Full text of DOI:10.1079/PNS2001123

Proceedings of the Nutrition Society (2001), 60, 449–456
© The Authors 2001
DOI:10.1079/PNS2001123
CP485©NA6SBN1uI2ntr3tieDtrionienattaSioroyncainlePftylNu2eS0nP0croe1sceoendminugcsoosfatlhiemNmuutnrittiyoLn.SSo.cHieatrybi(g2e00a1n)d4©B.NAu. tCri.tiFoinshSeorc4i4e9ty 2001 600029-6651
Dietary fatty acid modulation of mucosally-induced tolerogenic immune
responses
Laurence S. Harbige and Benjamin A. C. Fisher
School of Chemical and Life Sciences, University of Greenwich, Wellington Street, London SE18 6PF, UK
Dr L. S. Harbige, fax +44 20 8331 8305, email L.Harbige@gre.ac.uk
Immunological unresponsiveness or hyporesponsiveness (tolerance) can be induced by feeding
protein antigens to naïve animals. Using a classical oral ovalbumin gut-induced tolerance protocol
in BALB/c mice we investigated the effects of dietary n-6 and n-3 polyunsaturated fatty acids
(PUFA) on high-and low-dose oral tolerance (and in non-tolerised animals, i.e. effects of antigen
challenge alone) in relation to lymphoproliferative, cytokine and antibody responses. Fish oil rich
in long-chain n-3 fatty acids decreased both T-helper (Th) 1- and Th2-like responses. In contrast,
borage (Borago officinalis) oil rich in n-6 PUFA, of which γ-linolenic acid is rapidly metabolised
to longer-chain n-6 PUFA, increased Th1-like responses and decreased Th2-like responses, and
possibly enhanced suppressor cell or Th3-like activity. These findings are in general agreement
with other studies on the effects of long chain n-3 PUFA on immune system functions, and
characterise important differences between long-chain n-3 and n-6 PUFA, defining more precisely
and broadly the immunological regulatory mechanisms involved. They are also discussed in
relation to autoimmune disease.
Dietary n-3 and n-6 fatty acids: Cytokine and antibody responses: Tolerogenic and antigenic
immune responses
EAE, experimental autoimmune encephalomyelitis; IFN-γ, interferon-γ; Ig, immunoglobulin; IL, interleukin; OVA, ovalbumin; PG, prostaglandins; PUFA, polyunsaturated fatty acids; TGFβ, transforming growth factor β; Th, T-helper
The normal gut immune system can discriminate between
harmless food antigens (tolerance) and those associated with
pathogenic and non-pathogenic microbes. How this discrim-
ination is achieved, however, is not fully understood, but
fundamentally involves regulation of the balance between
tolerance (suppression) and sensitisation (priming). Factors
that influence this balance include: genetic, e.g. major histo-
compatibility complex (Lamont et al. 1988); environmental,
e.g. microbial flora (Chin et al. 2000; Matsuzaki & Chin,
2000); developmental or physiological, e.g. gut closure,
hormonal, lactation (Kolb & Pozzilli, 1999; Heyman, 2001).
Increased understanding of the underlying regulatory
mechanisms in tolerance have demonstrated the importance
(often in relation to antigen dose) of CD4+ T-cells, clonal
deletion, clonal anergy, T-cell cytokine profiles (i.e. T-helper
(Th) 1 v. Th2), active suppression (e.g. Th3 cytokines),
apoptosis, lack of co-stimulatory molecules on antigen-
presenting cells (e.g. B7.1/7.2 (CD80)), and the local
environment (Mowat et al. 1982; Melamed & Friedman,
1993; Friedman & Weiner, 1994; Garside et al. 1995a;
Vidard et al. 1995; Bailey et al. 2001; Strobel, 2001). The
role of T-cell subpopulations secreting different cytokines
has received considerable, and often controversial, attention
in relation to the mechanisms underlying high- and low-dose
tolerance induction, and to tolerance induction during the
neonatal period, which is said to be a critical window for
tolerance induction (Burstein & Abbas, 1993; Garside et al.
1995b; Forsthuber et al. 1996; Ridge et al. 1996; Sarzotti
et al. 1996; Weiner, 1997).
Lymphocyte subsets, cytokines and immunoregulation
Two different patterns of cytokine secretion by T-cells have
been identified, which lead to different functional responses
and are referred to as Th1 and Th2 T-cells (Mosman &
Coffman, 1989). Th1 cells produce interleukin (IL) 2, inter-
feron-γ (IFN-γ) and tumour necrosis factor-β, which are not
synthesised by Th2 cells. In contrast, Th2 cells (but not Th1
cells) produce IL-4, IL-5, IL-10 and others. Th1 cells
enhance cell-mediated inflammatory activity (such as
delayed hypersensitivity) and induce B-cell antibody-class
switching to immunoglobulin (Ig) G2a, whereas Th2 cells
synthesise cytokines that help B-cells develop into IgE
(mediates immediate hypersensitivity) and IgG1 antibody-
producing cells. There are also T-cells able to produce both
Th1 and Th2 cytokines, referred to as Th0 cells (Mosman &
Abbreviations: EAE, experimental autoimmune encephalomyelitis; IFN-γ, interferon-γ; Ig, immunoglobulin; IL, interleukin; OVA, ovalbumin; PG,
prostaglandins; PUFA, polyunsaturated fatty acids; TGFβ, transforming growth factor β; Th, T-helper.
*Corresponding author: Dr L. S. Harbige, fax +44 20 8331 8305, email L.Harbige@gre.ac.uk

450
L. S. Harbige and B. A. C. Fisher
Sad, 1996). Weiner (1997) has further characterised a Th3
T-cell subset that primarily produces transforming growth-
factor β (TGFβ), provides help for B-cell IgA production,
has potent suppressive properties, and is involved in oral
tolerance. Importantly, the overall balance of cytokine
production by Th1, Th2, and Th3 cells affects the type of
immune response generated (Mosman & Sad, 1996).
and docosapentaenoic (22 : 5n-6) acids, and the n-3 longer-
chain PUFA eicosapentaenoic (20 : 5n-3, e.g. found in
seafoods, fish and fish oils), docosapentaenoic (22 : 5n-3) and
docosahexaenoic (22 : 6n-3, e.g. found in seafoods, fish and
fish oils) acids (Crawford, 1983; Sinclair, 1990). Although
the same enzyme systems are responsible for the alternative
desaturation and elongation of the PUFA families, they have
differing affinities, n-3 > n-6 > n-9 (Sinclair, 1990). Some of
the PUFA containing three, four and five C=C are precursors
for the bioactive eicosanoids (prostaglandins (PG), throm-
boxanes, leukotrienes and lipoxins). The precursor PUFA
20 : 3n-6, 20 : 4n-6 and 20 : 5n-3 are cyclised and incorporate
molecular oxygen, at a specific site in the molecule, to
produce unstable peroxidised intermediates such as PG.
Depending on whether the C20 PUFA precursor is 20 : 3n-6,
20 : 4n-6 or 20 : 5n-3, PGE is referred to as E1, E2, or E3 res-
pectively, and thromboxane A as A1, A2 or A3 respectively;
although there is no PGI1, there is a PGI2 (prostacyclin) and
PGI3. Metabolism by other enzymic routes produces the
leukotrienes such as B4 and B5 by lipoxygenase activity and
the 5-, 12- and 15-hydroxy acids by hydroperoxidase activity
(Sinclair, 1990). Dietary-derived n-6 PUFA and their metab-
olites are essential for the normal development and structural
and functional integrity of the immune system (Clausen &
Moller, 1967; Selivonchick & Johnston, 1975; Scott et al.
1980; Goldyne & Stobo, 1981; Goodwin & Webb, 1981;
Hopkins et al. 1981; Tripp et al. 1986; Shipman et al. 1988;
Roper et al. 1990; Betz & Fox, 1991; Dvorak & Stepankova,
1991; Phipps et al. 1991; Ushikubi et al. 1993; Goetzl et al.
1995; Harbige, 1998). Dietary, PUFA (n-3, n-6 and n-9) are
known to affect the membrane fatty acid composition of cells
of the immune system, and through dietary PUFA manipu-
lation systemic immune system functions as well as
autoimmune disease can be modulated (Calder, 1998;
Harbige, 1998; de Pablo & Alvarez de Cienfuegos, 2000). In
view of the importance of PUFA in immune system functions
we undertook studies to investigate the effects of different
dietary PUFA on tolerogenic immune responses of the gut
mucosal immune system generated via oral dosing with oval-
bumin (OVA).
Metabolism and immunobiology of polyunsaturated
fatty acids
Little is known about the regulatory effects, impact and/or
requirements of the gut mucosal immune system in relation
to the essential macro- and micronutrient components of the
diet. The macronutrient component of the diet includes the
polyunsaturated fatty acids (PUFA). There are several
families of PUFA, the major ones are the n-9, n-6 and n-3
(Sinclair, 1990). Both plants and animals can synthesise the
n-9 parent PUFA oleic acid (18 : 1) from stearic acid (18 : 0)
by the action of the enzyme delta-9-desaturase. Importantly,
plants are able to make the n-3 and n-6 parent PUFA α-
linolenic acid (18 : 3n-3) and linoleic acid (18 : 2n-6) respect-
ively from oleic acid (18 : 1) due to the possession of
appropriate desaturase enzymes. Animals, however, cannot
synthesise the parent n-3 and n-6 PUFA (Sinclair, 1990).
Nevertheless, they can elongate and desaturate the parent n-3
and n-6 PUFA which occur in the diet (Fig. 1), and which are
classically referred to as essential fatty acids (Mead, 1981;
Rivers & Frankel, 1981; Crawford, 1983; Sinclair, 1990).
The n-6 parent PUFA linoleic acid (e.g. found in leafy
vegetables, sunflower- and safflower-seed oils) and the n-3
parent PUFA α-linolenic acid (e.g. found in leafy vegetables
and linseed oil) require delta-6-desaturase enzyme activity
for conversion to γ-linolenic acid (18 : 3n-6, e.g. found in
blackcurrant and borage- (Borago officinalis) seed oils) and
stearidonic acid (18 : 4n-3) respectively (Fig. 1). Further
chain elongation through elongase enzymes and desaturation
by delta-5 and delta-4 desaturation enzymes produce the
longer-chain n-6 metabolites such as dihomo-γ-linolenic
(20 : 3n-6), arachidonic (20 : 4n-6), docosatetraenoic (22 : 4n-6)
Linoleic acid
18 : 2n-6
∆6
18 : 3n-3 α-Linolenic acid
∆6
γ-Linolenic acid
18 : 3n-6
18 : 4n-3 Stearidonic acid
Dihomo-γ-linolenic acid
20 : 3n-6
∆5
20 : 4n-3
∆5
Eicosanoids
Docosanoids
Arachidonic acid
20 : 4n-6
20 : 5n-3 Eicosapentaenoic acid
22 : 4n-6
∆4
22 : 5n-3 Docosapentaenoic acid
∆4
Docosapentaenoic acid
22 : 5n-6
22 : 6n-3 Docosahexaenoic acid
Fig. 1. An outline of essential fatty acid metabolism. Dietary precursor fatty acids are alternatively
desaturated and elongated to produce longer-chain fatty acids with more double bonds. Eicosanoids
are produced by oxygenation via either cyclooxgenase or lipoxygenase enzymic pathways and
include: prostaglandins, e.g. prostaglandin E2; thromboxanes, e.g. thromboxane A2; leukotrienes, e.g.
leukotriene B4; the lipoxins, e.g. lipoxin A4. Docosanoids are less well characterised.

Dietary influences on mucosal immunity
451
30
25
20
15
*
††
10
5
**
*
0
Dietary treatment . . .
Base diet
n-6-rich Borage oil
n-3-rich Fish oil
Fig. 2. Effect of dietary n-6 and n-3 fatty acids on spleen lymphocyte proliferative responses to ovalbumin
(OVA) in both tolerised and non-tolerised animals. BALB/c mice were fed one of three diets; normal chow (low
fat, 2·5% (w/w)), normal chow plus 10% (w/w) n-3 fatty acid-rich fish oil or normal chow plus 10% (w/w) n-6
fatty acid-rich borage oil. After 14 d animals were either tolerised (orally by gavage) with 2 ( ]) or 25(1)mg
OVA respectively or sham-operated (\). After 7d post-tolerisation animals were immunised with OVA in
complete Freund’s adjuvant (100 µg in 50 µl). Spleen cells were obtained 7 d post-immunisation and cultured
in the presence of 1 mg OVA/ml and the lymphoproliferative responses measured as described previously
(Harbige et al. 2000). Results are expressed as a stimulation index (SI), i.e. the mean count/min for
OVA-stimulated cultures divided by the mean count/min for cultures without OVA. Values are means and
standard deviations represented by vertical bars for four animals per group. Mean values within dietary
groups were significantly different: *P<0·05, **P<0·01. Mean value was significantly different from the
corresponding value for base diet: ††P<0·01.
n-6 and n-3 Fatty acids, lymphoproliferative, cytokine
compared with the corresponding control was greatest in the
borage oil group and lowest in the fish oil group (Fig. 3). The
reason for this difference is unclear, but it may be due to
changes in the generation of suppressor cells, even though in
this study there was no discrimination between active
suppression and anergy. Initial proliferation induced by
feeding OVA occurs in the mesenteric lymph nodes, and is
similar in magnitude to that seen in the popliteal lymph nodes
of mice immunised with OVA and complete Freund’s
adjuvant (Williamson et al. 1999; Nagler-Anderson, 2000).
Furthermore, in experimental autoimmune encephalomy-
elitis (EAE) using myelin basic protein-specific T-cell
receptor-transgenic mice, a significant increase in cells
secreting IL-2, IFN-γ, IL-4 and IL-5 is seen in the mesenteric
lymph nodes on day 1 after feeding myelin basic protein
(Benson & Whitacre, 1997). If feeding fish oil reduces this
initial proliferation, then it could be postulated that the
number of Th3 or Tr1 (T-regulatory 1 e.g. produce IL-10,
TGFβ) cells generated would also be fewer. Different mech-
anisms have been shown for suppression of proliferation by
fish oil, including lower major histocompatibility complex
class II expression on dendritic cells and other antigen-
presenting cells, and therefore decreased antigen presen-
tation (Fujikawa et al. 1992; Hughes et al. 1998; Sanderson
et al. 1997). It is also possible that there may be increased
apoptosis rather than simply decreased proliferation
(Fernandes et al. 1996). The increased apoptosis of
lymphocytes observed by Fernandes et al. (1996) in fish
oil-fed mice is likely to be the result of increased oxidative
stress (Buttke & Sandstrom, 1994) induced by the highly-
and tolerogenic responses
BALB/c male mice were fed one of three diets: normal
chow (which is low in fat), normal chow plus 10% (w/w) n-
3 PUFA-rich fish oil or normal chow plus 10% (w/w) n-6
PUFA-rich borage oil. After 14 d the mice in each group
were either tolerised orally by gavage with 2 or 25 mg OVA
to induce low- and high-dose tolerance respectively, or
sham-operated. After 7 d post-tolerisation all mice were
immunised with OVA in complete Freund’s adjuvant.
Spleen cells were harvested from some of the animals at day
7 post immunisation to investigate OVA-specific lympho-
proliferative responses and cytokine production, and at day
14 serum was collected from the remaining animals for
estimation of OVA-specific antibodies.
In all dietary groups high-dose oral tolerance decreased
lymphocyte proliferation to a greater extent than low-dose
tolerance (Fig. 2). In the fish oil group, however, OVA-
specific proliferation was significantly reduced (P<0·01) in
the control group. This finding contrasts with the borage oil
control where there was increased proliferation (Fig. 2). The
decreased proliferation seen in the fish oil group was asso-
ciated with decreased IL-2 production (Fig. 3). These
findings are in accordance with previous work showing that
fish oils decrease lymphoproliferation and IL-2 production in
animal and human studies (Calder, 1996, 1998). In the
present study low-dose tolerance, and to a greater extent
high-dose tolerance, both decreased IL-2 production in all
dietary groups (Fig. 3). However, the percentage reduction

452
L. S. Harbige and B. A. C. Fisher
100
80
60
40
20
*
**
**
0
Dietary treatment . . .
Base diet
n-6-rich Borage oil
n-3-rich Fish oil
Fig. 3. Effect of dietary n-6 and n-3 fatty acids on ovalbumin (OVA)-specific interleukein (IL) 2 production by
spleen lymphocytes in response to OVA in both tolerised and non-tolerised animals. BALB/c mice were fed
one of three diets; normal chow (low fat, 2·5 % (w/w)), normal chow plus 10 % (w/w) n-3 fatty acid-rich fish oil
or normal chow plus 10 % (w/w) n-6 fatty acid-rich borage oil. After 14 d animals were either tolerised (orally
by gavage) with 2 (]) or 25(1)mg OVA respectively or sham-operated (\). After 7 d post-tolerisation
animals were immunised with OVA in complete Freund’s adjuvant (100 µg in 50 µl). Spleen cells were
obtained 7 d post-immunisation and cultured in the presence of 1 mg OVA/ml for 24 h and the supernatant
fractions removed and stored at −70°C. Mouse IL-2 was measured in supernatant fractions using a
commercially available ELISA system (Harbige et al. 2000). Values are means and standard deviations
represented by vertical bars for four animals per group. Mean values within dietary groups were significantly
different: *P<0·05, **P<0·02.
peroxidisable long-chain n-3 PUFA (Harbige et al. 1990;
Avula et al. 1999).
borage oil, then there are important implications for the
study of oral tolerance in autoimmune disease. Since a wide
variety of autoantigens are implicated in many autoimmune
diseases through epitope spreading, the concept of bystander
suppression is pivotal, whereby tolerised cells are able to
suppress responses to other antigens in their vicinity,
through the elaboration of cytokines such as IL-10 and
TGFβ. Similar mechanisms could explain our observations
in EAE where n-6 PUFA-rich seed oils are protective and
n-3 PUFA-rich fish oils increase severity and delay recovery
(Harbige, 1993, 1998; Harbige et al. 1995, 2000). Natural
recovery in EAE is mediated by expansion of suppressor
lymphoid cells (Adda et al. 1977), some of which have been
characterised as TGFβ-producing CD4+ T cells by Karpus
& Swanborg (1991). The protective effect of borage oil in
EAE is linked to increased TGFβ transcription and
production by T-cells and/or monocytes and/or PGE2-
producing monocytes (Harbige et al. 2000). The increased
severity and delayed recovery in EAE induced by feeding
fish oil may be the consequence of an inhibitory effect on
the proliferative expansion of the suppressor cells (or
decrease in PGE2 from suppressor monocytes?) involved in
the natural recovery from EAE. Given the immuno-
suppressive effects of fish oils and their suppressive effects
in spontaneous autoantibody-mediated disease (Harbige,
1998), why there should be lack of suppression in EAE in
the first place is difficult to explain. One possibility is that
the potent immunogenic nature of the central nervous
Immunosuppression following fish oil feeding has been
shown in several animal studies. For example, diets rich in
fish oil increase the mortality of mice to whom Salmonella
typhimurium has been administered orally (Chang et al.
1992), and prolong graft survival (Grimm et al. 1998). In
contrast, the effects of n-6 fatty acids are less clear. The lack
of lymphoproliferative suppression in the borage oil control
(non-tolerised) group suggests a maintenance of response to
novel antigens, whereas there are enhanced effects of oral
tolerance on lymphoproliferation and IL-2, i.e. with low
dose. It is therefore interesting to note that in the borage oil
control group the production of IFN-γ was significantly
(P<0·05) enhanced (LS Harbige and BAC Fisher, unpub-
lished results). This finding is also consistent with those of
previous studies in which feeding borage oil was observed
to be associated with a trend of increasing IFN-γ production,
despite suppressing EAE, a CD4+ Th1-mediated disease
(Harbige et al. 2000; LS Harbige, L Layward, MM Morris-
Downes, DC Dumonde and S Amor, unpublished results).
These findings could be further tested by deriving lympho-
cytes (including Peyer’s patch and mesenteric lymphocytes)
from orally-tolerised mice fed n-3 or n-6 PUFA, and deter-
mining the concentration needed to suppress the antigen-
specific proliferative response of cells from animals
immunised with OVA. If fewer suppressor cells are
generated through feeding fish oil and more with feeding

Dietary influences on mucosal immunity
453
system antigen mixture with adjuvant used to induce EAE
may be sufficient to overcome any mild immunosuppressive
effect of fish oil, as milder EAE disease induction can be
suppressed by fish oil (Mertin, 1993).
PUFA feeding (Calder, 1998) and the increase in PGE2
production by n-6 PUFA (Harbige et al. 2000) could give
differential immunological responses. Interestingly, it has
been shown recently that the maternal dietary n-6:n-3
balance can influence the intestinal functions/ontogeny of
the offspring (Jarocka-Cyrta et al. 1998). In the context of
‘tolerance’ the Jarocka-Cyrta et al. (1998) findings not only
indicate that modification of intestinal function by dietary
PUFA might be important, but that there may also be
important effects on the fetal and neonatal periods of devel-
opment. Dietary PUFA effects on tolerance and thymic
education during the fetal and neonatal period could,
therefore, have implications for the development of immu-
nological disorders, and would be in keeping with current
views on nutrition and ‘fetal programming’ (Barker, 1998).
Furthermore, it is interesting to note that human breast milk
contains both preformed 20 : 4n-6 and 22 : 6n-3, whereas
formula milks do not (Drury & Crawford, 1990; Crawford
et al. 1997)
In the PUFA and oral-tolerance studies with OVA our
preliminarly findings indicate that we are unable to detect
OVA-specific TGFβ production. However, it is important to
note that we have only used spleen-derived lymphoid cells,
as other researchers have been able to detect TGFβ-
secreting cells from Peyer’s patches but not from popliteal
lymph nodes in BALB/c mice (Shi et al. 1999). This finding
may be due to there being low numbers of antigen-specific
TGFβ-secreting cells in the periphery. It is also possible to
induce oral tolerance in TGFβ1-null mice with both low-
and high-OVA doses (Barone et al. 1998), but these animals
are abnormal, requiring injections of LFA-1 (leucocyte
function-associated antigen-1) for survival, and may not
therefore reflect normal immune system physiology.
There are further conceptual reasons why active
suppression:apparent anergy might differ between the
dietary n-6 and n-3 PUFA groups. PG have important roles
in mucus secretion and in maintaining small intestine
epithelial tight junctions (for review, see Mohajer & Ma,
2000). Alteration of the eicosanoid milieu by alteration of
the proportions of precursor PUFA might therefore affect
the amount of tolerogen reaching the peripheral circulation
where anergy is induced. In addition, PGE2 is important in
lymph node ‘shutdown’, allowing an adequate proliferative
response to antigen within the lymph node (Hopkins et al.
1981). Thus, the known decrease in PGE2 production by n-3
n-6 and n-3 Fatty acids, antibody and tolerogenic
responses
The effects of n-3 PUFA-rich fish oil and n-6 PUFA-rich
borage oil on OVA-specific antibody production was inves-
tigated as described earlier. In all dietary groups IgG1
production was decreased by high-dose tolerance to a
greater extent than by low-dose tolerance (Fig. 4). The most
interesting finding, however, was that OVA-specific IgG1
was significantly reduced (P<0·01) in the borage oil control
.
0 8
.
0 7
*
.
0 6
††
††
.
††
0 5
††
††
0.4
††
*
.
0 3
.
0 2
.
0 1
.
0 0
Dietary treatment . . .
Base diet
n-6-rich Borage oil
n-3-rich Fish oil
Fig. 4. Effect of dietary n-6 and n-3 fatty acids on ovalbumin (OVA)-specific serum (1 : 100 dilution) immuno-
globulin G1 in both OVA-tolerised and non-tolerised animals. BALB/c mice were fed one of three diets; normal
chow (low fat, 2·5 % (w/w)), normal chow plus 10 % (w/w) n-3 fatty acid-rich fish oil or normal chow plus 10 %
(w/w) n-6 fatty acid-rich borage oil. After 14 d animals were either tolerised (orally by gavage) with 2 (]) or
25(1)mg OVA respectively or sham-operated (\). After 7 d post-tolerisation animals were immunised with
OVA in complete Freund’s adjuvant (100 µg in 50 µl). Serum was collected 14 d post-immunisation and OVA-
specific subclass antibodies detected using an ELISA format. Values are means and standard deviations
represented by vertical bars for four animals per group. Mean values within groups were significantly different:
*P<0·05. Mean values were significantly different from the corresponding value for the base group diet:
††P<0·01.

454
L. S. Harbige and B. A. C. Fisher
.
1 2
.
1 0
.
0 8
0.6
*
*
*
††
*
.
0 4
.
0 2
.
0 0
Dietary treatment . . .
Base diet
n-6-rich Borage oil
n-3-rich Fish oil
Fig. 5. Effect of dietary n-6 and n-3 fatty acids on ovalbumin (OVA)-specific serum (1 : 100 dilution) immuno-
globulin G2a in both OVA-tolerised and non-tolerised animals. BALB/c mice were fed one of three diets;
normal chow (low fat, 2·5% (w/w)), normal chow plus 10% (w/w) n-3 fatty acid-rich fish oil or normal chow
plus 10 % (w/w) n-6 fatty acid-rich borage oil. After 14 d animals were either tolerised (orally by gavage) with
2 (]) or 25(1)mg OVA respectively or sham-operated (\). After 7 d post-tolerisation animals were
immunised with OVA in complete Freund’s adjuvant (100 µg in 50 µl). Serum was collected 14 d post-
immunisation and OVA-specific subclass antibodies detected using an ELISA format. Values are means and
standard deviations represented by vertical bars for four animals per group. Mean values within groups were
significantly different: *P<0·05. Mean value was significantly different from the corresponding value for the
base group diet: ††P<0·01.
group, and to a greater extent in the fish oil control (Fig. 4).
IgG1 class switching is mediated by IL-4, and is part of the
Th2 response. The implication is that fish oil, and to a lesser
extent borage oil, inhibit Th2 responses. This finding is in
contrast to some studies that have shown increased
production of IgE to OVA in rats fed on a high fish oil diet
(Calder, 1998). This inconsistency may be due to the
amount of PUFA fed or species differences, or indicate that
there is a differential regulation of IgG1 and IgE in the
model used. The decrease in IgG1 could result from B-cell
apoptosis (see Fig. 2) or from decreased production. The
effect on IgG2a production is less marked, although low-
dose tolerance appears to be enhanced in the fish oil group
(Fig. 5). Taking into account the decreased IL-2 with fish oil
feeding (all groups), our data support the view that fish oil
feeding results in suppression of both Th1 and Th2
responses. In contrast to our findings with n-6 PUFA-rich
borage oil, Cinader et al. (1983) found loss of ‘suppressor
activity’ to the tolerance inducibility of rabbit γ-globulin in
mice fed on an n-6 linoleic acid-rich diet. The difference
between the Cinader et al. (1983) observations and our
findings is likely to be due to the use of the parent n-6 PUFA
linoleic acid, rather than its desaturated product which is
rapidly metabolised to longer-chain n-6 PUFA, i.e. 20 : 3n-6
and 20 : 4n-6 (Harbige et al. 1995; Harbige, 1998).
physical intestinal barrier and the immune cells themselves.
We have shown that dietary lipids can modulate gut
mucosal-induced immune responses. Long-chain n-3
PUFA-rich fish oil decreases both Th1- and Th2-like
responses with and without oral tolerogen. Fish oils rich in
long-chain n-3 PUFA can be viewed as being generally
immunosuppressive at high doses, which is in accordance
with most published studies (Calder, 1998; Harbige, 1998).
One of the mechanisms behind some of the effects of fish oil
appears to be peroxide-induced apoptosis of lymphocytes. It
would be interesting, therefore, to know if vitamin E admin-
istration could reverse fish oil-induced lymphocyte
apoptosis and the fish oil-induced increase in disease
severity and delayed recovery in EAE. In contrast, n-6
PUFA-rich borage oil has a more complex role. There is an
increase in proliferative response (in control OVA-immu-
nised but not tolerised animals) to antigen with a
concomitant increase in IFN-γ production (Th1 response).
In addition, borage oil decreases the level of IgG1 (in all
groups; a Th2 response) and enhances the tolerogenic
response to OVA, i.e. decreased proliferative and IL-2
responses, which suggests increased suppressor cell or
Th3-like activity. It appears, therefore, that n-6 PUFA-rich
borage oil can increase Th1-like responses, and possibly
Th3-like responses, and decrease Th2-like responses, and
may therefore be considered immunoregulatory (immune-
deviation?) rather than immunosuppressive in our experi-
mental system. Other mechanisms may also be important in
PUFA and eicosanoid effects on gut-induced tolerance;
e.g. intestinal functions such as barrier, absorptive and
Conclusion
The intestinal immune system is a complex interaction
between factors such as the diet, the local microflora, the

Dietary influences on mucosal immunity
455
Chin J, Turner B, Barchia I & Mullbacher A (2000) Immune
response to orally consumed antigens and probiotic bacteria.
Immunology and Cell Biology 78, 55–66.
Cinader B, Clandinin MT, Hosokawa T & Robblee NM (1983)
Dietary fat alters the fatty acid composition of lymphocyte
membranes and the rate at which suppressor capacity is lost.
Immunology Letters 6, 331–337.
Clausen J & Moller J (1967) Allergic encephalomyelitis induced
by brain antigen after deficiency in polyunsaturated fatty
acids during myelination. Acta Neurologica Scandinavica 43,
375–388.
Crawford MA (1983) Background to essential fatty acids
and their prostanoid derivatives. British Medical Bulletin 39,
210–213.
Crawford MA, Costeloe K, Ghebremeskel K, Phylactos A,
Skirvin L & Stacey F (1997) Are deficits of arachidonic and
docosahexaenoic acids responsible for neural and vascular
complications of preterm babies? American Journal of Clinical
Nutrition 66, Suppl., 1032S–1041S.
de Pablo MA and Alvarez de Cienfuegos (2000) Modulatory effects
of dietary lipids on immune system functions. Immunology and
Cell Biology 78, 31–39.
processing functions and mesenteric lymph node shutdown.
Dietary PUFA could also have important effects during the
fetal and neonatal periods in relation to tolerance induction
and thymic education.
The disappointing results of oral tolerance trials in estab-
lished human autoimmune disease may require more than
manipulation of antigen dose. Indeed, whey proteins from
cow’s milk, when given in single doses equivalent to a toler-
ogenic dose of OVA, are able to induce active immunity and
tolerance in the same animal (Strobel & Mowet, 1998).
Furthermore, the gut microflora and their products have
been shown to influence the induction and maintenance of
oral tolerance (Wardrop & Whitacre, 1999; Isolauri et al.
2001), and the gut microflora is known to vary between
individuals. The impact of diet on the human gut microflora
is also largely unknown. Any therapeutic approach might
therefore consider using multiple targets and exploring, for
example, the use of oral tolerance with probiotic and/or
PUFA manipulation.
Drury PJ & Crawford MA (1990) Essential fatty acids in
human milk. In Clinical Nutrition of the Young Child, vol. 2,
pp. 289–312 [A Ballabring, O Brunser, J Dobbing, M Gracey and
J Senterre, editors]. New York: Nestec Ltd, Raven Press Ltd.
Dvorak B & Stepankova R (1992) Effects of dietary essential fatty
acid deficiency on the development of the rat thymus and
immune system. Prostaglandins Leukotrienes and Essential
Fatty Acids 46, 183–190.
Fernandes G, Chandrasekar B, Luan X & Troyer DA (1996)
Modulation of antioxidant enzymes and programmed cell death
by n-3 fatty acids. Lipids 31, Suppl., S91–S96.
Forsthuber T, Yip HC & Lehmann PV (1996) Induction of TH1 and
TH2 immunity in neonatal mice. Science 271, 1728–1730.
Friedman A & Weiner HL (1994) Induction of anergy or active
suppression following oral tolerance is determined by antigen
dosage. Proceedings of the National Academy of Sciences USA
91, 6688–6692.
Acknowledgements
The authors would like to thank the Henry Smith Charity,
the Multiple Sclerosis Society of Great Britain and Northern
Ireland and the Wellcome Trust for support.
References
Adda DH, Beraud E & Depieds R (1977) Evidence for suppressor
cells in Lewis rats’ experimental allergic encephalomyelitis.
European Journal of Immunology 7, 620–623.
Avula CP, Zaman AK, Lawrence R & Fernandes G (1999)
Induction of apoptosis and apoptotic mediators in Balb/C splenic
lymphocytes by dietary n-3 and n-6 fatty acids. Lipids 34,
921–927.
Bailey M, Plunkett FJ, Rothkotter HJ, Vega-Lopez MA, Haverson
K & Stokes CR (2001) Regulation of mucosal immune responses
in effector sites. Proceedings of the Nutrition Society 60,
427–435.
Fujikawa M, Yamashita N, Yamazaki K, Sugiyama E, Suzuki H &
Hamazaki T (1992) Eicosapentaenoic acid inhibits antigen-
presenting cell function of murine splenocytes. Immunology 75,
330–335.
Barker DJ (1998) In utero programming of chronic disease. Clinical
Science 95, 115–128.
Garside P, Steel M, Liew FY & Mowat AM (1995a) CD4+ but not
CD8+ T cells are required for the induction of oral tolerance.
International Immunology 7, 505–504.
Barone KS, Tolarova DD, Ormsby I, Doetschman T & Michael JB
(1998) Induction of oral tolerance in TGF-β1 null mice. Journal
of Immunology 161, 154–160.
Benson JM & Whitacre CL (1997) The role of clonal deletion and
anergy in oral tolerance. Research in Immunology 148, 533–541.
Betz M & Fox BS (1991) Prostaglandin E2 inhibits production of
Th1 lymphokines but not of Th2 lymphokines. Journal of
Immunology 146, 108–113.
Burstein HJ & Abbas AK (1993) In vivo role of interleukin 4 in T
cell tolerance induced by aqueous protein antigen. Journal of
Experimental Medicine 177, 457–463.
Buttke TM & Sandstrom PA (1994) Oxidative stress as a mediator
of apoptosis. Immunology Today 15, 7–10.
Calder PC (1996) Immunomodulatory and anti-inflammatory
effects of n-3 polyunsaturated fatty acids. Proceedings of the
Nutrition Society 55, 737–774.
Calder PC (1998) Dietary fatty acids and lymphocyte functions.
Proceedings of the Nutrition Society 57, 487–502.
Chang HR, Dulloo AG, Vladoianu IR, Piguet PF, Arsenijevic D,
Girardier L & Pechere JC (1992) Fish oil decreases natural
resistance of mice to infection with Salmonella typhimurium.
Metabolism 41, 1–2.
Garside P, Steel M, Worthey EA, Satoskar A, Alexander J,
Bluethmann H, Liew FY & Mowat AM (1995b) T helper 2 cells
are subject to high dose oral tolerance and are not essential for its
induction. Journal of Immunology 154, 5649–5655.
Grimm H, Grimminger F, Korom S & Seeger W (1998) Use of fish
oil to prevent graft rejection. Proceedings of the Nutrition Society
57, 577–585.
Goodwin JS & Webb DR (1981) Regulation of the immune
response by prostaglandins. Clinical Immunology and Immuno-
pathology 15, 106–122.
Goetzl EJ, An S & Smith WL (1995) Specificity of expression and
effects of eicosanoid mediators in normal physiology and human
disease. FASEB Journal 9, 1051–1058.
Goldyne ME & Stobo JD (1981) Immunoregulatory role of prosta-
glandins and related lipids. CRC Critical Reviews in Immunology
2, 189–223.
Harbige LS (1993) Essential fatty acids, experimental autoimmune
encephalomyelitis and multiple sclerosis. PhD Thesis, Brunel
University.

456
L. S. Harbige and B. A. C. Fisher
Harbige LS (1998) Dietary n-6 and n-3 fatty acids in immunity and
autoimmune disease. Proceedings of the Nutrition Society 57,
555–562.
Harbige LS, Ghebremeskel K, Williams G & Summers P (1990)
N-3 and N-6 phosphoglyceride fatty acids in relation to hydrogen
peroxide induced in vitro erythrocyte haemolysis in captive
marmosets (Callithrix jacchus) Comparative Biochemistry and
Physiology 97B, 167–170.
Harbige LS, Layward L, Morris-Downes MM, Dumonde DC &
Amor S (2000) The protective effects of omega-6 fatty acids in
experimental autoimmune encephalomyelitis (EAE) in relation
to transforming growth factor-beta 1 (TGF-β1) up-regulation and
increased prostaglandin E2 (PGE2) production. Clinical and
Experimental Immunology 122, 445–452.
Nagler-Anderson C (2000) Tolerance and immunity in the
intestinal immune system. Critical Reviews in Immunology 20,
103–120.
Phipps RP, Stein SH & Roper RL (1991) A new view of prosta-
glandin E regulation of the immune response. Immunology
Today 12, 349–352.
Ridge JP, Fuchs EJ & Matzinger P (1996) Neonatal tolerance
revisited: turning on newborn T cells with dendritic cells. Science
271, 1723–1726.
Rivers JPW & Frankel TL (1981) Essential fatty acid deficiency.
British Medical Bulletin 37, 59–64.
Roper LR, Conrad DH, Brown DM, Warner GL & Phipps R (1990)
Prostaglandin E2 promotes IL-4-induced IgE and IgG1 synthesis.
Journal of Immunology 145, 2644–2651.
Harbige LS, Yeatman N, Amor S & Crawford MA (1995)
Prevention of experimental autoimmune encephalomyelitis in
Lewis rats by a novel source of γ-linolenic acid. British Journal
of Nutrition 74, 701–715.
Heyman M (2001) How dietary antigens access the mucosal
immune system. Proceedings of the Nutrition Society 60,
419–426.
Hopkins J, McConnell I & Pearson JD (1981) Lymphocyte traffic
through antigen-stimulated lymph nodes II: role of prostaglandin
E2 as a mediator of cell shutdown. Immunology 42, 225–231.
Hughes DA (1998) In vitro and in vivo effects of n-3 poly-
unsaturated fatty acids on human monocyte function.
Proceedings of the Nutrition Society 57, 521–525.
Sanderson P, MacPherson G, Jenkins CH & Calder PC (1997)
Dietary fish oil diminishes the antigen presentation activity of rat
dendritic cells. Journal of Leukocyte Biology 62, 771–777.
Sarzotti M, Robbins DS & Hoffman PM (1996) Induction of
protective CTL responses in newborn mice by a murine retro-
virus. Science 271, 1726–1728.
Scott WA, Zrike JM, Hamill AL, Kempe J & Cohn ZA (1980)
Regulation of arachidonic acid metabolites in macrophages.
Journal of Experimental Medicine 152, 324–335.
Selivonchick DP & Johnston PV (1975) Fat deficiency in rats
during development of the central nervous system and suscepti-
bility to experimental allergic encephalomyelitis. Journal of
Nutrition 105, 288–300.
Isolauri E, Sutas Y, Kankaanpaa P Arvilommi H & Salminen S
(2001) Probiotics: effects on immunity. American Journal of
Clinical Nutrition 73, Suppl., 444S–450S.
Jarocka-Cyrta E, Perin N, Keelan M, Wierzbicki E, Wierzbicki T,
Clandinin MT & Thomson ABR (1998) Early dietary experience
influences ontogeny of intestine in response to dietary lipid
changes in later life. American Journal of Physiology 275,
G250–G258.
Shi HN, Grusby MJ & Nagler-Anderson CN (1999) Orally
induced peripheral nonresponsiveness is maintained in the
absence of functional Th1 or Th2 cells. Journal of Immunology
162, 5143–5148.
Shipman PM, Schmidt RR & Chepenik KP (1988) Relation
between arachidonic acid metabolism and development of
thymocytes in fetal thymic organ cultures. Journal of
Immunology 140, 2714–2720.
Karpus WJ & Swanborg RH (1991) CD4+ suppressor cells inhibit
the function of effector cells of experimental autoimmune
encephalomyelitis through a mechanism involving transforming
growth factor-β. Journal of Immunology 146, 1163–1168.
Kolb H & Pozzilli P (1999) Cow’s milk and type I diabetes: the
gut immune system deserves attention. Immunology Today 20,
108–110.
Lamont AG, Mowat AM, Browning MJ & Parrott DMV (1988)
Genetic control of oral tolerance to ovalbumin in mice.
Immunology 63, 737–739.
Matsuzaki T & Chin D (2000) Modulating immune responses with
probiotic bacteria. Immunology and Cell Biology 78, 67–73.
Meade JF (1981) The essential fatty acids: past, present and future.
Progress in Lipid Research 20, 1–6.
Melamed D & Friedman A (1993) Direct evidence for anergy in
T lymphocytes tolerized by oral administration of ovalbumin.
European Journal of Immunology 23, 935–942.
Sinclair HM (1990) Essential fatty acids – an historical perspective.
Transactions of the Biochemical Society 18, 756–761.
Strobel S (2001) Immunity induced after a feed of antigen during
early life: oral tolerance versus sensitisation. Proceedings of the
Nutrition Society 60, 437–442.
Strobel S & Mowat AM (1998) Immune responses to dietary
antigens: oral tolerance. Immunology Today 19, 173–181.
Tripp CS, Wyche A, Unanue ER & Needleman P (1986) The
functional significance of the regulation of macrophage Ia
expression by endogenous arachidonate metabolites in vitro.
Journal of Immunology 137, 3915–3920.
Ushikubi F, Aiba Y, Nakamura K, Namba T, Hirata M, Mazda O,
Katsura Y & Narumiya S (1993) Thromboxane A2 receptor is
highly expressed in mouse immature thymocytes and mediates
DNA fragmentation and apoptosis. Journal of Experimental
Medicine 178, 1825–1830.
Vidard L, Colarusso LJ & Benacerraf B (1995) Specfic T-cell
tolerance may refelct selective activation of lymphokine
synthesis. Proceedings of the National Academy of Sciences USA
92, 2259–2262.
Mertin J (1983) Omega-6 and omega-3 polyunsaturates and the
immune system. British Journal of Clinical Practice 31, Suppl.,
111–114.
Mohajer B & Ma TY (2000) Eicosanoids and the small intestine.
Prostaglandins and Other Lipid Mediators 61, 125–143.
Mosman TR & Coffman RL (1989) Th1 and Th2 cells: different
patterns of lymphokine secretion lead to different functional
properties. Annual Review of Immunology 7, 145–173.
Mosman TR & Sad S (1996) The expanding universe of T cell
subsets: Th1, Th2 and more. Immunology Today 17, 138–146.
Mowat AM, Strobel S, Drummond HE & Ferguson A (1982)
Immunological responses to fed protein antigens in mice. 1.
Reversal of oral tolerance to ovalbumin by cyclophosphamide.
Immunology 45, 105–113.
Wardrop RM & Whitacre CC (1999) Oral tolerance in the treatment
of inflammatory autoimmune diseases. Inflammation Research
48, 106–119.
Weiner HL (1997) Oral tolerance: immune mechanisms and
the treatment of autoimmune diseases. Immunology Today 18,
335–343.
Williamson E, O’Malley JM & Viney JL (1999) Visualizing the
T-cell response elicited by oral administration of soluble protein
antigen. Immunology 97 565–572.