Dental caries enamel composition and subsistence among prehistoric Amerindians of Ohio.код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 71:95-102 (1986) Dental Caries, Enamel Composition, and Subsistence Among Prehistoric Amerindians of Ohio KIM N. SCHNEIDER Department of Anthropology, The Wichita State Uniuersity, Wichita, Kansas 67208 KEY WORDS Trace elements, Acquired dental pathology, Correspondence analysis ABSTRACT Six populations of prehistoric Amerindians from Ohio are sampled to establish the relationship of enamel composition and dental caries experience. The populations used included groups practicing hunting-gathering-fishing and maize horticulture, and they represent a t least two major cultural traditions, the Late Archaic/Glacial Kame (1000-500 BC) and the Fort Ancient and Sandusky Bay Traditions (AD) 1200-1600). Characterization of enamel composition is achieved using scanning electron microscopy energydispersive X-ray analysis. Thirteen elements present are quantified, and they are analyzed with respect to each population’s subsistence base using correspondence analysis. Evaluations of cariogenic and cariostatic effects of elements are made on the basis of caries frequency comparisons among the populations. Results indicate that zinc, copper, iron, nickel, and calcium-phosphorus ratio distinguish populations exhibiting high, moderate, and low caries frequency as well as discriminate hunter-gatherers from maize horticulturalists. The study of dental disease among prehistoric human populations has become a n important focus of paleoepidemiological research (Rose, 1977; Pfeiffer, 1979; Cook and Buikstra, 1979; Sciulli, 1978; and many others). Generally, these studies have documented the frequency of dental caries, abscesses, premortem tooth loss, and periodontal disease among prehistoric samples, but they have not associated these pathological conditions with dietary, soil, or compositional parameters. Epidemiological studies of extant populations have long suggested that caries frequency is associated with soil and water sources of particular elements (Hercus, 1925; Dunning, 1953; Nichols and McNall, 1957; Durfor and Becker, 1962). Flourine has received enormous attention (for a recent review, see Hurny, 1978), but other elements known to be present in dental enamel have yielded equivocal results. Curzon and Cutress (1983) have summarized the current status of our understanding, and, clearly, more basic investigative research is needed to elucidate cariostatic or cariogenic effects for many elements. 01986 ALAN R. LISS. INC Acquired dental pathologies are the result of complex processes including factors such as the presence of oral bacteria, plaque, carbohydrates, and enamel solubility. In that enamel is composed of carbonate hydroxyapatite, it is a naturally accomodating structure and can be altered by the inclusion of a wide variety of ions. These substitutions may alter the crystal’s size or behavior. Hence, variations, in enamel composition may explain, to some degree, variation in susceptibility to carious lesions between individuals or populations. Additionally since the inclusion of intrusive ions depends on bioavailability during enamel formation, their presence or absence in enamel may produce a nutritional history for a n individual or a nutritional profile for a population. Elemental intake is via a soil/water-plant/ animal-human food chain. Food sources are recognized as providing the largest source of elements in human nutrition (Pike and Brown, 1975).Plant foods are limited in their Received, July 14, 1985; revised February 20, 1986 accepted February 26,1986. 96 K.N. SCHNEIDER contribution by their ability to extract nutrients and soil-derived elements, whereas fish and animal resources are affected by both the quality of plant andlor animal food eaten and the organism’s ability to store or eliminate nutrients. Human intervention may compromise the elemental composition of foods eaten via storage, processing, and cooking. Although the food chain is complex, general patterns of food use among human populations should be distinguishable if they depend on significantly different resources. Carbohydrates are recognized as important etiological factors with respect to both balanced human nutrition and the expression of dental caries. High levels of carbohydrate provide important sources of energy in many nonindustrialized societies (Guthrie, 1979), and their presence in the oral cavity provides acidogenic bacteria in dental plaque with fermentable substrates. The disolution of enamel by bacterial byproducts thus is enhanced by the presence of sugars and starches (Cox, 1952). Additionally, many of these diets are characterized by a lack of “detergent” foods requiring extensive mastication and aiding in stimulating salivary flow and oral clearance. Plaque formation can increase, thereby providing a permeable tooth coating for bacterial acids. Examination of dental caries and dental enamel composition may address at least two issues of paleoepidemiological interest. First, associations of particular elements and dental health status may substantiate their proposed cariogenic or cariostatic action. Second, profiles of elemental differences might allow independent tests among archaeologically defined subsistence strategies. The application of such profiles may allow inferences regarding subsistence pattern even when archaeological preservation of food remains is poor or absent. It is important to stress that this approach does not endeavor to reconstruct prehistoric diets but rather attempts to test the utility of compositional analysis in identifying similarities and differences between populations practicing different subsistence strategies. MATERIALS AND METHODS Six populations of prehistoric Amerindians from Ohio are used in the study (Fig. 1). Chronologically the earliest site sampled is Williams Cemetery (33Wo7a1, located approximately 12.5 km from Toledo in Wood County, Ohio. The site is a large cremation cemetery consisting of 18 separate burial features representing over 850 individuals (Schneider and Sciulli, unpublished data). The site also has yielded at last 75 noncremated individuals, and it is from this subgroup that the reseach sample is drawn. The site has been described and included in a number of recent investigations (Stothers, 1973; 1979; Sciulli et al., 1982, 1983, 1984) and has been dated radiometrically from multiple samples at 850 45 years BC. The population represented at Williams Cemetery is thought to have been composed of nomadic, small groups who exploited seasonally variable resources by hunting, gathering and fishing the riverine or lacustrine areas of northwestern Ohio. Three samples are included from the early Fort Ancient Tradition (AD 950-1250). The State Line site (33Ha58) is located on the Ohio-Indianaborder, 1.5 km north of the Ohio River. The presence of charred maize, nuts, seeds, and deer bones indicates that these people were at least rudimentary horticulturalists (Wheelersburg, 1982). The Turpin site (33HA19) is located on the east bank of the Little Miami River, 2 km north of its confluence with the Ohio River. The site has a Fort Ancient component, which has been dated radiometrically at AD 1175 f 150 years mraun, 19821, and large quantities of maize were found in association (Perzigian, personal communication).The Anderson Village site (33Wa4) is located on the east bank of the Little Miami River in Washington Township, Ohio. The site is designated as the typesite for the Anderson phase of Fort Ancient and has been well investigated (Essenpreis, 1978; Griffin, 1966; Barber, 1978). These three populations represent sedentary communities focused on maize horticulture while continuing to exploit local resources. Pearson Village (33Sa9) is located 4 km southwest of Sandusky Bay. The area of the site from which the research sample is taken has been dated radiometrically at AD 1410 f 65 years (Bowen, personal communication). With approximately 75% of the site excavated, more than 400 individual burials have been recovered. Faunal and botanical remains suggest a mixed subsistence base dependent on deer, elk, fish, maize, beans, and nuts. The most recent site included is the Indian Hills site (33W04). The site has been radiocarbon dated at AD 1610 100 years (Stothers, 1973)and is located less than 1km north * * DENTAL CARIES A N D ENAMEL COMPOSITION 97 U Fig. 1. Location of sites included in this study. of Williams Cemetery. The site represents a large, stockaded village whose population depended on locally available fauna and tended gardens of both native and tropically domesticated cultigens (Stothers and Pratt, 1981). The breadth of both subsistence bases and distance among the sites allows spatial as well as temporal changes in enamel compositions and caries frequencies to be examined. The location of these sites reflects an important geological boundary in the state. The northern sites are located in an area that was subjected to glaciation, whereas the southern sites are in an unglaciated region. Associated soil profiles would be relevant data to correlate to enamel composition, particularly if diagenetic exchange was significant (Lambert et al., 19791, but these were not available. The role of soil variability, however, can be assessed indirectly. If local soil composition contributes to the compositional differences among these samples, the northern sites (which are represented by both pre- and post horticulturalists) should cluster together, and the southern sites should be separated from them. Caries determination was made by the author for all samples except State Line and Turpin sites. Caries data for Turpin were provided by D.J. Braun (1982) and were unavailable for the State Line site. Caries were recorded for available succedaneous teeth, and determination was limited to macroscopic lesions. A carious lesion was recorded if any surface of the tooth had a macroscopic necrotic defect. Each tooth was examined under strong light and lesions were detected with the aid of a dental explorer. "he incidence of pre- and postmortem tooth loss and abscesses also was recorded when alveolar bone was present. Caries frequency is expressed as the number of caries observed as a percent of the total number of each tooth and tooth type. Frequencies were grouped by location-maxillary anterior and posterior mandibular anterior and posterior after initial tests of homogeneity. Table 1 was entered as a matrix for correspondenceanalysis, the result of which was then compared to the display produced based on elemental composition. Although it is recognized that the method- 98 K.N. SCHNEIDER TABLE 1. Caries frequency and sample size per category Maxillary ant. Caries n Maxillary post. Caries n Mandibular ant. Caries n Mandibular post. Caries n Total caries N Williams Cemetery Turpin' Anderson Village Indian Hills Pearson Village' 3 224 33 239 13 352 9 123 48 662 1 263 78 224 50 244 28 190 105 662 0 190 16 259 15 262 5 143 30 761 6 259 109 231 46 307 30 206 86 687 10 956 236 953 124 1,165 72 662 269 2,772 'Braun, 198216. 'Inludes all teeth collected through the 1983 excavation TABLE 2. Canine samples used in elemental composition study Sample Side (r/l) Age range (years) Williams Cemetery Turpin State Line Anderson Village Pearson Village Indian Hills 10112 710 315 414 15/15 3/7 9-21 10-19 10-19 12-21 10-21 9-19 ology used has limitations, for purposes of comparability to the Turpin data, no corrections were made for pre- or postmortem tooth loss (Hardwick, 1960) or multiple carious lesions on individual teeth. Numerous studies have detailed the composition of bone (e.g., Sillen, 1981; Aufderheide et al., 1985; Lambert et al., 1982, 1983). Despite the common crystallographic structure of bone and enamel, significant difference preclude their compositional comparison. For example, turnover rates (Bryant and Loutit, 1964), mineralized content (Bhaskar, 19801, and processes of elemental uptake (Vaughan, 1970) are different in these tissues. For the purposes of this study, bone samples from these populations were not included. Mandibular succedaneous canines were selected for compositional analysis (Table 2). This tooth was chosen for a number of reasons. First, the tooth has a long developmental period (6-7 years) before the enamel is completed (Bhaskar, 19801, allowing restricted sampling. Second, the mandibular sex F - M Unknown - 22 - - 8 2 2 1 24 9 4 - - 7 6 canine is morphologically less complex on the buccal surface than other teeth, so the tooth requires little alteration for energy-dispersive X-ray analysis. Only canines with unworn, disease-free, complete enamel surfaces were included. They were photographed, cast, measured, and cleaned with distilled water. The teeth then were polished with a series of silicon grit papers and diamond paste on a Struers DP-U2 polishing device to produce planar buccal surfaces. The specimens were cut a t the cementoenamel junction, and the polished crowns were mounted on copper easels and aluminum stubs and carbon coated. An S4-10 Cambridge scanning electron microscope (SEM) was used for characterization of enamel composition. The SEM is equipped with a n X-ray detector and multichannel analyzer, which records elements present in the structure by collecting, amplifying, and storing incoming X-ray pulses. Thirteen elements were designated as regions of interest. The rationale for this particular suite of elements rests on their presence in dental DENTAL CARIES AND ENAMEL COMPOSITION enamel, detectability, and nutritional role. The elements recorded included major essential elements (Ca, P), trace essential elements (Mg, Mo, Cu, Mn, Se,Zn, Fe, Ni), trace (inactive) nonessential elements (Sr, Al) and trace (toxic) nonessential elements (Pb). A small pilot investigation was made before the entire sample of canines was analyzed. Five mandibular canines not associated with a specific burial were used (three from Williams Cemetery and two from Pearson Village). A series of polishes corresponding to the depths of 50, 100, 200, 300, and 500 p was made sequentially on each tooth, and the compositional analysis was performed. Each depth-associated data set was run through the complete data correction analysis to determine if particular elements are associated with increasing distance from the surface of the tooth, potentially reflecting diagenesis. Data from depths of less than 200 p appeared most stable among aliquots of each tooth, and the 100-200 p depth was chosen for the study. Additionally, A1 was eliminated as an element of interest for the ensuing study because it was significantly correlated t o superficial layers and is considered to reflect postdepositional contamination. Elemental peak, associated background, and nonassociated background were collected for three separate analyses per tooth. The working distance of the sample to the detector was 11 mm, with the polished surface angled at 38"-40" inclination. The instrument was set at 20 kV saturation, and a 40 sec real-time interval was used for each of the three runs per tooth. Data smoothing and background correction were achieved using the BEDAMAC computer program (Foreman and Awadalla, 1978). The BEDAMAC printout includes the regression coefficients of the observed-to-expected peak and background values. R2 values of less than 0.95 were considered to reflect the presence of spectral artifacts and poor elemental peak data quality, and such aliquots were eliminated from the ensuing analysis. This resulted in the deletion of 21 aliquots from the initial 255 analyses (8.24%). Conversion to weight percent per element as compared to standards was done using the Colby (1971) MAGIC IV program. Mean weight percents of each element were converted to parts per million using SAS software (SAS Institute, 1982). These data were then entered as a matrix for correspondence 99 analysis using SPEAKEASY 111 software (Cohen and Peiper, 1979). Correspondence analysis allows the information contained in a matrix to be retained and displayed in lowvector space (Greenacre, 1984), in effect allowing all the samples and elements to be displayed simultaneously. RESULTS Figure 2 is the result of the correspondence analysis using values of all elements collected. The first two principal axes of inertia explain 54.98% and 23.95% of the variance present, respectively. Clearly, the huntinggathering-fishing sample, Williams Cemetery, is separated from the other samples. Zinc, Cu, and Fe, and to a lesser degree Mn, Mo, and Pb, contribute to its position. The horticultural samples are more dispersed than would be expected if their subsistence bases are the same. Rather, it appears that these groups are, in some cases, as different from one another as they are from Williams Cemetery. The position of Turpin is of particular interest; it is not closely associated with any other sample. Major essential elements (Ca, P) appear to make a major contribution to the pattern of variation presented in Figure 2. The calciumphosphorus ratios of these samples range from 1.50 (Turpin) to 1.90 (State Line). The ratio value is 1.62 in Williams Cemetery, 1.66 in Anderson Village, 1.57 in Indian Hills, and 1.67 in Pearson Village. The correspondence display's first axis appears to disperse populations of high and low ratios. The second axis delineates the samples geographically. Southwestern sites (Anderson Village, State Line, and Turpin) have positive coordinates with respect to the origin, and northwestern sites (Indian Hills, Pearson Village, and Williams Cemetery) have negative second axis coordinates. This observation suggests that geographic location and associated local conditions play important roles in discriminating these groups. Figure 3 represents the graphic results of correspondence analysis using dental caries frequency (Table 1).The first and second axes explain 48.32% and 35.80%of the variance, respectively. The sample with the lowest caries frequency is the hunting-gathering-fishing population, Williams Cemetery. Similarly, the sample with the highest caries frequency, Turpin, is separated clearly from all other samples. Comparing the two correspon- 100 K.N. SCHNEIDER mTU SL. Tixi? Fig. 2. Correspondence display of all elements and populations sampled. IH, Indian Hills; WC, Williams Cemetery; PE, Pearson Village; AN, Anderson Village; SL, State Line; TU, Turpin; all elements are represented by their chemical symbol. d ml Fig. 3. Correspondence analysis display of caries frequency by quadrant using Table 1data matrix of five populations sampled. xant, Maxillary anterior; xpost, maxillary posterior; dant, mandibular anterior; dpost, mandibular posterior. DENTAL CARIES AND ENAMEL COMPOSITION dence analysis displays, the relative positions of the archaeological sites in each independent test are very similar. 101 tested using alternative analytic approaches such as Cl2/CI3 ratios of skeletal samples from the site. The conclusion derived from these data is that, apparently, differential DISCUSSION AND CONCLUSIONS use of horticultural products was a local pheThe results presented indicate significant nomenon and not one prescribed by the larger variation in enamel composition correspond- cultural network. ing to different caries experience among the ACKNOWLEDGMENTS populations examined. This suggests that particular elements (Zn, Cu, and Fe) have a I would like to thank the following individcariostatic effect and that others (especially uals and institutions for allowing me to study Nil have a cariogenic effect when present in materials in their care: Paul W. Sciulli and dental enamel. This observation corroborates Jonathan Bowen (The Ohio State Univercurrent evaluations of these elements' effects sity), David Stothers (University of Toledo), when present in dental enamel (Curzon and Martha Potter-Otto (Ohio Historical Center), Cutress, 1983). and Anthony J. Perzigian (University of CinThe shift from a hunting-gathering-fishing cinnati). "hanks are also due Weldon Padsubsistence base to one focused on horticul- gett I11for the illustrations, the Instructional ture not only increased caries by introducing and Research Computing Center at the Ohio readily fermentable carbohydrate foodstuffs State University for providing computing (which could have increased bacterial de- funds, and the two reviewers who offered struction of dental enamel) but also affected constructive comments. A portion of this reacquired dental pathology expression by al- search was supported by a Presidential Feltering the availability of elements present lowship from the Ohio State University and during enamel calcification. This appears to a University Sponsored Research Grant from be expressed particularly strongly by the The Wichita State University. various calcium-phosphorus ratios observed LITERATURE CITED among these samples. The tentative conclusion from these data, then, is that the intro- Aufderheide, AC, Angel JL, Kelley, JO, Outlaw, AC, Outlaw, MA, Rapp G, and Wittmers, LE (1985) Lead duction and later the dependence on in bone 11. Prediction of social correlates from skeletal horticulture (particularly maize) had at least lead content in four colonial American populations these two major results. Dental enamel (Catoctin Furnace, College Landing, Governor's Isamong horticulturalists may be dissolved land, and Irene Mound). Am. J. Phys. Anthropol., 66:353-361. more readily by bacterial acid because of imFJ, and Loutit, JF (1964)The entry of strontiumpaired crystal structure. Research is now Bryant, 90 into human bone. Proc.R. Soc.Land. [Biol.] 159:449being conducted using X-ray diffraction to 465. establish if and how the unit cell of dental Barber, RJ (1978) A lithic analysis of the Anderson Vilenamel is altered by these types of ionic lage Site (33Wa4),A Fort Ancient culture site in Warren County, Ohio. Mid-Cont. J. Archeol3:189-213. substitutions. SN (1980) Orban's Oral Histology and EmFive of the samples used herein are consid- Bhaskar, bryology, 9th ed. St. Louis. C.V. Mosby. ered as having practiced horticultural sub- Braun, DJ (1982)A study of two dental pathologies in a sistence, each participating in a regional Ft. Ancient population. Paper presented at the Central exchange and cultural system (the Fort AnStates Anthropological Society meeting, April 10,1982. cient or Sandusky Bay Tradition). This study Cohen, S, and Pieper, SC (1979) Speakeasy Manual OSi VS Version. Chicago: Speakeasy Computing documents intersite/intracultural complex Corporation. diversity among these populations in both Colby, JW (1971)MAGIC-Microprobe Analysis General enamel composition and dental caries expe- Intensity Corrections-PROGRAM. Allentown PA. Bell Telephone Laboratories, Inc. rience. The cultural diversity within Fort Ancient (Essenpreis, 1978) appears to be ex- Cook, DC, and Buikstra, JE (1979) Health and differential survival in prehistoric populations: Prenatal den. pressed by various patterns of subsistence tal defects. Am. J. Phys. Anthropol. 51:649-664. strategy probably caused by selective inclu- Cox, GJ, (1952) Oral environment and dental caries, In: sion or exclusion of local or nonlocally de- A Survey of the Literature of Dental Caries. Washingrived cultigens. This scenario suggests that ton, DC: Academy of Sciences, National Research Council. pp. 243-324. the people represented at the Turpin site were involved more extensively in horticul- Curzon, MEJ, and Cutress, TW (eds) (1983) Trace Elements and Dental Disease. Boston: John Wright. tural activities (i.e., were more maize-depen- Dunning, J M (1953) The influence of latitude and disdent) than were the other Fort Ancient tance from the seacoast on dental disease. J. Dent. Res. groups examined. This hypothesis can be 322311-829. 102 K.N. SCHNEIDER Durfor, CN, and Becker, E (1962) Public Water Supplies of the 100 Largest Cities of the U.S.A. Water Supply Paper 1812. Washington, DC US. Government Printing Office. Essenpreis, P (1978) Fort Ancient settlement: Differential response at a Mississippian-Late Woodland interface. In: BD Smith (ed): Mississippian Settlement Systems. New York: Academic Press; pp 141-167. Foreman DW, and Awadalla, E (1978) Background corrections for electron microscopy and X-ray analysis. Scan. Electron Microsc. 1:109-132. Greenacre, MJ (1894) Theory and Applications of Correspondence Analysis. New York Academic Press. Griffin, JB (1966) Eastern North American archaeology: A summary. Science I56:175-191. Guthrie, HA (1979) Introductory Nutrition. St. Louis: C.V. Mosby. Hardwick, JL.(1960) The incidence and distribution of caries throughout the ages in relationship to the Englishman’s diet. Br. Dent. J. 108:9-17. Hercus, C (1925)Epidemic goiter in New Zealand and its relation to the soil iodine. J. Hygiene 24:321-402. Hurny, TA (1978) Fluoride and teeth. In B Covrvoisier, A Donath, and CA Baud (eds): Fluoride and Bone. Bern; Hans Hubert Publishers, pp. 119-124. Lambert, JB,Simpson, SV, Buikstra, JE,and Hanson, D (1983) Electron microprobe analysis of elemental distribution in excavated human femurs. Am. J. Phys. Anthropol. 62:409-423. Lambert, JB, Szpunar, CB, and Buikstra, JE (1979) Chemical analysis of excavated human bone from Middle and Late Woodland sites. Archaeometry 21:115129. Lambert, JB,Vlasak, SM, Thometz, AC and Buikstra, JE (1982) A comparative study of the chemical analysis of ribs and femurs in Woodland populations. Am. J. Phys. Anthropol. 59:289-294. Nichols, MS and McNall, DR (1957) Strontium content of Wisconsin municipal waters. Am. Water Works Assoc. J 49:1493-1501. Pfeiffer, S (1979) The relationship of buccal pits to caries formation and tooth loss. Am. J. Phys. Anthropol. 50:35-38. Pike, RL, and Brown M (1975) Nutrition: An Integrated Approach. 2nd ed. New York: John Wiley and Sons. Rose, JC (1977)Defective enamel histology of prehistoric teeth from Illinois. Am. J. Phys. Anthropol. 46:439446. SAS Institute (1982) SAS User’s Guide: Basics, 1982 ed. Cary, NC SAS Institute, Inc. Sciulli, PW (1978) Developmental abnormalities of the permanent dentition in prehistoric Ohio Valley Amerindians. Am. J. Phys. Anthropol. 48:193-198. Sciulli, PW, Aument, BW, and Piotrowski, LR (1982) The Williams (33Wo7a)Red Ocher Cemetery: Preliminary descriptive and comparative analysis of acquired dental pathology. Penna. Archeol. 52:17-24. Sciulli, PW, Piotrowski, LR, and Aument, BW (1983) The Williams (33Wo7a) Red Ocher Cemetery: Preliminary investigation of stature. Penna. Archaeol., 53:42-46. Sciulli, PW, Piotrowski, LR, and Stothers, D (1984) The Williams Cemetery: Biological variation and affinity with three Glacial Kame groups. N. Am. Archaeol. 5:139-170. Sillen, A (1981) Strontium and diet at Hayonim Cave. Am. J. Phys. Anthropol. 56:131-137. Stothers, DM (1973) Radiocarbon dating the culture chronology of the western Lake Erie Basin. Toledo Area Aboriginal Res. Bull. 226-42. Stothers, DM (1979) The Western Basin Tradition: Algonquin or Iroquois? Penna. Archaeol. 49:(3):13-30. Stothers, DM, and Pratt, GM (1981) New perspectives on the Late Woodland cultures of the western Lake Erie region. Mid-Cont. J. Archeol. 6:91-121. Vaughan, JM (1970) The Physiology of Bone. New York: Academic Press, p. 99. Wheelersburg, RP (1981)Paleoethnobotanical Analyses of State Line, Hahilton County, Ohio (33 Ha58). MA thesis, University of Cincinnati.