**1. Introduction**

Tuberculosis is an ancient disease. It was recognised and described by Hippocratus (460–390 BCE) and Galen (2nd–3rd century CE) in the western Classical World (Xarchus & Bourandas, 2003), ancient Egypt, India and the Far East (Morse, 1961). The obvious symptoms that attracted attention were the late outcomes of skeletal tuberculosis, where collapsed vertebrae led to scoliosis and Pott's disease, plus the symptoms associated with pulmonary tuberculosis, such as fever, weight loss and haemoptysis (coughing up blood). In the UK, tubercular lesions of the lymph glands (cervical lymphadenitis) were formerly termed scrofula, or the King's evil, and tubercular skin lesions were described as Lupus vulgaris or tuberculous chancre. The palaeopathology of ancient skeletal remains, together with classical and historical reports, demonstrate that tuberculosis occurred in prehistory. However, tuberculosis is still the greatest cause of death from any single infectious disease in the world today, with over one third of the global population infected and an estimated 1.7 million deaths from the disease in 2009 (WHO, 2010). Therefore it is essential to understand the nature of tuberculosis in the past: its distribution, spread and relationship to human society.

The disease is caused by members of a group of very closely related bacteria, termed the *Mycobacterium tuberculosis* complex (MTBC). These are obligate parasites and have the ability to subvert the cell-mediated immune system of the host and to survive and multiply within macrophages. Most human infections are caused by *Mycobacterium tuberculosis* and are usually acquired via the aerosol route from an active case of pulmonary tuberculosis. Infectious aerosols lodge in the alveoli but, in the majority of cases, the bacilli are controlled by the host immune system to form a granuloma and the disease remains latent. Infection can also occur by ingestion – milk or meat from an infected animal can give rise to human zoonotic cases of tuberculosis caused by *Mycobacterium bovis* or other members of the MTBC. In endemic areas, infection takes place in early life and may remain latent throughout a lifetime or become re-activated due to lowered host resistance caused by physical or mental stress, immunosuppression or extreme age (Rustad et al., 2009). Active primary tuberculosis, estimated to occur in 2–5% of cases, normally causes lymphadenitis and subsequent spread via the blood stream can cause meningitis or miliary tuberculosis (Grange & Zumla, 2009). Post-primary tuberculosis is estimated to occur in a similar proportion of people and these

Molecular Biomarkers for Ancient Tuberculosis 5

Within living cells, DNA is subjected to enzymatic repair processes, but this ceases after death. Thereafter, host DNA is rapidly degraded by enzymes derived both from the host and the macro and microbial flora that form part of the natural decay process (Pääbo et al., 2004). As a result of the cumulative changes over time (diagenesis) ancient DNA may develop hydrolytic and oxidative lesions. The breakdown of the N-glycosyl bond between the sugar and the base, in the presence of water, leads to hydrolytic cleavage and DNA fragmentation. Hydrolytic depurination causes a preferential loss of guanine and adenine, whereas the pyrimidines cytosine and thymine are 40-fold more susceptible to hydrolytic deaminisation (O'Rourke et al., 2000). Oxidative damage, especially to pyrimidines, can result in the formation of substances such as hydantoins, that block extension during PCR (Höss et al., 1996). DNA strands may also become chemically cross-linked due to Maillard products (Poinar et al., 1998), formed by condensation reactions between sugars and primary amino-groups in proteins and nucleic acids (Pääbo et al., 2004). Local environmental conditions have a strong impact on the persistence of aDNA, such as the temperature, the pH at the site, the availability of water and oxygen and fluctuations of all these factors over time (Poinar, 2003). Indeed, these factors outweigh the impact of the chronological age of samples. For example, a 200C decrease in temperature reduces base degradation 10- to 25-fold (Höss et al., 1996). Mycobacterial DNA is more robust than that of mammals, but its persistence depends not only upon the local environmental conditions but also the nature of the infection at the time of death of its host. Therefore, *M. tuberculosis* aDNA is often highly localized and DNA extraction protocols may have to be optimized for

In the early days of palaeomicrobiology, the criteria drawn up by researchers working on ancient mammalian DNA were recommended. These included preliminary screening tests, such as using the degree of amino acid racemization (Poinar & Stankiewicz, 1999) or collagen yield (Götherström et al., 2002), as an indication of the extent of DNA preservation. However, it appears that these are not reliable indicators, even of mammalian aDNA (Fernañdez et al, 2009; Kaestle & Horsburgh, 2002). More recently, Ottoni et al. (2009) discovered that there was better recovery of aDNA from animal bones that showed evidence of cooking, concluding that parameters based on protein diagenesis are not always useful for predicting ancient DNA survival. Work on other microbial pathogens, such as *Yersinia pestis* – the cause of bubonic plague – demonstrate that the dental pulp cavity in sound adult teeth is an excellent source of aDNA (Drancourt et al., 1998). It is believed that adsorption to hydroxyapatite increases the stability of aDNA (Götherström et al., 2002; Tuross, 1994) and any microorganisms present in the blood will potentially be present

For the examination of material for tuberculosis, the most common specimens available are bones. Most active cases of tuberculosis present as a lung disease, so ribs are a good source of *M. tuberculosis* DNA. Tuberculous lesions in the ribs arise by extension from spinal

**2. Ancient DNA (aDNA) from the** *M. tuberculosis* **complex (MTBC):** 

**Background and basics** 

specimens from different sites.

(Donoghue, 2008a).

**2.2 Selection of specimens and sampling** 

**2.1 DNA degradation and persistence** 

individuals in ancient and historical times would be the recognisable cases of skeletal tuberculosis. Therefore, it is highly significant that these historical cases, diagnosed by skeletal pathology, represent only around 5% of the total number of individuals with the disease.

Because of their very slow growth-rate and clinical significance, the MTBC was one of the first groups of microorganisms to benefit from the introduction of the polymerase chain reaction (PCR) and molecular diagnostics. This led to a an understanding of the epidemiology of tuberculosis (Reed et al., 2009; Smith et al., 2006), the evolution of the MTBC (Brosch et al., 2002; Ernst et al., 2007; Gordon et al., 2009; Gutierrez et al., 2005) and to the realisation that particular lineages of *M. tuberculosis* are associated with the country of origin of their human hosts (Hershberg et al., 2008; Hirsh et al. 2004; Wirth et al., 2008). Total sequencing of the genomes of *M. tuberculosis* (Cole et al., 1998), *M. bovis* (Garnier et al., 2003) and attenuated *M. bovis* Bacille Calmette-Guérin (BCG) vaccine strains (Pan et al., 2011; Seki et al., 2009) has elucidated the relationship between MTBC strain, lineage and pathogenicity. We now understand that the MTBC represents a clonal expansion of pathogenic strains or ecotypes (Smith et al., 2006) each of which is associated with a parallel clonal expansion of their mammalian hosts (Maiden, 2009).

The MTBC is distinct from the large number of environmental mycobacteria, which are generally non-virulent or cause opportunist infections in hosts, especially those with increased susceptibility. A characteristic feature of the mycobacteria is their cell envelope, which contains a high proportion of lipid-rich molecules, such as mycolic acids and phthiocerol dimycocerosate waxes (Minnikin, 1982; Minnikin et al., 2002). These result in a hydrophobic bacterial cell wall with decreased permeability and susceptibility to degradation, that may partially explain the very slow growth rate of the MTBC and persistence of viable organisms after the death of the host (Sterling et al., 2000; Weed & Baggenstoss, 1951). The mycobacteria are members of a taxonomic clade typified by organisms with a high percentage of guanidine and cytosine residues in their DNA. It is believed that the DNA of GC-rich bacteria is structurally more stable than that of other microbes because of the additional hydrogen bond cross-links between the DNA strands.

The wealth of information on the genomics of *M. tuberculosis* strains present in the world today, coupled with our understanding of the co-evolution of *M. tuberculosis* with its human host, has attracted interest in determining the origins and timescale of this relationship. Relevant information can be obtained from archaeology, anthropology and palaeopathology, which provide details on past human populations, societies and the occurrence of infectious diseases. The relative robustness of *M. tuberculosis* biomarkers enables the well-established molecular methods used in diagnostic clinical microbiology to be applied, with appropriate modification, to the study of historical and archaeological remains. Originally the emphasis was on the detection and characterisation of *M. tuberculosis*  ancient DNA (aDNA), as this enables the evolution of this group of pathogenic bacteria to be directly investigated by the detection and characterisation of their DNA (Fletcher et al., 2003a, 2000b; Matheson *et al.* 2009; Zink et al., 2001). However, it was soon appreciated that the unique lipid biomarkers found in the MTBC, in addition to enabling the independent verification of aDNA studies (Donoghue et al., 1998, 2010a; Gernaey et al., 2001; Hershkovitz *et al.,* 2008), have the potential to illuminate deep into human prehistory due to their particular stability (Gernaey & Minnikin, 2000; Redman *et al.,* 2009).

individuals in ancient and historical times would be the recognisable cases of skeletal tuberculosis. Therefore, it is highly significant that these historical cases, diagnosed by skeletal pathology, represent only around 5% of the total number of individuals with the

Because of their very slow growth-rate and clinical significance, the MTBC was one of the first groups of microorganisms to benefit from the introduction of the polymerase chain reaction (PCR) and molecular diagnostics. This led to a an understanding of the epidemiology of tuberculosis (Reed et al., 2009; Smith et al., 2006), the evolution of the MTBC (Brosch et al., 2002; Ernst et al., 2007; Gordon et al., 2009; Gutierrez et al., 2005) and to the realisation that particular lineages of *M. tuberculosis* are associated with the country of origin of their human hosts (Hershberg et al., 2008; Hirsh et al. 2004; Wirth et al., 2008). Total sequencing of the genomes of *M. tuberculosis* (Cole et al., 1998), *M. bovis* (Garnier et al., 2003) and attenuated *M. bovis* Bacille Calmette-Guérin (BCG) vaccine strains (Pan et al., 2011; Seki et al., 2009) has elucidated the relationship between MTBC strain, lineage and pathogenicity. We now understand that the MTBC represents a clonal expansion of pathogenic strains or ecotypes (Smith et al., 2006) each of which is associated with a parallel clonal expansion of

The MTBC is distinct from the large number of environmental mycobacteria, which are generally non-virulent or cause opportunist infections in hosts, especially those with increased susceptibility. A characteristic feature of the mycobacteria is their cell envelope, which contains a high proportion of lipid-rich molecules, such as mycolic acids and phthiocerol dimycocerosate waxes (Minnikin, 1982; Minnikin et al., 2002). These result in a hydrophobic bacterial cell wall with decreased permeability and susceptibility to degradation, that may partially explain the very slow growth rate of the MTBC and persistence of viable organisms after the death of the host (Sterling et al., 2000; Weed & Baggenstoss, 1951). The mycobacteria are members of a taxonomic clade typified by organisms with a high percentage of guanidine and cytosine residues in their DNA. It is believed that the DNA of GC-rich bacteria is structurally more stable than that of other microbes because of the additional hydrogen bond cross-links between the DNA strands. The wealth of information on the genomics of *M. tuberculosis* strains present in the world today, coupled with our understanding of the co-evolution of *M. tuberculosis* with its human host, has attracted interest in determining the origins and timescale of this relationship. Relevant information can be obtained from archaeology, anthropology and palaeopathology, which provide details on past human populations, societies and the occurrence of infectious diseases. The relative robustness of *M. tuberculosis* biomarkers enables the well-established molecular methods used in diagnostic clinical microbiology to be applied, with appropriate modification, to the study of historical and archaeological remains. Originally the emphasis was on the detection and characterisation of *M. tuberculosis*  ancient DNA (aDNA), as this enables the evolution of this group of pathogenic bacteria to be directly investigated by the detection and characterisation of their DNA (Fletcher et al., 2003a, 2000b; Matheson *et al.* 2009; Zink et al., 2001). However, it was soon appreciated that the unique lipid biomarkers found in the MTBC, in addition to enabling the independent verification of aDNA studies (Donoghue et al., 1998, 2010a; Gernaey et al., 2001; Hershkovitz *et al.,* 2008), have the potential to illuminate deep into human prehistory due to their

particular stability (Gernaey & Minnikin, 2000; Redman *et al.,* 2009).

disease.

their mammalian hosts (Maiden, 2009).

### **2. Ancient DNA (aDNA) from the** *M. tuberculosis* **complex (MTBC): Background and basics**

#### **2.1 DNA degradation and persistence**

Within living cells, DNA is subjected to enzymatic repair processes, but this ceases after death. Thereafter, host DNA is rapidly degraded by enzymes derived both from the host and the macro and microbial flora that form part of the natural decay process (Pääbo et al., 2004). As a result of the cumulative changes over time (diagenesis) ancient DNA may develop hydrolytic and oxidative lesions. The breakdown of the N-glycosyl bond between the sugar and the base, in the presence of water, leads to hydrolytic cleavage and DNA fragmentation. Hydrolytic depurination causes a preferential loss of guanine and adenine, whereas the pyrimidines cytosine and thymine are 40-fold more susceptible to hydrolytic deaminisation (O'Rourke et al., 2000). Oxidative damage, especially to pyrimidines, can result in the formation of substances such as hydantoins, that block extension during PCR (Höss et al., 1996). DNA strands may also become chemically cross-linked due to Maillard products (Poinar et al., 1998), formed by condensation reactions between sugars and primary amino-groups in proteins and nucleic acids (Pääbo et al., 2004). Local environmental conditions have a strong impact on the persistence of aDNA, such as the temperature, the pH at the site, the availability of water and oxygen and fluctuations of all these factors over time (Poinar, 2003). Indeed, these factors outweigh the impact of the chronological age of samples. For example, a 200C decrease in temperature reduces base degradation 10- to 25-fold (Höss et al., 1996). Mycobacterial DNA is more robust than that of mammals, but its persistence depends not only upon the local environmental conditions but also the nature of the infection at the time of death of its host. Therefore, *M. tuberculosis* aDNA is often highly localized and DNA extraction protocols may have to be optimized for specimens from different sites.

#### **2.2 Selection of specimens and sampling**

In the early days of palaeomicrobiology, the criteria drawn up by researchers working on ancient mammalian DNA were recommended. These included preliminary screening tests, such as using the degree of amino acid racemization (Poinar & Stankiewicz, 1999) or collagen yield (Götherström et al., 2002), as an indication of the extent of DNA preservation. However, it appears that these are not reliable indicators, even of mammalian aDNA (Fernañdez et al, 2009; Kaestle & Horsburgh, 2002). More recently, Ottoni et al. (2009) discovered that there was better recovery of aDNA from animal bones that showed evidence of cooking, concluding that parameters based on protein diagenesis are not always useful for predicting ancient DNA survival. Work on other microbial pathogens, such as *Yersinia pestis* – the cause of bubonic plague – demonstrate that the dental pulp cavity in sound adult teeth is an excellent source of aDNA (Drancourt et al., 1998). It is believed that adsorption to hydroxyapatite increases the stability of aDNA (Götherström et al., 2002; Tuross, 1994) and any microorganisms present in the blood will potentially be present (Donoghue, 2008a).

For the examination of material for tuberculosis, the most common specimens available are bones. Most active cases of tuberculosis present as a lung disease, so ribs are a good source of *M. tuberculosis* DNA. Tuberculous lesions in the ribs arise by extension from spinal

Molecular Biomarkers for Ancient Tuberculosis 7

inhibitors (Donoghue, 2008a). Hot-start PCR and excess enzyme can also drive the reaction and overcome residual inhibitors. Detection of amplicon has traditionally been based on agarose gel electrophoresis. Detection is also possible by hybridization of labelled amplicons to a membrane, using a dot block technique, for example. However, real-time PCR enables amplified product with an incorporated fluorescent marker or probe to be monitored directly via a computer screen and the methodology facilitates quantification. For the future, *M. tuberculosis* diagnostics is moving towards isothermal and array technology. Microarrays are not ideal for the direct detection of ancient *M. tuberculosis* in crude extracts, due to the extensive fragmentation of target sequences. However, the introduction of new platforms based on surface interactions and nanotechnology offer exciting possibilities for the future.

In recent years, the field of molecular diagnostics of tuberculosis has expanded rapidly and a wide variety of new techniques have been introduced, or are currently being validated for clinical use. This is beginning to have an impact on the specialized field of palaeomicrobiology (Section 3.2 below) and, no doubt, there will be many more studies in

Real-time PCR (RT-PCR) or, more correctly, quantitative PCR (qPCR) enables amplified product with an incorporated fluorescent marker or probe to be monitored directly via a computer screen. The underlying principle of non-specific double-stranded DNA binding dye chemistry is that fluorescent dyes such as SYBR Green intercalate with any double stranded DNA. This enables the progress of the amplification to be followed as it progresses, whereas conventional PCR relies upon the detection of amplicon once the reaction has completed. By use of standards and specific primers, the number of copies of amplicon or the absolute amount of DNA can be quantified. Normally a series of peaks are visible on the computer screen, so on completion a melt analysis is performed and the temperature at which strand separation occurs (Tm) is used to determine the targeted sequence. Greater clarity and specificity is conferred by the use of specific DNA probes, which incorporate a fluorescent reporter that is normally quenched (Nazarenko et al., 1997). The fluorescence is only released once the probe has bound to the specific target sequence.

Several isothermal target amplification methods, which avoid the use of a thermocycler machine, have been developed in the two past decades (Karami et al., 2011). Loop-mediated isothermal amplification (LAMP) is an isothermal molecular method of DNA amplification that has been successfully implemented in the detection of *M. tuberculosis* in clinical specimens (Notomi et al., 2000; Neonakis et al., 2011). The reaction is driven by outer primers leading to strand displacement DNA synthesis, production of a single-stranded template, further DNA synthesis initiated by additional primers, and hybridization to the other end of the target sequence to produce a loop. In subsequent cycles further strand displacement leads to multiple copies of the target sequence. LAMP has several advantages,

**2.4.1 Real-time PCR methods of MTBC aDNA detection and quantification** 

**2.4.2 Other methods of MTBC aDNA detection in liquid systems** 

such as rapidity, high sensitivity, ease of application and cost-effectiveness.

**2.4 Further developments in molecular diagnosis** 

the future that will be based on such technology.

lesions, from haematogenous spread from some remote soft tissue focus, or by direct spread from disease in the lungs, pleura, or chest wall lymphatic system (Mays et al., 2002). Initially, only bones with lesions were examined (Spigelman & Lemma, 1993), but it is now clear that in the majority of ancient cases of tuberculosis there are no lesions, but *M. tuberculosis* aDNA is present due to haematologous spread or by direct contact with infected tissue (Donoghue, 2011; Donoghue et al., 2011). Mummified tissue (Salo et al., 1994), skin (Faerman et al., 1997; Konomi et al., 2002), dental pulp (Faerman et al, 1997; Matheson et al, 2009) and calcified pleura (Donoghue et al., 1998) have also yielded MTBC aDNA.

Unfortunately, many published protocols based on human and animal aDNA research recommend that bones are pre-treated with bleach, ultraviolet light, or the outer bone surface is removed. The aim is thereby to remove surface contamination, but such procedures may inadvertently remove the very *M. tuberculosis* aDNA that is being sought (Donoghue, 2008a).

### **2.3 MTBC aDNA extraction and detection by conventional PCR**

The amount of material examined by different investigators varies greatly (Donoghue et al., 2009), but the subsequent extraction procedures follow a similar pattern. Mineralized tissue is powdered and demineralized, enzymes are used to remove proteins, the samples are disaggregated with agents such as phenol-chloroform or guanidium thiocyanate, and DNA is captured by silica, on to filters or membranes, or simply precipitated with isopropanol. Due to the persistent mycobacterial cell wall, robust techniques such as bead beating, freezethaw cycles in liquid nitrogen, and incubation for longer time periods and at temperatures such as 560C are often used. The reagent N-phenacylthiazolium bromide may be used to overcome the problem of Maillard products and enable strand separation (Pääbo et al., 2004). Reagents used for DNA capture may need modification to allow for DNA fragments that are <200 base pairs (bp) in length. In poorly preserved samples the fragment length may be less than 100 bp. Experience has shown that aDNA is unstable in aqueous solution so extraction preparations are best stored as dried silica or precipitates and only reconstituted immediately prior to examination. Thereafter, the aDNA extracts should be aliquoted and stored at -800C, if possible, so the number of freeze-thaw cycles can be minimized.

DNA amplification by PCR enables targeted and specific recovery of informative genetic loci. Amplification of MTBC aDNA is usually based on MTBC-specific regions of repetitive sequences in the genome of all members of the complex, such as IS*6110* (Eisenach et al, 1990). The sensitivity of conventional PCR can be increased by further amplification of the amplified PCR product via a nested reaction (Taylor et al., 1996). The primers devised by these two primer sets give rise to amplicons of 123 bp and 92 bp, respectively. IS*6110* may have up to 24 copies/cell in *M. tuberculosis* (Tanaka et al., 2000), although a small percentage of strains have no copies at all and *M. bovis* strains have only a single copy per cell. The alternative specific PCR, based on IS*1081* (Taylor et al., 2005), is preferable in most cases, as there are six copies/cell in each member of the MTBC so quantification is possible. Optimization of the PCR is necessary and modification of the PCR reaction mix is recommended for work with aDNA. Inclusion of stabilizers such as bovine serum albumin is often beneficial, probably due to a variety of effects such as masking non-specific binding sites, stabilizing DNA fragments and binding or otherwise inactivating co-purified PCR

lesions, from haematogenous spread from some remote soft tissue focus, or by direct spread from disease in the lungs, pleura, or chest wall lymphatic system (Mays et al., 2002). Initially, only bones with lesions were examined (Spigelman & Lemma, 1993), but it is now clear that in the majority of ancient cases of tuberculosis there are no lesions, but *M. tuberculosis* aDNA is present due to haematologous spread or by direct contact with infected tissue (Donoghue, 2011; Donoghue et al., 2011). Mummified tissue (Salo et al., 1994), skin (Faerman et al., 1997; Konomi et al., 2002), dental pulp (Faerman et al, 1997; Matheson et al,

Unfortunately, many published protocols based on human and animal aDNA research recommend that bones are pre-treated with bleach, ultraviolet light, or the outer bone surface is removed. The aim is thereby to remove surface contamination, but such procedures may inadvertently remove the very *M. tuberculosis* aDNA that is being sought

The amount of material examined by different investigators varies greatly (Donoghue et al., 2009), but the subsequent extraction procedures follow a similar pattern. Mineralized tissue is powdered and demineralized, enzymes are used to remove proteins, the samples are disaggregated with agents such as phenol-chloroform or guanidium thiocyanate, and DNA is captured by silica, on to filters or membranes, or simply precipitated with isopropanol. Due to the persistent mycobacterial cell wall, robust techniques such as bead beating, freezethaw cycles in liquid nitrogen, and incubation for longer time periods and at temperatures such as 560C are often used. The reagent N-phenacylthiazolium bromide may be used to overcome the problem of Maillard products and enable strand separation (Pääbo et al., 2004). Reagents used for DNA capture may need modification to allow for DNA fragments that are <200 base pairs (bp) in length. In poorly preserved samples the fragment length may be less than 100 bp. Experience has shown that aDNA is unstable in aqueous solution so extraction preparations are best stored as dried silica or precipitates and only reconstituted immediately prior to examination. Thereafter, the aDNA extracts should be aliquoted and stored at -800C, if possible, so the number of freeze-thaw cycles can be

DNA amplification by PCR enables targeted and specific recovery of informative genetic loci. Amplification of MTBC aDNA is usually based on MTBC-specific regions of repetitive sequences in the genome of all members of the complex, such as IS*6110* (Eisenach et al, 1990). The sensitivity of conventional PCR can be increased by further amplification of the amplified PCR product via a nested reaction (Taylor et al., 1996). The primers devised by these two primer sets give rise to amplicons of 123 bp and 92 bp, respectively. IS*6110* may have up to 24 copies/cell in *M. tuberculosis* (Tanaka et al., 2000), although a small percentage of strains have no copies at all and *M. bovis* strains have only a single copy per cell. The alternative specific PCR, based on IS*1081* (Taylor et al., 2005), is preferable in most cases, as there are six copies/cell in each member of the MTBC so quantification is possible. Optimization of the PCR is necessary and modification of the PCR reaction mix is recommended for work with aDNA. Inclusion of stabilizers such as bovine serum albumin is often beneficial, probably due to a variety of effects such as masking non-specific binding sites, stabilizing DNA fragments and binding or otherwise inactivating co-purified PCR

2009) and calcified pleura (Donoghue et al., 1998) have also yielded MTBC aDNA.

**2.3 MTBC aDNA extraction and detection by conventional PCR** 

(Donoghue, 2008a).

minimized.

inhibitors (Donoghue, 2008a). Hot-start PCR and excess enzyme can also drive the reaction and overcome residual inhibitors. Detection of amplicon has traditionally been based on agarose gel electrophoresis. Detection is also possible by hybridization of labelled amplicons to a membrane, using a dot block technique, for example. However, real-time PCR enables amplified product with an incorporated fluorescent marker or probe to be monitored directly via a computer screen and the methodology facilitates quantification. For the future, *M. tuberculosis* diagnostics is moving towards isothermal and array technology. Microarrays are not ideal for the direct detection of ancient *M. tuberculosis* in crude extracts, due to the extensive fragmentation of target sequences. However, the introduction of new platforms based on surface interactions and nanotechnology offer exciting possibilities for the future.

#### **2.4 Further developments in molecular diagnosis**

In recent years, the field of molecular diagnostics of tuberculosis has expanded rapidly and a wide variety of new techniques have been introduced, or are currently being validated for clinical use. This is beginning to have an impact on the specialized field of palaeomicrobiology (Section 3.2 below) and, no doubt, there will be many more studies in the future that will be based on such technology.

#### **2.4.1 Real-time PCR methods of MTBC aDNA detection and quantification**

Real-time PCR (RT-PCR) or, more correctly, quantitative PCR (qPCR) enables amplified product with an incorporated fluorescent marker or probe to be monitored directly via a computer screen. The underlying principle of non-specific double-stranded DNA binding dye chemistry is that fluorescent dyes such as SYBR Green intercalate with any double stranded DNA. This enables the progress of the amplification to be followed as it progresses, whereas conventional PCR relies upon the detection of amplicon once the reaction has completed. By use of standards and specific primers, the number of copies of amplicon or the absolute amount of DNA can be quantified. Normally a series of peaks are visible on the computer screen, so on completion a melt analysis is performed and the temperature at which strand separation occurs (Tm) is used to determine the targeted sequence. Greater clarity and specificity is conferred by the use of specific DNA probes, which incorporate a fluorescent reporter that is normally quenched (Nazarenko et al., 1997). The fluorescence is only released once the probe has bound to the specific target sequence.

#### **2.4.2 Other methods of MTBC aDNA detection in liquid systems**

Several isothermal target amplification methods, which avoid the use of a thermocycler machine, have been developed in the two past decades (Karami et al., 2011). Loop-mediated isothermal amplification (LAMP) is an isothermal molecular method of DNA amplification that has been successfully implemented in the detection of *M. tuberculosis* in clinical specimens (Notomi et al., 2000; Neonakis et al., 2011). The reaction is driven by outer primers leading to strand displacement DNA synthesis, production of a single-stranded template, further DNA synthesis initiated by additional primers, and hybridization to the other end of the target sequence to produce a loop. In subsequent cycles further strand displacement leads to multiple copies of the target sequence. LAMP has several advantages, such as rapidity, high sensitivity, ease of application and cost-effectiveness.

Molecular Biomarkers for Ancient Tuberculosis 9

8. biochemical preservation – use indirect assessment of the extent of DNA preservation by assessing the amount of diagenic change in other biomolecules, such as amino acids

9. associated remains such as those of animals can be used to check for comparable aDNA

10. phylogenetic sense – sequences should be compared with others in appropriate

Several of these recommendations have been accepted by palaeomicrobiologists, but a few

1. The suggestion of a separate building and clean rooms with filtered air supplies may be advisable for work on human aDNA, where every investigator is a potential source of contamination with modern DNA. However, the MTBC has no environmental reservoir, so provided that researchers do not suffer from tuberculosis it is quite possible to perform palaeomicrobiological research by following good microbiological practice, including plentiful negative controls and independent verification (Donoghue

2. Findings may not be reproducible because aDNA from tubercle bacilli will be localized. Even repeat samples from the same specimen may not yield a positive result in every case. Therefore, an additional criterion for work on a DNA of pathogenic microorganisms is proposed – that samples should be taken from sites appropriate to what is known of the natural history of the infection (Donoghue & Spigelman, 2006). 3. There is no evidence that cloning is necessary for verification of mycobacterial aDNA. Indeed, the opposite is true as work on *Mycobacterium leprae* (Taylor et al., 2006) showed that cloning gave no added value to data obtained by direct sequencing, but did introduce some errors, which were ascribed to *Taq* polymerase error and slipped strand mispairing. Similar conclusions have been reached in a recent study of mammalian

4. Independent replication of aDNA may give discordant results due to localization of pathogen biomarkers within samples (see point (5) above). MTBC-specific lipid biomarkers may be more sensitive and can verify aDNA data without the need for amplification (Section 4 below) even though determined sceptics (Wilbur et al., 2009)

5. Comparison with different host biomolecular markers is discussed above (Section 2.2) and the conclusion is that aDNA can be found even in samples where other

6. Comparison with the recovery of aDNA from associated faunal remains is inappropriate for MTBC aDNA for at least two reasons. First, mycobacterial DNA appears to be more robust than mammalian DNA (Section 2.1). In addition, faunal remains are often a poorer source of aDNA than associated human remains (Mays et al, 2001), possibly due to treatment of carcasses after death and the absence of burial

The earliest molecular studies on aDNA of the MTBC demonstrated proof of principle, but also answered historical questions about the occurrence of tuberculosis in the pre-colonial

or lipids;

are problematic:

et al., 2009).

survival to human DNA;

databases to ensure authenticity.

aDNA (Winters et al., 2010).

biomolecules are damaged.

(Taylor et al., 2010).

may ignore this (Donoghue et al., 2009).

**3. Palaeomicrobiology of tuberculosis** 

**3.1 Early studies 1993–2002 and initial conclusions** 

The change of scale by the use of nanoparticles reduces the need for multiple rounds of DNA amplification. For example, direct examination of clinical samples has successfully demonstrated *M. tuberculosis* DNA after an initial round of PCR, using a colorimetric method based on an *M. tuberculosis* probe linked to gold nanoparticles (Baptista et al., 2006).

#### **2.4.3 Detection of non-amplified DNA**

An alternative approach to conventional PCR is to directly detect non-amplified DNA by amplification of the detection system such as labelled probes (Bhatt et al., 1999). However, the technology has now been developed to enable direct detection of sequences in nonamplified genomic DNA by means of various sensors. In one example, a piezoelectric biosensor enables real-time and label-free detection of the hybridization reaction between an immobilized probe and the complementary sequence in solution. The DNA probe is immobilized on the sensing surface (10 MHz quartz crystals), while the complementary sequence is present in the genomic DNA, previously fragmented with restriction enzymes (Minunni et al., 2005). This approach has been developed for the detection of *M. tuberculosis* (Kaewphinit et al., 2010). Another specific DNA detection method uses fluorescent semiconductor quantum dots and magnetic beads for fast detection of mycobacteria without any DNA amplification. Two biotinylated oligonucleotide probes are used to recognize and detect specific complementary mycobacterial target DNA through a sandwich hybridization reaction. Quantum dots conjugated with streptavidin and specific probes are used to produce a fluorescent signal. Magnetic beads, conjugated with streptavidin and a genusspecific probe are used to isolate and concentrate the DNA targets (Gazouli et al., 2010). Surface primer extension reactions may also be used to quantitatively detect unamplified, double-stranded genomic DNA (Martins et al., 2010). This methodology, by eliminating the need for pre-target labeling or amplification procedures, constitutes an alternative for the direct detection of genomic DNA from solution.

#### **2.5 Authentication and precautions**

Lists of precautions to take when working with mammalian aDNA have dominated palaeomicrobiology even though the recommendations may not be appropriate. For example, this is a summary of the "top ten list" drawn up by Poinar (2003):


The change of scale by the use of nanoparticles reduces the need for multiple rounds of DNA amplification. For example, direct examination of clinical samples has successfully demonstrated *M. tuberculosis* DNA after an initial round of PCR, using a colorimetric method based on an *M. tuberculosis* probe linked to gold nanoparticles (Baptista et al., 2006).

An alternative approach to conventional PCR is to directly detect non-amplified DNA by amplification of the detection system such as labelled probes (Bhatt et al., 1999). However, the technology has now been developed to enable direct detection of sequences in nonamplified genomic DNA by means of various sensors. In one example, a piezoelectric biosensor enables real-time and label-free detection of the hybridization reaction between an immobilized probe and the complementary sequence in solution. The DNA probe is immobilized on the sensing surface (10 MHz quartz crystals), while the complementary sequence is present in the genomic DNA, previously fragmented with restriction enzymes (Minunni et al., 2005). This approach has been developed for the detection of *M. tuberculosis* (Kaewphinit et al., 2010). Another specific DNA detection method uses fluorescent semiconductor quantum dots and magnetic beads for fast detection of mycobacteria without any DNA amplification. Two biotinylated oligonucleotide probes are used to recognize and detect specific complementary mycobacterial target DNA through a sandwich hybridization reaction. Quantum dots conjugated with streptavidin and specific probes are used to produce a fluorescent signal. Magnetic beads, conjugated with streptavidin and a genusspecific probe are used to isolate and concentrate the DNA targets (Gazouli et al., 2010). Surface primer extension reactions may also be used to quantitatively detect unamplified, double-stranded genomic DNA (Martins et al., 2010). This methodology, by eliminating the need for pre-target labeling or amplification procedures, constitutes an alternative for the

Lists of precautions to take when working with mammalian aDNA have dominated palaeomicrobiology even though the recommendations may not be appropriate. For

1. a physically isolated work area, preferably a separate building where no genetic work is

2. PCR control amplifications, including non-template PCRs, multiple DNA and

3. molecular behaviour i.e. an inverse relationship between amount of PCR amplicon (bp)

5. reproducibility – results should be repeatable from both the same and different DNA

6. clone – direct sequencing should be confirmed by cloning amplicons and sequencing at least 10 clones to check for damage-induced errors and the ratio of endogenous to

7. independent replication – preferably by the independent examination of separate

example, this is a summary of the "top ten list" drawn up by Poinar (2003):

4. quantification – the copy number of DNA should be assessed;

samples of the same specimen in independent laboratories;

**2.4.3 Detection of non-amplified DNA** 

direct detection of genomic DNA from solution.

**2.5 Authentication and precautions** 

and length of target sequence;

carried out;

extraction controls;

extracts of a specimen;

exogenous sequences;


Several of these recommendations have been accepted by palaeomicrobiologists, but a few are problematic:


#### **3. Palaeomicrobiology of tuberculosis**

#### **3.1 Early studies 1993–2002 and initial conclusions**

The earliest molecular studies on aDNA of the MTBC demonstrated proof of principle, but also answered historical questions about the occurrence of tuberculosis in the pre-colonial

Molecular Biomarkers for Ancient Tuberculosis 11

BP, placing the remains within the Iron Age period. Further work on the same specimens used qPCR to detect, quantify and characterize the *M. bovis* DNA (Murphy et al., 2009).

The use of qPCR with specific fluorescent reporters should enable the detection of highly fragmented aDNA. This was demonstrated by the detection of a 63 bp *IS6110* target sequence specific for the MTBC in a pre-Hispanic (900–1100 CE) adult from the north coast of Peru (Klaus et al., 2010). Both conventional and qPCR were used to examine skeletal material from western Hungary with palaeopathology suggestive of tuberculosis (Évinger et al., 2011). Samples were dated from 800–1200 CE and the qPCR with a specific 75 bp IS*6110* target sequence was positive in six cases including two from the 9th century, whereas conventional PCR was negative. However, conventional PCR with a 113 bp target sequence for IS*1081* was positive in two of these cases plus one other, but a qPCR probe with a 72 bp target sequence was negative, thus demonstrating the lack of consistency when seeking

There has been no systemic examination of archaeological or historical material for coinfections, but our current understanding is that a pre-existing infection can increase susceptibility to another. A recent historical example is the influenza pandemic of 1918 where a major cause of death is believed to have been secondary bacterial pneumonia

For example, parallel developments in the molecular detection of *M. tuberculosis* and *M. leprae* aDNA enabled co-infected individuals to be identified. These were cases of lepromatous leprosy with very typical palaeopathology, who were subsequently discovered to have systemic *M. tuberculosis* aDNA in their skeletal remains (Donoghue et al., 2005). An extensive literature search revealed that such co-infections had been reported in historical times prior to the introduction of chemotherapy; the findings led to a hypothesis that tuberculosis might have been a major factor in the elimination of leprosy from Western

An example of an association of tuberculosis with a parasite infection comes from precolonial northern Peru, where Chaga's disease, caused by the protozoan parasite *Trypanosoma cruzii*, was widespread (Aufderheide et al., 2004). Palaeopathology and aDNA analysis demonstrated both Chaga's disease and tuberculosis in the population and one 12 year-old girl from 910-935 BP was shown to have a co-infection (Arriaza et al., 2008). Another such association between tuberculosis and *Leishmania* spp. infection, possibly also linked to nutritional stress, was reported in preliminary data from early Christian Nubia

There are many examples of increased susceptibility to infection associated with poor nutrition, a compromised immune system e.g. in neonates or the elderly, physical or mental stress due to wars and relocation, and underlying other medical conditions. An example of an association of tuberculosis with reduced lung function due to a massive vertebral

(Morens et al., 2008). It is very likely that additional examples will be found.

aDNA from microbial pathogens in human tissue.

**3.3.1 Co-infections** 

Europe.

(Spigelman et al., 2005).

**3.3.2 Co-morbidity** 

**3.3 Association of tuberculosis with other diseases** 

Far East (Spigelman & Lemma, 1993) and whether tuberculosis occurred in the Americas before Columbus (Salo et al., 1994). During the first decade of such research it was demonstrated that MTBC aDNA could be found in bone and mummified tissue, from body sites in specimens without lesions and of a broad age range, from locations around the world (Donoghue, 2011). Additional methods of examination included pathology, microscopy and radiology. Authentication was provided by the direct detection of MTBCspecific cell wall lipid markers (Donoghue et al., 1998, Gernaey et al., 2001). Use of additional PCR target sites, including *rpoB*, *mtp40*, *oxyR* and spoligotyping – which uses a dot-blot method based on the MTBC Direct Repeat (DR) region (Kamerbeek et al., 1997), enabled confirmation of the principal human pathogen *M. tuberculosis sensu stricto* (Taylor et al., 1999). The oldest confirmed case of tuberculosis was reported in a Pleistocene bison (17,870 BP) from the Natural Trap Cave, Wyoming, USA. A metacarpal showed suggestive pathology and spoligotyping indicated that the infecting organism was a member of the MTBC, but the species was not confirmed at the time (Rothschild et al., 2001).

#### **3.2 Recent findings and increased understanding**

The increased understanding arising from total genome sequencing of *M. tuberculosis* (Cole et al., 1998) led to an appreciation that this group of organisms exhibits sequential deletions that can be used to distinguish between strains and lineages. Therefore, molecular typing protocols were developed based on a combination of synonymous single nucleotide polymorphisms (SNPs) in the *katG* codon 463 (*katG*463), *gyrA* codon 95 (*gyrA*95) and deletions (Brosch et al., 2002). The TbD1 deletion was identified as specific to the human pathogen *M. tuberculosis* and a significant marker of "ancestral" and "modern" strains. Therefore, both SNP typing and deletion analysis have been incorporated into MTBC aDNA studies, provided that the DNA preservation was sufficiently good for such single-copy markers to be amplified and detected.

The next decade included population studies and early epidemiological findings. A welldocumented group of over 200 naturally mummified individuals from the 18th century was discovered in a church crypt in Vác, Hungary (Fletcher et al., 2003a). DNA preservation was particularly good and there was a high level of both active and presumed latent infections (Donoghue et al., 2011). It was possible to perform molecular fingerprinting and genotyping based on SNPs and to identify the *M. tuberculosis* aDNA as of "modern" strains. These techniques were used to demonstrate that in a small family group each person was infected with a different *M. tuberculosis* strain (Fletcher et al., 2003b). Interim epidemiological data have also been obtained from an on-going study of early Christian Nubians (550–750 and 750–1500 CE) and it is clear that tuberculosis was widespread, although there are no contemporaneous records and the DNA preservation is much less good (Donoghue, 2008b; Spigelman et al., 2005). Meanwhile, Zink, Nerlich and colleagues have produced a series of papers from a long-term study of burials in Thebes-West in ancient Egypt (Zink & Nerlich, 2004; Zink et al., 2003a, 2003b, 2004), spanning the pre-Dynastic period (5500–3100 BCE) to the New Kingdom (1550–1070 BCE). Molecular typing and spoligotyping indicated human *M. tuberculosis* and there was also evidence of another member of the MTBC, *Mycobacterium africanum*. However, no *M. bovis* was found. Indeed, there has only been one reported case of human tuberculosis associated with *M. bovis* aDNA (Taylor et al., 2007). This was found in a small group of pastoralists in south Siberia, dating from approximately 1761 to 2199 years BP, placing the remains within the Iron Age period. Further work on the same specimens used qPCR to detect, quantify and characterize the *M. bovis* DNA (Murphy et al., 2009).

The use of qPCR with specific fluorescent reporters should enable the detection of highly fragmented aDNA. This was demonstrated by the detection of a 63 bp *IS6110* target sequence specific for the MTBC in a pre-Hispanic (900–1100 CE) adult from the north coast of Peru (Klaus et al., 2010). Both conventional and qPCR were used to examine skeletal material from western Hungary with palaeopathology suggestive of tuberculosis (Évinger et al., 2011). Samples were dated from 800–1200 CE and the qPCR with a specific 75 bp IS*6110* target sequence was positive in six cases including two from the 9th century, whereas conventional PCR was negative. However, conventional PCR with a 113 bp target sequence for IS*1081* was positive in two of these cases plus one other, but a qPCR probe with a 72 bp target sequence was negative, thus demonstrating the lack of consistency when seeking aDNA from microbial pathogens in human tissue.
