Determination of optimal rehydration, fixation and staining methods for histological and immunohistochemical analysis of mummified soft tissues
A-M Mekota1 , M Vermehren2
Department of Biology I, Biodiversity Research/Anthropology1 and Department of Veterinary Anatomy II2 , Ludwig-Maximilians University Munich, Germany
Submitted January 8, 2002; revised May 4, 2004; accepted August 12, 2004
During an excavation headed by the German Institute for Archaeology, Cairo, at the tombs of the nobles in Thebes-West, Upper Egypt, three types of tissues from different mummies were sampled to compare 13 well known rehydration methods for mummified tissue with three newly developed methods. Furthermore, three fixatives were tested with each of the rehydration fluids. Meniscus (fibrocartilage), skin, and a placenta were used for this study. The rehydration and fixation procedures were uniform for all methods. The stains used were standard hematoxylin and eosin, elastica van Gieson, periodic acid-Schiff, and Grocott, and five commercially obtained immunohistochemical stains including pancytokeratin, vimentin, alpha-smooth-muscle-actin, basement membrane collagen type IV, and S-100 protein. The sections were examined by transmitted light microscopy. Our study showed that preservation of the tissue is dependent on the quality and effectiveness of the combination of the rehydration and fixation solutions, and that the quality of the histological and histochemical stains is dependent on the tissue quality. In addition, preservation of the antigens in the tissues is dependent on tissue quality, and fungal permeation had no influence on the tissue. Finally, the results are tissue specific. For placenta the best solution combination was Sandison and solution III (both fixed with formaldehyde) while results for skin were best with Ruffer I (using formaldehyde and Schaffer as fixatives), Grupe et al. (using formaldehyde as a fixative) and solution III (in combination with formaldehyde and Bouin fixatives). Ruffer II (using formaldehyde as a fixative) and solution III (in combination with Schaffer fixative) gave the best results for fibrocartilage.
Keywords: Egyptian mummy, fixation, histology, histochemistry, immunohistochemistry, mummified tissue, rehydration
Mummification, whether spontaneous or artificial, has provided soft tissue materials that allow in- sight into ancient cultures as well as diseases afflicting earlier populations. Palaeohistology is the study of histological structures found in pre- historic skeletal remains and mummified tissues. Unfortunately, dehydration and time have ren- dered the mummified tissues hard, brittle, and difficult to manage. Therefore, it is crucial to use optimal rehydration and fixation procedures to optimize morphological detail. Previous studies have employed a variety of fluids to reconstitute this material, but a systematic comparison to determine the optimal procedure has not been made.
Correspondence to: Anna-Maria Mekota, Department of Biology I, Biodiversity Research/Anthropology, Ludwig-Maximilians- University Mu¨nchen, Germany, Grosshadernerstr. 2, 82152 Martinsried, Germany. Tel: 0049-89-2180-74319; Fax: 0049-89- 2180-74 331; E-mail: email@example.com Biological Stain Commission Biotechnic & Histochemistry 2005, 80(1): 7Á/13. DOI:10.1080/10520290500051146 7
The aim of our study was to test the quality of 13 well known methods of rehydration for mummified tissue (Fulcheri et al. 1985, Giacometti and Chiarelli 1968, Gordon and Bradbury 1977, Graf 1949, Grupe et al. 1997, Kleiss and Simons- berger 1984, Piepenbrink and Herrmann 1988, Ruffer 1921, Sandison 1955, Turner and Holtom 1981, Wiest et al. 1994, Wilder 1904) and to compare these with three newly developed techniques (Table 1). Furthermore, three fixation fluids were tested with each rehydration solution. Each procedure was assessed for degree of tissue conservation, histological and histochemical stain- ing properties, and specificity of immunohisto- chemical staining methods.
Materials and methods In 1997, the German Institute for Archaeology headed an excavation of the tombs of the nobles in Thebes-West, Upper Egypt. At this time, three types of tissues were sampled from different mummies: meniscus (fibrocartilage), skin, and placenta. Archaeological findings suggest that the mummies dated from the New Kingdom (approxi- mately 1550Á/1080 BC). Three adjacent samples, approximately 0.5 cm3 , taken from each organ were placed in 100 ml of each rehydration fluid and left on a rotary mixer for a specific amount of time depending on the hardness and density of the tissue to be examined. Softening placenta samples required 24 h, skin samples required 36 h and meniscus samples required 48 h. After rehydration, samples of each tissue were taken from each rehydration solution and immersed in each of the following three fixatives for an additional 24 h: 4% formaldehyde, modified Schaffer solution (2:1 80% ethanol: 36% aqueous formaldehyde), or Bouin solution (15:15:1 satur- ated picric acid: 30% aqueous formaldehyde: concentrated acetic acid). After the samples were embedded in blocks of paraffin wax, they were cut with a microtome (Microm, HM440E) into both cross and longitudinal sections 4 mm thick. To ensure standard conditions for comparison, rehydration, fixation, embedding and staining procedures were kept uniform. The following histochemical stains were applied to the three tissues: hematoxylin and eosin (H & E; nuclei blue, cytoplasm pink), elastica van Gieson (EvG; elastic tissue black, connective tissue red), periodic acid-Schiff (PAS; nuclei blue, glycoproteins red) and Grocott (fungi black, cytoplasm green) (Bo¨ck 1989).
Table 1. Methods and ingredients of rehydration solutions
Method Ingredients Fulcheri et al. (1985) undiluted, inactivated human blood serum at 48 C Giacometti and Chiarelli (1968) 0.9% sodium chloride Gordon and Bradbury (1979) 70 ml ethanol (70%), 30 ml glycerin, 1g dithionite Graf (1949) 5 parts glycerin, 5 parts 10% acetic acid Grupe et al. (1997) 5% DMSO in Tris buffer, pH 7.6 Kleiss and Simonsberger (1984) 2% aqueous sodium carbonate Piepenbrink and Herrmann (1988) 15% glucose solution Ruffer I (1921) 5 parts distilled water, 3 parts absolute ethanol, 2 parts 5% aqueous sodium carbonate Ruffer II (1921) 97 parts tap water, 2 parts 5% aqueous sodium carbonate, 1 part 4% formaldehyde Sandison (1955) 5 parts 1% aqueous formaldehyde, 3 parts 96% ethanol, 2 parts 5% aqueous sodium carbonate Turner and Holtom (1981) 0.2% solution of "Comfort" fabric softer (Lever Bros.) in normal saline Wiest et al. (1994) 9.5 parts formaldehyde (2%), 0.5 parts Brij solution Wilder (1904) 1% solution of caustic potash Solution I 4 parts undiluted inactivated human blood serum at 48 C, 1 part 5% sodium carbonate Solution II 5 parts distilled water, 3 parts 15% saccharose solution, 2 parts 2% sodium carbonate Solution III 8 parts 0.2% solution of "Comfort" fabric softer (Lever Bros.) in 5% sodium carbonate, 2 parts aqueous formaldehyde (4%) 8 Biotechnic & Histochemistry 2005, 80(1): 7Á/13
Five commercially obtained immunohistochem- ical markers also were applied to the three tissues: pancytokeratin (Progen Biotechnik GmbH, Heidel- berg, Germany); vimentin (DAKO); alpha-smooth- muscle-actin (Boehringer Mannheim GmbH, Mannheim, Germany); basement membrane col- lagen type IV (Labo Nova, Giessen, Germany); and S-100 protein (DAKO). Immunohistochemistry was performed using enzymatic pre-digestion (pepsin) or heat-induced epitope retrieval in citrate buffer, pH 6.0, for enhancement of immunoreactivity, application of primary antibodies, and visualiza- tion by labeled-streptavidin-biotin (LSAB)/horse- radish peroxidase (DAKO) followed by DAB' (DAKO). Simultaneously, control experiments were performed with corresponding modern tis- sues. All sections were prepared in the Pathological Institute of the Ludwig-Maximilians-Universita¨t, Mu¨nchen, Germany. The sections were inspected by transmitted light microscopy.
We devised a grading system to compare and to rate each procedure for its degree of tissue con- servation, histological and histochemical staining properties, the specificity of immunohistochemical staining methods, and the degree of fungal pene- tration into the tissue.
Skin sections showed particularly good tissue preservation, although cellular outlines were never distinct. Although much of the epidermis had already separated from the dermis, the remaining epidermis often was preserved well (Fig. 1). The basal epithelial cells were packed with melanin as expected for specimens of Negroid origin. In the dermis, the hair follicles, hair, and sebaceous and sweat glands were readily apparent (Fig. 2). Blood vessels, but no red blood cells, and small peripheral nerves were identified unambiguously (Fig. 3). The subcutaneous layer showed loose connective tissue fibers attached to the dermis, and fat cell remnants were observed. To evaluate the influence of postmortum tissue decay by micro-organisms, the samples were tested for the presence of fungi using silver staining. Fungi were observed in some samples and were widespread in both epidermis and dermis. The molecular preservation of the antigen determinants, due to tissue preser- vation, determines the accuracy of the immunohis- tological stains. Depending on the rehydration or fixation procedure, specific immunohisto- chemical detections of single antigens were specific. Fig. 1. Ancient Egyptian skin. Epidermis, dermis and sweat glands rehydrated with solution III and fixed with formaldehyde. H & E. 200)/. Fig. 2. Ancient Egyptian skin. Sebaceous gland rehy- drated with solution III.
The placenta showed a relatively firm, uniform parenchyma containing a large amount of connec- tive tissue (Fig. 4). Unfortunately, only some cotyledons, blood vessels and placental villi could be identified due to significant tissue damage. In addition, fungi had penetrated the already spongy and web-like tissue. Stains yielded good results only in the parts of the connective tissue that were fairly well preserved. The best results were achieved using the H & E stain, although they were significantly different from fresh tissues as is frequently the case with ancient specimens (Herrmann et al. 1990). The H & E stain routinely provides variations in red hues, but appeared violet and brown in the historical slides. The EvG preparation displayed variations in shades of reddish brown, but no elastic fibers were identified. After applying the PAS stain, the historical speci- mens were stained a variety of blue tones, while the modern specimens produced shades of red. After double staining with Grocott, black fungi could be seen clearly against a green background. This stain was always specific and comparable to those of modern placentas. Owing to extensive decomposi- tion of the material, all immunohistochemical stains were ambiguous. There was no specific staining in any slides. The extreme background staining due to nonspecific protein fragments gave false-positive reactions; thus the results are incon- clusive and not useful for analysis.
Although it was no longer possible to categorize the individual knee joints of the menisci, because they were probably separated from the knee joints either during the excavation, by grave robbers, or simply by decomposition, the tissues showed a similar decomposition history. H & E preparations served well as an overview stain, differing only slightly from modern samples. Many lacunae could be identified, but no chondrocytes or cell nuclei were found (Fig. 5). The EvG stain often was accurate when suitable structures such as arteries and elastic fibers were present. Histologically, the collagen fibers (consisting of collagen type I) of the fibrocartilage could be stained positively by PAS. The fungi, stained with Grocott, usually appeared on the periphery and rarely penetrated the tissue. In the case of the menisci, the antibodies were expected to yield negative results, because there are no antigens preserved in the tissue. Therefore, if there was no staining, cross reactions did not occur. Most samples were not stained. Within the framework of our study, mucus de- generation was determined by alcian blue staining (Fig. 6). Alcian blue stains acidic mucus substan- ces blue. Cell nuclei, neutral glycosaminoglycans, Fig. 4. Collapsed cotyledons in ancient Egyptian human placenta rehydrated with solution III and fixed with formalin. C, collapsed cotyledon. EvG. 400)/. Fig. 3. Ancient Egyptian skin. Nerve tissue rehydrated with solution III and fixed with formaldehyde. H & E. 400)/ . 10 Biotechnic & Histochemistry 2005, 80(1): 7Á/13 carbohydrates and glycogen stain red. This applied to recent and archeological samples alike. Mucus degeneration occurs by the increase of atypical proteoglycans of the connective tissue. Mucopoly- saccharide pools in the affected connective tissue and disrupts the continuity of the collagen fibers so that the tear-resistance of the tissue is lost (Riede and Schaefer 1995).
Our study was based on the premise that all sections of a tissue series are either sampled from the same organ (placenta and skin) or from tissues with a similar decomposition history (meniscus), thereby ensuring that death and decomposition history as well as the time elapsed after the organ was buried are identical. Furthermore, all methods of tissue processing (embedding and staining) were identical; the samples were exposed only to differ- ing rehydration and fixating methods. These fac- tors strengthen the assumption that the observed differences in tissue quality can be traced solely to the different rehydration and fixation solutions and therefore lend themselves to assessing efficacy of the rehydration and fixation solutions. Comparison of the 16 rehydration and the corresponding three fixation methods using three different ancient Egyptian tissues (placenta, skin and fibrocartilage) revealed that the preservation of the tissue for current study is dependent on the quality and effectiveness of the combination of the rehydration and fixation solution. Furthermore, the quality of the histological and histochemical stains and the preservation of the antigens in the tissues, i.e. the specifity of the immunohistochem- ical stains, are dependent on the quality of the tissue provided by the combination of the rehydra- tion and fixation methods. Fungal penetration has no influence on the tissue, probably because the fungi permeated the tissues post mortem. The results are tissue-specific. The best solution combinations for the organs in question are summarized in Table 2. Depending on the rehydration and fixation solutions, good to very good results were achieved with Ruffer I (1921), Fig. 6. Ancient Egyptian meniscus rehydrated with Grupe et al. (1997) and fixed with formaldehyde. Alcian blue. 400)/. Table 2. Recommended rehydration and fixation solu- tions Organ Best rehydration and fixation fluids Placenta Sandison (1955) (formaldehyde) Solution III (formaldehyde) Skin Ruffer I (1921) (formaldehyde) Grupe et al. (1997) (formaldehyde) Solution III (formaldehyde) Ruffer I (1921) (Schaffer solution) Solution III (Bouin solution) Meniscus Ruffer II (1921) (formaldehyde) Solution III (Schaffer solution) Fig. 5. Ancient Egyptian meniscus rehydrated with solution III and fixed with Schaffer. EvG. 400)/. Mummified tissue analysis 11 Grupe et al. (1997), and Solutions II and III. The solutions of Ruffer II (1921), Sandison (1955), Gordon and Bradbury (1977), Graf (1949), Turner and Holtom (1981) and Wiest et al. (1994) yielded good results. Mediocre results were achieved using Giacometti and Chiarelli (1968), Fulcheri et al. (1985) and Solution I. The poorest rehydration resulted from Kleiss and Simonsberger (1984) and Piepenbrink and Herrmann (1988). The Wilder (1904) solution dissolved the samples so quickly that they had to be removed from the study; therefore, it proved entirely unsuitable.
What is necessary for a ``perfect'' rehydrating agent?
A ``perfect'' rehydration solution should diffuse quickly and evenly into the dry and brittle tissue, remove the brown color of the tissue, give the material the strength and hardness necessary for histological studies, and it should arrest decom- positon. To achieve this ``perfect'' solution, it must be an emulsifying agent, a solvent, a firming agent and a preservative. A good emulsifying agent should decrease the surface tension of water and be able to invade the material homogeneously. This can be achieved by glycerine, DMSO, glucose, saccharose, ``Comfort'' softener, sodium chloride, sodium carbonate, human serum, dithionite and Brij. A good rehydration agent should include a strong solvent that can moisturize the material and remove mummifying agents. Examples include ethanol, formaldehyde and acetic acid. An optimal rehydrating solution should give the tissue stability to ensure that it does not collapse. The stabilizing agents in the solutions tested here were ethanol, formaldehyde and acetic acid. Finally, a good rehydrating solution should include a strong preservative to inactivate bacteria by rendering the medium inhospitable to these organisms. This can be done by acetic acid, ethanol, dithionite or formaldehyde. Histological study of mummified tissue is com- plicated by many problems resulting from an advanced decomposition and from the type of mummification practice applied. When working with mummified material to diagnose pathologies, artifacts resulting from decomposition during the long period after the tissue was buried can become problematic and can result in so-called pseudopathologies. The main problem is the chemical composition of the mummified material. When creating analyzable sections of the mummi- fied material, surmounting the problems described above is of paramount importance for enabling reliable technical analysis to diagnose possible pathologies accurately that might have afflicted the people of antiquity. The molecular preservation of optimally prepared historical tissue must be comparable to modern material. Research efforts to prepare historical human tissue for histological research have been conducted previously in very narrow terms. Unsuitable rehy- dration and fixation solutions make identification of tissue structures and pathological diagnosis impossible. These pathological findings are neces- sary for referencing epidemiological questions. We hope the results of our study will contribute to the expansion of the scientific basis for researching the pathologies of our ancestors.
This research was funded by a Ph.D. grant from the Ludwig-Maximilians-Universita¨t Mu¨nchen. Tech- nical assistance was provided through a grant by the German Science Foundation. We are deeply grateful to Prof Dr G. Grupe for making this paper possible.
Bo¨ck P (1989) Romeis Mikroskopische Technik. 17th ed. Urban & Schwarzenberg. Fulcheri E, Massa ER, Fenoglio E (1985) Improvement in the histological technique for mummified tissue. Verh. Dtsch. Ges. Pathol. 69: 471. Giacometti L, Chiarelli B (1968) The skin of Egyptian mummies. A study in survival. Arch. Dermatol. 97: 712Á/ 716. Gordon K, Bradbury P (1977) Theory and Practice of Histological Techniques. Churchill-Livingstone, London. Graf W (1949) Preserved histological structures in Egyptian mummy tissues and ancient Swedish skeletons. Acta Anat. 8: 236Á/250. Grupe G, Schmid E, Soika J (1997) Identification of medieval human soft tissue remains in an advanced state of decomposition. Biotech. & Histochem. 72: 135Á/140. Herrmann B, Grupe G, Hummel S, Piepenbrink H, Schutkowski H (1990) Pra¨historische Anthropologie: Leitfa- den der Feld- und Labormethoden. Springer Verlag, Berlin. pp. 283Á/292. Kleiss E, Simonsberger P (1984) Zur Technik der Pala¨ohistologie. Mikroskopie 41: 91Á/107. Piepenbrink H, Herrmann G (1988) Behandlung u¨ber- dauerter Weichgewebe. In: Anthropologie. Handbuch der vergleichenden Biologie des Menschen. Band 1, Knubmann R, Ed. Gustav Fischer Verlag, Stuttgart. pp. 571Á/576. Riede U, Schaefer H (1995) Allgemeine und spezielle Pathologie. Georg Thieme Verlag, Stuttgart. pp. 49Á/76. 12 Biotechnic & Histochemistry 2005, 80(1): 7Á/13 Ruffer MA (1921) Histological studies on Egyptian mummies. In: Studies in the Paaleopathology of Egypt. University of Chicago Press, Chicago. pp. 49Á/89. Sandison AT (1955) The histological examination of mummified material. Stain Technol. 30: 277Á/283. Turner PJ, Holtom DB (1981) The use of a fabric softener in the reconstitution of mummified tissue prior to paraffin wax sectioning for light microscopical examina- tion. Stain Technol. 56: 35Á/38. Wiest I, Parsche F, Nerlich A (1994) Technical require- ments and limitations of collagen immunohistochemistry in historical bone and mummy tissue. Homo 45 Suppl: 141. Wilder HH (1904) The restoration of dried tissues with special reference to human remains. Am. Anthropol. 6: 1Á/ 17.