(PDF) Structural characterization and decontamination of dental calculus for ancient starch research

Archaeological and Anthropological Sciences (2019) 11:4847–4872 https://doi.org/10.1007/s12520-019-00830-7 ORIGINAL PAPER Structural characterization and decontamination of dental calculus for ancient starch research María Soto 1 & Jamie Inwood 1 & Siobhán Clarke 1 & Alison Crowther 2,3 & Danielle Covelli 4 & Julien Favreau 1 & Makarius Itambu 1,5 & Steve Larter 6 & Patrick Lee 1,7 & Marina Lozano 8 & Jason Maley 4 & Aloyce Mwambwiga 1,9 & Robert Patalano 1 & Ramaswami Sammynaiken 4 & Josep M Vergès 8,10 & Jianfeng Zhu 4 & Julio Mercader 1 Received: 8 January 2019 / Accepted: 19 March 2019 / Published online: 11 April 2019 # Springer-Verlag GmbH Germany, part of Springer Nature 2019 Abstract Ancient dental calculus research currently relies on destructive techniques whereby archeological specimens are broken down to determine their contents. Two strategies that could partly remediate a permanent loss of the original sample and enhance future analysis and reproducibility include (1) structural surface characterization through spectroscopy along with crystallographic and spectroscopic analysis of its molecular structure, and (2) surface decontamination protocols in which the efficacy of cleaning dental calculus prior to extraction is demonstrated. Dental calculus provides ancient starch research a niche where granules may be adsorbed to minerals, coated, overgrown, entrapped, and/or protected from chemical degradation. While encapsulation offers protection from degradation, it does not shield the sample’s surface from contamination. The most common approach to retrieving microbotanical particles from archeological calculus has been the direct decalcification of the sample, after a cleaning stage variously consisting of immersion in water, acids, and mechanical dislodgment via gas, sonication, and/or toothbrushes. Little is known about the efficiency of these methods for a complete removal of sediment/soil and unrelated microbotanical matter. In this paper, controlled laboratory exper- imentation leads to chemical structural characterization and a decontamination protocol to eradicate starch granules. Several concentrations of acids, bases, and enzymes were tested at intervals to understand their potential to gelatinize and fully destroy starch granules; arriving at a procedure that effectively eradicates modern starch prior to dissolution without damaging the matrix or entrapped starch microremains. This is the first attempt at creating synthetic calculus to understand and systematically test effective decontamination protocols for ancient starch research. Keywords Structural chemical characterization . Raman . XPS . P-XRD . Ancient dental calculus . Ancient starch research . Decontamination prior to decalcification . Starch contamination * Julio Mercader 6 Petroleum Reservoir Group (PRG), Department of Geoscience,

University of Calgary, 2500 University Drive NW, Calgary, AB T2N 1N4, Canada 1 Department of Anthropology and Archeology, University of Calgary, 7 2500 University Drive NW, Calgary, AB T2N 1N4, Canada Department of Anthropology, University of Toronto, 19 Russell 2 Street, Toronto, ON M5S 2S2, Canada School of Social Science, The University of Queensland, St Lucia, QLD 4072, Australia 8 Institut Català de Paleoecologia Humana i Evolució Social (IPHES), 3 Department of Archeology, Max Planck Institute for the Science of Zona Educacional 4 – Campus Sesceslades URV (Edifici W3), Human History, 07745 Jena, Germany 43007 Tarragona, Spain 4 Saskatchewan Structural Sciences Centre, University of 9 National Natural History Museum, PO Box 2160, Arusha, Tanzania Saskatchewan, 110 Science Place, Saskatoon, SK S7N 5C9, Canada 5 10 Department of Archeology and Heritage Studies, University of Dar Àrea de Prehistòria, Universitat Rovira i Virgili, Avinguda de es Salam, PO Box 35050, Dar es Salaam, Tanzania Catalunya 35, 43002 Tarragona, Spain 4848 Archaeol Anthropol Sci (2019) 11:4847–4872 Introduction floatation and dispersal reagents. Because starch granules are pervasive in environments where calculus samples are Dental calculus analysis has a long tradition of research in excavated, curated, and processed, it is critical to conduct prehistory (see early work by Brothwell and Brothwell systematic pre-screening to characterize exogenous starch 1969; Armitage 1975; Dobney and Brothwell 1986). Often granules before decalcifying, and to consider effective de- containing plant phytoliths (Middleton and Rovner 1994; contamination strategies that will aid in establishing au- Lalueza-Fox et al. 1996; Henry and Piperno 2008; Dudgeon thenticity. While in vivo dental calculus forms in a shel- and Tromp 2014; Madella et al. 2014), starch granules (Hardy tered environment, there is no empirical basis to assume et al. 2009; Li et al. 2010; Wesolowski et al. 2010; Henry et al. that archeological calculus is less prone to contamination 2011; Mickelburgh and Pagan-Jimenez 2012; Buckley et al. than lithics, ceramics, or sediments. As such, ancient cal- 2014; Power et al. 2014, 2015, 2018; Cristiani et al. 2016, culus is subject to incidental starch contamination prior to 2018), DNA (Weyrich et al. 2015, 2017), microbiomics burial, while residing in entombing soil and sediment, dur- (Warinner et al. 2015, 2017), proteins (Hendy et al. 2018), ing excavation and sampling, and in the laboratory. We and chemical compounds amenable to identification through note that geochemical variables within the burial environ- fingerprinting (Hardy et al. 2012, 2015, 2016b; Warinner et al. ment, as well as the size of the dental calculus deposit 2014), archeological mineralized plaque offers insights into itself, might play a role in differential preservation, but at ancient human ecology, foodways, health, genomics, and present time, there is a lack of controlled experimentation palaeodiets. to show their precise impact on dental calculus. Ancient dental calculus research currently relies on de- In this article, we characterize and compare artificial structive techniques whereby archeological specimens are and archeological dental calculus to elucidate mineral and broken down to determine their contents. Two strategies elemental composition, surface chemistry governing ad- that could partly remediate a permanent loss of the orig- sorption, crystallinity, formation period, and constituency inal sample and enhance future analysis and reproducibil- of trapped organic materials. The lessons learned from ity include (1) structural surface characterization through this comparative characterization formed the basis to de- spectroscopy along with crystallographic and spectroscop- velop a decontamination protocol. Archeological samples ic analysis of its molecular structure, and (2) surface de- of known provenance and curation history served as a contamination protocols in which the efficacy of cleaning proof of concept for ancient starch research. dental calculus prior to extraction is demonstrated. In microbotanical research, direct decalcification of the sam- ple, after a cleaning stage variously consisting of soaking Materials and methods in water, immersion in acids, and mechanical dislodgment via gas, sonication, and/or toothbrushes is common Synthetizing calculus (Armitage 1975; Lalueza-Fox et al. 1996; Coil et al. 2003; Boyadjian et al. 2007; Huang et al. 2007; Henry and Piperno All the experimental variables to create artificial calculus 2008; Piperno and Dillehay 2008; Hardy et al. 2009, 2012; Li are presented in Table 1. We first collected unmodified et al. 2010; Wesolowski et al. 2010; Henry et al. 2011; Buckley potato starch (Red Mill) with the tip of a Pasteur pipette et al. 2014; Horrocks et al. 2014; Warinner et al. 2014; Power (volumetric, 3 cm) and added it to calcium chloride et al. 2015, 2016; Tao et al. 2015; Tromp and Dudgeon 2015; [CaCl2] (Fisher C79-500) previously prepared at five dif- Tromp et al. 2017). Yet, Weyrich et al. (2015, p. 120) have ferent concentrations (0.01 g, 0.1 g, 0.2 g, 0.5 g, and 1 g) noticed that microbotanical analysis of calculus “has been in 9 ml of water (15 ml tube). We vortexed the prepara- performed (…) without rigorous decontamination or de- tions for 30 s. The volume was brought to 10 ml with calcification. Key questions remain about reproducibility previously boiled reverse osmosis deionized (RODI) wa- and accuracy of results, as well as the difference between ter, and pH was adjusted to 7–8 using sodium hydroxide innate dental calculus particles and the levels and contri- [NaOH] (Home Hardware Canada #3226-431). The tube butions of decontamination”. Detailed records of starch was vortexed for 2 min. Ammonium phosphate contamination introduced at the time of excavation through [NH4H2PO4] (Fisher A684-500) at five different concen- personnel’s clothing, motion, wind action, digging imple- trations (0.01 g, 0.1 g, 0.2 g, 0.5 g, and 1 g) received ments, sample bags, gloves, and paper products (Mercader glycine (Sigma-Aldrich, G7126) (0.01 g, 0.1 g, 0.2 g, et al. 2017) expand on this point. Laboratory contamina- 0.5 g, and 1 g) and sodium hydroxide to aid in both tion reported by Crowther et al. (2014) shows that starch increasing the pH and binding properties of the eventual contamination originate from non-powdered and powdered precipitate with starch. examination gloves, centrifuge tubes, aluminum foil, mi- Secondly, we used calcium carbonate blocks (Acros- croscope slides, coverslips, pipettes, air circulation, and Organics, CAS: 471.34-1) as a substrate to facilitate calcium Archaeol Anthropol Sci (2019) 11:4847–4872 4849 Table 1 Parameters used in the production of calculus Calcium Calcium Starch (w/ Glycine Ammonium Amount of pH pH Final carbonate chloride v) added phosphate sodium (solution) (precipitation) weight (g) (w/v) (w/v) hydroxide (g) (0.05) added (ml) 2.64 0.001 0.001 Y 0.01 N/A 6.30 6.30 2.75 4.28 0.001 0.001 N 0.1 N/A 6.02 6.02 5.09 2.84 0.01 0.01 N 0.1 > 20.00 7.37 8.35 8.94 2.06 0.01 0.01 N 0.1 0.70 6.11 6.11 2.73 2.72 0.02 0.02 N 0.02 3.00 5.47 6.11 2.78 3.83 0.02 0.02 Y 0.02 0.01 6.29 6.92 3.93 5.81 0.05 0.05 N 0.05 N/A 6.40 7.00 5.89 5.42 0.05 0.05 Y 0.05 0.13 6.80 7.00 5.49 7.66 0.05 0.05 Y 0.05 0.60 7.99 6.80 8.36 9.58 0.05 0.05 Y 0.05 0.60 7.04 7.11 9.94 7.02 0.05 0.05 Y 0.05 0.90 7.21 6.75 7.20 9.48 0.05 0.05 Y 0.05 0.60 6.64 6.65 9.65 7.52 0.05 0.05 Y 0.05 0.60 6.40 6.69 7.65 4.01 0.05 0.05 Y 0.05 3.50 6.94 7.00 4.40 6.85 0.05 0.05 Y 0.05 3.00 6.25 7.00 7.89 6.84 0.05 0.05 Y 0.05 3.55 6.59 7.30 8.05 3.61 0.05 0.05 Y 0.05 3.50 6.50 7.00 4.16 9.17 0.05 0.05 Y 0.05 2.50 6.73 7.00 10.36 5.19 0.05 0.05 Y 0.05 2.75 7.20 7.00 6.81 3.88 0.10 0.10 Y 0.10 0.31 7.00 6.47 4.78 4.65 0.10 0.10 Y 0.10 0.31 7.00 6.84 4.84 4.09 0.10 0.10 Y 0.10 1.31 5.69 7.04 4.25 4.09 0.10 0.10 Y 0.10 0.31 7.00 6.63 4.24 4.09 0.10 0.10 Y 0.10 0.31 7.00 6.27 4.50 3.72 0.10 0.10 Y 0.10 1.00 6.45 6.45 3.96 2.80 0.10 0.10 Y 0.10 N/A 6.91 6.91 3.39 phosphate precipitation. These blocks were weighed, rinsed w/v concentration for all experiments going forward: structur- twice with water, and air-dried. Each mixture prepared during al characterization, comparison with actual dental calculus, step 1 was poured onto the carbonate blocks inside a 50 ml and decontamination. tube, then vortexed for 5 min. Tubes were stored upright for 24 h. The decantation of the supernatant left us with a precip- Archeological controls itate at the bottom of the tube, which was transferred to a Petrie dish for 72 h of air-drying along with the carbonate The controls used to optimize our decontamination protocol block. We recorded the mass of the precipitated material in consisted of two archeological dental calculus samples from grams, appearance, cohesiveness, and starch trapping capaci- mandibular bone. One tooth (Fig. 1a) came from an individual ty. Microscopic analysis of the artificial calculus was carried excavated at the site of El Mirador (Atapuerca, Burgos, Spain) out at 10–40× with two systems: Motic BA310E/Olympus dated to 4760–4200 BP (Vergès et al. 2016). Archaeobotanical BX-51. studies of the occupation layer suggest a mosaic landscape with Several solutions of ammonium phosphate and calcium forested areas, crop fields, and pastures (Euba et al. 2016; chloride (0.1%, 1%, and 2%) did not yield cohesive precipi- Expósito and Burjachs 2016; Rodríguez et al. 2016). The hu- tates capable of trapping starch granules. Two solutions (5% man remains were exhumed in 2010, but the calculus for this and 10%) delivered high cohesion, trapping potato starch study was extracted in September 2018. The excavation proto- granules successfully. However, the 10% solution generated cols did not include anti-contamination controls and the re- excessive crystalline growth. Therefore, we selected the 5% mains were stored at the Institut Català de Paleoecologia 4850 Archaeol Anthropol Sci (2019) 11:4847–4872 Fig. 1 Dental sample from El Mirador. a Calculus in-situ. b–n Exogenous starch granules. o Calculus decontaminated after immersion in sodium hydroxide. p–u Exogenous phytoliths Humana i Evolució Social (Tarragona, Spain). Another tooth Archeological calculus from the two control sets was re- (Fig. 2a) came from a human skeleton dating to 810 BP from moved while wearing full body cleanroom coats, hairnets, the rockshelter site of Matangai Turu NW (Democratic masks, and starch-free gloves, via autoclaved dental picks Republic of the Congo) (Mercader 2002). The dental anthro- over oven-cleaned (400 °C) aluminum foil. The specimens pology was studied by Mercader et al. (2001), and the were placed in previously sterilized centrifuge tubes. By fol- microbotanical contents of the entombing sediment suggest lowing this procedure, we are certain that no modern starch densely forested environments (Mercader et al. 2000). The hu- was introduced by us during the brief time (approximately two man remains were excavated in 1993 following standard con- hours) during which the sample was removed from the tooth ditions at the time, but without the awareness that starch con- in uncontrolled spaces. Post-removal handling took place in a taminants are pervasive in the field (Mercader et al. 2017). As cleanroom (HEPA class H14) at the University of Calgary, such, dedicated excavation tools, isolated workspace, clean at- Earth Sciences Building, room no. 811, by placing 0.5 g of tire, and proven starch-free gloves and sample bags were not the extracted sample in a 15 ml autoclaved centrifuge tube used. In 1997, the remains were deposited at Universidad Complutense de Madrid, at the Department of Zoology and Fig. 2 Dental sample from Matangai Turu NW. (a) calculus in-situ, (b–„ Physical Anthropology’s collection. The dental materials were aw) exogenous starch granules, (ax) calculus decontaminated after curated under standard museum conditions, until sampled for immersion in sodium hydroxide, (ay) calculus with entrapped starch ancient calculus research in June 2017. granules (white arrows) (az-bc) exogenous phytoliths Archaeol Anthropol Sci (2019) 11:4847–4872 4851 4852 Archaeol Anthropol Sci (2019) 11:4847–4872 with 1 ml of previously boiled, RODI water. This was shaken (1486.6 eV) source and combined hemi-spherical analyzer at 90 rpm for 5 min, after which we extracted an aliquot of (HSA) and spherical mirror analyzer (SMA). A spot size of 0.15 ml containing both calculus fragments and aqueous so- 300 × 700 μm was used for both synthetic samples, as well as lution for microscopy at × 40 using an Olympus BX51. The the calculus from El Mirador (ATA10MIR201no.609: processing of control samples under cleanroom conditions ATA10MIR). The dental calculus from Matangai Turu NW allows us to infer that, should there be contaminants on these (MTNW) was too small to use such a large beam size, and control samples, it would come from burial sediments and/or thus a diameter slit of 110 μm was used. All survey scan excavation, handling, and curation prior to our work spectra were collected in the -5-1200 binding energy range (Crowther et al. 2014; Mercader et al. 2017; Mercader et al. in 1 eV steps with a pass energy of 160 eV. High-resolution 2018). scans of four regions were also conducted using 0.1 eV steps with a pass energy of 20 eV. An accelerating voltage of Structural characterization of synthetic 15 keV and an emission current of 15 mA were used for the and archeological dental calculus analysis when the large spot size was used, and an acceleration voltage of 15 keV with an emission current of 25 mAwas used Three methods determined structure and chemistry: Raman for the MTNW sample that used a 110-μm spot size. Data was spectroscopy (Raman), X-ray photoelectron spectroscopy analyzed using CasaXPS (Version 2.3.18PR1.0). (XPS), and powder X-ray diffraction (P-XRD). Raman de- P-XRD was utilized for the identification of crystalline livered structural information near the surface; XPS is also a phase and overall elucidation of 3-D atomic, geometric struc- surface technique penetrating to a depth of a few nanome- ture. The study was conducted at the Saskatchewan Structural ters, while XRD illuminated the geometry of the bulk ma- Sciences Centre on a Rigaku Ultima IV X-ray diffractometer terial but remained insensitive to surface characterization. equipped with a Cu source (1.54056 Å), a CBO optical, and a Raman and P-XRD were conducted at room conditions (at- Scintillation Counter detector. The diffractometer was operat- mospheric pressure, humidity, and temperature), but XPS ed at 40 kV and 44 mA. The measurements were carried out was executed at room temperature under ultra-high vacuum on the Multipurpose Attachment, with parafocusing mode. A (UHV) conditions, and therefore surface structures that Kβ filter (Ni foils) was placed at the receiving end. The syn- depended on the presence of water were changed. thetic samples were ground to fine powder (0.2 g). The Raman was carried out at the Saskatchewan Structural archeological samples were loosely loaded onto the sample Sciences Centre (University of Saskatchewan, Canada) on holder without any treatment. Then, 2θ was scanned from 3° a Renishaw InVia Reflex Raman microscope using a to 90° for all the samples (diffraction patterns were reported solid-state diode laser (Renishaw Inc.) operating at from 10° to 90°, since no sharp peaks were present but a broad 785 nm and a 1200 line/mm grating. The microscope feature originated from the sample holder). The scan rates was focused onto the sample using a Leica 20X NPLAN were 4° per minute (step size: 0.02°) for the synthetic samples, (NA = 0.40) objective, and the backscattered Raman sig- 1° per minute (step size: 0.02°) for the MTNW sample, and nals were collected with a Peltier cooled CCD detector. 0.1° per minute (step size: 0.01°) for the ATA10MIR sample. Measurements were collected using extended scan or stat- ic scan with a 5–10 s detector time. The laser power was Contamination and decontamination of synthetic 7.8 mW measured at the sample (archeological samples). calculus The instrument was calibrated using an internal Si (110) sample, which was measured at 520 cm−1. The lumines- To test the decontamination efficacy of different cleaning pro- cence profile of the different samples was collected using tocols and rid calculus surfaces of starch contaminants, we the Ar+ laser (Modulaser Stellar-Pro) operating at 514 nm. first proceeded to contaminate synthetic calculus previously The microscope was focused onto the sample using a precipitated on calcite blocks. The experimental contamina- Leica 20X NPLAN (NA = 0.40) objective, and the tion took place as follows: an aqueous solution of 1 ml con- backscattered luminescence signal were collected with a taining 0.02 g of cornstarch and 0.02 g glycine coated the Peltier cooled CCD detector. The laser power measured at calcium phosphate precipitate and allowed to dry for 24 h. the sample was 4.7 μW. We video-recorded and photographed the outcome of the con- XPS is a relative-quantitative spectroscopic method that tamination sequence at 10–40x via a Moticam 5+ and ana- characterizes chemical elemental composition, constituent ra- lyzed through Motic Images Plus 3.0 and Olympus SC50 tios, structure, and surface functional groups. All measure- (cellSens). ments were collected at the Saskatchewan Structural We experimented with four starch-gelatinizing agents Sciences Centre using a Kratos (Manchester, UK) AXIS previously used in the starch industry (Maher 1983; Supra system under UHV conditions. This system is equipped Yamada et al. 1986; Ragheb et al. 1995; Otsuka et al. with a 500-mm Rowland circle monochromated Al K-α 2001; Wang and Wang 2002): Archaeol Anthropol Sci (2019) 11:4847–4872 4853 & Sodium hydroxide [NaOH] (0.1%, 1%, and 2% w/v con- centration: Home Hardware Canada #3226-431). Then, 0.01 g, 0.1 g, and 0.2 g pellets were added to 9 ml of water (15 ml tubes) and vortexed (30 s). The volume was brought up to 10 ml with water. Vials were vortexed again for 30 s. & Calcium hydroxide [Ca(OH)2] (1%, 2%, and 5% w/v con- centration: Fisher Scientific Canada #C97-10). Then, 0.1 g, 0.2 g, and 0.5 g CaOH2 was added to 9 ml of water (15 ml tubes) and vortexed (30 s). The volume was brought to 10 ml with water. Vials were vortexed for 30 s. & EDTA [C10H14N2Na2O8·2H2O] (0.1%, 1%, and 2% w/v: disodium salt, dihydrate, crystal, J.T. Baker #4040). Then, 0.01 g, 0.1 g, and 0.2 g of EDTA were added separately to 9 ml of water (15 ml tubes) and vortexed (30 s). The Fig. 4 Raman peaks for synthetic and dental calculus (full width at half volume was brought up to 10 ml with water, and pH ad- maximum) justed to 8 using sodium hydroxide pellets. Vials were vortexed until the EDTA went into solution. & α-Amylase (0.1%, 1%, and 2% w/v concentration, 20 °C, Olympus SC50 (cellSens). Recording took place by scan- 500–1500 units/MG, Bacillus licheniformis, Sigma- ning 1 transect five times, repeated over 10 separate tran- Aldrich #A4551). Then, 0.01 g, 0.1 g, and 0.2 g amylase sects, totaling 50 scans per slide. The measurement of pH were added to 9 ml of water (15 ml tubes) and vortexed in all solutions was through a ThermoFisher Orion 320 (30 s). The volume was brought up to 10 ml with water, PerpHecT LogR meter, calibrated daily. The Ag/AgCl and pH adjusted to 7–8 using sodium hydroxide pellets. electrode was rinsed after calibration and in between Vials were vortexed for 2 min. samples. Experimentation with these four decontamination so- Unmodified potato (Red Mill) and corn starch (Bakers) lutions revealed that the best agent was sodium hydrox- were tested in sterilized 15 ml tubes, which received 1 ml ide (1 ml, 2% w/v), as it completely destroyed contam- of their respective decontaminating solutions. The results inant granules, while entrapped granules in the calculus were analyzed at time zero (T+0), one hour (T+1 h), six matrix were left intact. The precipitate was immersed hours (T+6 h), 24 hours (T+24 h), and seven days (T+ and left to evaporate over 24 h, then rinsed with water 7 d). Microscopy was carried out with a Motic BA310E/ (5 ml) and centrifuged 5 min (3000 rpm) three times. Olympus BX-51: 20–40×. Micrographs were collected The effectiveness of decontamination was confirmed using a Moticam 5+ (Motic Images Plus 3.0), and an through light microscopy. The original observation Fig. 3 Raman spectra for synthetic and archeological dental calculus Fig. 5 Comparison of Raman luminescence spectral profiles for dental from El Mirador and Matangai Turu NW calculus and synthetic samples 4854 Archaeol Anthropol Sci (2019) 11:4847–4872 Table 2 XPS elemental breakdown available online (doi: https://doi.org/10.31219/osf.io/ Element ATA10MIR MTNW Synthetic 6.84 Synthetic 6.85 b4fk6). (%) (%) (%) (%) C1s 19.3 58.4 13.6 37.1 O1s 45.4 21.7 53.1 17.6 Results Ca2p 14.1 10.2 14.2 2.8 P2p 6.8 6.1 13.4 2.8 Characterization: the structure of synthetic N1s 0.8 0.9 3.7 15.6 and dental calculus Si2p 10.7 1.3 N/A N/A Al2p 1.6 1.5 N/A N/A The Raman spectra for synthetic and dental calculus ap- Cl2p N/A N/A 1.8 20.7 pear in Fig. 3 with all samples showing a characteristic Na1s N/A N/A 0.2 3.0 ν3(P-O) symmetric stretch for the HPO42- hydroxyapatite S2p N/A N/A N/A 0.5 band at approximately 960 cm−1 (Koutsopoulos 2002). Two additional regions located at 370–460 cm −1 and 550–630 cm−1 correspond to the ν2(O-P-O) and ν4(O-P- matrix consisting of 5197 images is with the Federated O) bending modes of the PO43− group. The dental calcu- Research Data Repository (doi: https://doi.org/10.20383/ lus also showed a very broad peak (full width at half 101.0121). In addition, a preprint of this article has maximum, FWHM > 100 cm −1 ) centered at 982 cm −1 been uploaded to the Open Science Framework and is (Fig. 4). This broad feature combines vibrational bands Fig. 6 XPS survey spectra for dental and synthetic calculus. a El Mirador. b Matangai Turu NW. c 6.85. d 6.84 Archaeol Anthropol Sci (2019) 11:4847–4872 4855 from small molecules and perhaps brushite. The dental is 2:1–1:7 which could indicate hydroxyapatite. On the calculus generated a much larger luminescence spectral other hand, synthetic samples have Ca:P ratios close to profile compared to synthetic calculus (Fig. 5), also indi- 1 suggesting a di-calcium phosphate such as brushite cating differences in small molecule composition. The [CaHPO4 or CaHPO42H2O]. High-resolution XPS was synthetic calculus samples (Figs. 3, 4, and 5) showed collected for the following primary regions C1s (carbon) brushite with a ν3(P-O) symmetric stretch centered at (Fig. 7), O1s (oxygen) (Fig. 8), Ca2p (calcium) (Fig. 9), 987 cm−1. The FWHM for this peak was 8–16 cm −1. and P2p (phosphorus) (Fig. 10) (Table 3). In both The bending mode vibrations is located within the same archeological and synthetic samples, a peak around region as the hydroxyapatite. There is also a series of 284.8 eV indicates C-C/CH, and in one of the two syn- bands typical of carbohydrates derived from the trapped thetic samples (6.84), the C-O-C and/or C-N bonds are potato starch granules (Fig. 3), including a sharp band at very prominent (286.1 eV). (The other samples exhibit 478 cm−1 (ring skeletal mode), peaks at 860 cm−1, and this bond, but not to the same extent.) All materials show 940 cm−1 (C-O-C/C-O-H stretching vibrations for α-1,4 an abundance of carbon (288–289 eV), which is indica- glycosidic linkages) and a band at 2910 cm −1 (C-H tive of COOH groups. These organic functional groups stretching) (Mathlouthi and Koenig 1987; De Gelder are present in carbohydrates and proteins. A small amount et al. 2007). of CO3 also appears in one archeological (ATA10) and XPS survey of synthetic and actual dental calculus one synthetic sample (6.85). The O1s spectra of all sam- shows their elemental breakdown (Table 2): carbon, oxy- ples are fairly similar-essentially two peaks, with the peak gen, calcium, and phosphorus. Aluminum, silicon, nitro- at the lowest energy (~ 531 eV) being the most intense: gen, sodium, and chlorine appear in small quantities This is the oxygen that would be found in calcium phos- (Fig. 6). The Ca:P ratio for archeological dental calculus phate [Ca x(PO 4 ) y]. The second and less intense peak Fig. 7 XPS spectra for carbon (C1s). a El Mirador. b Matangai Turu NW. c 6.85. d 6.84 4856 Archaeol Anthropol Sci (2019) 11:4847–4872 Fig. 8 XPS spectra for oxygen (O1s). a El Mirador R. b Matangai Turu NW. c 6.85. d 6.84 (532–533 eV) represents oxygen attached to carbon in shows that our brushite-rich synthetic precipitates mimic CO, COOH, and COC. The phosphorus and calcium the early mineralization process that occurs in the oral spectra of all samples are essentially the same. They in- cavity prior to the stabilization of hydroxyapatite crystal- dicate one type of P and Ca [Cax(PO4)y]. lizations (Hayashizaki et al. 2008). P-XRD patterns of calculus appear in Fig. 11. The major peaks in synthetic calculus are very similar to each Experimental contamination of synthetic calculus other (Fig. 11a, b), and so are those from archeological and de-facto contamination of archeological calculus samples (Fig. 11c, d); thus, synthetic and archeological samples support different crystalline phases. The peaks in Synthetic brushite fragments (x ̅ 1.12 mm; n = 100) were ancient calculus are broad, which is likely due to a large deposited in sterilized petri dishes for contamination with crystal size. To identify these phases, the P-XRD pattern 1 ml of corn-glycine mix (2% w/v). Corn granules cov- was loaded to MDI Jade 2010, which performed baseline ered > 95% of the surface matrix across variable topog- corrections and background noise removal. The software raphies: elevated, flat, depressed, and fissured micro- then matched peaks against the incorporated database regions (Fig. 13). Except for crevices, we saw no evi- (COD, AMCSD, MDI-500) (Downs and Hall-Wallace dence that contaminant corn granules penetrated the 2003; Grazulis et al. 2009). A whole pattern fitting carbonatic matrix where the resident potato starch gran- (WPF) and a Rietveld Refinement determined the com- ule population are consistently distinguished from the position of each phase (Table 4, Fig. 12). These show adsorbed, overlying contaminants. that synthetic samples are brushite (Brown et al. 1962; As for the contamination potential of archeological Schroeder et al. 1977; Schofield et al. 2004), while the dental calculus by starch granules and other exogenous archeological calculus is hydroxyapatite (Sudarsanan and materials, in two samples we recorded 204 particles that Young 1969). In sum, the materials characterization are consistent with contamination, with 90% of these Archaeol Anthropol Sci (2019) 11:4847–4872 4857 Fig. 9 XPS spectra for calcium phosphate (Ca2p). a El Mirador. b Matangai Turu NW. c 6.85. d 6.84 contaminants being starch granules (Fig. 1 and 2). phosphate matrix. Even in cases where starch gran- Contamination was detected through microscopy of ules appear to adhere to the periphery of calculus two aliquots from each control: at Matangai Turu NW, Fig. 2(d–f, m, t, y, z, aa, ab, ag, ai, at, au, av, aw), we tallied 170 remains (starch granules, n = 166; this is an optical effect, with granules floating away phytoliths, n = 4), while we recorded 34 particles on from the calculus when a probe causes their dis- the sample from El Mirador, of which 18 are starch lodgment Fig. 2(at, au). granules, 15 are phytoliths, and 1 micro-remain is char- 3. The maize and wheat granules display morphomet- coal. Four lines of evidence support the exogenous na- rics unlike those from the starch granules genuinely ture of these particles: trapped in the dental calculus Fig. 2(ay), and opti- cally display the features that come from intact am- 1. Homogenous starch assemblage with morphometric ylopectin crystallites in pristine conditions: full bire- and textural qualities consistently overlapping with fringence and Maltese cross (e.g., Fig. 2 u, ad). The demonstrated field and lab contaminants such as starches from El Mirador display the typical damage maize and wheat Fig. 2(b–as, at) (Crowther et al. features from microbial decay that happen during 2014; Mercader et al. 2017) in the case of burial (Haslam 2004) (Fig. 1b–e): disruption of Matangai Turu NW. The calculus sample from El crystallites, centric implosion, pitting, and fissuring. Mirador was coated with granules (Fig. 1) resem- 4. Demonstrated transfer of other microbotanical particles bling natural starches from soil assemblages such as phytoliths Fig. 2(az-bc) (Fig. 1p–v) (Matangai (Mercader et al. 2017). Turu NW, n = 4; El Mirador, n = 15) from the entombing 2. Systematic lack of association with dental calculus: sediment to the calculus surface: they remain loose on the the starch granules are not trapped in the calcium surface, not trapped inside calculus. 4858 Archaeol Anthropol Sci (2019) 11:4847–4872 Fig. 10 XPS spectra for phosphorus. a El Mirador. b Matangai Turu NW. c 6.85. d 6.84 Decontamination effect from calcium hydroxide, & Potato starch granules: marked swelling and partial gela- sodium hydroxide, EDTA, and α-amylase tinization occurred even at low concentration. At 0.1% and 1% w/v, starch granule ghosts become apparent at Calcium hydroxide: this solution was tested using three T+24 h (1% w/v) and T+7 d (0.01% w/v). (The external concentrations (1%, 2%, and 5% w/v) (Table 5), with pH layers of starch granules form granule envelopes when 12.45–12.55: gelatinized, which in the field of carbohydrate polymers are referred to as “ghosts”: e.g., Atkin et al. 1998; Zhang & Potato starch granules (Fig. 14): All concentrations caused et al. 2014.) Complete gelatinization happened at 1% w/v clefting, radial fissuring, and implosion of the granules for and leaching at T+0 ≥ 2% w/v (Table 5). immersions up to 24 h. Implosion and disarticulation were & Corn starch granules: at the lowest concentration (0.1% w/ extreme in the longest immersion (T+7 d), noticing gela- v), granules underwent minor radial fissuring and gelati- tinization and leaching (Table 5). nization at T+7 d, while gelatinization took place at 1% w/ & Corn starch granules (Fig. 15): almost all granules v, and leaching at 2% w/v (T+0) (Table 5). (> 99%) remained intact while a very low number (0.81%, 49 granules; n = 6000) showed minor radial fissur- EDTA: mixes at 0.1%, 1%, and 2% w/v received 0.05– ing and implosion in the experiments that lasted 24 h/7 days 1.25 g of sodium hydroxide to stabilize pH at 8.0, given that (2–5% w/v) (Table 5). EDTA must be deprotonated to go into solution: & Potato starch granules (Fig. 16): 150 granules (11%, n = Sodium hydroxide: tested at 0.1%, 1%, and 2% w/v con- 1344) suffered radial fissuring and implosion in all con- centrations with pH 12.55–13.35: centrations >1 h (Table 5). Archaeol Anthropol Sci (2019) 11:4847–4872 4859 Percentage (%) Energy (eV) Assignment Percentage (%) Energy (eV) Assignment Percentage (%) 66.7 33.3 66.7 33.3 66.7 33.3 69.0 31.0 Cax(PO4)y Cax(PO4)y Cax(PO4)y Cax(PO4)y Cax(PO4)y Cax(PO4)y Cax(PO4)y Cax(PO4)y 133.3 134.1 133.1 133.9 132.9 133.8 133.1 134.0 P2p 67.2 32.8 67.1 32.9 67.9 32.1 66.9 33.1 Fig. 11 P-XRD patterns of synthetic and dental calculus. a Sample 6.84. b Sample 6.85, c MTNW, d El Mirador Cax(PO4)y Cax(PO4)y Cax(PO4)y Cax(PO4)y Cax(PO4)y Cax(PO4)y Cax(PO4)y Cax(PO4)y & Corn starch granules (Fig. 17): no damage was observed. Slight gelatinization occurred in 91 granules (n = 2000, 5%) that were immersed several hours (T+1 h, T+6 h) at the highest concentration (2% w/v) (Table 5). 347.5 347.4 347.1 Ca2p 350.9 347.3 351.1 350.6 350.9 α-amylase: the starch degradation potential of this enzyme was tested at 0.1%, 1%, and 2% w/v concentrations (pH 6.12– 6.72) at 20 °C. Higher temperatures (30–70 °C) conducive to 63.6 67.2 32.8 65.9 Na KLL Auger peak 8.1 36.4 34.1 61.4 30.5 maximum activity (Helbert et al. 1996; Goyal et al. 2005) were not used, given the structural changes heat induces to COOH/COC/SiOx COOH/COC/SiOx COOH/COC/SiOx COOH/COC/SiOx Cax(PO4)y, COH Cax(PO4)y, COH Cax(PO4)y, COH Cax(PO4)y, COH starch. Energy (eV) Assignment Percentage (%) Energy (eV) Assignment & Potato starch granules (Fig. 18): we noted radial fissuring, implosion, lamellae disruption, disarticulation, and loss of birefringence in 5–20 min (all concentrations). Although the structural damage seemed extreme, especially for the longest immersion times, no granule ghosts could be de- 530.8 531.1 536.2 531.6 533.0 532.5 531.4 532.4 532.6 O1s tected (Table 5). & Corn starch granules (Fig. 19): in our lowest concentration (0.1% w/v), granules were not affected except for T+24 h and T+7 d. Solutions at 1% and 2% w/v, however, led to High-resolution XPS results and fitting radial fissuring and disarticulation in 0.1% of the granules 63.3 8.8 1.8 16.1 6.4 21.2 68.0 18.9 1.0 46.1 12.4 34.6 11.1 26.1 44.1 19.9 (n = 1000). We did not notice granule ghosts (Table 5). COC/CN COC/CN COC/CN COC/CN C-C/CH C-C/CH C-C/CH C-C/CH COOH COOH COOH COOH COH CO3 CO3 CO Discussion 288.7 290.2 286.1 286.3 287.2 288.5 288.5 285.3 286.1 286.1 288.5 289.9 Synthetic 6.84 284.7 Synthetic 6.85 284.7 284.8 284.6 C1s This is the first attempt at creating synthetic calculus to under- stand and systematically test effective decontamination proto- ATA10MIR cols for ancient starch research. We understand the limitations MTNW Sample Table 3 of our microcosm, and the important differences between nat- ural, synthetic, and archeological calculus. Dental plaque is an 4860 Archaeol Anthropol Sci (2019) 11:4847–4872 Table 4 Diffraction peaks for brushite, hydroxyapatite, Brushite Hydroxyapatite Fluorapatite Whitlockite Octacalcium whitlockite, and octacalcium phosphate phosphate 2θ (°) Intensity 2θ (°) Intensity 2θ (°) Intensity 2θ (°) Intensity 2θ (°) Intensity 11.65 100.00 10.78 18.00 25.87 36.80 13.74 19.10 4.70 100.00 20.96 94.00 25.84 36.80 28.14 11.60 17.14 30.40 9.31 8.20 29.31 69.10 28.88 14.80 29.10 14.80 20.40 12.30 9.70 7.90 30.55 47.10 31.73 100.00 31.92 100.00 22.04 10.10 24.17 6.40 34.19 38.30 32.16 46.40 32.242 38.80 25.10 27.30 25.92 9.00 34.44 25.30 32.86 61.20 33.09 57.40 28.05 49.50 31.28 8.00 36.93 13.30 34.03 23.40 34.13 26.60 29.92 13.00 33.37 5.30 37.14 12.80 39.76 21.20 40.02 21.90 31.31 100.00 41.60 18.20 46.67 29.90 46.86 28.50 32.75 17.50 42.08 16.10 48.06 12.90 48.25 13.30 34.68 65.20 48.53 15.00 49.47 32.60 49.54 35.10 35.91 13.70 50.20 12.00 50.46 17.40 50.74 16.40 41.47 11.30 50.27 13.20 51.24 11.80 51.55 13.70 42.07 15.70 52.06 12.10 52.27 13.70 47.41 22.20 53.20 14.80 53.14 16.90 48.44 17.40 48.86 14.10 53.47 26.90 Fig. 12 P-XRD fitting for synthetic and dental calculus. a El Mirador [i, ii, fitting; iii, difference]. d Synthetic sample 6.84 [i, XRD pattern; ii, XRD pattern; ii, fitting; iii, difference]. b Matangai Turu NW [i, XRD fitting; iii, difference] pattern; ii, fitting; iii, difference]. c Synthetic sample 6.85 [i, XRD pattern; Archaeol Anthropol Sci (2019) 11:4847–4872 4861 Fig. 13 Experimental contamination: a corn granules (orange arrows) are pervasively adsorbed onto the surface, and also b infilling crevices. Note the underlying calcium phosphate matrix with potato starch granules (white arrows) open system mediated by bacteria (> 500 strains) resulting in a morphologically, in terms of crystalline phase, but also in highly heterogeneous structure and a variety of calcium phos- the limited amount of impurities such as aluminum, sodi- phate mineral phases (Rizzo et al. 1963; Grøn et al. 1967; um, and chlorine (Hoyer et al. 1984). Although Raman Rosan and Lamont 2000). Moreover, environmental condi- indicated apatitic phases in the synthetic materials (cf. tions such as pH and temperature determine elemental com- Tsuda and Arends 1993), it must be noted that brushite position, crystallization rates, and recrystallization (Hoyer is acidic and unstable in aqueous solutions and readily et al. 1984; Abraham et al. 2005; Hayashizaki et al. 2008). transforms to apatitic forms in basic solutions (Tas However, the materials science characterization we con- 2016). This would explain the identification of brushite ducted shows that synthetic calculus is a good proxy for through XPS under UHV (water absent), while Raman archeological dental calculus research along the following identified brushite/apatite (water preserved under room lines: conditions). The archeological calculus is mostly hy- droxyapatite [Ca10(PO4)6(OH)2] with traces of brushite, 1. Elemental composition and surface functional groups on since the tail of its band covers the 980 cm−1 region. synthetic and archeological surfaces dictate molecular ad- Apatite in dental calculus materializes between 8 months sorption and therefore the attachment of contaminants. All to a year (Jensen and Danø 1954; Jensen and Rowles samples show a peak at 284.8 eV indicative of C-C/CH. 1957; Rowles 1964; Schroeder and Bambauer 1966; Artificial and natural calculus both show an abundance of Swärdstedt 1966; Driessens and Verbeeck 1988; Kodaka carbon around 288–289 eV [COOH groups]. A small et al. 1988; Abraham et al. 2005; Charlier et al. 2010). No amount of CO3 also appears in archeological and synthet- octacalcium phosphate [Ca 8 H 2 (PO 4 ) 6 *5H 2 O] or ic samples. The O1s spectra are similar. In addition, the whitlocktite [(CaMg)3(PO4)2] were detected. phosphorus and calcium spectra of all samples are practi- 3. Dental calculus has large crystals as shown by Raman, cally identical. XPS, and P-XRD, and therefore it is highly mineralized 2. Our artificial matrices are dominated by the mineral and presents low solubility compared to samples of syn- brushite [CaHPO4*2H2O] and are equivalent to calculus thetic brushite (Tas 2016). Crystallinity governs perme- formed during early depositional stages, up to 3 months; ation and chelation with outside solutes. That is, at higher Table 5 Degradation effects of decontamination admixtures on potato and corn starch granules 4862 Amylase Effects EDTA (w/v) Effects Calcium hydroxide Effects Sodium Effects (w/v) (w/v) hydroxide (w/v) Potato T0 0.001 Radial fissuring/ 0.001 + (NaOH None observed/radial 0.01 None observed/clefting/ 0.001 Gelatinization/ birefringence loss/ 1 ml) fissuring implosion birefringence disarticulation loss/disarticulation T0 0.01 Clefting/radial fissuring/ 0.01 + (NaOH None observed/ 0.02 None observed/clefting/ 0.01 Gelatinization/ gelatinization/birefringence 0.06 g) birefringence loss/ implosion/disarticulation birefringence loss/disarticulation disarticulation loss/disarticulation T0 0.02 Clefting/radial fissuring/ 0.02 + (NaOH None observed/implosion/ 0.05 None observed/clefting 0.02 Birefringence disarticulation 0.15 g) gelatinization loss/leachate T1hr 0.001 Radial fissuring/ 0.001 + (NaOH None observed 0.01 None observed/implosion birefringence loss 1 ml) T1hr 0.01 Radial fissuring/implosion/ 0.01 + (NaOH None observed/ Birefringence 0.02 None observed/implosion/ birefringence loss/ 0.06 g) loss/gelatinization/ gelatinization disarticulation disarticulation T1hr 0.02 Radial fissuring/ 0.02 + (NaOH None observed/radial 0.05 None observed birefringence loss 0.15 g) fissuring/birefringence loss/ gelatinization/disarticulation T6hr 0.001 Radial fissuring/ 0.001 + (NaOH None observed/radial 0.01 None observed/implosion birefringence loss/ 1 ml) fissuring disarticulation T6hr 0.01 Radial fissuring/implosion/ 0.01 + (NaOH None observed/radial 0.02 None observed/implosion/ birefringence loss/ 0.06 g) fissuring gelatinization Disarticulation T6hr 0.02 Radial fissuring/ 0.02 + (NaOH None observed 0.05 None observed/radial birefringence loss/ 0.151 g) fissuring disarticulation T24hr 0.001 Radial fissuring/disarticulation 0.001 + (NaOH None observed/radial fissuring/ 0.01 None observed/implosion/ 1 ml) gelatinization birefringence loss/ disarticulation T24hr 0.01 Radial fissuring/implosion/ 0.01 + (NaOH None observed/ 0.02 Implosion/birefringence loss disarticulation 0.06 g) radial fissuring T24hr 0.02 Radial fissuring/implosion/ 0.02 + (NaOH None observed/radial 0.05 Clefting/implosion/ birefringence loss/ 0.15 g) fissuring/gelatinization/ birefringence loss disarticulation disarticulation T7d 0.001 Radial fissuring/implosion/ 0.001 + (NaOH None observed 0.01 Gelatinization/implosion/ birefringence loss/ 1 ml) birefringence loss/ disarticulation molecular ghosts T7d 0.01 None observed/implosion/ 0.01 + (NaOH None observed/ 0.02 Implosion/birefringence birefringence loss/ 0.06 g) radial fissuring loss/gelatinization disarticulation T7d 0.02 Radial fissuring/birefringence 0.02 + (NaOH None observed/radial 0.05 Radial fissuring/implosion/ loss/disarticulation 0.15 g) fissuring birefringence loss/disarticulation Corn T0 0.001 None observed 0.001 + (NaOH None observed 0.01 None observed 0.001 Birefringence loss/ 1 ml) disarticulation Archaeol Anthropol Sci (2019) 11:4847–4872 Table 5 (continued) Amylase Effects EDTA (w/v) Effects Calcium hydroxide Effects Sodium Effects (w/v) (w/v) hydroxide (w/v) Potato T0 0.01 None observed/radial fissuring/ 0.01 + (NaOH None observed 0.02 None observed 0.01 Gelatinization/ disarticulation 0.06 g) birefringence loss/disarticulation T0 0.02 None observed 0.02 + (NaOH None observed 0.05 None observed 0.02 Birefringence 0.15 g) loss/leachate T1hr 0.001 None observed/radial 0.001 + (NaOH None observed 0.01 None observed fissuring 1 ml) Archaeol Anthropol Sci (2019) 11:4847–4872 T1hr 0.01 None observed/radial fissuring/ 0.01 + (NaOH None observed 0.02 None observed/ disarticulation 0.06 g) gelatinization T1hr 0.02 None observed/radial fissuring/ 0.02 + (NaOH None observed 0.05 None observed disarticulation 0.15 g) T6hr 0.001 None observed/radial 0.001 + (NaOH None observed 0.01 None observed fissuring 1 ml) T6hr 0.01 None observed/gelatinization 0.01 + (NaOH None observed 0.02 None observed 0.06 g) T6hr 0.02 None observed/radial fissuring/ 0.02 + (NaOH None observed 0.05 None observed disarticulation 0.15 g) T24hr 0.001 None observed/radial 0.001 + (NaOH None observed 0.01 None observed fissuring 1 ml) T24hr 0.01 None observed/birefringence 0.01 + (NaOH None observed 0.02 None observed loss/disarticulation 0.06 g) T24hr 0.02 None observed/radial fissuring/ 0.02 + (NaOH None observed 0.05 None observed birefringence loss/disarticulation 0.15 g) T7d 0.001 None observed/radial fissuring 0.001 + (NaOH None observed 0.01 None observed 1 ml) T7d 0.01 None observed/radial fissuring 0.01 + (NaOH None observed 0.02 None observed 0.06g) T7d 0.02 None observed/implosion/ 0.02 + (NaOH None observed 0.05 None observed/ birefringence loss/ 0.15 g) disarticulation disarticulation 4863 4864 Archaeol Anthropol Sci (2019) 11:4847–4872 Fig. 14 Effect of calcium hydroxide on potato starch granules crystallinity in calcium phosphates, the greater the cohe- sample’s surface from contamination. When designing our siveness and resistance to damage and stoichiometric so- experimental contamination process, we pursued a holistic lution will be. approach inclusive of diverse contamination scenarios and vectors in ancient starch research (Laurence et al. 2011; Archeological dental calculus provides ancient starch re- Crowther et al. 2014; Mercader et al. 2017, 2018). We used search a niche where granules may be adsorbed to minerals, native starches admixed with co-concurrent molecules (e.g., coated, overgrown, entrapped, or protected from chemical amino acids) to follow current expectations when conceptual- degradation (Mercader et al. 2018). While encapsulation of- izing starch diagenetic processes (see review in Mercader et al. fers protection from degradation, it does not shield the 2018). Potato and corn are well known structurally (Pérez Fig. 15 Effect of calcium hydroxide on corn starch granules Archaeol Anthropol Sci (2019) 11:4847–4872 4865 Fig. 16 Effect of EDTA on potato starch granules et al. 2009; Huang et al. 2015), are accessible for reproduc- The various chemical decontamination mixtures we ibility, and have variable crystallinity (Whittam et al. 1990; tested caused visible changes to starch granules: implo- Perez and Bertoft 2010). In addition, the granules from these sion, clefting, loss of birefringence, radial fissuring, dis- two species are distinguishable from one another by morpho- articulation, gelatinization, leaching, and granule ghosts. metrics (Jane et al. 1994), are structurally resistant (Gallant Does the proposed decontamination protocol permeate et al. 1997), and known widespread contaminants (Crowther and thus damage the endogenous starch granules trapped et al. 2014; Power et al. 2014; García-Granero et al. 2016; inside calcium phosphate matrices? The permeability of Mercader et al. 2017; Copeland and Hardy 2018). dental calculus has long been demonstrated, and reports Fig. 17 Effect of EDTA on corn starch granules 4866 Archaeol Anthropol Sci (2019) 11:4847–4872 Fig. 18 Effect of α-amylase on potato starch granules of solute infiltration confirm an open system (Kleinberg The sodium hydroxide-based protocol put forward in 1970; Baumhammers et al. 1973). Hydrothermally medi- this paper triggers two concurrent reactions: (1) gelatini- ated hydrolysis of calcium phosphate in the presence of zation of exogenous starch granules located outside or on water and bases is also well-established in dental calculus the surface of dental calculus (Fig. 20a, b) and (2) trans- research (Tas 2016), especially in the presence of sodium formation of brushitic surface into higher order, apatitic hydroxide (Furutaka et al. 2006; Abdel-Aal et al. 2013), recrystallization (Fig. 21e–h). These reactions consume which hydrolyses brushite to apatite, octacalcium phos- most of the sodium hydroxide from the system and dis- phate, and whitlocktite almost immediately: Raman dem- able its starch gelatinizing capability shortly, which im- onstrated the apatitic trend of brushite when mixed with plies that starch granules trapped inside calculus or those water. that developed a Ca:P cast (Fig. 21a–f) maintain their Fig. 19 Effect of α-amylase on corn starch granules Archaeol Anthropol Sci (2019) 11:4847–4872 4867 Fig. 20 Effects of decontamination protocol proposed in this paper. a granules have remained trapped (white arrows) while corn starch granules Before: synthetic matrix with embedded potato granules (white arrows) have gelatinized and leached beyond detection by optical microscopy contaminated with corn granules (orange arrows). b After: potato starch Fig. 21 a–c Calcium phosphate casts coated potato starch granules from synthetic calculus: notice how the granule areas without the protection of a cast (white arrows) gelatinize (a, b: cross-polarized images). e, f Shows phosphate crystallizations along the edges of the synthetic matrix and discrete granules. g–j Calculus permeation by iodine: co-precipitated potato starch granules stained purple and confirmed retention of biochemical functionality, original size, crystallite arrangement, native birefringence, Maltese cross, and lamellae 4868 Archaeol Anthropol Sci (2019) 11:4847–4872 Fig. 22 Effects of experimental cleaning of calculus with a dry (remaining granules, n = 164). c Corn starch coats the synthetic matrix toothbrush. a Corn granules coat most of the synthetic matrix (before (before brushing 5 min). d The elimination of contaminant granules is brushing, 1 min). b Incomplete elimination of contaminant granules partial (granules remaining, n = 334) structure, showing unaltered native features such as size, Little is known about the efficiency of these methods birefringence, lamellae, and Maltese cross (Fig. 21h–j). for a complete removal of sediment/soil and unrelated Contrarily, exogenous starch granules that do not have a microbotanical matter. Generally, mechanical cleaning protective Ca:P shield become targets for gelatinization such as the one accomplished with a toothbrush is unlikely (Fig. 20b). to remove exogenous starches: brushing failed to remove Not all researchers describe their decontamination methods the totality of maize starch granules from synthetic matri- and the effect of their cleaning protocols on contaminant ces (Fig. 22). Even partial surface dissolution by acid over starch granules. Cleaning standards include: short times has unknown effectiveness in removing all starch granules that exist over the entire surface topogra- phy, including deep fissures. Apatite and whitlocktite have 1. Immersion in water to remove phytolith/starch-bearing soil structural weaknesses (Dorozhkin 2012). Acid is likely to particles from the excavation context (Armitage 1975), etch microtopographic irregularities of varying cohesive- sometimes aided by a brush (Lalueza-Fox et al. 1996; Li ness differentially, affecting some spots but not others; et al. 2010; Mickelburgh and Pagan-Jimenez 2012) and/or because of crevices (> 200 μm) and inter-crystal pore size ultrasonics (Li et al. 2010; Madella et al. 2014). and orientation, anomalies in crystallization, different 2. Mechanical dislodgment of loose particles with brushes calcium/phosphate ratios, and intra-crystalline nucleation (Wesolowski et al. 2010; Blatt et al. 2011; Dudgeon and of Fe and Mg (Wolstenholme and O'Connor 2009; Tromp 2014; Power et al. 2014, 2015) plus compressed Dorozhkin 2011). Thus, without direct evidence that a air (Charlier et al. 2010). Note that brushing is often con- given sample was not contaminated before decalcifying, ducted dry (Wesolowski et al. 2010; Blatt et al. 2011; it is impossible to predict how effective cleaning methods Power et al. 2014, 2015, 2018). that use the dissolution of carbonate-bound starch contam- 3. “Chemical peel” via immersion in hydrochloric acid (5– inants from calculus surfaces are; especially when using 20%) for several minutes (Hardy et al. 2009, 2012, 2016a; weak acid solutions for only a few minutes. We cannot Blatt et al. 2011; Buckley et al. 2014). recommend water rinses alone as a decontamination Archaeol Anthropol Sci (2019) 11:4847–4872 4869 method, such as the one put forward by Tavarone et al. During our experimentation with decontaminating agents, we (2018). While this method may work in situations where found that EDTA, which is sometimes used in decalcification, the only proven contamination source are cornstarch gran- damages some starch granules. As such, for studies focusing ules from powdered gloves (cf. Laurence et al. 2011; on ancient starch, we do not recommend the use of EDTA to Crowther et al. 2014; Mercader et al. 2017, 2018), we note dissolve calculus. Future work should characterize dental cal- that the starch utilized in the glove industry are modified, culus at the structural level before engaging in destructive not native. They are cross-linked (Osman and Jensen techniques. This creates a permanent record for future use that 1999; Swanson and Ramalingam 2002), and the industry’s will enhance reproducibility. Materials science will help us goal when chemically changing corn starch for gloves is to select the best possible method to dissolve variable calcium have water-washable granules with altered viscosity and phosphate surface chemistry, mineralogy, and crystallinity, absorption, while maintaining lubrication. That is, water each with unique stoichiometric properties. In addition, re- rinses might not detach contamination from native starch, searchers must screen samples under the microscope to docu- which can adsorb to surfaces and is reported to require ment contaminant starch granules prior to the dissolution of s e v e r a l u l t r a s o n i c a t i o n c y c l e s to d i s l o d g e f r o m calculus, recording results before and after decontamination. archeological and experimental tools (Pedergnana et al. Dental calculus is also a source of proxies other than the 2016; Cnuts and Rots 2017; Mercader et al. 2017). We microbotanical markers studied in this paper; as it was beyond also wanted certainty in decontamination efficiency, the scope of this research, it is undetermined if the methodol- through a worst-case scenario that would include varied ogies presented here are directly transferrable to studies in potential contamination sources, not just gloves, and ancient DNA or proteomics, and this is an avenue for future chemical adsorption to surfaces, versus superficial coating work. from modified, strengthened starches industrially designed to dust off by rinsing (cf. Tavarone et al. 2018). Funding This work was sponsored by the Canadian Social Sciences and Humanities Research Council under its Partnership Grant Program no. 895-2016-1017. The Saskatchewan Structural Sciences Centre (SSSC) is Conclusion acknowledged for providing facilities to conduct this research. Canada Foundation for Innovation, Natural Sciences and Engineering Research Council of Canada and the University of Saskatchewan support research This work is a first step toward developing a starch decontam- at the SSSC. The following Spanish institutions and grants made this ination protocol that is safe for archeological dental calculus. work possible: MINECO/FEDER: CGL2015-65387-C03-1-P, Synthetic calculus allows for isolating, and therefore under- Generalitat de Catalunya: 2017SGR1040 (URV: 2016PFR-URVB2-17). Junta de Castilla y León, and Fundación Atapuerca. standing, all potential variables from the onset. This is op- posed to introducing the multitude of unknowns typical of dental calculus, which is characteristically highly variable with a polygenic matrix that varies from individual to individ- References ual in regard to mineralogy, crystalline system, and micro- structure. We propose to experiment on proxies and provide Abdel-Aal EA, El-Sayed D, Shoeib M, Kandil AT (2013) Enhancing coating of brushite/hydroxyapatite layer on titanium alloy implant a full characterization of the materials; thus recording elemen- surface with additives. Appl Surf Sci 285:136–143 tal composition, surface chemistry, formation period, and or- Abraham J, Grenon M, Sánchex HJ, Pérez C, Barrea R (2005) A case ganic constituency of trapped materials. Efficient decontami- study of elemental and structural composition of dental calculus nation of calculus samples prior to decalcification requires a during several stages of maturation using SRXRF. J Biomed Mater Res A 75:632–628. https://doi.org/10.1002/jbm.a.30484 protocol that eradicates starch granules across surfaces and Armitage RA (1975) The extraction and identification of opal phytoliths crevices within the open microcrystalline structure of dental from the teeth of ungulates. J Archaeol Sci 2:187–197. https://doi. calculus. We arrived at a procedure that disintegrates exoge- org/10.1016/0305-4403(75)90056-4 nous starch granules from all faces and fissures without dam- Atkin NJ, Abeysekera RM, Robards AW (1998) The events leading to the aging the calcium phosphate matrix or the entrapped formation of ghost remnants from the starch granule surface and the contribution of the granule to the gelatinization endotherm. microbotanical remains. This method calls for a 24-h immer- Carbohydr Polym 36:193–204. https://doi.org/10.1016/s0144- sion of dental calculus in 1 ml of sodium hydroxide of 2% w/v 8617(98)00002-2 solution. The decontamination process is neutralized by im- Baumhammers A, Conway JC, Saltzberg D, Matta RK (1973) Scanning mersion in 14 ml of water three times, prior to decalcification. electron microscopy of supragingival calculus. J Periodontol 44:92– Our synthetic precipitates are comparable to those occurring 95 Blatt SH, Redmond BG, Cassman V, Sciulli PW (2011) Dirty teeth and in the oral cavity before the hydroxyapatite crystals mature, ancient trade: evidence of cotton fibres in human dental calculus and are weakly mineralized. Therefore, we expect no damage from Late Woodland, Ohio. Int J Osteoarchaeol 21:669–678. to archeological specimens given their higher cohesiveness. https://doi.org/10.1002/oa.1173 4870 Archaeol Anthropol Sci (2019) 11:4847–4872 Boyadjian CHC, Eggers S, Reinhard K (2007) Dental wash: a problem- Atapuerca, Burgos, Spain). Quat Int 414:258–271. https://doi.org/ atic method for extracting microfossils from teeth. J Archaeol Sci 10.1016/j.quaint.2016.01.051 34:1622–1628. https://doi.org/10.1016/j.jas.2006.12.012 Furutaka K, Monma H, Okura T, Takahashi S (2006) Characteristic reac- Brothwell DR, Brothwell P (1969) Food in antiquity: a survey of the diet tion processes in the system brushite-NaOH solution. J Eur Ceram of early peoples. Thames and Hudson, London Soc 26:543–547 Brown WE, Smith JP, Lehr JR, Frazier AW (1962) Octacalcium phos- Gallant DJ, Bouchet B, Baldwin PM (1997) Microscopy of starch: evi- phate and hydroxyapatite: crystallographic and chemical relations dence of a new level of granule organization. Carbohydr Polym 32: between octacalcium phosphate and hydroxyapatite. Nature 196: 177–191. https://doi.org/10.1016/s0144-8617(97)00008-8 1050–1055 García-Granero JJ, Lancelotti C, Madella M, Ajithprasad P, Crowther A, Buckley S, Usai D, Jakob T, Radini A, Hardy K (2014) Dental calculus Korisettar R, Weisskopf A (2016) Millet and herders: the origins of reveals unique insights into food items, cooking and plant process- plant cultivation in semi-arid North Gujarat (India). Curr Anthropol ing in prehistoric Central Sudan. PLoS One 9:e100808. https://doi. 57:149–173 10.1086-685775 org/10.1371/journal.pone.0100808 Goyal N, Gupta JK, Soni SK (2005) A novel raw starch digesting ther- Charlier P, Huynh-Charlier I, Munoz O, Billard M, Brun L, de la mostable α-amylase from Bacillus sp. I-3 and its use in the direct Grandmaison GL (2010) The microscope (optical and SEM) exam- hydrolysis of raw potato starch. Enzym Microb Technol 37:723– ination of dental calculus depositis (DCD). Potential interest in fo- 734. https://doi.org/10.1016/j.emzmitec.2005.04.017 rensic anthropology of a bio-archaeological method. Legal Med 12: Grazulis S, Chateigner D, Downs RT, Yokochi AT, Quiros M, Lutterotti 162–171. https://doi.org/10.1016/j.legalmed.2010.03.003 L, Manakova E, Butkus J, Moeck P, Le Bail A (2009) Cnuts D, Rots V (2017) Extracting residues from stone tools for optical Crystallography open database: an open-access collection of crystal analysis: towards in experiment-based protocol. Archaeol Anthropol structures. J Appl Crystallogr 42:726–729 Sci 10:1717–1736. https://doi.org/10.1007/s12520-017-0484-7 Grøn P, Van Campen GJ, Lindstrom I (1967) Human dental calculus: Coil J, Korstanje MA, Archer S, Hastorf CA (2003) Laboratory goals and inorganic chemical and crystallographic composition. Arch Oral considerations for multiple microfossil extraction in archaeology. J Biol 12:829–837. https://doi.org/10.1016/0003-9969(67)90105-7 Archaeol Sci 30:991–1008. https://doi.org/10.1016/s0305-4403(02) Hardy K, Blakeney T, Copeland L, Kirkham J, Wrangham R, Collins M 00285-6 (2009) Starch granules, dental calculus and new perspectives on Copeland L, Hardy K (2018) Archaeological starch. Agronomy 8:4. ancient diet. J Archaeol Sci 36:248–255 https://doi.org/10.3390/agronomy801000 Hardy K, Buckley S, Collins M, Estalrrich A, Brothwell D, Copeland L, Cristiani E, Radini A, Edinborough M, Borić D (2016) Dental calculus García-Taberno A, García-Vargas S, de la Rasilla M, Lalueza-Fox C, reveals Mesolithic foragers in the Balkans consumed domesticated Huguet R, Bastir M, Santamaría D, Madella M, Wilson J, Cortés AF, plant foods. Proc Natl Acad Sci 113:10298–10303. https://doi.org/ Rosas A (2012) Neanderthal medics? Evidence for food, cooking, 10.1073/pnas.1603477113 and medicinal plants entrapped in dental calculus. Cristiani E, Radini A, Borić D, Robson HK, Caricola I, Carra M, Mutri G, Naturwissenschaften 99:617–626 Oxilia G, Zupancich A, Šlaus M, Vujević D (2018) Dental calculus Hardy K, Brand-Miller J, Brown KD, Thomas MG, Copeland L (2015) and isotopes provide direct evidence of fish and plant consumption The importance of dietary carbohydrate in human evolution. Q Rev in Mesolithic Mediterranean. Sci Rep 8:8117 Biol 90:251–268 10.1086-682587 Crowther A, Haslam M, Oakden N, Walde D, Mercader J (2014) Hardy K, Radini A, Buckley S, Blasco R, Copeland L, Burjachs F, Girbal Documenting contamination in ancient starch laboratories. J J, Yll R, Carbonell E, María Bermúdez de Castro J (2016a) Diet and Archaeol Sci 49:90–104 environment 1.2 million years ago revealed through analysis of den- De Gelder J, De Gusseum K, Vandenabeele P, Moens L (2007) Reference tal calculus from Europe’s oldest hominin at Sima del Elefante, database of Raman spectra of biological molecules. J Raman Spain. Sci Nat 104 doi:https://doi.org/10.1007/s00114-016-1420-x Spectrosc 38:1133–1147 Hardy K, Radini A, Buckley S, Sarig R, Copeland L, Gopher A, Barkai R Dobney K, Brothwell D (1986) Dental calculus: its relevance to ancient (2016b) Dental calculus reveals potential respiratory irritants and diet and oral ecology. In: Olsen S (ed) Teeth and anthropology, vol ingestion of essential plant-based nutrients at Lower Palaeolithic 291. Archeopress, Oxford, pp 55–81 Qesem Cave, Israel. Quat Int 398:129–135. https://doi.org/10. Dorozhkin SV (2011) Calcium orthophosphates: occurrence, properties, 1016/j.quaint.2015.04.033 biomineralization, pathological calcification and biomimetic appli- Haslam M (2004) The decomposition of starch grains in soils: implica- cations. Biomatter 1:121–164. https://doi.org/10.4161/biom.18790 tions for archaeological residue analyses. J Archaeol Sci 31:1715– Dorozhkin SV (2012) Amorphous calcium orthophosphates: nature, 1734 chemistry and biomedical applications. Int J Mater Chem 2:19–46. Hayashizaki J, Ban S, Nakagaki H, Okumura A, Yoshii S, Robinson C https://doi.org/10.5923/j.ijmc.2012201.04 (2008) Site specific mineral composition and microstructure of hu- Downs RT, Hall-Wallace M (2003) The American mineralogist crystal man supra-gingival dental calculus. Arch Oral Biol 53:168–174. structure database. Am Mineral 88:247–250 https://doi.org/10.1016/j.archoralbio.2007.09.003 Driessens FCM, Verbeeck R (1988) Relation between physio-chemical Helbert W, Schülien M, Henrissat B (1996) Electron microscope investi- solubility and biodegradability of calcium phosphates. In: Putter C gation of the diffusion of Bacillus licheniformis α-amylase into corn (ed) Implant materials in biofunction. Elsevier, Amsterdam, pp 105– starch granules. Int J Biol Macromol 19:165–169. https://doi.org/10. 111 1016/041-8130(96)01123-3 Dudgeon JV, Tromp M (2014) Diet, geography and drinking water in Hendy J, Warinner C, Bouwman A, Collins MJ, Fiddyment S, Fischer R, Polynesia: microfossil research from archaeological human dental Hagan R, Hofman CA, Holst M, Chanes E, Klaus L (2018) calculus, Rapa Nui (Easter Island). Int J Osteoarchaeol 24:634–648. Proteomic evidence of dietary sources in ancient dental calculus. https://doi.org/10.1002/oa.2249 Proc R Soc Lond (Biol) 285:20180977. https://doi.org/10.1098/ Euba I, Allue E, Burjachs F (2016) Wood uses at El Mirador Cave rspb.2018.0977 (Atapuerca, Burgos) based on anthracology and dendrology. Quat Henry AG, Piperno DR (2008) Using plant microfossils from dental Int 414:285–293. https://doi.org/10.1016/j.quaint.2015.08.084 calculus to recover human diet: a case study from tell al-Raqa'i, Expósito I, Burjachs F (2016) Taphonomic approach to the palynological Syria. J Archaeol Sci 35:1943–1950. https://doi.org/10.1016/j.jas. record of burnt and unburnt samples from El Mirador cave (Sierra de 2007.12.005 Archaeol Anthropol Sci (2019) 11:4847–4872 4871 Henry AG, Brooks AS, Piperno DR (2011) Microfossils in calculus dem- of Congo: the rock shelter site of Matangai Turu NW. Am J Phys onstrate consumption of plants and cooked foods in Neanderthal Anthropol 115:24–37 diets (Shanidar III, Iraq; Spy I and II, Belgium). Proc Natl Acad Mercader J, Abtosway M, Baquedano E, William Bird RW, Diez-Martin Sci 108:486–491. https://doi.org/10.1073/pnas.1016868108 F, Domínguez-Rodrigo M, Favreau J, Itambu M, Lee P, Mabulla A, Horrocks M, Nieuwoudt MK, Kinaston R, Buckley H, Bedford S (2014) Patalano R, Pérez-González A, Santonja M, Tucker L, Walde D Microfossil and Fourier transform infrared analyses of Lapita and (2017) Starch contamination landscapes in field archaeology: post-Lapita human dental calculus from Vanuatu, Southwest Pacific. Olduvai Gorge, Tanzania. Boreas 46:918–934. https://doi.org/10. J Roy Soc New Zeal 44:17–33. https://doi.org/10.1080/03036758. 1111/bor.12241 ISSN 0300-9483 2013.842177 Mercader J, Akeju T, Brown M, Bundala M, Collins MJ, Copeland L, Hoyer I, Gaengler P, Bimberg R (1984) In vivo remineralization of hu- Crowther A, Dunfield P, Henry A, Inwood J, Itambu I, Kim J-J, man enamel and dental calculus formation. J Dent Res 63:1136– Larter S, Longo L, Oldenburg T, Patalano R, Sammynaiken R, 1139. https://doi.org/10.1177/00220345840630090801 Soto M, Tyler R, Xhauflair H (2018) Exaggerated expectations in Huang ZQ, Lu JP, Li XH, Tong ZF (2007) Effect of mechanical activation ancient starch research and the need for new taphonomic and au- on physico-chemical properties and structure of cassava starch. thenticity criteria. Facets 30:777–798. https://doi.org/10.1139/ Carbohydr Polym 68:128–135. https://doi.org/10.1016/j.carbpol. facets-2017-0126 2006.07.017 Mickelburgh HL, Pagan-Jimenez JR (2012) New insights into the con- Huang TT, Zhou DN, Jin ZY, Xu XM, Chen HQ (2015) Effect of sumption of maize and other food plants in the pre-columbian debranching and heat-moisture treatments on structural characteris- Caribbean from starch granules trapped in human dental calculus. tics and digestibility of sweet potato starch. Food Chem 187:218– J Archaeol Sci 39:2468–2478. https://doi.org/10.1016/j.jas.2012.02. 224. https://doi.org/10.1016/j.foodchem.2015.04.050 020 Jane J-L, Kasemsuwan T, Leas S, Zobel H, Robyt JF (1994) Anthology of Middleton WD, Rovner I (1994) Extraction of opal phytoliths from her- starch granule morphology by scanning electron microscopy. bivore dental calculus. J Archaeol Sci 32:469–473. https://doi.org/ Starch-Stärke 46:121–129. https://doi.org/10.1002/star. 10.1006/jasc.1994.1046 19940460402 Osman MO, Jensen SL (1999) Surgical gloves: current probelms. World J Jensen AT, Danø M (1954) Crystallography of dental calculus and the Surg 23:630–637. https://doi.org/10.1007/PL00012360 precipitation of certain calcium phosphates. J Dent Res 33:741–750. Otsuka M, Moritaka H, Fukuba H, Kimura S, Ishihara M (2001) Effects https://doi.org/10.1177/00220345540330060201 of calcium carbonate and calcium hydroxide on gelatinization and Jensen AT, Rowles SL (1957) Magnesian whitlockite, a major consituent retrodegradation of corn starch. J Jpn Soc Food Sci 48:751–758. of dental calculus. Acta Odontol Scand 15:121–139. https://doi.org/ https://doi.org/10.3136/nskkk.48.751 10.3109/00016355709041096 Pedergnana A, Asryan L, Fernández-Marchena JL, Ollé A (2016) Kleinberg I (1970) Biochemistry of the dental plaque. Adv Oral Biol 4: Modern contaminants affecting microscopic residue analysis on 43–90 stone tools: a word of caution. Micron 86:1–21. https://doi.org/10. Kodaka T, Debari K, Higashi S (1988) Magnesium-containing crystals in 1016/j.micron.2016.04.003 human dental calculus. Microsc 37:73–80. https://doi.org/10.1093/ Perez S, Bertoft E (2010) The molecular structures of starch components oxfordjournals.jmicro.a050666 and their contribution to the architecture of starch granules: a com- Koutsopoulos S (2002) Synthesis and characterization of hydroxyapatite prehensive review. Starch-Stärke 62:389–420 crystals: a review study on the analytical methods. J Biomed Mater Res A 62:600–612 Pérez S, Baldwin PM, Gallant DJ (2009) Structural features of starch granules I. In: BeMiller J, Whistler R (eds) Starch: chemistry and Lalueza-Fox C, Juan J, Albert RM (1996) Phytolith analysis on dental technology. 3rd edn. Academic Press, Burlington, pp 149–192. calculus, enamel surface and burial soil: information about diet and https://doi.org/10.1016/B978-0-12-746275-2.00005-7 paleoenvironment. Am J Phys Anthropol 101:101–113 Laurence AR, Thoms AV, Bryant VM, McDonough C (2011) Airborne Piperno DR, Dillehay TD (2008) Starch grains on human teeth reveal starch granules as a potential contamination source at archaeological broad crop diet in northern Peru. Proc Natl Acad Sci 105:19622– sites. J Ethnobiol 31:213–232. https://doi.org/10.2993/0278-0771- 19627. https://doi.org/10.1073/pnas.0808752105 31.2.213 Power RC, Salazar-García DC, Wittig RM, Henry AG (2014) Assessing Li M, Yang Z, Wang H, Wang Q, Jia X, Ge QS (2010) Starch grains from use and suitability of scanning electron microscopy in the analysis of dental calculus reveal ancient plant foodstuffs at Chenqimogou site, micro remains in dental calculus. J Archaeol Sci 49:160–169. Gansu Province. Sci China Earth Sci 53:694–699. https://doi.org/10. https://doi.org/10.1016/j.jas.2014.04.016 1007/s11430-010-0052-9 Power RC, Salazar-García DC, Straus LG, Morales MRG, Henry AG Madella M, García-Granero JJ, Out WA, Ryan P, Usai D (2014) (2015) Microremains from El Mirón cave human dental calculus Microbotanical evidence of domestic cereals in Africa 7000 years suggests a mixed plant-animal subsistence economy furing the ago. PLoS One 9:e110177. https://doi.org/10.1371/journal.pone. Magdalenian in Northern Iberia. J Archaeol Sci 60:39–46. https:// 0110177 doi.org/10.1016/j.jas.2015.04.003 Maher GG (1983) Alkali gelatinization of flours. Starch-Stärke 35:271– Power RC, Salazar-García DC, Henry AG (2016) Dental calculus evi- 276 dence of Gravettian diet and behavior at Dolni Vestonice and Mathlouthi M, Koenig JL (1987) Vibrational spectra of carbohydrates. P a v l o v. I n : S v o b o d a J A ( e d ) D o l n í V ě s t o n i c e I I : Adv Carbohydr Chem Biochem 44:7–89 Chronostratigraphy, Paleoethnology, Paleoanthropology. Academy Mercader J (2002) Forest people: the role of African rainforests in human of Sciences of the Czech Republic, Institute of Archeology, Brno, pp evolution and dispersal. Evol Anthropol 11:117–124 345–352 Mercader J, Runge F, Vrydaghs L, Doutrelepont H, Ewango CEN, Juan- Power RC, Salazar-García DC, Rubini M, Darlas A, Havarti K, Walker Tresseras J (2000) Phytoliths from archaeological sites in the tropi- MJ, Hublin JJ, Henry AG (2018) Dental calculus indicates wide- cal Forest of Ituri, Democratic Republic of Congo. Quat Res 54: spread plant use within the stable Neanderthal dietary niche. J Hum 102–112 Evol 119:27–41. https://doi.org/10.1016/j.jhevol.2018.02.009 Mercader J, Garralda MD, Pearson O, Bailey R (2001) 800 year old Ragheb AA, Abd El-Thalouth I, Tawfik S (1995) Gelatinization of starch human remains from the Ituri tropical forest, Democratic Republic in aqueous alkaline solutions. Starch-Stärke 47:338–345 4872 Archaeol Anthropol Sci (2019) 11:4847–4872 Rizzo AA, Scott DB, Bladen HA (1963) Calcification of oral bacteria. Vergès JM, Alluè E, Fontanals M, Morales JI, Martín P, Carrancho Á, Ann N Y Acad Sci 109:14–22. https://doi.org/10.1111/j.1749-6632. Expósito I, Guardiola M, Lozano M, Marsal R, Oms X, Euba I, 1963.tb13458.x Rodríguez A (2016) El Mirador cave (Sierra de Atapuerca, Rodríguez A, Allue E, Buxó R (2016) Agriculture and livestock economy Burgos, Spain): a whole perspective. Quat Int 414:236–243. among prehistoric herders based on plant macro-remains from El https://doi.org/10.1016/j.quaint.2016.01.044 Mirador (Atapuerca, Burgos). Quat Int 414:272–284. https://doi. Wang Y-J, Wang L (2002) Characterization of acetylated waxy maize org/10.1016/j.quaint.2016.01.045 starches prepared under catalysis by different alkali and alkaline- Rosan B, Lamont RJ (2000) Dental plaque formation. Microbes Infect 2: earth hydroxides. Starch-Stärke 54:25–30 1599–1607. https://doi.org/10.1016/s1286-4579(00)01316-2 Warinner C, Hendy J, Speller C, Cappellini E, Fischer R, Trachsel R, Rowles SL (1964) Biophysical studies on dental calculus in relation to Arneborg J, Lynnerup N, Craig OE, Swallow DM, Fotakis A, periodontal disease. Dent Pract Dent Rec 15:2–7 Christensen RJ, Olsen JV, Liebert A, Montalva N, Fiddyment S, Schofield PF, Knight KS, van der Houwen JAM, Valsami-Jones E (2004) Charlton S, Mackie M, Canci A, Bouwman A, Ruhi F, Gilbert The role of hydrogen bonding in the thermal expansion and dehy- MTP, Collins M (2014) Direct evidence of milk consumption from dration of brushite, di-calcium phosphate dihydrate. Phys Chem ancient human dental calculus. Nat Sci Rep 4:1–6 Miner 31:606–624 Warinner C, Speller C, Collins MJ (2015) A new era in Schroeder HE, Bambauer HU (1966) Stages of calcium phosphate palaeomicrobiology: prospects for ancient dental calculus as a crystallisation during calculus formation. Arch Oral Biol 11:1–14. long-term record of the human oral microbiome. Philos Trans R https://doi.org/10.1016/0003-9969(66)90112-9 Soc B 370:20130376. https://doi.org/10.1098/rstb.2013.0376 Schroeder LW, Dickens B, Brown WE (1977) Crystallographic studies of Warinner C, Herbig A, Mann A, Fellows Yates JA, Weiß CL, Burbano the role of Mg as a stabilizing impurity in ß-Ca3 (PO4)2. II. HA, Orlando L, LKrause J (2017) A robust framework for microbial Refinement of Mg-containing ß-Ca3(PO4)2. J Solid State Chem archaeology. Annu Rev Genomics Hum Genet 18:321–356. https:// 22:253–262 doi.org/10.1146/annurev-genom-091416-035526 Sudarsanan K, Young RA (1969) Significant precision in crystal struc- Wesolowski V, de Souza SMFM, Reinhard KJ, Ceccantini G (2010) tural details: Holly Springs hydroxyapatite. Acta Cryst 25:1534– Evaluating microfossil content of dental calculus from Brazilian 1543. https://doi.org/10.1107/S0567740869004298 sambaquis. J Archaeol Sci 37:1326–1338 Swanson MC, Ramalingam M (2002) Starch and natural rubber allergen interaction in the production of latex gloves: a hand-held aerosol. J Weyrich LS, Dobney K, Cooper A (2015) Ancient DNA analysis of Allergy Clin Immunol 110:S15–S20. https://doi.org/10.1067/mai. dental calculus. J Hum Evol 79:119–124. https://doi.org/10.1016/j. 2002.125338 jhevol.2014.06.018 Swärdstedt T (1966) Odontological aspects of a Medieval population in Weyrich LS, Duchene S, Soubrier J, Arriola L, Llamas B, Breen J, Morris the province of Jämtland, mid-Sweden. University of Lund, AG, Alt KW, Caramelli D, Dresely V, Farrell M, Farrer AG, Tao D, Zhang J, Zheng W, Cao Y, Sun K, Jin S (2015) Starch grain Francken M, Gully M, Haak W, Hardy K, Harvati K, Held P, analysis of human dental calculus to investigate Neolithic consump- Holmes EC, Kaidonis J, Lalueza-Fox C, de la Rasilla M, Rosas A, tion of plants in the middle Yellow River Valley, China: a case study Semal P, Soltysiak A, Townsend G, Usai D, Wahl J, Huson DH, on Gouwan site. J Archaeol Sci Rep 2:485–491. https://doi.org/10. Dobney K, Cooper A (2017) Neanderthal behavior, diet, and disease 1016/j.jasrep.2015.05.003 inferred from ancient DNA in dental calculus. Nature 544:357–372 Tas AC (2016) Transformation of brushite (CaHPO4.2H2O) to Whittam MA, Noel TR, Ring SG (1990) Melting behaviour of A- and B- whitlockite (Ca9Mg (HPO4)(PO4)6) or other CaPs in physiologi- type crystalline starch. Int J Biol Macromol 12:359–362. https://doi. cally relevant solutions. J Am Ceram Soc 99:1200–1206 org/10.1016/0141-8130(90)90043-A Tavarone A, de los Milagros Colobig M, Passeggi E, Fabra M (2018) Wolstenholme GEW, O'Connor M (2009) Caries-resistant teeth vol 964. Cleaning protocol of archaeological dental calculus: a methodolog- John Wiley and Sons, ical proposal for vegetable microremains analysis. Am J Phys Yamada T, Morimoto Y, Hisamatsu M (1986) Effect of citric acid on Anthropol 167:416–422. https://doi.org/10.1002/ajpa.23630 potato starch gelatinization. Starch-Stärke 38:264–268 Tromp M, Dudgeon JV (2015) Differentiating dietary and non-dietary Zhang B, Dhital S, Flanagan BM, Gidley MJ (2014) Mechanism for microfossils extracted from human dental calculus: the importance starch granule ghost formation deduced from structural and enzyme of sweet potato to ancient diet on Rapa Nui. J Archaeol Sci 54:54– digestion properties. J Agric Food Chem 62:760–771. https://doi. 63. https://doi.org/10.1016/j.jas.2014.11.024 org/10.1021/jf404697v Tromp M, Buckley H, Geber J, Matisoo-Smith E (2017) EDTA decalci- fication of dental calculus as an alternate means of microparticle extraction from archaeological samples. J Archaeol Sci Rep 14: 461–466. https://doi.org/10.1016/j.jasrep.2017.06.035 Tsuda H, Arends J (1993) Raman spectra of human dental calculus. J Publisher’s note Springer Nature remains neutral with regard to juris- Dent Res 72:1609–1613 dictional claims in published maps and institutional affiliations.