Health authorities advocate restricting red meat in our diet, even though they can’t agree what red meat is. If pressed about what it is about red meat that makes it a health risk (e.g. for colorectal cancer, CRC), the common answer is haem-iron (HI). There is a neat convenience about this answer, because plant matter does not contain HI — plants contain iron, but not in the haem form. In this post, I will look more closely at HI, non-haem-iron (nHI) and iron absorption.
What is haeme iron?
Haem iron is a ring-shaped molecule composed of carbon, hydrogen and oxygen, with four nitrogens in the centre holding an iron atom:
The iron, in turn, binds or releases molecular oxygen according to circumstances. HI is necessary and plentiful in the blood of all animals that breathe oxygen. It is present to varying degrees in meat from those animals, and it is also present in internal organs such as liver and kidney.
The HI complex is itself attached to larger protein structures (collectively known as globulins) that allow the HI to be water-soluble. There are a number of HI-plus-protein (haemoprotein) combinations that perform multiple critically-important functions in our bodies, but the two of most interest to the present topic are haemoglobin in circulating red blood cells (RBCs) and myoglobin in muscle cells. In general, HI is responsible for the colour (usually redness) of blood or muscle.
Haemoglobin is not important to the topic of this post (I will explain why later). However, it is a good introduction to HI.
Oxygen is essential for the metabolism of nearly every cell in our body (although not RBCs), and it must be delivered (from the lungs) to those cells through the blood circulation. However, oxygen is not very soluble in water, and so it must be carried around the circulation by a transporter — haemoglobin. The HI in haemoglobin attaches to oxygen in the environment of the lung, and releases the oxygen to cellular machinery that need it.
As oxygen is burned for energy by those cells, carbon dioxide (CO2) and water are produced, and haemoglobin carries the CO2 back to the lungs to be exhaled (but not bound to the iron, it is attached elsewhere on the globulin structure). The water can be utilised if needed, or eliminated in various ways (sweat, urine, exhalation or faeces). Breathing out CO2 is the main mechanism whereby our weight can reduce — we do not reduce weight by burning fat or glucose for energy as such, but rather by eliminating the byproducts of that metabolism from our bodies.
About 70% of the HI in our bodies is contained in circulating haemoglobin. There are some 5 million RBCs per millilitre of blood in an adult human (a number worth pondering); each RBC contains 200–300 million molecules of haemoglobin, and each molecule of haemoglobin contains 4 HI complexes (and can thus carry 4 oxygen molecules). This equates to about 30 quadrillion HI complexes in our circulatory system at any one time, constantly interacting with all oxygen-dependent cells in our bodies. This exposes every cell that has a blood supply to systemic HI, including those of the colon wall, at least in some way.
Because of the importance of HI, our bodies constantly synthesise it. In the case of haemoglobin, this occurs primarily in bone marrow (mostly in the pelvis, spine and sternum). The process in marrow involves: making haem, binding it to iron to make HI, attaching globulin proteins to make haemoglobin and finally producing RBC precursors from stem cells and releasing them into the circulation, bundled with a cargo of haemoglobin, where they mature into RBCs.
RBCs are almost unique in the body, because they jettison much of their cellular machinery (including mitochondria and the DNA-containing nucleus) to make more room for haemoglobin. The RBCs effectively cannot replicate because they have no DNA with which to do so. They just keep working at their task until they wear out or become damaged (after about 3 months on average). They are then directed to the spleen, which is responsible for breaking down RBCs and their haemoglobin for recycling. The gobulin proteins are broken down to their amino acids. The iron in HI is stripped off the haem, packaged safely and sent to the liver for recycling or storage. The remaining haem is not recycled and is potentially damaging (if haem has a health risk, the organ at most risk should be the spleen). The spleen modifies the haem and sends it to the liver which converts into a harmless form (bilirubin) and sends it to the gall bladder, where bilirubin combines with bile and discharges into the small intestine, ultimately being eliminated in faeces. Some bilirubin also goes to the kidney for elimination.
Bilirubin is responsible for the colour of both faeces and urine. If the liver has a problem with eliminating bilirubin, it can build up and show as a yellow colouration to skin and eyes (jaundice). Bilirubin is also responsible for the yellow around bruises — haem in damaged RBCs is being broken down to bilirubin by cells of the immune system, from the outside in.
Myoglobin is closely-relate to haemoglobin, and of the most relevance to this post. The main differences are that there is only one (not 4) HI units per myoglobin molecule, they bind oxygen more strongly because they are an oxygen store, and they don’t circulate in the blood and are found in skeletal muscle cells. Their purpose is to release stored oxygen if circulating RBCs cannot meet demand for oxygen, such as with prolonged muscle activity.
As a consequence, the amount of myoglobin in a muscle will depend on what that muscle is mostly used for. For example, a cow carries significant mass and its legs, in particular, are continually active, and it must support its trunk against gravity to stop it sagging. Thus cow muscle, and that of other grazing animals, tends to be high in myoglobin and its meat reddish in colour. Pigs are less active and their flesh is lower in myoglobin and less red (although pork is still classified as a red meat in various schemes). Chickens tend to walk rather than fly, hence their leg muscles are higher in myoglobin (and thus darker) than their breast muscles. Ducks fly long distances and their flesh is high in myoglobin and deeply red. Fish are supported buoyantly by water, and their muscle tends to be low in myoglobin and their flesh can be quite white (or grey). Exceptions are migratory fish such as tuna that swim for extended periods. Sea-going mammals (such as whales) have large myoglobin stores for a different reason — to supply oxygen during deep dives. The colour of salmon is due to diet, not myoglobin.
A more scientific explanation is that meat colour is determined by the proportions of type 1, type 2a and type 2b muscle fibres in different muscles. These fibre types have different roles — sustained or fast contraction, fatiguable or fatigue resistant, for example. The fibre types have different metabolisms and needs, and therefore differ in oxygen requirements and myoglobin content.
3. Why haemoglobin doesn’t matter for diet
Haemoglobin is important for live muscle and other cells, however, it is not important in meat. In a living animal, most of the HI is in the blood and perfused tissues. However, the HI we ingest when we eat meat from those animals comes from the myoglobin in muscle cells, not from haemoglobin, because there is virtually no blood left in meat after slaughter. It has been removed in a process known as exsanguination. The average residual blood/haemoglobin content of beef is negligible (~0.3%). The red juice that comes out of a lightly-cooked steak is myoglobin dissolved in water. There’s no blood in meat, red or otherwise.
4. The colours of meat
In living blood and muscle, the primary role of HI is to bind or release oxygen. As it does this, HI changes shape and reflects light differently. With bound oxygen, HI appears red, whereas if oxygen has been released it takes on a more purple hue. When oxygen is released, it is replaced by a molecule of water.
This also has an impact for meat derived from muscle. The cellular machinery can continue for a while after slaughter, sustaining itself with oxygen from myoglobin. If meat is then packaged for a time in an oxygen-free environment, such as vacuum-packed, it will not appear red because the HI in myoglobin is deoxygenated. The meat should return to a more red colour when the package is opened and the meat surface is exposed to air (and oxygen). If it is held in its deoxygenated state for an extended period, the iron atom in HI undergoes a transformation (from ferrous to ferric — see later) and the meat will take on a more brown colour. This is also what happens when meat is cooked — first HI becomes deoxygenated and then the iron transforms to the ferric form.
Oxygen is not the only molecule that can bind to HI and alter its colour. Carbon monoxide (CO) binds to HI much more strongly than oxygen itself. This is what makes breathing CO so deadly, even in the presence of oxygen — it displaces oxygen from HI in haemoglobin and effectively suffocates us. Nevertheless, we do have some CO in our bodies, for example as a byproduct from the breakdown of haemoglobin in the spleen. CO can be useful to us, and at low doses it acts as a cellular messenger and can lower blood pressure by vasodilation.
As it happens, CO also causes myoglobin to be perceived as red. Consequently, some manufacturers package meat with a little CO (less than 0.4%) to maintain the appearance of freshness. This is allowed in some countries (e.g. USA) because it reduces food wastage — packaged meat can lose its colour and appeal to a consumer well before there is any risk of microbial spoilage, in turn leading to avoidance and wastage. The practice is banned in other countries (e.g. Australia) for the same reason — meat can still look fresh and edible long after microbial spoilage has occurred.
Nitric oxide (NO) also binds to HI, and creates a pink tinge. We most often encounter this with meats cured with nitrites (nitrous oxide), which accounts for their usually pink colour. The NO is not released from the myoglobin by cooking, and cured meats retain their colour with heat. NO also acts as a vasodilator.
Finally, the sulphur atom from hydrogen sulphide (H2S, ‘rotten egg’) can bind to HI (producing sulfmyoglobin), which has a greenish colour, while the H2S contributes an off smell. Over time, and even at refrigeration temperatures, benign microbial activity on the surface of the meat can produce H2S from sulphur-containing amino acids in meat proteins.
Egg whites also contain H2S. With over-cooking the sulphur bonds with iron in the yolk, producing green-grey ferrous sulphide and discolouration of the egg.
Perhaps surprisingly, our bodies naturally produce low-levels of H2S, it is part of our physiology and has a protective role in multiple organ systems. There is even discussion of using it for therapeutic purposes.
5. Greening and sous-vide cooking — an aside
I am not certain, however sulfmyoglobin may be part of the greenish exudate that can sometimes appear with long-duration, moderate-temperature, sous-vide cooking. I have not seen that this is a health concern, although it has received little attention. It can be abolished with a post-cook high-temperature sear. Greening (and smell) also can be avoided by pre-searing the meat to kill surface microbes (and also add flavour). Alternatively, the meat can be vacuum-packed and then blanched in boiling water (if the plastic is boil-safe). This avoids the possibility of recontamination after searing and before packaging. However, although boiling (100C) will kill live bacteria, the temperature is not high enough to kill spores. Spores are only a potential problem with long cooks.
So far I have focussed on HI from animal sources. There are also non-haem-iron (nHI) sources, sometimes known as inorganic iron, in our diet.
Meat contains both HI and nHI (in beef in about equal proportions, for example). However, all plants contain only nHI. Further, nHI can be in two forms: ferrous and ferric. The gut can only absorb the ferrous form of nHI, however, the nHI in our diet is ferric.
Plants need iron for multiple reasons, one being to help in the synthesis of chlorophyll. Chlorophyll is haem-like in its structure, however instead of an iron core it has a magnesium core (haem-magnesium gives chlorophyll its green colour). Iron deficiencies in the soil reduce chlorophyll production and result in yellowing leaves. There are other causes though. Iron deficiency in soil is a world-wide problem.
Both HI and nHI are absorbed in the first third of the small intestine (duodenum), near where it attaches to the stomach. Here, stomach acid that empties with the food itself creates an environment suited to iron absorption. This means that iron absorption in the gut occurs as far away from the colon and rectum as is anatomically possible — how this could give rise to CRC is far from obvious.
1. HI absorption
The globulin proteins surrounding the HI in myoglobin are broken down by digestive enzymes in the stomach. The HI enters the small intestine where it is transported into the cells lining the gut wall (this is an active transport, not a passive diffusion). There, the iron is stripped off the haem and converted to the ferric form if necessary (oxygenated HI is in the ferrous form). The ferric iron is either stored in the gut cell in a large protein container (ferritin), or transported out of the cell into the blood through a portal (ferroportin) and wrapped up in another protein called transferrin to travel through the blood. Transferrin carries the iron to where it is useful, principally the liver (for storage) and the bone marrow (for making haemoglobin). At all times the iron needs to be bundled in proteins, because ’free’ iron is damaging to us.
The haeme remaining in the gut cells, now without its iron core, undergoes the same conversion to bilirubin that occurs in the spleen/liver. It is released into the circulation and transported to the liver, where it joins the bilirubin from the spleen and undergoes the same fate.
2. NHI absorption
The ferric nHI from plants (or animal sources) is also stored in ferritin, which is broken down during digestion. It then must be converted to ferrous iron by enzymes in the gut before it can be absorbed by cells on the gut wall. Ferrous iron is doubly charged, and these channels are not specific for ferrous iron — they open for other doubly-charged metals that may be good for us (e.g. manganese or magnesium) or not (e.g. lead and cadmium).
Once in the gut cell, the iron is converted back to its ferric form, and its subsequent utilisation is identical to the ferric iron from HI. From here on, the body does not know whether the iron came from haem or non-haem sources.
3. Co-nutrients and iron absorption
A number of dietary factors influence nHI absorption. Significantly, especially for meat, nHI absorption is increased in the presence of HI. Antioxidants, such as vitamin C, help with the conversion from ferric to ferrous and thus help nHI absorption. Many compounds in plants inhibit nHI absorption, including tannins (wine, coffee), and a host of other compounds, many of which are otherwise beneficial to us (e.g. calcium). Staple crops such as wheat, corn and rice contain high levels of phytates, which are anti-nutrients that bind to iron making it unavailable for absorption by the gut. This is thought to be a factor in iron deficiencies in some people eating a high-fibre diet. Spinach is high in nHI, but contains oxalates that inhibit its absorption.
In contrast, the absorption of HI is not greatly affected by plant co-nutrients, although absorption is increased in the presence of protein. HI seems to be our preferred source for dietary iron, presumably driven by the importance of carnivory in our evolutionary biology. In round numbers, about 25–30% of dietary HI is absorbed, compared to 5–15% for nHI from plant sources. This poses a risk for people following plantarian-style diets, unless they practice vigilance.
Given that iron absorption does not occur in the colon, it follows that the iron that is not absorbed and that continues on to the colon is more likely to be relevant to CRC. In which case, the poorly absorbed nHI should pose a greater threat, if there even is one.
We don’t have a dietary requirement for haem, we make it and manage it ourselves (although we do also absorb some if it’s in the diet, it is bound to another protein, hemopexin, for transport around the body).
However, we do have a dietary requirement for iron.
Iron is an essential nutrient, but what makes iron different is that we have no means of eliminating it from our bodies in a controlled way once it is inside us. Normally, we absorb as much of a nutrient as we can (e.g. glucose), then try to regulate it from within our bodies. With iron, we regulate how much we let in.
We keep iron levels steady in the body by recycling iron that we already have, maintaining an iron reserve in the liver to compensate for transient deficiencies, and, importantly, by controlling the amount we absorb through the intestine irrespective of the amount consumed in the diet.
We need iron in the diet because we can lose some iron incidentally (for example by shedding skin, gut and hair cells) or sporadically (for example by accidental bleeding, menstruation or blood donation). We carry around 3–4 g of iron in our body, and lose only about 1–2 milligrams (mg) in an average day (i.e. incidental loss). This means that we manage to recycle, or keep in store, over 99.9% of our iron. Recommended daily intake of iron is about 10 times the daily incidental loss (depending on gender, age etc) to provide a safety-factor, and to allow for poor iron absorption with high-plant low-animal diets, such as that promoted by current dietary guidelines.
The liver is the master regulator. Iron transported to it by transferrin is extracted and stored in the liver as ferritin (just like in gut cells or plant cells). The liver senses the level of iron in the body and regulates iron absorption in the gut by releasing (or withholding) a small protein (peptide) called hepcidin. If iron levels are high, the liver releases hepcidin that blocks absorption (by blocking the ferroportin channel in gut cells). If iron levels are low, the liver releases iron from its stores to make up for the immediate shortfall, and reduces hepcidin thus enabling greater iron absorption by the gut. This is the sole mechanism available to manage body iron — the gut will take precisely what iron the body needs, and any surplus, whether from HI, nHI, natural or supplemented, will travel through the intestines and be eliminated.
Iron deficiency is the most common dietary deficiency in the developing world. Iron deficiency is said to affect up to 2 billion people, is a particular concern for children, and can cause many other health problems. However, even in the US, the prevalence of iron-deficiency anaemia is around 6% and increasing (it has doubled from 2003–4 to 2011–12), and the presence of iron deficiency (without manifest anaemia) is at least double that.
The problem is likely to be dietary advice to reduce red meat in the diet and replace its calories with starchy carbohydrates. An obvious approach might be to rethink that advice. However, this has not happened. Instead, health authorities have resorted to fortifying refined carbohydrates with iron and other trace nutrients.
In the US, refined grain products are usually enriched, and to be called enriched they are required by law to be fortiﬁed with the B-group vitamins thiamin (vitamin B1), riboﬂavin (B2), niacin (B3), folate (B9), as well as iron and zinc. Wholegrains (whatever that means) have no such requirement, and despite their good press, they cannot provide the necessary levels for these nutrients. For example, a serving of cooked brown rice might only contain ~7% of the recommended daily requirement for thiamin, whereas a serving of enriched breakfast cereal may contain 100%.
This is why authorities do not urge people to consume wholegrains to the exclusion of refined grains — they know that this could result in nutrient deficiencies. Instead, they use a disingenuous approach — without admitting that their diet is deficient, the guidelines recommend that ‘at least half of your grains should be wholegrain’. Half, not all.
Iron supplements can be helpful for certain individuals. However, the presence and cause of iron deficiency should be established, bearing in mind that supplementation will be ineffective unless the gut is directed by the liver to absorb the iron — iron absorption is not a passive process.
The most recent trend is biofortification — grains genetically modified to be higher in iron (in the form of ferritin). As grains also naturally contain large amounts of the anti-nutrient phytate, there are efforts to engineer reduced phytate levels in staple crops such as wheat, maize, rice and barley. Also, supplements are being developed that have high bio-availability even when consumed with these inhibitory crops.
Much of this is relevant to iron deficiency in developing countries, particularly those that depend on crops for food. However, it seems absurd for most people in developed countries with access to more varied food choices, and it would make more sense to provide varied food choices in developing countries. Current guidelines promote a high-grain diet, low in bio-available iron. Authorities do not recommend animal sources for readily absorbed iron. Instead, they rely on artificial fortification of nutrient-poor grains and cereals (either by enrichment or genetic engineering) to achieve a nutrient balance.
Then there are our gut microbes. This is an emerging field, however iron is as relevant to bacteria as it is to us. Current thinking seems to be that excess iron in the gut may promote pathogenic bacteria over beneficial bacteria. Many beneficial bacteria utilise manganese over iron, whereas many pathogenic bacteria utilise, and depend on, iron. It is for this reason that the liver reduces iron levels in our body when we are suffering from an infection — to stress the pathogen. Further, iron (or the bacterial byproducts of utilising it) may be pro-inflammatory in excess. In relation to CRC (a disease of the colon), the bulk of our gut microbes reside in the colon and iron that is poorly absorbed by the small intestine is more likely to reach the colon. Arguably, this might be from nHI, which also might be more bio-available to microbes because it is not bundled with haem. This is a massive topic and I’m not going there. However, our gut microbiota adds another level of complexity to the story of iron and CRC.
What does any of this mean for HI and CRC? Nothing that I can deduce — I don’t see anything that stands out. HI seems to be an explanation of convenience for a hypothesis (i.e. red meat increases the risk of CRC) that is yet to be proven.
I don’t deny that there are ways that haem can be deleterious (mostly through inducing oxidative stress). However, haem also has many beneficial roles (other than oxygen transport). As well, iron in excess can be deleterious too (whether it comes from HI or nHI). As can oxygen (free radical formation, requiring antioxidants). As can glucose (chronic-elevation produces the symptoms of type 2 diabetes). And so on.
And, a system so complex as haem and iron regulation can go wrong. However, guidelines to reduce red meat consumption are directed at the general public, most of whom will have healthy regulatory systems. Indeed, healthy and health-conscious individuals may be the group most likely to listen to this advice.
There are other suggestions. For example, there is an obscure sugar that can be found in mammal meat that we have lost the ability to metabolise. Again, that sounds like an explanation of convenience, and perhaps a sign of desperation.
It might be argued that we don’t need an explanation if the science could show, definitively, that eating red meat (or mammal meat) does cause CRC. However, this has not been shown to be true. A discussion of that topic will need to wait for another post.