Cholesterol degradation in archaeological pottery mediated by fired clay and fatty acid pro-oxidants

Article abstract of DOI:10.1016/j.tetlet.2018.10.071

Cholesterol is generally absent in animal fat residues preserved in archaeological ceramic vessels. It is known from edible oil refining that during bleaching with activated clay sterols are degraded, largely via oxidation. Laboratory heating experiments

Full text of DOI:10.1016/j.tetlet.2018.10.071

Accepted Manuscript
Cholesterol degradation in archaeological pottery mediated by fired clay and
fatty acid pro-oxidants
Simon Hammann, Mathilda Whittle, Lucy J.E. Cramp, Richard P. Evershed
PII:
S0040-4039(18)31312-1
https://doi.org/10.1016/j.tetlet.2018.10.071
TETL 50381
DOI:
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
17 August 2018
26 October 2018
30 October 2018
Please cite this article as: Hammann, S., Whittle, M., Cramp, L.J.E., Evershed, R.P., Cholesterol degradation in
archaeological pottery mediated by fired clay and fatty acid pro-oxidants, Tetrahedron Letters (2018), doi: https://
doi.org/10.1016/j.tetlet.2018.10.071
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Cholesterol degradation in archaeological pottery mediated by fired
clay and fatty acid pro-oxidants
Simon Hammann1,2, Mathilda Whittle2, Lucy J. E. Cramp1, Richard P. Evershed2
1 Department of Anthropology and Archaeology
University of Bristol
43 Woodland Road
Bristol BS81UU
United Kingdom
2 Organic Geochemistry Unit
School of Chemistry
University of Bristol
Cantock’s Close
United Kingdom
Bristol BS81TS
Abstract:
Cholesterol is generally absent in animal fat residues preserved in archaeological
ceramic vessels. It is known from edible oil refining that during bleaching with
activated clay sterols are degraded, largely via oxidation. Laboratory heating
experiments using fired clay from replica pottery vessels promoted rapid degradation
of cholesterol via oxidation. Furthermore, heating cholesterol with fatty acids
(saturated and unsaturated) revealed additional degradation to occur independently
of the ceramic matrix. As both conditions are met in archaeological pottery during
animal (and plant) product processing involving heating, the very rare detection of
sterols in organic residues can be explained.
1

Keywords: cholesterol; fatty acids; archaeology; degradation; clay; organic residue
Unglazed ceramic pots found commonly at archaeological sites are carriers of
valuable information about past human diet and food procurement. This information
resides in the chemical residues of food constituents, mainly lipids, that become
absorbed in the porous ceramic matrix during food processing.1, 2 Once absorbed,
these lipids are relatively protected against ground water leaching and microbial
degradation, and owing to their hydrophobicity, can survive over many millennia.2
The lipids recovered from archaeological pottery rarely retain the composition
originally found in the commodities from which they derive. The changes observed
can be separated into pre-burial, use-related modifications and post-burial (mainly
microbial) changes. In particular, high heat (>270 ºC) during vessel use has been
shown to lead to formation of a range of diagnostic degradation products, e.g. long-
chain ketones from the condensation of fatty acids, or vicinal dihydroxy fatty acids
and ω-(o-alkylphenyl)-alkanoic acids formed from monounsaturated and
polyunsaturated fatty acids, respectively.2-6 Pre- and post-burial changes are
apparent from the hydrolysis of acyl lipids (such as triacylglycerols) to free fatty acids
and di- and monoacylglycerols, as well as the preferential loss of polar compounds
due to water leaching.7, 8
Sterols, such as cholesterol in animals and ß-sitosterol in plants, are essential
membrane compounds and therefore present in any living eukaryotic organism and
consequentially virtually in every food commodity. Yet, gas chromatography-mass
spectrometry (GC-MS) analysis of several thousand lipid extracts has shown sterols
are almost entirely absent in archaeology pottery. Indeed, so rare are sterols that
where they are detected, modern contamination is suspected to be the most likely
source, e.g. cholesterol from fingerprints.9, 10 Since we have recently shown that
sterols are reasonably resistant to post-burial degradation,11 this makes degradation
pre-burial, for example during cooking, the most likely explanation for the absence of
sterols from organic residues.
2

Commercial edible oil refining, particularly the bleaching process, is known to lead to
the degradation of sterols.12-14 Interestingly, acid-activated clay suspensions are the
most common solid phases used in oil bleaching.12, 15 Furthermore, the heating of
sterols with fatty acids and triacylglycerols has been found to promote lipid
degradation.16 Active clay surfaces and abundant acyl lipids would have been
ubiquitous in archaeological ceramic cooking pots, and cooking processes would
have involved applying substantial heat. Therefore, we rationalized that this might
cause the degradation of sterols and explain their general absence in archaeological
food residues in pottery. To test this hypothesis, laboratory heating experiments
were performed using known lipids spiked on powdered clay ceramics. The progress
of the experiments was monitored by GC analysis of trimethylsilylated solvent
extracts of the ceramics, with quantification using internal standards.
In initial experiments cholesterol (212 µg) and oleic acid (58 µg) were applied to 0.5
g of powdered clay ceramic and heated (100 ºC, 14 h) in open glass vials. GC
analysis showed that 73(±4)% of the cholesterol was degraded by this treatment
(Experiment 1 in Table 1, Fig. 1a).
3

Figure 1. GC-FID chromatograms of cholesterol and cholesterol oxidation products extracted from
experiments 1 (a), 4 (b) and 3 (c). The internal standards (ISTD1 and ISTD2) are at the same
concentrations in all three chromatograms. The magnified section of chromatogram in panel a) shows
the cholesterol oxidation products eluting between 9 and 12 min (identified by GC-MS).
The same pattern of degradation products and loss was observed when only
cholesterol was spiked on clay. However, heating cholesterol without clay resulted in
almost the complete recovery of cholesterol (Experiment 4 in Table 1, Fig. 1b). The
addition of oleic acid and heating without clay led to minimal degradation of
cholesterol and the formation of only low abundances of degradation products
(Experiment 3 in Table 1, Fig. 1c). Previous studies suggested that cholesterol is
stable at temperatures of 100 ºC, but the addition of clay clearly led to cholesterol
degradation even at this relatively low temperature.17, 18 Cholesterol degradation
followed approximately first-order kinetics for the first 24 h (R(Chol)=184.6e-0.071t, Fig.
2).
Figure 2. Cholesterol (blue) and cholesterol oxidation products (orange) recovered from clay after
heating 212 µg cholesterol and 58 µg oleic acid at 100 ºC for 1, 2, 4, 6, 14, and 24 h (experiments 5-
10 in Table 1).
Interestingly, GC-amenable oxidation products did not accumulate to account for
cholesterol loss, such that after 24 h almost 180 µg of cholesterol was degraded, but
4

only ca. 30 µg of cholesterol degradation products could be accounted for. Most
likely, the oxidation products underwent further reactions and were subsequently lost
as volatile products from the reaction vessel or were polymers that had too high
molecular weight to be analysed by the GC method used in this study.19
The main degradation products (identified by GC/Q-TOF MS) were cholesterol-
dehydration products (cholestadiene and cholestatriene isomers), as well as the
oxidation products cholest-4-en-3-one and 7-keto-cholesterol. Although almost 30
degradation products of cholesterol were detected (see ESI Table S1), the
distribution did not correspond to the usually observed thermooxidation products or
degradation products of cholesterol expected and reported in bleaching earth, i.e.
steradienes, 7-α/β-hydroxycholesterol, 5-α/β-epoxycholesterol and cholestanetriol
alongside 7-keto-cholesterol.14, 17, 20, 21 While 7ß-hydroxycholesterol and 7-keto-
cholesterol were detected they were not the main degradation products, which
indicates different degradation pathways compared to those observed in edible oil
refining. These differences could be due to different treatments of the clay in both
processes, i.e. acid-activation for refining of oil and high-temperature firing for the
production of pottery, or different availability of oxygen.
Table 1. Cholesterol recovery, heating time and initial composition of the
experimental mixtures. The amount of cholesterol added was 212 µg.
Experiment Clay Fatty acid,
(µg spiked)
Ratio
of Time heated Cholesterol
at 100 ºC (h) recovered (%)
acid (uncertainty
cholesterol:
fatty
(w/w)
from
duplicates
experiments
unless stated
otherwise)
1
2
3
4
Yes Oleic acid (58 µg)
4:1
-
14
14
14
14
27 (±4, n=4)
30 (±5, n=4)
89 (±3)
Yes
No
-
Oleic acid (58 µg)
-
4:1
-
No
98 (±1)
5
6
7
Yes Oleic acid (58 µg)
Yes Oleic acid (58 µg)
Yes Oleic acid (58 µg)
4:1
4:1
4:1
1
2
4
86 (±2)
70 (±1)
69 (±1)
5

8
Yes Oleic acid (58 µg)
Yes Oleic acid (58 µg)
Yes Oleic acid (58 µg)
4:1
4:1
4:1
6
48 (±1)
30 (±3)
17 (±1)
9
14
24
10
11
12
13
Yes Stearic acid (56 µg)
Yes Oleic acid (58 µg)
Yes Linoleic acid (58 µg)
4:1
4:1
4:1
14
14
14
27 (±2)
30 (±4, n=4)
27 (±2)
14
15
16
17
18
Yes Oleic acid (104 µg)
Yes Oleic acid (214 µg)
Yes Oleic acid 416 µg)
Yes Oleic acid (2.1 mg)
Yes Oleic acid (21 mg)
2:1
14
14
14
14
14
28 (±1)
28 (±1)
25 (±3)
1 (±1)
1:1
1:2
1:10
1:100
1 (±1)
19
20
21
22
23
24
25
26
Yes Stearic acid (412 µg) 1:2
14
14
14
14
14
14
14
14
20 (±1)
1 (±1)
Yes Stearic acid (2.1 mg
Yes Stearic acid (21 mg)
1:10
1:100
1:100
1:10
2 (±1)
No
No
No
No
No
Stearic acid (21 mg)
Oleic Acid (2.1 mg)
Oleic Acid (21 mg)
Trioleate (2.1 mg)
Trioleate (21 mg)
49 (±9)
1 (±0)
1:100
1:10
1 (±0)
86 (±2)
88 (±2)
1:100
To test the effect of fatty acids as potential pro-oxidative agents, further heating
experiments with fatty acids of different degrees of unsaturation, i.e. stearic acid
(fully saturated), oleic acid and linoleic acid (mono-, and diunsaturated, respectively)
were conducted using a ratio of 1:4 w/w fatty acid to cholesterol (Experiments 11-13
in Table 1). No significant differences were seen in the recovery of cholesterol
between the fatty acids or a control without fatty acids. This showed that at these low
ratios fatty acids had only a minimal effect on cholesterol degradation and the
degree of unsaturation did not markedly affect reactivity. Accordingly, the clay was
deemed to be the major influence factor on the observed degradation.
Cholesterol was then heated with different ratios of oleic acid, from the initially used
ratio of 1:4 (fatty acid to cholesterol) to 100:1 w/w (experiments 14-18 in Table 1). It
was observed that when the ratio reached 10:1 w/w cholesterol became almost
6

completely degraded, and the GC chromatograms showed a wide range of new
peaks (Fig. 3a+b). The degradation products differed from those observed during
experiments involving heating cholesterol with only clay or lower concentrations of
fatty acids. Some of the most prominent new products formed were C6-C9 ω-diacids
(identified by GC/Q-ToF MS) together with unidentified degradation products eluting
from the GC later than cholesterol. The diacids likely formed from hydroperoxides
produced through autoxidation at the double-bond in oleic acid (ESI, Fig. 1).22 None
of the potentially formed unsaturated diacids (i.e. deca-2-en-1,10-dioic acid and
undeca-2-en-1,11-dioic acid from C-10 and C-11 peroxides) were observed, most
likely due to further reactions leading to their degradation.
7

Figure 3. GC-FID chromatograms of the products of experiments 17 and 18 (Table 1) corresponding
to oleic acid and cholesterol in ratios of a) 10:1 and b) 100:1 w/w showing the almost complete
degradation of cholesterol, as well as c) and d) the chromatograms of the products of experiments 21
and 22 (stearic acid: cholesterol 100:1 w/w) with and without clay, respectively. Peaks labelled with
C6-C9 are α,ω-diacids with 6-9 carbons.
Critically, the degradation seen in these experiments was observed even when oleic
acid was replaced with the saturated n-alkanoic acid, stearic acid (Experiments 19-
21 in Table 1). The conclusion is, therefore, that we are observing a fatty acid-
dependent reaction independent from the oxidative reactivity of the clay surface
observed before. The pattern of degradation products was very different when
stearic acid was used, rather than oleic acid, with none of the diacids being observed
and higher residual concentration of stearic acid (Fig. 3c). This was expected due to
the absence of double bonds in stearic acid and the consequential lower
susceptibility to autoxidation. When cholesterol alone was heated with an excess of
stearic acid (100:1 w/w) in the absence of clay powder a moderate decrease of
cholesterol was observed, but also the formation of a sizable peak which matched
the retention time of cholesterol stearate (Fig. 3d). This suggests that significant
esterification takes place under these conditions and the observed decrease of
cholesterol is not only through oxidation in this case. While the respective oleic acid
esters in the other experiments were possibly formed as intermediates, they were
apparently not stable under the experimental conditions and underwent further
degradation reactions. Noteworthy, heating of cholesterol with trioleate instead of
oleic acid lead only to a moderate decrease in cholesterol (12-14%, experiments 25-
26 in Table 1). Considering the evidence that sterol esters are more prone to
oxidation than free sterols this could indicate that esterification and further
degradation of the esters could also be an important reaction pathway here.18, 23
In summary, we have shown that both clay and fatty acids (present in high relative
abundances, compared to cholesterol, consistent with those seen in animal fats)
independently promote the degradation of cholesterol when heated at only 100 ºC.
Free fatty acids promoted cholesterol degradation more strongly than
triacylglycerols, but free fatty acids can be readily liberated from the TAGs during
cooking steps and after absorption into the ceramic matrix. Since both fatty acids
8

and clay would have been commonplace during cooking in archaeological pottery
(through the presence of the pot matrix and co-occurring lipids in foodstuffs), this
could explain the general absence of cholesterol in preserved lipid residues. The
latter makes the general survival and occurrence of cholesterol in archaeological
cooking residues unlikely, and its detection should always be interpreted with
caution.
Acknowledgments: This work was supported by NERC (NE/N011317/1). GC/Q-ToF
MS (CC010NERC capital grant) analyses were performed at the NERC Life
Sciences Mass Spectrometry Facility (Bristol Node). The authors thank the Natural
Environment Research Council, UK for partial funding of the mass spectrometry
facilities at Bristol (Contract No. R8/H10/63; http://www.lsmsf.co.uk).
References
1
Evershed, R. P. World Archaeology 2008, 40, 26.
2
Evershed, R. P. Archaeometry 2008, 50, 895.
3
4
Evershed, R. P.; Copley, M. S.; Dickson, L.; Hansel, F. A. Archaeometry 2008, 50, 101.
Hansel, F. A.; Copley, M. S.; Madureira, L. A. S.; Evershed, R. P. Tetrahedron Letters 2004, 45,
2999.
5
6
Hansel, F. A.; Evershed, R. P. Tetrahedron Letters 2009, 50, 5562.
Raven, A.; Van Bergen, P.; Stott, A.; Dudd, S.; Evershed, R. Journal of Analytical and Applied
Pyrolysis 1997, 40, 267.
7
Charters, S.; Evershed, R. P.; Quye, A.; Blinkhorn, P. W.; Reeves, V. Journal of Archaeological
Science 1997, 24, 1.
8
Dudd, S. N.; Regert, M.; Evershed, R. P. Organic Geochemistry 1998, 29, 1345.
9
Evershed, R. P. World Archaeology 1993, 25, 74.
10
Roffet-Salque, M.; Dunne, J.; Altoft, D. T.; Casanova, E.; Cramp, L. J. E.; Smyth, J.; Whelton, H.
L.; Evershed, R. P. Journal of Archaeological Science: Reports 2016.
11
12
13
Hammann, S.; Cramp, L. J. E. Journal of Archaeological Science 2018, 93, 74.
Zschau, W. European Journal of Lipid Science and Technology 2001, 103, 505.
Bortolomeazzi, R.; De Zan, M.; Pizzale, L.; Conte, L. S. Journal of Agricultural and Food
Chemistry 2000, 48, 1101.
14
15
16
17
Ubhayasekera, S. J.; Dutta, P. C. Journal of the American Oil Chemists' Society 2012, 89, 1427.
Hussin, F.; Aroua, M. K.; Daud, W. M. A. W. Chemical Engineering Journal 2011, 170, 90.
Kim, S.; Nawar, W. Journal of the American Oil Chemists Society 1991, 68, 931.
Osada, K.; Kodama, T.; Yamada, K.; Sugano, M. Journal of Agricultural and Food Chemistry
1993, 41, 1198.
18
Lehtonen, M.; Kemmo, S.; Lampi, A.-M.; Piironen, V. European Food Research and
Technology 2011, 232, 255.
19
Lampi, A.-M.; Kemmo, S.; Mäkelä, A.; Heikkinen, S.; Piironen, V. European Journal of Lipid
Science and Technology 2009, 111, 1027.
20
Barriuso, B.; Otaegui-Arrazola, A.; Menendez-Carreno, M.; Astiasaran, I.; Ansorena, D. Food
Chemistry 2012, 135, 706.
9

21
Journal of Lipid Science and Technology 2002, 104, 331.
22 Passi, S.; Picardo, M.; De Luca, C.; Nazzaro-Porro, M.; Rossi, L.; Rotilio, G. Biochimica et
Biophysica Acta (BBA)-Lipids and Lipid Metabolism 1993, 1168, 190.
23 Lehtonen, M.; Lampi, A.-M.; Ollilainen, V.; Struijs, K.; Piironen, V. European Food Research
Verleyen, T.; Cortes, E.; Verhe, R.; Dewettinck, K.; Huyghebaert, A.; De Greyt, W. European
and Technology 2011, 233, 51.
Highlights:
-
Cholesterol was degraded when dosed on powdered clay and heated at 100
ºC
-
-
Dehydration products and cholesterol oxidation products could be detected
Fatty acids in tenfold or higher excess promoted drastic cholesterol
degradation
-
This explains the absence of dietary sterols in archaeological lipid residues
10

11