SEARCH

SEARCH BY CITATION

Keywords:

  • immunohistochemistry;
  • mummification model;
  • mummified remains;
  • protein preservation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Concluding remarks
  8. Acknowledgements
  9. References

Immunohistochemistry is an important tool in the investigation of ancient mummified remains because of its ability not only to detect proteins but also to isolate their location to specific tissues and thereby improve confidence that the results are genuine. A mouse model of Egyptian mummification has been used to demonstrate that the survival of proteins, judged by the retention of immunohistochemical staining, varies markedly. Some survive the process well, whereas others become barely detectable despite the morphology of the tissue being excellently preserved. The results obtained show that protein preservation is multi-factorial, with tissue type and degradation, and the properties of the protein itself all having significant effects. Proteins forming large, multi-subunit complexes such as collagen IV appear to be more resistant to degradation than those that do not, such as S-100. Although modern modelling studies cannot replicate the full extent of degradative processes and taphonomic changes experienced by real mummies, the results obtained can be useful for guiding research that requires ancient tissues.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Concluding remarks
  8. Acknowledgements
  9. References

The use of animal models for studying ancient Egyptian mummification techniques as described in ancient sources such as Herodotus (1996) and Diodorus Siculus (1933) has a long history, stretching back at least to the 1930s and Lucas’s experiments on pigeons (Lucas, 1932). However, most, if not all, of these models are primarily concerned with the final success or failure of the technique in question, treating the preservation of the body as a single event rather than a process. Although such models have proved invaluable for demonstrating the plausibility of ancient anthropogenic methods and the details of the preparative steps, there are very few examples of the use of these models to study the effects of mummification on the preservation of the tissue at the molecular level.

Immunohistochemistry has proved to be a valuable tool for studying mummified remains, both from Egypt and from other cultures, mainly for palaeopathology (Fulcheri 1995) but also to demonstrate the presence of neurochemicals (Hoyle et al. 1997) and to determine the success of different tissue processing techniques (Mekota & Vermehren, 2005a). It is especially popular, as not only can it show the presence of proteins in appropriate tissues, which increases confidence that the results are not due to contamination, but it can also use samples that are prepared for standard histological staining. Histology is a vital stage in the investigation of mummified tissue as it often provides the only method of determining the nature of the tissue sample with confidence, and, because it uses only tiny amounts of tissue for multiple tests, the demand for valuable and irreplaceable mummified tissue is reduced.

Employing a mouse model that mimicked the main parts of the best method described by Herodotus, specifically evisceration and salting with natron, we used immunohistochemistry to study the preservation of different proteins over the course of mummification.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Concluding remarks
  8. Acknowledgements
  9. References

Eighteen mice obtained from a colony cull had their intestines and stomachs removed through a full-length ventral incision. Two were chosen at random and sampled immediately to act as controls. Artificial natron was made from laboratory chemicals in proportions that would result in effective mummification [50% w/w anhydrous sodium carbonate, 30% w/w sodium bicarbonate, 10% w/w sodium chloride, 10% w/w sodium sulphate (Garner, 1979)], and was packed into the abdominal cavity of eight mice selected at random, which were then buried in more natron (approximately 500 g per mouse). The remaining mice were placed in a box and allowed to decompose normally at room temperature.

Pairs of mice were taken from each group after 5, 10, 20 and 40 days, and samples of skin and liver removed from each. The tissue samples were fixed overnight in a 10% formalin solution with 1% washing up liquid added to aid rehydration. The samples were then embedded in paraffin wax according to standard histological techniques.

Sections of 5 μm were cut and mounted onto 3-aminopropyltriethoxy silane (APES)-coated slides. The primary polyclonal antibodies were against:

  •  S-100 (Dako, raised in a rabbit and used at a 1 : 400 dilution) – found in Schwann cells in the peripheral nervous system;
  • • 
    α-smooth muscle actin (Abcam, raised in a rabbit and used at a 1 : 50 dilution) – a molecule found in smooth muscle cells, particularly those associated with vascular basement membranes);
  •  collagen IV (Abcam, raised in a rabbit and used at 1 : 50 dilution) – a component of all basement membranes.

Pre-treatment and optimal concentrations were determined using the manufacturer’s recommendations as a guide. The secondary antibody was biotinylated anti-rabbit IgG (Dako) used at a concentration of 1 : 500. The third component of the staining reaction, which allowed stain amplification, was horseradish peroxidise (HRP)-conjugated streptavidin (Vector Labs). The disclosing agent targeted at the bound peroxidise was diaminobenzidine (DAB, Sigma-Aldrich), following quenching of native peroxidise with 1% v/v H2O2. Negative controls with the primary antibody omitted were included alongside each reaction.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Concluding remarks
  8. Acknowledgements
  9. References

Decomposition controls

In the decomposition controls, many of the tissues had degraded to a point where there were no morphological or immunodetectable elements. Fat, muscle and nerve cells decomposed rapidly, reducing the samples to a combination of collagen and degradation products, and no positive results were seen in any of the skin samples after day 10, even in blood vessels that remained histologically recognisable. The liver samples included vessels that varied in size from large to small. The larger vessels had thicker basement membranes which retained positive staining of smooth muscle actin throughout the experiment (e.g. Fig. 1), indicating an improved resistance to degradation compared with the smaller vessels. However, the same sites were only positive for collagen IV until day 10. Most of the structure of the liver cell plates was lost during decomposition. Whereas sinusoidal membranes were positive for collagen IV in fresh tissue, the decomposition controls were all negative. However, there was some intracellular staining seen for collagen IV. Liver nerve fibres were S-100-negative throughout all the experiments.

image

Figure 1.  Liver sample from decomposition control taken after 20 days showing weak positive staining for smooth muscle actin around the basement membrane of a blood vessel. The dashed line outlines the deformed blood vessel, and the arrows point to some areas of staining. Scale bar: 100 μm.

Download figure to PowerPoint

Mummified tissues

The mummified skin samples retained positive staining of both nerves and vascular basement membranes throughout the experiment. However, the intensity of the S-100 stain diminished significantly over time, despite a lack of visible morphological damage. Mummified liver samples were positive for smooth muscle actin throughout, but showed no positive staining for S-100 at any time. Although sinusoids remained collagen IV-positive throughout, some intracellular staining similar to that seen in the decomposition controls was seen in samples taken at days 20 and 40 (Fig. 2). The results for all samples are summarised in Table 1.

image

Figure 2.  Liver sample from mummified sample taken after 20 days showing positive intracellular and sinusoidal collagen IV staining. Scale bar: 25 μm.

Download figure to PowerPoint

Table 1.   Summary of positive staining seen during the experiment.
 Control samples at dayMummified samples at day
  1. S, positive staining for S-100; SMA, positive staining seen for smooth muscle actin; C, positive staining seen for collagen IV; –, no staining seen.

Feature in skin51020405102040
Schwann cellsSSSSSS
Blood vessel basement membraneSMA CSMA CSMA CSMA CSMA CSMA C
Feature in liver51020405102040
Schwann cells
Blood vessel basement membraneSMA CSMA CSMASMASMA CSMA CSMA CSMA C
SinusoidsCCCC
‘Collapsed’ sinusoidsCCCCCC

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Concluding remarks
  8. Acknowledgements
  9. References

The results indicate that protein preservation is dependent on which tissue it is expressed in, how well preserved the tissue is, and the nature of the protein itself. Negative results in the later decomposition skin samples for collagen IV and smooth muscle actin were a function of wide-scale tissue degradation rather than simply being due to compromised antibody binding sites. Positive results in some of the later liver controls indicate that proteins can survive in a recognisable form for longer than indicated by the skin samples.

However, as shown by the rapid loss of S-100 from the controls and mummified liver samples, and the marked reduction in the density of reaction product seen in the mummified skin samples with time, some proteins may not survive mummification even when the tissue shows excellent morphological preservation and the presence of other proteins. The nature of the tissue is also likely to be important, as tissues with a higher enzymatic activity will show more rapid and extensive autolytic degradation of proteins. Other methods of mummification, especially those that involve rapid freezing, may show better preservation, as putrefaction is retarded at lower temperature. Different methods of desiccating tissue have also been shown to affect preservation of proteins, with collagen IV showing greater preservation in air-dried than in natron-preserved tissue, and smooth muscle actin showing the opposite trend (Jeziorska et al. 2005).

Given the nature of the proteins under investigation, it is unlikely that there is a single set of circumstances in which staining is reduced or lost. Both smooth muscle actin (Woodrum et al. 1975) and collagen IV (Hudson et al. 1993) form large polymeric complexes, whereas S-100 is dimeric (Fano et al. 1995), making it more soluble in its native state and therefore more readily lost alongside the fluid that seeps out during decomposition or is drawn out during mummification. This is probably the best explanation for the loss of positive staining seen in nerves that are morphologically unaffected by degradation.

For all three proteins, autolysis and putrefaction will adversely affect antibody binding, and hence staining. The size reduction and fragmentation caused by endogenous proteolytic enzymes and bacteria will break up the three-dimensional structure of the proteins, thereby destroying both conformational and sequential epitopes. Such damage to the tertiary structure has been suggested as a forensic method for determining time since death by immunohistochemical detection of insulin (Wehner et al. 1999), thyroglobulin (Wehner et al. 2000) and calcitonin (Wehner et al. 2001). Extensive decomposition has previously been blamed for inconclusive results in ancient mummified tissues, such as placenta (Mekota et al. 2005b). Reaction with water may also cause fragmentation and can chemically alter the amino acid side chains (Ambler & Daniel, 1991), making binding sites unrecognisable, although the extent of hydrolysis in the mummified samples will be limited by the removal of water from the tissue. Similarly, autolysis and putrefaction will have mainly been limited to the decomposition samples, as these processes have profound effects on morphology that are not seen in any of the mummified samples. Different tissues will undergo autolysis at a rate dependent on the concentration of proteolytic enzymes present, with the effect on specific proteins varying according to their properties. The combination of these factors may explain why the liver samples in both groups were S-100-negative throughout.

Concluding remarks

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Concluding remarks
  8. Acknowledgements
  9. References

The results obtained indicate that mummification preserves antigens in an immunologically recognisable form but that the extent of preservation is dependent on a range of factors. In addition, although specific tissue structures may be well preserved, histological preservation is not a reliable indicator of protein preservation. However, determining the state of preservation is important for research involving tissue-specific proteins, as the tissue obviously must be present for the experiment to stand a chance of success. The variability of preservation seen in the proteins used here indicates that the use of experimental models for preliminary studies into the viability of research that relies on the preservation of a specific protein should be considered before ancient material is used. Demonstrating whether the protein in question survives the mummification procedure may prevent irreplaceable ancient samples being destroyed when there is little or no hope of achieving a positive result with today’s technologies. Perhaps more importantly, diagnoses of diseases or conditions based on the absence of a protein (e.g. Fulcheri et al. 1992) should be approached cautiously at best.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Concluding remarks
  8. Acknowledgements
  9. References

This research was funded by an internal scholarship. The authors would like to thank Rosalie David, John Denton and Maria Jeziorska for their helpful advice and assistance.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Concluding remarks
  8. Acknowledgements
  9. References
  • Ambler RP, Daniel M (1991) Proteins and molecular palaeontology. Philos Trans R Soc Lond B Biol Sci 333, 381389.
  • Fano G, Biocca S, Fulle S, et al. (1995) The S-100: a protein family in search of a function. Prog Neurobiol 46, 7182.
  • Fulcheri E, Baracchini P, Rabino Massa E (1992). Immunocytochemistry in histopaleopathology (abstract). In Proceedings of First World Congress of Mummy Studies, Vol. 2, pp. 559. Tenerife.
  • Fulcheri E (1995) Immunohistochemistry: a new outlook in histopaleopathology. Boll Soc Ital Biol Sper 71, 105110.
  • Garner R (1979) Experimental mummification. In The Manchester Museum Mummy Project (ed. David AR), pp. 1924. Manchester: Manchester University Press.
  • Herodotus (1996) Herodotus: The Histories. London: Penguin.
  • Hoyle CH, Thomas PK, Burnstock G, et al. (1997) Immunohistochemical localisation of neuropeptides and nitric oxide synthase in sural nerves from Egyptian mummies. J Auton Nerv Syst 67, 105108.
  • Hudson BG, Reeders ST, Tryggvason K (1993) Type IV collagen: structure, gene organization, and role in human diseases. Molecular basis of Goodpasture and Alport syndromes and diffuse leiomyomatosis. J Biol Chem 268, 2603326036.
  • Jeziorska M, Wade R, Walker MG, et al. (2005) Egyptian versus natural mummification: tracking the differences in loss of tissue antigenicity. J. Biol Res 80, 229232.
  • Lucas A (1932) The use of natron in mummification. J Egypt Archaeol 18, 125140.
  • Mekota AM, Vermehren M (2005a) Determination of optimal rehydration, fixation and staining methods for histological and immunohistochemical analysis of mummified soft tissues. Biotech Histochem 80, 713.
  • Mekota AM, Grupe G, Zimmerman MR, et al. (2005b) First identification of an ancient Egyptian mummified human placenta. Int J Osteoarchaeol 15, 5160.
  • Siculus D (1933) Library of History. London: Heinemann.
  • Wehner F, Wehner HD, Schieffer MC, et al. (1999) Delimitation of the time of death by immunohistochemical detection of insulin in pancreatic beta-cells. Forensic Sci Int 105, 161169.
  • Wehner F, Wehner HD, Schieffer MC, et al. (2000) Delimitation of the time of death by immunohistochemical detection of thyroglobulin. Forensic Sci Int 110, 199206.
  • Wehner F, Wehner HD, Subke J (2001) Delimitation of the time of death by immunohistochemical detection of calcitonin. Forensic Sci Int 122, 8994.
  • Woodrum DT, Rich SA, Pollard TD (1975) Evidence for biased bidirectional polymerization of actin filaments using heavy meromyosin prepared by an improved method. J Cell Biol 67, 231237.