zeballos RJ, Martin TW, et al. Effect of Army basic training in
sickle-cell trait. Arch Intern Med 1988;148:1140–1144.
32. Harris KM, Haas TS, Eichner ER, et al. Sickle cell trait associated with
sudden death in competitive athletes. Am J Cardiol 2012;110:1185–1188.
33. Kark JA, Ward FT. Exercise and hemoglobin S. Semin Hematol 1994;
31:181–225.
34. Nelson DA, Deuster PA, Carter R 3rd, etal. Sicklecell trait, rhabdomyolysis,
and mortality among U.S. Army soldiers. NEnglJMed2016;375:435–442.
35. Davis AM. Sickle-cell trait as a risk factor for sudden-death in physical-
training—reply. NEnglJMed1987;317:781–787.
36. Harmon KG, Drezner JA, Klossner D, et al. Sickle cell trait associated with a
RR of death of 37 times in National Collegiate Athletic Association football
athletes: a database with 2 million athlete-years as the denominator. Br J
Sports Med 2012;46:325–330.
37. O'Connor FG, Bergeron MF, Cantrell J, et al. ACSM and CHAMP summit
on sickle cell trait: mitigating risks for warfighters and athletes. Med Sci
Sports Exerc 2012;44:2045–2056.
38. Mitchell BL. Sickle cell trait and sudden death—bringing it home. JNatl
Med Assoc 2007;99:300–305.
39. Guindo A, Traore K, Diakite S, et al. An evaluation of concur rent G6PD
(A-) def iciency and sickle cell trait in Malian populations of children
with severe or uncomplicated P. falciparum malaria. Am J Hematol
2011;86:795–796.
40. Ouattara AK, Yameogo P, Diarra B, et al. Molecular heterogeneity of
glucose-6-phosphate dehydrogenase deficiency in Burkina Faso: G-6-PD
Betica Selma and Santamaria in people with symptomatic malaria in
Ouagadougou. Mediterr J Hematol Infect Dis 2016;8:e2016029.
41. Sebastiani P, Farrell JJ, Alsultan A, et al. BCL11A enhancer haplotypes
and fetal hemoglobin in sickle cell anemia. Blood Cells Mol Dis 2015;54:
224–230.
42. Sebastiani P, Solovieff N, Hartley SW, et al. Genetic modifiers of the severity
of sickle cell anemia identified through a genome-wide association study.
Am J Hematol 2010;85:29–35.
43. Help and Documentation. Cary, NC: SAS Institute; 2002–2004.
44. Akinbami AO, Campbell AD, Han ZJ, et al. Hereditary persistence of fetal
hemoglobin caused by single nucleotide promoter mutations in sickle cell
trait and Hb SC disease. Hemoglobin 2016;40:64–65.
45. Hariharan P, Sawant M, Gorivale M, et al. Synergistic effect of two βglobin
gene cluster mutations leading to the hereditary persistence of fetal
hemoglobin (HPFH) phenotype. MolBiolRep2017;44:413–417.
46. Al-Allawi NA, Puehringer H, Raheem RA, et al. Genetic modifiers in
β-thalassemia intermedia: a study on 102 Iraqi Arab patients. Genet Test
Mol Biomarkers 2015;19:242–247.
47. Sokal R, Rohlf F. Biometry: The Principles and Practice of Statistics in
Biological Research. New York: WH Freeman; 2012.
48. Bacchetti P, Deeks SG, McCune JM. Breaking free of sample size dogma to
perform innovative translational research. Sci Transl Med 2011;3:87ps24.
Flansburg et al •Fetal Hemoglobin Modulators and SCT in Football Players
294 © 2019 The Southern Medical Association
Copyright © 2019 The Southern Medical Association. Unauthorized reproduction of this article is prohibited.
