Vitamin A and Copper: The Retinol Connection in the Root Cause Protocol

Table of Contents

1. 1. The Claim in Robbins’s Own Framing
2. 2. The Copper Machinery: ATP7A, ATP7B and Three Enzymes
3. 3. Evidence Check: Does Retinoic Acid Switch On Copper Handling?
4. 4. Evidence Check: Does Vitamin D Block Vitamin A?
5. 5. Retinol vs Beta-Carotene: The BCMO1 Question
6. 6. Cod Liver Oil and the Retinol-to-D Ratio
7. 7. The Best Retinol Foods, by the Numbers
8. 8. Safety: Upper Limit, Pregnancy, Liver and Bone
9. 9. What Holds Up, What Is Plausible, What Is Unproven
10. Key Research Papers
11. Connections
12. Featured Videos

1. The Claim in Robbins’s Own Framing

Robbins describes a “copper–retinol axis” that works in opposition to what he calls the “iron–sugar axis.” In interviews he puts it plainly: there are “little tiny pumps” in the body named ATP7A and ATP7B, “they’re activated by retinoic acid,” and retinoic acid is made from retinol — “you can’t get there from beta carotene.” In his telling, vitamin A’s most important job is to be “the clinical and metabolic factor to make copper bioavailable.” Without retinol, copper may be present in the diet and in the blood, but it is not loaded into the proteins that put it to work, above all ceruloplasmin, the copper protein that escorts iron (see Ceruloplasmin and Bioavailable Copper).

The claim has four working parts:

1. Retinol → retinoic acid → copper pumps. Preformed vitamin A becomes retinoic acid, which switches on ATP7A and ATP7B so copper gets loaded into ceruloplasmin, lysyl oxidase (the enzyme that cross-links collagen and elastin) and Cu/Zn superoxide dismutase (SOD1, an antioxidant enzyme).
2. High-dose vitamin D is a brake. Robbins says that vitamin D supplements “block vitamin A uptake,” so a population “drowning in vitamin D” is functionally short of retinol and therefore of usable copper. This is one plank of his vitamin D position.
3. Cod liver oil over D-only pills. He calls cod liver oil “a beautiful source of A and D” because it carries “at least 10 times more retinol” than vitamin D, so the A comes with its own counterweight.
4. Eat the retinol foods. His list: beef liver, cod liver oil, grass-fed butter and heavy cream, and pastured (“yard”) eggs with deep orange yolks. He also argues that modern advice has overstated retinol toxicity.

Each part is a different kind of claim. The first is about molecular biology, the second about a nutrient interaction, the third about a ratio in a food, and the fourth about diet and safety. They do not stand or fall together, so the sections below test them one at a time.

Back to Table of Contents

2. The Copper Machinery: ATP7A, ATP7B and Three Enzymes

The pumps Robbins names are real and important. ATP7A and ATP7B are copper-transporting ATPases — membrane proteins that spend energy to move copper out of the cell’s fluid and into the compartment where new proteins are finished (the secretory pathway), where copper is built into copper-dependent enzymes. Losing ATP7A causes Menkes disease, a fatal copper-starvation disorder of infancy; losing ATP7B causes Wilson’s disease, in which copper piles up in the liver and brain (Lutsenko 2007, review).

Where the textbook picture differs from the Root Cause Protocol version is in the details of which pump feeds which enzyme:

• Ceruloplasmin is made in liver cells and gets its copper from ATP7B. That is why people with Wilson’s disease typically have low ceruloplasmin even while copper overloads their liver.
• Lysyl oxidase, the collagen and elastin cross-linker, is a secreted enzyme served by ATP7A in most other tissues.
• SOD1 is different. It lives in the cell’s fluid, not the secretory pathway, and it receives copper from a separate escort protein, the copper chaperone for superoxide dismutase (CCS). In mice bred without CCS, SOD1 activity fell sharply while copper delivery to the other copper enzymes was normal (Wong 2000, animal). So SOD1 is a copper enzyme, but it is not loaded by ATP7A or ATP7B.

The diagram follows the claim step by step. The green arrows from each pump or chaperone to its enzyme are textbook biology. The cyan arrows from retinoic acid to the pumps have support in a 1987 rat study and a 2009 tumour-cell study, covered in the next section. The red dashed vitamin D brake is the weakest link.

Back to Table of Contents

3. Evidence Check: Does Retinoic Acid Switch On Copper Handling?

This is the strongest of Robbins’s claims, and the literature does contain a real thread behind it.

• Rats, 1987 (animal). Barber and Cousins injected rats with 13-cis retinoic acid. Ceruloplasmin activity rose within 24 hours and, with daily injections, reached about four times control levels after 4 days; new ceruloplasmin synthesis rose about 1.5-fold. Two details matter. In copper-deficient rats, retinoic acid did nothing unless copper was also given — retinoic acid cannot make ceruloplasmin out of thin air. And when vitamin A-deficient rats were given retinoic acid, their ceruloplasmin activity and synthesis rose compared with untreated deficient rats. Evidence tier: animal, pharmacological doses.
• Tumour cells, 2009 (in vitro). In neuroblastoma cells, Bohlken and colleagues found that ATP7A is a retinoid-responsive gene, switched on through the retinoic acid receptor RARβ2, and that this changed copper efflux from the cells. This is the closest direct support for “retinoic acid activates ATP7A” — but it was shown in a childhood cancer cell line, studied to understand retinoid cancer drugs, not in healthy human tissue.
• Patient-derived liver cells, 2022 (in vitro). Song and colleagues grew liver cells from stem cells of people with Wilson’s disease. Those cells made and secreted less ceruloplasmin, showed abnormal retinoid signalling, and a drug screen found that retinoids rescued ceruloplasmin secretion. A striking result — and still a dish of cells, not a treatment trial.

What this does and does not show. Retinoic acid can raise ceruloplasmin in animals and can switch on ATP7A in at least one cell type. Vitamin A status and copper handling are connected. What has not been shown is the practical claim: that ordinary people eating a typical diet have under-loaded copper enzymes because they lack retinol, or that eating more retinol raises their ceruloplasmin. No human trial has tested that. One more caution: ceruloplasmin is also an acute-phase protein that rises with inflammation, so a change in it is not automatically a sign that copper is being “used” better. Evidence tier for the overall claim: plausible mechanism, animal and cell support, untested in people.

Back to Table of Contents

4. Evidence Check: Does Vitamin D Block Vitamin A?

Vitamins A and D do interact — their receptors partner with the same helper receptor inside the cell, and researchers have studied the tug-of-war for decades. The question is direction and dose.

• Vitamin A blunting vitamin D is the better-documented direction. In weanling rats, rising doses of retinyl acetate progressively reduced bone ash and, at the higher doses, stopped vitamin D from raising blood calcium (Rohde 1999, animal). In people, a small double-blind crossover study gave 9 healthy volunteers 15 mg of retinyl palmitate — about what one serving of liver delivers — and found it lowered blood calcium on its own and blunted the calcium response to active vitamin D (Johansson 2001, human, acute, very small).
• Vitamin D lowering vitamin A has animal support. In young broiler chickens, adding 25-hydroxy vitamin D or active vitamin D to the feed reduced plasma vitamin A, and active vitamin D also reduced liver vitamin A (Aburto 1998, animal). The same experiments showed the familiar reverse effect: a high vitamin A diet of 45,000 IU per kg of feed interfered with vitamin D and raised the incidence of rickets. Chicken feed doses do not translate directly to human supplements.
• The one human trial that looked did not find it. In a randomized, double-blind, placebo-controlled trial, 85 adults took 800 IU of vitamin D3 a day, with or without calcium, for 6 months; there was no significant effect on serum retinol (Chai 2012, human RCT). Two caveats keep this from being the last word: 800 IU is a modest dose, far below the 5,000–10,000 IU some people take, and serum retinol is held steady by the liver until stores are badly depleted, so it is a blunt measure.
• Population data. In 21,774 older Norwegians — a population with high intakes of both vitamins, largely from cod liver oil — there was no significant interaction between blood vitamin D and blood retinol on hip fracture risk (Holvik 2015, cohort).

Verdict. “Vitamin D supplements block vitamin A uptake” is not established in humans. The animal evidence points both ways, the clearer human signal runs the other way (vitamin A blunting vitamin D), and the only human trial found no drop in retinol at 800 IU. Whether very high-dose D3 taken for years depletes liver retinol stores is a fair research question that has not been answered. That is why the brake is drawn dashed in the diagram above.

Back to Table of Contents

5. Retinol vs Beta-Carotene: The BCMO1 Question

Robbins says “you can’t get there from beta carotene.” Taken literally, that is not right: the body cuts beta-carotene into retinal using the enzyme BCMO1 (beta-carotene 15,15′-monooxygenase), and retinal becomes both retinol and retinoic acid. Plant carotenoids are the main vitamin A source for much of the world.

But there is a real point underneath. Conversion is highly variable. Leung and colleagues (2009, human volunteers plus enzyme studies) noted that up to 45% of healthy people are “poor converters” and identified two common variants in the BCMO1 gene. Carriers of one variant converted 32% less beta-carotene, and carriers of both converted 69% less. So for some people, a carrot-and-sweet-potato diet delivers far less vitamin A than the label arithmetic suggests — a fair reason to include some preformed retinol from animal foods. A more accurate version of the claim: beta-carotene works, but unreliably for a large minority, and retinol bypasses that step.

One related claim does not hold up. Robbins says a pale yellow yolk means the egg “does not have enough retinol” and that “retinol is going to turn it orange.” Retinol itself is nearly colourless at food concentrations; yolk colour comes mainly from lutein and zeaxanthin, plant pigments in the hen’s feed that the body does not convert to vitamin A. A deep orange yolk says a lot about what the hen ate and is a reasonable sign of a pastured diet, but it is not a retinol gauge. See Eggs.

Back to Table of Contents

6. Cod Liver Oil and the Retinol-to-D Ratio

Robbins’s arithmetic checks out against the government food database. USDA FoodData Central lists cod liver oil (FDC 173577) at 30,000 mcg retinol and 250 mcg vitamin D per 100 g. One teaspoon (4.5 g) therefore holds about 1,350 mcg retinol and about 11 mcg (450 IU) of vitamin D. Converted to international units, 1,350 mcg of retinol is about 4,500 IU — so the retinol-to-D ratio is about 10 to 1 in IU, which is the figure he quotes. A tablespoon (13.6 g) carries about 4,080 mcg retinol, which by itself is above the adult upper limit.

Three things to know before acting on that:

• Products vary widely. The USDA entry is a single reference value. Many commercial cod liver oils are refined, which strips vitamins, and some then add vitamins back at chosen levels. Read the label for retinol (or “vitamin A as retinyl palmitate”) and vitamin D per serving rather than assuming a ratio. Robbins recommends particular cod liver oil brands; this site does not name or rank brands.
• The dose is small. At these values, a teaspoon a day already supplies more than the 900 mcg adult RDA; a tablespoon a day exceeds the 3,000 mcg upper limit before any food is counted.
• Pairing A with D has a reasonable basis — vitamin A and D act together on bone and the immune system — and the Norwegian cohort authors concluded that cod liver oil should not be discouraged as a vitamin D source (Holvik 2015). But that is an argument for moderate cod liver oil, not evidence for the copper mechanism.

For the food itself, see Cod Liver.

Back to Table of Contents

7. The Best Retinol Foods, by the Numbers

All values below are retinol (preformed vitamin A) from USDA FoodData Central’s reference database, multiplied out to an ordinary serving. Pasture-raised products may differ from these averages; the USDA figures are for conventional foods.

• Beef liver, pan-fried — FDC 168627: 7,728 mcg per 100 g, so a 3-ounce (85 g) serving gives about 6,569 mcg — more than twice the daily upper limit in one meal. The same serving also carries about 12.4 mg of copper (14.588 mg per 100 g), above the 10 mg daily copper upper limit. That is fine as an occasional meal; it is not a daily food. See Beef Liver.
• Chicken liver, pan-fried — FDC 174491: 4,293 mcg per 100 g, about 3,649 mcg per 3-ounce serving, also above the upper limit. See Chicken Liver.
• Cod liver oil — FDC 173577: about 1,350 mcg per teaspoon (4.5 g).
• Heavy whipping cream — FDC 170859: 405 mcg per 100 g, about 122 mcg per 2 tablespoons (30 g).
• Butter — FDC 173410: 671 mcg per 100 g, about 95 mcg per tablespoon (14.2 g).
• Egg yolk — FDC 172184: 371 mcg per 100 g, about 63 mcg per large yolk (17 g).

The chart shows the practical point: retinol is extremely concentrated in liver and in cod liver oil, and only modest in dairy fat and yolks (the axis is marked at 0, 2,000, 4,000 and 6,000 mcg). Butter, cream and egg yolks are good foods, but they add little vitamin A per serving — about 95, 122 and 63 mcg. The adult RDA is 900 mcg RAE a day for men and 700 for women; the adult upper limit for preformed vitamin A is 3,000 mcg. Eating liver once or twice a week, rather than daily, keeps the weekly average comfortably inside the limit for most adults. More on food sources: Vitamin A Sources.

Back to Table of Contents

8. Safety: Upper Limit, Pregnancy, Liver and Bone

Robbins argues that the fear of retinol has been overdone. Part of that is fair: most documented harm comes from high-dose supplements taken daily for months or years, not from eating liver now and then, and beta-carotene from food carries none of these risks. But retinol toxicity is real and well documented, and the limits below come from the US National Academies’ Dietary Reference Intakes for Vitamin A, Copper and other nutrients (2001).

• Upper limit. For adults the Tolerable Upper Intake Level for preformed vitamin A (retinol and retinyl esters) is 3,000 mcg a day (10,000 IU). Beta-carotene from food does not count toward it. The copper upper limit is 10 mg a day, and the adult copper RDA is 900 mcg.
• Pregnancy — the most important caution. Retinol causes birth defects. In 22,748 pregnant women, those taking more than 10,000 IU a day of preformed vitamin A from supplements had about 4.8 times the rate of cranial-neural-crest birth defects compared with women taking 5,000 IU or less, with the risk concentrated in exposure before the seventh week; the authors estimated about 1 baby in 57 was affected by the supplement at that dose (Rothman 1995, cohort). Anyone who is or may become pregnant should stay under the upper limit from all preformed sources combined, which means being careful with both liver and cod liver oil. Beta-carotene does not carry this risk.
• Liver damage. In 41 patients with vitamin A liver toxicity, 17 had cirrhosis; the smallest continuous intake that led to cirrhosis was 25,000 IU a day for 6 years (Geubel 1991, case series). That is about 7,500 mcg a day — roughly a daily 3-ounce serving of beef liver — which is why liver belongs in the weekly menu rather than the daily one.
• Bone — mixed evidence. In 2,322 Swedish men followed for 30 years, those in the highest fifth of blood retinol had a 1.64-fold risk of any fracture and 2.47-fold risk of hip fracture compared with the middle fifth (Michaëlsson 2003, cohort). The larger Norwegian study of 21,774 older adults found no increase in hip fracture risk at high blood retinol (Holvik 2015, cohort). The honest reading: chronically high retinol may harm bone in some populations, and there is no reason to push intake above the upper limit. See Osteoporosis.
• Wilson’s disease. Anyone with Wilson’s disease must not follow copper-raising advice; beef liver, at about 12.4 mg copper per serving, is one of the richest copper foods there is.
• Anemia is a diagnosis to finish, not a reason to guess. If you have true iron-deficiency anemia, a cause has to be found — most often blood loss from heavy periods or the gut. Adding retinol or withholding iron does not replace that workup.

Full details: Vitamin A Toxicity.

Back to Table of Contents

9. What Holds Up, What Is Plausible, What Is Unproven

Well established:

• ATP7A and ATP7B are essential copper pumps; ATP7B loads ceruloplasmin in the liver and ATP7A serves secreted copper enzymes such as lysyl oxidase.
• Beef liver and cod liver oil are the richest common sources of retinol; dairy fat and yolks contain modest amounts.
• Cod liver oil carries roughly 10 times more vitamin A than vitamin D in IU, at USDA reference values.
• Beta-carotene conversion varies a great deal between people, partly because of common BCMO1 gene variants.
• Preformed vitamin A has a 3,000 mcg adult upper limit and is a proven cause of birth defects at high intake.

Plausible but unproven in people:

• That retinoic acid meaningfully regulates copper handling in healthy humans. It raised ceruloplasmin in rats (1987), switched on ATP7A in tumour cells (2009) and rescued ceruloplasmin secretion in Wilson’s-disease liver cells (2022) — but no human study has shown that more dietary retinol raises ceruloplasmin or improves copper status.
• That very high-dose vitamin D taken for years depletes vitamin A stores. Chicks show it; an 800 IU human trial did not.

Not supported as stated:

• That SOD1 depends on the ATP7A/ATP7B pumps — it gets copper from its own chaperone, CCS.
• That beta-carotene cannot produce retinoic acid at all — it can, though unreliably for many people.
• That an orange yolk signals retinol — the colour comes mostly from non-vitamin-A carotenoids.

The practical overlap between Robbins’s advice and mainstream nutrition is larger than the disagreement: eat some liver occasionally, use real butter and eggs, and do not rely on carrots alone if you may be a poor converter. Where they part ways is the dose and the certainty. For the wider protocol, see The Root Cause Protocol, and for the official programme description, the Root Cause Protocol website.

Back to Table of Contents

Key Research Papers

1. Barber EF, Cousins RJ (1987). Induction of ceruloplasmin synthesis by retinoic acid in rats: influence of dietary copper and vitamin A status. J Nutr. — PubMed PMID: 3655940
2. Bohlken A, Cheung BB, Bell JL, et al. (2009). ATP7A is a novel target of retinoic acid receptor beta2 in neuroblastoma cells. Br J Cancer. — PubMed PMID: 19127267
3. Song D, Takahashi G, Zheng YW, et al. (2022). Retinoids rescue ceruloplasmin secretion and alleviate oxidative stress in Wilson’s disease-specific hepatocytes. Hum Mol Genet. — PubMed PMID: 35388883
4. Lutsenko S, Barnes NL, Bartee MY, Dmitriev OY (2007). Function and regulation of human copper-transporting ATPases. Physiol Rev. — PubMed PMID: 17615395
5. Wong PC, Waggoner D, Subramaniam JR, et al. (2000). Copper chaperone for superoxide dismutase is essential to activate mammalian Cu/Zn superoxide dismutase. Proc Natl Acad Sci U S A. — PubMed PMID: 10694572
6. Rohde CM, Manatt M, Clagett-Dame M, DeLuca HF (1999). Vitamin A antagonizes the action of vitamin D in rats. J Nutr. — PubMed PMID: 10573558
7. Aburto A, Edwards HM Jr, Britton WM (1998). The influence of vitamin A on the utilization and amelioration of toxicity of cholecalciferol, 25-hydroxycholecalciferol, and 1,25 dihydroxycholecalciferol in young broiler chickens. Poult Sci. — PubMed PMID: 9565243
8. Johansson S, Melhus H (2001). Vitamin A antagonizes calcium response to vitamin D in man. J Bone Miner Res. — PubMed PMID: 11585356
9. Chai W, Bostick RM, Ahearn TU, et al. (2012). Effects of vitamin D3 and calcium supplementation on serum levels of tocopherols, retinol, and specific vitamin D metabolites. Nutr Cancer. — PubMed PMID: 22149065
10. Leung WC, Hessel S, Méplan C, et al. (2009). Two common single nucleotide polymorphisms in the gene encoding beta-carotene 15,15′-monoxygenase alter beta-carotene metabolism in female volunteers. FASEB J. — PubMed PMID: 19103647
11. Rothman KJ, Moore LL, Singer MR, et al. (1995). Teratogenicity of high vitamin A intake. N Engl J Med. — PubMed PMID: 7477116
12. Geubel AP, De Galocsy C, Alves N, Rahier J, Dive C (1991). Liver damage caused by therapeutic vitamin A administration: estimate of dose-related toxicity in 41 cases. Gastroenterology. — PubMed PMID: 2019375
13. Michaëlsson K, Lithell H, Vessby B, Melhus H (2003). Serum retinol levels and the risk of fracture. N Engl J Med. — PubMed PMID: 12540641
14. Holvik K, Ahmed LA, Forsmo S, et al. (2015). No increase in risk of hip fracture at high serum retinol concentrations in community-dwelling older Norwegians: the Norwegian Epidemiologic Osteoporosis Studies. Am J Clin Nutr. — PubMed PMID: 26377161

PubMed Topic Searches

1. PubMed: Retinoic acid, ceruloplasmin and copper
2. PubMed: Vitamin A and vitamin D interaction
3. PubMed: BCMO1 and beta-carotene conversion

Back to Table of Contents

Connections

• All Notable Doctors
• Morley Robbins
• The Root Cause Protocol
• Ceruloplasmin & Bioavailable Copper
• Copper–Iron Dysregulation
• The Vitamin D Controversy
• Whole-Food Copper Sources
• Vitamin A
• Vitamin A Toxicity
• Vitamin D3
• Copper
• Ceruloplasmin Test
• Beef Liver
• Cod Liver
• Wilson’s Disease

Back to Table of Contents

Featured Videos