Why sugar doesn’t make you fat – the ultimate guide on lipogenesis

Ever had a potato turn into a slab of butter?

Some people think that it’s possible in the body where carbs are converted to fat and then it makes you fat real fast, so they avoid carbs like the plague.

The process is called de novo lipogenesis (DNL) – the creation of new fats.


This process starts in the mitochondria of the cell. Carbs are broken down to pyruvate, which enters the TCA cycle and is converted to citrate. When there is too much pyruvate streaming in and the TCA cycle is moving too slowly and citrate is not converted fast enough, the citrate is transporter out of the mitochondria into the cytoplasm of the cell where it undergoes DNL.

The first set is where it’s converted to malonyl-CoA, which is an inhibitor of the enzyme carnitine palmitoyltransferase I (CPT1). CPT1 transports fatty acids into the mitochondria for β-oxidation, so malonyl-CoA controls the amount of fats being burned. This is a good thing, because when a lot of carbs are coming in, they get preference and when they don’t…the person’s diabetic.

Next malonyl-CoA is further converted to palmitic and stearic acid, but mostly palmitic.  This can then be further elongated to very long chain fats, or be unsaturated by the SCD (stearyl-CoA desaturase) enzyme to create palmitoleic acid from palmitic acid and oleic acid from stearic acid.

Furthermore, DNL is involved in the synthesis of phospholipids for cell membranes (which is necessary for proper cellular function and muscle strength), lipoic acid, sebum, brain development, muscle cell fuel selection,
hypothalamic appetite control (malonyl-CoA reduces fat oxidation and promotes satiety), generation of endogenous ligands for the nuclear receptor PPARα in liver, pancreatic β-cell signaling, and more.

FAHFAs (branched fatty acid esters of hydroxy fatty acids) created by DNL have beneficial metabolic effects, including enhancing insulin-stimulated glucose transport and glucose-stimulated glucagon like peptide 1 (GLP1) and insulin secretion, as well as powerful anti-inflammatory effects. These anti-inflammatory effects of one of these FAHFAs, namely PAHSAs (palmatic acid ester of hydroxy stearic acid) may be mediated through the GPR120 receptor similar to the anti-inflammatory effects of ω-3 fatty acids (R).

Adipose tissue dysfunction, which promotes excessive inflammation and lipolysis, is associated with low levels of PAHSAs. Could dysfunctional adipose tissue be due to insufficient adipose tissue DNL? This study shows that DNL is actually too low in dysfunctional adipose tissue.

Interestingly, a high fat diet (predominantly PUFAs) reduces DNL in the liver and adipose tissue but not in muscle (R), which could lead to adipose dysfunction.

DNL’s role in the body still remains largely unknown and new research are just beginning to explore that area.

DNL can occur mostly everywhere in the body, but three of the major sites are skeletal muscle, liver and adipose tissue.

DNL in adipocytes where, unlike in liver cells, it has beneficial metabolic effects since it improves insulin sensitivity, enhances glucose transporter-2 (GLUT2) receptor expression and glucose uptake.

DNL is muscles ensure cellular integrity and promote muscular strength. Intramuscular fat is not harmful if it’s regularly depleted by exercise, however it will promote insulin resistance and cellular dysfunction is it’s not. Also, dietary fat can induce obesity and promote diabetes, but endogenous fat production (DNL) is not thought to affect skeletal muscle insulin resistance, an antecedent of metabolic disease. Elevated lipolysis causes intramuscular fat accumulation, insulin resistance and dysfunctional cells.

In the liver, excess DNL is another story. In pretty severe conditions such as nonalcoholic steatohepatitis, inhibiting DNL has been useful, but that’s because DNL was elevated several fold above normal (R), by things like excess inflammation, ROS, insulin, serotonin, prolactin, etc.

Furthermore, excess DNL and fat accumulation in the liver results in hepatic steatosis, fibrosis and non-alcoholic steatohepatitis (NASH); fat in the pancreas is associated with impaired insulin secretion, β-cell dysfunction and apoptosis; excess intramyocardial fat leads to cardiomyopathy, coronary heart disease and sudden death; in the skeletal muscles, intramyocellular triglycerides are associated with insulin resistance and impaired glucose uptake. But I’d say the excess fat is due to elevated lipolysis and to a smaller degree DNL.

In mouse studies, inhibiting DNL protects the mice from fatty liver and insulin resistance, but not from obesity (R). However, inactivation of SCD-1 in the liver, which unsaturates the fats, prevents obesity induced by high carbohydrate feeding. But those are animal studies and we need more human studies to confirm the same in humans.

Let’s look at the rates of DNL. How much fat is actually created from carbs.

Donnelly et al. (R) were able to conclude that 59% of triglycerides (TGs) in the livers of patients with non-alcoholic fatty liver disease (NAFLD) were from lipolysis, 26% from DNL and 15% from the diet. 26% isn’t even that high, but compared to the 2–5% in normal subjects consuming a typical Western diet, it’s pretty high (R).

De novo lipogenesis pie chart

So DNL contributes to only 2-5% fat in the liver in healthy people, and that shoots up to 20-26% in people with fatty liver and even higher in people with hepatitis, HIV and AIDS.

Furthermore, a “simple” way to check if DNL is elevated is to check respiratory quotient (RQ). If only fats are burned, it’s 0.7, if it’s 1, only carbs are burned, if it’s over 1, DNL is activated. With that in mind, a single large meal containing up to 500g of carbs does not cause RQ to exceed 1.0, which indicates DNL isn’t elevated (R).

Another thing to keep in mind, if DNL did contribute to a significant amount to fat gain, it would reflect in the adipose tissue. Meaning, the type of fat created would be high in the adipose tissue. Yet it’s not. The adipose tissue reflects the fats consumed in the diet, and not DNL.

Let’s look at more specific rates:

The amount of fat synthesized from carbs are normally less than 10g and sometimes less than 1g per day (R), but this might increase a little with a caloric surplus and a low fat diet.

Reducing the amount of fat in the diet can increase DNL. A very low fat high carb diet (10% fat/70% carbs) for three weeks increased DNL by 30% (R). That would increase total fat synthesized from 10g to 13g daily. Wow right? Nope.

With overfeeding or really high carb intake, DNL doesn’t immediately shoot up, but it takes about 3-5 days for it to maximally upregulate. It takes a very high carb diet at about 8000 calories per day before DNL is fully induced.

But even with a 50% surplus diet with 800g carbs daily, absolute DNL in the liver accounts for < 5g fatty acids synthesized per day.

CHO overfeeding DNL

A high carb day increased DNL by 1-5%. This represents a maximum of only 2.5 grams palmitate per day (R). Pretty minuscule.

The reason only a small amount of carbs is converted to fat is because carbs are stored as glycogen and the rest is burned off rapidly. With surplus high carb intake, whole-body glucose oxidation increases six fold and fat oxidation decreases by > 90% (R).

A study by Buul et al. found that overfeeding daily with 150–200g fat and 750–1000g carbohydrates led to a DNL of 5g fat per day, equivalent to less than 3% of the total fat consumed (R).

As we can see, DNL from glucose is pretty small, but what about fructose?

Let’s talk about the F word: Fructose

Fructose is thought to be highly lipogenic and much more fattening than glucose because it stimulates DNL to a greater extent than glucose…or so people think.

Here are a few studies we can look at:

  • DNL from fructose contributes to less than 1% of circulating fatty acids (R).
  • There’s no difference between liver fat or triglycerides in glucose or fructose overfeeding (R), meaning, DNL is stimulated equally by both.
  • Both glucose and fructose cause equal DNL in the liver but glucose actually created more muscle fat (intramyocellular fat) than fructose (R).
  • Less than 10% of a bolus of 250g of fructose was converted to fats after 6 hours, which means that less than 1g of fat is created per hour from a massive unnatural amount of fructose (R).
  • A 2-month high-fruit vs. high-nut diets in 30 healthy participants resulted in a 3-fold increase in fructose intake in the fruit group and was not associated with a change in liver fat (R).
  • Long-term fructose in the diet (15% of calories for three weeks in a eucaloric diet; which would be 94g of fructose on a 2500 calorie diet) did not alter fractional or absolute DNL in the fasted state or in response to acute fructose ingestion (R).
  • The type of carbohydrate overfeeding (sucrose or glucose) had no significant effect on DNL in either subject group (R).

There you have it, glucose and fructose stimulate DNL to a similar extent and fructose is not more fattening than glucose in regards to DNL.



Under normal conditions DNL is not a major way to increase body fat stores, because it represented only a few grams per day.

However, under conditions of excess insulin and inflammation, lipogenesis becomes upregulated, but it isn’t lipogenesis that’s to blame, but rather excess lipolysis and adipose dysfunction.

Excess insulin and inflammation occur first, which is why substances that block excess lipolysis, such as niacin and aspirin, reduce fat accumulation in the liver and the rest of the body.

Niacin reduces liver fat (R) and there is a new study with niacinamide for treating fatty liver which’s results I’m awaiting, but I’m sure it will be very promising (R). The most likely reason is due to lipolysis inhibition and boosting NAD levels (R).

Additionally, daily aspirin use is associated with reduced risk for fibrosis progression in patients with NAFLD (R) and it’s also inversely correlated with NAFLD (R).

But each individual is different and DNL is highly variable between individuals (R).

Overfeeding DNL

As you can see from the graph above, some people actually have higher DNL from glucose than fructose, whereas others are the opposite.

So how do you find what works for you?

Look back in your life when you were lean and what you ate then. If you never were lean or were eating junk at the time, don’t worry we can still find out how.

Here’s what you do. Pick 3-4 carb sources that you’d like and think you’d do good on. Maybe it’s sweet potato, milk, honey and some fruits, like oranges and bananas.

Next, substitute all the carbs that you’re eating currently for the above sources. Eat a generous amount of it each day for a week and keep an eye on how you feel when you eat it, how is your digestion and what happens to your weight.

Eating the right food for you will help improve digestion, reduce weight, improve energy and so on, so it will be easy to identify with a little consistency.

A final word on fructose:

I think the fructose in fruits and honey is great and not a problem and even normal white sugar can be acceptable for some. However, white sugar is awful for others.

In this study the researchers reduced the sugar content from obese children’s diets (from 28% to 10%) for 9 days and replaced it with equal amounts of refined glucose so that calories stayed more or less the same.

As a result, DNL dropped by almost 60%, L-lactate (a product of glycolysis that promotes fatigue) by 33%, D-lactate (a detox product of methylglyoxal which causes glycation products such as HbA1c) by 50%, liver fat by 37% and insulin by 30% in just 9 days (R).

I think the main reason for this benefit is because high fructose with high PUFA intake is synergistically detrimental (R). Cutting out fructose would have the fastest “beneficial” effect, whereas cutting out PUFAs would take longer to see benefits, but that would definitely be better for health in the long run.

This just shows that when the conditions aren’t perfect in the body, fructose might just make everything worse as indicated by the studies above.

But at the end of the day, experimentation is key to see which carbs you do best on, and if you don’t do good on fructose for now, reduce it to minimum levels and try to increase it later on as your health improves and PUFA levels drop in the body.

A final word on saturated fat:

Many think that saturated fat is evil and causes fatty liver and because DNL creates saturated fat, it’s also evil. Well, a new study showed that intrahepatic lipids were negatively correlated with VLDL-triglyceride secretion rates and that VLDL-triglyceride secretion rates were positively correlated with hepatic concentrations of saturated diacylglycerol (DAG) (R). Meaning saturated DAG promotes VLDL-triglyceride secretion from the liver and prevents fatty liver. Furthermore, saturated fats just require a high protein and choline intake to prevent accumulation in the liver.

It’s polyunsaturated fat that’s the bad guy that causes fatty liver, liver damage and later liver dysfunction (R).

So how do we inhibit excess lipogensis?

Things that inhibit lipogenesis:

  • Vitamin D (R)
  • Cholesterol (R)
  • Testosterone (R)
  • DHEA (R)
  • DHT (R)
  • Phytic acid and myo-inositol (R). The myo‑inositol (MI) ring of PA is thought to be the active one.
  • Magnesium and selenium co-administration (R)
  • Caffeine (R)
  • Carbonic anhydrase inhibitors (R, R). Vitamin B1 is a good inhibitor.
  • α-lipoic acid (R)
  • Biotin (R)
  • Uncoupling. UCP1 deficiency elevates adipose stearoyl-CoA desaturase (SCD) gene expression (R)
  • Intracellular calcium promote lipogenesis, whereas calcium supplementation can help reduce intracellular calcium (R).
  • Palmitate. Possibly in a negative feedback loop.
  • Vitamin C (R)
  • Mountain celery (R)
  • FXR activation by bile acids (Ursodeoxycholic acid (UDCA) (R), TUDCA (R), taurine) and androsterone (R).
  • Aspirin (R)
  • EGCG (R)
  • Cocoa (R)
  • Alcohol (R)
  • And many others which I will not list because this list will become waaayyy too long…

Things that increases lipogenesis:

  • Thyroid hormones (R), but it also promotes lipolysis and fat oxidation, so there is a net negative fat balance.
  • Iron. Lipogenesis correlates with iron levels in diet and tissue (R)
  • Inflammation (R)
  • Serotonin (A)
  • Prolactin (A)
  • Endotoxins (R)
  • Estrogen (only weakly affects DNL)
  • Hyperinsulinemia (can boost DNL by 3 fold)
  • Hyperuricemia (R, R)
  • Lactate (R). Lactate is also a major substrate for DNL itself, so lowering lactate will lower DNL.
  • Viral infections (R)
  • Corticotropin-releasing hormone (CRH), ACTH and cortisol excess (R, R, R). Testosterone and growth hormone were the only two hormones that induces CRH negative feedback.
  • People with already fatty liver. Subjects with high liver fat had more than 2-3.5 fold (23.2% vs 10.1%) higher rates of DNL than subjects with low liver fat (R).

Sign up for my FUN FACT FRIDAY Newsletter

Where I share a weekly dose with my readers of small things I did that week; things I found interesting, maybe a good book I’m reading, something I’m experimenting with, an inspirational quote etc., and will also give you a link to the article I did that week.

So don’t miss out!

Thanks for signing up. Great to have you!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.