Monty :
For the present, I would like to request that you put aside for the
present the major 'controversial' questions about detailed membrane
structure;
basic or conditional essentiality; and the fragility of PUFA.

Supposing you will grant that, here is my question which I feel will
be of significant interest to the group.

Assuming that current mainstream recommendations for n6-n3 levels
(and  ratio) are observed, there appears to be a large amount of
accumulated evidence indicating significant health benefits.
Please *notice* that I am not saying a typical western values for
these, but currently "expert" international dietary recommendations.
Please assume that the suggested PUFA levels are obtained from both
natural
and supplementary sources, and that the sources are pure and have
not been cooked or otherwise overheated.

The following 3 year old "Minireview" summarizes the so called
benefits in a compact form, with references. The full text version
is also
available at no charge if you will open the JBC link.

I respectfully request that you take your time and give us a
step by step critique of each item, indicating where these
researchers are correct or may have missed the boat, especially in
clinical terms.
For my part I promise not to poke fun at your comments, (as I have
been known to do) and to approach them as one who has a sincere wish
to learn and benefit from all points of view.
Again, would you please make the focus the presence or absence of
health benefits. Pointing out long term risks is of course expected.
Thanks
mikeV

**************
Full Text Article: Journal of Biological Chemistry:
http://www.jbc.org/cgi/content/full/277/11/8755

Originally published In Press as doi:10.1074/jbc.R100062200 on
December 17, 2001
J. Biol. Chem., Vol. 277, Issue 11, 8755-8758, March 15, 2002

MINIREVIEW
The Biochemistry of n-3 Polyunsaturated Fatty Acids*,

Donald B. Jump
From the Departments of Physiology, Biochemistry, and Molecular
Biology, Michigan State University, East Lansing, Michigan 48824

Dietary n-3 polyunsaturated fatty acids (n-3 PUFA)1 have effects on
diverse physiological processes impacting normal health and chronic
disease, such as the regulation of plasma lipid levels (1-4),
cardiovascular (5-7) and immune function (8), insulin action (9,
10), and neuronal development and visual function (11) (see Table I,
supplemental material). Ingestion of n-3 PUFA will lead to their
distribution to virtually every cell in the body with effects on
membrane composition and function, eicosanoid synthesis, and
signaling as well as the regulation of gene expression (11-14).
However, cell-specific lipid metabolism as well as the expression of
fatty acid-regulated transcription factors likely plays an important
role in determining how cells respond to changes in PUFA
composition. In this minireview I will highlight some of the recent
advances in our understanding of n-3 PUFA effects on cells with an
emphasis on those mechanisms likely to have a broad physiological
impact.

         Synthesis and Metabolism of n-3 and n-6 PUFA in Mammals

n-3 and n-6 PUFA are the two major classes of PUFA encountered in
the diet, and both classes of fatty acids are required for normal
human health (15). Linoleic acid (18:2n-6) is the predominant
plant-derived dietary PUFA and is a precursor for arachidonic acid
(20:4n-6) and eicosanoids (see Fig. 1, supplemental
material). -Linolenic acid (18:3n-3) is the predominant
plant-derived dietary n-3 PUFA and is a precursor for 22:6n-3.
Linoleic acid, 20:4n-6 and 22:6n-3, are prominent PUFA in cellular
phospholipids (11).

Non-esterified fatty acids (NEFA) enter cells via fatty acid
transporters and are rapidly converted to fatty acyl-CoA thioesters
(FA-CoA) by acyl-CoA synthetases (see Fig. 2, supplemental
material). Intracellular NEFA and FA-CoA are low (<10 ?M), and a
major fraction of these lipids is bound to specific proteins, i.e.
fatty acid-binding protein and FA-CoA-binding protein. FA-CoAs are
substrates for neutral lipid (triglycerides, cholesterol esters) and
polar lipid (phospholipids (PS, PE, PC), sphingolipids, and
plasmalogens) synthesis as well as elongation,
desaturation, -oxidation, and protein acylation reactions.

22:6n-3 is the most abundant n-3 PUFA in most tissues and is found
at the sn-2 position of phospholipids (11, 16). Deficiencies of n-3
PUFA lead to a loss of 22:6n-3 from brain and retina rod outer
segment (ROS) phospholipids with a compensatory replacement by
22:5n-6 (11, 17-19). This minor change in membrane phospholipid
structure is sufficient to lead to memory loss, learning
disabilities, and impaired visual acuity. Metabolic studies with
healthy humans have shown that in contrast to 20:5n-3, 18:3n-3 is
not efficiently converted to 22:6n-3. 18:3n-3 is preferentially
utilized by the skin and is more rapidly oxidized than the 20- and
22-carbon n-3 PUFA (20). 22-Carbon PUFA requires prior
peroxisomal -oxidation before entering the mitochondrial -oxidation
spiral (21). This metabolic partitioning decreases the availability
of 18:3n-3 for conversion to the 20- and 22-carbon PUFA in the liver
(20).

In rodents maintained on chow diets, 18:3n-3 and 20:5n-3 are minor
PUFAs in the phospholipid fraction. However, supplementing diets
with fish oil, a rich source of 20:5n-3 and 22:6n-3, significantly
increases tissue levels of 20:5n-3, 22:5n-3, and 22:6n-3; these
changes occur at the expense of 20:4n-6. Such diets also induce
hepatic microsomal, peroxisomal, and mitochondrial fatty acid
oxidation while suppressing fatty acid synthesis (12). Differences
in how 18- versus 20- and 22-carbon n-3 PUFA are metabolized in
cells likely contribute to their effects on cellular regulatory
processes. These effects extend beyond differential -oxidation to
include PUFA assimilation into neutral lipids. CoA thioesters of
20:5n-3 are poor substrates for diacylglycerol acyltransferase, the
last step in triglyceride synthesis (22-24). Neither 20- nor
22-carbon n-3 PUFAs are good substrates for cholesterol ester
synthesis.2 Thus, the unique properties of 20:5n-3 and 22:5n-3
likely affect intracellular NEFA or CoA thioester levels, factors
that will impact several regulatory mechanisms (see Fig. 2,
supplemental material). Sorting out these mechanisms represents one
of the major challenges in this field.

         n-3 PUFA Effects on Membrane Structure and Function
     TOP
     INTRODUCTION
     Synthesis and Metabolism of...
     n-3 PUFA Effects on...
     n-3 PUFA Effects on...
     n-3 PUFA Effects on...
     Concluding Remarks
     REFERENCES

n-3 PUFA Effects on the Retina and CNS-- The retina contains very
high levels of 22:6n-3. In fact, ~50% of all acyl chains in the ROS
phospholipids (both sn-1 and sn-2) are 22:6n-3 (in PC, PE, and PS).
Minor phospholipids, like phosphatidylinositol and phosphatidic
acid, contain predominantly 20:4n-6 (25). Thus, ROS represents an
excellent model to define the role of 22:6n-3 in membrane structure
and function. Rhodopsin is an integral membrane protein in the ROS.
When excited by light, rhodopsin enters a metastable equilibrium
between two conformation states, i.e. MI and MII (26-28). MII binds
and activates the G-protein, transducin (Gt), and catalyzes a
GDP-GTP exchange that activates cGMP-specific phosphodiesterase.
Activated phosphodiesterase catalyzes cGMP hydrolysis and triggers
closure of the cGMP-gated Na+/Ca2+ channels leading to
hyperpolarization of the ROS plasma membrane and the visual
response.

Using reconstituted membranes, Litman and colleagues (26-28)
reported that the equilibrium constant (Keq) for the formation of
MII is related directly to the degree of phospholipid acyl chain
unsaturation. Phospholipids with one or more 22:6n-3 acyl chains
increase both the formation of MII and binding to Gt. Moreover,
22:6n-3-containing phospholipids attenuated the inhibitory effect of
cholesterol on both the formation of MII and MII-Gt association.
Thus membrane composition plays a critical role in the temporal
response of the G-protein-coupled signaling system in the retina
ROS.

n-3 PUFA deficiency is associated with memory loss and diminished
cognitive function (11, 17-20). Two observations might account for
this effect. First, 22:6n-6 activates RXR in cultured neuronal cells
(29). RXR is a heterodimer partner to many class II nuclear
receptors, some of which, like T3 receptors, have a major impact on
CNS development. Astrocytes, but not neurons, make 22:6n-3, thus
providing a potential source of 22:6n-3 to activate RXR (15).
Second, n-3 PUFA deficiency is correlated with a decline in PS and
increased neuronal apoptosis (11, 30, 31). Neuronal apoptosis is
linked to Raf-1 and plasma membrane PS content. Neuro 2A cells
treated with 22:6n-3 increase PS in the inner leaflet of the plasma
membrane and enhance Raf-1 translocation to the membrane. Raf-1
association with the plasma membrane down-regulates caspase-3
activity and prevents apoptotic cell death (31).

n-3 PUFA Effects on Lipid Rafts-- Outside the CNS, retina, and
testes, 22:6n-3 rarely exceeds 10% of the total fatty acid in
membrane phospholipids; 20:5n-3 and 22:5n-3 are at even lower
levels. However, treating cells with 20:5n-3 increases 20:5n-3,
22:5n-3, and possibly 22:6n-3 in both the phospholipid and protein
component of membranes. Lipids rafts are one target for n-3 PUFA
effects on membrane function. Lipid rafts are regions within the
exoplasmic leaflet of the plasma membrane that are enriched in
cholesterol and sphingolipids (32). Rafts selectively incorporate
proteins and govern protein-protein and protein-lipid interactions.
These membrane microdomains contribute to the structure and function
of caveolae, plasma membrane invaginations, signal transduction,
endocytosis, transcytosis, and cholesterol trafficking as well as
tyrosine kinase and sphingolipid cell signaling. Proteins acylated
with saturated fatty acids (14:0 or 16:0) partition into the inner
leaflet of the plasma membrane with high affinity for rafts, perhaps
because of the unusually long saturated acyl chains on sphingolipids
(33). In some cases palmitoylation of proteins requires prior
myristoylation, and palmitoylation is required for membrane
targeting. A number of proteins involved in cell signaling are found
in lipid rafts, including G-proteins (s, q), members of the Src
kinase family, caveolin, and Gap43. Lipid rafts are sensitive to
modification by PUFA. These modifications occur to both the
phospholipid component as well as to proteins associated with rafts.

A well defined model for studying PUFA effects on lipid rafts is
T-cell activation (33-37). Both n-3 and n-6 PUFAs affect raft
composition and function through an eicosanoid-independent mechanism
(35). n-3 PUFAs (20:5n-3 and 22:6n-3) are used clinically as
immunosuppressive agents because they rapidly alter cellular
phospholipid compositions without enhancing inflammatory eicosanoid
production associated with n-6 PUFA (see below) (38). Maximal T-cell
activation requires the T-cell receptor (TCR)-CD3 complex plus other
co-stimulatory signals. These co-stimulatory molecules are attached
to the plasma membrane via glycosylphosphatidylinositol anchors
clustered in lipid rafts (35). The Src kinase family of protein
tyrosine kinases plays an important role in T-cell activation. All
Src kinases have myristate (14:0) covalently attached through an
amide linkage at a glycine at position 2 of the protein. Seven of
the Src kinases are also palmitoylated (16:0) at a cysteine
(Myr-Gly-Cys). Myristoylation and palmitoylation are required for
targeting Src kinases to rafts (34, 37).

The two Src family kinases, Lck and Fyn, are concentrated on the
cytoplasmic side of lipid rafts. Stimulation of the TCR triggers Lck
and Fyn activation that leads to increased Ca2+ signaling, ERK
activation, and other downstream signaling events (34-37). Lipid
rafts from Jurkat T-cells treated with 20:4n-6 or 20:5n-3 display
reduced Lck and Fyn content and a decline in both calcium signaling
and ERK activation. In contrast, the
glycosylphosphatidylinositol-anchored proteins, (CD59 and CD48), the
ganglioside GM1, and caveolin remain in rafts after PUFA enrichment.
Treatment of Jurkat T-cells with 20-carbon PUFA enriches both outer
(sphingomyelin, PC) and inner leaflet (PE) phospholipids with 20-
and 22-carbon PUFA. The more unsaturated lipid environment of rafts
and, in particular, the cytoplasmic leaflet may account for the
displacement of acylated proteins from rafts in PUFA-treated T-cells
(36).

PUFA treatment of T-cells also results in PUFA acylation of Fyn (34,
37). In fact, the profile of Fyn acylation parallels the fatty acid
composition supplied to the cells. Saturated (14:0, 16:0, 18:0),
monounsaturated (18:1n9) and polyunsaturated (20:4n-6, or 20:5n-3)
fatty acids can be covalently bound to Fyn. Apparently, palmitoyl
acyltransferase is a promiscuous enzyme that will covalently attach
a broad spectrum of fatty acids to proteins. Replacement of 16:0
with any PUFA results in the failure of Fyn to specifically interact
with rafts. Failure of Fyn or other Src kinases to interact with
rafts uncouples the TCR from activating downstream signaling
pathways, e.g. calcium signaling or ERK activation.

These studies illustrate how changes in membrane phospholipid
composition as well as protein acylation affect signaling from the
plasma membrane. In particular, the effect of PUFA on membrane
protein acylation and protein targeting to the membrane and
microdomains, like rafts, is likely to have a broad impact on
signaling pathways. It will be important to determine whether n-3
PUFA acylation of membrane proteins extends beyond Src kinases to
include effects on G-protein receptors, membrane channels, or
sphingolipid signaling.

         n-3 PUFA Effects on Eicosanoid Signaling
     TOP
     INTRODUCTION
     Synthesis and Metabolism of...
     n-3 PUFA Effects on...
     n-3 PUFA Effects on...
     n-3 PUFA Effects on...
     Concluding Remarks
     REFERENCES

Non-esterified PUFAs released from the sn-2 position of the membrane
phospholipids by the action of specific phospholipases
(phospholipase A2) are substrates for cyclooxygenases (COX-1, a
constitutive enzyme or COX-2, an inducible enzyme), lipoxygenases
(5-, 12-, or 15-LOX), or cytochrome P450 monooxygenases (CYP) (6, 8,
14, 39-41). Cyclooxygenase products of 20:4n-6 give rise to
prostanoids and thromboxanes (14, 41). The LOX pathway catalyzes the
insertion of molecular oxygen into arachidonic acid as the first
step in the formation of leukotrienes and hydroxyeicosatetraenoic
acids. COX and LOX products exit cells and act locally at nanomolar
levels through autocrine or paracrine processes on cell surface
receptors linked to G-proteins. Activation of G-protein-associated
receptor leads to changes in intracellular cAMP or calcium, which
serve as second messengers that activate signaling mechanisms that
have pronounced effects on various cellular functions. The COX
products are modulators of thromboregulatory, inflammatory, and
chemotaxic responses, whereas the LOX products are involved in
vascular permeability, vasoconstriction, and bronchoconstriction (6,
8, 14).

When compared with 20:4n-6, 20:5n-3 and 22:6n-3 are poor substrates
for the COX and LOX reactions (6, 8, 42, 43). Structural analysis of
COX-1 reveals a strained configuration when 20:5n-3 binds (44). This
configuration misaligns carbon 13 with respect to Tyr-385, the
residue that abstracts hydrogen from substrate fatty acids and leads
to a 7-fold reduction in oxygenation efficiency relative to 20:4n-6.
Moreover, most eicosanoids resulting from COX and LOX action on
20:5n-3 have a bioactivity weaker than the 20:4n-6 product. n-3
PUFAs also enhance eicosanoid catabolism by increasing its
peroxisomal degradation (43). Clearly n-3 PUFAs are important
modulators of eicosanoid signaling through several mechanisms. The
effects of n-3 PUFA on the synthesis, bioactivity, and metabolic
clearance of eicosanoid (COX and LOX) products accounts, at least in
part, for the anti-inflammatory properties of n-3 PUFA.

A third route for eicosanoid production involves microsomal
cytochrome P450-linked monooxygenases. These enzymes are members of
a large superfamily of enzymes that catalyzes the NADPH-dependent
oxidation of a diverse array of lipophilic compounds including fatty
acids, hormones, drugs, and xenobiotics (39, 40). CYP-mediated
oxidation of 20:4n-6 yields a variety of eicosanoids, including
epoxides, midchain hydroxy fatty acids, -hydroxy fatty acids, and
dihydroxy fatty acids. Of these oxidized lipids, the epoxy
derivatives of 20:4n-6 are reported to modulate calcium signaling,
channel activity, transporter function, mitosis, and impact
hypertension. n-3 PUFAs are converted to both epoxy and hydroxy
fatty acids by cytochrome P450-linked monooxygenases (45). This
mechanism may be important in cells where there is little COX or LOX
activity, e.g. hepatic parenchymal cells (46). Unfortunately, little
information is available on the bioactivity of 20:5n-3-derived
monooxygenase products on biological systems.

         n-3 PUFA Effects on Gene Expression
     TOP
     INTRODUCTION
     Synthesis and Metabolism of...
     n-3 PUFA Effects on...
     n-3 PUFA Effects on...
     n-3 PUFA Effects on...
     Concluding Remarks
     REFERENCES

The effects of fatty acids on gene expression have received
considerable attention because this represents a direct route for
fatty acids to regulate gene function (12, 13). n-3 PUFAs have rapid
effects on gene expression; changes in mRNAs encoding several
lipogenic enzymes can be detected within hours of feeding animals
diets enriched in n-3 PUFA (47, 48). Moreover, these effects are
sustained so long as the n-3 PUFAs remain in the diet. In these
cases, the fatty acid acts like a hormone to control the activity or
abundance of key transcription factors.

PPAR was the first transcription factor identified as a prospective
fatty acid receptor (49, 50). Studies with PPAR knockout mice have
shown that PPAR is required for many of the effects of fatty acids
on gene expression (51-53). PPAR plays a role in the regulation of
an extensive network of genes involved in glucose and lipid
metabolism including fatty acid transport, fatty acid-binding
proteins, fatty acyl-CoA synthesis, microsomal, peroxisomal, and
mitochondrial oxidation, ketogenesis, and 5, 6, and 9 desaturation
(50); the expression of genes for at least one glycolytic enzyme,
i.e. L-pyruvate kinase (52) and several apolipoproteins, e.g. apoCII
and -CIII, are influenced by PPAR (53). However, studies with the
PPAR null mouse have shown that PPAR is not the sole transcription
factor involved in mediating fatty acid effects on gene
transcription. In addition to the PPAR family (PPAR, -, -1, and -2)
several other transcription factors have been identified as targets
for fatty acid regulation, including hepatic nuclear factor-4,
SREBP-1c, LXR and -, RXR, and NFB (6, 12, 13, 29, 54-60).

At present, ligand binding and structural properties are best
defined for the PPARs. All PPAR subtypes bind 20:5n-3 (IC50 (or Kd)
of ~1-4 ?M). Structural analysis of PPAR shows that 20:5n-3 occupies
300 ?3 of the 1300 ?3 hydrophobic binding pocket (61). The acid
group and the first 8 carbons are buried in the binding pocket. The
hydrophobic -end is bent into the upper arm of the Y-shaped pocket,
and 20:5n-3 is not exposed to solvent. Binding of 20:5n-3 or other
natural ligands shifts the equilibrium to the active configuration,
one that stabilizes the AF-2 helix (helix 12) through H-bonding and
hydrophobic interactions with the ligand. This stable conformation
of AF-2 permits co-activator recruitment to the receptor, a
requisite event in ligand-mediated effects on gene activation. Fatty
acids 14 carbons or longer than 20 carbons do not fit in this
docking mode and would be exposed to solvent. These configurations
do not stabilize the AF-2 helix, lessening the likelihood for
co-activator recruitment. Based on this structural analysis, 20:5n-3
is an endogenous ligand for PPARs. Activation of PPARs by 22:6n-3
will likely require prior retro conversion to 20:5n-3, a process
that requires peroxisomal -oxidation (21).

Although PPARs can bind many fatty acids in vitro, differential
lipid metabolism imposes physiological discrimination at the
cellular level. PPAR is the predominant PPAR subtype in rat hepatic
parenchymal cells. PPAR binds 18:1n9 and 20:5n-3 with nearly equal
affinity, i.e. 0.6 versus 1.1 ?M (61). Yet, 20:5n-3, but not 18:1n9,
activates PPAR in rat primary rat hepatocytes (51). The simplest
explanation for this difference is that the intracellular NEFA pool
available to activated PPAR is subject to metabolic regulation. When
compared with 18:1n9-CoA, 20:5n-3-CoA is a poor substrate for
diacylglycerol acyltransferase (24). A decrease in the rate of
20:5n-3 assimilation into neutral lipids might lead to an elevation
in intracellular 20:5n-3 sufficient to activate PPAR. The notion
that slowly or poorly metabolized fatty acids activate PPAR is
supported by studies with modified fatty acids, e.g. bromo- and
sulfur-substituted fatty acids (50). The fibrate class of
lipid-lowering drugs (Lopid?, Pfizer; Tricor?, Abbott) was
synthesized originally as metabolically stable analogs of branched
chain fatty acids (62). When compared with 20:5n-3, fibrates are
strong activators of PPAR (12, 50-52).

PPAR also binds 20:5n-3 (Kd ~4 ?M) and is expressed in many tissues
including adipose, muscle, and vascular cells. Activated PPAR
induces lipoprotein lipase and fatty acid transporters (CD36) and
enhances adipocyte differentiation as well as inhibiting NFB
function and cytokine and COX-2 expression (8, 50). The glitazones,
e.g. troglitazone, pioglitazone, and rosiglitazone, are
pharmacological PPAR agonists and are used in the treatment of
insulin resistance. Pharmacological activation of PPAR and PPAR
reduces lipid levels in muscle and adipose tissue and improves
insulin sensitivity in these tissues (63, 64). Although n-3 PUFAs
are weak agonists of PPARs, when compared with pharmacological
agonists n-3 PUFAs have significant effects on insulin sensitivity
in various tissues, particularly skeletal muscle (9, 10). Thus n-3
PUFA action on insulin responsiveness in these tissues may extend
beyond its regulation of PPAR activity.

Recently, the liver X receptors (LXR and LXR) were identified as
targets for fatty acid regulation (60, 64). LXRs bind oxysterols and
regulate the expression of genes involved in hepatic bile acid
synthesis (65). Unsaturated fatty acids antagonize oxysterol
activation by LXR in Hek 293 and hepatoma cell lines by interfering
with oxysterol binding. Although such studies suggest that changes
in hepatic PUFA levels might affect bile acid synthesis in vivo,
feeding studies with mice have yet to support this view. Acting
through LXR, oxysterols induce CYP7A1, the rate-limiting enzyme for
bile acid synthesis (66). Interestingly, hepatic 7-hydroxylase
(CYP7A) activity or mRNACYP7A levels are not suppressed in mice fed
diets supplemented with unsaturated fatty acids (67, 68).

LXRs also play a major role in lipogenesis through the regulation of
transcription of the gene encoding sterol regulatory element-binding
protein-1c (SREBP-1c) (69-71). SREBP-1c is a helix-loop-helix
transcription factor required for the insulin-mediated induction of
hepatic fatty acid and triglyceride synthesis (71-75). SREBP-1c
binds sterol regulatory elements in promoters of many genes involved
in fatty acid and triglyceride synthesis, including citrate lyase,
acetyl-CoA carboxylase, fatty acid synthase, stearoyl-CoA
desaturase-1, S14 protein, and glycerophosphate acyltransferase but
not L-pyruvate kinase (58, 75). In contrast to bile acid synthesis,
there is ample evidence for PUFA suppression of hepatic lipogenic
gene expression (12). Feeding animals diets supplemented with corn
oil, walnut oil, or fish oils suppresses the transcription of many
genes involved in de novo lipogenesis including fatty acid synthase,
stearoyl-CoA desaturase-1, L-pyruvate kinase, and S14 protein (48,
56-59). PUFA suppresses the nuclear content of SREBP-1c (56, 58).
Overexpression of the nuclear form of SREBP-1c overrides the
suppressive effect of PUFA on lipogenic gene expression (58, 59).
Thus, PUFA regulation of nuclear form SREBP-1c levels may account
for many of the suppressive effects of PUFA on hepatic lipogenesis.
PPAR is not required for PUFA suppression of SREBP-1c or the mRNAs
encoding the lipogenic genes (58).

In contrast to PPAR and LXR, fatty acid regulation of SREBP-1c may
not involve direct fatty acid binding but rather control the nuclear
abundance of SREBP-1c. Like other members of the SREBP family,
SREBP-1c is translated as a large precursor protein (~125 kDa)
tethered to the endoplasmic reticulum and Golgi complex (75, 76).
The precursor is proteolytically processed to a mature form that
moves to the nucleus where it binds as a dimer to sterol regulatory
elements in promoters of responsive genes. In rat liver, both n-3
and n-6 PUFAs suppress the cellular level of mRNASREBP-1c as well as
the precursor and nuclear forms of SREBP-1c (56-59). The hierarchy
for fatty acid regulation of mRNASREBP-1c is 20:5n-3 = 20:4n-6 >
18:2n-6 > 18:1n-9. The mechanism for PUFA regulation of hepatic
mRNASREBP-1c involves an enhanced rate of mRNASREBP-1c turnover
rather than inhibition of gene transcription (77). Specific
cis-regulatory elements within the transcript that are targeted by
unsaturated fatty acids have eluded identification.2 In contrast to
liver and primary hepatocytes, established cell lines display a more
complicated response to PUFA that involves effects at the
transcriptional level, mRNASREBP-1c turnover, and conversion of
precursor SREBP-1c to the nuclear form (60, 65). PUFA functions as a
feedback regulator of fatty acid synthesis. Coupling this action
with the PUFA-mediated induction of PPAR-regulated genes shifts
hepatic metabolism away from lipid synthesis and storage to lipid
oxidation (51, 58). This mechanism prevents lipotoxicity associated
with lipid overload.

         Concluding Remarks
     TOP
     INTRODUCTION
     Synthesis and Metabolism of...
     n-3 PUFA Effects on...
     n-3 PUFA Effects on...
     n-3 PUFA Effects on...
     Concluding Remarks
     REFERENCES

The 20- and 22-carbon n-3 PUFAs are unique lipids that when added to
the diet or to cells can alter membrane phospholipid composition,
impact eicosanoid synthesis and action, and regulate transcription
factor activity and abundance. n-3 PUFAs affect diverse
physiological processes including cognitive functions and visual
acuity, immunosuppressive and anti-inflammatory actions, and
anti-thrombotic and anti-arrhythmia activities along with having
major effects on whole body glucose and lipid metabolism (see Table
I, supplemental material). It is important to distinguish those
effects that are specific for n-3 PUFA from those effects that are
seen with unsaturated fatty acids in general. Whereas n-6 PUFAs
stimulate, n-3 PUFAs inhibit eicosanoid synthesis and signaling and
NFB activation. This feature accounts for the anti-inflammatory and
anti-thrombotic action of n-3 PUFA. There is also a strict
requirement for 22:6n-3 over 22:5n-6 for normal CNS development and
function. In contrast, PUFA effects on membrane raft composition as
well as the regulation of transcription factors like PPARs, LXRs, or
SREBP-1c are determined more by changes in cellular levels of
unsaturated fatty acids rather than specific effects of n-3 PUFA.
The modest resistance of n-3 PUFA to -oxidation or assimilation into
neutral lipids might be sufficient to elevate intracellular n-3 NEFA
or PUFA-CoA levels allowing these factors to serve as regulatory
ligands for transcription factors or substrates for protein
acylation. Unfortunately no direct evidence has been reported to
support this concept.

Some effects of n-3 PUFA on physiological processes remain poorly
defined. The rapid attenuation of arrhythmias in cardiomyocytes
treated with n-3 PUFA involves changes in the activity of several
membrane channels (7, 78). Whether this effect involves changes in
membrane phospholipid composition or targeting of membrane channel
proteins to specific microdomains is unknown. Nevertheless,
considerable progress has been made in understanding how n-3 PUFAs
affect cell function. Many mechanisms have been described, and new
mechanisms are likely to be discovered that will better define how
these unique lipids impact human health and disease.

         ACKNOWLEDGEMENTS

I am grateful for many members of the laboratory who have
participated in this research. I thank Norman Salem Jr. at the
National Institutes of Health and William Smith at Michigan State
University for many helpful discussions and for a critical review of
the manuscript.