University of Southern California Dissertations and Theses

Genetic and dietary determinants of nonalcoholic fatty liver disease in Hispanic children

(USC Thesis Other)

Genetic and dietary determinants of nonalcoholic fatty liver disease in Hispanic children

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content ! "!
Genetic and dietary determinants of nonalcoholic fatty liver disease in Hispanic children

by
Ryan William Walker

A dissertation presented to the faculty of the
University of Southern California Graduate School
In partial fulfillment of the requirements for the degree
Doctor of Philosophy
(Integrative Biology of Disease)

August 2013

! #!

! $!
Table of Contents
Page 4. List of tables
Page 5. List of Figures
Page 6. Acknowledgments
Page 7. Abstract

Page 8. Chapter 1: Backround; Genetic and Carbohydrate Related Factors
Affecting Liver Fat Accumulation
Page 8. Introduction
Page 8. Sugars And Liver Fat Accumulation
Page 9. Genetic Contributors To Nonalcoholic Fatty Liver Disease
Page 11. Nutrigenetic Interactions
Page 11. Biological Mechanisms
Page 13. Conclusion
Page 15. Chapter 1 References

Page 18. Chapter 2: Genetic and Clinical Markers of Elevated Liver Fat Content in
Overweight and Obese Hispanic Children
Page 18. Abstract
Page 19. Introduction
Page 20. Methods!
Page 20. Results!
Page 25. Discussion
Page 39. Chapter 2 References

Page 43. Chapter 3: Fructose content of sugar sweetened beverages and juices
Page 43. Abstract
Page 44. Introduction
Page 44. Methods and Procedures
Page 47. Results
Page 49. Discussion
Page 59. Chapter 3 References

Page 63. Chapter 4: Adipose tissue inflammation and fibrosis may mediate liver
damage and diabetes risk in overweight and obese Italian children
Page 63. Abstract
Page 64. Introduction
Page 66. Methods
Page 67. Results
Page 70. Discussion
Page 81. Chapter 4 References

Page 85. Chapter 5: High rates of fructose malabsorption are associated with
reduced liver fat in obese African Americans
Page 85. Abstract
Page 86. Introduction
Page 86. Methods
Page 88. Results
Page 89. Discussion
Page 97. Chapter 5 References

Page 101. Chapter 6: Summary of findings, future directions and conclusion
Page 112. Chapter 6 References
! %!

List of Tables
Page 14. Table 1-1. Key Background Points
Page 30. Table 2-1. The Clinical Characteristics of the Study Population
Page 31. Table 2-2. Individual Effects of SNPs on Liver Fat Content.
Page 32. Table 2-3. Performance Parameters of Clinical and Genetic Risk Factors on
ELF Predicton.
Page 33. Table 2-4. ALT, BMI Percentile and GRS as Predictors of Liver Fat Content.
Page 34. Table 2-5. Power calculation for six SNPs previously associated with liver fat
content.
Page 53. Table 3-1. Formulas.
Page 54. Table 3-2. Percent of total sugar in excess of labeled total sugar
Page 74. Table 4-1. Subject Characteristics
Page 75. Table 4-2. Adipose and Liver Immunohistological Measures
Page 95. Table 5-1. Clinical Characteristics

! &!

List of Figures
Page 35. Figure 2-1: Distribution of BMI percentile, ALT and AST by liver fat content.
Page 36. Figure 2-2: Combined genetic effects on liver fat content.
Page 37. Figure 2-3: Utility of a GRS to predict Elevated Liver Fat Content.
Page 38. Figure 2-4. Combined genetic effects six SNPs on liver fat content.
Page 55. Figure 3-1. Mean sugar comparison of sodas across three independent methods.
Page 57. Figure 3-2. Fructose concentration and Fructose to Glucose Ratio: Sodas/Sports
Beverages
Page 58. Figure 3-3. Fructose concentration and Fructose to Glucose Ratio: Juices.
Page 76. Figure 4-1. Adipose cell size, CLS presence and collagen ratio as a function of
obesity.
Page 77. Figure 4-2. Liver histology: Steatosis, fibrosis and inflammation.
Page 78. Figure 4-3A. Mean adipose collagen ratio impacts disposition index.
Page 79. Figure 4-3B. Glucose and insulin dynamics are affected by SAT fibrosis.
Page 80. Figure 4-4. Mean adiponectin decreases as a function of increasing steatosis.
Page 96. Figure 5-1a. Mean H
2
area under the curve (AUC) for all breath collection time
points over the 180-minute procedure. !
Page 97. Figure 5-1b. Liver fat percent vs. H
2
production in African Americans and
Hispanics. !
Page 98. Figure 5-1c. Liver fat percent vs. Breath H
2
at 180 minutes in African
Americans and Hispanics.
Page 99. Figure 5-2. Mean Incremental H
2
AUC at 150-180 minutes by ethnicity
Page 101. Figure 6-1. General model for the role of genetic and dietary components in
NAFLD.
Page 104. Figure 6-2. Historic trends in daily fructose consumption.
Page 105. Figure 6-3. A dietary intervention targeting at-risk genotypes in Hipsanic
children.
Page 107. Figure 6-4. Proposed mechanism by which fructose malabsorption may
prevent hepatic fat accumulation.
Page 110. Figure 6-5. Summary of relationships between obesity inflammation, liver
disease and metabolic risk.
!

! '!

Acknowledgments
This body of work is the direct result of expert mentoring and the unselfish sharing of
opportunity provided by both of my mentors while at the University, Michael Goran and
Hooman Allayee. I cannot put into words how invaluable I consider my time as a student
in your labs. The skills and critical thinking that I have been able to develop after four
years of your mentorship have prepared me for a career in research. I cannot thank you
both enough and only hope I can achieve a level of expertise in my future work that is
comparable to what you both afforded me as mentors.

I would like to thank my wife, Kathryn Sullivan, for her unwavering support throughout
this process. Kathryn, thank you for keeping me focused, providing critical feedback and
guidance and for loving me regardless of my successes or failures. You kept me on
track through some difficult stages and for that this degree is as much yours as it is
mine. I love you.

Thank you to my parents, Allyson and James Walker, for instilling in me at a very early
age the desire to ask and obtain answers to questions. My interest in the world of
science began at an early age and I attribute that to the way that you both raised me,
and applaud you for nurturing it as I grew. My success in this endeavor is a reflection of
your success as parents. I love you both.

! (!
Abstract
The purpose of this dissertation is to summarize our recent findings that have examined
dietary, genetic and inflammatory factors that contribute to fat accumulation and
nonalcoholic fatty liver disease (NAFLD) in Hispanic children. In addition, this
dissertation discusses how some of these contributions to liver fat may vary across the
population in terms of ethnic-specific effects and future treatments that may result from
the findings presented.

Recent genome wide studies have identified several polymorphisms that contribute to
increased liver fat accumulation, with some of these genes relating to dietary
carbohydrate and sugar consumption. In particular, a variant of the PNPLA3 gene, which
is highly prevalent in Hispanics, contributes to excessive liver fat beginning at a young
age, especially in the context of high sugar consumption. We show that a genetic risk
score comprised of variants associated with liver fat improve detection of NAFLD when
added to current clinical measures. Additionally, dietary carbohydrate, and especially
fructose, has been shown to contribute to increased liver fat accumulation due to the
lipogenic potential of fructose during liver metabolism. We present data that
demonstrates certain beverages amongst sodas and juices contain higher
concentrations of fructose than previously thought, suggesting population estimates of
fructose consumption in children and adults are underestimated. Lastly, we present
recent finding suggesting that inflammation and fibrosis in the subcutaneous adipose
tissue of children may impact liver disease progression and increase risk for type 2
diabetes.

Dietary sugar contributes to liver fat accumulation, with this being explained by de novo
lipogenesis from fructose in the liver. Certain genetic factors, including PNPLA3, GCKR
and APOC3 contribute to increased liver fat accumulation, with these effects being
manifested at an early age. Hispanics in particular are at elevated risk for liver fat
accumulation due to the higher frequency of genetic variants such as PNPLA3 and
GCKR as well as an interaction between the PNPLA3 and dietary sugar. Additionally,
adipose inflammation caused by the obese state may incur elevated risk for liver disease
and type 2 diabetes. The summation of these factors represents, and potentially
explains, the added risk and high prevalence of NAFLD among the Hispanic pediatric
population.
! )!
Chapter 1: Background; Genetic and Carbohydrate Related Factors Affecting Liver
Fat Accumulation

Introduction
Nonalcoholic fatty liver disease (NAFLD) is characterized by the accumulation of large
droplets of triglycerides within hepatocytes in the absence of chronic alcohol
consumption. NAFLD may lead to nonalcoholic steatohepatitis, cirrhosis, and eventually
hepatocarcinoma [1,2]. The purpose of this chapter is to summarize recent findings that
have examined dietary, genetic and gene–diet interactions that contribute to fat
accumulation in the liver during growth and development, with particular focus on
contributions relating to dietary carbohydrate consumption. In addition, another focus of
this review is to highlight how some of these contributions to liver fat vary across the
population in terms of ethnic-specific effects.

Sugars And Liver Fat Accumulation
Several recent studies support the concept that carbohydrate intake, and more
specifically sugar, is a major culprit in liver fat deposition, due primarily to the high
lipogenic potential of fructose. Maersk et al. [3] randomly assigned 47 overweight
individuals to receive 1 liter per day for 6 months of cola, milk (same calories as the
cola), sugar-free cola or water. There were no changes in total fat mass across groups
but the cola group had significant increases in liver fat (~35%) as well as increased
visceral fat (~25%) and triglycerides (32%). Le et al. [4] conducted a study in 16 male
offspring of type 2 diabetes and eight controls who received either 7 days of an
isocaloric diet or a hypercaloric diet using fructose to increase daily energy by 35% in a
crossover design. The high-fructose diet liver fat increased by 76% in controls and 79%
in offspring type 2 diabetes. Stanhope et al. [5] examined overweight and obese
individuals who consumed glucose or fructose sweetened beverages for 10 weeks at
25% of daily energy requirements under closely controlled conditions. No data on liver
fat are available yet from this study, but the results showed that despite similar weight
gain in the two groups, the fructose group had a significant increase in visceral fat (14
vs. 3% increase in glucose group), and hepatic de-novo lipogenesis (+75 vs. +27%),
both indicating the likelihood of greater liver fat accumulation. Using more invasive liver
biopsies, one study has identified a link between dietary fructose and the severity of liver
damage [6]. Despite this accumulating evidence, not all studies have demonstrated a
! *!
relationship between increased dietary fructose and increased liver fat. For example,
Silbernagel et al. [7] conducted a trial over 4 weeks with 20 healthy non-obese
individuals who were placed on a weight maintaining diet and either 150 g per day of
fructose or glucose (non-crossover design). There were no significant effects of fructose
or glucose on insulin resistance, visceral fat or liver fat, possibly due to the smaller
sample size and shorter feeding period of 4 weeks. Even though there were no changes
in liver fat, this study did find that circulating triglycerides were 44% higher after fructose
with no change after glucose.

Genetic Contributors To Nonalcoholic Fatty Liver Disease
The emergence of genome-wide association studies (GWAS) has led to the identification
of several loci associated with NAFLD and/or hepatic inflammation, and some of these
loci relate to carbohydrate metabolism. Of the three GWAS studies to date that have
been undertaken to specifically identify potential genetic variants that impact hepatic fat
content [8,9,10], two have identified specific loci that purportedly play a role in hepatic
triglyceride content. The first of these GWAS encompassed over 9000 nonsynonymous
sequence variant single nucleotide polymorphisms (SNPs) that were tested in adult
Hispanic (n=383), African–American (n=1032) and white (n=696) participants [8], from
the Dallas Heart Study, for associations with hepatic fat levels measured by
spectroscopy [11]. This study revealed that a SNP (rs738409; C > G) in the patatin-like
phospholipase domain-containing protein 3 (PNPLA3) gene was strongly associated
with both hepatic fat content and inflammation (as determined by elevated serum liver
enzymes) in all three ethnic groups. However, the frequency of the variant allele was
highest in Hispanics (49%) with individuals homozygous for the variant having
approximately two-fold higher hepatic triglyceride content than non-carriers.
The association of PNPLA3 with NAFLD has been replicated in numerous subsequent
studies, thus confirming this gene as an important genetic determinant of hepatic fat
accumulation. Of note, several recent studies have also shown that the effect of this
gene is manifested in children [12,13,14]. One study in over 300 Hispanic children
showed that liver fat in homozygous variant carriers of the rs738409 SNP was almost
2.5 times higher than non-carriers, and that this effect extended to the youngest children
(8 – 10years)[12]. Another more recent GWAS was conducted in over 7100 adult
individuals participating in several population-based cohort studies [10]. A subset of 592
patients, with biopsy-proven non-alcoholic steatohepatitis, were used to validate the 45
! "+!
loci that were associated with hepatic triglyceride content and five SNPs were found to
be associated with NAFLD [computed tomography or histologically defined]; PNPLA3
(rs738409), neurocan (rs2228603), lysophospholipase-like 1 (rs12137855), glucokinase
regulatory protein (GCKR; rs780094) and protein phosphatase 1, regulatory subunit 3B
(PPP1R3B; rs4240624). The aforementioned effect of the PNPLA3 variant on hepatic fat
was replicated in this study with the rs738409 SNP being highly associated with NAFLD
as determined by both CT and histology. Novel associations with the remaining four
SNPs and NAFLD were also reported but, with the exception of GCKR, have yet to be
confirmed in independent studies. For example, GCKR has also been associated with
NAFLD in an adult Chinese population [15] as well as in Hispanic children [16].
Analogous to PNPLA3, the at–risk variant is fairly frequent in Hispanics (36%),
suggesting that GCKR may further contribute to the increased genetic risk for high liver
fat and NAFLD in this population.

More recently, Petersen et al. [17] carried out a candidate gene study and reported that
variants of APOC3 (rs2854117 and rs2854116) were associated with NAFLD in Asian–
Indian men, with findings replicated in a larger validation group of white men. Among
variant allele carriers, 38% were positive for NAFLD, whereas the prevalence of NAFLD
in wild-type homozygotes was 0% [17]. Several studies thereafter have been unable to
replicate these associations with hepatic fat in multi-ethnic adult populations [16,18–21]
and only two studies have been conducted thus far in children [16,20]. One group was
unable to show an association with APOC3 SNPs and the severity of liver damage,
independent of the effects of PNPLA3 [20]; however, the study was conducted only in
Italian participants. The second study consisted of a larger group of 455 children and
also did not observe any associations with liver fat in Hispanics, African–Americans or
whites [16]. Further studies with larger sample sizes will be required to better understand
the potential contributions of these APOC3 polymorphisms to the complex pathological
progression of liver disease [22]. These efforts should also include His- panics, who are
the most at-risk group for developing NAFLD, as the effects of the APOC3 variants may
either be ethnic-specific or stronger in certain populations.

! ""!
Nutrigenetic Interactions
Given the adverse effects that dietary sugar has in promoting accumulation of fat in the
liver, an interesting extension of genetic studies with NAFLD has been the identification
of gene – dietary interactions. For example, a nutrigenetic analysis with the PNPLA3
rs738409 variant in Hispanic children revealed that hepatic fat was positively related to
carbohydrate (r = 0.38, P = 0.02) and total sugar (r = 0.33, P = 0.04) intakes but only in
the homozygous variant group [23]. These findings suggest that Hispanic children with
two copies of the PNPLA3 variant (GG) are particularly susceptible to increased liver fat
in the context of high dietary sugar, whereas such a nutrigenetic effect is not apparent in
CC and CG individuals. Notably, the results of a recent short-term intensive dietary
intervention are consistent with the observed nutrigenetic association with PNPLA3 [24].
In this study, 18 adults with NAFLD (matched for elevated liver fat content) were
preselected on the basis of PNPLA3 genotype and placed on a 6-day hypocaloric, low-
carbohydrate diet. Although hepatic fat content significantly decreased in both GG (n = 8)
and CC (n = 10) individuals, the reduction in liver fat was 2.5-fold greater in GG
homozygotes. This differential effect was observed even though weight loss was
similarly marginal in the two genotype groups (3%). Thus, this latter study demonstrates
that PNPLA3 variant homozygotes respond better to a low-carbohydrate diet with
respect to liver fat reduction, which can be observed in as few as eight individuals.

Biological Mechanisms
One important consideration in genetic studies of NAFLD is an understanding of the
underlying biological mechanisms, particularly for genes whose functional role in hepatic
triglyceride metabolism is not initially evident. In this regard, APOC3 is thought to
potentially influence NAFLD by delaying the metabolism of triglyceride-rich lipoprotein
particles, which increases their uptake by the liver [17]. GCKR inhibits glucokinase
activity, thereby regulating glucose storage/disposal and increasing substrate availability
for de-novo lipogenesis [25]. PPP1R3B, a serine/threonine phosphatase involved in
hepatic glycogen synthesis could thus modulate risk of NAFLD through a similar
mechanism as GCKR. Thus, it is possible that the association of these genes with
NAFLD is mediated through their effects on adiposity, lipid metabolism and/or glucose
homeostasis.

! "#!
By comparison, less is known about the function of PNPLA3 and, as this gene is the
most strongly associated genetic risk factor for NAFLD, we will focus most of our
discussion on functional studies of this protein. PNPLA3, or adiponutrin, encodes a
protein in hepatocytes that has been reported to have lipase-like activity and promote
hepatic triglyceride hydrolysis [26,27]. In addition, PNPLA3 has been shown to have
specific hydrolase activity against triglycerides but not other lipid substrates, such as
phospholipids, cholesteryl ester and retinyl esters [27]. On the basis of these studies,
this putative enzyme does not appear to promote de-novo hepatic triglyceride synthesis,
although it is possible that PNPLA3 has other functional properties that have yet to be
determined [27]. Animal studies have also shown that expression of hepatic PNPLA3
mRNA levels is low during fasting and increases approximately 90-fold in response to
carbohydrate feeding [28]. This effect occurs as a secondary effect of insulin-mediated
up-regulation of sterol regulatory element binding protein 1 (SREBP-1) and liver X
receptor, which are important transcription factors responsible for fat metabolism in the
liver. Although this transcriptional effect would presumably not differ across genotypes,
recent studies have shown that the G variant substitutes a methionine at position 148 in
the protein and abolishes PNPLA3 hydrolase activity [26], which would presumably
increase risk of NAFLD by inhibiting hepatic fat mobilization. This notion would also be
consistent with the nutrigenetic interactions described above because homozygous GG
individuals would consequently be more susceptible to the effects of dietary sugar as
transcriptional up-regulation of PNPLA3 would still result in a protein with severely
reduced function.

Other animal studies, however, do not support this biological model. For example, two
groups have independently knocked out PNPLA3 in mice using gene targeting and do
not observe any effects on hepatic triglyceride accumulation, glucose homeostasis, lipid
levels or body composition [29,30]. These surprising results were observed even after
the mice were placed on either a genetically induced obesity background (i.e. leptin
deficiency) or on a variety of diets, including those with high sucrose content. Two other
interesting observations from these studies were that expression of PNPLA3 is regulated
in a nutrient-specific manner, consistent with previous observations, and that PNPLA5 is
up- regulated several fold in adipose tissue of PNPLA3-deficient mice. The latter findings
suggest that the increased expression of PNPLA5 in adipose, which is a paralog of
PNPLA3, may serve as a compensatory mechanism for the lack of PNPLA3, although it
! "$!
was also observed in both studies that PNPLA5 expression was either very low or not
detectable in liver. By comparison, another study used the adenovirus system to perturb
PNPLA3 expression [31]. Whereas overexpression of PNPLA3 in mouse primary
hepatocytes increased intracellular triglycerides, knockdown of PNPLA3 suppressed
SREBP-1-stimulated lipid accumulation. Thus, despite the strong and compelling genetic
evidence in humans that PNPLA3 influences the development of NAFLD, these
important animal studies suggest that species-specific differences may exist with respect
to how this putative triglyceride hydrolase mediates liver fat accumulation and that
additional studies will be needed to understand the precise role that PNPLA3 (and the
rs738409 variant) plays in hepatic lipid metabolism.

Conclusion
New studies point to the contributing role of dietary sugar in the accumulation of fat in
the liver, with this being explained by the potential of fructose to serve as a substrate for
de-novo lipogenesis in the liver. In addition, genetic variants associated with liver fat
accumulation are being identified with mechanism of action in some cases being closely
related to carbohydrate metabolism. Hispanics in particular are at elevated risk for liver
fat accumulation due to the higher frequency of variants such as PNPLA3 and GCKR.
Given the high frequency of the PNPLA3 rs738409 variant and the increasing
prevalence of NAFLD and obesity in Hispanic children and adolescents, high levels of
added sugar intake likely play a vital role in the pathogenesis of the disease in
Hispanics. This gene–diet interaction is a specific example with translational
implications, as this finding suggests that specific interventions based on reducing
dietary sugar intake in genetically predisposed individuals may lead to more effective
therapeutic outcomes for fatty liver. Although previous studies have shown that weight
loss alone can lead to significant and rapid reduction in liver fat in children and adults,
this strategy may not be sustainable and may not be effective in certain subgroups
depending on genotype. Therefore, other adjunct strategies are required that may need
to target individual genotype, to prevent and/or treat the accumulation of fat in the liver
for overall improvement of metabolic health.

! "%!
Table 1-1. Key background points.

! "&!
Chap 1: References
1. Starley BQ, Calcagno CJ, Harrison SA. Nonalcoholic fatty liver disease and
hepatocellular carcinoma: a weighty connection. Hepatology 2010; 51:1820 – 1832.
2. PreissD,SattarN.Nonalcoholic fatty liver disease: an overview of prevalence,
diagnosis, pathogenesis and treatment considerations. Clin Sci (Lond) 2008; 115:141 –
150.
3. Maersk M, Belza A, Stodkilde-Jorgensen H, et al. Sucrose-sweetened beverages
increase fat storage in the liver, muscle, and visceral fat depot: a 6-mo randomized
intervention study. Am J Clin Nutr 2012; 95:283 – 289.
4. Le KA, Ith M, Kreis R, et al. Fructose overconsumption causes dyslipidemia and
ectopic lipid deposition in healthy subjects with and without a family history of type 2
diabetes. Am J Clin Nutr 2009; 89:1760–1765.
5. Stanhope KL, Schwarz JM, Keim NL, et al. Consuming fructose-sweetened, not
glucose-sweetened, beverages increases visceral adiposity and lipids and decreases
insulin sensitivity in overweight/obese humans. J Clin Invest 2009; 119:1322 – 1334.
6. Abdelmalek MF, Suzuki A, Guy C, et al. Increased fructose consumption is associated
with fibrosis severity in patients with nonalcoholic fatty liver disease. Hepatology
2010;51:1961 – 1971.
7. Silbernagel G, Machann J, Unmuth S, et al. Effects of 4-week very-high-
fructose/glucose diets on insulin sensitivity, visceral fat and intrahepatic lipids:
anexploratory trial. Br J Nutr 2011;106:79 – 86.
8. Romeo S, Kozlitina J, Xing C, et al. Genetic variation in PNPLA3 confers susceptibility
to nonalcoholic fatty liver disease. Nat Genet 2008; 40:1461–1465.
9. Chalasani N, Guo X, Loomba R, et al. Genome-wide association study & identifies
variants associated with histologic features of nonalcoholic fatty liver disease.
Gastroenterology 2010; 139:1567–1576. Subsequent GWAS that confirmed the
association of PNPLA3 with NAFLD.
10. Speliotes EK, Yerges-Armstrong LM, Wu J, et al. Genome-wide association analysis
identifies variants associated with nonalcoholic fatty liver disease that have distinct
11. Victor RG, Haley RW, Willett DL, et al. The Dallas Heart Study: a population-
based probability sample for the multidisciplinary study of ethnic differences in
cardiovascular health. Am J Cardiol 2004; 93:1473–1480.
12. Goran MI, Walker R, Le K-A, et al. Effects of PNPLA3 on liver fat and metabolic
profile in Hispanic children and adolescents. Diabetes 2010; 59:3127–3130.
! "'!
13. Romeo S, Sentinelli F, Cambuli VM, et al. The 148M allele of the PNPLA3
gene is associated with indices of liver damage early in life. J Hepatol 2010;
53:335 – 338.
14. Valenti L, Alisi A, Galmozzi E, et al. I148M patatin-like phospholipase domain-
containing 3 gene variant and severity of pediatric nonalcoholic fatty liver disease.
Hepatology 2010; 52:1274 – 1280.
15. Yang Z, Wen J, Tao X, et al. Genetic variation in the GCKR gene is associated
with nonalcoholic fatty liver disease in Chinese people. Mol Biol Rep 2011;
38:1145 – 1150.
16. Santoro N, Zhang CK, Zhao H, et al. Variant in the glucokinase regulatory
protein (GCKR) gene is associated with fatty liver in obese children and
adolescents. Hepatology 2012; 55:781 – 789.
17. Petersen KF, Dufour S, Hariri A, et al. Apolipoprotein C3 gene variants in
nonalcoholic fatty liver disease. N Engl J Med 2010;
362:1082 – 1089.
18. Kozlitina J, Boerwinkle E, Cohen JC, Hobbs HH. Dissociation between
APOC3 variants, hepatic triglyceride content and insulin resistance.
Hepatology 2011; 53:467 – 474.
19. Sentinelli F, Romeo S, Maglio C, et al. Lack of effect of apolipoprotein C3
polymorphisms on indices of liver steatosis, lipid profile and insulin resistance in obese
Southern Europeans. Lipids Health Dis 2011; 10:93.
20. Valenti L, Nobili V, Al-Serri A, et al. The APOC3 T-455C and C-482T promoter
region polymorphisms are not associated with the severity of liver damage
independently of PNPLA3 I148M genotype in patients with nonalcoholic fatty liver. J
Hepatol 2011; 55:1409 – 1414.
21. Hyysalo J, Stojkovic I, Kotronen A, et al. Genetic variation in PNPLA3 but not
APOC3 influences liver fat in NAFLD. J Gastroenterol Hepatol 2011. [Epub ahead of
print]. doi: 10.1111/j.1440-1746.2011.07046.x
22. Romero-Go!mez M. APOC3 polymorphisms and nonalcoholic fatty liver disease:
resolving some doubts and raising others. J Hepatol 2011; 55:1184 – 1186.
23. Davis JN, Le KA, Walker RW, et al. Increased hepatic fat in overweight Hispanic
youth influenced by interaction between genetic variation in PNPLA3 and high dietary
carbohydrate and sugar consumption. Am J Clin Nutr 2010; 92:1522 – 1527.
! "(!
24.Sevastianova K, Kotronen A, Gastaldelli A, et al. Genetic variation in PNPLA3
(adiponutrin) confers sensitivity to weight loss-induced decrease in liver fat in humans.
Am J Clin Nutr 2011; 94:104–111.
25. Beer NL, Tribble ND, McCulloch LJ, et al. The P446L variant in GCKR
associated with fasting plasma glucose and triglyceride levels exerts its effect through
increased glucokinase activity in liver. Hum Mol Genet 2009; 18:4081 – 4088.
26. He S, McPhaul C, Li JZ, et al. A sequence variation (I148 M) in PNPLA3 associated
with nonalcoholic fatty liver disease disrupts triglyceride hydrolysis. J Biol Chem 2010;
285:6706 – 6715.
27. Huang Y, Cohen JC, Hobbs HH. Expression and characterization of a PNPLA3
protein isoform (I148 M) associated with nonalcoholic fatty liver disease. J Biol Chem
2011; 286:37085 – 37093.
28. Huang Y, He S, Li JZ, et al. A feed-forward loop amplifies nutritional regulation
ofPNPLA3. Proc Natl Acad Sci U S A 2010; 107:7892 –
7897.
29. Chen W, Chang B, Li L, Chan L. Patatin-like phospholipase domain-containing
3/adiponutrin deficiency in mice is not associated with fatty liver disease. Hepatology
2010; 52:1134–1142.
30. Basantani MK, Sitnick MT, Cai L, et al. Pnpla3/Adiponutrin deficiency in mice does
not contribute to fatty liver disease or metabolic syndrome. J Lipid Res 2011; 52:318–
329.
31. Qiao A, Liang J, Ke Y, et al. Mouse patatin-like phospholipase domain-
containing 3 influences systemic lipid and glucose homeostasis. Hepatology 2011;
54:509–521.

! ")!
Chapter 2: Genetic and Clinical Markers of Elevated Liver Fat Content in
Overweight and Obese Hispanic Children

Abstract
Objective: Genetic variation in six genes has been associated with elevated liver fat and
nonalcoholic fatty liver disease in adults. We sought to determine the influence of these
genes on liver fat and whether a genetic risk score (GRS) would improve upon the ability
of common clinical risk factors to predict elevated liver fat content (ELF) in Hispanic
children.

Design and Methods: 223 obese Hispanic children were genotyped for six SNPs. MRI
was used to measure liver fat. A GRS was tested for association with ELF using
multivariate linear regression. Predictors were assessed via ROC curves and pair-wise
analysis was used to determine significance alone and combined with clinical markers.

Results: Only variants in PNPLA3 and APOC3 genes were associated with liver fat
(p<0.001, p=0.01, respectively). Subjects with a GRS=4 had ~3-fold higher liver fat
content than subjects with GRS of 0 (15.1±12.7% vs. 5.1±3.7%, p=0.03). While the
addition of the GRS to a model containing BMI and liver enzymes increased ROC AUC
from 0.83 to 0.85 [95% CI, 0.79-0.89], (p=0.01), it does not improve detection of ELF
from a clinical perspective.

Conclusions: Only PNPLA3 and APOC3 were related to ELF and a GRS comprised of
these susceptibility alleles did not add to the discriminatory power of traditional
biomarkers for clinical assessment of liver fat.

! "*!
Introduction
As national rates of overweight and obesity continue to increase, so does the prevalence
of nonalcoholic fatty liver disease (NAFLD). This trajectory, in terms of both increasing
obesity and NAFLD, is also present in the pediatric population (1, 2). NAFLD, a condition
characterized by an elevated liver fat content (ELF) of greater than 5.5 percent, which
can be determined by imaging and/or histology (3), represents the proximal end of a
broad spectrum of liver disease including nonalcoholic steatohepatitis (NASH), cirrhosis,
and ultimately hepatocellular carcinoma. NAFLD represents the leading cause of chronic
liver disease amongst all children in the United States (4, 5) and Hispanics are at an
elevated risk of developing NAFLD, due in part to high obesity rates and potentially
genetic factors.

Early clinical detection of ELF can be difficult, as the condition is usually asymptomatic.
Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST), as well
as obesity status are currently used in the clinical evaluation of ELF and NAFLD (6, 7).
MRI and magnetic resonance spectroscopy (MRS) can quantitatively assess, with a high
degree of accuracy, hepatic fat content but are expensive clinical tests and not used
routinely. By comparison, ultrasound is commonly used to assess liver fat content for
clinical purposes and has limited utility due to the qualitative nature of the measure.
While these three imaging techniques are commonly used, the “gold standard” remains
liver biopsy, which can present an added risk for complications, particularly in the
pediatric population. These current assessment methods represent a significant added
burden and expense when compared to anthropometric and plasma assay measures in
combination with genetic testing.

Recent genome-wide association studies (GWAS) and candidate gene approaches in
adults have implicated six genes in the development of NAFLD and/or NASH, including
patatin-like phospholipase 3 (PNPLA3), glycogen binding subunit of protein
phosphatase1 (PPP1R3B), neurocan (NCAN), glucokinase regulatory protein (GCKR),
lysophospholipase-like 1 (LYPLAL1), and apolipoprotein C3 (APOC3) (8-10). Previously,
we validated the effect of PNPLA3 on hepatic triglyceride levels in 188 Hispanic children
and demonstrated that this effect manifests as early as 8 years of age. However, the
independent and combined effects of the five other variants on hepatic fat content in this
pediatric population are not known. Therefore we sought to evaluate the association of
! #+!
these six genetic variants on hepatic fat content in obese Hispanic children and
determine whether cumulative genetic burden improves the prediction of ELF beyond
traditional clinical risk factors.

Methods
Participants
This study was a cross-sectional analysis of 223 overweight and obese (BMI of 25-45
kg/m
2
), Hispanic children (41% male) between the ages of 8-17 years old who were
recruited from several Districts in Los Angeles County. Participants were defined as
Hispanic if they reported both parents and all four grandparents as Hispanic. All
participants had medical and family history screening to ensure eligibility criteria were
met. Patients were not eligible for the study if the following conditions were indicated: i)
met any diagnostic criteria for diabetes; ii) the use of medications or supplements or the
past or present diagnosis of other syndromes or diseases known to influence liver
function, insulin action or lipid levels; iii) previous diagnosis of any major illness since
birth; or iv) smoking (currently smoked or had smoked greater than 100 cigarettes in
their lifetime) or drinking alcohol on a regular basis (in excess of 2 drinks per week as
determined by questionnaire. Participants were not eligible for the study if they had
current or past involvement with any weight loss/exercise/sports program in the six
months prior to participation. Informed written child assent and parental consent were
obtained from all patients. The Institutional Review Board of the USC Keck School of
Medicine approved the study.

Anthropometry and Fasting Glucose
Height and weight were measured in triplicate for all patients, and BMI was expressed
as kg body weight per m
2
height. Fasting glucose was screened at the first outpatient
visit, as previously described (11). If fasting blood glucose measured "126 mg/dl, an
additional sample was immediately drawn to confirm the value and an additional visit
was scheduled to confirm the diagnosis of diabetes. Diagnosis of diabetes resulted in
exclusion from the study and referral to an endocrinologist.

! #"!
Body Composition and Liver Fat
Whole body fat and soft lean tissue mass was estimated by dual energy x-ray
absorptiometry (DEXA) using a Hologic QDR 4500W (Hologic, Bedford, MA). MRI was
carried out on a General Electric 1.5-Tesla magnet (GE Healthcare, Waukesha, WI) to
assess subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT) and hepatic
fat fraction using a modification of the Dixon 3-point technique as previously described
(12). ELF was defined as >5.5% hepatic fat fraction as determined by MRI, a value
commonly attributed to likely NAFLD (12) and used in previous genetic studies with
adults.

Alanine aminotransferase (ALT) and aspartate aminotransferase (AST)
measurements
ALT and AST levels were measured in plasma samples from 208 participants via the
kinetic rate method using a Synchron CX analyzer (Beckman Coulter Inc. Fullerton, CA)
(13). All samples were prepared in accordance with manufacturer specifications and
analyzed in duplicate. The analytical range for both ALT and AST was 5-400 IU/L. ALT
levels were considered abnormal when they exceeded recently established population-
based upper limits of normal (ULN) values (14). An ALT normal reference range of 5-
23.5 U/L was utilized in this study, with 23.5 representing the mean of the following ULN
cutoff values established from NHANES data in boys (>22 U/L) and girls (>25 U/L) (14).

Molecular Genetic Analyses
Genomic DNA was extracted from whole blood. SNPs were selected based upon
previously identified associations with NAFLD and/or hepatic steatosis from GWAS (9)
and other association studies. The following 8 SNPs were genotyped in the study
population: PNPLA3 (rs738409), APOC3 (rs2854116 and rs2854117), GCKR (rs780094
and rs1260326), NCAN (rs2228603), LYPLAL1 (rs12137855) and PPP1R3B
(rs4240624). Initial genotyping for the two variants in APOC3 (rs2854116 and
rs2854117) and GCKR (rs780094 and rs1260326) showed these SNPs to be in high
linkage disequilibrium (r
2
>0.8, data not shown). Therefore, subsequent analyses were
only conducted with rs2854117 for APOC3 and rs780094 for GCKR. Genotyping was
performed with the TaqMan Allelic Discrimination System. Each 96-well DNA plate
contained the same four control DNA samples, two from International HAPMAP Project
! ##!
and two randomly chosen DNAs from our Hispanic study cohort. Replicate quality
control samples yielded 99% concordance, and the overall call rate was 95%.

Calculations and Statistical Analyses
All variants were first tested for Hardy-Weinberg equilibrium (HWE) using a #
2
test prior
to analysis. General linear models were used to compare mean values of quantitative
traits across groups, with adjustment for covariates. Liver fat values were natural log-
transformed for analyses. A genetic risk score (GRS) was constructed for each
participant based upon the sum of the PNPLA3 and APOC3 risk alleles that individually
showed a significant association with elevated liver fat fraction. A weighted GRS
(wGRS) was also calculated by multiplying the effect estimate (beta) on liver fat,
obtained from separate linear regression analyses of each included SNP in the present
study, adjusting for age and gender, by the number of risk alleles for that corresponding
variant (0,1, or 2), and summing these values. GRS groups were defined as 1 (0-1 risk
alleles), 2 (2-3 risk alleles) and 3 (4 risk alleles). Analysis of variance (ANOVA),
controlling for age and gender, was used to determine between group effects with post
hoc Bonferroni adjustment for multiple tests. Logistic regression was used to determine
the impact of a composite score comprised of BMI percent and ALT (model A) or BMI
percent, ALT and GRS (model B) on the prediction of ELF, controlling for age and
gender. Resultant classification tables were used to calculate model performance.
Receiver Operating Characteristic (ROC) curves were constructed for the models that
included BMI and ALT/AST as conventional predictors with and without the GRS. The
state variable for the test was defined by ELF status (liver fat " 5.5%). The resultant area
under the curve (AUC) for each ROC was obtained and pair wise AUC comparisons
were conducted. Data analysis was carried out using SAS software (version 9.2 of the
SAS System, Cary, NC. USA) and SPSS software (version 18, SPSS Inc., Chicago, IL.
USA).

Results
Clinical Characteristics of the Study Population
The clinical characteristics of the participants are shown in Table 1. Subjects with ELF,
as defined by greater than 5.5% liver fat, were more likely to be male and had nearly
four-fold higher liver fat content compared to normal liver fat participants (15.2 ±9.4 vs.
! #$!
3.7±1.1, p=5.4x10
-32
). Similarly, measures of adiposity were elevated in ELF subjects,
most notably BMI percentile (97.9±3.6 vs. 92.3±10.8, p=1.5x10
-6
) (Table 2-1). The
association between BMI percentile and ELF is further illustrated in Figure 1A where
90% of subjects with liver fat over 5.5% met the criteria for obesity, as defined by BMI
"95
th
percentile, compared to 64% of the normal liver fat content participants. Subjects
with ELF also had significantly higher serum ALT and AST levels than normal liver fat
content subjects (ALT: 20.7±12.3 vs. 10.8±5.4 IU/L, p=2.2x10
-13
; AST: 24.8±10.3 vs.
17.9±4.4 IU/L, p=4.4x10
-10
) (Table 2-1). However, 69% and 77% of ELF subjects had
ALT and AST levels, respectively, that were within the accepted normal reference
ranges for these analytes (Figure 2-1B and C). When standard ALT cutoffs (5-35 U/L) for
ULN were utilized, 84% of ELF subjects fell within the normal reference range. Taken
together, these data indicate that while BMI percentile-defined obesity accurately reflects
the presence of ELF, a cutoff using the clinical definition of elevated liver enzymes does
not provide similar discrimination.

Genetic Effects on Liver Fat Content
To explore the genetic determinants of liver fat content in this pediatric population, we
genotyped six previously reported variants in PNPLA3, PPP1R3B, NCAN, GCKR,
LYPLAL1, and APOC3. All SNPs were in HWE and, with the exception of PNPLA3,
LYPLAL1, and PPP1R3B, the effect allele frequencies (EAF) of the SNPs in APOC3,
GCKR, and NCAN in this Hispanic cohort were similar to those previously reported in
Caucasians (Table 2-2).

Of the variants tested, those in PNPLA3 and APOC3 demonstrated significant dose-
dependent associations with liver fat content even after adjustment for age, sex, and
VAT (Table 2). For example, each copy of the risk allele for PNPLA3 rs738409 and
APOC3 rs2854117 increased liver fat by ~40-50% (p
trend
=4.2x10
-6
) and 20-30%
(p
trend
=0.01), respectively. In addition, the association of APOC3 with liver fat remained
significant (p=0.05) after conditioning on the effect of PNPLA3, but there was no was no
evidence for an interaction between the two genes (p
interaction
=0.40). However, it should
be noted that p-value we obtained for the association of APOC3 with liver fat is slightly
above the threshold for significance based on a Bonferroni correction for multiple tests
(0.05/6=0.008).
! #%!

We next evaluated the combined effects of the PNPLA3 and APOC3 SNPs on liver fat
content. For each subject, an unweighted GRS was calculated for rs738409 and
rs2854117. As shown in Figure 2, the GRS was normally distributed in this study
population and there was a significant stepwise increase in liver fat content as a function
of the number of risk alleles carried (p
trend
=3.12x10
-5
). Participants with 4 risk alleles had
3-fold higher liver fat than those with a GRS of 0 (15.1±12.7% vs. 5.1±3.7%, p=0.03).
These associations remained significant when controlling for age, sex, and VAT. An
analysis using the wGRS yielded similar results (p
trend
=1.3x10
-4
; data not shown). We
also constructed a comprehensive GRS with all 6 SNPs. In this analysis, the maximum
number of risk alleles carried by any one individual did not exceed 9 in our population,
Using tertiles of risk alleles (T1=0-3, n=22; T2=4-6, n=143; and T3=7-9, n=44), a clear,
stepwise increase in liver fat was observed by GRS tertile (Figure 2-4), but the overall
association did not reach significance (p
trend
=0.08).

Discriminatory Analysis
We next determined whether cumulative genetic burden could improve upon the
discriminatory ability of common clinical risk factors to detect ELF. Based on our
observed clinical associations described above, and high correlation between ALT and
AST levels (r=0.72, p<0.001), BMI percentile and ALT were chosen as representative
conventional risk factors in these analyses. The model including GRS had higher
discriminatory ability than BMI and ALT alone, (#
2
=80.6, p<0.001 with df=2 vs. #
2
=72.7,
p<0.001 with df=1, respectively) and overall prediction success was marginally, but
significantly, improved in the full model (0.76 vs. 0.75; p<0.01).

We next carried out ROC curves for the predictors independently and in combination
(Figure 2-3). Compared to BMI percentile alone, the AUC for the model with BMI + ALT
was significantly higher (0.77, 95% CI, 0.70-0.82 vs. 0.84, 95% CI, 0.78-0.89; p=0.03),
indicating that ALT levels significantly improve discrimination of ELF. A model that
included BMI percentile and the GRS also significantly improved discrimination (0.80
95% CI 0.73-0.86) compared to BMI percentile alone, although not to the same extent as
the addition of ALT. PNPLA3 alone demonstrated a similar ability to discriminate ELF
(0.64 95% CI, 0.56-0.72) compared to the GRS, however addition of APOC3 to the
model did not improve AUC values over PNPLA3 alone. Finally, inclusion of all three risk
! #&!
factors (BMI percentile, ALT, and GRS) slightly increased the discriminatory power to
detect ELF to 0.85 (95% CI, 0.79-0.89; p=0.01). The performance of the wGRS for ELF
discriminatory ability was nearly identical to that of the GRS (data not shown). As shown
in Table 3, the addition of GRS to BMI percentile and ALT improved sensitivity and
specificity parameters. Taken together, these data indicate that the addition of GRS
marginally, but significantly, improves overall ELF discrimination (15).

To test the ability of ALT, BMI percentile and GRS to predict liver fat on a continuous
scale, we conducted linear regression analyses. All three variables were significant,
independent predictors of liver fat percent, with ALT explaining the most variance in liver
fat (39%). A per-unit increase in ALT, BMI and GRS corresponded to a significant
increase in liver fat of 0.39, 0.81 and 2.1%, respectively. In multiple regression
analyses, ALT, BMI percentile and GRS were tested as predictors of liver fat percent in a
stepwise model of regression. The full prediction model explained the most variance in
liver fat percent (48%), with an overall model significance of p=0.001(Table 2-4).

Discussion
In the present study, we evaluated the association of commonly used clinical measures
with ELF and examined the contribution of genetic variants to hepatic lipid content in a
Hispanic pediatric population. We further assessed whether the combination of BMI,
plasma liver enzyme levels, and cumulative genetic burden could serve as potential
clinical tool for re-classification of ELF.

Our data show that almost 90% of children with MRI-defined ELF fell above the 95
th

percentile for BMI, illustrating the strong effect of obesity on hepatic fat content.
However, 64% of the subjects with normal liver fat content were also above the 95
th

percentile for BMI, suggesting that this measure alone is not sufficient for indicating ELF.
Similarly, ALT levels correlated significantly with liver fat, but 69% of ELF subjects fell
within the recently established pediatric clinical reference range considered to be normal
(5-23.5 IU/L). This high false negative rate could be problematic for diagnosing NAFLD
in larger populations, as ALT is a commonly used primary indicator of potential NAFLD
in children (16). It is established that the upper limits of normal levels for ALT vary
between populations (17, 18) and abnormal ALT levels have been shown to occur in
about 36% of obese Hispanics (19). In this regard, Ruhl et al. suggest that ALT upper
! #'!
limit ranges should be both ethnic specific and adjusted to reflect the growing prevalence
of NAFLD in the adult population (20). Our data support this recommendation. Reducing
the current normal reference range for ALT could also potentially decrease false
negative diagnoses and improve the sensitivity of ALT as a predictor of ELF in obese
Hispanic children.

Of the genes tested, only PNPLA3 and APOC3 demonstrated significant associations
with liver fat content. Since being identified (8), PNPLA3 has been reproducibly
associated with NAFLD and related phenotypes in numerous studies with both adults
and children (5, 21-25), including one by our group where the effect of PNPLA3 was
observed in Hispanic children as young as 8 years of age (5). Although the underlying
molecular mechanism for how PNPLA3 promotes triglyceride accumulation in the liver is
not entirely known, it is clear that PNPLA3 rs738409 represents the strongest validated
genetic risk factor for NAFLD identified thus far. By comparison, we did not observe any
evidence for association of GCKR, NCAN, PPP1R3B and LYPLAL1 with ELF in this
study sample. Given that these genes were identified through a GWAS of over 7000
Caucasian adults (9), it is likely that our study was underpowered to detect their smaller
effects on liver fat. To formally evaluate this possibility, we used effect sizes recently
reported in Hispanic-American adults for 5 SNPs that were associated with liver fat
content, as measured by computed tomography (26), to conduct a post hoc power
analysis (Quanto V1.2.4). Assuming 80% power, an $-error of 0.05, and the allele
frequencies observed in our Hispanic study population, we had sufficient power to detect
a significant association with PNPLA3 rs738409, which is not surprising given the robust
effect of this variant on liver fat observed in numerous studies. However, our study
sample of 223 subjects did not provide sufficient power to detect associations with the
GCKR, LYPLAL1, NCAN, and PPP1R3B variants (Supplemental Table 1). Interestingly,
GCKR was associated with liver fat fraction in Chinese adults (27) as well as in a recent
study of Caucasian, African American, and Hispanic children, in which an effect was
observed in all three ethnicities (28). Thus, it remains to be determined whether the
effect of this gene is stronger in adults than in children.

Our study also demonstrated an association of APOC3 with liver fat content. APOC3 is
transported on circulating triglyceride-rich lipoproteins and inhibits both lipoprotein lipase
and hepatic lipase, thus delaying the catabolism of triglyceride-rich particles (29, 30).
! #(!
Interestingly, over-expression of APOC3 in mice leads to diet-induced ELF and hepatic
insulin resistance (31). Petersen et al. identified two tightly linked promoter
polymorphisms in APOC3 that were associated with NAFLD in a relatively small
candidate gene study of Asian-Indian men (10). Of note, the association of APOC3 with
liver fat content in our Hispanic pediatric population was observed even after controlling
for the strong effect of PNPLA3. However, other studies in different ethnicities have
failed to detect an association with APOC3 (27, 32-35), including two in children (27, 35).
Such discrepancies could be due to ethnic and age-specific genetic effects (36),
differences in the study populations, and/or the methods used to quantitate liver fat. For
example, to our knowledge there are no published studies reporting the effect of APOC3
on quantitative measures of liver fat content, thus precluding a more direct comparison
with our results. Additional studies, particularly in larger samples sizes and multiple
ethnicities, will be needed in order to clarify the genetic contribution of APOC3 to hepatic
lipid content.

Using a GRS comprised of PNPLA3 and APOC3 susceptibility alleles, we showed that
individuals carrying all 4 risk alleles have almost 3-fold higher hepatic fat content than
non-variant carriers, illustrating the joint effects of PNPLA3 and APOC3 on liver fat
content. Together, the additive effects of PNPLA3 rs738409 and APOC3 rs2854117
account for 12% of the variance in liver fat in our study population, although we did not
obtain evidence for a genetic interaction between these variants. However, the GRS
only resulted in a modest improvement in ELF discrimination beyond BMI percentile and
ALT levels. A model containing ALT, BMI percentile and the GRS was able to explain
47% of the variability in liver fat, whereas BMI percentile and ALT alone explained 41%.
Other studies have used similar approaches to predict NAFLD by using biomarkers and
anthropometric measures, which have been extended to the prediction of NASH and
fibrosis (37, 38).

Hispanic children are vulnerable to the current obesogenic environment and poised to
benefit the most from early ELF detection. However, clinically useful predictors of ELF or
NAFLD, other than liver enzymes and BMI, are lacking in this population. One study
showed that ALT and BMI-z score were both independent predictors of NAFLD and
together they accounted for most of NAFLD prediction in obese Italian children (39).
However this study also noted that ALT alone should not be used as a surrogate marker
! #)!
of NAFLD in obese children (39), supporting existing literature in adults (40) as well as
our current findings. Maffeis et al. were able to develop a highly discriminatory NAFLD
predictive model in obese Italian children using waist-to-height ratio, ALT, adiponectin,
and insulin resistance (41). While encouraging, this would require administration of oral
glucose tolerance tests and additional biomarker assays.

Our approach with genetic factors, which in previous studies have been shown to
explain up to 25% of the variability in hepatic fat content (28), was statistically significant
but did not result in a substantial improvement in clinical utility over two common
predictors of ELF. Additionally, PNPLA3 appears to be driving the utility of the GRS in
this population, which is not surprising given such a strong association of the gene
variant with both NAFLD and NASH in other studies (5, 21-25). Thus, it is not clear
whether such an incremental increase in ELF discrimination merits the clinical
implementation of a PNPLA3 and APOC3 GRS in the detection of ELF at the present
time. Furthermore, the increased discrimination we observe is specific to Hispanic
children and it remains to be determined whether a GRS similarly improves risk
classification in adults and/or other ethnicities. We did not detect associations of liver fat
content with the four other validated NAFLD variants in PPP1R3B, NCAN, GCKR, and
LYPLAL1, presumably due to our relatively small sample size. Thus, it is possible that a
more comprehensive GRS could still provide additional prognostic value for the early
detection of ELF, which could be addressed in the future through prospective study
designs with larger numbers of subjects.

While potentially relevant from a clinical perspective, our study has limitations, including
defining ELF and/or NAFLD as liver fat content greater than 5.5%. This cut-point is
generally accepted in the literature but is an arbitrary level based on previously
published work. Although the use of MRI to detect percent liver fat fraction correlates
highly with values derived through histology, liver biopsy remains the gold standard in
steatosis quantification. However, we were not able to assess the association of a GRS
with the level of histological damage (i.e. NASH) in this study, which is an important
consideration since steatosis often does not progress to fibrotic conditions. Therefore,
further studies in populations where biopsy samples are available will be necessary to
assess the utility of a GRS in predicting the degree of liver inflammation and/or fibrosis in
addition to liver fat percent. In this regard, the development of an accurate and
! #*!
inexpensive GRS test for predicting abnormal liver fat content may still represent an
alternative to costly and invasive biopsies or imaging techniques.

In conclusion, we have shown that variants of PNPLA3 and APOC3 independently and
in an additive manner contribute to elevated hepatic triglyceride content in Hispanic
children. Although BMI percentile and ALT were useful predictors of ELF in this
population, they lack sensitivity when used alone. Moreover, risk prediction models with
BMI percentile that incorporated a GRS with PNPLA3 and APOC3, either alone or in
combination with ALT, marginally improved upon the ability to discriminate ELF however
did not improve detection of ELF from a clinical perspective. Larger study populations
may allow a more comprehensive evaluation of the clinical utility of a GRS, which could
lead to early detection and treatment of ELF or NAFLD in at-risk subjects.

! $+!
Table 2-1. The Clinical Characteristics of the Study Population

Trait
All Participants
(n=223)
Normal Liver fat
(n=126)
ELF
(n=97)
p-value
Age (year) 13.5 ± 2.9 13.5 ± 3.1 13.5 ± 2.8 NS
Male/Female (n) 93/130 43/83 50/47 0.005
Height (cm) 157.3 ± 16.6 157.4 ± 18.2 157.2 ± 11.6 NS
Weight (kg) 77.9 ± 28.2 72.1 ± 27.4 85.4 ± 27.6 <0.001
BMI (kg/m
2
) 30.5 ± 7.6 28.2 ± 7.1 33.4 ± 7.4 <0.001
BMI percentile 94.5 ± 10.3 92.3 ± 10.8 97.9 ± 3.6 <0.001
SAT (L) 12.1 ± 7.1 10.4 ± 6.6 14.2 ± 7.2 <0.001
VAT (L) 1.8 ± 1.3 1.4 ± 0.9 2.2 ± 1.4 <0.001
Total Fat (kg) 29.1 ± 12.1 26.1 ± 11.9 32.9 ± 11.5 <0.001
Liver fat (%) 8.8 ± 8.5 3.7 ± 1.1 15.2 ± 9.4 <0.001
ALT (IU/L)
‡
14.9 ± 10.2 10.8 ± 5.4 20.7 ± 12.3 <0.001
AST (IU/L)
‡
20.8 ± 8.2 17.9 ± 4.4 24.8 ± 10.3 <0.001
TAG (mg/dL)
†
107.6 ± 52.7 96.6 ± 45.5 121.8 ± 58.1 0.001
Total
Cholesterol
(mg/dL)
†

140.9 ± 29.6 138.4 ± 28.6 143.9 ± 30.7 NS
HDL (mg/dL)
†
37.7 ± 9.4 39.0 ± 10.1 35.9 ± 8.3 0.024
LDL (mg/dL)
†
85.2 ± 28.4 84.4 ± 28.8 86.2 ± 28.1 NS
Data are shown as mean ± SD. ELF was defined as liver fat content greater than 5.5%
by MRI. NS=not significant. p-values are given for comparison between ELF and normal
liver fat content groups and were obtained via independent t-tests. SAT = subcutaneous
adipose tissue, VAT = visceral adipose tissue. ELF = elevated liver fat content, TAG =
triacylglycerol.
‡
n=208,
†
n=183.

! "#!
Table 2-2. Individual Effects of SNPs on Liver Fat Content.
Effect Allele Copy Number
Gene SNP Alleles
a
EAF
b
Reported
EAF
c

0 1 2 p-value
d

PNPLA3 rs738409 G/C 0.52 0.239

5.6±4.3
(n=53)
8.0±8.3
(n=109)
12.3±9.8
(n=60)
4.2x10
-6

APOC3 rs2854117 A/G 0.35 0.302 7.1±7.6
(n=91)
9.4±8.4
(n=103)
11.3±10.5
(n=26)
0.01
GCKR rs780094 A/G 0.32 0.391 8.6±9.0
(n=101)
8.5±7.4
(n=102)
10.7±9.9
(n=22)
0.44
NCAN rs2228603 T/C 0.03 0.09 8.7±8.5
(n=212)
9.2±7.8
(n=11)
NA 0.96
LYPLAL1 rs12137855 C/T 0.92 0.79 9.7±0
(n=1)
7.8±8.4
(n=36)
8.9±8.5
(n=184)
0.50
PPP1R3B rs4240624 A/G 0.67 0.92 8.7±7.9
(n=101)
8.8±9.6
(n=88)
8.5±7.8
(n=28)
0.38
Data are shown as mean liver fat content (%) ± SD as a function of carrying 0, 1, or 2 copies of the
effect alleles for selected SNPs.
a
Effect/Other allele.
b
EAF, effect allele frequency in Hispanics.
c
Reported effect allele frequency in Caucasians (HapMap-CEU).
d
p-values are obtained from multiple
linear regression using natural log-transformed values, adjusted for age, sex and VAT.
! "#!
Table 2-3. Performance Parameters of Clinical and Genetic Risk Factors on ELF
Prediction.
!
Parameter GRS ALT BMI BMI+ALT BMI+GRS BMI+ALT+GRS
Sens 0.69 0.76 0.65 0.75 0.66 0.76
Spec 0.60 0.68 0.78 0.75 0.73 0.75
Prev 0.23 0.30 0.58 0.40 0.52 0.40
PPV 0.34 0.51 0.80 0.67 0.73 0.68
NPV 0.86 0.87 0.61 0.81 0.67 0.82
DLR+ 1.71 2.35 2.91 2.94 2.48 3.11
DLR- 0.52 0.36 0.46 0.34 0.46 0.31
The estimated sensitivity (Sens), specificity (Spec), prevalence (Prev), positive and
negative predictive values (PPV and NPV, respectively) and diagnostic likelihood ratio
(DLR+ and DLR-) values for the predicted probability of ELF are listed by different
models (columns). ELF = elevated liver fat content, GRS = genetic risk score, ALT =
alanine aminotransferase, BMI = body mass index.

! ""!
Table 2-4. ALT, BMI Percentile and GRS as Predictors of Liver Fat Content.
Model Predictor(s) r
2
SEE
!
p
1 ALT 0.39 6.9 0.39 <0.001
2 GRS 0.1 8.3 2.07 <0.001
3 BMI 0.09 8.2 0.81 <0.001
4 ALT + GRS 0.44 6.7 - <0.001
5 ALT + GRS + BMI 0.48 6.5 - 0.001

Output values from multiple regression analyses. ALT, GRS and BMI percentile were
tested as predictors of liver fat content (expressed as a percentile) independently and in
stepwise models of regression. All variables were tested for collinearity and outliers. The
full prediction model explained the most variance in liver fat content (48%), with an
overall model significance of p=0.001. GRS = genetic risk score, ALT = alanine
aminotransferase, BMI = body mass index, SEE = standard error of the estimate, ! =
beta regression coefficient, p = model significance.

! "$!
Table 2-5. Power calculation for six SNPs previously associated with liver fat
content.
!
Gene SNP ID Effect
Palmer
EAF
Walker
n
Required
PNPLA3 rs738409 0.28† 0.52 94
PPP1R3B rs4240624 0.15† 0.67 392
NCAN rs2228603 0.27† 0.03 891
GCKR rs780094 0.10† 0.32 901
LYPLAL1 rs12137855 0.07† 0.92 5328
!
Assuming 80% power, an !-error of 0.05, and the observed allele frequencies in our
Hispanic study population (EAF
Walker
), our sample size was sufficient to detect a
significant association with PNPLA3 utilizing regression coefficients from Palmer er al.
(†=Effect
Palmer
). However, there was insufficient power to detect association of liver fat
content with the GCKR, LYPLAL1, NCAN, and PPP1R3B variants in our study
population. EAF=effect allele frequency. Calculations performed in Quanto Ver 1.2.4.

! "%!
Figure 2-1: Distribution of BMI percentile, ALT and AST by liver fat content.

Legend: 90% of subjects with ELF had a BMI "95
th
percentile (dark shaded box),
compared to 64% of participants with normal liver fat content (A). 69% and 77% of ELF
subjects had ALT (B) and AST (C) levels, respectively, that were within the normal
reference ranges (ALT = 5-23.5 U/L, AST = 5-35 U/L) indicated by the light shaded box.
Normal = normal liver fat content, Yes= ELF. ELF = elevated liver fat content defined as
" 5.5%.

! "&!
Figure 2-2: Combined genetic effects on liver fat content.

Legend: GRS category (0-4) is shown along the x-axis and mean liver fat ± 95% CI
(open circles with bars) is plotted on the left y-axis. The right y-axis denotes the number
of subjects in each GRS category, represented by the non-shaded bars. The distribution
of GRS in the population was normal. Mean liver fat significantly increased as a function
of GRS (p
trend
=3.12 x 10
-5
).

! "'!
Figure 2-3: Utility of a GRS to predict Elevated Liver Fat Content.

Legend: The ROC curves for the different models are shown where the state variable
was defined by ELF. Test variables were BMI percentile, ALT, GRS and combinations
thereof (listed next to line legend with AUC [CI]). BMI+ALT+GRS (solid red line) was a
significantly better discriminator of ELF than BMI+ALT (AUC=0.85; 95% CI, 0.79-0.89
vs. AUC=0.84; 95% CI, 0.78-0.89, respectively [p=0.01]). The reference line (AUC=0.5)
represents no predictive ability. ELF = elevated liver fat content " 5.5%.

! "(!
Figure 2-4. Combined genetic effects six SNPs on liver fat content.

Legend: Comprehensive 6-SNP GRS tertile category (1-3) is shown along the x-axis and
mean liver fat is plotted on the y-axis. The comprehensive GRS was based on tertiles of
risk alleles (T1=0-3, n=22;T2= 4-6, n=143; and T3=7-9, n=44) and tested for an
association with liver fat. Mean liver fat was not significantly increased as a function of
comprehensive GRS (p
trend
=0.08).

! ")!
Chapter 2 references:
1. Barshop NJ, Sirlin CB, Schwimmer JB, Lavine JE. Review article: epidemiology,
pathogenesis and potential treatments of paediatric non-alcoholic fatty liver disease.
Aliment Pharmacol Ther 2008;28(1):13-24.
2. Patton HM, Sirlin C, Behling C, Middleton M, Schwimmer JB, Lavine JE. Pediatric
nonalcoholic fatty liver disease: a critical appraisal of current data and implications for
future research. J Pediatr Gastroenterol Nutr 2006;43(4):413-427.
3. Chalasani N, Younossi Z, Lavine JE, Diehl AM, Brunt EM, Cusi K, et al. The
Diagnosis and Management of Non-alcoholic Fatty Liver Disease: Practice Guideline by
the American Gastroenterological Association, American Association for the Study of
Liver Diseases, and American College of Gastroenterology. Gastroenterology
2012;142(7):1592-609.
4. Lavine JE, Schwimmer JB. Nonalcoholic fatty liver disease in the pediatric population.
Clin Liver Dis 2004;8:549-558.
5. Goran MI, Walker R, Le KA, Mahurkar S, Vikman S, Davis JN, et al. Effects of
PNPLA3 on liver fat and metabolic profile in Hispanic children and adolescents. Diabetes
2010;59(12):3127-3130.
6. Prati D, Colli A, Conte D, Colombo M. Spectrum of NAFLD and diagnostic implications
of the proposed new normal range for serum ALT in obese women. Hepatology
2005;42(3):650-656.
7. Reid AE. Nonalcoholic steatohepatitis. Gastroenterology 2001;121(3):710-723.
8. Romeo S, Kozlitina J, Xing C, Pertsemlidis A, Cox D, Pennacchio LA, et al. Genetic
variation in PNPLA3 confers susceptibility to nonalcoholic fatty liver disease. Nat Genet
2008;40(12):1461-1465.
9. Speliotes EK, Yerges-Armstrong LM, Wu J, Hernaez R, Kim LJ, Palmer CD,
Gudnason V, et al. Genome-wide association analysis identifies variants associated with
nonalcoholic fatty liver disease that have distinct effects on metabolic traits. PLoS Genet
2011;7(3);!e1001324.
10. Petersen KF, Dufour S, Hariri A, Nelson-Williams C, Foo JN, Zhang XM, et al.
Apolipoprotein C3 gene variants in nonalcoholic fatty liver disease. N Engl J Med
2010;362(12):1082-1089.
11. Hasson RE, Adam TC, Davis JN, Weigensberg MJ, Ventura EE, Lane CJ, et al.
Ethnic differences in insulin action in obese African-American and Latino adolescents. J
Clin Endocrinol Metab 2010;95(8)4048-4051.
! $*!
12. Szczepaniak LS, Nurenberg P, Leonard D, Browning JD, Reingold JS, Grundy S, et
al. Magnetic resonance spectroscopy to measure hepatic triglyceride content:
prevalence of hepatic steatosis in the general population. Am J Physiol Endocrinol
Metab 2005;288(2):462-468.
13. Henry RJ, Chiamori N, Golub OJ, Berkman S. Revised spectrophotometric methods
for the determination of glutamic-oxalacetic transaminase, glutamic-pyruvic
transaminase, and lactic acid dehydrogenase. Am J Clin Pathol 1960;34:381-398.
14. Schwimmer JB, Dunn W, Norman GJ, Pardee PE, Middleton MS, Kerkar N, et al.
SAFETY study: alanine aminotransferase cutoff values are set too high for reliable
detection of pediatric chronic liver disease. Gastroenterology 2010;138:1357–1364.
15. Sackett, D. Evidence-Based Medicine: How to Practice and Teach EBM. 2nd ed. ed.
2000: Churchill Livingstone.
16. Kim WR, Flamm SL, Di Bisceglie AM, Bodenheimer HC; Public Policy Committee of
the American Association for the Study of Liver Disease. Serum activity of alanine
aminotransferase (ALT) as an indicator of health and disease. Hepatology
2008;47(4):1363-1370.
17. Prati D, Taioli E, Zanella A, Della Torre E, Butelli S, Del Vecchio E, et al. Updated
definitions of healthy ranges for serum alanine aminotransferase levels. Ann Intern Med
2002;137(1):1-10.
18. Kim HC, Nam CM, Jee SH, Han KH, Oh DK, Suh I. Normal serum aminotransferase
concentration and risk of mortality from liver diseases: prospective cohort study. BMJ
2004;328(7446):983-983.
19. Schwimmer JB, McGreal N, Deutsch R, Finegold MJ, Lavine JE. Influence of gender,
race, and ethnicity on suspected fatty liver in obese adolescents. Pediatrics
2005;115(5):561-565.
20. Ruhl, CE, Everhart JE. Upper limits of normal for alanine aminotransferase activity in
the United States population. Hepatology 2012;55(2):447-454.
21. Kotronen A, Johansson LE, Johansson LM, Roos C, Westerbacka J, Hamsten A, et
al. A common variant in PNPLA3, which encodes adiponutrin, is associated with liver fat
content in humans. Diabetologia 2009;52(6):1056-1060.
22. Romeo S, Sentinelli F, Cambuli VM, Incani M, Congiu T, Matta V, et al. The 148M
allele of the PNPLA3 gene is associated with indices of liver damage early in life. J
Hepatol 2010;53(2):335-338.
! $+!
23. Rotman Y, Koh C, Zmuda JM, Kleiner DE, Liang TJ; NASH CRN. The association of
genetic variability in patatin-like phospholipase domain-containing protein 3 (PNPLA3)
with histological severity of nonalcoholic fatty liver disease. Hepatology 2010;52(3):894-
903.
24. Sookoian S, Castaño GO, Burgueño AL, Gianotti TF, Rosselli MS, Pirola CJ. A
nonsynonymous gene variant in the adiponutrin gene is associated with nonalcoholic
fatty liver disease severity. J Lipid Res 2009;50(10):2111-2116.
25. Valenti L, Alisi A, Galmozzi E, Bartuli A, Del Menico B, Alterio A, et al. I148M patatin-
like phospholipase domain-containing 3 gene variant and severity of pediatric
nonalcoholic fatty liver disease. Hepatology 2010;52(4):1274-1280.
26. Palmer ND, Musani SK, Yerges-Armstrong LM, Feitosa MF, Bielak LF, Hernaez R et
al. Characterization of European-ancestry NAFLD-associated Variants in individuals of
African and Hispanic descent. Hepatology. 2013 Apr 8. doi: 10.1002/hep.26440. [Epub
ahead of print]
27. Yang Z, Wen J, Tao X, Lu B, Du Y, Wang M, et al. Genetic variation in the GCKR
gene is associated with non-alcoholic fatty liver disease in Chinese people. Mol Biol Rep
2011;38(2):1145-1150.
28. Santoro N, Zhang CK, Zhao H, Pakstis AJ, Kim G, Kursawe R, et al. Variant in the
glucokinase regulatory protein (GCKR) gene is associated with fatty liver in obese
children and adolescents. Hepatology 2011;55(3):781-9.
29. Aalto-Setälä K, Fisher EA, Chen X, Chajek-Shaul T, Hayek T, Zechner R, et al.
Mechanism of hypertriglyceridemia in human apolipoprotein (apo) CIII transgenic mice.
Diminished very low density lipoprotein fractional catabolic rate associated with
increased apo CIII and reduced apo E on the particles. J Clin Invest 1992;90(5):1889-
1900.
30. Ito Y, Azrolan N, O'Connell A, Walsh A, Breslow JL. Hypertriglyceridemia as a result
of human apo CIII gene expression in transgenic mice. Science 1990;249(4970):790-
793.
31. Hyysalo J, Stojkovic I, Kotronen A, Hakkarainen A, Sevastianova K, Makkonen J, et
al. Genetic variation in PNPLA3 but not APOC3 influences liver fat in non-alcoholic fatty
liver disease. J Gastroenterol Hepatol 2012;27(5):951-6
32. Lee HY, Birkenfeld AL, Jornayvaz FR, Jurczak MJ, Kanda S, Popov V, et al.
Apolipoprotein CIII overexpressing mice are predisposed to diet-induced hepatic
steatosis and hepatic insulin resistance. Hepatology 2011;54(5):1650-1660.
! $#!
33. Kozlitina J, Boerwinkle E, Cohen JC, Hobbs HH. Dissociation between APOC3
variants, hepatic triglyceride content and insulin resistance. Hepatology 2011;53(2):467-
474.
34. Sentinelli F, Romeo S, Maglio C, Incani M, Burza MA, Scano F, et al. Lack of effect
of apolipoprotein C3 polymorphisms on indices of liver steatosis, lipid profile and insulin
resistance in obese Southern Europeans. Lipids Health Dis 2011;10:93-93.
35. Valenti L, Nobili V, Al-Serri A, Rametta R, Leathart JB, Zappa MA, et al. The APOC3
T-455C and C-482T promoter region polymorphisms are not associated with the severity
of liver damage independently of PNPLA3 I148M genotype in patients with nonalcoholic
fatty liver. J Hepatol 2011;55(6):1409-1414.
36. Duseja A, Aggarwai R. APOC3 and PNPLA3 in non-alcoholic fatty liver disease:
need to clear the air. J Gastroenterol Hepatol 2012;27(5):848-51.
37. McPherson S, Stewart SF, Henderson E, Burt AD, Day CP. Simple non-invasive
fibrosis scoring systems can reliably exclude advanced fibrosis in patients with non-
alcoholic fatty liver disease. Gut 2010;59(9):1265-9.
38. Younossi ZM, Page S, Rafiq N, Birerdinc A, Stepanova M, Hossain N, et al. A
biomarker panel for non-alcoholic steatohepatitis (NASH) and NASH-related fibrosis.
Obes Surg 2011;21(4):431-9.
39. Sartorio A, Del Col A, Agosti F, Mazzilli G, Bellentani S, Tiribelli C et al. Predictors of
non-alcoholic fatty liver disease in obese children. Eur J Clin Nutr 2007;61(7):877-83.
40. Bedogni G, Miglioli L, Masutti F, Tiribelli C, Marchesini G, Bellentani S. Prevalence of
and risk factors for nonalcoholic fatty liver disease: the Dionysos nutrition and liver study.
Hepatology 2005;42(1):44-52.
41. Maffeis C, Banzato C, Rigotti F, Nobili V, Valandro S, Manfredi R et al. Biochemical
parameters and anthropometry predict NAFLD in obese children. JPGN 2011;53(9):590-
3.

! $"!
Chapter 3: Fructose content of sugar sweetened beverages and juices

Abstract
Objective: Excess fructose consumption is associated with metabolic disease risk.
Actual fructose consumption levels are difficult to estimate because of the unknown
quantity of fructose in beverages. We determined the actual fructose content in
beverages made with and without HFCS as an added sweeter.

Design and Methods: Ten previously tested sugar sweetened beverages (SSBs) were
analyzed via metabalomics to determine sugar profiles. Additionally, twenty randomly
selected juices were analyzed via gas chromatography to obtain sugar profiles.

Results: For SSBs, three independent laboratory methods showed consistent and
reproducible results. In popular SSBs made with HFCS as a sweetener, fructose
constituted 60.6±2.7% of sugar content. Two 100% apple juices had fructose to glucose
ratios of ~67% and mean fructose concentration of 65.35 g/L, the highest in the study.

Conclusion: Our results provide evidence of high amounts fructose in beverages
suggesting that current dietary analyses may underestimate actual fructose exposure.
Popular beverages made with HFCS have 50% more fructose than glucose, indicating
this HFCS is vastly different from sucrose, in which fructose and glucose are equivalent.
Given the metabolic consequences of excess fructose consumption, our results support
the need for nutrition labels to incorporate more specific information relating to fructose
sources and amounts.

! $$!
Introduction
Assessment of fructose content in foods and beverages is an important public health
issue to consider, as excess fructose consumption has been linked to fatty liver disease

(1), dyslipidemia

(2), type 2 diabetes

(3), obesity

(4) and gout

(5). Previous work from our
group has shown that the fructose content of sugar sweetened beverages (SSBs)
sweetened with HFCS can be as high as 65% of total sugar content, higher than that
suggested by the fructose content of HFCS-55

(6), contributing to “hidden” fructose in
the diet. Despite clear evidence for the role of excess fructose consumption in disease,
others have argued that any additional fructose in HFCS is minimal, without any adverse
health effects, and that HFCS-55 is “essentially” the same as sucrose in terms of sugar
content

(7). A growing body of clinical evidence suggests that fructose consumption
plays a direct role in metabolic disease risk

(8, 9) and may have adverse effects on
central appetite regulation compared to glucose

(10). Despite this evidence, current
food-labeling practices do not provide information on fructose content in foods and
beverages made with HFCS, fruit juice concentrate or crystalline fructose, all of which
contain fructose and are being used in increasing amounts as added sugar in the food
supply

(11). Since there are currently no disclosures of fructose content in foods and
beverages

(11), it is challenging to accurately determine actual fructose consumption
levels.

To better understand and document the actual fructose content in foods and beverages
that are made with and without HFCS as an added sweetener we sought to: 1)
reproduce and confirm higher than expected fructose content in our previously tested
beverages using two additional assay methods, 2) compare data across all methods to
determine the actual free fructose to glucose ratio in beverages made with HFCS, and 3)
expand the analysis to determine fructose content in commonly consumed juice
products.

Methods and Procedures
Based on product popularity

(12), we selected 10 of the 23 beverages that we previously
tested using liquid chromatography (LC)

(6), for follow-up analysis using two alternative
methods to determine sugar content: 1) metabolomic (MET) analysis based on mass
spectrometry with combined liquid and gas chromatography and 2) gas chromatography
! $%!
(GC). In addition we extended the use of GC to analyze a selection of juice products, as
described below.

Metabolomics Analysis
Popular sugar sweetened beverages (SSBs) were purchased from retailers in East Los
Angeles in 2012. Beverages were selected to replicate the sampling used in our
previous study

(6), in which the selection of beverages was based upon consumption
frequencies of children in our past studies. Nutrition label information and serving size
data were recorded. Immediately after opening bottled/canned beverages, 500ul
samples were aliquoted and transferred to Eppendorf Cryotubes. All samples were held
under refrigeration and sequentially flash frozen in liquid nitrogen within 1 hour of the
initial transfer. Samples were stored at -20 degree Celsius overnight before shipment.
Glucose, fructose, sucrose and maltose standard solutions were created from research
grade reagents (Sigma-Aldrich, St. Louis. MO) to serve as controls. Ten grams of the
sucrose, fructose and glucose reagents were added to 100mL of Millipore water and
brought into solution. Two concentrations of maltose were prepared, 10g/L and 1g/L.
Finally, a 50:50 solution of fructose and sucrose was prepared by combining 5 grams of
each reagent with 100mL of water. These sugar standard concentrations were chosen to
replicate the approximate sugar content equivalents found in most sweetened beverages
with the two maltose preparations representing trace amounts of this sugar that may be
found in sweetened beverages. For all standards 500uL aliquots were taken and
prepared as described above. All samples were shipped overnight packed in dry ice to
Metabolon (Research Triangle Park, Durham NC). Metabolon was blinded to the source
of all samples and standards and samples were analyzed according to previously
described methodologies using a metabolomics approach to examine a broad array of
sugars

(13). Samples were split into equal parts for analysis on the gas
chromatography/mass spectrometry (GC/MS) and liquid chromatography/mass
spectrometry (LC/MS) platforms based on previously published methodology (14). The
GC column was 5% phenyl and the temperature ramp was from 40° to 300° C in a 16
minute period. Samples were analyzed on a Thermo-Finnigan Trace DSQ fast-scanning
single-quadrupole mass spectrometer using electron impact ionization. The LC/MS
portion of the platform was based on a Waters ACQUITY UPLC and a Thermo-Finnigan
LTQ mass spectrometer, which consisted of an electrospray ionization source and linear
ion-trap mass analyzer. Compounds were identified by comparison to library entries of
! $&!
purified standards or recurrent unknown entities. Identification of known chemical
entities was based on comparison to metabolomic library entries of purified standards.
The combination of chromatographic properties and mass spectra gave an indication of
a match to the specific compound or an isobaric entity.

Gas Chromatography Analysis
The ten SSBs analyzed in the MET analysis were again selected along with four
additional randomly selected SSBs and 20 other juice products. Online shopping
databases for Walmart, SuperValu, and Safeway were accessed in order to select
samples. To control for location and inventory, online store inventories were selected
within a defined zip code region (90033). Twenty juices were randomly selected by
choosing every tenth product in the retailers’ databases until 10 products made with
HFCS and 10 products made without HFCS, according to package ingredients labels,
were selected. One juice product was omitted from the analysis due to handling error
resulting in 19 products that proceeded to assay. All samples were aliquotted to sterile,
sealed containers and sample weights were determined and recorded. Samples were
packaged and shipped overnight on dry ice to Covance Laboratories (Madison, WI) for
subsequent blinded analysis via gas chromatography according to previously published
methods

(15-17). The sugar profile analysis conducted at Covance was applicable to the
determination of fructose, galactose, glucose, sucrose, lactose, and maltose in as little
as 10g of food products, syrups, and beverages using gas chromatography as described
below. Once received, samples were prepared in accordance with Covance procedures
and sugars were extracted from the homogenized sample with water. Aliquots were
dried under inert gas and reconstituted with a hydroxylamine hydrochloride solution in
pyridine containing phenyl-#-D-glucoside as the internal standard. The resulting oximes
were converted to silyl derivatives with hexamethyldisilazane (HMDS) and trifluoracetic
acid (TFA) treatment and analyzed by gas chromatography

(18, 19) using a flame
ionization detector. The limit of quantitation for most matrices is 0.1%. The relative
standard deviations, on a cereal matrix, for fructose, glucose, sucrose, and maltose were
4.9%, 7.4%, 3.2%, and 6.4%, respectively. Specific gravity testing was conducted

(20)
on all liquid samples to allow the reporting of sugar content in appropriate units of
measure.

! $'!
Data Analysis
Examination of sugar composition in 10 beverages across 3 different methods:
A mean with standard deviation and coefficient of variation (CV) were calculated for
fructose, glucose, sucrose and maltose across methodologies to assess consistency
across the independent methods. Percent of total sugar was calculated for all measured
sugars in the SSBs analyzed via the three methodologies.

Juice analysis:
Data for individual sugars were reported in the following formats; percent of total sugar
(%TS), concentration of each sugar in grams per liter (g/L) and grams per serving (g/s).
Actual total sugar (TS
Actual
) was calculated and compared to total label sugar (TS
Label
).
Free fructose to glucose ratios (F:G) and the concentration of free fructose (F
concentration
)
in each product were also assessed. The raw F:G was adjusted (F:G
Adjusted
) to account
for the additional glucose that the disaccharide maltose may contribute to the overall
sugar profile of the products. F:G values were reported using the first number,
representing fructose, as the referent (ex. F:G of 60:40; reported as 60). Formulas used
to obtain these values are presented in Table 3-1.

Results
Fructose Content of SSBs: Methodological comparison
We first compared the fructose content of the original 10 beverages, as measured by
three independent laboratories (LC

(6), MET and GC), which are displayed in Figure 3-
1. Results were consistent across all three methodologies for percent fructose, glucose
(Figure 3-1), sucrose and maltose, remaining relatively similar across methodologies
and within samples. Free fructose content was notably consistent across methodologies
with standard deviations remaining below 3.6%, with the exception of Gatorade
(SD=4.5%). Mean free fructose content, expressed as a percent of all sugars, for
beverages listing HFCS as an ingredient was 60.6±2.7%. In all remaining beverages, the
mean free fructose content, expressed as a percent of all sugars, was 35.5±15.4%.
Mexican Coca-Cola consistently contained ~49% of total sugar as free fructose despite
HFCS or fructose not being listed on the label. Additionally, Pepsi Throwback, Gatorade
and Sierra Mist, all which list neither HFCS nor fructose as added sweeteners, contained
fructose as a percent of total sugar in ~50, 40 and 8%, respectively. Analyses confirmed
that only trace amounts of maltose (not greater than 1.7% of sugars) were present in the
! $(!
sampled beverages. The CV values for fructose and glucose were consistently below
0.12 and 0.1, indicating reliability between measures. Mexican Coca Cola had a glucose
CV of 0.2% and an artificially elevated sucrose CV of 0.9% due to the original analysis
detecting no sucrose resulting in a very high SD. Sierra Mist was not assayed in the
original analysis, therefore no CV was reported. The CV of sucrose in other products
was in all cases less than 0.2%. In the MET and GC analyses, maltose was only
detected in 4 and 3 of the 10 beverages, respectively. CV values ranged from 0.1 to 0.3,
likely due to the trace amounts detected via the 2 methods. Maltose was not measured
in the initial study.

Sugar Analysis Using Gas Chromatography
SSBs and Juices
Seven beverages listing HFCS as an ingredient had a F:G
Adjusted
of > 55, with a mean
value of 59.6±0.5 (Figure 3-2). Among products not listing HFCS as a sweetener, the
mean F:G
Adjusted
was 50.7±0.6. F:G
Raw
values were not altered when adjusted for
disaccharides. Mean F
concentration
in products listing HFCS was 59.4±8.9 g/L vs. 30.8±19.5
g/L for non-HFCS products (Figure 3-2). Values obtained by GC showed that beverages
contained 0.01 to 11% less total sugar than reported on the label (Table 3-2). Sprite, Dr.
Pepper and Pepsi had free fructose accounting for 60% or more of total sugar. Several
SSBs that did not list HFCS or fructose as an ingredient on the nutrition label had
F
concentration
substantially greater than zero (Mexican Coca-Cola; 51 g/L, Pepsi Throwback;
42 g/L, Gatorade; 23 g/L, Sierra Mist; 7 g/L). Pepsi lists sucrose as an included
ingredient, however no sucrose was detected in Pepsi using GC methodology and its
F:G
Adjusted
was 60. Maltose levels did not exceed 2% of sugars in any of the beverages.

Minute Maid and Juicy Juice 100% apple juices had F:G
Adjusted
values of ~67 with a mean
F
concentration
of 65.35 g/L, the highest in the study, despite no added HFCS (Figure 3-3).
Five other juices had F:G
Adjusted
values of greater than 55. Hawaiian Punch had the
highest F:G
Adjusted
value (61.5) amongst the products listing HFCS as an ingredient.
Mean F:G
Adjusted
and F
concentration
for HFCS products were 52.1±5.9 and 45.7±10.6 g/L,
respectively. Mean F:G
Adjusted
and F
concentration
for non-HFCS products were 56.7±6.9 and
45.2±16.6 g/L, respectively. Juices ranged from 79.5 to 108.5% of total sugar listed on
the nutrition label (Table 3-2).

! $)!
Discussion
This is the first comprehensive study determining fructose content and sugar profiles in a
broad selection of SSBs and juice products. The results of the multi-method sugar profile
analysis were strikingly similar in terms of fructose content. Our prior work
demonstrated that in popular SSBs, fructose constituted 65% of the total sugar

(6),
however this initial analysis may have been methodologically limited
7
in that maltose,
which may potentially alter the fructose to glucose ratio, was not measured. In the
current study we used 2 additional and independent assays that were capable of
detecting the presence of trace sugars and disaccharides, including maltose, and
extended our analysis beyond SSBs to juices.

The clearest and most consistent finding in our study was that the top 5 most popular

(12) HFCS-sweetened sodas made by companies that comprise ~90% of the annual
beverage market share (12) (Coca-Cola, Pepsi, Dr Pepper, Mountain Dew and Sprite)
have fructose to glucose ratios of ~60:40, meaning they contain ~50% more fructose
than glucose. This fructose content differs dramatically from the 50:50 ratio found in
sucrose and from the assumed ratio in HFCS-55. These findings, which were confirmed
by three independent laboratories, were maintained after adjusting for the presence of
trace sugars, and support our initial report

(6) providing further evidence for the elevated
fructose to glucose ratios in the most popular SSBs made with HFCS. We can only
speculate as to why our analysis indicates a higher than expected fructose content in
these beverages made with HFCS. HFCS can be manufactured to have variable
fructose contents

(21) and is also available in higher concentrations up to HFCS-90

(22)
(90% fructose). One possible explanation of the higher fructose content may be the
blending of HFCS-90 with glucose syrup

(21) to create products with fructose contents
higher than HFCS-55, thereby optimizing sweetness and flavoring to achieve what has
been termed the “bliss point” in food product development and production

(23). This
strategy is both feasible and allowable under current regulations, as current FDA
guidelines for use of HFCS-55 as an ingredient only require it to be a “minimum” of 55%
fructose

(24, 25) and allow the unrestricted sales and use of HFCS-90

(25). Without
specification of the actual fructose content and blend of HFCS used, it is unclear exactly
how much actual fructose is contained in food and beverage products. Given the higher
than expected ratio of fructose to glucose in some HFCS containing products, it is not
accurate to consider HFCS-55 nutritionally identical to sucrose

(26) especially when
! %*!
considering the strong evidence attributing a host of metabolic consequences

(27) to
excess fructose consumption. These findings support our belief that food labels should
include, in addition to listing HFCS and total sugar amounts, specific sources and
amounts of fructose.

Sugars not listed on the nutrition labels were detected in several of the SSBs analyzed.
For example, Mexican Coca Cola had a high free fructose concentration (51g/L), despite
no source of free fructose being listed as an ingredient. This finding suggests that
fructose or HFCS was used as an ingredient. Alternatively, sucrose could conceivably
break down during storage into fructose and glucose; however, it is unlikely that this
would fully explain the high fructose

value. Similarly, Pepsi, which lists sugar in addition
to HFCS as a sweetener, contained no sucrose and had a fructose to glucose ratio of
60:40 suggesting the presence of HFCS. In this case the hydrolysis of sucrose to
monosaccharides would not explain the elevated ratio of fructose to glucose.

In the analysis of juices we found that the mean fructose concentration among all juices
was 45.5g/L, which is comparable to that of all sodas (50.4 g/L). Minute Maid and Juicy
Juice 100% apple juices had the highest fructose to glucose ratios. These juices were
not sweetened with HFCS, but still had a higher fructose concentration than most sodas.
Many juices not containing HFCS utilize fruit juice concentrate as a sweetener, which is
the most commonly listed sweetener in 100% fruit juice products

(11). Juice products are
often advertised as a healthy alternative to SSBs; however, in terms of free fructose
content, our data suggests that certain juice products may contribute to daily fructose
exposure equivalent to or greater than that of sodas. Sunny D and Ocean Spray 100%
cranberry juice also had a fructose to glucose ratio

of ~60, again suggesting 50% more
fructose than glucose in these products. These products likely contain natural fruit
sugars; however, the overall sugar profiles are strikingly similar to those of SSBs
sweetened with HFCS. When total fructose exposure is considered (free fructose +
fructose from sucrose), juices contained a mean concentration of fructose

almost
equivalent to that of sodas (51.4±15.4 vs. 55.7±11.7g/L, respectively). Although sodas
are the most consumed source of SSBs in adults and children, juice consumption has
increased in adolescent and minority populations in recent years

(28). Considering larger
serving sizes, higher daily consumption rates of juices and the more common use of fruit
juice concentrate or HFCS in these products, there is likely a higher than expected daily
! %+!
fructose exposure in the population that warrants further research on the metabolic
consequences of chronic high fructose intake.

Taken together, our analyses indicate that in many cases SSB and juice products can
contain upwards of 5-15% more free fructose than would be expected based on use of
HFCS-55. This suggests that prior population based studies reporting fructose or HFCS
consumption

(29-31) likely underestimate fructose exposure and that US consumers,
especially children, are exposed to higher daily fructose amounts than expected. Based
on national data children often consume > 70 g/d

(32) and can consume up to 121g of
fructose per day

(29); 2 times greater than when assessed in 1978

(30) and 10 times
greater than the 6 g/d per capita value used to determine the safety of consumption in
1976

(33). Fructose exposure at these levels has been shown to be metabolically
deleterious in humans

(34-36). It is plausible that additional, unlabeled amounts of
fructose contained in SSBs and juices can add up and, in combination with other
commonly consumed high fructose containing foods, can easily result in fructose intake
of > 100g/day. Four servings of certain sodas or juices per day can represent upwards of
25% of the total caloric content of a 2000 kcal/d diet. Given that high daily fructose
consumption is known to promote visceral fat expansion and hepatic steatosis

(37, 38)
as well as the progression from nonalcoholic fatty liver disease to nonalcoholic
steatohepatitis

(39), chronic, long-term consumption of beverages high in fructose
represents a significantly greater risk to some populations

(40), suggesting that informing
the consumer of the sugar content of foods through appropriate labeling could lead to
responsible consumption of sweetened foods

(41) and assist in preventing adverse
metabolic outcomes.

In conclusion, this study strengthens our previous findings regarding the fructose content
of SSBs and provides new information on the sugar composition and overall fructose
content of commonly consumed SSB and juice products. The results support our initial
findings

(6), suggesting the most popular sodas made with HFCS as the sole added
sweetener have a fructose to glucose ratio of 60:40, indicating a meaningful difference
from the equivalent ratio of fructose to glucose observed in table sugar. Certain fruit
juices contain more fructose than sodas, often with 50% more fructose than glucose.
Although SSBs are a major source of fructose in the diet of Americans, our results
demonstrate that juice products may contribute substantially to total daily fructose
! %#!
consumption. Based on these findings, current population estimates of fructose
consumption determined via existing food nutrient data are likely underestimated due to
this additional ‘hidden’ fructose. Therefore, the addition of fructose sources and amount
to soda and juice labels may be a viable means to inform consumers and reduce chronic
exposure to high levels of fructose in the diet.

! %"!

Table 3-1. Formulas.

Definitions: F=fructose, G=glucose, GAL=galactose, M=maltose, L=lactose, S=sucrose
1. TS
Actual
per 100g sample = S + L + M + G + F + GAL
2. Total % sugar in sample = (X g sugar TS
Actual
per 100g sample) * 100
3. TS
Actual
per serving size = TS
Acrtual
per 100g (100 / X g/s)
4. Difference in actual vs. label sugar amount = TS
Acrtual
– TS
Label
5. Amount individual sugars per serving = X grams per 100g (100 X g/s)
7. Grams G from disaccharide = grams M + (0.5 * grams L)
8. Percent over label sugar = [(TS
Acrtual
– TS
Label
) TS
Label
] *100
9. F:G (raw) = [F grams (F grams + G grams)] * 100
10. F:G (adjusted) =
[F grams (F grams + G grams +grams G from disaccharide )] * 100

*Sugar Calculations (based upon lab results provided in g/100g sample format)

! %$!

Table 3-2. Percent of total sugar in excess of labeled total sugar

Sodas/Sports Drinks Difference From Label (%)
*Mug -11.18
*Wal-Mart Ginger Ale -7.33
*Mountain Dew -6.69
*7-Up -6.25
*Pepsi -6.2
*Dr. Pepper -5.79
*Super Chill Cola -4.96
Coca-Cola (Mexican) -4.32
*Coca-Cola -4.31
Sierra Mist Natural -3.84
*Sprite -2.42
Gatorade -1.33
Pepsi Throwback -1.27
*Arizona Iced Tea -0.16
Juices
Ocean Spray 100% Cranberry -20.5
Ocean Spray Light Cranberry -13.12
*Minute Maid Premium Fruit Punch -11.66
*V8 Splash Berry Blend -8.43
*Kern's Peach -6.74
Sunny D -6.59
*Hawaiian Punch -6.03
*Kool-Aid Jammers -5.85
Great Value Ruby Red Grapefruit -5.15
Great Value Cranberry -5.08
*Welch's Passion Fruit -1.87
Juicy Juice 100% Apple -1.11
Ocean Spray Blueberry -0.58
Minute Maid 100% Apple 0.67
*Clamato 0.86
*Kern's Strawberry Banana 1.09
*Caprisun 1.8
Tropicana 100% Orange 8.51

Asterisk represents products listing HFCS as an ingredient.

! %%!
Figure 3-1. Mean sugar comparison of sodas across three independent methods

! %&!

Legend: Mean sugar comparison of sodas across three independent methods.
Panel A displays percent of total sugar shown to be free fructose in soda/sports drink
products. The dashed line represents 55% fructose expected of HFCS-55. Panel B
displays percent of total sugar shown to be free glucose in soda/sports drink products.
The dashed line represents 45% glucose expected of HFCS-55. Bars represent
methodology used to determine sugar profiles: LC=liquid chromatography,
MET=metabolomics and GC=gas chromatography.
* denotes products with HFCS listed as an ingredient on the label.

! %'!
Figure 3-2. Fructose concentration and Fructose to Glucose Ratio: Sodas/Sports Beverages

Legend: The concentration of fructose (g/L) in soda/sports drink products is displayed on
the left y-axis (open bars) and the F:G
Adjusted
is shown on the right y-axis (solid bars).
* denotes products with HFCS listed as an ingredient on the label.

*Mountain Dew
*Mug Root Beer
*Pepsi
*Coca-Cola
*Sprite
*Dr. Pepper
*Arizona Iced Tea
*Super Chill Cola
*7-Up
*Wal-Mart Ginger Ale
Coca-Cola (Mexican)
Pepsi Throwback
Gatorade
Sierra Mist Natural
0
10
20
30
40
50
60
70
80
40
45
50
55
60
65 HFCS on label No HFCS on label
Figure 2. Fructose concentration and Fructose to Glucose Ratio:
Sodas/Sports Beverages
! Free Fructose (g/L)
" F:G (adjusted)
! %(!
Figure 3-3. Fructose concentration and Fructose to Glucose Ratio: Juices.

Legend: The concentration of fructose in juices is displayed on the left y-axis (open bars)
and the F:G
Adjusted
is shown on the right y-axis (solid bars).
* denotes products with HFCS listed as an ingredient on the label.

*Hawaiian Punch
*V8 Splash Berry Blend
*Minute Maid Premium Fruit Punch
*Welch's Passion Fruit
*Caprisun
*Caprisun Pacific Cooler
*Clamato
*Kern's Peach
*Kern's Strawberry Banana
*Kool-Aid Jammers
Juicy Juice 100% Apple
Minute Maid 100% Apple
Sunny D
Oceanspary 100% Cranberry
Ocean Spray Light Cranberry
Oceanspray Blueberry Cocktail
Tropicana 100% Orange
Great Value Ruby Red Grapefruit
Great Value Cranberry
0
10
20
30
40
50
60
70
80
0
10
20
30
40
50
60
70
HFCS on label No HFCS on label
Figure 3. Fructose concentration and Fructose to Glucose Ratio:
Juices.
! Free Fructose (g/L)
" F:G (adjusted)
! %)!
Chapter 3 References
1. Vos MB, Lavine JE. Dietary fructose in nonalcoholic fatty liver disease. Hepatology.
2013 Feb 6. doi: 10.1002/hep.26299. [Epub ahead of print]
2. Stanhope KL. Role of fructose-containing sugars in the epidemics of obesity and
metabolic syndrome. Annu Rev Med. 2012;63:329-43.
3. Goran MI, Ulijaszek SJ, Ventura EE. High fructose corn syrup and diabetes
prevalence: a global perspective. Glob Public Health. 2013;8(1):55-64.
4. Basu S, Yoffe P, Hills N, Lustig RH. The relationship of sugar to population-level
diabetes prevalence: an econometric analysis of repeated cross-sectional data. PLoS
One. 2013; 8(2):e57873.
5. Choi HK, Willett W, Curhan G. Fructose-rich beverages and risk of gout in women.
JAMA. 2010 Nov 24;304(20):2270-8.
6. Ventura EE, Davis JN, Goran MI. Sugar content of popular sweetened beverages
based on objective laboratory analysis: focus on fructose content. Obesity (Silver
Spring). 2011 Apr;19(4):868-74.
7. Hobbs L, Kreuger D. Response to “Sugar Content of Popular Sweetened Beverages
Based on Objective Laboratory Analysis: Focus on Fructose Content”. Obesity (Silver
Spring). 2011;19(4):687.
8. Stanhope KL, Schwarz JM, Havel PJ. Adverse metabolic effects of dietary fructose:
results from the recent epidemiological, clinical, and mechanistic studies. Curr Opin
Lipidol. 2013 Apr 16. [Epub ahead of print]
9. The InterAct consortium. Consumption of sweet beverages and type 2 diabetes
incidence in European adults: results from EPIC-InterAct. Diabetologia. 2013 Apr 26.
10. Page KA, Chan O, Arora J, Belfort-Deaguiar R, et al. Effects of fructose vs glucose
on regional cerebral blood flow in brain regions involved with appetite and reward
pathways. JAMA. 2013 Jan 2;309(1):63-70.
11. Ng SW, Slining MM, Popkin BM. Use of caloric and noncaloric sweeteners in US
consumer packaged foods, 2005-2009. J Acad Nutr Diet. 2012 Nov;112(11):1828-34
12. John Sicher (Editor & Publisher). “Special Issue: Top-10 CSD Results for 2010.”
Beverage Digest. 17 Mar. 2010: 59(5): 1-2.
13. Milburn MV, Lawton KA. Application of metabolomics to diagnosis of insulin
resistance. Annu Rev Med. 2013 Jan 14;64:291-305.
14. Milburn MV, Lawton KA. Application of metabolomics to diagnosis of insulin
resistance. Annu Rev Med. 2013 Jan 14;64:291-305.
! &*!
15. Mason, BS, Slover, HT. A Gas Chromatographic Method for the Determination of
Sugars in Foods. Journal of Agricultural and Food Chemistry, 19(3): 551-554, (1971).
(Modified)
16. Brosbt, K. Gas Liquid Chromatography of Trimethylsilyl Derivations. Methods in
Carbohydrate Chemistry, 6:3-8, Academic Press, New York, NY (1972). (Modified)
17. United States Pharmacopeia, Twenty-Sixth Revision, <841>, United States.
Pharmacopeial Convention, Inc.: Rockville, Maryland (2003)
18. Mason, BS, Slover, HT. A Gas Chromatographic Method for the Determination of
Sugars in Foods. Journal of Agricultural and Food Chemistry, 19(3): 551-554, (1971).
(Modified)
19. Brosbt, K. Gas Liquid Chromatography of Trimethylsilyl Derivations. Methods in
Carbohydrate Chemistry, 6:3-8, Academic Press, New York, NY (1972). (Modified)
20. United States Pharmacopeia, Twenty-Sixth Revision, <841>, United States.
Pharmacopeial Convention, Inc.: Rockville, Maryland (2003)
21. Parker K, Salas M, Nwosu VC. High fructose corn syrup: Production, uses and public
health concerns. Biotechnology and Molecular Biology Review. 2010; 5(5):71-8.
22. ADM. (n.d.). Cornsweet® 90. Product Database. Retrieved May 8
th
, 2013 from
http://www.adm.com
23. Moss, Michael. Salt, Sugar, Fat: How the Food Giants Hooked Us. New York:
Random House, 2013. Print.
24. Food chemical codex. 4
th
Edn National Academy Press: Washington DC, 1996
pp191-192
25. United States. Corn Refiners Association. Letter to Division of Dockets and
Management (HFS-305), Food and Drug Administration. Re: Docket No. FDA-2012-
P0904. Washington: Jan. 30, 2013. Letter.
26. Food and Drug Administration. (n.d.). High Fructose Corn Syrup: Questions and
Answers. Food Additives and Ingredients. Retrieved April 12
th
, 2013 from
http://www.fda.gov/
27.Tappy L, Mittendorfer B. Fructose toxicity: is the science ready for public health
actions? Curr Opin Clin Nutr Metab Care. 2012 Jul;15(4):357-61.
28. Han E, Powell LM. Consumption patterns of sugar-sweetened beverages in the
United States. J Acad Nutr Diet. 2013 Jan;113(1):43-53
29. Marriott BP, Cole N, Lee E. National estimates of dietary fructose intake increased
from 1977 to 2004 in the United States. J Nutr. 2009 Jun;139(6):1228S-1235S.
! &+!
30. Park YK, Yetley EA. Intakes and food sources of fructose in the United States. Am J
Clin Nutr. 1993 Nov;58(5 Suppl):737S-747S.
31. Bray GA, Nielsen SJ, Popkin BM. Consumption of high-fructose corn syrup in
beverages may play a role in the epidemic of obesity. Am J Clin Nutr. 2004
Apr;79(4):537-43.
32. Vos MB, Kimmons JE, Gillespie C, Welsh J, Blanck HM. Dietary fructose
consumption among US children and adults: the Third National Health and Nutrition
Examination Survey. Medscape J Med. 2008;10(7):160.
33. United States, Department of Health and Human Services. (1996). Direct food
substances Affirmed as Generally Recognized as Safe; High Fructose Corn Syrup
[docket no. 85N-0548, 21 CFR parts 182 and 184]. Federal Register Vol. 61, No. 165.
Retrieved February 12
th
from www.gpo.gov
34. Livesey G, Taylor R. Fructose consumption and consequences for glycation, plasma
triacylglycerol, and body weight: meta-analyses and meta-regression models of
intervention studies. Am J Clin Nutr. 2008 Nov;88(5):1419-37.
35. Teff KL, Grudziak J, Townsend RR, et al. Endocrine and metabolic effects of
consuming fructose- and glucose-sweetened beverages with meals in obese men and
women: influence of insulin resistance on plasma triglyceride responses. J Clin
Endocrinol Metab. 2009 May;94(5):1562-9.
36. Hudgins LC, Parker TS, Levine DM, Hellerstein MK. A dual sugar challenge test for
lipogenic sensitivity to dietary fructose. J Clin Endocrinol Metab. 2011 Mar;96(3):861-8.
37. Le KA, Ith M, Kreis R, et al. Fructose overconsumption causes dyslipidemia and
ectopic lipid deposition in healthy subjects with and without a family history of type 2
diabetes. Am J Clin Nutr 2009; 89:1760-1765.
38. Stanhope KL, Schwarz JM, Keim NL, et al. Consuming fructose-sweetened, not
glucose-sweetened, beverages increases visceral adiposity and lipids and decreases
insulin sensitivity in overweight/obese humans. J Clin Investig 2009; 119:1322-1334.
39. Kohli R, Kirby M, Xanthakos SA, et al. High-fructose, medium chain trans fat diet
induces liver fibrosis and elevates plasma coenzyme Q9 in a novel murine model of
obesity and nonalcoholic steatohepatitis. Hepatology 2010; 52:934.
40. Goran MI, Walker R, le K-A, et al. Effects of PNPLA3 on Liver Fat in Obese Hispanic
Children and Adolescents Diabetes 2010; 59: 3127-3130.
! &#!
41. Food and Drug Administration. Code of Federal Regulations Title 21. Direct Food
substances Affirmed as Generally Recognized as Safe. Database of Select Committee
on GRAS Substances (SCOGS) Reviews. Report no. 50, 1979.

! &"!
Chapter 4: Adipose tissue inflammation and fibrosis may mediate liver damage and
diabetes risk in overweight and obese Italian children

Abstract
Background: Obesity in childhood is associated with risk of diabetes and liver disease,
which are both linked to an inflammatory state in adipose tissue and liver. However, no
prior study has examined potential associations between the similar pathologies that
occur in adipose tissue inflammation and liver disease or which elements of tissue
damage are associated with type 2 diabetes risk.

Objective: To determine if inflammation and fibrosis in subcutaneous adipose tissue in
children is related to the extent of observed liver disease or type 2 diabetes risk.

Methods: Subcutaneous adipose tissue (SAT) and liver biopsies were collected from 33
Italian children (mean BMI 28.1±5.1 kg/m
2
and mean age 11.6±2.2 years) with confirmed
non-alcoholic fatty liver disease. Histology and immunohistochemistry were conducted to
assess CD68+ macrophages, crown like structures (CLS) and fibrosis in the adipose
tissue, while liver tissue was scored for fibrosis, steatosis, ballooning and inflammation.

Results: The presence of SAT CLS was significantly related to extent of liver fibrosis
(CLS+; 1.7±0.7 vs. CLS-; 1.2±0.7, p=0.04) independent of BMI. High levels of SAT
collagen yielded lower insulin during an OGTT and lower disposition index (r=-0.48,
p=0.006). No other adipose measures correlated with liver disease parameters.

Conclusion: SAT inflammation is associated with the extent of scored liver fibrosis
independent of obesity and SAT collagen may contribute to insulin resistance. SAT
inflammation and fibrosis may contribute to liver health and type 2 diabetes risk in
children and should be studied further.

! &$!
Introduction
Obesity is marked by excess storage of lipids and a chronic pro-inflammatory state in
adipocytes (1). The obesity-associated low-grade inflammation is sustained by the
increase of macrophages in adipose tissue (2) and is strongly associated with increased
ectopic fat accumulation in the liver (3), which can lead to non-alcoholic fatty liver
disease (NAFLD), a process that begins early in life (4). Pediatric obesity prevalence
continues to rise internationally (5) and is a strong risk factor for NAFLD, which is now
the primary cause of liver disease in children (6). Adipose and liver tissues demonstrate
similar yet independent relationships between excess fat deposition and the ensuing
cell-signalling responses and resultant inflammation, however there is very little research
that examines how inflammation in these two tissue sites may be linked. This is
surprising given that the prevalence of NAFLD amongst obese children can be as high
as 70% (7) and even at an early age some children may progress to NASH, marked by
the presence of fibrosis and inflammation (4). Increases in visceral adipose tissue (VAT)
have been linked to metabolic dysregulation (8) and increased ectopic lipid deposition in
the liver (1), the latter which promotes pro-inflammatory mediators (9) that are thought to
activate specific liver-resident macrophages known as Kupffer cells, which may mediate
the progression from NAFLD to NASH (10) by inducing hepatic injury and fibrosis (11).
Similarly, amongst adipocytes, a pro-inflammatory state perturbs normal extracellular
matrix (ECM) remodelling and favors fibrous collagen deposition (12) that can result in
insulin resistance (13) and a diminished capacity for fat mass expansion (14). This
inability of adipose fat mass to expand under fibrotic conditions may force ectopic
deposition of fat into other organs such as the liver (15), which suggests a potential link
between inflammation in adipose and hepatic tissues, yet the specific role of SAT
inflammation in metabolic or liver disease is unclear.

To investigate potential links between obesity-induced adipose inflammation and liver
disease in children, we conducted immunohistochemical analyses on biopsied tissue
from both subcutaneous adipose (SAT) depots and liver. The aim of this analysis was to
determine and characterize the extent of inflammation and fibrosis in the adipose tissue
of overweight and obese children and examine their relationship with liver disease and
type 2 diabetes risk.

! &%!
Methods
Patients:
A total of 40 consecutive patients with biopsy-proven NAFLD seen at Bambino Gesù
Children’s Hospital were prospectively included in the study from January 2011 to
December 2012. The study was approved by the Ethics Committee of the Bambino
Gesù Children’s Hospital and Research Institute, Rome, Italy. Inclusion criteria and
exclusion criteria were the same as previously reported (16) Autoimmune liver disease,
metabolic liver disease, Wilson’s disease, celiac disease, and alpha-1-antitrypsin
deficiency were ruled out using standard clinical, laboratory, and histological criteria. Of
the 40 children enrolled, 33 had complete data for all measures and are the subjects of
the present study.

Anthropometric measures:
Weight and height were measured using standard procedures. Body mass index (BMI)
(kg/m
2
) and its standard deviation score (Z score) were calculated using US reference
data (17, 18). Waist circumference (WC) was measured at the highest point of the iliac
crest.

Laboratory assessment:
Alanine and aspartate transferases, gamma-glutamyl-transpeptidase, total triglycerides,
and total low density (LDL) and high-density lipoprotein (HDL) cholesterol were
evaluated using standard laboratory methods. Plasma insulin was measured using a
radioimmunoassay (Myria Technogenetics, Milan, Italy). All participants underwent a
standard oral glucose tolerance test (OGTT) performed with 1.75 grams of glucose per
kilogram of body weight (up to 75g), and glucose and insulin were measured at 0, 30,
60, 90 and 120 minutes. The degree of insulin sensitivity/resistance was determined via
the homeostatic model assessment (HOMA) (19) and by the OGTT-derived insulin
sensitivity index (ISI) (20). Both the HOMA and the OGTT-derived ISI have a significant
correlation with the gold standard euglycemic hyperinsulinemic glucose clamp
technique. A HOMA value>2 or ISI value<6 were considered an indication of insulin
resistance.

! &&!
Pro-inflammatory markers and adipocytokines:
Serum C-reactive protein (CRP) was determined via a high sensitivity latex agglutination
method on HITACHI 911 Analyser (Sentinel Ch., Milan). The kit had a minimum
detection of less than 0.05 mg/L, and a measurable concentration range up to 160 mg/L.
The intra-assay and inter-assay variation coefficients were, respectively, 0.8–1.3 and
1.0–1.5%. Serum tumor necrosis factor (TNF)-" and interleukin (IL)-6 were measured by
sandwich ELISA (R&D System Europe Ltd, Abingdon, UK). For TNF-", the kit had a
sensitivity of 0.12 pg/mL in a 200-µL sample size and a range of 0.5 to 32 pg/mL. The
intra-and inter-assay coefficients of variation were 5.9% and 12.6%, respectively. For IL-
6, the kit had a sensitivity of 0.25 pg/mL in a 50-µL sample size and a range of 3.9 to
250 ng/mL. The intra- and inter-assay coefficients of variation were 3.4% and 5.8%,
respectively. Serum adiponectin, leptin and retinol binding protein (RBP)-4 were
measured by ELISA kit according to the manufacturer's protocol (Ray Biotech, Norcross,
GA, USA)

Liver histology:
The clinical indication for biopsy was either to assess the presence of NASH and degree
of fibrosis and/or to rule out potential other liver diseases. Liver biopsy was performed in
all children after an overnight fast, using an automatic core biopsy 18 gauge needle
(Biopince, Amedic, Sweden) under general anesthesia and ultrasound guidance. A
Sonoline Omnia ultrasound machine (Siemens, Munich, Germany) equipped with a 5-
MHz probe (5.0 C 50, Siemens) and a biopsy adaptor were employed. The length of liver
specimen was recorded and only samples with a length #15 mm and including at least
5–6 complete portal tracts were considered adequate for the purpose of the study.
Biopsies were routinely processed (ie, formalin-fixed and paraffin-embedded) and
sections of liver tissue were stained with hematoxylin-eosin, Van Gieson, Periodic acid-
Schiff diastase and Prussian blue stain. Biopsies were evaluated by a single
hepatopathologist who was blinded to clinical and laboratory data. Steatosis,
inflammation, hepatocyte ballooning and fibrosis were scored using the NAFLD Clinical
Research Network (CRN) criteria (21): Briefly, steatosis was graded on a 4-point scale:
grade 0 = steatosis involving <5% of hepatocytes; grade 1 = steatosis involving up to
33% of hepatocytes; grade 2 = steatosis involving 33–66% of hepatocytes; and grade 3
= steatosis involving >66% of hepatocytes. Lobular inflammation was graded on a 4-
point scale: grade 0 = no foci; grade 1 = <2 foci per 200$field; grade 2 = 2–4 foci per
! &'!
200$ field; and grade 3 = >4 foci per 200$ field. Hepatocyte ballooning was graded from
0 to 2: 0 = none; 1= few balloon cells; and 2 =many/prominent balloon cells. The stage
of fibrosis was quantified using a 5-point scale: stage 0 = no fibrosis; stage 1 =
perisinusoidal or periportal (1a = mild, zone 3, perisinusoidal; 1b = moderate, zone 3,
perisinusoidal; 1c = portal/periportal); stage 2 =perisinusoidal and portal/periportal; stage
3 = bringing; and stage 4 = cirrhosis.
Features of steatosis, lobular inflammation, and hepatocyte ballooning were combined to
obtain the NAFLD activity score (NAS). As recently recommended by the NASH Clinical
Research Network (21), a microscopic diagnosis based on overall injury pattern
(steatosis, hepatocyte ballooning, inflammation) as well as the presence of additional
lesions (e.g. zonality of lesions, portal inflammation and fibrosis) was assigned to each
case. Accordingly, biopsies were subdivided into: not-NASH and definite NASH
subcategories (22).

Adipose Histology:
Following liver biopsy, an abdominal subcutaneous adipose tissue biopsy was
performed in all children after an overnight fast, using an automatic core biopsy 18
gauge needle (Biopince, Amedic, Sweden). Biopsies were routinely processed (ie,
formalin-fixed and paraffin-embedded) and sections of adipose tissue were stained with
hematoxylin-eosin and CD68 antibody (Leica Biosystems, Newcastle, UK) or picrosirius
red. The CD68 antibody incubation time was 15 min and antigen retrieval was at PH 8
for 20 minutes on the Leica bond. Biopsies were evaluated by a single technician who
was blinded to clinical and laboratory data. For the CD68 stained sections, 4 consecutive
5-micron sections were obtained for each subject. Two independent fields at 20x
magnification were captured for each section, for a total of 8 images per subject.
Adipose cell size (micron
2
), adipose cell count, isolated macrophages and crown like
structure (CLS) counts were obtained for each field captured using Fiji quantitative
microscopy software (23). For each field, isolated macrophage per cm
2
and CLS per cm
2

were calculated by dividing by the area of the field measured. For each subject, the
mean values for adipose cell size, macrophages per cm
2
and CLS cm
2
were obtained.
For the collagen analysis, 3 non-serial 5 micron sections were obtained for each subject
and stained with picrosirius red and prepared according to Bedossa et al. (24). Two
independent fields at 20x magnification were captured for each section. Fiji software was
used to obtain total collagen area and a collagen to adipose ratio.
! &(!
Data analysis:
Values are expressed mean ± standard deviation. Variables were assessed for normality
and non-normal data was log (ln) transformed for analyses. Wilcoxon test was used to
compare between-group differences. Non-parametric Spearman’s correlations were
used to examine correlations and partial correlations were conducted controlling for age,
gender and BMI as covariates. Statistical significance was set at p<0.05. All analyses
were performed on SPSS v18 (SPSS Inc, Chicago IL, USA).

Results
Anthropometric associations with SAT and liver tissue measures:
A total of 33 children participated in the study. Anthropometric characteristics are
reported in Table 1, together with a panel of adipocytokines and additional subject
descriptors. Mean BMI was 28.1±5.1 kg/m
2
with a mean BMI Z-score of 2.3±0.76 (lean,
n=7; overweight, n=16, obese, n=10). Twenty of the 33 children were male and the
mean age was 11.6±2.2 years.

Obese children had a higher mean adipocyte area than the lean participants
(1924.2±364.9 cm
2
vs. 2742±800.8 cm
2
; p=0.03, respectively) and quantity of CLS per
cm
2
was positively correlated with BMI and WC (r=0.43, p=0.01 and r=0.41, p=0.01,
respectively). There were no significant differences in insulin and glucose parameters by
obesity status.

Scores for liver steatosis (1.8±0.6 vs. 2.2±0.7), ballooning (1.1±0.6 vs. 1.5±0.6) and NAS
(4.4±1.5 vs. 5.2±1.3) were lower in lean vs. overweight/obese children, but these trends
did not reach significance. Measures of liver disease were not associated with any
glucose or insulin parameters.

SAT morphology, liver disease and relationships between fat and liver inflammation:
Of the 33 participants, 14 had one or more CLS present in SAT sections (Table 2).
Microscopy identified clear clinical differences in adipocyte cell size, CLS presence and
fibrosis by obesity classification (Figure 1). Adipose cell area was significantly positively
associated quantity of CLS/cm
2
(r=0.4, p=0.01). Obese participants exhibited less SAT
collagen than their lean counterparts, however the correlation did not reach statistical
significance (r=-0.3, p=0.11).
! &)!

NASH was indicated by a NAS score of " 5 in over 50% of the study participants (Table
2) and histology confirmed advanced steatosis and fibrosis (Figure 2). No study
measures were significantly different when means were compared by NAS category (0-4
and " 5). Serum pro-inflammatory cytokines (CRP, IL-6 and TNF-") were not related to
adipose inflammation and fibrosis or NAFLD features in this cohort.

The presence of SAT CLS demonstrated significantly higher liver fibrosis scores (CLS+;
1.7±0.7 vs. CLS-; 1.2±0.7, p=0.04), independent of BMI. Similarly, CLS+ participants
had a tendency for greater hepatocyte ballooning (1.4±0.7 vs. 1.0±0.5, p=0.07)
compared to participants with no CLS. There were no other statistically significant
relationships between indicators of SAT inflammation or fibrosis by NASH status.

SAT morphology, liver disease and glucose and insulin measures:
Adipose collagen ratio was significantly inversely correlated with insulinogenic index (r=-
0.37, p=0.03) and disposition index (DI) (r=-0.48, p=0.006), but not with HOMA or fasting
plasma insulin or glucose. The relationship between collagen and DI was maintained
(r=-0.31, p=0.05) after adjusting for age, gender and BMI (Figure 3). To interrogate the
relationship between SAT collagen and DI further, the study population was
dichotomized based on values > (high fibrosis) or < (low fibrosis) the mean collagen ratio
of 21.1. The High fibrosis group demonstrated a marginally higher glucose AUC with
lower insulin AUC in response to an OGTT, however these differences did not reach
statistical significance (Figure 3B). Additionally, the high fibrosis group had a lower
mean Matsuda Index (3.4±0.5 vs. 4.3±0.09) and Insulinogenic Index (1.4±0.04 vs.
4.1±1.9) than the low fibrosis group. Serum adiponectin was not related to any aspects
of adiposity or SAT inflammation, however mean serum adiponectin significantly differed
between steatosis scores (p
trend
<0.001) (Figure 4) and displayed a highly significant
negative correlation with NAS score (r=-0.62, p<0.001, respectively) even when
controlling for BMI z-score and gender (-0.31, p=0.05).

Discussion
! '*!
In a study of 33 Italian children with biopsy-proven NAFLD, our results indicate a
significant relationship between SAT CLS presence and liver fibrosis. Additionally, SAT
collagen was significantly associated with impaired DI, which was supported by the
higher glucose AUC and lower insulin AUC values observed in the high fibrosis group.
Obesity was significantly associated with SAT inflammation and leaner children tended
to have lower qualitative liver disease scores than obese counterparts. These data
suggest that increasing obesity is strongly related to adipose tissue damage and liver
damage with some evidence of a link between CLS presence in SAT and liver disease.

As mentioned above, nearly half of the children in the present study had detectable CLS
in their SAT and the number of CLS was positively related to obesity status and adipose
cell size. This is consistent with another study (14) in which the number of CLS
increases with adipocyte hypertrophy, indicating greater adipocyte cell death and a
heightened inflammatory state. However, a recent study in children did not detect CLS in
the abdominal adipose tissue of healthy, overweight children (25). Tordjman et al. (26)
recently demonstrated that there are no observable differences in macrophage number
or CD68 gene expression in SAT by NASH status in adults, however they were able to
show highly significant differences by liver disease status in both of these measures
when examining deep SAT or VAT. Increased delivery of pro-inflammatory factors from
macrophages to the liver along with free fatty acid flux may be one possible link between
VAT and liver disease (27), but this observation may not extend to SAT depots (26).
Given that CLS+ subjects had significantly greater liver fibrosis scores, our data
suggests that SAT inflammation may play a role in modulating inflammation in the liver.
In a study composed of obese adults, subjects with CLS in their SAT had elevated liver
fat, as measured by proton magnetic resonance spectroscopy (28), and similar to our
study there was no association between adipocyte cell size and steatosis. It is entirely
possible that quantitative measures of liver disease, such as liver fat fraction, would
demonstrate a stronger relationship to inflammatory SAT measures in children and other
adipose regions may be linked to liver disease, however our study was not able to test
these hypotheses directly as the measures were not available.

Obese participants in our study showed less SAT fibrosis than lean counterparts which
is supported, in part, by a recent study in adults in which obese subjects had a higher
degree of pericellular adipose fibrosis in both SAT and omental adipose tissue than lean
! '+!
participants (29), without a significant difference in total fibrosis by obesity status.
Similarly, in healthy children, obese individuals have significantly less collagen that lean
peers (25) and this collagen was inversely associated with both adipose cell size and the
presence of macrophages in adipose tissue with collagen decreasing as a function of
age. Our findings support the relationships between collagen and adipose measures, yet
age did not mediate these associations in our population, perhaps because our study
included slightly older children. SAT collagen also exhibited an inverse relationship with
disposition index, which is the ability to compensate for insulin resistance through
increased insulin secretion, in our study. Fibrosis in adipose tissue has been previously
associated with insulin resistance (30) and although SAT fibrosis did not correlate
significantly with any other measures of glucose and insulin dynamics in our population,
an observation also noted in adults (29), those with greater fibrosis had a higher glucose
AUC and lower insulin AUC in response to an OGTT. Higher collagen deposition in
adipose tissue is thought to inhibit adipose expansion, which points to a potential
relationship between adipose extracellular matrix (ECM) and inflammation (25). Our
findings suggest that higher adipose fibrosis may impair glucose and insulin dynamics
supporting the proposed role of adipose fibrosis in increasing inflammation and thereby
insulin resistance and risk for type 2 diabetes (13). Collagen is also strongly associated
with BMI and VAT content (13) and limits the ability to store lipids in SAT, which was
observed in our population, supporting the role of fibrosis as a key factor in regulating
metabolic health. Our finding show higher SAT collagen in the leaner participants, thus it
remains possible that the relationships between specific adipose depots and glucose
homeostasis are very different in the developing adipose tissues of children than that of
adults.

The adipokine adiponectin has a protective role in adipose inflammation by reducing pro-
inflammatory cytokine expression (31, 32) while also regulating the oxidation of fatty
acids in the liver and protecting against steatosis (33, 34). We demonstrate no
association of adiponectin with measures of inflammation in adipose tissue or plasma
pro-inflammatory cytokines. However, we did observe a trend towards decreased
adiponectin with increasing SAT collagen. Additionally, qualitative measures of liver fat
increased as adiponectin levels decreased, supporting the protective role of adiponectin
against NAFLD even in children. During childhood, adipose tissue remodelling is high
and our data suggests that SAT collagen production may perturb adiponectin production
! '#!
and hinder the protective benefits of adiponectin on hepatic steatosis. Further studies
are warranted to determine the full extent to which inflammation and fibrosis in other
adipose depots impact liver health.

We were able to detect significant relationships among measures within adipose and
liver tissue, but our relatively small sample size likely prevented us from finding stronger
associations between adipose and liver inflammation. Furthermore, these relationships
may be very different based on age and ethnicity. Additionally, relative SAT and VAT
amounts were not available for our population and our collagen analysis was restricted
to total rather than subtypes of collagen, which limits the application of our findings to
relationships between SAT and total collagen with liver disease. Recently, diet-induced
obesity had demonstrated that macrophages and the expression of pro-fibrogenic matrix
metalloproteinases (MMPs) were increased in VAT, but not in liver tissue, suggesting a
differential relationship between adiposity and inflammation in SAT and liver (35). Our
data supports this notion, as only SAT CLS was associated with liver disease, but future
work should include the measure of MMPs. Higher glucose and lower insulin AUCs in
the high-fibrosis group is suggestive of collagen contributing to insulin resistance,
however our measures of insulin resistance were based on data from OGTT rather than
intravenous glucose tolerance tests or clamp methods. Additionally, we did not assay c-
peptide, which could be used to further verify insulin resistance, beyond DI, and test for
associations with SAT collagen.

Obesity leads to pro-inflammatory conditions in the adipose tissue promoting fibrotic
conditions and preventing normal adipose tissue expansion. This would presumably
promote hepatic steatosis and subsequent fibrosis. In biopsied tissue from overweight
and obese children we observed an abundance of inflammatory markers in the SAT,
including CLS and fibrosis, and independent relationships between BMI and SAT as well
as BMI and liver scores were detected. We were able to observe a positive association
between CLS status and liver fibrosis, suggesting a role for SAT in the mediation of liver
inflammation. Additionally, higher levels of SAT collagen appear to effect glucose
homeostasis indicating that SAT inflammation and fibrosis could be risk factors for
insulin resistance and type 2 diabetes. It is likely that SAT inflammation in children may
not impact all aspects of liver disease. However, given the relationship between SAT
! '"!
CLS and liver fibrosis in this population, the role of adipose inflammation on liver health
in children should be investigated further.

! '$!
Table 4-1. Subject Characteristics
Population Characteristics Mean (SD)
N 33
Age (yrs) 11.6 (2.2)
Gender (M/F) 20/13
BMI (kg/m
2
) 28.1 (5.1)
BMI z-score 2.3 (.76)
WC (cm) 89.3 (8.5)
Biochemical Measures Mean (SD)
Fasting Plasma Glucose (mg/dL) 80.2 (13)
Fasting Plasma Insulin (mg/dL) 16.7 (11.2)
Matsuda Index 3.9 (3.1)
HOMA-IR 3.3 (2.3)
Insulinogenic Index (IGI) 2.9 (6.4)
Disposition Index (IGI*ISI) 9 (16.5)
Cholesterol (mg/dL) 161.2 (26.7)
HDL (mg/dL) 42.5 (7.5)
LDL (mg/dL) 101.2 (15.6)
Triglyceride (mg/dL) 111.3 (44.1)
AST (IU/L) 29.2 (8.8)
ALT (IU/L) 34.4 (15.6)
GGT (ng/dL) 20.6 (9.2)
Adiponectin (ng/dL) 21.9 (2.3)
CRP (ng/dL) 1.6 (.4)
IL-6 (ng/dL) 9.5 (4.5)
TNF-" (ng/dL) 6.5 (1.9)

! '%!
Table 4-2. Adipose and Liver Immunohistological Measures
Adipose Histology (n=33) Mean (SD)
Macrophage count 4.9 (3.6)
Macrophages per cm
2
0.51 (0.35)
CLS (-/+) 19/14
CLS per cm
2
(among CLS+) 0.56 (0.42)
Adipose Cell Area (µ
2
) 2267.1 (704.4)
Collagen Ratio 21.1 (14.5)
Liver Histology (n=33) Frequency(%)
Fibrosis 0-1 17 (60.6)
2-3 13 (39.4)
Steatosis 0-1 25 (75.7)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!"2
8 (24.2)
Lobular Inflammation 0-1 15 (44.5)
2 18 (54.5)
Portal Inflammation 0-1 19 (57.5)
2 14 (42.4)
Ballooning 0-1 23 (69.7)
2 10 (30.3)
NAS 0-2 1 (3)
3-4 15 (45.5)
"5 (NASH)
17 (51.5)

! "#!
Figure 4-1. Adipose cell size, CLS presence and collagen ratio as a function of obesity status

Legend: Histological analysis of SAT adipose sections (Panel A) at 20x magnification demonstrated a significant increase in
adipocyte cell area (µ
2
) between lean and obese participants (Panel B). CD68 immunihistochemical staining for macrophages
identified CLS (arrow) in 14 of 33 participants (Panel C) and a clinical increase in CLS per cm
2
by obesity was observed (Panel D).
Picrosirius red staining was used to assess fibrosis (arrow) in SAT through obtaining a collagen ratio (Panel E). A trend in decreasing
collagen ratio as a function of obesity was observed (Panel F). CLS=crown-like structures, SAT= subcutaneous adipose tissue.
! ""!
Figure 4-2. Liver histology: steatosis, fibrosis and inflammation

Legend: Panel A (10x magnification), is a representative image of NASH with steatosis,
and fibrosis score F3. Panel B Represents a child with NASH (steatosis grade 2,
ballooning grade 1, and inflammation grade 1. Panel C depicts steatosis score 2 and
hepatic inflammation as evidenced by CD163+ Kupfer cell staining.

! "#!
Figure 4-3A. Mean adipose collagen ratio impacts disposition index

Legend: Mean adipose collagen ratio impacts disposition index
Mean adipose collagen ratio is displayed on the x-axis and mean disposition index is on
the y-axis. Disposition index decreased as a function of increasing adipose fibrosis.
Graphical data are presented as raw, unadjusted values and the correlation (-0.31,
p=0.05) persisted after controlling for age, gender and BMI.

! "$!
Figure 4-3B. Glucose and insulin dynamics are affected by SAT fibrosis

Legend: Glucose and insulin dynamics are affected by SAT fibrosis
The study participants were dichotomized by SAT collagen amount into high (!21.1;
solid line) and low (<21.1; dashed line) fibrosis. Glucose and insulin AUCs in response to
an OGTT were compared. The high fibrosis group had a higher glucose AUC with lower
insulin AUC.

! #%!
Figure 4-4. Serum adiponectin decreases as a function of steatosis score

Legend: Mean adiponectin decreases as a function of increasing steatosis.
Bars represent mean adiponectin (ng/dL) by steatosis graded score. Adiponectin
significantly decreased as steatosis score climbed (p
trend
<0.001). Error bars represent
95% confidence interval.

! #&!
Chapter 4 References:
1. Lumeng, CN, Saltiel, AR. Inflammatory links between obesity and metabolic disease.
J Clin Invest. 2011 June 1; 121(6): 2111–2117.
2. Chen Y, et al. Variations in DNA elucidate molecular networks that cause disease.
Nature. 2008; 452(7186):429–435.
3. Suganami T, Tanaka M, Ogawa Y. Adipose tissue inflammation and ectopic lipid
accumulation. Endocr J. 2012;59(10):849-57. Epub 2012 Aug 9.
4. Angulo P. Nonalcoholic fatty liver disease. N Engl J Med. 2002 Apr 18;346(16):1221-
31.
5. Finucane MM, Stevens GA, Cowan MJ et al. National, regional, and global trends in
body-mass index since 1980: systematic analysis of health examination surveys and
epidemiological studies with 960 country-years and 9.1 million participants. Lancet 2011.
377, 557–567.
6. Feldstein AE, Charatcharoenwitthaya P, Treeprasertsuk S, Benson JT, Enders FB,
Angulo P. The natural history of non-alcoholic fatty liver disease in children: a follow-up
study for up to 20 years.Gut. 2009 Nov;58(11):1538-44.
7. Dunn W, Schwimmer JB. The obesity epidemic and nonalcoholic fatty liver disease in
children.Curr Gastroenterol Rep. 2008 Feb;10(1):67-72.
8. Kursawe R, et al. Cellularity and adipogenic profile of the abdominal subcutaneous
adipose tis- sue from obese adolescents: association with insulin resistance and hepatic
steatosis. Diabetes. 2010;59(9):2288–2296.
9. Ekstedt M, Franzén LE, Mathiesen UL, Thorelius L, Holmqvist M, Bodemar G,
Kechagias S. Long-term follow-up of patients with NAFLD and elevated liver enzymes.
Hepatology. 2006 Oct;44(4):865-73.
10. Alisi A, Manco M, Vania A, Nobili V. Pediatric nonalcoholic fatty liver disease in
2009.J Pediatr. 2009 Oct;155(4):469-74.
11. Spruss A, Kanuri G, Wagnerberger S, Haub S, Bischoff SC, Bergheim I. Toll-like
receptor 4 is involved in the development of fructose-induced hepatic steatosis in
mice.Hepatology. 2009 Oct;50(4):1094-104.
12. Nakajima I, Muroya S, Tanabe R, Chikuni K. Extracellular matrix development during
differentiation into adipocytes with a unique increase in type V and VI collagen. Biol Cell.
2002 Jun;94(3):197-203.
! #'!
13. Pasarica M, Gowronska-Kozak B, Burk D, Remedios I, Hymel D, Gimble J, Ravussin
E, Bray GA, Smith SR. Adipose tissue collagen VI in obesity. J Clin Endocrinol Metab.
2009 Dec;94(12):5155-62.
14. Cinti, S., G. Mitchell, G. Barbatelli, I. Murano, E. Ceresi, E. Faloia, S. Wang, M.
Fortier, A. S. Greenberg, and M. S. Obin. 2005. Adipocyte death defines macrophage
localization and function in adipose tissue of obese mice and humans. J. Lipid Res.
46:2347–2355.
15. Duval C, Thissen U, Keshtkar S, Accart B, Stienstra R, Boekschoten MV, Roskams
T, Kersten S, Müller M. Adipose tissue dysfunction signals progression of hepatic
steatosis towards nonalcoholic steatohepatitis in C57BL/6 mice. Diabetes. 2010
Dec;59(12):3181-91.
16. Alkhouri N, De Vito R, Alisi A, Yerian L, Lopez R, Feldstein AE, Nobili V.
Development and validation of a new histological score for pediatric non-alcoholic fatty
liver disease.J Hepatol. 2012 Dec;57(6):1312-8. doi: 10.1016/j.jhep.2012.07.027. Epub
2012 Aug 4.
17. Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, et al. CDC growth charts: United
States. Adv Data 2000:1–27
18. Cole TJ, Bellizzi MC, Flegal KM, et al. Establishing a standard definition for child
overweight and obesity worldwide: international survey. BMJ 2000;320:1240–1243
19. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner
RC. Homeostasis model assessment: insulin resistance and beta-cell function
from fasting plasma glucose and insulin concentrations in man. Diabetologia
1985;28:412-419
20. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose
tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care
1999;22:1462–1470
21. Kleiner DE, Brunt EM, Van Natta M, Behling C, Contos MJ, Cummings OW, Ferrell
LD, Liu YC, Torbenson MS, Unalp-Arida A, Yeh M, McCullough AJ, Sanyal AJ;
Nonalcoholic Steatohepatitis Clinical Research Network. Design and validation of a
histological scoring system for nonalcoholic fatty liver disease. Hepatology. 2005;41:
1313-21
22. Brunt EM, Kleiner DE, Wilson LA, Belt P, Neuschwander-Tetri BA. Nonalcoholic
fatty liver disease (NAFLD) activity score and the histopathologic diagnosis in NAFLD:
distinct clinicopathologic meanings. Hepatology 2011;53:810-820.
! #(!
23. Johannes Schindelin, Ignacio Arganda-Carreras, Erwin Frise, Verena Kaynig, Mark
Longair, Tobias Pietzsch, Stephan Preibisch, Curtis Rueden, Stephan Saalfeld,
Benjamin Schmid, Jean-Yves Tinevez, Daniel James White, Volker Hartenstein, Kevin
Eliceiri, Pavel Tomancak and Albert Cardona (2012) Fiji: an open-source platform for
biological-image analysis, Nature Methods 9(7): 676-682
24. Bedossa P, Lemaigre G, Bacci J, Martin E: Quantitative estimation of the collagen
content in normal and pathologic pancreas tissue. Digestion 1989, 44:7-13.
25. Tam CS, Tordjman J, Divoux A, Baur LA, Clément K. Adipose tissue remodeling in
children: the link between collagen deposition and age-related adipocyte growth. J Clin
Endocrinol Metab. 2012 Apr;97(4):1320-7.
26. Tordjman J, Divoux A, Prifti E, Poitou C, Pelloux V, Hugol D, Basdevant A, Bouillot
JL, Chevallier JM, Bedossa P, Guerre-Millo M, Clement K. Structural and inflammatory
heterogeneity in subcutaneous adipose tissue: relation with liver histopathology in
morbid obesity. J Hepatol. 2012 May;56(5):1152-8.
27. Fontana L, Eagon JC, Trujillo ME, Scherer PE, Klein S. Visceral fat adipokine
secretion is associated with systemic inflammation in obese humans. Diabetes. 2007
Apr;56(4):1010-3.
28. Jansen HJ, Vervoort GM, van der Graaf M, Stienstra R, Tack CJ. Liver fat content is
linked to inflammatory changes in subcutaneous adipose tissue in type 2 diabetes
patients. Clin Endocrinol (Oxf). 2012 Nov 20. doi: 10.1111/cen.12105. [Epub ahead of
print]
29. Divoux A, Tordjman J, Lacasa D, Veyrie N, Hugol D, Aissat A, Basdevant A, Guerre-
Millo M, Poitou C, Zucker JD, Bedossa P, Clément K. Fibrosis in human adipose tissue:
composition, distribution, and link with lipid metabolism and fat mass loss. Diabetes.
2010 Nov;59(11):2817-25.
30. Spencer M, Yao-Borengasser A, Unal R, Rasouli N, Gurley CM, Zhu B, Peterson
CA, Kern PA. Adipose tissue macrophages in insulin-resistant subjects are associated
with collagen VI and fibrosis and demonstrate alternative activation. Am J Physiol
Endocrinol Metab. 2010 Dec;299(6):E1016-27.
31. Buechler C, Wanninger J, Neumeier M. Adiponectin, a key adipokine in obesity
related liver diseases. World J Gastroenterol. 2011 Jun 21;17(23):2801-11.
32. Moschen AR, Wieser V, Tilg H. Adiponectin: key player in the adipose tissue-liver
crosstalk. Curr Med Chem. 2012;19(32):5467-73.
! #)!
33. Li L, Wu LL. Adiponectin and interleukin-6 in inflammation-associated disease. Vitam
Horm. 2012;90:375-95.
34. Finelli C, Tarantino G. What is the role of adiponectin in obesity related non-alcoholic
fatty liver disease? World J Gastroenterol. 2013 Feb 14;19(6):802-12.
35. Kang HS, Liao G, DeGraff LM, Gerrish K, Bortner CD, Garantziotis S, Jetten AM.
CD44 plays a critical role in regulating diet-induced adipose inflammation, hepatic
steatosis, and insulin resistance. PLoS One. 2013;8(3):e58417.

! #*!
Chapter 5: High rates of fructose malabsorption are associated with reduced liver fat in
obese African Americans.

Abstract
Objective: African Americans commonly have lower liver fat accumulation than
Hispanics, despite a similar propensity for obesity. Both ethnicities exhibit high
consumption of fructose-containing beverages, which has been associated with high
liver fat due to the lipogenic properties of fructose. Therefore, differences in fructose
absorption may be an important factor in regulating liver fat deposition. We hypothesized
that fructose malabsorption in African Americans may reduce hepatic delivery of
fructose, thus contributing to lower liver fat deposition compared to Hispanics.
Methods: Thirty-seven obese young adults aged 21.4±2.1 years (16 African
American/21 Hispanic) underwent a 3-hour hydrogen (H
2
) breath test to assess fructose
malabsorption. MRI was used to determine visceral (VAT) and subcutaneous adipose
tissue (SAT) volume and liver fat. Fructose malabsorption was expressed as an area
under the curve for H
2
production (H
2
AUC).
Results: Compared to Hispanics, African Americans had lower liver fat (5.4±5.0% vs.
8.9±2.3%, p=0.02) and had a higher prevalence of fructose malabsorption (75.0% vs.
42.9%; p=0.05) Liver fat was negatively related to the extent of fructose malabsorption in
African Americans (r=-0.53, p=0.03), and this relationship was independent of total fat,
SAT and VAT. There were no significant relationships between liver fat and fructose
malabsorption in Hispanics.
Conclusion: African Americans have both a higher prevalence and a greater magnitude
of fructose malabsorption than Hispanics. In African Americans, fructose malabsorption
was negatively correlated with liver fat, which may be protective against fatty liver
disease.

! #+!
Introduction:
The national and worldwide increase in the prevalence of obesity has been mirrored by
elevated dietary fructose intake

[1]. Elevated fructose and added sugar intake has been
implicated in enhancing liver fat accumulation, which can lead to the eventual
development of non-alcoholic fatty liver disease (NAFLD)

[2-4]. Currently NAFLD affects
20%-30% of the adult population

[4], and varies in incidence among populations of
different ethnicities. In the United States, the prevalence of NAFLD is higher among
Hispanics than those of African descent

[5-7], despite similar propensities for obesity and
insulin resistance, which are the main predictors of NAFLD.

The increased intake of dietary fructose, either in the form of high fructose corn syrup
(HFCS) or sucrose in sugar sweetened beverages

[8] and other food products, may be
associated with the development of visceral adiposity

[9] and an increase in hepatic de
novo lipogenesis (DNL)

[10]. Fructose is a naturally occurring monosaccharide absorbed
by the small intestine and converted to glucose in hepatocytes. Fructose absorption
occurs at the enterocyte brush border via two facilitative transporters; GLUT5, a
fructose-specific transporter, and GLUT2 a non-specific transporter. Uptake is
modulated by the luminal concentration of fructose and is physiologically limited.
Excessive fructose consumption may overwhelm the ability of the fructose-specific
hexose transporters to absorb the monosaccharide resulting in malabsorption

[11]. The
relatively lower load of fructose entering the portal circulation, due to malabsorption, may
limit the availability of fructose as a substrate for hepatic DNL and contribute as a
protective factor against the development of NAFLD.

The aim of this study was to investigate the frequency of fructose malabsorption among
obese Hispanics and African American young adults. We hypothesized that fructose
malabsorption would be negatively associated with liver fat accumulation and that this
relationship may explain, in part, some of the ethnic difference in liver fat between
Hispanics and African Americans.

Methods:
Participants
This cross-sectional analysis of 37 obese (BMI of 30-45 kg/m
2
), African American and
Hispanic participants (43% AA, 59% female), aged 18-25 years, was conducted based
! #"!
on data collected from February to November, 2009. Participants were obtained based
upon previously published recruiting methods [12]. Participants were defined as African
American or Hispanic if they reported both parents and all four grandparents as African
American or Hispanic, respectively. All participants had medical and family history
screening to ensure eligibility. Participants were excluded from the study if medications
known to influence body composition, insulin action/secretion and/or hepatic fat content
were indicated or if any current or past history of diseases affecting body composition,
fat distribution and/or liver fibrosis and steatosis were reported. Participants were not
eligible for the study if they had current or past involvement with any weight
loss/exercise/sports program in the six months prior to participation. If participants
currently smoked or had smoked greater than 100 cigarettes in their lifetime or if
participants consumed alcohol in excess of 2 drinks per week, they were excluded from
participation. Informed written consent was obtained from all patients and the study was
approved by the Institutional Review Board of the University of Southern California,
Health Sciences Campus.
Medical History, Physical Examination, Anthropometry and Fasting Glucose.
A physician, physician assistant or nurse practitioner completed a thorough family
history and physical examination. Height and weight were measured in triplicate for all
participants using a beam medical scale and wall-mounted stadiometer, and BMI was
calculated based upon the following formula: kg/m
2
. Fasting glucose was screened at
the first outpatient visit, via a frequently sampled intravenous glucose tolerance test
(IVGTT), as previously described [13]. If fasting blood glucose measured !120 mg/dl, an
additional sample was drawn to confirm diagnosis of diabetes. Diagnosis of diabetes
resulted in exclusion from the study and referral to endocrinologist.
Fructose Tolerance/Hydrogen Breath Test
Following an overnight fast, participants were given a 50 g fructose diluted into a 250mL
water solution. Breath hydrogen (H
2
) was expressed in parts per million (ppm) from
breath exhaled into foil pouches every 30 minutes for a total of 180 minutes [14, 15].
Breath hydrogen analysis was performed using a Quintron Microlyzer (Quintron,
Milwaukee WI). A positive fructose malabsorption test was defined as a sustained
increase in exhaled H
2
of greater than or equal to 20ppm over baseline or three or more
successive incremental increases in H
2
of greater than or equal to 5ppm over the
! ##!
preceding value

[14-16]. Overall breath hydrogen production, as measured during the
180-minute procedure, was expressed as an area under the curve (H
2
AUC).
Body Composition and Liver Fat
Whole body fat and soft lean tissue mass was estimated by dual energy x-ray
absorptiometry (DXA) using a Hologic QDR 4500W (Hologic, Bedford, MA). Magnetic
resonance imaging, on a General Electric 1.5-Tesla magnet (GE Healthcare, Waukesha,
WI) was used to assess abdominal fat distribution and hepatic fat fraction using a
modification of the Dixon 3-point technique as previously described

[17,18].
Calculations and Statistical Procedures
Independent t-tests and Analysis of Variance (ANOVA) were used to detect between
group differences in anthropometrics, adiposity measures and breath test data. The
hepatic fat fraction and H
2
AUC variables exhibited non-normal distribution therefore log
transformed values were used for all statistical analyses. Partial correlations were used
to detect significant relationships between breath test and adiposity measures using
VAT, SAT, percent body fat and total fat as covariates. The incremental area under the
curve (iAUC) for breath hydrogen was calculated via the Wolever method [19]. All
statistical analyses were performed using SPSS version 13.0, (SPSS, Chicago IL) and
significance levels were set at p"0.05.

Results:
African Americans and Hispanics demonstrated similar characteristics, with the
exception of visceral adipose tissue (1.9±1.3 vs. 3.7±1.3 L, p<0.001) and liver fat
(5.4±5.0 vs. 8.9±2.3 %, p=0.02), which were both significantly higher in Hispanics (Table
5-1). The prevalence of fructose malabsorption was higher in African Americans
compared to Hispanics (75% vs. 43%, p=0.05). African Americans exhibited significantly
higher mean H
2
production than Hispanics at both 150 and 180 minutes (p=0.01 and
p<0.05, respectively) and a trend towards higher mean H
2
AUC than Hispanics
(116.2±101.5 ppm vs. 78.5±67.5 ppm, p=0.08) (Figure 5-1a). Liver fat was negatively
related to fructose malabsorption (H
2
AUC) in African Americans (r=-0.53, p=0.03),
(Figure 5-1b) and this relationship was independent of SAT (r=-0.56, p=0.01), VAT (r=-
0.46, p=0.04) and total body fat (r=-0.48, p=0.03). Liver fat was negatively related to H
2

production at 180 minutes in African Americans (r=-0.628, p=0.03), and this relationship
was also independent of SAT (r=-0.57, p=0.03) and total body fat (r=-0.62, p=0.02), but
! #$!
not VAT (Figure 5-1c). H
2
AUC displayed no significant correlations with any other
measures of adiposity in African Americans. H
2
production at 180 minutes was modestly
correlated with VAT (r=-0.05, p=0.04) in African Americans and there were no
relationships between liver fat and H
2
AUC or mean H
2
at 180 minutes and measures of
adiposity in Hispanics. The iAUC from 150-180 minutes was significantly higher in
African Americans than in Hispanics (915 vs. 146, respectively; p=0.006) (Figure 5-2).
There was no significant difference in overall iAUC values between the two groups and
no iAUC measures were significantly associated with liver fat.

Discussion:
The main finding of this study is that African Americans show higher rates of fructose
malabsorption over a 180-minute hydrogen breath test compared to Hispanics. The
extent of the elevated level of malabsorption is significantly related to lower mean
percent liver fat in African Americans, independent of other indices of fat distribution.

Fructose malabsorption in the adult population is a condition that has been
predominately reported in the context of other gastro-intestinal disorders. In healthy
Caucasian participants up to 40% of a Dutch population demonstrated malabsorption of
a 25g fructose load, via breath hydrogen testing

[20]. Another study with 89 patients,
who ingested a 50g fructose load, reported a 70% malabsorption rate

[21]. Such
discrepancies may be attributed to the various fructose loads used during the breath
test. Indeed, it was reported that malabsorption prevalence increases as a function of
increasing fructose dosage

[14], however this observation does not affect the results, as
both groups received identical fructose dosage. Further variability in malabsorption rates
has also been shown with the administration of the same standard 50g fructose load,
with prevalence in normal, healthy male and female volunteers ranging from 38%-80%

[22,23]. The data from this study demonstrates that 75% of African Americans and 43%
of Hispanics are malabsorbing fructose when challenged with a 50g fructose load.
Furthermore, the magnitude of fructose malabsorption is higher in African Americans
than Hispanics, as indicated by overall and incremental AUC (at 180 minutes) for
intestinal hydrogen production. Participants in the study did not report antibiotic use (a
potential mediator of intestinal bacteria) or a history of lactose intolerance, which has the
potential to influence intestinal hydrogen production.

! $%!
Upon ingestion, fructose is transported across the intestinal mucosa in a transporter-
mediated facilitative diffusion process primarily by the fructose-specific hexose
transporter, GLUT5. Fructose is then metabolized into triose-phosphates in the liver,
which can be: a) directly oxidized, b) contribute to glycogen and glucose synthesis via
gluconeogenesis or c) promote triglyceride synthesis and liver fat accumulation via
stimulation of the DNL pathway. Fructose metabolism is not rate-limited by the
availability of phosphofructokinase, as is the metabolism of glucose

[24], therefore
fructose can more readily enter the gluconeogenic and glycogen synthesis pathways.
Hepatic glycogen storage is limited, therefore excess fructose may be rapidly diverted to
other metabolic pathways, such as DNL, which contributes to the production of hepatic
triglyceride. This may eventually lead to liver fat accumulation and disease, as the
primary lipogenic response of the liver to fructose abundance has been implicated in the
pathogenesis of NAFLD

[10].

Fructose intake has been shown to be twice as high (90g/day) in patients with NAFLD
than in healthy patients

[10]. The main source of fructose in all age classes in the US is
soft drinks

[25,26]. Although the present study did not include dietary recall
questionnaires to assess fructose intake, it has been shown that sugar intake is
relatively high (>100g/d), accounting for nearly 50% of overall daily carbohydrate intake
and 25% of caloric intake,

in both African American and Hispanic adolescents living in
Los Angeles [25].

Although the mechanism by which fructose malabsorption protects against hepatic fat
accumulation remains unknown, the consumption of added sugars, particularly fructose
containing sugar-sweetened beverages, is similar between African American and
Hispanic adolescents

[27, 28]. Our data suggests an acute ethnic variation in the ability
to absorb dietary fructose. It is likely that chronic overconsumption of fructose containing
beverages facilitates hepatic lipid deposition via DNL [10] and that variation in fructose
malabsorption may play a role in explaining individual variation in liver fat accumulation.
It is thus plausible to hypothesize that the portal delivery of fructose to the liver, in
African Americans with low fructose absorption, may be reduced, thus lowering
substrate availability for DNL resulting in lower liver fat levels. Further studies will be
required to investigate such mechanisms.

! $&!
Another potential explanation for the divergence in fructose absorption between African
Americans and Hispanics may involve GLUT5 transporter expression levels. During the
prenatal and infant periods of human development, GLUT5 mRNA and fructose
transport levels are low

[29,30] and malabsorption is observed when fructose is
introduced to children less than 1 year old

[31]. In animal studies, the early introduction
of dietary fructose in rat pups has been shown to alter gene expression (Slc2a5) and up-
regulate GLUT5 protein expression, thereby improving fructose absorption

[32].
Differential expression of the fructose transporter, GLUT5, could potentially play a role in
divergent fructose absorption between Hispanics and African Americans. In addition to
potential gene expression differences, the existence of various single nucleotide
polymorphisms (SNPs) in the Slc2a5 fructose transporter gene have been reported in
humans, which may possibly alter the function or activity of the protein

[33]. There are no
studies that have investigated ethnic differences in intestinal GLUT5 transporter
expression and/or function, but the authors speculate that further investigation along
these avenues may help explain ethnic differences in fructose absorption capacity.
This study is limited by its relatively small sample size (16 African American and 21
Hispanic). The study was a cross-sectional examination of data, therefore despite a
significant association between higher malabsorption rates and lower liver fat values in
African Americans, no causal relationship can be established. Although there is
currently no standardized fructose load in the literature, based on previously published
research a 50g fructose load was utilized [21,22], however it is important to acknowledge
that breath hydrogen testing is limited in its diagnostic utility. The subjects were all obese
young adults, therefore any conclusions resultant from this study must be limited to
these groups. However, because African Americans and Hispanics represent the two
most divergent groups in terms of hepatic fat fraction, and obesity confers an elevated
risk in the development of NAFLD, this study presented a unique opportunity to examine
differences in fructose malabsorption related to hepatic fat fraction in two at-risk groups.
In summary, we found that African Americans have both a higher prevalence and a
greater magnitude of fructose malabsorption than Hispanics. In African Americans,
fructose malabsorption was negatively correlated with liver fat, which, through
mechanisms that warrant further investigation, may be protective against fatty liver
disease. A better understanding of the physiological mechanisms underlying this
relationship may help explain the divergence in NAFLD prevalence between African
! $'!
Americans and Hispanics as well as inform novel dietary recommendations and/or
therapeutic targets for the reduction of liver fat in Hispanics.

! $(!
Table 5-1. Clinical Characteristics

Legend: Values are presented in mean±SD and were obtained by independent t-test
with equal variance assumed. Gender and fructose malabsorption frequency were
analyzed with the Pearson chi-square test. SAT=subcutaneous adipose tissue,
VAT=visceral adipose tissue, HFF=hepatic fat fraction, H
2
AUC=hydrogen area under
the curve.

! $)!
Figure 5-1A. Mean H
2
area under the curve (AUC) for all breath collection time points
over the 180-minute procedure.

Legend: The solid circles represent African Americans and the hollow circles represent
Hispanics. African Americans exhibited significantly higher mean H
2
production than
Hispanics at 150 and 180 minutes (p<0.01 and p<0.05, respectively). African Americans
also showed a trend towards higher mean overall H
2
AUC than Hispanics (116.2±101.5
ppm vs. 78.5±67.5 ppm).

!!,
!!!!!!!,
! $*!

Figure 5-1B. Liver fat percent vs. H
2
production in African Americans and Hispanics.

Legend: Liver fat and hydrogen area under the curve data was log transformed. Liver fat
was negatively related to fructose malabsorption in African Americans (r=-0.56, p=0.03).
There were no significant relationships between liver fat and fructose malabsorption in
Hispanics. AA=African American, HIS=Hispanic.

! $+!
Figure 5-1C. Liver fat percent vs. Breath H
2
at 180 minutes in African Americans and
Hispanics.

Legend: Liver fat data and breath hydrogen at 180 minutes data were log transformed.
Liver fat was negatively related to H
2
production at 180 minutes in African Americans
(r=-0.628, p=0.03). There were no significant relationships between liver fat and H
2

production in Hispanics. AA=African American, HIS=Hispanic.

! $"!
Figure 5-2. Mean Incremental H
2
AUC at 150-180 minutes by ethnicity.

Legend: The mean incremental H
2
AUC from 150 to 180 minutes was significantly higher
in African Americans than in Hispanics (915 vs. 146, respectively; p=0.006). Bars
represent standard error.

! $#!
Chapter 5 References:
1. Bray GA, Nielsen SJ, Popkin BM: Consumption of high-fructose corn syrup in
beverages may play a role in the epidemic of obesity. Am J Clin Nutr 80:1090, 2004.
2. Le KA, Ith M, Kreis R, Faeh D, Bartolloti M, Tran C, Boesch C, Tappy L: Fructose
overconsumption causes dyslipidemia and ectopic lipid deposition in healthy subjects
with and without a family history of type 2 diabetes. Am J Clin Nutr 89:1760–1765, 2009.
3. Koebnick C, Getahun D, Reynolds K, Coleman KJ, Porter AH, Lawerence JM,
Punyanitna M, Quinn VP, Jacobsen SJ: Trends in nonalcoholic fatty liver disease-related
hospitalizations in US children, adolescents, and young adults. J Pediatr Gastroenterol
Nutr 48:597-603, 2009.
4. Browning JD, Szczepaniak LS, Dobbins R, Nuremberg P, Horton JD, Cohen JC,
Grundy SM, Hobbs HH: Prevalence of hepatic steatosis in an urban population in the
United States: impact of ethnicity. Hepatology 40:1387-1395, 2004.
5. Weston SR, Leyden W, Murphy R, Bass NM, Bell BP, Manos MM, Terrault, NA:
Racial and ethnic distribution of nonalcoholic fatty liver in persons with newly diagnosed
chronic liver disease. Hepatology 41: 372–379, 2005.
6. Solga SF, Clark JM, Alkhuraishi AR. Torbenson M, Tabesh A, Schweitzer M, Diehl
AM, Magnuson TH: Race and comorbid factors predict nonalcoholic fatty liver disease
histopathology in severely obese patients. Surg Obes Relat Dis 1:6–11, 2005.
7. Giday SA, Ashiny Z, Naab T, Smoot D, Banks A: Frequency of nonalcoholic fatty liver
disease and degree of hepatic steatosis in African-American patients. J Natl Med Assoc
98:1613–1615, 2006.
8. Nielsen SJ, Popkin BM: Changes in beverage intake between 1977 and 2001. Am J
Prev Med 27:205-210, 2004.
9. Stanhope KL, Schwarz JM, Keim NL, Griffen SC, Bremer AA, Graham JL, Hatcher B,
Cox CL, Dyachenko A, Zhang W, McGahan JP, Seibert A, Krauss RM, Chiu S, Schaefer
EJ, Ai M, Otokozawa S, Nakajima K, Nakano T, Beysen C, Hellerstein MK, Berglund L,
Havel PJ: Consuming fructose-sweetened, not glucose-sweetened, beverages increases
visceral adiposity and lipids and decreases insulin sensitivity in overweight/obese
humans. J Clin Invest 119:1322-1334, 2009.
10. Ouyang X, Cirillo P, Sautin Y, McCall S, Bruchette JL, Diehl AM, Johnson RJ,
Abdelmalek MF: Fructose consumption as a risk factor for non-alcoholic fatty liver
disease. J Hepatol 48:993–999, 2008.
! $$!
11. Riby JE, Fujisawa T, Kretchmer N. Fructose absorption: Am J Clin Nut 58:748S-
753S, 1993.
12. Kim JS, Lê K-A, Mahurkar S, Davis JA, Goran MI: Relationships Between Liver Fat,
Circulating Adipocytokines and Insulin Resistance in Obese Hispanic Adolescents.
Pediatric Obesity 7:158-164, 2012.
13. Lê KA, Mahurkar S, Alderete TL, Hasson RE, Adam TC, Kim JS, Beale E, Xie C,
Greenberg AS, Allayee H, Goran MI: Subcutaneous adipose tissue macrophage
infiltration is associated with ectopic fat deposition, hyperinsulemia and stimulation of
NF-kappa-B stress pathway. Diabetes 60:2802-2809, 2011.
14. Rao SS, Attaluri A, Anderson L, Stumbo P: Ability of the normal human small
intestine to absorb fructose: Evaluation by breath testing. Clinical Gastroenterology and
Hepatology 5:959-963, 2007.
15. Braden B: Methods and functions: Breath tests. Best Practice and Research Clinical
Gastroenterology 23:337-352, 2009.
16. Urita Y, Ishihara S, Akimoto T, Kato H, Hara N, Honda Y, Nagai Y, Nakanishi K,
Shimada N, Sugimoto M, Miki K: Seventy-five gram glucose tolerance test to assess
carbohydrate malabsoprtion and small bowel bacterial overgrowth. World J
Gastroenterol 12:3092-3095, 2006.
17. Szczepaniak LS, Nurenberg P, Leonard D, Browning JD, Reingold JS, Grundy S,
Hobbs HH, Dobbins RL: Magnetic resonance spectroscopy to measure hepatic
triglyceride content: prevalence of hepatic steatosis in the general population. Am J
Physiol Endocrinol Metab 288:E462-468, 2005.
18. Davis JN, Lê KA, Walker RW, Vikman S, Spruijt-Metz D, Weigensberg MJ, Allayee
H, Goran MI: Increased hepatic fat in overweight Hispanic youth influenced by
interaction between genetic variation in PNPLA3 and high dietary carbohydrate and
sugar consumption. Am J Clin Nutr 92:1522-1527, 2010.
19. Wolever TMS & Jenkins DJA: The use of the glycemic index in predicting the blood
glucose response to mixed meals. Am J Clin Nutr 43:167–172, 1986.
20. Nelis GF, Vermeeren MA, Jansen W: Role of fructose-sorbitol malabsorption in the
irritable bowel syndrome. Gastroenterology 99:1016-1020, 1990.
21. Choi YK, Johlin FC Jr, Summers RW, Jackson M, Rao SS: Fructose intolerance: An
under-recognized problem. Am J Gastroenterol 98:1348-1353, 2003.
22. Ravich WJ, Bayless TM: Fructose: incomplete intestinal absorption in humans.
Gastroenterology 84:26-29, 1983.
! &%%!
23. Rumessen JJ, Gudemand-Hoyer E: Absorption capacity of fructose in healthy adults.
Comparison with sucrose and its constituent’s monosaccharides. Gut 27:1161-1168,
1986.
24. Spruss A, Bergheim I: Dietary fructose and intestinal barrier: potential risk factor in
the pathogenesis of nonalcoholic fatty liver disease. Journal of Nutritional Biochemistry
20:657-662, 2009.
25. Putnam JJ & Allshouse, JE: Food consumption, prices, and expenditures, 1970–97.
Food and Rural Economic Division, Economic Research Service. USDA Stat. Bull 965,
1999.
26. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM: Prevalence
of overweight and obesity in the United States, 1999--2004. JAMA 295:1549-1555,
2006.
27. Hasson RE, Adam TC, Davis JN, Kelly LA, Ventura EE, Byrd-Williams CE, Toledo-
Corral CM, Roberts CK, Lane CJ, Azen SP, Chou CP, Spruijt-Metz D, Weigensberg MJ,
Berhane K, Goran MI: Randomized controlled trial to improve adiposity, inflammation
and insulin resistance in obese African American and Latino youth. Obesity 20:811-818,
2011.
28. Thompson FE, McNeel TS, Dowling EC, Midthune D, Morrissette M, Zeruto CA:
Interrelationships of added sugars intake, socioeconomic status, and race/ethnicity in
adults in the United States: National Health Interview Survey, 2005. Journal of the
American Dietetic Association 109:1376-1383, 2009
29. Buddington RK, Diamond JM: Ontogenetic development of intestinal nutrient
transporters. Annu Rev Physiol 51:601–619, 1989.
30. Ferraris RP: Dietary and developmental regulation of intestinal sugar transport.
Biochem J 360:265–276, 2001.
31. Nobigrot T, Chasalow FI, Lifshitz F: Carbohydrate absorption from one serving of
fruit juice in young children: age and carbohydrate composition effects. J Am Coll Nutr
16:152–158, 1997.
32. Douard V, Cui XL, Soteropoulos P, Ferraris RP: Dexamethasone sensitizes the
neonatal intestine to fructose induction of intestinal fructose transporter (Slc2A5)
function. Endocrinology 149:409–423, 2008.
33. Wasserman D, Hoekstra JH, Tolia V, Taylor CJ, Kirschner BS, Takeda J, Bell GI,
Taub R, Rand EB. Molecular analysis of the fructose transporter gene (GLUT5) in
isolated fructose malabsorption. J Clin Invest 98:2398-2402, 1996.
! &%&!
Chapter 6: Summary, future directions and concluding remarks

This chapter focuses on a summary of findings and future directions will be discussed
with an emphasis on utilizing the key findings of the studies in this dissertation as a
means to inform potential therapies for NAFLD. The chapter will close with an overall
conclusion of this dissertation.

The purpose of this dissertation was to examine how certain genetic polymorphisms and
dietary factors, specifically sugar, may impact risk for NAFLD in primarily Hispanic
children. The fundamental hypothesis was that certain genetic variants related to
triglyceride hydrolysis in the liver may contribute to elevated risk for NAFLD in
overweight and obese Hispanic children and exposure to dietary factors, such as
fructose (at levels known to promote increases in liver fat), in a population of children
already at risk for NAFLD due to high obesity rates, would further increase risk of
developing the disease. These factors are all tightly liked and many are interrelated, as
highlighted in the figure below (Figure 6-1), and represent somewhat of a “perfect storm”
for Hispanics.

Figure 6-1. General model for the role of genetic and dietary components in NAFLD.

! &%'!
Findings in this dissertation shed some light on several key relationships in the paradigm
surrounding obesity and NAFLD in Hispanic children, some of which are outlined and
expanded upon below.

Genetic Susceptibility:
In paper 1, “Genetic and Clinical Markers of Elevated Liver Fat Content in Overweight
and Obese Hispanic Children”, we showed that variants of PNPLA3 and APOC3
independently and in an additive manner contribute to elevated hepatic triglyceride
content in Hispanic children. Moreover, risk prediction models that incorporated a GRS
marginally improved upon the ability to discriminate ELF. Our study was underpowered
to detect associations of the other SNP variants with liver fat, thus it remains to be
determined if the inclusion of said SNPs into a more comprehensive GRS could improve
prediction of NAFLD better. From a biological perspective, the variants interrogated in
Chapter 2 exert their potential effects on liver accumulation in varied ways. APOC3 is
thought to potentially influence NAFLD by delaying the metabolism of triglyceride-rich
lipoprotein particles, which increases their uptake by the liver (1) and over-expression of
the gene product in animal models exacerbates this function. The frequency of the risk
allele of APOC3 is slightly higher in our Hispanic population than in Caucasian
populations, thus it is likely that this may in part explain the lipogenic role it plays in
NALFD. However, several other studies, as noted in Chapter 2, fail to show an
independent relationship with liver fat and the role of APOC3 is further complicated by
potentially different effects of the variant in children vs. adults. The GCKR variant inhibits
glucokinase activity, thereby regulating glucose storage/disposal and increasing
substrate availability for hepatic de-novo lipogenesis (2) and similarly PPP1R3B, a
serine/threonine phosphatase involved in hepatic glycogen synthesis, could thus
modulate risk of NAFLD through a similar mechanism as GCKR. Since the association
of these genes with NAFLD is mediated through their effects on adiposity, lipid
metabolism and/or glucose homeostasis, it is also possible that we were unable to see
differential between genotype effects that were as strong as expected because all of the
participants in the analysis were overweight to obese. Including lean subjects in follow
up analyses may indicate stronger variant allele effects on liver fat content.

PNPLA3 has the most robust effect on liver fat in this, and other, populations. Although
we were able to show an effect of APOC3 independent of PNPLA3, is it apparent that
! &%(!
the G allele of PNPLA3 is driving the risk for elevated liver fat that we observed in the
population. Similar results were recently presented in a larger analysis of Mexican
Americans by Palmer et al. (3) The authors were able to again replicate the effects of
PNPLA3 on liver fat (#=0.28, CI, 0.18-0.38) while also showing positive effects of GCKR
(NS) and PPP1R3B and negative effects of NCAN or LYPLAL1 on liver fat. Again, this
study was both larger and conducted in an adult population, however it is likely that
these significant gene variants are playing a role, albeit small in comparison to PNPLA3,
in risk for NAFLD in Hispanics.

As a follow up to the study presented in Chapter 2, genetic admixture will be conducted
on the cohort to determine actual vs. self-reported ethnicity. It is possible that some
participants that self-identify as Hispanic possess genetic markers that link them to
different ancestries, such as African American, Asian or Caucasian. Post hoc analyses
stratified by ethnicity may further elucidate the risk that these gene variants confer for
NAFLD on distinct ethnic groups. Although the subgroups may be much smaller, it is
possible that even with limited power some variants may achieve significance with liver
fat and that a comprehensive GRS may confer more benefit amongst a certain ethnicity
that the others. Validations of such findings in larger study populations may allow a more
comprehensive evaluation of the clinical utility of a GRS. Given the steady increase in
NAFLD prevalence in the US and internationally, a GRS could be a useful tool in the
earlier detection and treatment of NAFLD in at-risk subjects.

Dietary sugar as a risk factor:
Paper 2, “Fructose content of sugar sweetened beverages and juices”, strengthens
previous findings regarding the fructose content of SSBs and provides new information
on the sugar composition and overall fructose content of commonly consumed SSB and
juice products. The most popular sodas made with HFCS as the sole added sweetener
had a fructose to glucose ratio of 60:40, indicating a meaningful difference from the
equivalent ratio of fructose to glucose observed in table sugar and certain fruit juices
contained more fructose than sodas, often with 50% more fructose than glucose.
Although SSBs are a major source of fructose in the diet of Americans, these results
demonstrate that juice products may contribute substantially to total daily fructose
consumption.

! &%)!
Fructose per-day consumption has risen substantially over the last century (4), as
demonstrated in figure 6-2. Some adolescents consume as much as 125g/d (5), which
can represent more than 25% of total caloric intake.

Figure 6-2. Historic trends in daily fructose consumption.

Given that high daily fructose consumption is known to promote visceral fat expansion
and hepatic steatosis as well as the progression from nonalcoholic fatty liver disease to
nonalcoholic steatohepatitis, chronic, long-term consumption of beverages high in
fructose represents a significantly greater risk to Hispanic populations (6), suggesting
that interventions focused on 1) informing the consumer of the sugar content of foods
and 2) reduced consumption of sweetened beverages may assist in preventing adverse
metabolic outcomes. The concept of collapsing a dietary intervention into an intervention
that specifically targets improvement in a genetically susceptible group is a natural
progression from the findings presented in this dissertation. A positive association
between dietary sugar and liver fat, which was significant only in GG carriers of
PNPLA3, has been observed in Hispanic children (7). These novel data suggest that
overweight and obese Hispanic children with the GG genotype may be more susceptible
to diets high in sugar, and thus, would hypothetically be more responsive to reduction in
dietary sugar in terms of liver fat reduction (Figure 6-3).
! &%*!

Figure 6-3. A dietary intervention targeting at-risk genotypes in Hipsanic children.

This hypothesis is consistent with the current, though limited, understanding of PNPLA3
function, which is thought to produce a protein with lipase-like activity. Therefore, it is
prudent to directly test the hypothesis of a specific diet by gene interaction for potential
treatment of NAFLD in obese Hispanic children and adolescents. This approach allows a
unique opportunity to directly test for a specific diet by gene interaction and therefore the
potential to identify dietary approaches for reducing fatty liver disease that may be
specific to an individual’s genotype. Sugar reduction may be a particularly effective
strategy for GG homozygotes of the PNPLA3 at-risk variant, a hypothesis that is
supported by a biologically plausible mechanism. A randomized clinical trial of dietary
sugar reduction in children with fatty liver disease would facilitate the examination of
whether there are differential effects based on genotype of a single amino acid
substitution in the PNPLA3 gene that is highly prevalent in Hispanics and associated
with significantly elevated liver fat. The overall study design would be to therefore recruit
120 Hispanic children (10-18 years) with clinically diagnosed NAFLD and randomize
them to one of two 12-week interventions; Group 1 (standard of care control group):
! &%+!
Dietary intervention focused on healthy eating (n=60) or Group 2; (sugar reduction):
Dietary intervention focused on healthy eating and sugar reduction (focus on reduction
of sugary beverages, juices and added sugar towards a goal of 10% of daily calories;
n=60). Overall, the results of such a study would generate efficacy data for a novel gene
by diet strategy for treatment of NAFLD and improvement of associated metabolic
outcomes in obese Hispanic children and adolescents. Consequently, this study would
have the potential to impact personalized dietary recommendations for treatment and
prevention of NAFLD as a function of genetic predisposition, and lead to a better
understanding of the mechanisms controlling liver fat in Hispanics and other at-risk
populations.

Another strategy to further investigate novel roles of fructose in promoting metabolic
disease is via examining the clinical malabsorption of fructose and the reaction of the gut
microbiome to excess fructose as a substrate. With dietary ingestion of fructose-
containing products, fructose is transported across the intestinal mucosa in a
transporter-mediated facilitative diffusion process by the fructose-specific hexose
transporter, GLUT5, with additional contributions by other transporters such as GLUT2
(8) and GLUT7 (9). Absorbed fructose is then delivered to the liver via portal circulation,
where it is metabolized into triose-phosphates. These intermediates can be directly
oxidized, or contribute to glycogen and glucose synthesis via gluconeogenesis, as well
as triglyceride synthesis and liver fat accumulation via stimulation of the de novo
lipogenesis pathway. Fructose metabolism is not rate-limited by the availability of
phosphofructokinase, as is the metabolism of glucose (10), therefore fructose can more
readily enter the gluconeogenic and glycogen synthesis pathways. Hepatic glycogen
storage is limited, therefore excess fructose is rapidly diverted to other metabolic
pathways, such as de novo lipogenesis, which contributes to the production of hepatic
triglyceride and further secretion as very low-density lipoprotein triglyceride. This may
eventually lead to liver fat accumulation and disease, as the primary lipogenic response
of the liver to fructose abundance has been implicated in the pathogenesis of NAFLD
(11). With impaired fructose absorption, as found amongst African Americans in Chapter
5, portal delivery of fructose to the liver may be reduced, thus lowering substrate
availability for de novo lipogenesis resulting in lower liver fat accumulation. GLUT5
fructose-specific transporters are likely overwhelmed by excess fructose in small
intestinal lumen (Figure 6-4). Excess fructose does not enter portal circulation and
! &%"!
delivery of fructose to the liver is perturbed due to this malabsorption. Fructose is
transported to the large intestine and undergoes anaerobic bacterial fermentation
producing the bi-products H2, CH4, CO2 and short chain fatty acids (SCFA). H2 diffuses
into circulation and is exhaled in the breath where it can be measured and used to
determine the extent of fructose malabsorption.

Figure 6-4. Proposed mechanism by which fructose malabsorption may prevent hepatic
fat accumulation.

In addition to the production of measurable gasses that may be measured as a means of
determining fructose malabsorption, the gut microbiota also produce endotoxin as a
biproduct of fructose fermentations. Lipopolysachharide (LPS), is one such biproduct
(12). LPS has been shown to be delivered to the liver via portal circulation and induce
both inflammation and insulin resistance vie the activation of toll-like receptor 4 proteins
(13). Although very little is known about the role of fructose-derived endotoxins in liver
health, a clear relationship exists and merits further investigation. Additionally, the mere
! &%#!
presence of chronically high levels of fructose in the intestine may serve to alter
microbiome population dynamics. Fructose drives the population of large intestinal
microbiota towards a more “Westernized” phenotype, which is marked by altered
metabolic activity and strong associations with obesity-related disease (14). A loss or
reduction of specific microbiome populations as a consequence of this “Western diet”
substrate conditioning (ex. fructose) may promote dysbiosis or loss of several specific
bacterial niches in the gut, which could contribute to the previously observed differences
between bacterial populations in obese and normal weight subjects (15, 16).

The fact that current population estimates of fructose consumption are likely
underestimated due to ‘hidden’ fructose means that consumers are exposed to
significantly higher levels of daily fructose. This supports the hypothesis that excess
fructose consumption may promote long-term changes in metabolism via constant
malabsorption and the passing of substrate to the large intestine. The role of gut bacteria
in metabolic health is a rapidly emerging field and the abovementioned changes in gut
host bacteria in response to dietary exposures may represent hallmark periods in which
interventions could potentially prevent “Westernization” of the microbiome and
subsequent disease risk. A longitudinal analysis measuring changes in the gut bacterial
populations, as well as biproducts of fructose and other substrates, in response to
temporal changes in dietary patterns (i.e. the shift from breast feeding to solid foods or
the introduction of sodas and juices into the diet) in children would shed light on these
relationships and potentially highlight critical periods in development in which dietary
choices impact future health. Coupled with new methodology that is under development
to more accurately assess sugar consumption from SSBs (17), there is an abundance of
opportunity test the effect of interventions that reduce SSB intake on alterations to gut
bacteria and resultant metabolic health.

Obesity inflammation as a mediator of liver disease:
In Chapter 4 we showed that in biopsied tissue from Italian children an abundance of
inflammatory markers in the SAT were observed, including CLS and fibrosis. We were
able to observe a positive association between CLS status and liver fibrosis and
ballooning scores and no other links between adipose tissue inflammation or fibrosis and
liver disease were detected. Additionally, SAT collagen was significantly associated with
impaired DI, which was supported by the higher glucose AUC and lower insulin AUC
! &%$!
values observed in the high fibrosis group. Other studies, as outlined in Chapter 4, have
successfully established links between other fat depots and liver disease and glucose
homeostasis, yet the role of SAT in these outcomes remains unresolved. Some of our
findings are outlined in figure 6-5. Despite our findings, our relatively small sample size
likely prevented us from finding stronger associations between adipose and liver
inflammation. Obtaining biopsied adipose and liver tissue in children is challenging. This
makes the direct measure of inflammation in adipose depots and/or liver tissue
extremely difficult. Because the discussed relationships between the two tissue sites
may be very different based on age, it is important to understand how inflammation and
fibrosis contribute to insulin resistance and NAFLD in children vs. adults. NAFLD
prevalence continues to rise internationally, and liver biopsies to diagnose NASH may
become a more common occurrence. However, it is nearly impossible to justify the
biopsy of visceral or omental adipose regions in children not already undergoing
abdominal surgery of some type. Alternatively, deep SAT (dSAT) has become a depot of
interest due to similar inflammatory profiles compared to VAT (18). Novel computed
tomography or ultrasound-guided biopsy techniques (19) present an ideal opportunity to
test associations of measures of inflammation and fibrosis in other fat depots with
aspects of liver disease. Furthermore, it is entirely possible that SAT inflammatory
measures would correlate differently with quantitative measures of liver fat such as MRI,
as shown recently by Jansen et al. (20). Follow up analyses of the study presented in
Chapter 4 could include histological quantification of liver content, via oil red-o staining,
in the biopsy sections. Additionally, future studies of links between liver adipose
inflammation and liver disease should include quantitative scoring of liver disease in
addition to MRI-based assessment of steatosis.

The collagen analysis utilized in chapter 4 was restricted to total rather than subtypes of
collagen, which limits the application of our findings to relationships between SAT and
total collagen with liver disease. Other specific collagen types, such as pericellular
collagen and collagen VI, have been more specifically linked to diabetes and
inflammation. Furthermore, adipose extracellular matrix is considered to be somewhat
dysfunctional in obesity and this is thought to contribute to metabolic syndrome (21).
Pasarica et al. recently demonstrated that collagen VI mRNA is upregulated in obesity in
humans, contributes to macrophage recruitment to adipose regions and higher levels of
macrophage inflammatory proteins. Given the ties we observed between total collagen
! &&%!
and diabetes risk, it is likely that the ECM plays a role in regulating insulin resistance and
that this role may be mediated specifically by subtypes of collagen. This would help
explain the observation of less total collagen in obese than leaner participants that we
observed and lend credit to the hypothesis that collagen type is more important than
absolute amount. Collagen is also thought to reduce adipose oxygenation (22), which is
associated with inflammation, suggesting that collagen VI development may be the
primary step in driving an unhealthy metabolic phenotype in adipose tissue. Subsequent
studies examining fat depot biopsies should aim to assay for collagen mRNA subtype to
assess relationships with measures of metabolic health such as insulin resistance.

Figure 6-5. Summary of relationships between obesity inflammation, liver disease and
metabolic risk.

Conclusion:
As with any complex disease, NAFLD is impacted and driven by a myriad of factors, of
which obesity is likely the most robust. As we have shown, certain genetic
polymorphisms such as PNPLA3, APOC3 and others, perturb normal triglyceride
hydrolysis in the livers of carriers of these variants. Additionally, excess sugar intake, in
particular fructose, stresses normal hepatic pathways of metabolism and drives these
substrates towards a lipogenic fate resulting in increased liver fat. Underlying obesity
! &&&!
leads to pro-inflammatory conditions in the adipose tissue in which adipokines such as
adiponectin are reduced and fibrotic conditions are favored. This environment prevents
normal adipose tissue expansion and likely forces ectopic spill-over of lipids into the
liver, which promotes hepatic steatosis, NAFLD and subsequently more serious liver
pathologies.

Obesity sets the stage for the onset of fatty liver in children. When viewed individually,
any one of the abovementioned factors is a significant contributor to the prevalence of
liver disease in Hispanic children. However, these contributors cannot be thought of in
isolation of one another. In many cases all three mechanisms are working
simultaneously and synergistically to drive the development of NAFLD. The clear answer
to halting NAFLD is through the prevention of obesity, yet this has proved to be an
extremely difficult task. Current weight loss strategies in children are exercise and
calorie-restriction based, yet due to a host of reasons are relatively ineffective and
difficult to maintain. The high frequency of the PNPLA3 variant, the chronic
overconsumption of fructose-containing foods and beverages, and the underlying
prevalence of overweight and obesity amongst Hispanic children present alternative
points at which to intervene and develop evidence-based interventions for NAFLD
prevention and reversal. Future effective therapies to address NAFLD in this population
will need to employ strategies that target those most at risk. Our findings provide both a
starting point from which to begin identifying the most at-risk children as well as specific
dietary components on which to focus intervention.

! &&'!
Chapter 6 references:
1. Petersen KF, Dufour S, Hariri A, et al. Apolipoprotein C3 gene variants in nonalcoholic
fatty liver disease. N Engl J Med 2010; 362:1082 – 1089.
2. Beer NL, Tribble ND, McCulloch LJ, et al. The P446L variant in GCKR
associated with fasting plasma glucose and triglyceride levels exerts its effect through
increased glucokinase activity in liver. Hum Mol Genet 2009; 18:4081 – 4088.
3. Palmer ND, Musani SK, Yerges-Armstrong LM, Feitosa MF, Bielak LF, Hernaez R et
al. Characterization of European-ancestry NAFLD-associated Variants in individuals of
African and Hispanic descent. Hepatology. 2013 Apr 8. doi: 10.1002/hep.26440. [Epub
ahead of print]
4. Lim JS, Mietus-Snyder M, Valente A, Schwarz JM, Lustig RH. The role of fructose in
the pathogenesis of NAFLD and the metabolic syndrome. Nat Rev Gastroenterol
Hepatol. 2010 May;7(5):251-64.
5. Marriott BP, Cole N, Lee E. National estimates of dietary fructose intake increased
from 1977 to 2004 in the United States. J Nutr. 2009 Jun;139(6):1228S-1235S.
6. Goran MI, Walker R, le K-A, et al. Effects of PNPLA3 on Liver Fat in Obese Hispanic
Children and Adolescents Diabetes 2010; 59: 3127-3130.
7. Davis JN, Lê KA, Walker RW, Vikman S, Spruijt-Metz D, Weigensberg MJ, Allayee H,
Goran MI. Increased hepatic fat in overweight Hispanic youth influenced by interaction
between genetic variation in PNPLA3 and high dietary carbohydrate and sugar
consumption. Am J Clin Nutr. 2010 Dec;92(6):1522-7.
8. Kellett GL, Brot-Laroche E, Mace OJ et al. Sugar absorption in the intestine: the role
of GLUT2. Annu Rev Nutr, 2008. 28: p. 35-54.
9. Drozdowski LA, Thomson AB. Intestinal sugar transport. World J Gastroenterol, 2006.
12(11): p. 1657-70.
10. Spruss A, Bergheim I. Dietary fructose and intestinal barrier: potential risk factor in
the pathogenesis of nonalcoholic fatty liver disease. J Nutr Biochem, 2009.20(9): p. 657-
62
11. Ouyang X, Cirillo P, Sautin Y et al. Fructose consumption as a risk factor for non-
alcoholic fatty liver disease. J Hepatol 48: 993–999, 2008.
12. Rao SS, Attaluri A, Anderson L, Stumbo P. Ability of the normal human small
intestine to absorb fructose: evaluation by breath testing. Clin Gastroenterol Hepatol
2007;5:959-963.
! &&(!
13. Vos MB, Lavine JE. Dietary fructose in nonalcoholic fatty liver disease. Hepatology.
2013 Jun;57(6):2525-31.
14. Payne AN, Chassard C, Lacroix C. Gut microbial adaptation to dietary consumption
of fructose, artificial sweeteners and sugar alcohols: implications for host-microbe
interactions contributing to obesity. Obes Rev. 2012 Sep;13(9):799-809.
15. Ley RE, Backhed F, Turnbaugh P, Lozupone CA, Knight RD, Gordon JI. Obesity
alters gut microbial ecology. Proc Natl Acad Sci U S A 2005; 102: 11070–11075.
16. Payne AN, Chassard C, Zimmermann MB, Mü ller P, Stinca S, Lacroix C. The
metabolic activity of gut microbiota in obese children is increased compared with normal-
weight children and exhibits more exhaustive substrate utilization. Nutr Diabetes 2011;
1: e12. doi:10.1038/nutd.2011.8.
17. Choy K, Nash SH, Kristal AR, Hopkins S, Boyer BB, O'Brien DM. The carbon isotope
ratio of alanine in red blood cells is a new candidate biomarker of sugar-sweetened
beverage intake. J Nutr. 2013 Jun;143(6):878-84. doi: 10.3945/jn.112.172999. Epub
2013 Apr 24.
18. Tordjman J, Divoux A, Prifti E, Poitou C, Pelloux V, Hugol D, Basdevant A, Bouillot
JL, Chevallier JM, Bedossa P, Guerre-Millo M, Clement K. Structural and inflammatory
heterogeneity in subcutaneous adipose tissue: relation with liver histopathology in
morbid obesity. J Hepatol. 2012 May;56(5):1152-8.
19. de Bazelaire C, Sabatier F, Pluvinage A, de Kerviler É. CT-guided percutaneous
biopsies. J Radiol. 2011 Sep;92(9):842-59.
20. Jansen HJ, Vervoort GM, van der Graaf M, Stienstra R, Tack CJ. Liver fat content is
linked to inflammatory changes in subcutaneous adipose tissue in type 2 diabetes
patients. Clin Endocrinol (Oxf). 2012 Nov 20. doi: 10.1111/cen.12105. [Epub ahead of
print]
21. Pasarica M, Gowronska-Kozak B, Burk D, Remedios I, Hymel D, Gimble J, Ravussin
E, Bray GA, Smith SR. Adipose tissue collagen VI in obesity. J Clin Endocrinol Metab.
2009 Dec;94(12):5155-62.
22. Pasarica M, Sereda OR, Redman LM, Albarado DC, Hymel DT, Roan LE, Rood JC,
Burk DH, Smith SR. Reduced adipose tissue oxygenation in human obesity: evidence
for rarefaction, macrophage chemotaxis, and inflammation without an angiogenic
response. Diabetes. 2009 Mar;58(3):718-25.

Abstract (if available)

Abstract The purpose of this dissertation is to summarize our recent findings that have examined dietary, genetic and inflammatory factors that contribute to fat accumulation and nonalcoholic fatty liver disease (NAFLD) in Hispanic children. In addition, this dissertation discusses how some of these contributions to liver fat may vary across the population in terms of ethnic-specific effects and future treatments that may result from the findings presented. ❧ Recent genome wide studies have identified several polymorphisms that contribute to increased liver fat accumulation, with some of these genes relating to dietary carbohydrate and sugar consumption. In particular, a variant of the PNPLA3 gene, which is highly prevalent in Hispanics, contributes to excessive liver fat beginning at a young age, especially in the context of high sugar consumption. We show that a genetic risk score comprised of variants associated with liver fat improve detection of NAFLD when added to current clinical measures. Additionally, dietary carbohydrate, and especially fructose, has been shown to contribute to increased liver fat accumulation due to the lipogenic potential of fructose during liver metabolism. We present data that demonstrates certain beverages amongst sodas and juices contain higher concentrations of fructose than previously thought, suggesting population estimates of fructose consumption in children and adults are underestimated. Lastly, we present recent finding suggesting that inflammation and fibrosis in the subcutaneous adipose tissue of children may impact liver disease progression and increase risk for type 2 diabetes. ❧ Dietary sugar contributes to liver fat accumulation, with this being explained by de novo lipogenesis from fructose in the liver. Certain genetic factors, including PNPLA3, GCKR and APOC3 contribute to increased liver fat accumulation, with these effects being manifested at an early age. Hispanics in particular are at elevated risk for liver fat accumulation due to the higher frequency of genetic variants such as PNPLA3 and GCKR as well as an interaction between the PNPLA3 and dietary sugar. Additionally, adipose inflammation caused by the obese state may incur elevated risk for liver disease and type 2 diabetes. The summation of these factors represents, and potentially explains, the added risk and high prevalence of NAFLD among the Hispanic pediatric population.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The influence of dietary fructose on genetic effects of GCK and GCKR in Mexican Americans

PDF

Quantity versus quality: how adipose tissue accumulation and immune cell profile associate with risk for type 2 diabetes in minority children and adults

PDF

Ectopic fat and adipose tissue inflammation in overweight and obese African Americans and Hispanics

PDF

Dietary fats, atherosclerosis, and cardiovascular disease

PDF

Genetic studies of inflammation and cardiovascular disease

PDF

The association between the PNPLA3 rs738409 variant and hepatic fat in Mexican Americans

PDF

Role of sympathetic nervous system in metabolic function

PDF

Study of genetic variation in Asian Indians -- applications to complex diseases and endogamic exogamy

PDF

Genetic risk factors in multiple myeloma

PDF

Identification of gene-exposure interactions for risk of cardiovascular disease

PDF

Understanding acute lymphoblastic leukemia in different ethnic groups in the United States

PDF

Hispanic ethnicity in multiple sclerosis

PDF

The contribution of arachidonate 5-lipoxygenase to inflammatory response in coronary artery disease

PDF

Genetic variations in gene from the cytochrome P450 family may account for differential response to pioglitazone therapy in the Hispanic women

PDF

Influence of genetic background on the effect of traffic-related air pollution on obesity phenotypes

PDF

The role of adipocytes in acute lymphoblastic leukemia cell migration and survival against daunorubicin

PDF

Association of single nucleotide polymorphisms in GCK, GCKR and PNPLA3 with type 2 diabetes related quantitative traits in Mexican-American population

PDF

Functional characterization of colorectal cancer GWAS loci

PDF

Air pollution, mitochondrial function, and growth in children

PDF

Gene expression and genetic variation of ERG is associated with inflammation in endothelial cells and risk of coronary artery disease in humans

Asset Metadata

Creator Walker, Ryan William (author)

Core Title Genetic and dietary determinants of nonalcoholic fatty liver disease in Hispanic children

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Integrative Biology of Disease

Publication Date 08/01/2013

Defense Date 06/10/2013

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Children,dietary,fat,genetic,HFCS,Hispanic,liver,liver fat,NAFLD,OAI-PMH Harvest,obesity,polymorphism,SNP,Sugar

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Allayee, Hooman (committee chair), Goran, Michael (committee member), Sinatra, Frank (committee member), Watanabe, Richard M. (committee member)

Creator Email rwwalker@usc.edu,ylwgto@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-310859

Unique identifier UC11292827

Identifier etd-WalkerRyan-1924.pdf (filename),usctheses-c3-310859 (legacy record id)

Legacy Identifier etd-WalkerRyan-1924.pdf

Dmrecord 310859

Document Type Dissertation

Format application/pdf (imt)

Rights Walker, Ryan William

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA