Human populations constantly share theecosystems with a wide variety of microorganisms(bacteria, viruses and protozoa) establishing withthem a delicate equilibrium; changes in theenvironment, transitions in human ecology ormutations and adaptations in microorganisms oftenaffect this delicate interaction.

Most microorganisms are benevolent, and someare symbiontic, because their presence is beneficialto both host and microbe, whereas some adverselyaffect the host’s biology (McMichael, 2001). In thelatter case we have an “infectious disease”.

Before the advent of molecular techniques, thediagnosis of a disease in antiquity wasfundamentally based on three sources ofinformation: artistic representations of pathologicalconditions, ancient descriptive tests, and directobservations of human remains (skeletal ormummified material) by radiologic or histologicalexaminations (Willcox, 2002). The first two

methods may be criticized, but the last source iswithout any doubt more reliable. In the earlynineties, Wood and colleagues (Wood et al., 1992)discussed “the osteological paradox” based on theevidence that in most cases the bones of thedeceased do not show signs of disease, simplybecause in many acute conditions the illness resultsin death before a skeletal response occurs. Themanifestations of diseases in bones are, generally,expressions of chronic conditions. For this reason,skeletons without signs of disease may belong eitherto healthy individuals or to individuals which havedied as a result of acute illness.

Hence to reconstruct the general condition ofhealth in the past, the information available inbones should be offset with analyses of the softtissues. Mummies (human or animal bodies thathave been preserved through artificial or accidentalmeans) are ideal for this purpose.

The analysis of the microbial DNA associatedwith human remains provides direct evidence of the

JASs Invited ReviewsJournal of Anthropological Sciences

Vol. 84 (2006), pp. 53-64

the JASs is published by the Istituto Italiano di Antropologia www.isita-org.com

The study of bacterial DNA in ancient human mummies

Franco Rollo, Luca Ermini, Stefania Luciani, Isolina Marota & Cristina Olivieri

Laboratorio di Archeo-Antropologia molecolare/DNA antico, c/o Dipartimento di Biologia MCA, UNICAM,62032 Camerino, Italy: e-mail: francougo.rollo@unicam.it

Summary – Analysis of ancient microorganism DNA represents one of the newest and most promisingbranches of molecular archaeology. In particular, microbial DNA associated with human remains can providedirect evidence of the occurrence and frequency of infectious diseases in historic times. Human mummiesrepresent very interesting subjects for palaeomicrobiological investigations as they retain soft tissues. Initial reportson the identification of ancient bacterial pathogens in human mummies using DNA analysis date back to theearly nineties and publications are steadily becoming more numerous. However, despite this favourable trend,the analysis of ancient microbial DNA is still a contentious issue. Among the difficulties in ancient bacterialDNA work, sample contamination due to aerosol in laboratories where the DNA of modern microbes ismanipulated and the ubiquitous dispersal of microorganisms seem to play the most prominent roles; hence theneed for authentication criteria. The present work critically reviews some of the most relevant papers on thissubject with special emphasis on the claims of Mycobacterium tuberculosis DNA identification in dynastic andpre-dynastic mummies from Egypt. The latest achievements in the emerging field of palaeomicrobiology of frozenmummies are also presented and discussed.

Keywords – Mummy, Bacteria, Ancient DNA, Permafrost, Intestinal flora, M. tuberculosis.

occurrence and frequency of infectious diseases andbears witness to the delicate relationship amonghumans, environment, and microbes in historictimes. Moreover, the analysis of bacterial DNA inancient tissue samples offers an interestingapproach for the study of the evolution ofmicroorganisms, their interaction with hosts, theorigin and spatial distribution of a disease, andaetiologies of epidemics.

In recent times, various old infectious diseases,such as tuberculosis, malaria and cholera, thatseemed to be eliminated, have been on the increase.Data from studies of past infectious diseases mayoffer insights into the present management ofpublic health, as well as an understanding of theemergence and reemergence of infectious disease.

This paper reviews the polymerase chainreaction (PCR) based molecular studies which haveaimed to detect and identify bacteria associatedwith archaeological human mummies.

Ancient DNA

In the past two decades the biological research onthe past has taken advantage of a new discipline,known as “ancient DNA” (aDNA). With this termwe indicate the study of any survived geneticmaterials in biological remains differing in origin,age and state of preservation such as human bones,embalmed animals, fixed tissues, coprolites, etc. Theunifying element is that the DNA isolated from thesesources is rare, short-sized, and almost inevitablycontaminated by modern human, bacterial andfungal DNA. These factors pose severe limitations tothe progress of ancient DNA research.

Various guidelines to assess the reliability of theaDNA have been proposed (Handt et al., 1994a;Cooper & Poinar, 2001; Hofreiter et al., 2001).

Currently, the aDNA approach is used in avariety of disciplines including anthropology,archaeology, conservation and evolutionary genetics,and palaeopathology (see for example the followingreviews: Willerslev & Cooper, 2005; Drancourt &Raoult, 2005; Pääbo et al., 2004).

Long-term persistence of bacterial DNA

The persistence of DNA throughout the courseof time remains a contentious issue. DNA molecules

are relatively unstable compared with other cellularcomponents and will degrade with time if notrepaired. DNA undergoes a spontaneousdepurination process that rapidly leads to strandbreakage. The speed of this process can vary as afunction of temperature and pH of the medium.Irrespective of the speed, however, the final result willalways be the same, i.e. the degradation of all DNAin the sample (Lindhal, 1993). Past reports (Höss etal., 1996; Leonard et al., 2000) indicate that whilesamples coming from Arctic and Antarctic regionshave relatively high success rates, this is not the casefor those coming from warm regions. On the basis ofthe extent of racemization of aspartic acid, Poinar etal. (1996) stated that “the survival of DNA is limitedto a few thousand years in warm regions such asEgypt and to roughly 105 years in cold regions.”

Marota et al., (2002) studied the DNA decay ratein papyri and human remains from Egyptianarchaeological sites. They established that the DNAhalf-life in papyri is about 19-24 years whichcorresponds to a maximum survival time of about530-670 years.

Theoretical calculations of DNA survival, basedon the Arrhenius equation and depurination kinetics(Lindhal & Nyberg, 1972), suggest (Fig. 1) that abacterial genome of 3.0 X 106 bp (alternating

54 Bacterial DNA in mummies

Fig. 1 - Long-term survival of 100 bp of DNA asa function of temperature. The calculations arebased upon a genome size of 3.0 X 106 bp, theArrhenius equation and depurination kinetics ofLindahl & Nyberg (1972) (i.e. a depurinationrate of 4 X 10-9 sites sec-1 at 70 °C, pH 7.4, anda constant activation energy of 31 kcal mol-1).The calculation is simplified assuming thatdamage is distributed equally over the genomeat all purine sites. (Willerslev et al., 2004a).

F. Rollo et al. 55

purines and pyrimidines in a ratio of 1:1, andassuming constant activation energy) will be brokeninto fragments of approximately ~100 bp in lengthwithin 500 years at 15 °C, 81,000 years at -10 °Cand 1.7 million years at -20 °C (Willerslev et al.,2004a).

Recently, Willerslev et al., (2004a) published astudy of DNA durability and degradation of a broadvariety of bacteria preserved under optimal freezingconditions. They investigated twelve permafrostsamples dated between 0 and 8.1 million of years.DNA was extracted from the samples and PCRamplified, using a variety of primer pairs designed tobind to different fragments of the 16S ribosomalRNA gene. Amplifications products of 120 bp and600 bp were obtained from samples 400,000-600,000 years old. Moreover, DNA concentrationsand taxonomic diversity were found to decrease withage. The results suggest that non-spore-forminggrampositive Actinobacteria are the most durable,outsurviving endospore formers such as Bacillaceaeand Clostridiaceae (Fig. 2).

Identifying ancient bacterial DNA

A fundamental issue in all palaeomolecularstudies is how to discriminate between ancient andmodern DNA. This problem is particularly relevantin the case of ancient microbes.

The following two criteria have been proposed.

First, use of PCR systems designed on the basisof DNA sequences which are specific for thepathogenic forms of a certain bacterium, but absentin free-living isolates. Examples of this are the questfor Mycobacterium tuberculosis (Dixon & Roberts,2001) and Yersinia pestis (Drancourt et al., 1998)using, respectively, the insertion sequence IS6110(M. Tuberculosis) and the RNA polymerase beta-subunit-encoding gene and the virulence-associatedplasminogen activator encoding gene (Y. pestis).

Second, application of the so-called criterion of“palaeoecological consistency” (Rollo & Marota,1999). According to this criterion, the remains arefirst tested for their diagenetic state by analyzing thepreservation of the original human mitochondrialDNA and the level of aspartic acid racemization(Poinar et al., 1996). If the response is favorable,DNA preparations obtained from different biopticsamples are amplified using “universal” PCRsystems designed on the basis of the 16S ribosomalRNA (16S rRNA) gene sequence. Followingcloning and sequencing of the amplified DNA, theancient bacteria are putatively identified bydatabase search. The results are finally evaluated forthe consistency of the (putative) bacterial type anddistribution with that of the living person.

The two criteria are by no means conflicting.The first is appropriate when the searchconcentrates on a single pathogen, normally on thebasis of the indications given by previousosteological or histological investigations. Thesecond, on the other hand, overcomes thelimitations imposed by the first when no previousindication is available. Practically, the secondcriterion can help us to determine which microbes,either symbionts, opportunists or pathogens, wereassociated with a certain historical individualduring his last days of life.

Authentication of the molecular datain palaeomicrobiology

Despite the growing number of papersaddressing this topic, the analysis of ancientmicrobial DNA is still a contentious issue. Twocases, in particular, have been the subject of debate.The first is the above cited detection of Y. pestis, theaetiological agent of plague, in human teeth fromMiddle Ages and Renaissance victims found in

Fig. 2 - Long term persistence of bacterial DNA inpermafrost. Relative frequency of gram-positiveActinobacteria, low-GC gram-positive bacteria,and gram-negative bacteria as derived from thescreening of 16S rRNA gene libraries (Willerslevet al., 2004b, modified).

Table 1 - Criteria for the authentication of molecular data in palaeomicrobiology (Drancourt & Raoult, 2005)

Absence of a positive control- The positive control should be removed from the laboratory in which ancient specimens are

processed.

Negativity of negative controls- Several negative controls should be analyzed in parallel with the specimens being processed.- Negative controls should be as similar as possible to the ancient specimens.- Negative controls should remain free of amplicons.

Sequencing of PCR amplicons- PCR alone does not ensure the specificity of the diagnosis, and amplicons have to be sequenced

to identify ancient microorganisms.

Targeting a new sequence in the laboratory- PCR should target a specific sequence that has not previously been amplified in the laboratory.

Amplification and sequencing of a second target- A positive result must be confirmed by amplification and sequencing of a second specific

molecular target.

Originality of the ancient sequence- The acquisition of an original sequence that differs from modern homologues in mutation or

deletion excludes contamination.

56 Bacterial DNA in mummies

European archaeological sites. While reports fromone laboratory (Drancourt et al., 1998; Raoult etal., 2000; Drancourt et al., 2004) claimed a veryhigh percentage of successful identifications, not asingle positive result could be obtained in twoothers (Gilbert et al., 2004a,b).

The second case is the identification of M.tuberculosis DNA in mummified human remainsfrom Egyptian archaeological sites (Nerlich et al.,1997; Zink et al., 2001, 2003a, 2003b, 2004;2005). These and the associated results on thepreservation of human nuclear DNA (Zink et al.,2004) were challenged on the basis of empirical andtheoretical calculations suggesting the unlikeness oflong-term survival of ancient DNA in the Egyptianenvironment (Marota et al., 2002; Willerslev &Cooper, 2005; Gilbert et al., 2005).

Among the difficulties in the ancient microbialDNA investigations, sample contamination due toaerosol in laboratories where the DNA of modernmicrobes is manipulated (Willerslev & Cooper,

2005) and the ubiquitous dispersal ofmicroorganisms (Finlay & Clarke, 1999) seem toplay some of the most prominent roles.

In addition to general criteria for identifyingancient DNA (Cooper & Poinar 2001; Pääbo et al.,2004), more specific protocols have been proposed(Rollo & Marota, 1999; Drancourt & Raoult,2005). It may however be worth remarking that thecriterion proposed by Drancourt & Raoult (2005)for the authentication of ancient pathogens focuseson the identification of the pathogen and pays littleor no attention at all to the consistency of the findwith the diagenetic level of the specimen (Table 1).In the past, we have suggested that molecular data inpalaeomicrobiology should provide an exhaustivedescription of the microbial ecology of thespecimen. In particular, they should take intoaccount the issue of the specimen’s recolonization bythe environmental flora and the erasing action of thelatter over the remains of the oldest genetic material(Rollo & Marota, 1999; Ubaldi et al., 1998).

F. Rollo et al. 57

Gilbert et al. (2005) recently proposed furtherauthentication criteria in the form of a series of keyquestions to be asked about ancient DNA (Table 2).

In 1993, Spigelmann and Lemma published thefirst molecular identification of a human microbialpathogen, M. tuberculosis, in an archaeologicalspecimen. The study was performed on 11 bonespecimens from Europe, Turkey and Borneo,ranging from 300-1400 years old, that hadpreviously been morphologically diagnosed withtuberculosis. To detect the pathogen’s DNA, a 123bp fragment of the insertion sequence IS6110 wasPCR amplified. This sequence (Eisenach et al.,1990) is specific to the M. tuberculosis complex(MTB) which includes M. tuberculosis, M. bovis,and M. africanum and is particularly suited forpalaeopathological investigations because it exists as10-16 copies per cell.

European mummiesA particularly comprehensive investigation was

reported by Fletcher et al. (2003). A large number(265) of burials dating from 1731 to 1838 A.D.was discovered in sealed crypts of the DominicanChurch, Vác, Hungary in 1994. Many bodies werenaturally mummified, so that both soft tissue andbones were available. Contemporary archivesenabled the determination of age at death, and theidentification of family groups. In some cases,symptoms before death were described and,

occasionally, occupation. Initial radiologicalexamination of a small number of individuals hadindicated calcified lung lesions and demonstrableacid-fast bacteria suggestive of tuberculosisinfection. Tuberculosis was endemic in 18th-19thcentury Europe, so human remains should containdetectable MTB DNA, enabling comparison withmodern isolates. Therefore, a detailed examinationof 168 individuals for the presence of MTB DNAwas undertaken. Specific DNA amplificationmethods for MTB showed that 55% of individualswere positive and that the incidence variedaccording to age at death and sampling site in thebody. Radiographs were obtained from 27individuals and revealed an association betweengross pathology and the presence of MTB DNA.There was an inverse relationship between PCRpositivity and MTB target sequence size. In somecases, the preservation of MTB DNA was excellent,and several target gene sequences could be detectedfrom the same sample, thus enabling the researchersto identify different genotypic groups of M.tuberculosis.

Pre-Columbian mummiesSalo and coworkers (Salo et al., 1994) published

the first finding of DNA of M. tuberculosis in a pre-Columbian mummy. They examined a right hilarlymph node and a lung lesion from a 40- to 45-yearold spontaneously mummified female body. Thetomb from which this body was exhumed was in aburial site (Chiribaya Alta) used by the Chiribaya,an agricultural population that occupied the lowerOsmore Valley (Peru) in about 1000-1300 A.D. In

Table 2 - Key questions to ask about ancient DNA (Gilbert et al., 2005).

• Do the age, environmental history and preservation of the sample suggest DNA survival?

• Do the biomolecular and/or macromolecular preservation of the sample, the amplified molecular target, the innate nature of the sample and its handling history suggest that contamination is a risk?

• Do the data suggest that the sequence is authentic, rather than the result of damage, jumping PCR, and contamination? Would patterns in the data indicate other artifacts, such as phantom mutations? Do the Authors offer sufficient additional proof that the sequences are authentic?

• Do the results make sense, and are there enough data to make the study useful and/or support the conclusions?

Identification of archaeological M. tuberculosis DNA

58 Bacterial DNA in mummies

this case M. tuberculosis DNA was identified thanksto the IS6100 primer system. This work is ofparticular importance as it gives the first molecularevidence of the disease before European contact.

Recently, Konomi et al. (2002) utilized tissuesamples from the genital area of 12 mummies in theAmerican Museum of Natural History collection inNew York in search of M. tuberculosis DNA. Themummies were excavated in the Andes Mountainregion of South America, and radiocarbon datingestimates that the mummies date from A.D. 140 to1200. DNA was successfully extracted from alltissues and was suitable for PCR analysis. PCR wascarried out to detect M. tuberculosis complex andmycobacteria other than M. tuberculosis (MOTB).The M. tuberculosis complex was detected in 2 outof 12 samples, and MOTB were detected in 7samples.

Egyptian mummiesThe identification of M. tuberculosis DNA in

Egyptian mummies has been reported several timessince 1997 (Nerlich et al., 1997; Crubézy et al.,1998; Zink et al., 2003a,b; 2005).

Crubézy et al. (1998) claimed the recovery ofM. tuberculosis DNA sequences from the almostcomplete skeleton of a 12-14-year-old child. Theindividual was exhumed from a tomb in thePredynastic necropolis of Adaïma, located in thedistrict of Esna (Upper Egypt), where more than200 graves have been excavated since 1990. Thefinding of a typical pottery of Nagada IID2associated to the skeleton allowed to date it at3400-3200 B.C. The child exhibited a kyphotic,“hunchback” spinal deformity consistent withPott’s disease and suggestive of tuberculousvertebral involvement.

In a study by Zink et al. (2003b), bone and softtissue samples from 85 ancient Egyptian mummieswere analyzed for the presence of ancient M.tuberculosis complex DNA and furthercharacterized by spoligotyping. The specimens wereobtained from individuals buried in different tombcomplexes in Thebes West, Upper Egypt, in whichupper class fellows were buried during the MiddleKingdom (since approximately 2050 B.C.) and theLate Period (until approximately 500 B.C.). A totalof 25 samples provided a specific positive signal forthe amplification of a 123-bp fragment of the

repetitive element IS6110, indicating the presenceof M. tuberculosis DNA. Further PCR-based testsfor the identification of subspecies failed due to lackof specific amplification products. Of these 25positive specimens, 12 were successfullycharacterized by spoligotyping. They were shown todisplay either an M. tuberculosis or an M. africanumpattern, but none revealed an M. bovis-specificpattern.

The claims of recovery of M. tuberculosis DNAfrom Egyptian mummies and skeletal remains haveraised considerable debate as they contradicttheoretical and empirical estimates of long-termDNA preservation in the Egyptian environment(Marota et al., 2002; Gilbert et al., 2005; Willerslevet al., 2004b). Further scrutiny of these results istherefore required. On the other hand, supportersof the authenticity put forward the particularbiology of mycobacteria. The pathogen oftuberculosis is characteristically localized incalcified lesions that might help to preserve residualDNA. Mycobacteria have DNA with a highproportion of guanine and cytosine; this increasesDNA stability and may aid its survival. In addition,the thick cell walls of these mycobacteria are lipid-rich, protecting DNA from the attack of lyticenzymes after autolysis and necrosis of the hostcells, microflora and fauna of the host upon death,and the first-stage decomposers that invade a bodyonce a person dies. M. tuberculosis has survived afterfixing in formalin and staining (Gerston et al.,1998), and has even been transmitted from thecorpse to a new host one year after burial (Sterlinget al., 2000).

The digestive tract of frozen mummiesas a model system forpalaeomicrobiological investigations

Given the close correlation between long-termpreservation of DNA and environmentalconditions, mummified human remains kept attemperatures below 0 °C are ideal subjects forpalaeomolecular studies. This kind of mummy isfound in the Arctic region, in Siberia and at highaltitudes in the Andes and the Alps (Zimmermann,1996; Notman & Beattie, 1996; Hart Hansen &Nordqvist, 1996; Horne, 1996; Spindler, 1995). Ingeneral, the bodies have all undergone a natural

F. Rollo et al. 59

freeze-desiccation (lyophilization) process eitherdue to water phase separation (permafrost) orsublimation (high altitude).

To date, most of the scientifically valuableinformation regarding frozen mummies derivesfrom histopathological, immunological andelectron microscopic observations. Relatively fewmolecular data are available.

The human digestive tract harbors a highlycomplex and abundant community ofmicroorganisms (Tannock 2000; Hooper &Gordon 2001; Hooper et al., 2002). Thecomposition and activity of the intestinal bacterialcommunity has a significant impact on the healthof the host because of its influence on nutrition,bowel habit, the physiology of the mucin barrier,and the ontogeny of the mucosal immune system(Moreau & Corthier 1988; Okada et al., 1994;Hooper & Gordon 2001). In normal conditions,the upper part of the small intestine (duodenum,jejunum) is considered to have a low microbialcontent, while its lower portion (ileum) ischaracterized by higher cell counts. The highestmicrobial content is found in the large intestine(Hill, 1995).

In the past, the composition of the intestinalmicroflora was studied using traditionalmicrobiological techniques. In the last 10 years,however, several 16S rRNA gene libraries have beenprepared and screened (Suau et al., 1999; Hold etal., 2002; Wang et al., 2003) providing animportant element of reference for thepalaeomicrobiological investigators.

As stressed above, the authentication of the datareferring to single bacterial species in archaeologicalspecimens may be problematic, due to the ease oflaboratory contamination and the ubiquity ofmicroorganisms. These issues are markedly reducedif the object of the investigation is not a singlebacterial species but a community of bacteria, suchas that inhabiting the digestive tract.

An early study on the preservation of theintestinal flora’s DNA in a frozen-desiccatedmummy was published in 1998 (Ubaldi et al.,1998). The mummy, at the present kept at theNational Museum of Anthropology in Florence,comes from the Peruvian Andes and has beenradiocarbon dated to the years 980-1170 A.D.. Theanalysis of 16S ribosomal RNA gene fragments in

bioptic samples from the small and large bowels ofthe mummy showed that the concentration of thefragments was considerably higher in the latter, asituation reminiscent of that currently found in theliving. In a further step (Luciani et al., 2006), 16SrRNA gene fragments from fecal residues were usedto construct a library. Screening of the library bysequencing and comparison with known bacterialsequences from Genbank, demonstrated that tracesof the original microbial flora were still present. Inparticular, it was possible to putatively identify theDNA of Haemophylus parainfluenzae. As suggestedby the name, meaning “blood lover” in Greek, thewhole genus Haemophylus consists of species thatare associated with warm-blooded animals. In thepast, the unfavorable growth characteristics of thisorganism and the poor specificity of traditionalmethods for species identification were responsiblefor inaccuracies in the diagnosis of infection causedby H. parainfluenzae and related organisms, leadingto a substantial underestimation of their pathogenicrole. More recently, thanks to the application ofmolecular techniques, H. parainfluenzae wasrecognized as a cause of serious invasive diseasessuch as endocarditis and community-acquiredpneumonia.

Currently, the most famous frozen mummy isthe so-called “Tyrolean Iceman”, or “SimilaunMan”, or Ötzi. This mummy was discovered on19th September 1991 at 3270 m above sea level inan Alpine glacier near the Austro-Italian border.The most relevant feature of the discovery,radiocarbon dated to 3350–3100 B.C. (Bonani etal., 1992), corresponding to the Late Neolithic orEarly Copper Age, is the excellent state ofpreservation of the body and the pieces of clothingand equipment found with it (Spindler, 1995).

Since his discovery, the Iceman has been theobject of molecular investigations aimed atdetermining the mitochondrial haplogroup (Handtet al., 1994b; Rollo et al., 2006), the composition ofthe last meals (Rollo et al., 2002), the species ofgrass utilized to manufacture some pieces of theequipment (Rollo et al., 1994, 1995a,b) and thetypes of microorganisms associated with the clothes(Ubaldi et al., 1996; Antonini et al., 2000).

A preliminary study of the intestinal microfloraDNA of the Iceman was performed in the latenineties by Cano and coworkers (2000). The poor

60 Bacterial DNA in mummies

quality of the specimens available at that time,however, made it possible to obtain only a verysmall number (17) of bacterial sequences. Theywere putatively attributed to the species Clostridiumperfringens, C. ghonii, C. sordellii, Eubacteriumtenue, and Bacteroides sp. Fortunately, newinvestigations could be carried out following thefirst complete defrosting of the body which tookplace on 25th September 2000 at the South TyrolArchaeological Museum in Bolzano, Italy (Fig. 3).

Last year, in addition, chance made it possibleto study another Alpine mummy of relatively recentorigin. On 22nd August 2004, three human bodieswere recovered at 3596 m a.s.l. in the Forni glacier(Alta Val di Pejo, northern Italy). Remains ofclothes and equipment made it possible to identifythe bodies as belonging to three Austrian soldiers ofthe 1st Regiment Tyroler Kaiserschützen (Fig. 4).They were killed by a burst of some shells on 3rdSeptember 1918 during a counterattack on theItalian lines. Because of the impediments posed bythe battle, the corpses were forsaken on the glacier(either thrown into a crevasse or covered withsnow) rather than being transported down thevalley. Before being transferred to the Merano warcemetery, to be buried with military honors, thecorpses were examined by Dr. Eduard Egarter-Viglof the Regional Hospital of Bolzano and found tobe perfectly mummified. As the naturalmummification process (freeze-drying) closelyresembled that of the Iceman, one of the bodies wassubjected to necroscopic examination, and biopticsamples of the internal organs were collected and

stored for DNA analysis.DNA was isolated from the intestinal (colon)

content of the 1918 mummy and the Iceman, andPCR amplified using different sets of primer pairsdesigned to bind to different fragments of the 16SrRNA gene. Amplified DNA was then cloned.Sequence analysis of the resulting two libraries (49sequences for the 1918 mummy, 119 sequences forthe Iceman) showed that the characteristiccomposition of the intestinal microflora of man(mainly members of the classes Alpha-, Beta- andGammaprotebacteria, Bacteroides and Clostridia)could still be traced in the recent mummy while, inthe Iceman, more than 90% of the flora wasrepresented by Clostridia (Fig. 5). Comparison ofthese data with recent reports describing thescreening of 16S rRNA gene libraries from otherparts of the Iceman’s body, such as muscle and skin(Rollo et al., 2000) and environmental samples suchas specimens of present-day and ancient Siberianpermafrost (Willerslev et al., 2004a) suggests that

Fig. 3 - The Iceman following complete de-frosting on 25th September 2000.

Fig. 4 - One of the three natural mummiesrecovered on 22nd August 2004 in the ForniGlacier (Alta Val di Pejo, northern Italy).

the change of bacterial DNA composition inside themummies may be attributed partly to theproliferation of the cadaveric flora and in part to thelower stability of the DNA of gram negative bacteriacompared to that of endospore-former low-GCgram positive bacteria.

Conclusions

The study of bacterial DNA in humanmummies is an emerging field. While its potentialfor palaeomicrobiological, palaeopathological andanthropological studies is undisputed, it is limited alack of standard protocols. It seems clear that thedevelopment of the field will benefit from advancedunderstanding of our knowledge of themechanisms responsible for DNA preservation. Inparticular, the emerging issue of the differentialpreservation of bacterial DNA in archaeologicalmaterials is likely to play a major role in futurepalaeomicrobiological research.

Acknowledgements

Work in the Authors’ laboratory was supported by theMinistero dell’Istruzione, dell’Università e dellaRicerca research project “Malattie, ambiente e societàalla Corte Granducale di Firenze: studio storico,archeologico e paleopatologico delle deposizioni funebridei Medici (secoli XVI-XVIII).

References

Antonini S., Luciani S., Marota I., Ubaldi M. &Rollo F. 2000. Compilation of DNA sequencesfrom the iceman’s grass clothing. InBortenschlager S. & Oeggl K. (eds), TheIceman and his Natural Environment; pp. 157-162. Springer, Wien.

Bonani G., Ivy S.D., Niklaus T.R., Suter M.,Housley R.A., Bronj C.R., van Klinken G.J. &Edges R.E.M. 1992. Altersbestimmung vonLilligrammproben der Ötzaler Gletscherleichemit der BeschleunigermassenapektrometrieMethode (AMS). In Höpfel F., Platzer W.,Spindler K., (eds), Der Mann im Eis 1; pp.108-116. Veröffentlichungen der UniversitätInnsbruck, Innsbruck.

Cano R.J., Tiefenbrunner F., Ubaldi M., Del CuetoC., Luciani S., Cox T. Orkand P., KunzelK.H., & Rollo F. 2000. Sequence analysis ofbacterial DNA in the colon and stomach of theTyrolean iceman. Am. J. Phys. Anthropol., 112:297-309.

Cooper A. & Poinar H. N. 2001. Ancient DNA:do it right or not at all. Science, 18: 289.

Crubézy E., Ludes B., Poveda J.D., Clayton J.,Crouau-Roy B. & Montagnon D. 1998.Identification of Mycobacterium DNA in anEgyptian Pott’s disease of 5,400 years old. C. R.Acad. Sci. Paris., 321: 941-951.

F. Rollo et al. 61

Fig. 5 - Composition of the intestinal (colon) microflora in the living and in two mummies as derivedfrom the screening of 16S rRNA gene libraries (*other sequences: Flavobacteria, Fusobacteria,Mollicutes, Verrucomicrobiae).

Dixon R.A. & Roberts C.A. 2001. Modern andancient scourges: the application of ancientDNA to the analysis of tuberculosis andleprosy from archaeologically derived humanremains. Ancient Biomolecules 3: 181-193.

Drancourt M., Aboudharam G., Signoli M.,Dutour O. & Raoult D. 1998. Detection of400-year-old Yersinia pestis DNA in humandental pulp: an approach to the diagnosis ofancient septicemia. Proc. Natl. Acad. Sci. USA,95: 12637-12640.

Drancourt M., Roux V., Dang L.V., Tran-Hung L.,Castex D., Chenal-Francisque V., Ogata H.,Fournier P.E., Crubezy E. & Raoult D. 2004.Genotyping, Orientalis-like Yersinia pestis, andplague pandemics. Emerg. Infect. Dis., 10:1585-1592.

Drancourt M. & Raoult D. 2005.Palaeomicrobiology: current issues andperspectives. Nature Rev. Microbiol., 3: 23–35.

Eisenach K.D., Cave M.D., Bates J.H. & CrawfordJ.T. 1990. Polymerase chain reactionamplification of a repetitive DNA sequencespecific for Mycobacterium tuberculosis. J.Infect. Dis., 161: 977-981.

Finlay B.J. & Clarke K.J. 1999 - Ubiquitousdispersal of microbial species. Nature 400: 828.

Fletcher H.A., Donoghue H.D., Holton J., Pap I.& Spigelman M. 2003. Widespreadoccurrence of Mycobacterium tuberculosisDNA from 18th-19th century Hungarians.Am. J. Phys. Anthropol., 120: 144-152.

Gerston K.F., BlumbergL. & Gafoor H. 1998.Viability of mycobacteria in formalin-fixedtissues. Int. J. Lung Dis.; 2: 521-523.

Gilbert M.T.P., Cuccui J., White W., Lynnerup N.,Titball R.W., Cooper A. & Prentice M.B.2004a. Absence of Yersinia pestis-specific DNAin human teeth from five Europeanexcavations of putative plague victims.Microbiol., 150: 341-354.

Gilbert M.T.P. (and 10 others) 2004b. Ancientmitochondrial DNA from hair. Curr. Biol., 14:R463-R464.

Gilbert M.T.P., Bandelt H-J., Hofreiter M. &Barnes I. 2005. Assessing ancient DNAstudies. Trends Ecol. Evol. 20: 541-544.

Handt O., Höss M., Krings M. & Pääbo S. 1994a.Ancient DNA: methodological challenges.

Experientia, 50: 524-529.Handt O., Richards M., Trommsdorff M., Kilger

C., Simanainen J., Georgiev O., Bauer K.,Stone A., Hedges R., Schaffner W., UtermannG., Sykes B. & Pääbo S. 1994b. Moleculargenetic analyses of the Tyrolean Ice Man.Science, 264: 1775-1778.

Hart Hansen J.P. & Nordqvist J. 1996.Themummy find from Qilakitsoq in northwestGreenland. In K. Spindler, H. Wilfing, E.Rastbichler-Zissernig, D. zur Nedden & H.Nothdurfter (eds), Human mummies ( Theman in the ice 3); pp. 107-121. Springer-Verlag, Wien, New York.

Hill M. J. 1995. The normal gut bacterial flora. InHill M. J. (ed.), Role of gut bacteria in humantoxicology and pharmacology; pp. 3-17. Taylorand Francis, London.

Hofreiter M., Serre D., Poinar H.N., Kuch M. &Pääbo S. 2001. Ancient DNA. Nature Rev.Genet., 2: 353-360.

Hold G.L., Pryde S.E., Russell V.J., Furrie E. &Flint H.J. 2002. Assessment of microbialdiversity in human colonic samples by 16SrDNA sequence analysis. FEMS Microbiol.Ecol., 39: 33-39.

Hooper L.V. & Gordon J.I. 2001. Commensalhost-bacterial relationships in the gut. Science,292: 1115-1118.

Hooper L.V., Midtvedt T. & Gordon J.I. 2002.How host microbial interactions shape thenutrient environment of the Mammalianintestine. Ann. Rev. Nut., 22: 283-307.

Horne P.D. 1996. The Prince of El Plomo: a frozentreasure. In K. Spindler, H. Wilfing, E.Rastbichler-Zissernig, D. zur Nedden & H.Nothdurfter (eds), Human mummies (The manin the ice 3); pp.153-157. Springer-Verlag,Wien, New York.

Höss M., Jaruga P., Zastawny T.H., Dizdaroglu M.& Pääbo S. 1996. DNA damage and DNAsequence retrieval from ancient tissues. NucleicAcids Res., 24: 1304-1307.

Konomi N., Lebwohl E., Mowbray K., Tattersall I.& Zhang D. 2002. Detection of mycobacterialDNA in Andean mummies. J. Clin. Microbiol.,40: 4738-4740.

Leonard J.A., Wayne R.K. & Cooper A. 2000.Population genetics of Ice Age brown bears.

62 Bacterial DNA in mummies

Proc. Natl. Acad. Sci. USA, 97: 1651-1654.Lindahl T. & Nyberg B. 1972. Rate of

depurination of native DNA. Biochem., 11:3610–3618.

Lindahl T. 1993 - Instability and decay of theprimary structure of DNA. Nature, 362:709–715.

Luciani S., Fornaciari G., Rickards O., MartìnezLabarga C & Rollo F. 2006. Molecularcharacterization of a pre-Columbian mummyand in situ coprolite. Am. J. Phys. Anthropol.,129: 620-629.

Marota I., Basile C., Ubaldi M. & Rollo F. 2002.DNA decay rate in papyri and human remainsfrom Egyptian archaeological sites. Am. J. Phys.Anthropol., 117: 310-318.

McMichael A.J. 2001 - Human culture, ecologicalchange and infectious disease. Ecosyst. Health,7: 107-115.

Moreau M.C. & Corthier G. 1988. Effect ofgastrointestinal microflora on induction andmaintenance of oral tolerance to ovalbumin inC3H/HeJ mice. Infect Imm., 56, 2766–2768.

Nerlich A.G., Haas C.J., Zink A., Szeimies U. &Hagedorn H.G. 1997.Molecular evidence fortuberculosis in an ancient Egyptian mummy.Lancet, 350: 1404.

Notman D. & Beattie O. 1996. The palaeoimagingand forensic anthropology of frozen sailorsfrom the Franklin Arctic expedition massdisaster (1845-1848): a detailed presentationof two radiological surveys. In K. Spindler, H.Wilfing, E. Rastbichler-Zissernig, D. zurNedden & H. Nothdurfter (eds), Humanmummies (The man in the ice 3); pp.93-106.Springer-Verlag, Wien, New York.

Okada Y., Setoyama H., Matsumoto S., Imaoka A.,Nanno M., Kawaguchi, M. & Umesaki Y.1994. Effects of fecal microorganisms andtheir chloroform-resistant variants derivedfrom mice, rats and humans onimmunological and physiologicalcharacteristics of the intestines of ex-germfreemice. Infect Imm., 62: 5442-5446.

Pääbo S., Poinar H., Serre D., Jaenicke-Després V.,Hebler J., Rohland N., Kuch M., Krause J.,Vigilant L. & Hofreiter M. 2004. Geneticanalyses from ancient DNA. Annu. Rev. Gen.38: 645-679.

Poinar H.N., Höss M., Bada L.J. & Pääbo S. 1996.Amino acid racemization and the preservationof ancient DNA. Science, 272: 864-866.

Raoult D., Aboudharam G., Crubezy E., LarrouyG., Ludes B. & Drancourt M. 2000.Molecular identification by “suicide PCR” ofYersinia pestis as the agent of medieval blackdeath. Proc. Natl. Acad. Sci. USA, 97: 12800-12803.

Rollo F., Asci W., Antonini S. & Ubaldi M. 1994.Molecular ecology of a Neolithic meadow, theDNA of the grass remains from thearchaeological site of the Tyrolean Iceman.Experientia, 50: 576-584.

Rollo F., Asci W., Marota I. & Sassaroli S. 1995a.DNA analysis of grass remains found at theiceman’s archaeological site. In Spindler K.,Rastbichler-Zissernig E., Wilfing H., NeddenD. zur and Nothdurfter H. (eds), Der Mannim Eis; pp. 91-105. Springer, Wien.

Rollo F., Sassaroli S. & Ubaldi M. 1995b.Molecular phylogeny of the fungi of theIceman’s grass clothing. Curr. Genet. 28: 289-297.

Rollo F. & Marota I. 1999. How microbial ancientDNA, found in association with humanremains, can be interpreted. Phil. Trans. R. Soc.Lond. B., 354: 111-119.

Rollo F., Luciani S., Canapa A. & Marota I. 2000.Analysis of bacterial DNA in skin and muscleof the Tyrolean iceman offers new insight intothe mummification process. Am. J. Phys.Anthropol., 111: 211-219.

Rollo F., Ubaldi M., Ermini L. & Marota I. 2002 –Ötzi’s last meals: DNA analysis of theintestinal content of the Neolithic glaciermummy from the Alps. Proc. Natl. Acad. Sci.USA, 99: 12594-12599.

Rollo F., Ermini L., Luciani S., Marota I., OlivieriC. & Luiselli D. 2006. Fine characterization ofthe Iceman’ s mtDNA haplogroup. Am. J.Phys. Anthropol., 130: 557-564

Salo W.L., Aufderheide A.C., Buikstra J. &Holcomb T.A. 1994. Identification ofMycobacterium tuberculosis DNA in a pre-Columbian Peruvian mummy. Proc. Natl.Acad. Sci. USA, 91: 2091-2094.

Spigelman M. & Lemma E. 1993. The use ofpolymerase chain reaction to detect

F. Rollo et al. 63

Mycobacterium tuberculosis in ancientskeletons. Papers from the Institute ofArchaeology, 7: 69-74.

Spindler K. 1995. The Man in the Ice. OrionLondon.

Sterling T.R., Pope D.S., Bishai W.R., HarringtonS., Gershon R.R. & Chaisson R.E. 2000.Transmission of Mycobacterium tuberculosisfrom a cadaver to an embalmer. N. Engl. J.Med., 342: 246-248.

Suau A., Bonnet R., Sutren M., Godon J.J., GibsonG.R., Collins M.D. & Doré J. 1999. DirectAnalysis of Genes Encoding 16S rRNA fromComplex Communities Reveals Many NovelMolecular Species within the Human Gut.Appl. Environ. Microbiol. 65: 4799-4807.

Tannock G.W. 2000. The intestinal microflora:potentially fertile ground for microbialphysiologists. Adv. Microbiol. Phys., 42: 25-46.

Ubaldi M., Sassaroli S. & Rollo F. 1996. RibosomalDNA Analysis of culturable Deuteromycetesfrom the iceman’s hay: comparison of livingand mummified fungi. Ancient biomolecules, 1:35-42.

Ubaldi M., Luciani S., Marota I., Fornaciari G.,Cano R. J. & Rollo F. 1998. Sequence analysisof bacterial DNA in the colon of an Andeanmummy. Am. J. Phys. Anthropol., 107: 285-295.

Wang X., Heazlewood S.P., Krause D.O. & FlorisT.H.J. 2003. Molecular characterization of themicrobial species that colonize human ilealand colonic mucosa by using 16S rDNAsequence analysis. J. Appl. Microbiol., 95: 508-520.

Willcox A.W. 2002. Mummies and molecules:molecular biology meets paleopathology. Clin.Microbiol. Newsl., 24: 57–60.

Willerslev E., Hansen A.J. & Poinar H.N. 2004a.Isolation of nucleic acids and cultures fromfossil ice and permafrost. Trends Ecol. Evol. 19:141-147.

Willerslev E., Hansen A.J., Rønn R., Brand T.B,Barnes I., Carsten W., Gilichinsky D.,Mitchell D. & Cooper A. 2004b. Long-term

persistence of bacterial DNA. Curr. Biol., 14:9-10.

Willerslev E. & Cooper A. 2005. Ancient DNA.Proc. R. Soc. B. 272: 3-16.

Wood J.W., Milner G.R., Harpending H.C. &Weiss K.M. 1992. The osteological paradox:problems of inferring prehistoric health fromskeletal samples. Curr. Anthropol., 33: 343-370.

Zimmermann M.R. 1996. Mummies of the Arcticregions. In K. Spindler, H. Wilfing, E.Rastbichler-Zissernig, D. zur Nedden & H.Nothdurfter (eds), Human mummies (The manin the ice 3); pp.83-92. Springer-Verlag, Wien,New York.

Zink A., Haas C.J., Reischl U., Szeimies U. &Nerlich, A.G. 2001. Molecular analysis ofskeletal tuberculosis in an ancient Egyptianpopulation. J. Med. Microbiol, 50: 355-366.

Zink A.R., Grabner W., Reischl U., Wolf H. &Nerlich A.G. 2003a. Molecular study onhuman tuberculosis in three geographicallydistinct and time delineated populations fromancient Egypt. Epidemiol. Infect., 130: 239-249.

Zink A.R., Sola C., Reischl U., Grabner W.,Rastogi N., Wolf H. & Nerlich, A.G. 2003b.Characterization of Mycobacterium tuberculosiscomplex DNAs from Egyptian mummies byspoligotyping. J. Clin. Microbiol., 41: 359-367.

Zink A.R., Sola C., Reischl U., Grabner W.,Rastogi N., Wolf H. & Nerlich, A.G. 2004.Molecular identification and characterizationof Mycobacterium tuberculosis complex inancient Egyptian mummies. Int. J.Osteoarchaeol, 14: 404-413.

Zink A.R., Grabner W. & Nerlich, A.G. 2005.Molecular identification of humantuberculosis in recent and historic bone tissuesamples: The role of molecular techniques forthe study of historic tuberculosis. Am. J. Phys.Anthropol., 126: 32-47.

64 Bacterial DNA in mummies