Detection, monitoring, and deleterious health effects of lipid oxidation

By Martin Grootveld, Victor Ruiz Rodado, and Christopher J.L. Silwood

In This Section

November/December 2014

  • Trans fatty acids have received considerable attention in the media and the scientific literature (together with their putative atherogenic and carcinogenic actions). The far more toxic aldehydes generated in thermally stressed culinary oils, however, have received little or none.
  • The most important chemical reaction involved is the self-sustaining, free radical-mediated oxidative deterioration of polyunsaturated fatty acids (PUFAs), which occurs during the heating of culinary oils (Fig. 1) and, to a much diminished level, saturated fatty acid (SFA)-rich fats. This oxidative degradation process can generate extremely toxic conjugated lipid hydroperoxydienes (CHPDs). These are unstable at standard frying temperatures (ca. 180°C) and are degraded to a broad range of secondary products, particularly saturated and unsaturated aldehydes, together with di- and epoxyaldehydes.
  • Such aldehydic fragments also have toxicological properties in humans owing to their high reactivity with critical biomolecules in vivo (proteins such as low-density lipoprotein, amino acids, thiols such as glutathione, DNA, etc.). Despite their reactivities, high levels of CHPDs can remain in PUFA-rich oils which have been subjected to routine frying practices.


 Figure 1

1H, two-dimensional 1H-1H, and 1H-13C NMR investigations of thermally stressed PUFA-containing culinary oils have found high levels of α,β-unsaturated aldehydes (including trans-2-alkenals, and cis,trans- and trans,trans-alka-2,4-dienals, the latter including the mutagen trans,trans-2,4-decadienal), and n-alkanals, together with their CHPD and hydroxydiene precursors (Fig. 2) (1–3). Indeed, samples of repeatedly used oils collected from fast-food retail outlets and restaurants have confirmed the production of aldehydic lipid oxidation products (LOPs) at levels exceeding 10-2 moles per kilogram (mol·kg-1) during “on-site” frying episodes.

Figure 2

As expected, the levels of total aldehydes generated increase proportionately with oil PUFA content, and over half are the more highly cytotoxic α,β-unsaturated classes (Fig. 3a; Table 1), which include acrolein and 4-hydroxy-trans-2-nonenal (HNE), as well as 4-hydroperoxy-, 4-hydroxy-, and 4,5-epoxy-trans-2-alkenals. Total α,β-unsaturated aldehyde concentrations in culinary oils (heated at 180°C for 30–90 minutes or longer) are often higher than 20 mmol·kg-1 and can sometimes approach 50 mmol·kg-1. Furthermore, relatively low concentrations of 1H NMR-detectable aldehydes and their CHPD precursors are even found in newly purchased unheated culinary oils. Key resonances for CHPD isomers are also readily detectable within the 5.5–6.7 ppm (parts per million) regions of high-resolution 1H NMR spectra. We can also employ these NMR techniques to monitor the corresponding  degradation of culinary oil PUFAs and monounsaturated fatty acids (MUFAs) during heating/frying (Fig. 2b).

2014 Nov/Dec figure 3a


Length of Thermal Stressing Episode trans-2-Alkenals ( trans,trans-Alka-2,4-Dienals (  cis,trans-Alka-2,4-Dienals
4-Hydroxy- trans-2-Alkenals
Further Unassigned Aldehydes (total:
Unheated (0 min. nd  nd  nd  nd  nd  nd 
30 min. 3.90 6.42 1.12 nd 2.90 0.05
60 min. 7.16


1.79 nd 5.22 0.03
90 min.  10.56 13.77 2.71 0.87 6.62 0.03 
Table. Aldehyde levels detectable in a sample of sunflower oil subjected to thermal stressing at a standard frying temperature of 180oC in the presence of atmospheric O2 (0-90 minutes). Typical results are shown. Abbreviations: nd, none detectable.     

 figure 3b

Portable bench-top NMR instruments clearly have potential for monitoring PUFA peroxidation and LOP generation in cooking oils at industrial sites or in restaurants.

Acrylamide (which can exert toxic effects on the nervous system and fertility, and may also be carcinogenic) can also arise from an acrolein source when asparagine-rich foods are deep-fried in PUFA-rich oils (4). The levels of acrylamide generated in foods during high-temperature cooking/frying processes are substantially lower than those recorded for aldehydes formed in PUFA-rich culinary oils during frying episodes (to date, the very highest reported levels are only ca. 4 ppm, equivalent to 56 µmol·kg-1). Should we not also have major concerns about the much higher concentrations of cytotoxic LOPs available for consumption by humans (see sidebar on toxicological actions)?

The concentrations of aldehydes generated in culinary oils during episodes of heating at 180°C represent only what remains in the oil: Owing to their low boiling points, many of the aldehydes generated are volatilized at standard frying temperatures. These represent inhalation health hazards, in view of their inhalation by humans, especially workers in inadequately ventilated fast-food retail outlets.

Figure 3b shows a colorimetric (TBARS test) illustration of aldehyde production in a sample of corn oil subjected to heating for 60 minutes.


In vivo absorption of dietary LOPs

Except for direct damage to the gastrointestinal epithelium, the toxicological actions exerted by LOPs depend on their rate and extent of absorption from the gut into the systemic circulation where they may cause damage to essential organs, tissues, and cells. Experiments in rats have demonstrated that trans-2-alkenals, which are generated in PUFA-containing culinary oils during thermal stressing episodes, are absorbed (5). Following absorption, these cytotoxic agents are metabolized by a process involving the primary addition (Michael addition reaction) of glutathione across their electrophilic carbon–carbon double bonds and finally excreted in the urine as C-3 mercapturate derivatives.


Estimating human dietary intake of LOPs

Estimates of dietary LOP intake should be ideally (but not exclusively) targeted at foods subjected to high-temperature frying/cooking. Indeed, the composition and content of hazardous LOPs available in fried foods depend on the identity of the frying/cooking oil and its PUFA content, the frying conditions employed, the length of the frying process, exposure of the frying medium to atmospheric O2, the reactivities of these agents with a range of other biomolecules (e.g., amino acids and proteins), and, to a limited extent, the antioxidant content of the frying matrix. Our experiments have shown that shallow frying gives rise to much higher levels of LOPs than deep frying under the same conditions (reflecting the influence of the surface area of the frying medium, its exposure to atmospheric O2, and the subsequent dilution of LOPs generated into the bulk medium).

Although the lipid (predominantly triacylglycerol, or TAG) content of deep-fried French fries is dependent on the nature of the frying process, its length, and temperature, these values generally range from 7% to 35% (w/w), of which SFAs are 28–46%, MUFAs 25–43%, and PUFAs 2–45%. On this basis, a 300-g serving of French fries (chips) with a fat content of 11.5% (w/w) contains 35 g of LOP-containing lipids per serving. If we estimate that the total level of aldehydic LOPs present in a typical thermally stressed PUFA-rich culinary oil is 10 millimoles per kilogram (mmol·kg-1) and assume they are not significantly consumed by reaction with food biomolecules, the aldehyde content of this serving of French fries will be ca. 0.35 mmol, of which over half are the extremely toxic α,β-unsaturated ones (Fig. 3a).

The New Zealand Heart Foundation estimated that a meal consisting of two pieces of battered fish plus an average serving of fries represents 95 g of fat. With an assumption of only 10 mmol·kg-1 total aldehydes in thermally stressed culinary oils used to fry such foods, there would be 0.95 mmol of such LOPs in an average fried meal—a not insignificant figure in view of the potent toxicological properties of such agents.


Acrolein represents the simplest and perhaps one of the more toxic α,β-unsaturated aldehydes generated from the lipid peroxidation process, and the WHO’s ‘tolerable’ intake level of this agent is only 7.5 μg (0.13 μmol) per day per kg of body weight, and therefore this would be equivalent to 525 μg (9.4 μmol) per day for an assumed (average) human body weight of 70 kg, a value which is substantially less than that estimated above for total aldehydes in typical fried food portions. If all of the above estimated 0.35 and 0.95 mmol of total food portion-associated aldehyde was acrolein, it would be equivalent to a content of 430 ppm in the food matrix to be consumed.  


Even if almost 50% or so of these aldehydes were the less chemically-reactive and toxic saturated ones, this is still very worrying. We should expect at least some reaction of these aldehydes with food biomolecules (amino acids, etc.) during the frying process (and presumably much less subsequently at ambient or lower temperatures). Nevertheless, culinary oil aldehyde and further LOP concentrations—which we monitor by high-resolution NMR analysis—are only those that remain in the oil matrix after frying/heating episodes.


Furthermore, with regard to the risk of inhalation of these aldehydes volatilised during frying practices by humans, we should consider the maximum US Occupational Safety and Health (OSHA) permissible exposure limit (PEL) for acrolein, which is an (atmospheric) level of 0.1 ppm (equivalent to only 1.8 µmol·kg-1 in our fried food model) for a time-weighted long-term (8 hour) exposure, and 0.3 ppm (5.4 µmol·kg-1)for a short-term (15 minute) one. This 15-minute exposure time can be considered to be less than the time taken to consume a typical fried meal!


Hence, we believe that many researchers and food scientists involved in this area should re-think their recommendations of the “health-promoting” properties of linoleoyl- and linolenoylglycerols in shallow and deep frying processes.

Acrolein is just one of the α,β-unsaturated aldehydes generated in thermally stressed PUFA-rich oils: Many others generated in this manner have comparable toxicological properties The foregoing considerations exclude possible toxicological properties of their isomeric CHPD precursors (also present in the high millimolar range in thermally stressed oils) in a typical fried food meal. Indeed, in one early investigation, a single intravenous dose of methyl linoleate hydroperoxide (20 mg·kg-1) administered to rats gave rise to a high mortality within 24 hours (animals dying from lung damage), although a higher dose given orally was without effect. This observation may reflect the limited in vivo absorption of these particular aldehyde precursors, in contrast to the known absorption of aldehydes.


Toxicological and pathogenic properties of dietary LOPS

Potential influence of dietary LOPS on metabolic pathways. As a consequence of theirabsorption from the gut into the systemic circulation, LOPs may penetrate cellular membranes, allowing their entry into particular intracellular sites/organelles where many critical metabolic processes occur. Literature evidence indicates that feeding thermally stressed or repeatedly used culinary oils to experimental animals induces significant modifications to key liver microsomal pathways and to the mitochondrial respiratory chain, for example. These effects are likely to occur via reactions of LOPs with key enzymes (and more especially their active sites), for example, the oxidation of active methioninyl and cysteinyl residues by CHPDs, or alteration of critical side-chain amino acid amine or thiol groups with aldehydes via Schiff base or Michael addition reactions.

Atherosclerosis. Investigations have revealed that dietary derived LOPs can accelerate all three stages of the development of atherosclerosis (i.e., endothelial injury, accumulation of plaque, and thrombosis). Animal studies have shown that diets containing thermally stressed, PUFA-laden (and hence LOP-rich) oils exhibit a greater atherogenicity than those containing unheated ones (6). Because cytotoxic aldehydes can be absorbed, they have the capacity to attack and structurally alter the apolipoprotein B component of low density lipoproteins (LDLs) (5). This mechanism can engender uptake of lipid-loaded LDLs by macrophages, which, in turn, transforms them to foam cells, the accumulation of which is responsible for the development of aortic fatty streaks, a hallmark of the aetiology of atherosclerosis and its pathological sequelae. More recently, our co-investigators found that aldehydic LOPs elevated the expression of the CD36 scavenger receptor of macrophages, a phenomenon that also promotes this process (7).

Mutagenic and carcinogenic properties. Since they are powerful electrophilic alkylating agents, α,β-unsaturated aldehydes can covalently modify DNA base units via a mechanistically complex process that may involve their prior epoxidation in vivo.Such chemically altered bases may therefore be of mutagenic potential. Additionally, these LOPs can inactivate DNA replicating systems, a process that can, at least in principle, elevate the extent of DNA damage. Hence, following cellular uptake, such aldehydes have the potential to cause both DNA and chromosomal damage.

An Intelligence Bulletin published by the US Center for Disease Control and Prevention’s National Institute for Occupational Safety and Health (NIOSH, 1992) concluded that malondialdehyde (MDA) and other aldehydes arising from lipid peroxidation (especially acrolein) present a serious carcinogenic hazard. Indeed, adenomas and carcinomas of the thyroid gland, together with adenomas of the pancreatic islet cells, were induced in rats by MDA in a prolonged gavage study; nasal and laryngeal cancers arose in rats and hamsters, respectively, during long-term acetaldehyde inhalation experiments. Hence, both these aldehydes satisfied the NIOSH criteria for classification as carcinogens, and therefore it has set exacting limits for their occupational exposure.

We have previously shown that MDA is also generated by thermally stressing culinary oils, although at concentrations much lower than those of the more reactive α,β-unsaturated aldehydes. Paradoxically, in view of its somewhat exaggerated importance as an aldehydic LOP, much of the occupational exposure to MDA appears to arise from its use in research laboratories investigating lipid peroxidation processes.

Volatile emissions from heated culinary oils used in Chinese-style cooking are indeed mutagenic; exposure to such indoor air pollution may render humans more susceptible to contracting lung or further cancers, together with rhinitis and diminished lung function (8–10). The high temperatures used in standard (especially Chinese) frying result in fumes that are rich in volatile LOPs, including acrolein.

An entire book (Chadwick, D. J., J. Goode (Novartis Foundation), Acetaldehyde-related pathology: Bridging the trans-disciplinary divide, ISBN 9780470057667) together with numerous governmental health and safety documents, has been dedicated to the pathology (most especially carcinogenicity) of the n-alkanal acetaldehyde (with a corresponding OSHA PEL of 200 ppm), which is much less hazardous than equivalent levels of α,β-unsaturated aldehydes such as acrolein (OSHA PEL 0.1 ppm).

Further investigations have focused on the possible association of prostate cancer with the consumption of fried foods (11) and the clastogenic potential of thermally stressed culinary oils (12). It is also conceivable that congenital malformations can arise from the reactions of DNA base adducts with aldehydes. Alternatively, DNA-polymerase and -repair enzymes can be covalently modified by these reactive aldehydes, processes giving rise to modifications in the rate and extent of DNA replication and repair, and hence chromosomal damage.

Teratogenic actions. In principle, if aldehydic LOPs induce DNA and chromosomal damage during embryo development, fetal malformations may arise. In 1996, Viana et al. (13) investigated the ability of the chain-breaking antioxidant α-tocopherol (α-TOH, vitamin E) to prevent the teratogenic effects of uncontrolled diabetes mellitus in rats (a study based on the hypothesis that diabetic animals have an elevated level of oxidative stress and therefore in vivo lipid peroxidation when expressed relative to that of healthy controls). They found that a PUFA-rich culinary oil (which served as a vehicle for oral administration of α-TOH) increased the rate of malformations and reabsorptions in both normal and diabetic pregnancies. Further investigations (14) revealed that subjection of this safflower oil vehicle to thermal stressing episodes (according to standard frying practices for a period of 20 minutes) markedly enhanced its teratogenic effects. That is, the evidence indicates that the LOPs therein are primarily responsible for these actions.

Further adverse health effects of dietary LOPs. Further documented health effects of LOPs include their pro-inflammatory and gastropathic properties (for the latter, oral administration of HNE to rats at a dose level of only 0.26 μmol·dm-3, a level similar to that of healthy human blood plasma, induced peptic ulcers), and also a significant elevation in systolic blood pressure and an impaired vasorelaxation observed in rats fed pre-heated soy oil (15). 


Potential solutions and future perspectives

The most obvious solution to the generation of LOPs in culinary oils during frying is to avoid consuming foods fried in PUFA-rich oils as much as possible. Indeed, consumers, together with those involved in the fast-food sector, could employ culinary oils of only a low PUFA content, or MUFA- (or even SFA)-rich alternatives such as olive, selected canola, palm, or coconut oils for frying (MUFAs such as oleoylglycerol adducts are much more resistant to peroxidative degradation than are PUFAs (1), and hence markedly lower levels of only selected classes of aldehydes are generated during frying).

One further potential solution to this problem is supplementation or further supplementation of culinary oils with antioxidants such as α-TOH. Note that concentrations of this antioxidant naturally present in cooking oils (for example, approximately 2 mmol·kg-1 in corn oil, but only ca. 0.3-0.4 and 0.2 in palm and soybean oils resepectively) appear to be ineffective in preventing the thermally induced generation of high levels of CHPDs and aldehydes in these products (1), and our results have shown that additional supplementation with such agents also fails to offer protection.

Previous studies that investigated the prospective health effects or benefits of dietary PUFAs (i.e., those involving feeding trials with humans or animals or, alternatively, related epidemiological ones) should be scrutinized. With hindsight, it seems to us that many of these experimental investigations were flawed since, in addition to some major design faults, they failed to take into account or even consider the nature and concentrations of any cytotoxic LOPs present in the oils or diets involved. Similarly, corresponding epidemiological (or meta-analysis-based) investigations incorporated only the (estimated) total dietary intake of selected PUFAs and further fatty acids, and ignored any LOPs derived or derivable from frying/cooking. Even if PUFA containing culinary oils are unheated, it is virtually impossible to rule out the presence of traces of LOPs within them (analysis of apparently pure PUFAs or their corresponding TAGs obtained from reputable commercial sources has revealed that these materials contain traces of CHPDs and/or aldehydes that are readily detectable by NMR).

Our original research investigations performed in the early 1990s in this key food toxicology/health research area have been repeated, replicated, and exemplified in many research group laboratories worldwide (e.g., 16) and have been available to the scientific community since then (1,2), Until just recently, this major problem has received scant or limited attention from the food industry and health researchers. Future clinical trial or epidemiological investigations aimed at determining relationships between the incidence of selected human diseases and dietary LOP consumption may serve to clarify the nature of such associations. We agree that completely pure, authentic, and essential PUFAs offer no threats to human health, but we point out that LOPs arising from the frequent and common use of PUFA containing frying/cooking media (or those produced during prolonged storage episodes) certainly do so.


Martin Grootveld is a Professor of Bioanalytical Chemistry and Chemical Pathology atLeicester School of Pharmacy, De Montfort University,in Leicester, United Kingdom. He can be contacted at

Victor Ruiz Rodado is a postgraduate doctoral training student also at Leicester School of Pharmacy, and Christopher J. L. Silwood is a scientific consultant with major research interests in the bioanlytical chemistry, and lipid analysis/oxidation areas.

We are very grateful to Andrew Claxson (formerly of St. Bartholomews and the Royal London School of Medicine and Dentistry, London, UK) for excellent technical and scientific assistance, Nigel Crossley of Oxford Instruments for the acquisition of spectra on the rare earth magnet bench-top NMR instrument, and to Nina Teicholz, author of The Big Fat Surprise: Why Butter, Meat & Cheese Belong in a Healthy Diet, whose research was helpful to us in writing this article.


Glossary     Toxicological actions of aldehydes
(and other lipid oxidation products)
  • Atherogenicity
  • Mutagenicity and carcinogenicity
  • Congenital malformations (causal)
  • Gastropathic actions
  • Pro-inflammatory effects
  • Teratogenicity
  • Hypertensivity
CHPD   conjugated lipid hydroperoxydienes
HNE    4-hydroxy-trans-2-nonenal
LDL   low density lipoprotein
LOP   lipid oxidation product
MDA   malondialdehyde
MUFA   monounsaturated fatty acid
NIOSH   US National Institute for Occupational Safety and Health
NMR   nuclear magnetic resonance
OSHA   US Occupational Safety and Health Administration
PEL   permissible exposure level
PUFA   polyunsaturated fatty acid
SFA   saturated fatty acid
TBARS   thiobarbituric acid reactive-substances  
TOH   α-tocopherol  



  1. Claxson, A.W.D., G.E. Hawkes, D.P. Richardson, D.P. Naughton, R.M. Haywood, et al., Generation of lipid peroxidation products in culinary oils and fats during episodes of thermal stressing: a high field 1H NMR study,FEBS Lett. 355:81–90 (1994).
  2. Haywood, R.M., A.W.D. Claxson, G.E. Hawkes, D.P. Richardson, D.P. Naughton, et al., Detection of aldehydes and their conjugated hydroperoxydiene precursors in thermally-stressed culinary oils and fats: investigations using high resolution proton NMR spectroscopy, Free Rad. Res. 22:441–482 (1995).
  3. Silwood, C.J.L., and M. Grootveld, Application of high-resolution two-dimensional 1H and 13C nuclear magnetic resonance techniques to the characterization of lipid oxidation products in autoxidized linoleoyl/linolenoyglycerols, Lipids 34:741–756 (1999).
  4. Ehling, S., M. Hengel, and T. Shibamoto, Formation of acrylamide from lipids, Adv, Exp. Med. Biol. 561:223–233 (2005).
  5. Grootveld, M., M.D. Atherton, A.N. Sheerin, J. Hawkes, D.R. Blake, et al., In vivo absorption, metabolism, and urinary excretion of α,β-unsaturated aldehydes in experimental animals. Relevance to the development of cardiovascular diseases by the dietary ingestion of thermally stressed polyunsaturated-rich culinary oils, J. Clin. Invest. 101:1210–1218 (1998).
  6. Staprãns, I., J.H.Rapp, X.-M. Pan, D.A. Hardman, and K.R. Feingold, Oxidized lipids in the diet accelerate the development of fatty streaks in cholesterol-fed rabbits, Arterioscler. Thromb. Vasc. Biol. 16:533–538 (1996).
  7. Viana, M., L. Villacorta, B. Bonet, A. Indart, A. Muntennu, et al., Effect of aldehydes on CD36 expression, Free Rad. Res. 39:973–977 (2005).
  8. Wu, S.C., and G.C. Yen, Effects of cooking oil fumes on the genotoxicity and oxidative stress in human lung carcinoma (A-549) cells, Toxicol. In Vitro 18:571–580 (2004).
  9. Young, S.C., L.W. Chang, H.L. Lee, L.H. Tsai, Y.C. Liu, et al., DNA damages induced by trans,trans-2,4-decadienal (tt-DD), a component of cooking oil fume, in human bronchial epithelial cells, Environ. Mol. Mutagen. 51:315–321 (2010).
  10. Dung, C.H., S.C. Wu, and G.C. Yen, Genotoxicity and oxidative stress of the mutagenic compounds formed in fumes of heated soybean oil, sunflower oil and lard, Toxicol. In Vitro 20:439–447 (2006).
  11. Stott-Miller, M., M.L. Neuhouser, and J.L. Stanford, Consumption of deep-fried foods and risk of prostate cancer, Prostate 73:960–969 (2013).
  12. Indart, A., M. Viana, S. Clapés, L. Izquierdo, and B. Bonet, Clastogenic and cytotoxic effects of lipid peroxidation products generated in culinary oils submitted to thermal stress, Food Chem. Toxicol. 45:1963–1967 (2007).
  13. Viana, M., E. Herrera, and B. Bonet, Teratogenic effects of diabetes mellitus in the rat. Prevention by vitamin E, Diabetologia 39:1041–1046 (1996).
  14. Indart, A., M. Viana, M.C. Grootveld, C.J.L. Silwood, I. Sanchez-Vera, et al., Teratogenic actions of thermally-stressed culinary oils in rats, Free Rad. Res. 36:1051–1058 (2002).
  15. Leong, X.-F., M.R. Mustafa, S. Das, and K. Jaarin, Association of elevated blood pressure and impaired vasorelaxation in experimental Sprague-Dawley rats fed with heated vegetable oil, Lipids Health Dis. 9:66 (2010).
  16. Guillén, M.D., and E. Goicoechea, Toxic oxygenated α,β-unsaturated aldehydes and their study in foods: a review, Crit. Rev. Food Sci. Nutr. 48:119–136 (2008).