AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 80:137-143 (1989)

Fluorochrome Labelling in Roman Period Skeletons From Dakhleh Oasis, Egypt

MEGAN COOK, EL MOLTO, AND C. ANDERSON Department of Pathology, University of Western Ontario, London, Ontario, Canada N6A 5CI (M.C., CA3; Department of Anthropology, Lakehead University, Thunder Bay, Ontario, Canada P7B 5El (E.M.)

KEY WORDS tion

Bone histomorphometry, Histomorphometry, Infec-

ABSTRACT A histological aging study of femoral midshafts in a late Roman period sample from the Dakhleh Oasis, Egypt, showed discrete fluoro- chrome labelling. The fluorochome is yellow-green in colour and fluoresces at a wavelength of 525 nm. The labelling occurs at the mineralization fronts and is so distinct that several histomorphometric measurements were made. The percent labelled surface values ranged from 6.03% to 59.34%, and the mean distance between labels ranged from 13.57 to 20.63 Fm. Teeth from several individuals were also labelled within the enamel matrix. Comparisons were made between the patterns of fluorescent labelling from this population and from patients treated with interval and continuous dosages of tetracycline. The results collectively indicate that the fluorochrome is most likely tetracy- cline that made its way into the bone (in vivo) via stored grain contaminated by Streptomycetes. The labelling was differential, much like that of the patient on interval doses of tetracycline, so it is argued that the tetracycline was ingested occasionally, probably on a seasonal basis. The lack of bone infection in the adults (n = 29) suggests that the tetracycline may have provided some antibiotic protection in this population.

In the 18th century, dye from the madder plant root (alizarin red) was known to label bone and was then used to study bone min- eralization (Hunter, n.d.). Today, several fluorochromes are used in bone studies [e.g., calcein blue, xylenol orange, calcein (green), synthetic alizarin red complexone, and tetra- cycline] (the following are but a few of the plethora of studies using these bone markers for diagnostic and research purposes: Frost et al., 1967; Sanger and Holt, 1965; Ander- son and Danylchuk, 1978; Jorch and Ander- son, 1980; Hendel et al., 1982; Hori et al., 1985; Tam and Anderson, 1980; Antonyshyn et al., 1987).

In 1958, Milch et al. reported the uptake of tetracycline antibiotics by actively miner- alizing bone tissue. Tetracycline is now the most extensively used fluorochrome in the study of both normal And abnormal (dis- eased) bone growth and the metabolic bone diseases. Because tetracycline forms com- plex, permanent compounds at the mineral-

ization fronts, static and dynamic histo- morphometric parameters have been established in modern samples, and these offer potential for investigating bone dynam- ics in past populations.

In 1980, Bassett et al. first reported the presence of what they believed was “in vivo” tetracycline in archeaological bone from a sample in Nubia (dated circa AD 350). This hypothesis has been challenged by research- ers favouring taphonomic processes to ex- plain the mode of fluorochrome incorpora- tion into ancient bone. In particular, Piepenbrink et al. (1983) have shown that the secondary metabolites from many soil microorganisms (e.g., the fungus Struchy- botrys) contaminate bone and produce a tet- racycline-like fluorescence. This paper ad- dresses this in vivo vs. in vitro mode of tetracycline labelling by comparing the pat-

Received November 14,1988; accepted January 23,1989.

@ 1989 ALAN R. LISS, INC.

138 M. COOK ET A L

tern of fluorochrome labelling in a recently excavated skeletal sample from Egypt to a modern tetracycline-treated clinical sample. Our data support the in vivo hypothesis, and we argue that this finding has significance for the interpretation of the palaeoepidemi- ology of this and other Middle East skeletal populations.

MATERIALS AND METHODS

The skeletal remains are from a late Ro- man period tomb (circa AD 400-500) in the Dakhleh Oasis, Egypt. The oasis is located 660 km SSW of Cairo at 25" 33' N, 28" SSE and is 100 km east to west by 25 km north to south. The crypt was excavated in the 1986 field season of the Dakhleh Oasis Project (D.O.P.) now in its eleventh year of studying the human biocultural adaptations to this Saharan ecozone. The 40 individuals ex- humed are relatively homogeneous in their morphological characteristics and likely shared very close kinship bonds (Molto, 1988). Standard observations of morphology and pathology were recorded in the field. Bone samples were obtained from the femo- ral heads and midshafts of adults for histo- morphometric analysis. Three permanent teeth from different individuals were also taken for analysis in the laboratory. It was during the processing of the adult femoral midshafts (for osteon aging) that the fluoro- chrome labelling was first noted.

The comparative modern sample was col- lected at University Hospital, London, On- tario, Canada. The sample consisted of indi- viduals (n = 25) in a metabolic bone disease research project. The histomorphometric analysis of the Dakhleh bone involved ex-

tracting 200 km of the femoral midshaft cross sections cut initially with the Isomet, then ground down to 25-50 km on the Maruto lapping machine. All sections were then analysed under ultraviolet light (light source HBO 200).

The following bone histomorphometric pa- rameters were measured using a semiauto- matic image analyser (Carl Zeiss, Toronto, Ontario, Canada; Video Plan programme). Table 1 summarizes the nomenclature, ab- breviations, and methodology used in this paper.

1. Distance between labels: This is mea- sured in micrometers and is the mean width between double fluorescent labels.

2. Mineral apposition rate: This value is obtained by dividing the distance between double Haversian labels by the time interval between tetracycline doses in modern clini- cal patients, calculated as Distance between labeldtime interval. The normal value is 1 p d d a y (Frost et al., 1961).

3. Total length of canals: This is the sum of all the lengths of the Haversian canals mea- sured in millimeters. 4. Percent surface labelled: This is the sum

of the lengths of double and single fluores- cent labels, with the resultant value giving the percent of the surface labelled. % Surface labelled = (First and second labeY2 + Single label x 100)/Total length of canal.

5 . Bone formation rate: This measures how many cubic millimeters of new bone is made per cubic millimeter of total bone per year and is presented as a percentage, calcu- lated as follows: Bone formation rate (BFR) (mm3/mm2/year) = Mineral apposition rate X % Labelled surface x 365/1,000.

TABLE I . Abbreuiations, nomenclature, and methodology used in the histomorphometric analysis

Parameters Units Abbreviations Method of calculation

Distance between w DBL Mean width between double fluorescent labels

Mineral appositional rate pm MA Distance between double labels divided by the

Total canal length' mm TLC Sum of the lengths of the Haversian canals Percent surface labelled % SL Sum of the length of the fluorescent labels,

calculated as follows: (First and second label/ 2 + Single label X 100)/Total canal length

Bone formation rate2 mm3/mm2/year BFR How many mm3 new bone is made per mm2 tissue surface per year calculated as: Mineral appositional rate X % Surface labelled X 365/1,000.

fluorescent labels'

time interval between tetracycline doses

'Fifty fields measured on the semiautomatic image analyser. 'Estimates only.

139 FLUOROCHROME IN ROMAN PERIOD SKELETONS

RESULTS The bone sections from Dakhleh fluo-

resced yellow-green at 525 nm, the same wavelength as tetracycline (see Table 2). Figures 1 and 2 illustrate the pattern of fluorescence in two specimens, an adult fe- male (aged 39 years) and a young adult (aged 19 years), respectively. Both show that there is differential labelling of the osteons, the actively mineralizing osteons being labelled most clearly. In the subadult, there is also subperiosteal drifting and labelled lamellar bone consistent with the modeling sequence of growing bone. Table 3, which summarizes the histomorphometrics of the adults, shows that the differential labelling of the osteons is a constant pattern of the Dakhleh speci- mens. In fact, every individual in the sample has fluorochrome labelling. The D.B.L. val-

TABLE 2. Fluorochromes

Substance Emission in bone (nm)

Calcein blue 445 Xylenol orange 615 Calcein 540 Alizarian complexone 625 Tetracycline 525

ues range from a low of 13.5 to a high of 20.67 pm, and the percent surface labelled values range from 6.04% to 52.35%. The latter pa- rameter in particular shows increased % surface labelled with young individuals whose bone is actively modeling (Fig. 2), although the state of an individual's health has a pervasive effect on each of these pa- rameters and must be considered in any interpretation of these data. This is best illustrated in individual Al , a female in her fifties, with an exceptionally low BFR (.033 mm3/mm2/yr). It must be emphasized that the BFR volumes are estimates only. As we noted previously, she likely suffered from hyperparathyroidism (Cook et al., 19881, and this disease profoundly alters the rate of bone mineralization.

Figures 3 and 4 illustrate the nature of the tetracycline labels in the clinical sample. The patient in Figure 3 (the one with the interval dosages) shows a pattern of label- ling very similar to the Dakhleh individuals. The patient with the continuous dosage (Fig. 4) also displays actively mineralizing os- teons, although the separate tetracycline la- bels are not distinct. Clearly this pattern differs from those of the Egyptian sample and the first patient.

Fig. 1. Differential osteon fluorochrome labelling in Egyptian remains (DK-31).

140 M. COOK ET AL.

Fig. 2. Young egyptian (DK-31) remains showing labelled lamellar bone.

TABLE 3. Dakhleh Oasis: Histomorphometric Values

Individual ium) (%) (mm3/mm2/vear) DBL SL BFR

Males

Y 14.81 18.68 ,068 N 20.67 6.03 ,022 G 17.96 45.53 ,166

M 19.13 39.92 ,145 A9 16.70 31.59 ,115 W 14.87 38.09 ,139 I 18.98 29.10 ,106 A1 15.73 9.19 ,033 S 16.95 21.58 ,078 H 18.50 18.95 ,069 A3 16.99 59.34 216 A6 15.96 40.66 ,148

Females

A7 Single labels 52.34 ,191

Figure 5 illustrates tetracycline labelling in an isolated tooth from an adult individual. Note that the labelling is within the enamel matrix and is not limited to the external margin. The other permanent teeth in the sample (n = 3) also show this pattern. Of additional importance is the fact that there

is no macroscopic evidence of tetracycline staining in the deciduous or permanent teeth.

DISCUSSION

As was noted, Bassett et al. (1980) re- ported in vivo tetracycline labelling in skele- tal remains from Nubia. They suggested that a tetracycline producing moldlike bacte- rium, Streptomycetes, contaminated stored grain, which was later consumed, eventually being incorporated into the bones of these people.

The critique of this hypothesis advanced by Piepenbrink et al. (1983) in their experi- mental study of postmortem infestation of bone centered on the fact that bone contami- nation by microorganisms can produce a tet- racyclinelike fluorescence. However, close analysis of their paper reveals salient differ- ences relative to the data reported here. As noted, the Egyptian bone without exception is differentially labelled. The latter no doubt reflects age variation in the rate of mineral- ization plus the fact that disease states that influence bone mineralization vary between the individuals. Regardless, there is no case of the diffuse labelling recorded by Piepen- brink and his colleagues, which we also ob-

FLUOROCHROME IN ROMAN PERIOD SKELETONS

Fig. 3. Differential osteon tetracycline labelling in a modern individual.

Fig. 4. Thick tetracycline bands in patient with acne treated with continuous doses of tetracycline.

141

142 M. COOK ET AL.

Fig. 5. Tooth from DK-31 showing fluorochrome labelling.

served in the patient with continuous dosage of tetracycline. The pattern in the Dakhleh bone clearly represents periodic labelling such as that found in the modern clinical patient (Fig. 3). The concentric differential labelling is unequivocal support for an in vivo mode of incorporation. The fact that the teeth were labelled within the matrix also supports the in vivo hypothesis, since enamel, unlike bone, is avascular and is more resistant to contamination deep in its structure by taphonomic processes.

The following questions emerge in consid- ering the in vivo hypothesis. How often and in what circumstances was the tetracycline consumed? Were the amounts of tetracycline sufficient to provide a therapeutic effect on the population? From the differential label- ling data, it appears that the tetracycline was ingested periodically. Because it is im- possible to determine the amount of time represented between labels in given individ- uals, we cannot equivocally determine if there are specific cycles when tetracycline was ingested. However, in considering the Bassett et al. (1980) hypothesis on the gene- sis of tetracycline in Middle East popula- tions, it is our opinion that the contaminated grain was likely consumed when fresh food

supplies were limited. It is during these periods when the contaminated grain in the bottom of the containers would have most likely been consumed. Ironically, this corre- sponds to the time when the tetracycline is most needed, namely, when nutritional and therefore disease stress are likely to be greatest.

The palaeoepidemiological data from DK31 show that bone infection was absent in the adults (n = 29) and was high in the subadults (n = 11). In the latter sample, each individual <4-5 years of age had unremod- elled periosteal reactions, whereas, in the older subadults, the lesions were remodel- ling and/or remodelled. According to current theory, reactive periostitis is a primary re- sponse to infectious agents (Mensforth et al., 19781, so it is reasonable to suggest that if the tetracycline did have a therapeutic role in this population, its influence was re- stricted to the older subadults and adults. The total lack of bone infection in the adults in a sample with considerable nutritional stress and other types of pathology (three adult males with well healed multiple frac- tures) is provisionally suggestive of a thera- peutic role for tetracycline. It is germane to point out that, although it is impossible to

FLUOROCHROME IN ROM

determine the amounts of tetracycline in- gested at a given time, these amounts would likely have been considerably less than mod- ern prescribed dosages. However, in this Egyptian sample, the tetracycline func- tioned in a preventative role rather than as a treatment for an existing infection. In pre- vention, smaller periodic doses could con- ceivably work with the immune system to halt the progression of infection, especially infections of an endogenous nature such as staphylococcal-caused osteomyelitis. We should add that to date over 150 adults from this cemetery have been analyzed, and no bone lesions that can be attributed to infec- tious disease processes have been identified. These data collectively support our hypothe- sis of antibiotic surveillance. An alternative hypothesis is that the desert’s arid environ- ment provided some degree of protection from infectious microorganisms. The ecology does not encourage the proliferation of di- verse microorganisms.

CONCLUSIONS

The differential pattern of fluorochrome labelling that characterizes the Egyptian bone specimens that were examined in the study supports the hypothesis of an in vivo mode of incorporation. In addition, wave- length fluorescence at 525 pm provides strong evidence that the specific fluoro- chrome is tetracycline. Information about the age-related distribution of periosteal re- actions in our sample suggests that ingested tetracyclines had an antibiotic andlor pro- phylactic effect on adults. If further research supports this hypothesis, then future pa- leoepidemiological studies of past Middle East populations must evaluate the extent to which natural antibiotics effected the health status and demographic characteristics of these earlier human groups.

ACKNOWLEDGMENTS

Our thanks to Prof. Anthony Mills, Direc- tor of the Dakhleh Oasis Project; to Betty

AN PERIOD SKELETONS 143

Gardiner for typing the manuscript; and to Dr. Mei-Shu Shih for advice and support.

LITERATURE CITED

Anderson C, Danylchuk K (1978) Appositional bone formation rates in the beagle. Am. J . Vet. Res.

Antonyshyn 0, Colcleugh RG, Anderson C (1987) Growth potential in onlay bone grafts: Acomparison of vascularized and free calvarial bone and suture bone grafts. Plast. Reconstruct. Surg. 79:12-23.

Bassett EJ, Keith M, Armelagos G, Martin D, Villanueba A (1980) Tetracycline-labeled human bone from an- cient Sundanese Nubia (AD 350) Science 209: 1532-1534.

Cook M, Molto E, Anderson C (1988) A possible case of hyperparathyroidism in a Roman skeleton from the Dakhleh Oasis, Egypt, diagnosed using bone histo- morphometry. Am. J. Phys. Anthropol. 75:23-30.

Frost HM, Villanueva A, Roth H, Stanisavljevic S (1961) Tetracycline bone labelling. J New Drugs 1 :206-216.

Hendel PM, Hattner RS, Rodrigo J , Buncke HJ (1982) The functional vascular anatomy of the rib. Plast. Reconstruct. Surg. 70:578-584.

Hori M, Takahashi H, Konno J , Inque J , Habra T (1985) A classification of in vivo bone labels after double labelling in canine bones. Bone 6:147-154.

Hunter J (n.d.) Archives of Welcome Museum, Royal College of Surgeons, Lincoln’s Inn Fields, London, England.

Jorch U, Anderson C (1980) Haversian bone remodelling measurements in young beagles. Am. J. Vet. Res.

Mensforth RP, Lovejoy CO, Lallo JW, Armelagos GJ (1978) The role of constitutional factors, diet and infectious disease in the etiology of prorotic hyper- osteotitis and periosteal reactions in prehistoric in- fants and children. Med. Anthropol. 259.

Milch R, Koll DP, Tobie JE (1958) Fluorescence of tetra- cycline antibiotics in bone. J. Bone Joint Surg

Molto JE (1988) Skeletal Remains from the Dakhleh Oasis, Egypt. Toronto: Society for the Study of Egyp- tian Antiquities Publication.

Piepenbrink H, Herrmann B, Hoffmann P (1983) Tetra- cyelintypische Fluoreszenzen an bodengelagerten Skeletteilen. Z. Rechtsmed. 91:71-74.

Recker R (ed) (1986) Bone Histomorphometry: Tech- nique and Interpretation. Boca Raton, FL: C.R.C. Press.

Sanger VL, Holt A (1965) Experimental tetracycline labelling in avian osteopetrosis. Can. J. Comp. Med. Vet. Sci. 29,245-252.

Tam T, Anderson W (1980) Tetracycline labelling ofbone in vivo. Calcif. Tissue Int. 30:121-125.

4~k907-9 10.

41 :1512-1515.

40Ac897-909.