SUPPLEMENT

The Science of Fatty Acids and Inflammation1–3

Kevin L Fritsche*
Animal Sciences Division, Department of Nutrition and Exercise Physiology, University of Missouri, Columbia, MO

ABSTRACT

Inflammation is believed to play a central role in many of the chronic diseases that characterize modern society. In the past decade, our
understanding of how dietary fats affect our immune system and subsequently our inflammatory status has grown considerably. There are
compelling data showing that high-fat meals promote endotoxin [e.g., lipopolysaccharide (LPS)] translocation into the bloodstream, stimulating
innate immune cells and leading to a transient postprandial inflammatory response. The nature of this effect is influenced by the amount
and type of fat consumed. The role of various dietary constituents, including fats, on gut microflora and subsequent health outcomes in the
host is another exciting and novel area of inquiry. The impact of specific fatty acids on inflammation may be central to how dietary fats affect
health. Three key fatty acid–inflammation interactions are briefly described. First, the evidence suggests that saturated fatty acids induce
inflammation in part by mimicking the actions of LPS. Second, the often-repeated claim that dietary linoleic acid promotes inflammation was not
supported in a recent systematic review of the evidence. Third, an explanation is offered for why omega-3 (n–3) polyunsaturated fatty acids
are so much less anti-inflammatory in humans than in mice. The article closes with a cautionary tale from the genomic literature that illustrates
why extrapolating the results from inflammation studies in mice to humans is problematic. Adv Nutr 2015;6:293S–301S.

Keywords: fatty acids, inflammation, endotoxin, linoleic acid, omega-3, microflora, lipopolysaccharide

Introduction (4), diabetes (4), Alzheimer and other neurologic diseases

Inflammation is an essential component of the host re- (5), and cancer (6). Inflammatory responses result in local
sponse to infection or injury (1, 2). The inflammatory re- and systemic production of numerous soluble products,
sponse involves the interactions among and between many including C-reactive protein (CRP)4, TNF-a, IL-6, serum
different cell types. The classic symptoms associated with in- amyloid A, and plasminogen activator inhibitor 1, among
flammatory responses include heat, redness, swelling, pain, others (7). Numerous epidemiologic studies have reported
and loss of function. Typically, inflammation is meant to associations between one or more of these biomarkers and
be transient, but under some circumstances the acute re- the risk of various chronic diseases. However, there is no
sponse can become chronic (3). Excess chronic inflamma- consensus regarding which inflammatory biomarker is
tion is an important etiologic factor in a wide range of best. Instead, it appears that many researchers measure mul-
common chronic diseases, including cardiovascular disease tiple biomarkers in any given study to increase the odds that
associations with clinical outcomes will emerge. Impor-
1
tantly, recent reports provide support for the idea that differ-
Published in a supplement to Advances in Nutrition. Presented at the University of
Massachusetts Roundtable on "Fats and Oils: Where Food Function Meets Health" held in
ent diseases are associated with specific inflammatory
Chicago, IL, 18–19 November 2013. The roundtable was sponsored by the University of biomarker profiles, which relates to dysregulation of specific
Massachusetts and supported by an unrestricted grant from the Alliance for Potato immune cell populations (8, 9). In the future one hopes that
Research and Education. The nonindustry roundtable speakers received travel funding and
an honorarium for participation in the meeting and manuscript preparation. The views
disease-specific inflammation biomarkers will be clearly de-
expressed are those of the authors. fined. In the meantime, readers interested in gaining a better
2
The author’s research program is supported by University of Missouri’s Food for the 21st understanding of the many issues surrounding biomarker
Century Program in the College of Agriculture, Food, and Natural Resources within the
University of Missouri and by grant P50AT006273 from the National Center for
selection are encouraged to read the report of the European
Complementary and Alternative Medicine (NCCAM), the Office of Dietary Supplements branch of the International Life Sciences Institute, which
(ODS), and the National Cancer Institute (NCI). The contents of this paper are solely the
responsibility of the author and do not necessarily represent the official views of the
NCCAM, ODS, NCI, or the NIH.
3
Author disclosure. KL Fritsche serves as a scientific advisor to the Lipids Committee for the 4
Abbreviations used: AA, arachidonic acid; ALA, a-linolenic acid; CRP, C-reactive protein;
International Life Sciences Institute–North America. GPR120, G protein–coupled receptor 120; LA, linoleic acid; LTB4, leukotriene B4; MF, milk fat;
* To whom correspondence should be addressed. E-mail: [email protected]. TLR, toll-like receptor.

ã2015 American Society for Nutrition. Adv. Nutr. 6: 293S–301S, 2015; doi:10.3945/an.114.006940. 293S
commissioned a review of the biomarkers for monitoring a number of metabolic processes associated with metabolic
inflammation in human nutritional studies (10). syndrome (e.g., hepatic TG accumulation, elevated fasting
There exists a large body of evidence suggesting that a va- insulin, visceral adipose tissue accumulation) in a manner
riety of dietary factors can enhance or diminish inflammation similar to infusion of LPS. Elevated expression of a num-
(11). The focus of this article is to describe how dietary fats ber of inflammation biomarkers, such as TNF-a, IL-1, and
affect inflammation and why it is an important human health IL-6, was observed in the liver, adipose, and muscle. These
consideration. Current dietary guidelines for fats provide in- responses were surprisingly similar in mice infused with
formation about both the amount as well as the types of fats LPS compared with those fed the high-fat diet. However,
that should be consumed (12). Because fats are energy dense, it was the authors’ use of mice carrying a deletion of
the guidelines set an upper limit of 35% of total calories from CD14, a critical component of the LPS receptor, which
fats. The primary goal of setting this upper limit is to reduce provided the most compelling evidence that many of the
the risk of developing obesity, a condition associated with el- adverse effects of high-fat feeding may be a consequence
evated concentrations of several inflammatory biomarkers, of activation of inflammatory signaling pathways.
including CRP and TNF-a (13). Current thinking is that the acute postprandial inflam-
Evidence suggests that, in addition to the amount of fat, matory response associated with fat consumption is medi-
the types of fats consumed can have a major impact on hu- ated by endotoxin, primarily derived from gut microflora
man health. Their impact on the risk of cardiovascular dis- (23). Others suggested, however, that many foods contain
ease has dominated the rationale for these recommendations endotoxin or related proinflammatory compounds that,
for the past 50 y (14–16). There is a growing recognition by upon absorption, directly stimulate inflammatory re-
leaders in the nutrition and health field that dietary fats can sponses (24). Those foods with the most in vitro inflam-
affect host inflammatory responses (17). After reviewing the matory activity included meats, cheeses, and dairy
evidence, it seems reasonable that the impact of dietary fats products. Regardless of source, once absorbed, endotoxin
on inflammatory status be factored into future dietary guide- is shuttled between chylomicrons and HDL particles via a
lines for these nutrients. specific binding proteins and soluble receptors present in
the circulation (25).
Current Status of Knowledge Rodent studies indicate that the structure/form of the di-
It appears that one way in which dietary fat is linked to in- etary fat affects how much endotoxin is absorbed (26).
flammation is by promoting the translocation of microbial There is a strong correlation between postprandial lipemia
products from the gut into the bloodstream. Conservative and net endotoxin absorption. Evidence suggests that chylo-
estimates suggest that there are >100 trillion commensal microns play a critical role in the absorption and transport
(i.e., normal) organisms in the gut (18). These microorgan- of endotoxin. Factors that promoted fat absorption, such
isms, collectively referred to as the “gut microbiome,” con- as emulsification, also enhanced endotoxin absorption. Re-
tain >1 g of LPS. LPS is also referred to as endotoxin, cently, Mani et al. (27) demonstrated that fat source affected
because it is an endogenous component of the cell wall of postprandial endotoxin absorption and transport in the sera
all gram-negative bacteria and it can have “toxic” effects in of domestic pigs. After overnight feed deprivation, pigs were
most mammals (19). LPS is a very potent stimulus of in- fed a meal containing 12.5% by weight (~25% of total en-
flammatory responses, with bioactivity in the microgram ergy) added fat or saline (control). The meal was consumed
per liter concentration range. Importantly, there exists con- within 10 min, then blood samples were taken hourly for
siderable diversity in the structure and bioactivity of LPS 5 h. Pigs fed coconut oil, rich in SFAs, had the highest cir-
from microbial species (20). This diversity suggests that in- culating concentrations of endotoxin followed by those that
dividual gut microbiomes may be more or less proinflam- were fed vegetable oil and then those fed fish oil. Additional
matory on the basis of how effectively innate immune testing on freshly isolated samples of ileum showed that the
receptors recognize these microbially derived agonists. fats had not affected overall intestinal integrity or permea-
Researchers have known for decades that gut microbes bility. The authors suggested that the endotoxin absorbed
could be a source of systemic bacterial infection leading to was most likely transported by way of lipid raft–mediated
sepsis and organ failure under a variety of medical circum- endocytosis.
stances (21). Yet, it was only recently that researchers reported However, not all fat challenge studies reported significant
that dietary fat could promote endotoxin absorption. Cani findings. Tousoulis et al. (28) examined the impact of a single
et al. (22) reported that feeding mice a diet very high in fat bout of fat/oil consumption on inflammation using soluble
(i.e., 72% of total energy) over 4 wk significantly elevated cir- vascular cell adhesion molecule 1 assessment. Healthy sub-
culating endotoxin concentrations compared with mice fed a jects (n = 37) were randomly assigned to receive 50 mL of
low-fat control diet. The data suggest that high-fat feeding re- water or oil (e.g., corn oil, extra-virgin olive oil, soy oil, or
sults in a chronic elevation in circulating endotoxin through- cod liver oil). The authors found no statistical differences
out the day and night. The authors reported that the high-fat between treatment groups for circulating soluble vascular cell
diet altered the distribution and numbers of some of the mi- adhesion molecule 1 either pre- or post-treatment. The authors’
crobial populations found in the gut. Interestingly, the au- reliance on a single inflammatory biomarker is an important
thors went on to demonstrate that high-fat feeding affected limitation of this study.

294S Supplement
 The selection of multiple inflammatory biomarkers, Recent findings suggest that dietary fats can influence gut
however, is still no guarantee that treatment effects will be microflora composition and that this can affect inflamma-
observed. For example, Voon et al. (29) monitored circulat- tory status in vivo (31). In a recent study the effects of 3 dif-
ing TNF-a, IL-1, IL-6, IL-8, and CRP in 45 healthy human ferent dietary fat sources on the normal microflora of
subjects who participated in a randomized crossover inter- C57BL/6 mice were examined (32). When compared with
vention study. Three different test fats were examined for lard, milk fat (MF) and a PUFA-rich fat had similar effects
their impact on circulating cholesterol and inflammatory on Bacteroidetes and Firmicutes; however, a large increase
status. The test fats included a palmitic acid (16:0, hexadeca- in a member of the Deltaproteobacteria, Bilophila wadswor-
noic acid)–rich palm oil, coconut oil (rich in 12:0 + 14:0, thia, was consistently observed only with MF. Interestingly,
dodecanoic 1 tetradecanoic acids), and virgin olive oil these microbiota changes differed from those induced by
[rich in oleic acid (18:1, octadecenoic acid)]. The hypothesis lard-based SFAs.
was that fats rich in SFA, such as coconut and palm oils, To demonstrate the clinical relevance of such a shift in
would enhance circulating biomarkers of inflammation, gut microflora, the researchers conducted a follow-up study
and thus the inflammatory status of these healthy subjects. with a strain of mice that are susceptible to developing in-
Fasting and nonfasting (2 h postprandial) blood samples flammatory bowel disease (i.e., Il10–null mice). Only MF
were collected 5 wk after the start of the dietary interven- consumption promoted colitis and inflammatory cytokine
tions. The macronutrient content of the test diets were expression in the distal colonic mucosa of these Il10–null
maintained at 20% of energy from protein, 30% from fat, mice. These researchers went on to show that the mecha-
and 50% from carbohydrates. The results failed to demon- nism by which MF affected gut microflora was related to
strate any significant impact of fat source on any of these cir- changes in hepatic bile acid production (see Figure 1)
culating biomarkers of inflammation. The most instructive (33). Whether dietary fats substantially affect inflammatory
aspect of these data was the significant inter- and intraindi- status of people by altering their gut microflora remains un-
vidual variation that was noted for these inflammation bio- tested, but with the rapid advances in the field, answers
markers. As a consequence of this variation, many such should be forthcoming.
studies suffer from being underpowered. Apart from their role in promoting the uptake of endo-
Clearly, not all fats under all circumstances promote toxin in the gut, many researchers believe that dietary fats
postprandial inflammation. There are insufficient data are able to affect inflammation by more direct means. For
to predict when and how specific fat sources will affect in- example, it has long been known that SFAs are an essential
flammatory status in people. One possible explanation for structural component of bacterial endotoxins (19). The lipid
the discrepancies in the literature is the variability in the A portion of all pathogenic LPS contain 6 ester- and/or amide-
types of microbes in the gastrointestinal tract of individ- linked saturated fatty acyl groups. The length of the acyl
uals being studied in these postprandial fat challenge chains in lipid A ranges from 12 to 16 carbons. The substi-
studies. Recent advances in our understanding of how tution of SFA with MUFA or PUFA eliminates the proin-
the gut microbiome adapts to changes in human ecology flammatory activity of LPS. Macrophages, and other cells
over time are particularly relevant in this context (30). It of the innate immune system, possess receptors [i.e., toll-
was suggested that the increasing incidence of allergic and like receptor (TLR) 4] that recognize LPS (34). LPS-mediated
metabolic disorders (e.g., obesity, type 2 diabetes) in im- signaling through TLR4 leads to the activation of NF-kB, a
migrant populations as they adopt Western dietary pat- transcription factor, that subsequently turns on the expression
terns is linked to shifts in gut microbial populations. of numerous proinflammatory cytokines, such as TNF-a,
Whether this can explain why some individuals are more IL-1, IL-6, and IL-8. TLR4 is part of a larger family of recep-
sensitive to systemic inflammation upon a fat challenge is tors responsible for recognizing pathogen-associated mo-
unknown. lecular patterns (35). More than a dozen TLRs have been

FIGURE 1 Pathway from

consuming milk fat to colitis by
way of altered bile acid
production and subsequent shifts
in gut microbial populations.
Reproduced from reference 33
with permission.

Dietary fat, fatty acids, and inflammation 295S

identified in humans and mice. These receptors are ex- acid (AA; 20:4n26, eicosatetraenoic acid) accumulation, 2)
pressed throughout the body and are critical for host de- enhanced synthesis of proinflammatory eicosanoids derived
fense against invading pathogens. from AA, 3) reduced conversion of a-linolenic acid (ALA;
In 2001, Lee et al. (36) were the first to demonstrate that 18:3n23, octadecatrienoic acid) into EPA (eicosapentae-
SFAs were able to directly stimulate inflammatory gene ex- noic acid; 20:5n–3) and/or DHA, and 4) diminished synthe-
pression by way of TLR4 signaling in vitro. The relative po- sis of anti-inflammatory eicosanoids from EPA and DHA.
tency of various SFAs varied with chain length, with lauric The experimental evidence supporting each step of this par-
acid (12:0) showing the greatest activity, whereas myristic adigm originated primarily from rodent and cell culture
acid (14:0) and stearic acid (18:0, octadecanoic acid) ap- studies. More recently, and in contrast with the multistep
peared to have surprisingly little proinflammatory activity. process described above, it was suggested that various oxi-
In contrast to SFAs, MUFAs and PUFAs failed to activate dized forms of LA are directly responsible for stimulating
TLR4 signaling. Interestingly, these researchers were able inflammation (43).
to show that pretreatment of cells for 3 h with a variety of Fortunately, the first evidence-based review of all the hu-
PUFAs or oleic acid (octadecaenoic acid; 18:1n–9) signifi- man clinical data available that addressed the impact of di-
cantly reduced the subsequent proinflammatory effect of etary LA on inflammation in healthy adults was recently
lauric acid treatment. They went on to show that the ability published (44). Fifteen studies (8 parallel and 7 crossover)
of PUFAs to block inflammatory responses induced by LPS met the inclusion criteria. The most important inclusion
or lauric acid was dependent on TLR4. criterion was that the only FA other than LA that was al-
A few years later, this same group examined the impact of lowed to differ substantially between the experimental and
FAs on TLR2 signaling (37). TLR2, like TLR4, is a potent control dietary interventions was oleic acid. The reason for
stimulator of inflammatory responses. Microbial compo- this was the existing evidence that suggested that oleic acid
nents that are potent agonists for TLR2 include di- and tri- had no impact on inflammation (45). In contrast, it is be-
acylated lipoproteins, peptidoglycans, and lipoteichoic acid. lieved that SFAs promote and n–3 PUFAs reduce inflamma-
It turns out that TLR2 only functions as part of a hetero- tion; thus, simultaneous changes in the intake of those FAs
dimer with either TLR1 or TLR6. The rationale for explor- along with LA could confound interpretation of the results.
ing the potential of FAs to affect TLR2 signaling arose Heterogeneity between these studies prevented meta-analysis.
from the results of a study these researchers conducted Regardless of this limitation, not one of the studies reported
with innate immune cells from TLR4-mutant mice. When a significant positive association between LA intake and cir-
bone marrow cells from TLR4-null mice were differentiated culating concentrations for a wide variety of inflammatory
into macrophages and then treated with lauric acid the markers. More often than not, higher LA intake was associ-
authors observed upregulation of cyclooxygenase 2 Ptgs2 ated with lower, not higher, inflammatory status in healthy
expression, an inflammatory gene. Because these cells did adults. In addition, the results from a randomized controlled
not express TLR4, LPS treatment was without effect on cy- trial, which was completed and published after the system-
clooxygenase 2 expression. These authors found that lauric atic review, indicated that increasing LA intake from 4%
acid stimulated TLR2-mediated signaling only when TLR2 to 13% of energy improved (i.e., diminished) biomarkers
was coexpressed with TLR1 or TLR6. In contrast to lauric of inflammation in obese subjects (46). Not surprisingly,
acid, DHA (22:6n–3, an n–3 PUFA) treatment tended to di- those subjects consuming extra SFAs from butter showed
minish microbial agonist-mediated signaling of a wide vari- elevations in plasma markers of inflammation.
ety of TLRs. More will be said about the anti-inflammatory So an important question is why did the evidence from
activity of DHA in the section on n–3 FAs. human clinical trials fail to support the theory that dietary
Linoleic acid (LA; 18:2n26, octadecadienoic acid) is an LA promotes inflammation? One reason might be that the
n–6 PUFA and an essential nutrient (38). LA comprises “LA-proinflammatory paradigm” relies on an overly simpli-
$50% of the most widely consumed vegetable oils in West- fied model of LA metabolism. Originally, most of the proin-
ern societies. For many decades it has been known that LA flammatory activity of dietary LA was thought to be a
helps reduce blood cholesterol concentrations and that consequence of an accumulation of AA, which leads to
substituting LA for SFAs lowers the risk of heart disease greater production and release of proinflammatory eicosa-
(39). Therefore, current recommendations from numerous noids, such as PGE2 and leukotriene B4 (LTB4). A review
expert bodies, including the Institute of Medicine and the of the clinical literature by Rett and Whelan (47) indicated
American Heart Association, are that people should consume that increasing LA up to 6-fold within the context of a typ-
between 5% and 10% of total energy as LA for a heart- ical Western diet failed to increase tissue AA. Surprisingly,
healthy diet (40). reducing dietary LA down to 10% of control was without ef-
However, a few members of the lipid research community fect on circulating AA. Therefore, for those who currently
have expressed concerned that LA-rich diets are unhealthy advocate for large reductions in dietary LA, their emphasis
and promote inflammation (41,42). The theoretical basis has shifted to the potential adverse effects of oxidized forms
for this concern over LA’s proinflammatory actions involve of this PUFA. Recent advances in analytical capabilities
a number of putative interrelated metabolic processes, includ- (i.e., lipidomics) have greatly expanded our knowledge of
ing the following: 1) dietary LA promoting tissue arachidonic LA-derived metabolites (48). Yet, our understanding of the

296S Supplement
 complex process that involves a programmed series of steps
has only recently become widely accepted. Importantly, in
addition to DHA and EPA, AA-derived lipid mediators (i.e., lip-
oxins) also play an important role in programming resolution.
Second, DHA is believed to affect lipid microdomains
within cell membranes (i.e., lipid rafts) that play a role in
immune cell signaling pathways critical to inflammation
(57). These researchers reported that, although lauric acid
promoted, DHA diminished the recruitment of TLR4 into
lipid raft fractions after LPS treatment. This DHA action re-
duced TLR4 homodimerization and subsequent signaling
through this key proinflammatory pathway. The ability of
DHA to affect the physical properties of cellular membrane
FIGURE 2 LA and AA metabolites play roles in both microdomains has been shown for a variety of cell types
inflammation and resolution. Solid lines indicate (58). DHA-mediated alterations in membrane structure
proinflammatory pathways, and dotted/dashed lines represent can induce apoptosis in cancer cells, but its effects are highly
anti-inflammatory/proresolving pathways. AA, arachidonic acid; dependent on the cell type and the molar concentration of
CYP450, cytochrome P450; EET, epoxyeicosatrienoic acids; HETE, DHA within membrane phospholipids.
hydroxyeicosatetraenoic acid; HODE, hydroxyoctadecadienoic A third possible mechanism of action for DHA (and EPA)
acid; LA, linoleic acid; LO, lipoxygenase; LTB4, leukotriene B4; LTX, relative to their anti-inflammatory activity recently emerged
leukotoxin; NO, nitrosylated; PGE2, prostaglandin E2. from the laboratory of Olefsky and coworkers (59). They re-
ported that a novel G protein–coupled receptor 120 (GPR120)
bioactivity and physiologic role of each of these novel me- serves as a receptor/sensor for DHA (and to a lesser extent,
tabolites remains incomplete. Figure 2 is meant to capture EPA). Using a mouse macrophage cell line (i.e., RAW 264.7
some of this complexity, at least as it relates to LA. Unfortu- cells) they demonstrated that stimulation of GPR120 with
nately, this author is unaware of a single publication that de- DHA or a chemical agonist resulted in a greatly diminished
scribes a research study in which all of the possible bioactive inflammatory response from these cells. By using small in-
metabolites of LA and other important FAs have been mea- terfering RNA–mediated knockdown of GPR120, they were
sured and accounted for. Two excellent reviews related to able to completely abrogate the anti-inflammatory activity
this topic were recently published (49,50). of in vitro DHA treatment. The fact that most of the data
Much has been written about the potential health bene- generated in these studies used 100 mmol/L DHA, a concen-
fits associated with increasing our intake of n–3 PUFAs, in- tration that greatly exceeds physiologic norms, raises some
cluding improved neurologic and cardiovascular health and concerns about the actual role of GPR120 in vivo. The evi-
diminished inflammation (51–53). n–3 PUFAs are a group dence that GPR120 is specific for DHA (and EPA) remains
of structurally related FAs. ALA is an 18-carbon n–3 PUFA uncertain in light of the limited nature of the actual dose-
from which all other n–3 PUFAs can be produced through a response curves presented and the fact that palmitoleic acid
series of metabolic steps. Like LA, ALA is considered to be (16:1n–7, hexadecenoic acid) was just about as effective as
an essential nutrient for humans and most animals. DHA is DHA (59).
the final end product of ALA elongation and desaturation Regardless of which mechanism might explain the anti-in-
(54). Although DHA is found in the highest concentrations flammatory role of DHA, it would be reasonable to presume
in the brain and retina, it can be found in every cell mem- that when exploring the impact of dietary interventions on
brane. Although consumption of ALA is an inefficient means DHA-mediated immune and inflammatory responses, the
for enriching cellular DHA content, consuming preformed molar concentration and the magnitude of change in mem-
DHA from various marine products (e.g., fish oil) can result brane DHA content would be critical factors in determining
in substantial increases in cellular DHA and EPA content (55). to what extent cellular responses are altered. It is here where
Importantly, DHA modulation of immune cell function there is an important difference between mice and humans.
and subsequent inflammatory response are thought to be a Specifically, immune cells from mice start out with higher con-
result of one or more of these 3 actions. First, DHA (and centrations of DHA and, upon exposure to dietary DHA, the
EPA) are precursors for anti-inflammatory, proresolving lipid levels of enrichment in immune cell membranes far exceed
mediators known as resolvins, docosatrienes, and protectins what is possible in human immune cells (see Figure 3) (60).
(56). The anti-inflammatory activity of these dual-acting The dramatic accumulation of DHA in murine immune cells
lipid mediators is a consequence of their promotion of neu- likely affects lipid rafts and cell signaling in ways that are not
trophil apoptosis and monocyte recruitment. These mono- reproduced in the more modestly enriched human immune
cytes differentiate into macrophages that efficiently engulf cells. Turk and Chapkin (61) illustrated how DHA enrichment
the apoptotic neutrophils and depart the inflammatory site might affect membrane-based lipid rafts (see Figure 4). These
by way of the lymphatic system. The novel concept that res- and other experts in the field (62) have discussed the dose-
olution is not simply the absence of inflammation but a dependent nature of n–3 FAs on a number of health conditions,

Dietary fat, fatty acids, and inflammation 297S

 inflammatory lipid mediators than that produced by murine
cells. Unfortunately, quantitative analyses of these lipid me-
diators have not been reported to date. These key species-
dependent differences may help explain why dietary fish oil, a
rich source of DHA, has such a powerful beneficial impact on
a variety of inflammatory conditions in mice, whereas human
clinical trials have shown much more modest benefits, if any.
Recently, a comprehensive analysis of the disparity be-
tween mouse and human responses relative to the potency
of dietary n–3 PUFAs to affect inflammatory conditions
was published (63). Consistent with the refractory nature
of human immune cells to incorporate DHA into their
membranes, one would predict that dietary intake of DHA
would need to be much higher than what would be required
FIGURE 3 Quantitative comparison of dietary DHA with immune to modulate immune cell function in mice, whose immune
cell DHA from mice, rats, and humans. The data were from studies cells are more easily enriched with n–3 PUFAs. In fact, that is
that met the following criteria: 1) dietary n–3 PUFA (i.e., EPA and/or exactly what the author concluded: “In adult humans, an
DHA) intake was a dependent variable in the study design, 2) the EPA plus DHA intake >2 g/d seems to be required to elicit
FA profile of an identifiable immune cell population was reported, anti-inflammatory actions.” Because such intake amounts
and 3) data were published and identified in PubMed (National cannot readily be obtained through dietary means, these ef-
Library of Medicine) through December 2005. n–3 PUFA intake is fects should be considered pharmacologic and not nutri-
expressed as a percentage of total energy consumed (i.e., en%). In
tional in nature.
most studies, daily caloric intake was not reported. Thus, the
Recent data from genomic screening experiments in mice
following assumptions were made: 1) human subjects consumed
2000 kcal/d and 2) rodents consumed the same calories across diet and humans suggest that there is another reason that results
treatment groups. Best-fit lines/curves with 95% CI displayed by from mouse feeding studies with n–3 PUFA as well as other
dotted lines were generated by using Prism software version 4.0b fat sources may have led to findings that were not predictive
(GraphPad). Reproduced from reference 60 with permission. of responses in humans (64). In this report, the authors
compared the temporal changes in the expression of thou-
including inflammation. In addition, the lower concentrations sands of genes from blood leukocytes isolated from humans
of cellular EPA and DHA in human immune cells would in or mice after 3 forms of serious trauma: burns, endotoxe-
all likelihood result in more modest production of the anti- mia, and blunt injury. The genomic responses in circulating

FIGURE 4 Putative model for the

effect of n–3 PUFAs on lipid rafts. Lipid
rafts are nanoscale regions of the plasma
membrane, enriched in cholesterol,
sphingomyelin, and phospholipids
containing saturated acyl chains. Both
transmembrane and peripheral
membrane proteins can be localized to
lipid rafts. Upon treatment with a
combination of n–3 PUFAs or DHA
alone, these PUFAs are incorporated into
phospholipids, which are inserted into
both raft and nonraft regions of the
plasma membrane. This results in
enhanced clustering of lipid raft regions,
which are depleted of cholesterol and
sphingomyelin. In addition, many lipid
raft–associated proteins “mislocalize” to
the bulk membrane domain. This results
in a suppression of lipid raft–mediated
processes, including T cell activation and
downstream signal transduction.
Reproduced from reference 61 with
permission.

298S Supplement
human leukocytes to these diverse forms of trauma were the possibility that fats can have both acute as well as chronic
surprisingly similar. Yet, among the genes that were changed effects on host inflammatory responses. Whereas cell culture
significantly in humans, the mouse orthologs failed to reflect and animal models play an important role in biomedical re-
similar changes (R2 between 0.0 and 0.1; see Figure 5). search, limitations inherent in these models suggest that data
These data suggest that mouse models poorly reflect the from human clinical trials will continue to have primacy in
physiologic responses seen in humans to systemic inflam- setting dietary recommendations for fats and FAs. In light of
matory challenges. In fact, similar conclusions were drawn the lack of consensus regarding which biomarker is best for
in a 2007 article, in which the clinical outcomes (e.g., circu- monitoring inflammatory status, it is recommended that as
lating cytokines, leukopenia, fever, changes in respiration) many inflammation biomarkers be measured as feasible and
associated with sepsis in humans and the various “relevant” that studies be appropriately powered in recognition of the
mouse models were compared (65). Researchers should highly variable nature of these biomarkers.
therefore exercise caution when relying solely on mouse
models for investigating the impact of dietary fats on inflam- Acknowledgments
matory responses/status in humans. The sole author had responsibility for all parts of the manuscript.

Conclusions References
Dietary fats have a major impact on human health. A growing 1. Henson PM. Dampening inflammation. Nat Immunol 2005;6:1179–81.
body of evidence suggests that inflammatory status should be 2. Nathan C. Points of control in inflammation. Nature 2002;420:
846–52.
included as one of the characteristics for which dietary fats are
3. Nathan C, Ding A. Nonresolving inflammation. Cell 2010;140:871–82.
evaluated relative to their impact on human health. At this 4. Dandona P, Aljada A, Bandyopadhyay A. Inflammation: the link be-
time, it is uncertain how dietary fats might affect inflamma- tween insulin resistance, obesity and diabetes. Trends Immunol 2004;
tory status, but current evidence suggests that the gut micro- 25:4–7.
biome is important in this regard. Studies should account for 5. Perry VH, Cunningham C, Holmes C. Systemic infections and inflam-
mation affect chronic neurodegeneration. Nat Rev Immunol 2007;7:
161–7.
6. Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and
cancer. Cell 2010;140:883–99.
7. Larsen GL, Henson PM. Mediators of inflammation. Annu Rev Immu-
nol 1983;1:335–59.
8. Nikolajczyk BS, Jagannathan-Bogdan M, Denis GV. The outliers be-
come a stampede as immunometabolism reaches a tipping point. Im-
munol Rev 2012;249:253–75.
9. Mraz M, Haluzik M. The role of adipose tissue immune cells in obesity
and low-grade inflammation. J Endocrinol 2014;222:R113–27.
10. Calder PC, Ahluwalia N, Albers R, Bosco N, Bourdet-Sicard R, Haller
D, Holgate ST, Jonsson LS, Latulippe ME, Marcos A. A consideration of
biomarkers to be used for evaluation of inflammation in human nutri-
tional studies. Br J Nutr 2013;109(Suppl 1):S1–34.
11. Basu A, Devaraj S, Jialal I. Dietary factors that promote or retard in-
flammation. Arterioscler Thromb Vasc Biol 2006;26:995–1001.
12. Kris-Etherton PM, Innis S, Ammerican DA. Position of the American
Dietetic Association and Dietitians of Canada: dietary fatty acids. J Am
Diet Assoc 2007;107:1599–611.
13. Trayhurn P, Wood I. Signalling role of adipose tissue: adipokines and
inflammation in obesity. Biochem Soc Trans 2005;33:1078–81.
FIGURE 5 Comparison of the genomic response in circulating 14. Hu FB, Manson JE, Willett WC. Types of dietary fat and risk of coro-
nary heart disease: a critical review. J Am Coll Nutr 2001;20:5–19.
leukocytes to severe acute inflammation from 6 distinct causes
15. Lichtenstein AH, Appel LJ, Brands M, Carnethon M, Daniels S, Franch
in human and murine models. GEO was queried for studies in HA, Franklin B, Kris-Etherton P, Harris WS, Howard B. Diet and life-
the white blood cells of severe acute inflammatory diseases (i.e., style recommendations revision 2006: a scientific statement from the
burns, endotoxemia, trauma, sepsis, ARDS, and infection) in American Heart Association Nutrition Committee. Circulation 2006;
humans and mice. The fold-change of each gene measured was 114:82–96.
calculated between patients and controls in a human study or 16. Chowdhury R, Warnakula S, Kunutsor S, Crowe F, Ward HA, Johnson
between treated and control groups in a murine model study; L, Franco OH, Butterworth AS, Forouhi NG, Thompson SG, et al. As-
and for a time-course data set, the maximum fold-change was sociation of dietary, circulating, and supplement fatty acids with coro-
calculated. The gene response in each data set was then nary risk: a systematic review and meta-analysis. Ann Intern Med 2014;
compared with the 5554 genes that were significantly changed 160:398–406.
17. Lee JY, Zhao L, Hwang DH. Modulation of pattern recognition recep-
in human trauma, burns, and endotoxemia. Shown are
tor-mediated inflammation and risk of chronic diseases by dietary fatty
correlations (x axis) and directionality (y axis) of gene response acids. Nutr Rev 2010;68:38–61.
from the resulting multiple published data sets in GEO 18. Guarner F, Malagelada J-R. Gut flora in health and disease. Lancet
compared with human burn injury. ARDS, acute respiratory 2003;361:512–9.
distress syndrome; GEO, Gene Expression Omnibus. Reproduced 19. Rietschel ET, Kirikae T, Schade FU, Mamat U, Schmidt G, Loppnow H,
from reference 64 with permission. Ulmer AJ, Zahringer U, Seydel U, Di Padova F. Bacterial endotoxin:

Dietary fat, fatty acids, and inflammation 299S

 molecular relationships of structure to activity and function. FASEB J 40. Flock MR, Kris-Etherton PM. Dietary Guidelines for Americans 2010:
1994;8:217–25. implications for cardiovascular disease. Curr Atheroscler Rep 2011;13:
20. Miller SI, Ernst RK, Bader MW. LPS, TLR4 and infectious disease di- 499–507.
versity. Nat Rev Microbiol 2005;3:36–46. 41. Lands WE. Dietary fat and health: the evidence and the politics of pre-
21. Deitch EA. Bacterial translocation of the gut flora. J Trauma 1990;30: vention: careful use of dietary fats can improve life and prevent disease.
S184–89. Ann N Y Acad Sci 2005;1055:179–92.
22. Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, Bastelica D, Neyrinck 42. Simopoulos AP. Essential fatty acids in health and chronic disease. Am J
AM, Fava F, Touhy KM, Chabo C. Metabolic endotoxemia initiates obe- Clin Nutr 1999;70(Suppl):560S–9S.
sity and insulin resistance. Diabetes 2007;56:1761–72. 43. Ramsden CE, Zamora D, Leelarthaepin B, Majchrzak-Hong SF, Faurot
23. Berg RD. Bacterial translocation from the gastrointestinal tract. In: KR, Suchindran CM, Ringel A, Davis JM, Hibbeln JR. Use of dietary
Mechanisms in the pathogenesis of enteric diseases. Vol. 2. Springer linoleic acid for secondary prevention of coronary heart disease and
Science1Business Media, New York; 1999. p. 11–30. death: evaluation of recovered data from the Sydney Diet Heart Study
24. Erridge C. The capacity of foodstuffs to induce innate immune activa- and updated meta-analysis. BMJ 2013;346;e8707.
tion of human monocytes in vitro is dependent on food content of 44. Johnson GH, Fritsche K. Effect of dietary linoleic acid on markers
stimulants of Toll-like receptors 2 and 4. Br J Nutr 2011;105:15–23. of inflammation in healthy persons: a systematic review of ran-
25. Yu B, Hailman E, Wright SD. Lipopolysaccharide binding protein and domized controlled trials. J Acad Nutr Diet 2012;112(7):1029–41,
soluble CD14 catalyze exchange of phospholipids. J Clin Invest 1997; e15.
99:315–24. 45. Baer DJ, Judd JT, Clevidence BA, Tracy RP. Dietary fatty acids af-
26. Laugerette F, Vors C, Géloën A, Chauvin M-A, Soulage C, Lambert- fect plasma markers of inflammation in healthy men fed controlled
Porcheron S, Peretti N, Alligier M, Burcelin R, Laville M. Emulsified diets: a randomized crossover study. Am J Clin Nutr 2004;79:
lipids increase endotoxemia: possible role in early postprandial low- 969–73.
grade inflammation. J Nutr Biochem 2011;22:53–9. 46. Bjermo H, Iggman D, Kullberg J, Dahlman I, Johansson L, Persson L,
27. Mani V, Hollis JH, Gabler NK. Dietary oil composition differentially Berglund J, Pulkki K, Basu S, Uusitupa M. Effects of n26 PUFAs com-
modulates intestinal endotoxin transport and postprandial endotoxe- pared with SFAs on liver fat, lipoproteins, and inflammation in abdom-
mia. Nutr Metab (Lond) 2013;10:6–15. inal obesity: a randomized controlled trial. Am J Clin Nutr 2012;95:
28. Tousoulis D, Papageorgiou N, Antoniades C, Giolis A, Bouras G, Gounari 1003–12.
P, Stefanadi E, Miliou A, Psaltopoulou T, Stefanadis C. Acute effects of 47. Rett BS, Whelan J. Increasing dietary linoleic acid does not increase tis-
different types of oil consumption on endothelial function, oxidative sue arachidonic acid content in adults consuming Western-type diets: a
stress status and vascular inflammation in healthy volunteers. Br J Nutr systematic review. Nutr Metab (Lond) 2011;8:36–51.
2010;103:43–9. 48. Dennis EA. Lipidomics joins the omics evolution. Proc Natl Acad Sci
29. Voon PT, Ng TKW, Lee VKM, Nesaretnam K. Diets high in palmitic USA 2009;106:2089–90.
acid (16:0), lauric and myristic acids (12:0+14:0), or oleic acid (18:1) 49. Delmastro-Greenwood M, Freeman BA, Gelhaus Wendell S. Redox-
do not alter postprandial or fasting plasma homocysteine and inflam- dependent anti-inflammatory signaling actions of unsaturated fatty
matory markers in healthy Malaysian adults. Am J Clin Nutr 2011;94: acids. Ann Rev Physiol 2014;76:79–105.
1451–7. 50. Morisseau C, Hammock BD. Impact of soluble epoxide hydrolase and
30. Walter J, Ley R. The human gut microbiome: ecology and recent evo- epoxyeicosanoids on human health. Annu Rev Pharmacol Toxicol 2013;
lutionary changes. Annu Rev Microbiol 2011;65:411–29. 53:37–58.
31. Burcelin R, Garidou L, Pomié C, editors. Immuno-microbiota cross 51. Wall R, Ross RP, Fitzgerald GF, Stanton C. Fatty acids from fish: the
and talk: the new paradigm of metabolic diseases. In: Seminars in im- anti-inflammatory potential of long-chain omega-3 fatty acids. Nutr
munology. Elsevier; 2012. Rev 2010;68:280–9.
32. Devkota S, Wang Y, Musch MW, Leone V, Fehlner-Peach H, Nadimpalli 52. Hooper L, Thompson RL, Harrison RA, Summerbell CD, Ness AR,
A, Antonopoulos DA, Jabri B, Chang EB. Dietary-fat-induced tauro- Moore HJ, Worthington HV, Durrington PN, Higgins J, Capps NE.
cholic acid promotes pathobiont expansion and colitis in IL102/2 Risks and benefits of omega 3 fats for mortality, cardiovascular disease,
mice. Nature 2012;487:104–8. and cancer: systematic review. BMJ 2006;332:752–60.
33. Turnbaugh PJ. Microbiology: fat, bile and gut microbes. Nature 2012; 53. Riediger ND, Othman RA, Suh M, Moghadasian MH. A systemic re-
487:47–8. view of the roles of n-3 fatty acids in health and disease. J Am Diet As-
34. Hoshino K, Takeuchi O, Kawai T, Sanjo H, Ogawa T, Takeda Y, Takeda soc 2009;109:668–79.
K, Akira S. Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are 54. Brenna JT. Efficiency of conversion of a-linolenic acid to long chain
hyporesponsive to lipopolysaccharide: evidence for TLR4 as the LPS n-3 fatty acids in man. Curr Opin Clin Nutr Metab Care 2002;5:
gene product. J Immunol 1999;162:3749–52. 127–32.
35. Takeda K, Kaisho T, Akira S. Toll-like receptors. Annu Rev Immunol 55. Arterburn LM, Hall EB, Oken H. Distribution, interconversion, and
2003;21:335–76. dose response of n23 fatty acids in humans. Am J Clin Nutr 2006;
36. Lee JY, Sohn KH, Rhee SH, Hwang D. Saturated fatty acids, but not un- 83(Suppl):1467S–76S.
saturated fatty acids, induce the expression of cyclooxygenase-2 medi- 56. Serhan CN, Chiang N, Van Dyke TE. Resolving inflammation: dual
ated through Toll-like receptor 4. J Biol Chem 2001;276:16683–9. anti-inflammatory and pro-resolution lipid mediators. Nat Rev Immu-
37. Lee JY, Zhao L, Youn HS, Weatherill AR, Tapping R, Feng L, Lee WH, nol 2008;8:349–61.
Fitzgerald KA, Hwang DH. Saturated fatty acid activates but polyunsat- 57. Wong SW, Kwon M-J, Choi AM, Kim H-P, Nakahira K, Hwang DH.
urated fatty acid inhibits Toll-like receptor 2 dimerized with Toll-like Fatty acids modulate Toll-like receptor 4 activation through regula-
receptor 6 or 1. J Biol Chem 2004;279:16971–9. tion of receptor dimerization and recruitment into lipid rafts in a
38. Das UN. Essential fatty acids: biochemistry, physiology and pathology. reactive oxygen species-dependent manner. J Biol Chem 2009;284:
Biotechnol J 2006;1:420–39. 27384–92.
39. Harris WS, Mozaffarian D, Rimm E, Kris-Etherton P, Rudel LL, Appel 58. Stillwell W, Shaikh SR, Zerouga M, Siddiqui R, Wassall SR. Docosahex-
LJ, Engler MM, Engler MB, Sacks F. Omega-6 fatty acids and risk for aenoic acid affects cell signaling by altering lipid rafts. Reprod Nutr Dev
cardiovascular disease: a science advisory from the American Heart 2005;45:559–79.
Association Nutrition Subcommittee of the Council on Nutrition, 59. Oh DY, Talukdar S, Bae EJ, Imamura T, Morinaga H, Fan W, Li P, Lu
Physical Activity, and Metabolism; Council on Cardiovascular Nurs- WJ, Watkins SM, Olefsky JM. GPR120 is an omega-3 fatty acid receptor
ing; and Council on Epidemiology and Prevention. Circulation 2009; mediating potent anti-inflammatory and insulin-sensitizing effects.
119:902–7. Cell 2010;142:687–98.

300S Supplement
60. Fritsche K. Important differences exist in the dose–response relation- 63. Calder PC. Mechanisms of action of (n-3) fatty acids. J Nutr 2012;142
ship between diet and immune cell fatty acids in humans and rodents. (Suppl):592S–9S.
Lipids 2007;42:961–79. 64. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker HV, Xu W,
61. Turk HF, Chapkin RS. Membrane lipid raft organization is uniquely Richards DR, McDonald-Smith GP, Gao H, Hennessy L. Genomic re-
modified by n-3 polyunsaturated fatty acids. Prostaglandins Leukot Es- sponses in mouse models poorly mimic human inflammatory diseases.
sent Fatty Acids 2013;88:43–7. Proc Natl Acad Sci USA 2013;110:3507–12.
62. Calder PC. Fatty acids and inflammation: the cutting edge between 65. Rittirsch D, Hoesel LM, Ward PA. The disconnect between animal
food and pharma. Eur J Pharmacol 2011;668:S50–8. models of sepsis and human sepsis. J Leukoc Biol 2007;81:137–43.

Dietary fat, fatty acids, and inflammation 301S