Search This Blog

Tuesday, 31 July 2012

The Metabolic Push-Me-Pull-You: HCV core protein, and the even-handed generosity of DM2: Gluconeogenesis and De Novo Lipogenesis, Oh My!

One of the features of HCV which I will probably return to again and again in this work-in-progress is that it represents a useful model of metabolic syndrome and DM2. HCV-infected hepatocytes run simultaneous gluconeogenesis and de novo lipogenesis, showing disordered insulin signalling (not just insulin resistance, which would decrease lipogenesis). The mechanisms involved (metabolic and immunologic) are being studied in detail and seem to have a lot to tell us about non-viral DM2. I would suggest that they are more relevant than drug-damage models of insulin deficiency or resistance; the interference is more subtle, the adjustments are clearly the results of adaptive processes, are highly effective, and are only mildly cytotoxic.

At the very least, HCV presents another angle from which to look at these problems.

The hepatitis C virus (HCV) induces lipid accumulation in vitro and in vivo. The pathogenesis of steatosis is due to both viral and host factors. Viral steatosis is mostly reported in patients with genotype 3a, whereas metabolic steatosis is often associated with genotype 1 and metabolic syndrome. Several molecular mechanisms responsible for steatosis have been associated with the HCV core protein, which is able to induce gene expression and activity of sterol regulatory element binding protein 1 (SREBP1) and peroxisome proliferator-activated receptor γ (PPARγ), increasing the transcription of genes involved in hepatic fatty acid synthesis. Steatosis has been also implicated in viral replication. In infected cells, HCV core protein is targeted to lipid droplets which serve as intracellular storage organelles. These studies have shown that lipid droplets are essential for virus assembly. Thus, HCV promotes steatosis as an efficient mechanism for stable viral replication. Chronic HCV infection can also induce insulin resistance.

(The hepatitis C virus has evolved to be transmitted from infected cells on the lipid transport system; therefore it “wants” infected cells to maximise triglycerides and release of (HCV-carrying) VLDL; the higher the TAG, the more HCV virions in serum and the greater the chance of infection from blood-to-blood  encounters with the host; also, the virus can ensure that it stays a step ahead of the host’s immune defences by regularly infecting naïve cells.)

Del Campo and Romero Gomez, the authors of this paper, are experimenting with the statins (mainly fluvastatin) as anti-HCV agents. A sensible deduction from the evidence, but quite possibly not as sensible as carbohydrate restriction and intermittent fasting. However I don’t doubt that it is easier to get funding for a drug trial than for a trial of an antiviral Atkins-type diet. Yup, I can see how that suggestion might go down at the funding board.
A suggestion might be that the virus is inducing IR at the gluconeogenesis end, while promoting those genes that normally respond to insulin at the lipogenesis end of hepatocyte metabolism.
But what is the advantage of promoting gluconeogenesis and elevated blood glucose?

HCV can perhaps replicate more effectively if TAGs are not being used as a fuel, i.e., if carbohydrate is the cell’s main energy substrate. If viral manipulation of cell processes to promote gluconeogenesis results in elevated blood glucose, this will tend to prevent lipids “sponsored” by the virus being oxidised to fuel cell processes.  Gluconeogenesis is underwriting lipogenesis.
(correction: officially, at least, hepatocytes only run on ketoacids (pyruvate and oxaloacetate) from amino acid catabolism. However, another reference (Best and Taylor) implies that these are fasting-state gluconeogenesis substrates, and states that the newborn liver is wholly dependent on sugars and lipids. It seems more likely that any cell uses a mix of energy substrates and that in the case of hepatocytes the preferential usage is ketoacids, if only because other cells cannot metabolize gluconeogenic amino acids. Anyhow, this suggestion can be left in the air for now.)

How do infected hepatocytes (about 2-25% of the total, greater in non-responders, in one study ) manage to produce this two-way excess? We are used to cells that convert glucose (or fructose) or lipids to triglycerides, or glycogen and amino acids (or fructose) to glucose, but how does a cell do both at once? Does the cell take in more substrate than it normally would – a pate de foie gras forced feeding under viral prompting – or does it neglect the many other duties of a hepatocyte and squander the ATP produced through its mitochondrial density on its guest, or both?

[update 10-12-2012: HCV infection can associated with a DROP in tryglycerides; see this recent post. The virus is monopolizing VLDL in the same way it monopolizes cholesterol.]

This even-handed generosity is explained in diabetic research by concurrent insulin sensitivity and insulin resistance.

“[Although] an impairment of insulin receptor signaling to Foxo1 can explain insulin's inability to restrain HGP, one would predict that, if the liver were wholly insulin resistant, triglyceride (TG) synthesis and assembly into ApoB-containing lipoproteins would also be impaired. But the opposite is true in the diabetic liver.
In recent years, the idea that the diabetic liver may harbor a noxious brew of insulin resistance and excessive insulin sensitivity has gained a second wind.”

The HCV toxin that manipulates cell processes is called core protein and yields a number of fractions.

Confusingly, HCV core protein is also an integral part of the viral coat or capsid. Like Batman’s utility belt, it performs an amazing array of functions in its interactions with lipoproteins, mitochondria, immune system pathways, cell surface receptors, and RNA copying mechanisms. Like a pushy talent agent it ruthlessly promotes the interests of its RNA wherever it goes.

HCV core protein (like that of the more benign Hepatitis G virus, GVB-V) shares genomic features with plant oleosins. These proteins are found associated with fatty droplets in grains and seeds, and sesame oleosin is a type1 (IgE) allergen. Presumably oleosin-like properties (unique to these two related viruses) allow the close association with VLDL-LDL that is characteristic of the HCV lifestyle.
Some of the transcription pathways involved are (as we might expect, and to return to our muttons) very close to those implicated in non-viral DM2. HCV core protein promotes gluconeogenesis by increasing activity of nuclear Fox01 transcription factor (Fox01 is to glucose what NF-KappaB is to cytokines) through inhibition of phosphorylation by mitochondrial ROS (inhibition of Mito Complex 1 by HCV core protein).

In Robert Lustig’s DM2 hyperglycaemia scenario, fructose plays the same role (perhaps decreased phosphorylation is the result of fructose depleting [P1] as in the text book extract I will end with).

 “Hepatic insulin resistance, made worse by elevated fructose concentrations, prevents the phosphorylation of FoXo1, which allows this protein to enter the nucleus and induce the transcription of enzymes that promote gluconeogenesis. “

This is the one paper than anyone curious about Robert Lustig’s ideas should read, especially the sections on dietary fat vs dietary carbohydrate as factors in DM2.


in 2007 my HCV viral load was 400,000

Earlier this year, after a few months of more-or-less ketogenic diet (25-75g carbs), VL was 26,000

After a month or so of higher carbs (but still low-carb – 50-150g) latest VL was 60,000


Trying to get these concepts and references into one smooth flow has been like herding cats, so I will summarize in plain language:

HCV (through its core protein) can promote both insulin resistance (elevating blood glucose) and/or insulin sensitivity (elevating fasting triglycerides). This both mimics and adds to dietary metabolic syndrome, and increases the risk of Type 2 Diabetes. And a diet and lifestyle that encourages DM2 will promote increases in viral load, disease pathology, and resistance to treatment.

Because HCV down-regulates GLUT2 to produce the gluconeogenic effect, reducing glucose uptake of infected hepatocytes, fructose becomes an ideal substrate for both gluconeogenesis and lipogenesis. Fructose consumption is predicted to optimize viral replication.

A carbohydrate-restricted version of the Paleo diet – Paleo-Atkins is a convenient shorthand for this – preferably with time-restricted feeding (16 hour daily fasts and 8 hour feeding windows, or at least no carbohydrate or protein outside of regular mealtimes),  is – or ought to be – the default diet for DM2 and fatty liver.

This diet ought to reduce the HCV viral load (as in my case), improve Hep C pathology (ditto), and improve the response to treatment (we'll see one day, maybe).

Research into HCV core protein effects on gluconeogenic and lipogenic regulation can provide insights into the mysterious aetiology of diabesity. For example,  tending to corroborate the theory that high-fructose diets play an important causative role in metabolic disease.

More on Fructose:

Not everyone metabolizes fructose the usual way, and not everyone who does clears it easily.
Everyone needs to metabolise glucose, and the genes for this have been conserved, but our ancestors sometimes survived for long periods without much fructose exposure, and these genes show more variety.
For example, the higher rate of gout in some South Pacific populations may be the result of adaptation to an ancestral diet low in fructose.
This classic account is from White, Handler and Smith's Principles of Biochemistry (c) 1954, the 1973 edition

Metabolism of Fructose

Although fructose can be phosphorylated in the 6 position at a slow rate by non-specific kinases, most of ingested fructose is phosphorylated in the liver by a fructokinase that specifically directs phosphorylation at the C-1 position of this ketose. No mutase is known that can catalyse conversion of fructose 1-phosphate to fructose 6-phosphate, nor can phosphofructokinase effect synthesis of fructose diphosphate from fructose 1-phosphate. The only pathway available to the latter is made possible by a specific aldose that catalyses the following reaction:

Fructose 1-phosphate ó dihydroxyacetone phosphate + glyceraldehyde

The further metabolism of glyceraldehyde requires reduction by NADH to glycerol, which is then phosphorylated by glycerol kinase, using ATP, and reoxidized by NAD+ to dihydroxyacetone phosphate. The latter then enters the usual glycolytic sequence.

Individuals who lack fructokinase excrete the major portion of ingested fructose in the urine. “Fructose intolerance” is a more serious illness, characterized by genetic lack of the aldose specific for fructose 1-phosphate, which accumulates after fructose ingestion and inhibits diverse enzyme systems. Even normal individuals may experience difficulty with large fructose intake. Although both the kinase and the special aldolase are present in large and equivalent activities, fructose 1-phosphate may accumulate in the liver for some time after a large fructose flux, e.g., after ingestion of a large quantity of sucrose.   

The explanation offered is as follows: Because of the effectiveness of the kinase, both [ATP] and hence [P1] are lowered in liver cells. The P1 inhibition of adenylate deaminase is thus released and inosinic acid [IMP] accumulates; in intact, perfused liver [IMP] increases seven-fold in ten minutes under such circumstances. However, IMP is a powerful inhibitor of the fructose 1-phosphate aldolase (K1 = O.1 mM and Km F-1-P = 0.18 mM), thus delaying further metabolism of this compound.
The presence of fructose and galactose in the intestine, with the amino acids, inhibits intestinal absorption of the latter.
Metabolic effects of androgenic compounds on the sex organs and tissues...include increased fructose production by seminal vesicles and utilization of this sugar by seminal plasma, with concomitant enhancement of the activity of both aldose reductase and ketose reductase.

Thursday, 26 July 2012

Fructose, Calories, Carbohydrate and de novo lipogenesis: What does "Hypercaloric" mean?

What does “hypercaloric” actually mean?

One possible objection to the prediction that fructose (and glucose from higher carbohydrate intakes) will enhance HCV replication by inducing DAGT1 and VLDL expression is the claim that is sometimes made that fructose only exerts strong effects on de novo lipogenesis in hypercaloric states.
Otherwise, it will be used to replenish glycogen stores or converted to energy.
The chemistry textbooks put this another way; glucose is converted to fat when it cannot be used to replenish glycogen or generate ATP. No mention of calories. The possibility is left open that other factors - micronutrient deficiencies, toxins, mitochondria defects, hormonal sensitivities - could influence the process.
(Note: I will skip between glucose and fructose in this essay; I think that anyone familiar with this topic will recognize that there is no sleight of hand going on. Fructose is a subset of carbohydrate, of which glucose is the main representative.)
The whole calories in, calories out controversy would take too long to review here. Suffice it to say that, in my opinion, counting calories in provides a both over-complicated and over-simplified way to do something that is more quickly and usefully done by counting (or estimating, rather) grams of fat, carbohydrate, and protein in a meal.
And counting calories out with any hope of accuracy is pretty much impossible unless you spend time in a laboratory or a high-tech gym, and the results there won’t tell you much about calorie expenditure in other conditions. That’s what your appetite ought to be doing…
The most sensible definition of a hypercaloric state is provided by Chris Kresser at Healthy Skeptic; basically, if you are gaining weight, your diet is hypercaloric. (Dr Kresser summarizes the case against fructose as a driver of DNL )

Unfortunately the loss and gain of body weight is a very protracted process and the weight of most people fluctuates from hour to hour and day to day.
Is there such a thing as a hypercaloric meal?
If I were to graph calories out in the course of a day, the baseline of expenditure (basal metabolic rate) would be quite high, with many peaks above it where I exerted my body or my mind; any troughs where the BMR dropped would be shallow.
If I were to graph calories in, you would see two huge curves around my two meals; the graph would be at zero at least half the time. There would only be four “isocaloric” points on the graph, two hypercaloric peaks (of perhaps 4 hours duration each) and the rest of the time, calories in would be less than calories out.
To confuse things even more, if you fast for a week no single day's eating can restore you to a "hypercaloric" state, but carbohydrate will be converted to fat as soon as glycogen stores are full...
Simply because that is how your body stores most of the energy it gets from carbohydrate. Glycogen stores cannot be expanded beyond about 150% of normal (and even that takes some doing - see the next link), whereas fat stores, as everyone knows by now, are pretty much capable of expanding indefinitely.

This makes it very hard to rely on the result of any particular piece of research unless we know a great deal about the surrounding conditions.
Lucas Tafur has done a worthwhile analysis of one DNL study.
Showing how things may not always be as they appear.

A further point is that fructose studies often use pure fructose, when what we should be concerned about is a) the combined effect of fructose and glucose, at the ratios similar to those found in sugar, HFCS, and fruit juice; and b) the combined effects of sugar and dietary carbohydrate from starch.
Recently Dr Peter Attia has published
some tables showing the difference in triglycerides after feeding of glucose, fructose, and HFCS.

A further complication is, that in the case of type 2 diabetes (a common complication of Hepatitis C), there is already a high blood glucose level due to increased hepatic glucose production.
Does this elevated blood sugar then mimic a “hypercaloric” state whenever additional sugars are fed? Any fed sugars have to compete with blood glucose for conversion to glycogen or ATP.
And DNL, according to the textbooks, occurs when sugars are surplus to amounts that can be converted to glycogen or ATP (and if DNL cannot take care of the sugars, they will be dumped through the kidneys, a convenience which is not conducive to good kidney function in the long haul, hence the connection between diabetes and kidney disease).
Calories, of course, can also come from fat or protein. Is a meal of fat likely to prevent replenishment of glycogen?
R. D. Feinman points out that fructose is easily converted to glycogen when total dietary carbohydrate is restricted.

That’s where we want to be. That’s the sweet spot, if you’ll pardon the expression. Restriction of carbohydrate means we don’t need to worry about the occasional fruit we eat, or the sugar in the pickle we put on our bacon and eggs, or the sugar in our dark chocolate, within reason. There’s fructose in beetroot, carrot, potato, and onion that no-one ever mentions.
Of course, carbohydrate can, at least in theory, cause fatty liver and high TG without any DNL whatsoever; if the liver is preoccupied with burning carbohydrate, it may be unable to convert all the fat you eat into ATP; in which case the fat will be recycled into triglycerides, to be stored in the liver or released as VLDL (or IDL).
You really do want to be primarily a fat-burner, either way.

The problem with the “hypercaloric” version of fructose-driven DNL is that it promises that “if you don’t eat too much, it doesn’t matter”. But the science doesn’t tell us how much to eat and when to achieve this miracle.
And restricting calories overall tends to increase appetite and mess with our best intentions.

Whereas R. D. Feinman’s prescription “if you don’t eat carbs (much) it doesn’t matter (much)” is more practical.
Or, in Professor Feinman's own words, "as carbohydrate and calories are reduced, any effect of fructose will be minimized. In the extreme, if you are on a very low carbohydrate diet, any fructose you do eat is likely to be turned into glucose".
Glucose, for glycogen storage or conversion to ATP, rather than fat.

A philosophical digression

There is a critical philosophical distinction between the type of notions that calories (from physics) and macronutrients (from chemistry) represent.

A sugar molecule is real, a gram is abstract, but you can look at, handle and taste a gram of sugar.
A calorie is abstract and amorphous (it could be applied to food, petrol, wood, coal, uranium, movement, noise, heat, radiation and so on).
There are glucose receptors, metabolites and so on. There is no receptor or enzyme that deals with calories. Saying "calories (in)" is often just a lazy and potentially misleading way of saying "so many grams (or ounces etc.) of macronutrients in such and such a ratio" (just as carbohydrate is a lazy way of saying "starch and/or sugar" which is itself a lazy way of saying something even more informative).
(If I say "a whale 100metres long" I am giving you extra information about the whale. You are not meant to be focusing on the implication that a metre is 1/100th the length of my whale.)

A calorie does not have a "nature". There is no natural history of the calorie as there is a natural history of every nutrient.
So why bother? Calories (or joules) are the only way to measure calories out for comparison with nutrition taken in. This is useful for calculations of work and food allowance - as in rationing, for example, or for planning calorie restricted or overfeeding diets.
But it does lead to a false assumption - that what is the only useful measurement of metabolic output must therefore be the only measurement of input worth considering.

Wednesday, 18 July 2012

Is a Diet High in Saturated Fat Good for the Liver?

If we follow the advice laid out in a previous post and increase fat while restricting PUFA, will the extra saturated fat be good or bad for our liver, HCV aside?

That depends, I suppose on whether you think alcohol and drugs and NAFLD are good models for virus-related liver damage. At present they are the only models we have. And we know that alcohol, acetaminophen, and fructose don't help people with Hep C. So I think they are very good models; all the mediators of liver damage from those causes are also present in the livers of people with Hep C (high ferritin, stellate cell activation, LPS sensitivity, lipid peroxidation, oxidative stress, and so on).
So personally I don't see it as any kind of leap to accept the relevance of papers like these. Besides, the proof of the suet pudding will appear in the eating thereof.

Hepatic Stellate Cells are the cells that make collagen in fibrosis. They are also called Ito cells, which explains why I missed all these Ito cell studies before (when I was collecting data on factors, including SFAs, that reduced hepatic stellate cell activity).
This is another line of evidence supporting the view that saturated fat in the diet is antifibrotic.
"When tallow was substituted for corn oil the Ito cells were not activated and the liver histology was normal".

Is it the PUFA restriction alone, or the addition of SFA? 
My reading of these papers is that both play a role. Even 5% calories as PUFA causes some fibrosis on a low-fat diet, but none when corn oil is added to beef fat or coconut MCT to give a similar ratio.

These may only be animal tests, but I assure you, if a supplement performed half as well as saturated fat in animals, it would out-sell Silymarin.
Besides, we are talking about something everyone eats already.

We used the intragastric feeding rat model for alcoholic liver disease to investigate the relationship between transforming growth factor (TGF)-beta 1 and inhibition of endothelial cell proliferation. Twelve groups of male Wistar rats (four to five rats per group) were fed ethanol or dextrose with either corn oil or saturated fat for 1-, 2-, and 4-week periods. All control animals were pair fed the same diets as ethanol-fed rats except that ethanol was isocalorically replaced by dextrose. In the ethanol-fed groups, nonparenchymal cells were isolated and TGF-beta 1 was measured in the nonparenchymal cell supernatant. Liver pathology and endothelial cell proliferation with an antibody to proliferating cell nuclear antigen were studied in all groups. Plasma TGF-beta 1 was measured in all rats. Pathological changes (fatty liver, necrosis, and inflammation) were observed only in the corn oil/ethanol-fed rats at 4 weeks. Significantly higher levels of TGF-beta 1 were seen in both plasma and nonparenchymal cell supernatant in rats fed corn oil and ethanol; plasma levels of TGF-beta 1 were not significantly different between the dextrose-fed controls and saturated fat/ethanol-fed rats. A significant inverse correlation (r = -0.89, P < 0.01) was seen between plasma TGF-beta 1 and the number of endothelial cells arrested at G1/S. Immunohistochemistry revealed the presence of TGF-beta 1 staining in interstitial macrophages only in rats fed corn oil and ethanol. The present study provides evidence for a role for TGF-beta 1 in inhibiting endothelial cell proliferation in experimental alcoholic liver disease. Arrest of endothelial cells may lead to their differentiation and/or to produce mediators that could stimulate other cells such as Ito cells. Sustained TGF-beta 1 may also lead to Ito cell production of extracellular matrix.

Alcohol Clin Exp Res. 1991 Dec;15(6):1060-6.
Effect of dietary fat on Ito cell activation by chronic ethanol intake: a long-term serial morphometric study on alcohol-fed and control rats.
Takahashi H, Wong K, Jui L, Nanji AA, Mendenhall CS, French SW.

Department of Internal Medicine, National Kurihama Hospital, National Institute on Alcoholism of Japan, Kanagawa.

We studied the effects of long-term ethanol ingestion and dietary fat on Ito cell activation morphometrically in rats. Sixteen pairs of Wistar male rats were divided into two groups, one fed tallow and the other fed corn oil as the source of dietary fat. Each group of rats were pair-fed a nutritional adequate liquid diet containing either corn oil (CF) or tallow (TF) as fat as well as protein and carbohydrate. Half of each group received ethanol, the rest were pair-fed isocaloric amounts of dextrose via an implanted gastric tube for up to 5 months. Morphometric analysis of the change in fat and rough endoplasmic reticulum (RER) of Ito cells was performed on electron micrographs obtained from monthly biopsies including baseline. Ito cell activation was assessed by a decrease in the ratio of fat/RER in Ito cells. The ratio of fat/RER in Ito cells of alcoholic rats fed the CF diet (CFA) gradually decreased. The ratio was found to be lower than in the pair-fed control rats (CFC) at 5 months of feeding. CFA: 1.74 +/- 0.57, vs. 7.46 +/- 2.05, respectively, p less than 0.05, mean +/- SE). Ito cell fat also significantly decreased at 5 months of feeding (p less than 0.05). The fat/RER ratio in CFA significantly decreased only subsequent to the development of fatty change, necrosis, and inflammation followed by fibrosis of the liver. In contrast, the TFA rats did not show a significant decrease in the fat/RER ratio in the Ito cells throughout the study, while TFC rats showed an increase in the fat/RER ratio.(ABSTRACT TRUNCATED AT 250 WORDS)

Alcohol Alcohol Suppl. 1991;1:357-61.
Ito cell activation induced by chronic ethanol feeding in the presence of different dietary fats.
French SW, Takahashi H, Wong K, Mendenhall CL.

Department of Pathology, Faculty of Medicine, University of Ottawa, Ontario, Canada.

Bronfenmajer et al. (1966) first studied Ito cells in alcoholic hepatitis (AH) by light microscopy (LM). The number of Ito cells and the number of fat droplets were increased. Okanoue et al. (1983) found that Ito cells were reduced by LM but increased by electron microscopy (EM) in scars in AH. Ito cells were activated in scars (increased RER and decreased fat in Ito cells with transition to fibroblasts). Minato et al. (1983) showed that increased RER in Ito cells correlated with increased collagen synthesis of liver biopsies in vitro. Mak et al. showed increased RER correlated with the degree of fibrosis in alcoholic baboons (1984) and alcoholic cirrhosis in man (1988). French et al. (1988b) showed morphometrically that Ito cell fat was decreased and RER was increased only in scars but not in normal sinusoids so that Ito cell activation was restricted to the scars. There was no correlation of sinusoidally located Ito cell fat or RER with the amount of perisinusoidal collagen. In rats fed ethanol and a nutritionally adequate diet including corn oil (25% of calories) by intragastric cannula for five months the fatty liver progressed to focal central fibrosis, and Ito cell activation (fat/RER) was increased. When tallow was substituted for corn oil the Ito cells were not activated and the liver histology was normal. Thus, the type of dietary fat and the local environment (scars) are important factors in the activation of Ito cells by alcohol in vivo.

This paper suggests that flaxseed oil may not be the best omega-3 supplement for alcoholics: linolenic acid is the vegetable form of omega-3.
Besides, the benefits from PUFAs in Hep C (antiviral and metabolic) are only seen with DHA, arachadonic acid, and EPA (in descending order of potency), and these are only found in animal fats; oily fish, fatty red meat, organ meat, dairy fats, and egg yolks.
Life Sci. 2003 Jul 18;73(9):1083-96.
The ethanol metabolite, linolenic acid ethyl ester, stimulates mitogen-activated protein kinase and cyclin signaling in hepatic stellate cells.
Li J, Hu W, Baldassare JJ, Bora PS, Chen S, Poulos JE, O'Neill R, Britton RS, Bacon BR.

Department of Internal Medicine, Division of Gastroenterology and Hepatology, The Brody School of Medicine, East Carolina University, 600 Moye Boulevard, Greenville, NC 27858-4354, USA.

Chronic ethanol consumption can result in hepatic fibrosis and cirrhosis. In addition to oxidative metabolism, ethanol can be metabolized by esterification with fatty acids to form fatty acid ethyl esters (FAEE) such as linolenic acid ethyl ester (LAEE). We have previously demonstrated that LAEE has promitogeinc and activating effects on hepatic stellate cells (HSC), but the mechanisms of these actions are not known. Intracellular signaling through MAP kinase pathways, including extracellular signal-regulated kinase (ERK) and c-Jun N-terminal kinase (JNK) can influence the activity of the transcription factor AP-1, while cell-cycle regulatory proteins such as cyclin E and cyclin-dependent kinase (CDK), play an important role in cell proliferation. In this study, we demonstrate that treatment of HSC with LAEE increases cyclin E expression and cyclin E/CDK2 activity, which may underlie the promitogenic effects of this compound. In addition, LAEE increases ERK and JNK activity, and these pathways play an important role in the activation of AP-1-dependent gene expression by LAEE. The stimulation of intracellular signaling pathways in HSC by this well-characterized ethanol metabolite may contribute to ethanol-induced hepatic fibrogenesis.

Dietary Saturated Fat Reduces Alcoholic Hepatotoxicity in Rats by Altering Fatty Acid Metabolism and Membrane Composition

Rats fed a saturated fat diet are protected from experimentally induced alcoholic liver disease, but the molecular mechanisms underlying this phenomenon remain in dispute. We fed male Sprague-Dawley rats intragastrically by total enteral nutrition using diets with or without ethanol. In 1 control and 1 ethanol group, the dietary fat was corn oil at a level of 45% of total energy. In other groups, saturated fat [18:82 ratio of beef tallow:medium-chain triglyceride (MCT) oil] was substituted for corn oil at levels of 10, 20, and 30% of total energy, while keeping the total energy from fat at 45%. After 70 d, liver pathology, serum alanine aminotransferase (ALT), biochemical markers of oxidative stress, liver fatty acid composition, cytochrome P450 2E1 (CYP2E1) expression and activity and cytochrome P450 4A (CYP4A) expression were assessed. In rats fed the corn oil plus ethanol diet, hepatotoxicity was accompanied by oxidative stress. As dietary saturated fat content increased, all measures of hepatic pathology and oxidative stress were progressively reduced, including steatosis (P < 0.05). Thus, saturated fat protected rats from alcoholic liver disease in a dose-responsive fashion. Changes in dietary fat composition did not alter ethanol metabolism or CYP2E1 induction, but hepatic CYP4A levels increased markedly in rats fed the saturated fat diet. Dietary saturated fat also decreased liver triglyceride, PUFA, and total FFA concentrations (P < 0.05). Increases in dietary saturated fat increased liver membrane resistance to oxidative stress. In addition, reduced alcoholic steatosis was associated with reduced fatty acid synthesis in combination with increased CYP4A-catalyzed fatty acid oxidation and effects on lipid export. These findings may be important in the nutritional management and treatment of alcoholic liver disease. 

LOOK AT THIS TABLE: ... nsion.html

In fact, this paper should be read in full and all the figures and tables studied, because there are 4 groups of control rats fed various fats without alcohol, and what happens to them is as interesting as the effects with alcohol.

The control rats fed most saturated fats gained the least weight. Corn oil (high PUFA) was significantly more fattening than a mixture of beef fat and coconut MCTs, in a breed of rat designed to get fat on a "high fat" diet. 

And here's one for the vegetarians: Olive oil is quite good at suppressing fibrosis, though perhaps not as brilliant as beef fat. (note that mutton, goat, venison, cocoa, and dairy fats ought to be similar to beef and coconut).

When I say "saturated fat" I really include monounsaturated fat, as it behaves chemically in pretty much the same way; it takes more than one unsaturated bond in close proximity (there are two such bonds near to one another in omega-6 lineolic acid) to promote lipid peroxidation.

J Gastroenterol. 2009;44(9):983-90. Epub 2009 Jun 9.
Dietary olive oil prevents carbon tetrachloride-induced hepatic fibrosis in mice.
Tanaka N, Kono H, Ishii K, Hosomura N, Fujii H.

First Department of Surgery, Faculty of Medicine, University of Yamanashi, 1110 Shimokato, Chuo, Yamanashi, 409-3898, Japan.

The specific purpose of this study was to investigate the effects of dietary olive oil on hepatic fibrosis induced by chronic administration of carbon tetrachloride (CCl(4)) in the mouse. In addition, the effects of oleic acid, a major component of olive oil, on activation of hepatic stellate cells (HSCs) were investigated in vitro.

Mice were fed liquid diets containing either corn oil (control, AIN-93) or olive oil (6.25 g/L) throughout experiments. Animals were treated with CCl(4) for 4 weeks intraperitoneally. The mRNA expression of TGF-beta1 and collagen 1alpha2 (col1alpha2) in the liver was assessed by reverse transcriptase-polymerase chain reaction (RT-PCR). The HSCs were isolated from mice, and co-cultured with either oleic acid (100 microM) or linoleic acid (100 microM) for 2 days. The expression of alpha-smooth muscle actin (alpha-SMA) was assessed by immunohistochemistry. In addition, the production of hydroxyproline was determined.

Serum alanine aminotransferase levels and the mRNA expression of TGF-beta and collalpha2 were significantly reduced by treatment of olive oil. Dietary olive oil blunted the expression of alpha-SMA in the liverand liver injury and hepatic fibrosis were prevented by treatment of olive oil. The number of alpha-SMA positive cells was significantly lower in HSCs co-cultured with oleic acid than in those co-cultured with linoleic acid. Concentration of hydroxyproline in culture medium was significantly lower in cells co-cultured with oleic acid than in the control.
Dietary olive oil prevents CCl(4)-induced tissue injury and fibrosis in the liver. Since oleic acid inhibited activation of HSCs, oleic acid may play a key role on this mechanism.