- Email: [email protected]
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena (Granada, Spain)
Journal of Archaeological Science (2002) 29, 167–175 doi:10.1006/jasc.2001.0701, available online at http://www.idealibrary.com on
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena (Granada, Spain) Jesu´s M. Torres, Concepcio´n Borja and Enrique G. Olivares* Unidad de Inmunologı´a, Instituto de Biotecnologı´a, Universidad de Granada, Spain (Received 3 April 2000, revised manuscript accepted 2 March 2001) We studied the presence of immunoglobulin G (IgG) in equid fossil bones from Venta Micena (Granada, Spain), dated at 1·6 Myr old, and from Atapuerca (Burgos, Spain) (0·12 Myr), and in ostensibly hominid fossils VM1960 and VM3691 (1·6 Myr) from Venta Micena and CV1 and CV2 (1·4 Myr) from Cueva Victoria (Murcia, Spain). Samples taken from the specimens were crushed to a fine powder and their mineral structure was dissolved in an ethylenediaminetetracetic acid (EDTA) solution. Fossil extracts were tested with antibodies against human IgG and against horse IgG with two independent immunological methods: dot-blotting (DB) and a modification of this latter method: quantitative dot-blotting (QDB). IgG was detected by DB and was quantifiable by QDB in some of the fossils tested. Equid fossils from Atapuerca and Venta Micena gave stronger reactions with the antibodies against horse IgG than with the antibodies against human IgG. Fossils VM3691 and VM1960 reacted more strongly with the antibodies against human IgG than with antibodies against horse IgG, whereas no IgG was detected in fossils CV1 and CV2. These findings show that species-specific IgG can be detected in fossils as old as 1·6 Myr. The immunological analysis of fossil proteins may help to solve palaeontological controversies. 2002 Elsevier Science Ltd. Keywords: IGG, HOMINID, EQUID, VENTA MICENA, ATAPUERCA, FOSSIL BONE, DOT-BLOTTING, QUANTITATIVE DOT-BLOTTING.
With respect to proteins, Bada et al. (1999), based on the analysis of Pleistocene mammalian bone from Olduvai Gorge, came to conclusion that ‘‘original protein components should be hydrolysed to free amino acids which then diffuse out of the fossil matrix and are lost over time-scales of 105–106 years in most environments’’. Nevertheless, this inference is contradicted by the many results that show that proteins, or at least pieces of them, can be identified by immunological methods in fossil bones millions of years old. Lowenstein, using a radioimmunoassay (RIA), detected species-specific collagen and serum factors in fossils as old as a 0·5-Myr-old Homo erectus, a 1·9Myr-old Australopithecus robustus (Lowenstein, 1981) and an 8-Myr-old Ramapithecus (Lowenstein, 1983). Furthermore, collagen has been found by dot-blotting in a 10-Myr-old bone (Rowley et al., 1986); and osteocalcin has been extracted from 13-Myr-old fossil bovid bones, from 30-Myr-old fossil rodent teeth (Ullrich et al., 1987), and from 75-Myr-old dinosaur bones (Muyzer et al., 1992). Osteocalcin from the fossil bovid bones still retained functionally active gammacarboxyglutamic acid residues after 13 Myr (Ulrich et al., 1987). Seventy-Myr-old molluscs also yielded positive immunological reactions (De Jong et al., 1974), and collagen fibrils were observed with electron microscopy in a 200-Myr-old dinosaur bone (Wykoff, 1972). A striking recent finding was the antisera
Introduction umerous reports have demonstrated that species-specific biomolecules can persist in archaeological material (Newman et al., 1996) and in some fossils (Lowenstein & Scheuenstuhl, 1991). These molecules are often scarce and difficult to identify with certainty; however, when they persist, analyses of proteins and DNA may be a reliable way to study archaeological sites (Newman et al., 1993) or to classify fossils, particularly when anatomical interpretations are discrepant (Smith & Littlewood, 1994). DNA has been detected in samples thousands of years old (Pa¨a¨bo et al., 1989; Hoss et al., 1996), and mitochondrial DNA of a Neanderthal bone (40,000 years old) has been sequenced (Krings et al., 1997, 1999). However, the rapid degradation of this biomolecule in bone has led some authors to suggest that the detectability of DNA in fossil bones is limited to specimens no older than a 100,000 years (Poinar et al., 1996; Lindhal, 1997). Nevertheless, several studies demonstrated the presence of species-specific DNA millions of years old in fossil plants (Poinar et al., 1993) and in insects preserved in amber (Cano et al., 1993). A more recent report suggested, however, that the presence of ancient DNA in amber fossils is doubtful (Austin et al., 1997).
N
*Tel. and fax: 34-958-243522; E-mail: [email protected]
167 0305–4403/02/020167+09 $35.00/0
2002 Elsevier Science Ltd.
168
J. M. Torres et al. Table 1. List of fossils studied in this work
Fossil
Assigned to:
Anatomical location
Age (Myr)
Found in:
VM1960 VM3691 CV1 CV2 VM3032 VM3064 VM3037 VM3008 VM3016 AT83 GSU-9. E.21–4 (AT83·1)* AT83 GSU-9. F.16–33 (AT83.2)* AT85
Hominid Hominid Hominid Hominid Equid Equid Equid Equid Equid Equid Equid Equid
Humerus fragment Humerus fragment Humerus fragment Humerus fragment Tibia fragment Tibia fragment Tibia fragment Metatarsal bone Metatarsal bone Right femur Mandible Sacrum
1·6 1·6 1·4 1·4 1·6 1·6 1·6 1·6 1·6 0·12 0·12 0·12
Venta Micena, Orce Venta Micena, Orce Cueva Victoria, Murcia Cueva Victoria, Murcia Venta Micena, Orce Venta Micena, Orce Venta Micena, Orce Venta Micena, Orce Venta Micena, Orce Atapuerca, Burgos Atapuerca, Burgos Atapuerca, Burgos
*In parentheses is the name given to the fossil in this article. Two independent samples were taken from the surface and interior of fossils AT83.1 and AT83.2. These samples were named AT83.1S and AT83.2S (superficial samples) and AT83.1I and AT83.2I (interior samples).
obtained in rats injected with dinosaurian tissues which gave positive reactions with purified avian and mammalian haemoglobins; this result showed that fragments of haemoglobin were preserved within the dinosaur tissues (Schweitzer et al., 1997a). In contradiction to the claim of Bada et al. (1999), all these results clearly show that proteins from some fossil bones, although fragmentary, can persist and retain enough of their immunological properties to provide useable genetic information for periods of millions of years. Preservation seems to be favoured by the fact that proteins are embedded in the mineral phase of the bone, as this provides considerable protection from degradation by environmental conditions (Smith et al., 1985; Berman et al., 1988). The mineral phase also appears to preserve, by diffusion, even the cartilage of some fossil bones (Franc et al., 1995). The observation of cells and histological structures preserved in dinosaur bones is also consistent with the presence of biomolecules (Schweitzer et al., 1997b; Pawlicki & Nowogrodzka-Zago´rska, 1998). Compared with collagen and osteocalcin—the most abundant bone proteins—albumin or immunoglobulin G (IgG), although present in lower concentrations in bone, provide much more evolutionary information given their faster rate of evolution (Wilson et al., 1977). Albumin has helped to resolve the disputed phylogenic affinities of extinct mammoths, mastodons, Steller’s sea cow, the Tasmanian wolf and the quagga (Lowenstein, 1981; Lowenstein et al., 1981; Lowenstein & Ryder, 1985; Rainey et al., 1984; Shoshani et al., 1985; Lowenstein & Scheuensthul, 1991). In some instances, plasma proteins appear to be better preserved than collagen in bone, so that the bone becomes enriched in plasma proteins above in vivo levels (Tuross, 1989; Tuross et al., 1989). In a previous study our group at the University of Granada and Lowenstein’s group at the University of California in San Francisco independently showed that species-specific albumin can
be detected by enzyme-linked-immunosorbant assay (ELISA) or RIA in fossils of different species collected at Venta Micena (Granada, Spain) (Borja et al., 1997). These independent studies supported the human assignation of controversial fossils such as skull fragment VM0 and humerus fragment VM1960, both dated at about 1·6 Myr (Zihlman & Lowenstein, 1996; Palmqvist, 1997; Gibert et al., 1998; Tobias, 1998). In the present study we show that immunoglobulin G (IgG) could also be detected by dot-blotting (DB) and a modification of this method called quantitative dotblotting (QDB) in 0·12-Myr-old equid fossil bones from Atapuerca and 1·6-Myr-old equid and hominid fossil bones from Venta Micena, Orce, Granada.
Materials and Methods Fossil and bone extracts The fossils studied are listed in Table 1. Aseptic conditions were rigorously maintained and disposable materials were used to avoid external contamination during the process of extraction and testing. Unless otherwise indicated, samples were taken from the interior of the fossils. The samples were ground to a fine powder and independently maintained in phosphate-buffered saline (PBS) for one week in continuous agitation at room temperature (washing). The extracts were centrifuged, and the supernatants were collected, divided and stored at 70C until use. The remaining bone sediment of each sample was then decalcified in a solution of 0·2 M EDTA, pH 7·4, following the same procedure (agitation, centrifugation) as before. Finally, the decalcified sediment was also extracted with 0·5 M acetic acid with the same procedure. Supernatants were examined under light microscope and those that showed bacterial growth were discarded. From each fossil, we therefore obtained PBS, EDTA and acetic acid aliquots. Equid
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena 169
fossils were extracted starting at 5 mg bone per ml of PBS, whereas hominid fossils were extracted at 3·5 mg/ ml. A 100-year-old vertebra of a donkey found in Orce (near Venta Micena) was extracted at 15 mg/ml and at 3 mg/ml in two independent processes.
Detection of IgG by quantitative dot-blotting Because DB gives only semiquantitative results (intensity of staining) we modified this method to obtain fully quantitative results. All samples (run in duplicate) were placed inside circles that were previously scored on the nitrocellulose membrane with a needle. We followed the same steps as in DB until the anti-IgG antibodies were added. We then cut out the circles of nitrocellulose within which the reaction took place and transferred each circle to a well of a microtitre plate. A solution of 0·4 mg/ml of orthophenylenediamine dihydrochloride (Sigma) in phosphate citrate buffer was then added and absorbance was read at 490 nm with a microplate autoreader. Instead of the fossil extracts, solutions used for fossil extractions were used as negative controls. The absorbances obtained with the fossil extracts in the immunological direct and cross reactions were extrapolated against human and horse IgG calibration curves to determine the amount of protein in the fossils. When the fossil corresponded to or was close to the species to which the antibodies were directed, the amount of IgG detected was larger than when antibodies to a non-related species were used (Figure 1). Although useful for comparisons between species, the amounts of protein detected in a fossil cannot be considered at face value, as these fossil proteins are probably split and denatured, in contrast with the intact native modern human and horse IgG that were
Absorbance (mU)
1200
900
600
300
0 0.01
0.1
10
100
1 10 IgG (ng-equivalents)
100
1 Horse IgG
1500
1200 Absorbance (mU)
Detection of IgG by dot-blotting Five microlitres of each fossil extract or of a solution of modern human or horse IgG was placed on a nitrocellulose membrane (Biorad, Richmond, CA, U.S.A.) and incubated for 30 min at room temperature. All samples were tested in duplicate. After washing with a solution of 0·05% PBS–Tween-20, the membrane was blocked with a solution of 1% gelatin in PBS for 1 h at 37C and overnight at 4C. The appropriate anti-IgG antibodies (affinity isolated goat anti-human IgG or rabbit anti-horse IgG polyclonal antibodies labelled with peroxidase, Sigma, St Louis, MO, U.S.A.) were then added and the membrane incubated for 1 h at room temperature. After incubation the membrane was washed again with PBS–Tween and the antigen– antibody reactions were revealed with a solution of 0·05% 3,3 -diaminobenzidine (Sigma) and 0·003% H2O2 in PBS for 5–15 min incubation time. The enzymatic reaction was stopped by washing with distilled water. Solutions used for fossil extractions (PBS, EDTA and acetic acid) were used as negative controls.
Human IgG 1500
900
600
300
0 0.01
0.1
Figure 1. Quantitative dot-blotting showing the reactions of anti-human IgG and anti-horse IgG antibodies with EDTA extracts of hominid fossil VM3691 ( ) and equid fossil VM3032 (). The absorbances obtained with the fossil extracts in the direct and cross reactions were extrapolated against human and horse IgG calibration curves to determine the amount of protein in the fossils. The dotted line (- - - -) represents the background. The amount of human IgG in the VM3691 detected in the direct reaction was larger than that of horse IgG detected in the cross-reaction. The opposite was observed in VM3032: the amount of direct horse IgG was clearly greater than those of cross-reacted human IgG.
used to obtain the calibration curves (Figure 1). Moreover, the human IgG quantified in the equid fossils and the horse IgG quantified in the hominid fossils reflected cross-reactivity of the antibodies to IgG of different species. We therefore used the term ‘‘ng-equivalents’’ to denote that the amounts reported are not absolute values (Bjorja et al., 1997).
170
J. M. Torres et al. 5 PBS 4
3
2
1
0
AT83.1S AT83.1I AT85 AT83.2S AT83.2I Control
5 EDTA
Figure 2. Dot-blotting showing the duplicate reactions of anti-horse IgG antibodies with PBS, EDTA and acetic acid extracts (arrows) of equid fossils from Atapuerca. The name of each fossil is at the top of its corresponding reactions. Reactions of anti-horse IgG antibodies with decreasing amounts of native modern horse IgG are at the top of the membrane.
IgG (ng-equivalents)
4
3
2
1
Results Detection of IgG in fossils from Atapuerca assigned to equids Fossils from Atapuerca assigned to equids were studied by DB and QDB. We obtained parallel results with these two methods. PBS extracts were negative in all cases. The anti-horse IgG antibodies gave positive reactions with the EDTA extract of AT83.1S and AT83.1I and much weaker or no reactions with the EDTA extracts of the other fossils tested. All the Atapuerca fossils were positive in the acetic acid extracts. The reactivity of the acetic acid extracts was higher than that of the EDTA extracts (Figures 2 & 3). With QDB, fossils were tested with both anti-horse IgG and antihuman IgG antibodies. The amount of horse IgG, detected by direct reaction, was larger than the amount of human IgG detected by cross-reaction (Figure 3). Cattaneo et al. (1995) have stressed the importance of bone integrity for the survival of proteins in ancient bone. These authors believed that direct exposure of the bone to the environment implies worse preservation of the proteins. We tested samples from two equid fossil bones from Atapuerca (AT83.1 and AT83.2), from the surface (more exposed, AT83.1S and AT83.2S) and from the interior (more protected AT83.1I and AT83.2I). All four samples were positive for horse IgG in QDB, however, no significant differences were found between the total amounts of IgG detected in these fossils (Figure 3); therefore no relation between the amount of IgG detected and the
0
AT83.1S AT83.1I AT85 AT83.2S AT83.2I Control
5 ACETIC ACID 4
3
2
1
0
AT83.1S AT83.1I AT85 AT83.2S AT83.2I Control
Figure 3. Quantities of human IgG (black bars) and horse IgG (white bars) detected in PBS, EDTA and acetic acid extracts of equid fossils from Atapuerca by quantitative dot-blotting. The amounts of horse IgG were larger than those of human IgG. The extraction solutions were used as negative controls.
place (surface or interior) where the sample was taken could be established in these fossils. Detection of IgG in fossils from Venta Micena assigned to equids Fossils from Venta Micena assigned to equids were studied by DB and QDB. In both methods the
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena 171 3.50 PBS 3.00 2.50 2.00 1.50 1.00 0.50 0.00
VM3032 VM3064 VM3037 VM3008
Control
3.50 EDTA
Figure 4. Dot-blotting showing the duplicate reactions of anti-horse IgG antibodies with PBS, EDTA and acetic acid extracts (arrows) of equid fossils from Vente Micena. The name of each fossil is at the top of its corresponding reactions. Reactions of the anti-horse IgG antibodies with decreasing amounts of native modern horse IgG are at the top of the membrane.
anti-horse IgG antibodies gave positive reactions with the EDTA extract from all the fossils tested. PBS extracts were negative in all cases and we found only weak positivity in the acetic acid extract of VM3032 (by DB and QDB) and of VM3037 (by DB) (Figures 4 & 5). With QDB, fossils were tested with both antihorse IgG and anti-human IgG antibodies. These latter antibodies cross-reacted only weakly with the EDTA extract of VM3037 and were negative with the other equid fossils. The amount of horse IgG, detected by direct reaction, was larger than the amount of human IgG detected by to cross-reaction (Figure 5).
IgG (ng-equivalents)
3.00 2.50 2.00 1.50 1.00 0.50 0.00
VM3032 VM3064 VM3037 VM3008
Control
3.50 ACETIC ACID 3.00 2.50 2.00 1.50
Detection of IgG in fossils from Venta Micena and Cueva Victoria assigned to hominids Fossils from Venta Micena and Cueva Victoria assigned to hominids were studied by DB and QDB. In both methods the anti-human IgG antibodies gave positive reactions with the EDTA extract from fossils VM3691 and VM1960. However, PBS extracts and acetic acid extracts of these fossils were negative. All the extracts from fossils CV1 and CV2 were negative for the presence of IgG (Figures 6 & 7). With QDB, fossils were also tested with anti-horse IgG antibodies. These antibodies also reacted, although weakly, with the EDTA extracts from VM3691 and VM1960, and like the anti-human IgG antibodies, were negative with the other extracts. Nevertheless the amounts of horse IgG detected in these two fossils were lower than the amounts of human IgG. Fossils CV1 and CV2 were negative with the anti-horse IgG antibodies (Figure 7).
1.00 0.50 0.00
VM3032 VM3064 VM3037 VM3008
Control
Figure 5. Quantities of human IgG (black bars) and horse IgG (white bars) detected in PBS, EDTA and acetic acid extracts of equid fossils from Venta Micena by quantitative dot-blotting. The amounts of horse IgG were larger than those of human IgG. The extraction solutions were used as negative controls.
Comparisons of the amounts of IgG detected in different specimens The total amounts of IgG (EDTA+acetic acid extracts) of the different fossils of this study were plotted against their age and compared with a more recent bone: a 100-year-old vertebra from a donkey.
172
J. M. Torres et al. 1.50 PBS 1.20
0.90
0.60
0.30
0.00
VM3691 VM1960
CV1
CV2
Control
1.50 EDTA
Figure 6. Dot-blotting showing the duplicate reactions of antihuman IgG antibodies with PBS, EDTA and acetic acid extracts (arrows) of hominid fossils from Venta Micena and Cueva Victoria. The name of each fossil is at the top of its corresponding reactions. Reactions of the anti-human IgG antibodies with decreasing amounts of native modern human IgG are at the top of the membrane.
IgG (ng-equivalents)
1.20
0.90
0.60
0.30
0.00
VM3691 VM1960
CV1
CV2
Control
1.50
On average, we observed a decrease in the amounts of IgG as the age of the fossils increased, nevertheless, we observed fossils from Venta Micena that contained similar amounts of IgG as the fossils from Atapuerca (e.g., VM3032 and VM3639) (Figure 8).
Discussion We demonstrated that IgG survived in 1·6-Myr-old equid and hominid fossils from Venta Micena and 0·12-Myr-old equid fossils from Atapuerca. We also show that the immunoreactivity of this protein was consistent with the species assignation of the fossils. The efficiency of DB in detecting proteins from fossils was first reported by Rowley et al. (1986), who identified collagen in a 10-Myr-old fossil. The small amount of sample needed for DB and QDB (5 l) also makes these methods appropriate for studies of proteins from fossils, especially when the amount of sample available is very small. The methodological modification introduced by us in the QDB technique to obtain fully quantitative results overcomes the main problem with DB, in which results can only be expressed semiquantitatively. The detection in the equid fossils of IgG closer immunologically to modern horse IgG than to human
ACETIC ACID 1.20
0.90
0.60
0.30
0.00
VM3691 VM1960
CV1
CV2
Control
Figure 7. Quantities of human IgG (black bars) and horse IgG (white bars) detected in PBS, EDTA and acetic acid extracts of hominid fossils from Venta Micena and Cueva Victoria by quantitative dot-blotting. The amounts of human IgG were larger than those of horse IgG. The extraction solutions were used as negative controls.
IgG makes it unlikely that the presence of this protein in the fossils was due to exogenous contamination. However, the fossils assigned to hominids may have been contaminated with human material during handling. This is none the less improbable in the light of the negative results obtained with the PBS extracts
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena 173 600
Donkey
IgG (ng-equivalents) / bone (mg)
500
400
Donkey
300
AT83.1I
200
AT83.1S 100
0 1 10
VM3032
AT83.2S AT85 AT83.2I CV1 CV2 2
10
3
10
4
10 Years
5
10
6
10
VM3639 VM1960 VM3037 VM3008 7
10
Figure 8. Total amounts of IgG (sum of the amounts found in the EDTA plus the acetic acid extracts) detected in a 100-year-old vertebra from a donkey found in Orce, and from equid and hominid fossils from Atapuerca, Cueva Victoria and Venta Micena. Open circles ( ) represent the amounts of IgG found in each fossil, and closed circles ( ) represent the average and standard deviation of the amounts detected in different fossils from each site.
of the fossils. Of the three consecutive extracts obtained from each equid or hominid fossil, IgG was detected in the EDTA and/or acetic acid extracts, whereas no IgG was found in the PBS extracts (Figures 2–7). These results indicate that the mineral structure needed to be dissolved for the fossil proteins to be released, i.e., that IgG was indigenous to the fossil bone, and not merely adsorbed to the bone as a result of exogenous contamination. The largest amounts of IgG were found in the EDTA extracts of both equid and hominid fossils from Venta Micena (Figures 4–7), whereas in the equid fossils from Atapuerca the largest amounts of IgG were detected in acetic acid fossil extracts (Figures 2 & 3). This reflects differences in the solubility of the proteins in fossils from different sites. Protein solubility is probably influenced by the physicochemical events that affected the specimens during the process of fossilization. Another possibility is that younger fossils from Atapuerca contained a higher proportion of collagen, and IgG was somehow associated to it. Collagen (and associated IgG) were not soluble in EDTA, but swelled in the acetic acid solution. In certain fossils, non-collagen proteins appear to be selectively protected over collagen (Master, 1987; Tuross, 1989; Muyzer et al., 1992). It seems that the ability of these non-collagen proteins (even serum protein) to be complexed with the mineral phase of the bone, and even to be encapsulated by the hydroxyapatite crystals, may account for the preservation of these molecules (Smith et al., 1985; Berman et al., 1988). The fact that our fossils needed to be demineralized by EDTA solution in order to detect
IgG, shows that this protein was indeed bound to the fossil mineral phase. We observed that the amount of IgG decreased with increasing age (Figure 8). However, the age of the fossil cannot be used as an absolute criterion, as in some instances older fossils were found to contain larger amounts of biomolecules than younger ones (Ulrich et al., 1987; Poinar et al., 1996). For example, we did not detect IgG in fossils CV1 and CV2, which are more recent than the fossils from Venta Micena. On the other hand, the amount of IgG in equid fossil VM3032 is closer to that of the fossils from Atapuerca (Figure 8). Environmental factors (humidity, pH, oxidizing conditions, bacterial attack) appear to be more critical than age for the preservation of biomolecules (Eglinton & Logan, 1991; Muyzer et al., 1992), although in each case, preservation must be assumed to be the result of a combination of qualitative (environmental) factors and extensive factors (age). Cattaneo et al. (1993) buried human and animal bones in a garden of Sheffield, U.K., for months and then tested them for the presence of proteins with an ELISA. These authors did not detect IgG in bones buried for longer than two months, leading Palmqvist (1997) to state that this protein cannot be detected in bones older than this age. In contrast, Tuross & Stathoplos (1993) showed that IgG from fossil bones can be separated by gel electrophoresis and identified apparently intact with Western blotting. In this connection, Poinar (1993) stated that in the study of ancient biomolecules, conclusions obtained from nonfossilized material should not be strictly applied to
174
J. M. Torres et al.
fossil specimens. A fossil bone is a unique specimen that becomes a fossil only under very particular conditions. These conditions are responsible not only for the existence of the fossil, but very probably also influence the preservation of biomolecules (Eglinton & Logan, 1991). Fossil bones thousands or millions of years old are by no means comparable with recent bones buried under garden conditions for two months. After the animal dies, proteins tend to disappear within a short period under normal (garden/cemetery) conditions. However, under the singular conditions that lead to fossilization, as at Venta Micena or Atapuerca, bone proteins are probably ‘‘frozen’’ into the mineral phase of the bone so that they are preserved for thousands or millions of years. Immunoglobulin G from equid fossils from Atapuerca and Venta Micena reacted as expected: we detected larger amounts of horse IgG than of crossreactive human IgG (Figures 3 & 5). Immunoglobulin G from fossils VM1960 and VM3691 reacted in accordance with their hominid assignation, whereas fossils CV1 and CV2 from Cueva Victoria were negative for IgG (Figure 7). Independent studies of fossils VM1960, CV1 and CV2 at the University of Granada and the University of California at San Francisco for the presence of species-specific albumin were concordant with our IgG results: VM1960 contained albumin immunologically close to modern human albumin, and CV1 and CV2 were negative for the presence of albumin (Borja et al., 1997). Our results demonstrate that IgG can under some circumstances be preserved for as long as 1·6 Myr in fossil bones.
Acknowledgements This work was supported by a grant from DGICYT and by La Diputacio´ de Barcelona. The authors thank J. Gibert from Institut Paleontolo`gic ‘‘Dr M. Crusafont’’ de Sabadell (Spain), for allowing us to take samples from Venta Micena and Cueva Victoria fossils and for palaeontological information about them; E. Aguirre and B. Sa´nchez from Museo Nacional de Ciencias Naturales, Madrid for allowing us to take samples from Atapuerca fossils, and A. Arribas from Museo Nacional de Ciencias Naturales, Madrid for palaeontological information about them. We also thank K. Shashok for improving the English in the manuscript.
References Austin, J. J., Ross, A. J., Smith, A. B., Fortey, R. A. & Thomas, R. H. (1997). Problems of reprudicibility. Does geologically ancient DNA survive in amber-preserved insects? Proceedings of the Royal Society of London. Series B: Biological Sciences 264, 467–474. Bada, J. L., Wang, X. S. & Hamilton, H. (1999). Preservation of key biomolecules in the fossil record: current knowledge and future challenges. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 354, 77–87.
Berman, A., Addadi, L. & Weiner, S. (1988). Interactions of urchin macromolecules with growing calcite crystals—a study of intracrystalline proteins. Nature 331, 546–548. Borja, C., Garcı´a-Pacheco, M., Olivares, E. G., Scheuenstuhl, G. & Lowenstein, J. M. (1997). Immunospecificity of albumin detected in 1·6 million-year-old fossils from Venta Micena in Orce, Granada, Spain. American Journal of Physical Anthropology 103, 433–441. Cano, R. J., Poinar, H. N., Pieniazek, N. J., Acra, A. & Poinar, G. O. (1993). Amplification and sequenzing of DNA from a 120–135 million-year-old weevil. Nature 363, 536–538. Cattaneo, C., Gelsthorpe, K., Phillips, P. & Sokol, R. J. (1993). Blood residues on stone tools: indoor and outdoor experiments. World Archaeology 25, 29–43. Cattaneo, C., Gelsthorpe, K., Phillips, P. & Sokol, R. J. (1995). Differential survival of albumin in ancient bone. Journal of Archaeological Sciences 22, 271–276. De Jong, E. W., Westbroek, P., Westbroek, J. F. & Bruning, J. W. (1974). Preservation of antigenic properties of macromolecules over 70 Ma. Nature 252, 63–64. Eglinton, G. & Logan, G. A. (1991). Molecular preservation. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 333, 315–328. Franc, S., Marzin, E., Boutillon, M. M., Lafont, R., Lecheˆne de la Porte, P. & Herbage, D. (1995). Immunochemical and biochemical analyses of 20000–25000-year-old fossil cartilage. European Journal of Biochemistry 234, 125–131. Gilbert, J., Campillo, D., Arque´s, J. M., Olivares, E. G., Borja, C. & Lowenstein, J. (1998). Hominid status of the Orce cranial fragment reasserted. Journal of Human Evolution 34, 203–217. Ho¨ss, M., Dilling, A., Currant, A. & Pa¨a¨bo, S. (1996). Molecular phylogeny of the extinct ground sloth Mylodon darwinii. Proceedings of the National Academy of Sciences of the United States of America 93, 181–185. Krings, M., Geisert, H., Schmitz, R. W., Krainitzki, H. & Pa¨a¨bo, S. (1999). DNA sequence of the mitochondrial hypervariable region II from the Neandertal type specimen. Proceedings of the National Academy of Sciences of the United States of America 96, 5581–5585. Krings, M., Stone, A., Schmitz, R. W., Krainitzki, H., Stoneking, M. & Pa¨a¨bo, S. (1997). Neanderthal DNA-sequences and the origin of modern humans. Cell 90, 19–30. Lindahl, T. (1997). Facts and artifacts of ancient DNA. Cell 90, 1–3. Lowenstein, J. M. (1981). Immunological reactions from fossil material. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 292, 143–149. Lowenstein, J. M. (1983). Fossil proteins and evolutionary time. Pontificiae Academia Scientiarum Scripta Varia 50, 151–162. Lowenstein, J. M. & Ryder, O. A. (1985). Immunological systematics of the extinct quagga (Equidae). Experientia 41, 1192–1193. Lowenstein, J. M., Sarich, V. M. & Richardson, B. J. (1981). Albumin systematics of the extinct mammoth and Tasmanian wolf. Nature 291, 409–411. Lowenstein, J. M. & Scheuenstuhl, G. (1991). Immunological methods in molecular palaentology. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 333, 375–380. Masters, P. (1987). Preferential preservation of noncollagenous proteins during bone diagenesis: Implications for chronometric and stable isotope measurements. Geochimica et Cosmochimica Acta 51, 3209–3214. Muyzer, G., Sandberg, P., Knapen, M. H. J., Vermeer, C., Collins, M. & Westbroek, P. (1992). Preservation of the bone protein osteocalcin in dinosaurs. Geology 20, 871–874. Newman, M. E., Ceri, H. & Kooyman, B. (1996). The use of immunological techniques in the analysis of archaeological materials—a response to Eisele; with report of studies at Head-Smashed-in Buffalo Jump. Antiquity 70, 677–682. Newman, M. E., Yohe, R. M. II, Ceri, H. & Sutton, M. Q. (1993). Immunological protein residue analysis of nonlithic archaeological material. Journal of Archaeological Science 20, 119–132.
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena 175 Pa¨a¨bo, S., Higuchi, R. G. & Wilson, A. C. (1989). Ancient DNA and the polymerase chain reaction. The emerging field of molecular archaeology. Journal of Biological Chemistry 264, 9709– 9712. Palmqvist, P. (1997). A critical reevaluation of the evidence for the presence of hominids in lower pleistocene times at Venta-Micena, southern Spain. Journal of Human Evolution 33, 83–89. Pawlicki, R. & Nowogrodzka-Zago´rska, M. (1998). Blood vessels and red blood cells preserved in dinosaur bones. Annals of Anatomy 180, 73–77. Poinar, G. O. Jr (1993). Reply to Tomas Lindahl: Recovery of antediluvian DNA. Nature 365, 700. Poinar, H. N., Cano, R. J. & Poinar, G. O. (1993). DNA from an extinct plant. Nature 363, 677. Poinar, H. N., Ho¨ss, M., Bada, J. L. & Pa¨a¨bo, S. (1996). Amino acid racemization and the preservation of ancient DNA. Science 272, 864–866. Rainey, W. E., Lowenstein, J. M., Sarich, V. M. & Magor, D. M. (1984). Sirenian molecular systematics including the extinct Steller’s sea cow (Hydrodamalis gigas). Naturwissenschaften 67, 343–346. Rowley, M. J., Rich, P. V., Rich, T. H. & Mackay, I. R. (1986). Immunoreactive collagen in avian and mammalian fossils. Naturwissenschaften 73, 620–623. Schweitzer, M. H., Marshall, M., Carron, K., Scott-Bohle, D., Busse, S. C., Arnold, E. V., Barnard, D., Horner, J. R. & Starkey, J. R. (1997a). Heme compounds in dinosaur trabecular bone. Proceedings of the National Academy of Sciences of the United States of America 94, 6291–6296. Schweitzer, M. H., Johnson, C., Zocco, T. G., Horner, J. R. & Starkey, J. R. (1997b). Preservation of biomolecules in cancellous
bone of Tyrannosaurus rex. Journal of Vertebrate Paleontology 17, 349–359. Shoshani, J., Lowenstein, J. M., Walz, D. A. & Goodman, M. (1985). Proboscidean origins of mastodon and mammoth demonstrated immunologically. Paleobiology 11, 429–437. Smith, A. J., Matthews, J. B., Wilson, C. & Sewell, H. F. (1985). Plasma proteins in human cortical bone: in vitro binding studies. Calcified Tissue International 37, 208–210. Smith, A. B. & Littlewood, D. T. J. (1994). Paleontological data and molecular phylogenetic analysis. Paleobiology 20, 259–273. Tobias, P. V. (1998). Some cumments on the case for early Pleistocene hominids in South-Eastern Spain. Human Evolution 13, 91–96. Tuross, N. (1989). Albumin preservation in the Taima-taima mastodon skeleton. Applied Geochemistry 4, 255–259. Tuross, N., Behrensmeyer, A. K., Eanes, E. D., Fisher, L. W. & Hare, P. E. (1989). Molecular preservation and crystallographic alterations in a weathering sequence of wildebeest bones. Applied Geochemistry 4, 261–270. Tuross, N. & Stathoplos, L. (1993). Ancient proteins in fossil bones. Methods in Enzymology 224, 121–129. Ulrich, M. M., Perizonius, W. R., Spoor, C. F., Sandberg, P. & Vermeer, C. (1987). Extraction of osteocalcin from fossil bones and teeth. Biochemical and Biophysical Research Communications 149, 712–719. Wilson, A. C., Carlson, S. S. & White, T. J. (1977). Biochemical evolution. Annual Review of Biochemistry 46, 573–639. Wyckoff, R. W. G. (1972). The Biochemistry of Animal Fossils. Bristol: Scientechnica. Zihlman, A. L. & Lowenstein, J. M. (1996). A spanish Olduvai? Current Anthropology 37, 695–697.
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena (Granada, Spain) Jesu´s M. Torres, Concepcio´n Borja and Enrique G. Olivares* Unidad de Inmunologı´a, Instituto de Biotecnologı´a, Universidad de Granada, Spain (Received 3 April 2000, revised manuscript accepted 2 March 2001) We studied the presence of immunoglobulin G (IgG) in equid fossil bones from Venta Micena (Granada, Spain), dated at 1·6 Myr old, and from Atapuerca (Burgos, Spain) (0·12 Myr), and in ostensibly hominid fossils VM1960 and VM3691 (1·6 Myr) from Venta Micena and CV1 and CV2 (1·4 Myr) from Cueva Victoria (Murcia, Spain). Samples taken from the specimens were crushed to a fine powder and their mineral structure was dissolved in an ethylenediaminetetracetic acid (EDTA) solution. Fossil extracts were tested with antibodies against human IgG and against horse IgG with two independent immunological methods: dot-blotting (DB) and a modification of this latter method: quantitative dot-blotting (QDB). IgG was detected by DB and was quantifiable by QDB in some of the fossils tested. Equid fossils from Atapuerca and Venta Micena gave stronger reactions with the antibodies against horse IgG than with the antibodies against human IgG. Fossils VM3691 and VM1960 reacted more strongly with the antibodies against human IgG than with antibodies against horse IgG, whereas no IgG was detected in fossils CV1 and CV2. These findings show that species-specific IgG can be detected in fossils as old as 1·6 Myr. The immunological analysis of fossil proteins may help to solve palaeontological controversies. 2002 Elsevier Science Ltd. Keywords: IGG, HOMINID, EQUID, VENTA MICENA, ATAPUERCA, FOSSIL BONE, DOT-BLOTTING, QUANTITATIVE DOT-BLOTTING.
With respect to proteins, Bada et al. (1999), based on the analysis of Pleistocene mammalian bone from Olduvai Gorge, came to conclusion that ‘‘original protein components should be hydrolysed to free amino acids which then diffuse out of the fossil matrix and are lost over time-scales of 105–106 years in most environments’’. Nevertheless, this inference is contradicted by the many results that show that proteins, or at least pieces of them, can be identified by immunological methods in fossil bones millions of years old. Lowenstein, using a radioimmunoassay (RIA), detected species-specific collagen and serum factors in fossils as old as a 0·5-Myr-old Homo erectus, a 1·9Myr-old Australopithecus robustus (Lowenstein, 1981) and an 8-Myr-old Ramapithecus (Lowenstein, 1983). Furthermore, collagen has been found by dot-blotting in a 10-Myr-old bone (Rowley et al., 1986); and osteocalcin has been extracted from 13-Myr-old fossil bovid bones, from 30-Myr-old fossil rodent teeth (Ullrich et al., 1987), and from 75-Myr-old dinosaur bones (Muyzer et al., 1992). Osteocalcin from the fossil bovid bones still retained functionally active gammacarboxyglutamic acid residues after 13 Myr (Ulrich et al., 1987). Seventy-Myr-old molluscs also yielded positive immunological reactions (De Jong et al., 1974), and collagen fibrils were observed with electron microscopy in a 200-Myr-old dinosaur bone (Wykoff, 1972). A striking recent finding was the antisera
Introduction umerous reports have demonstrated that species-specific biomolecules can persist in archaeological material (Newman et al., 1996) and in some fossils (Lowenstein & Scheuenstuhl, 1991). These molecules are often scarce and difficult to identify with certainty; however, when they persist, analyses of proteins and DNA may be a reliable way to study archaeological sites (Newman et al., 1993) or to classify fossils, particularly when anatomical interpretations are discrepant (Smith & Littlewood, 1994). DNA has been detected in samples thousands of years old (Pa¨a¨bo et al., 1989; Hoss et al., 1996), and mitochondrial DNA of a Neanderthal bone (40,000 years old) has been sequenced (Krings et al., 1997, 1999). However, the rapid degradation of this biomolecule in bone has led some authors to suggest that the detectability of DNA in fossil bones is limited to specimens no older than a 100,000 years (Poinar et al., 1996; Lindhal, 1997). Nevertheless, several studies demonstrated the presence of species-specific DNA millions of years old in fossil plants (Poinar et al., 1993) and in insects preserved in amber (Cano et al., 1993). A more recent report suggested, however, that the presence of ancient DNA in amber fossils is doubtful (Austin et al., 1997).
N
*Tel. and fax: 34-958-243522; E-mail: [email protected]
167 0305–4403/02/020167+09 $35.00/0
2002 Elsevier Science Ltd.
168
J. M. Torres et al. Table 1. List of fossils studied in this work
Fossil
Assigned to:
Anatomical location
Age (Myr)
Found in:
VM1960 VM3691 CV1 CV2 VM3032 VM3064 VM3037 VM3008 VM3016 AT83 GSU-9. E.21–4 (AT83·1)* AT83 GSU-9. F.16–33 (AT83.2)* AT85
Hominid Hominid Hominid Hominid Equid Equid Equid Equid Equid Equid Equid Equid
Humerus fragment Humerus fragment Humerus fragment Humerus fragment Tibia fragment Tibia fragment Tibia fragment Metatarsal bone Metatarsal bone Right femur Mandible Sacrum
1·6 1·6 1·4 1·4 1·6 1·6 1·6 1·6 1·6 0·12 0·12 0·12
Venta Micena, Orce Venta Micena, Orce Cueva Victoria, Murcia Cueva Victoria, Murcia Venta Micena, Orce Venta Micena, Orce Venta Micena, Orce Venta Micena, Orce Venta Micena, Orce Atapuerca, Burgos Atapuerca, Burgos Atapuerca, Burgos
*In parentheses is the name given to the fossil in this article. Two independent samples were taken from the surface and interior of fossils AT83.1 and AT83.2. These samples were named AT83.1S and AT83.2S (superficial samples) and AT83.1I and AT83.2I (interior samples).
obtained in rats injected with dinosaurian tissues which gave positive reactions with purified avian and mammalian haemoglobins; this result showed that fragments of haemoglobin were preserved within the dinosaur tissues (Schweitzer et al., 1997a). In contradiction to the claim of Bada et al. (1999), all these results clearly show that proteins from some fossil bones, although fragmentary, can persist and retain enough of their immunological properties to provide useable genetic information for periods of millions of years. Preservation seems to be favoured by the fact that proteins are embedded in the mineral phase of the bone, as this provides considerable protection from degradation by environmental conditions (Smith et al., 1985; Berman et al., 1988). The mineral phase also appears to preserve, by diffusion, even the cartilage of some fossil bones (Franc et al., 1995). The observation of cells and histological structures preserved in dinosaur bones is also consistent with the presence of biomolecules (Schweitzer et al., 1997b; Pawlicki & Nowogrodzka-Zago´rska, 1998). Compared with collagen and osteocalcin—the most abundant bone proteins—albumin or immunoglobulin G (IgG), although present in lower concentrations in bone, provide much more evolutionary information given their faster rate of evolution (Wilson et al., 1977). Albumin has helped to resolve the disputed phylogenic affinities of extinct mammoths, mastodons, Steller’s sea cow, the Tasmanian wolf and the quagga (Lowenstein, 1981; Lowenstein et al., 1981; Lowenstein & Ryder, 1985; Rainey et al., 1984; Shoshani et al., 1985; Lowenstein & Scheuensthul, 1991). In some instances, plasma proteins appear to be better preserved than collagen in bone, so that the bone becomes enriched in plasma proteins above in vivo levels (Tuross, 1989; Tuross et al., 1989). In a previous study our group at the University of Granada and Lowenstein’s group at the University of California in San Francisco independently showed that species-specific albumin can
be detected by enzyme-linked-immunosorbant assay (ELISA) or RIA in fossils of different species collected at Venta Micena (Granada, Spain) (Borja et al., 1997). These independent studies supported the human assignation of controversial fossils such as skull fragment VM0 and humerus fragment VM1960, both dated at about 1·6 Myr (Zihlman & Lowenstein, 1996; Palmqvist, 1997; Gibert et al., 1998; Tobias, 1998). In the present study we show that immunoglobulin G (IgG) could also be detected by dot-blotting (DB) and a modification of this method called quantitative dotblotting (QDB) in 0·12-Myr-old equid fossil bones from Atapuerca and 1·6-Myr-old equid and hominid fossil bones from Venta Micena, Orce, Granada.
Materials and Methods Fossil and bone extracts The fossils studied are listed in Table 1. Aseptic conditions were rigorously maintained and disposable materials were used to avoid external contamination during the process of extraction and testing. Unless otherwise indicated, samples were taken from the interior of the fossils. The samples were ground to a fine powder and independently maintained in phosphate-buffered saline (PBS) for one week in continuous agitation at room temperature (washing). The extracts were centrifuged, and the supernatants were collected, divided and stored at 70C until use. The remaining bone sediment of each sample was then decalcified in a solution of 0·2 M EDTA, pH 7·4, following the same procedure (agitation, centrifugation) as before. Finally, the decalcified sediment was also extracted with 0·5 M acetic acid with the same procedure. Supernatants were examined under light microscope and those that showed bacterial growth were discarded. From each fossil, we therefore obtained PBS, EDTA and acetic acid aliquots. Equid
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena 169
fossils were extracted starting at 5 mg bone per ml of PBS, whereas hominid fossils were extracted at 3·5 mg/ ml. A 100-year-old vertebra of a donkey found in Orce (near Venta Micena) was extracted at 15 mg/ml and at 3 mg/ml in two independent processes.
Detection of IgG by quantitative dot-blotting Because DB gives only semiquantitative results (intensity of staining) we modified this method to obtain fully quantitative results. All samples (run in duplicate) were placed inside circles that were previously scored on the nitrocellulose membrane with a needle. We followed the same steps as in DB until the anti-IgG antibodies were added. We then cut out the circles of nitrocellulose within which the reaction took place and transferred each circle to a well of a microtitre plate. A solution of 0·4 mg/ml of orthophenylenediamine dihydrochloride (Sigma) in phosphate citrate buffer was then added and absorbance was read at 490 nm with a microplate autoreader. Instead of the fossil extracts, solutions used for fossil extractions were used as negative controls. The absorbances obtained with the fossil extracts in the immunological direct and cross reactions were extrapolated against human and horse IgG calibration curves to determine the amount of protein in the fossils. When the fossil corresponded to or was close to the species to which the antibodies were directed, the amount of IgG detected was larger than when antibodies to a non-related species were used (Figure 1). Although useful for comparisons between species, the amounts of protein detected in a fossil cannot be considered at face value, as these fossil proteins are probably split and denatured, in contrast with the intact native modern human and horse IgG that were
Absorbance (mU)
1200
900
600
300
0 0.01
0.1
10
100
1 10 IgG (ng-equivalents)
100
1 Horse IgG
1500
1200 Absorbance (mU)
Detection of IgG by dot-blotting Five microlitres of each fossil extract or of a solution of modern human or horse IgG was placed on a nitrocellulose membrane (Biorad, Richmond, CA, U.S.A.) and incubated for 30 min at room temperature. All samples were tested in duplicate. After washing with a solution of 0·05% PBS–Tween-20, the membrane was blocked with a solution of 1% gelatin in PBS for 1 h at 37C and overnight at 4C. The appropriate anti-IgG antibodies (affinity isolated goat anti-human IgG or rabbit anti-horse IgG polyclonal antibodies labelled with peroxidase, Sigma, St Louis, MO, U.S.A.) were then added and the membrane incubated for 1 h at room temperature. After incubation the membrane was washed again with PBS–Tween and the antigen– antibody reactions were revealed with a solution of 0·05% 3,3 -diaminobenzidine (Sigma) and 0·003% H2O2 in PBS for 5–15 min incubation time. The enzymatic reaction was stopped by washing with distilled water. Solutions used for fossil extractions (PBS, EDTA and acetic acid) were used as negative controls.
Human IgG 1500
900
600
300
0 0.01
0.1
Figure 1. Quantitative dot-blotting showing the reactions of anti-human IgG and anti-horse IgG antibodies with EDTA extracts of hominid fossil VM3691 ( ) and equid fossil VM3032 (). The absorbances obtained with the fossil extracts in the direct and cross reactions were extrapolated against human and horse IgG calibration curves to determine the amount of protein in the fossils. The dotted line (- - - -) represents the background. The amount of human IgG in the VM3691 detected in the direct reaction was larger than that of horse IgG detected in the cross-reaction. The opposite was observed in VM3032: the amount of direct horse IgG was clearly greater than those of cross-reacted human IgG.
used to obtain the calibration curves (Figure 1). Moreover, the human IgG quantified in the equid fossils and the horse IgG quantified in the hominid fossils reflected cross-reactivity of the antibodies to IgG of different species. We therefore used the term ‘‘ng-equivalents’’ to denote that the amounts reported are not absolute values (Bjorja et al., 1997).
170
J. M. Torres et al. 5 PBS 4
3
2
1
0
AT83.1S AT83.1I AT85 AT83.2S AT83.2I Control
5 EDTA
Figure 2. Dot-blotting showing the duplicate reactions of anti-horse IgG antibodies with PBS, EDTA and acetic acid extracts (arrows) of equid fossils from Atapuerca. The name of each fossil is at the top of its corresponding reactions. Reactions of anti-horse IgG antibodies with decreasing amounts of native modern horse IgG are at the top of the membrane.
IgG (ng-equivalents)
4
3
2
1
Results Detection of IgG in fossils from Atapuerca assigned to equids Fossils from Atapuerca assigned to equids were studied by DB and QDB. We obtained parallel results with these two methods. PBS extracts were negative in all cases. The anti-horse IgG antibodies gave positive reactions with the EDTA extract of AT83.1S and AT83.1I and much weaker or no reactions with the EDTA extracts of the other fossils tested. All the Atapuerca fossils were positive in the acetic acid extracts. The reactivity of the acetic acid extracts was higher than that of the EDTA extracts (Figures 2 & 3). With QDB, fossils were tested with both anti-horse IgG and antihuman IgG antibodies. The amount of horse IgG, detected by direct reaction, was larger than the amount of human IgG detected by cross-reaction (Figure 3). Cattaneo et al. (1995) have stressed the importance of bone integrity for the survival of proteins in ancient bone. These authors believed that direct exposure of the bone to the environment implies worse preservation of the proteins. We tested samples from two equid fossil bones from Atapuerca (AT83.1 and AT83.2), from the surface (more exposed, AT83.1S and AT83.2S) and from the interior (more protected AT83.1I and AT83.2I). All four samples were positive for horse IgG in QDB, however, no significant differences were found between the total amounts of IgG detected in these fossils (Figure 3); therefore no relation between the amount of IgG detected and the
0
AT83.1S AT83.1I AT85 AT83.2S AT83.2I Control
5 ACETIC ACID 4
3
2
1
0
AT83.1S AT83.1I AT85 AT83.2S AT83.2I Control
Figure 3. Quantities of human IgG (black bars) and horse IgG (white bars) detected in PBS, EDTA and acetic acid extracts of equid fossils from Atapuerca by quantitative dot-blotting. The amounts of horse IgG were larger than those of human IgG. The extraction solutions were used as negative controls.
place (surface or interior) where the sample was taken could be established in these fossils. Detection of IgG in fossils from Venta Micena assigned to equids Fossils from Venta Micena assigned to equids were studied by DB and QDB. In both methods the
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena 171 3.50 PBS 3.00 2.50 2.00 1.50 1.00 0.50 0.00
VM3032 VM3064 VM3037 VM3008
Control
3.50 EDTA
Figure 4. Dot-blotting showing the duplicate reactions of anti-horse IgG antibodies with PBS, EDTA and acetic acid extracts (arrows) of equid fossils from Vente Micena. The name of each fossil is at the top of its corresponding reactions. Reactions of the anti-horse IgG antibodies with decreasing amounts of native modern horse IgG are at the top of the membrane.
anti-horse IgG antibodies gave positive reactions with the EDTA extract from all the fossils tested. PBS extracts were negative in all cases and we found only weak positivity in the acetic acid extract of VM3032 (by DB and QDB) and of VM3037 (by DB) (Figures 4 & 5). With QDB, fossils were tested with both antihorse IgG and anti-human IgG antibodies. These latter antibodies cross-reacted only weakly with the EDTA extract of VM3037 and were negative with the other equid fossils. The amount of horse IgG, detected by direct reaction, was larger than the amount of human IgG detected by to cross-reaction (Figure 5).
IgG (ng-equivalents)
3.00 2.50 2.00 1.50 1.00 0.50 0.00
VM3032 VM3064 VM3037 VM3008
Control
3.50 ACETIC ACID 3.00 2.50 2.00 1.50
Detection of IgG in fossils from Venta Micena and Cueva Victoria assigned to hominids Fossils from Venta Micena and Cueva Victoria assigned to hominids were studied by DB and QDB. In both methods the anti-human IgG antibodies gave positive reactions with the EDTA extract from fossils VM3691 and VM1960. However, PBS extracts and acetic acid extracts of these fossils were negative. All the extracts from fossils CV1 and CV2 were negative for the presence of IgG (Figures 6 & 7). With QDB, fossils were also tested with anti-horse IgG antibodies. These antibodies also reacted, although weakly, with the EDTA extracts from VM3691 and VM1960, and like the anti-human IgG antibodies, were negative with the other extracts. Nevertheless the amounts of horse IgG detected in these two fossils were lower than the amounts of human IgG. Fossils CV1 and CV2 were negative with the anti-horse IgG antibodies (Figure 7).
1.00 0.50 0.00
VM3032 VM3064 VM3037 VM3008
Control
Figure 5. Quantities of human IgG (black bars) and horse IgG (white bars) detected in PBS, EDTA and acetic acid extracts of equid fossils from Venta Micena by quantitative dot-blotting. The amounts of horse IgG were larger than those of human IgG. The extraction solutions were used as negative controls.
Comparisons of the amounts of IgG detected in different specimens The total amounts of IgG (EDTA+acetic acid extracts) of the different fossils of this study were plotted against their age and compared with a more recent bone: a 100-year-old vertebra from a donkey.
172
J. M. Torres et al. 1.50 PBS 1.20
0.90
0.60
0.30
0.00
VM3691 VM1960
CV1
CV2
Control
1.50 EDTA
Figure 6. Dot-blotting showing the duplicate reactions of antihuman IgG antibodies with PBS, EDTA and acetic acid extracts (arrows) of hominid fossils from Venta Micena and Cueva Victoria. The name of each fossil is at the top of its corresponding reactions. Reactions of the anti-human IgG antibodies with decreasing amounts of native modern human IgG are at the top of the membrane.
IgG (ng-equivalents)
1.20
0.90
0.60
0.30
0.00
VM3691 VM1960
CV1
CV2
Control
1.50
On average, we observed a decrease in the amounts of IgG as the age of the fossils increased, nevertheless, we observed fossils from Venta Micena that contained similar amounts of IgG as the fossils from Atapuerca (e.g., VM3032 and VM3639) (Figure 8).
Discussion We demonstrated that IgG survived in 1·6-Myr-old equid and hominid fossils from Venta Micena and 0·12-Myr-old equid fossils from Atapuerca. We also show that the immunoreactivity of this protein was consistent with the species assignation of the fossils. The efficiency of DB in detecting proteins from fossils was first reported by Rowley et al. (1986), who identified collagen in a 10-Myr-old fossil. The small amount of sample needed for DB and QDB (5 l) also makes these methods appropriate for studies of proteins from fossils, especially when the amount of sample available is very small. The methodological modification introduced by us in the QDB technique to obtain fully quantitative results overcomes the main problem with DB, in which results can only be expressed semiquantitatively. The detection in the equid fossils of IgG closer immunologically to modern horse IgG than to human
ACETIC ACID 1.20
0.90
0.60
0.30
0.00
VM3691 VM1960
CV1
CV2
Control
Figure 7. Quantities of human IgG (black bars) and horse IgG (white bars) detected in PBS, EDTA and acetic acid extracts of hominid fossils from Venta Micena and Cueva Victoria by quantitative dot-blotting. The amounts of human IgG were larger than those of horse IgG. The extraction solutions were used as negative controls.
IgG makes it unlikely that the presence of this protein in the fossils was due to exogenous contamination. However, the fossils assigned to hominids may have been contaminated with human material during handling. This is none the less improbable in the light of the negative results obtained with the PBS extracts
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena 173 600
Donkey
IgG (ng-equivalents) / bone (mg)
500
400
Donkey
300
AT83.1I
200
AT83.1S 100
0 1 10
VM3032
AT83.2S AT85 AT83.2I CV1 CV2 2
10
3
10
4
10 Years
5
10
6
10
VM3639 VM1960 VM3037 VM3008 7
10
Figure 8. Total amounts of IgG (sum of the amounts found in the EDTA plus the acetic acid extracts) detected in a 100-year-old vertebra from a donkey found in Orce, and from equid and hominid fossils from Atapuerca, Cueva Victoria and Venta Micena. Open circles ( ) represent the amounts of IgG found in each fossil, and closed circles ( ) represent the average and standard deviation of the amounts detected in different fossils from each site.
of the fossils. Of the three consecutive extracts obtained from each equid or hominid fossil, IgG was detected in the EDTA and/or acetic acid extracts, whereas no IgG was found in the PBS extracts (Figures 2–7). These results indicate that the mineral structure needed to be dissolved for the fossil proteins to be released, i.e., that IgG was indigenous to the fossil bone, and not merely adsorbed to the bone as a result of exogenous contamination. The largest amounts of IgG were found in the EDTA extracts of both equid and hominid fossils from Venta Micena (Figures 4–7), whereas in the equid fossils from Atapuerca the largest amounts of IgG were detected in acetic acid fossil extracts (Figures 2 & 3). This reflects differences in the solubility of the proteins in fossils from different sites. Protein solubility is probably influenced by the physicochemical events that affected the specimens during the process of fossilization. Another possibility is that younger fossils from Atapuerca contained a higher proportion of collagen, and IgG was somehow associated to it. Collagen (and associated IgG) were not soluble in EDTA, but swelled in the acetic acid solution. In certain fossils, non-collagen proteins appear to be selectively protected over collagen (Master, 1987; Tuross, 1989; Muyzer et al., 1992). It seems that the ability of these non-collagen proteins (even serum protein) to be complexed with the mineral phase of the bone, and even to be encapsulated by the hydroxyapatite crystals, may account for the preservation of these molecules (Smith et al., 1985; Berman et al., 1988). The fact that our fossils needed to be demineralized by EDTA solution in order to detect
IgG, shows that this protein was indeed bound to the fossil mineral phase. We observed that the amount of IgG decreased with increasing age (Figure 8). However, the age of the fossil cannot be used as an absolute criterion, as in some instances older fossils were found to contain larger amounts of biomolecules than younger ones (Ulrich et al., 1987; Poinar et al., 1996). For example, we did not detect IgG in fossils CV1 and CV2, which are more recent than the fossils from Venta Micena. On the other hand, the amount of IgG in equid fossil VM3032 is closer to that of the fossils from Atapuerca (Figure 8). Environmental factors (humidity, pH, oxidizing conditions, bacterial attack) appear to be more critical than age for the preservation of biomolecules (Eglinton & Logan, 1991; Muyzer et al., 1992), although in each case, preservation must be assumed to be the result of a combination of qualitative (environmental) factors and extensive factors (age). Cattaneo et al. (1993) buried human and animal bones in a garden of Sheffield, U.K., for months and then tested them for the presence of proteins with an ELISA. These authors did not detect IgG in bones buried for longer than two months, leading Palmqvist (1997) to state that this protein cannot be detected in bones older than this age. In contrast, Tuross & Stathoplos (1993) showed that IgG from fossil bones can be separated by gel electrophoresis and identified apparently intact with Western blotting. In this connection, Poinar (1993) stated that in the study of ancient biomolecules, conclusions obtained from nonfossilized material should not be strictly applied to
174
J. M. Torres et al.
fossil specimens. A fossil bone is a unique specimen that becomes a fossil only under very particular conditions. These conditions are responsible not only for the existence of the fossil, but very probably also influence the preservation of biomolecules (Eglinton & Logan, 1991). Fossil bones thousands or millions of years old are by no means comparable with recent bones buried under garden conditions for two months. After the animal dies, proteins tend to disappear within a short period under normal (garden/cemetery) conditions. However, under the singular conditions that lead to fossilization, as at Venta Micena or Atapuerca, bone proteins are probably ‘‘frozen’’ into the mineral phase of the bone so that they are preserved for thousands or millions of years. Immunoglobulin G from equid fossils from Atapuerca and Venta Micena reacted as expected: we detected larger amounts of horse IgG than of crossreactive human IgG (Figures 3 & 5). Immunoglobulin G from fossils VM1960 and VM3691 reacted in accordance with their hominid assignation, whereas fossils CV1 and CV2 from Cueva Victoria were negative for IgG (Figure 7). Independent studies of fossils VM1960, CV1 and CV2 at the University of Granada and the University of California at San Francisco for the presence of species-specific albumin were concordant with our IgG results: VM1960 contained albumin immunologically close to modern human albumin, and CV1 and CV2 were negative for the presence of albumin (Borja et al., 1997). Our results demonstrate that IgG can under some circumstances be preserved for as long as 1·6 Myr in fossil bones.
Acknowledgements This work was supported by a grant from DGICYT and by La Diputacio´ de Barcelona. The authors thank J. Gibert from Institut Paleontolo`gic ‘‘Dr M. Crusafont’’ de Sabadell (Spain), for allowing us to take samples from Venta Micena and Cueva Victoria fossils and for palaeontological information about them; E. Aguirre and B. Sa´nchez from Museo Nacional de Ciencias Naturales, Madrid for allowing us to take samples from Atapuerca fossils, and A. Arribas from Museo Nacional de Ciencias Naturales, Madrid for palaeontological information about them. We also thank K. Shashok for improving the English in the manuscript.
References Austin, J. J., Ross, A. J., Smith, A. B., Fortey, R. A. & Thomas, R. H. (1997). Problems of reprudicibility. Does geologically ancient DNA survive in amber-preserved insects? Proceedings of the Royal Society of London. Series B: Biological Sciences 264, 467–474. Bada, J. L., Wang, X. S. & Hamilton, H. (1999). Preservation of key biomolecules in the fossil record: current knowledge and future challenges. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 354, 77–87.
Berman, A., Addadi, L. & Weiner, S. (1988). Interactions of urchin macromolecules with growing calcite crystals—a study of intracrystalline proteins. Nature 331, 546–548. Borja, C., Garcı´a-Pacheco, M., Olivares, E. G., Scheuenstuhl, G. & Lowenstein, J. M. (1997). Immunospecificity of albumin detected in 1·6 million-year-old fossils from Venta Micena in Orce, Granada, Spain. American Journal of Physical Anthropology 103, 433–441. Cano, R. J., Poinar, H. N., Pieniazek, N. J., Acra, A. & Poinar, G. O. (1993). Amplification and sequenzing of DNA from a 120–135 million-year-old weevil. Nature 363, 536–538. Cattaneo, C., Gelsthorpe, K., Phillips, P. & Sokol, R. J. (1993). Blood residues on stone tools: indoor and outdoor experiments. World Archaeology 25, 29–43. Cattaneo, C., Gelsthorpe, K., Phillips, P. & Sokol, R. J. (1995). Differential survival of albumin in ancient bone. Journal of Archaeological Sciences 22, 271–276. De Jong, E. W., Westbroek, P., Westbroek, J. F. & Bruning, J. W. (1974). Preservation of antigenic properties of macromolecules over 70 Ma. Nature 252, 63–64. Eglinton, G. & Logan, G. A. (1991). Molecular preservation. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 333, 315–328. Franc, S., Marzin, E., Boutillon, M. M., Lafont, R., Lecheˆne de la Porte, P. & Herbage, D. (1995). Immunochemical and biochemical analyses of 20000–25000-year-old fossil cartilage. European Journal of Biochemistry 234, 125–131. Gilbert, J., Campillo, D., Arque´s, J. M., Olivares, E. G., Borja, C. & Lowenstein, J. (1998). Hominid status of the Orce cranial fragment reasserted. Journal of Human Evolution 34, 203–217. Ho¨ss, M., Dilling, A., Currant, A. & Pa¨a¨bo, S. (1996). Molecular phylogeny of the extinct ground sloth Mylodon darwinii. Proceedings of the National Academy of Sciences of the United States of America 93, 181–185. Krings, M., Geisert, H., Schmitz, R. W., Krainitzki, H. & Pa¨a¨bo, S. (1999). DNA sequence of the mitochondrial hypervariable region II from the Neandertal type specimen. Proceedings of the National Academy of Sciences of the United States of America 96, 5581–5585. Krings, M., Stone, A., Schmitz, R. W., Krainitzki, H., Stoneking, M. & Pa¨a¨bo, S. (1997). Neanderthal DNA-sequences and the origin of modern humans. Cell 90, 19–30. Lindahl, T. (1997). Facts and artifacts of ancient DNA. Cell 90, 1–3. Lowenstein, J. M. (1981). Immunological reactions from fossil material. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 292, 143–149. Lowenstein, J. M. (1983). Fossil proteins and evolutionary time. Pontificiae Academia Scientiarum Scripta Varia 50, 151–162. Lowenstein, J. M. & Ryder, O. A. (1985). Immunological systematics of the extinct quagga (Equidae). Experientia 41, 1192–1193. Lowenstein, J. M., Sarich, V. M. & Richardson, B. J. (1981). Albumin systematics of the extinct mammoth and Tasmanian wolf. Nature 291, 409–411. Lowenstein, J. M. & Scheuenstuhl, G. (1991). Immunological methods in molecular palaentology. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 333, 375–380. Masters, P. (1987). Preferential preservation of noncollagenous proteins during bone diagenesis: Implications for chronometric and stable isotope measurements. Geochimica et Cosmochimica Acta 51, 3209–3214. Muyzer, G., Sandberg, P., Knapen, M. H. J., Vermeer, C., Collins, M. & Westbroek, P. (1992). Preservation of the bone protein osteocalcin in dinosaurs. Geology 20, 871–874. Newman, M. E., Ceri, H. & Kooyman, B. (1996). The use of immunological techniques in the analysis of archaeological materials—a response to Eisele; with report of studies at Head-Smashed-in Buffalo Jump. Antiquity 70, 677–682. Newman, M. E., Yohe, R. M. II, Ceri, H. & Sutton, M. Q. (1993). Immunological protein residue analysis of nonlithic archaeological material. Journal of Archaeological Science 20, 119–132.
Immunoglobulin G in 1·6 Million-year-old Fossil Bones from Venta Micena 175 Pa¨a¨bo, S., Higuchi, R. G. & Wilson, A. C. (1989). Ancient DNA and the polymerase chain reaction. The emerging field of molecular archaeology. Journal of Biological Chemistry 264, 9709– 9712. Palmqvist, P. (1997). A critical reevaluation of the evidence for the presence of hominids in lower pleistocene times at Venta-Micena, southern Spain. Journal of Human Evolution 33, 83–89. Pawlicki, R. & Nowogrodzka-Zago´rska, M. (1998). Blood vessels and red blood cells preserved in dinosaur bones. Annals of Anatomy 180, 73–77. Poinar, G. O. Jr (1993). Reply to Tomas Lindahl: Recovery of antediluvian DNA. Nature 365, 700. Poinar, H. N., Cano, R. J. & Poinar, G. O. (1993). DNA from an extinct plant. Nature 363, 677. Poinar, H. N., Ho¨ss, M., Bada, J. L. & Pa¨a¨bo, S. (1996). Amino acid racemization and the preservation of ancient DNA. Science 272, 864–866. Rainey, W. E., Lowenstein, J. M., Sarich, V. M. & Magor, D. M. (1984). Sirenian molecular systematics including the extinct Steller’s sea cow (Hydrodamalis gigas). Naturwissenschaften 67, 343–346. Rowley, M. J., Rich, P. V., Rich, T. H. & Mackay, I. R. (1986). Immunoreactive collagen in avian and mammalian fossils. Naturwissenschaften 73, 620–623. Schweitzer, M. H., Marshall, M., Carron, K., Scott-Bohle, D., Busse, S. C., Arnold, E. V., Barnard, D., Horner, J. R. & Starkey, J. R. (1997a). Heme compounds in dinosaur trabecular bone. Proceedings of the National Academy of Sciences of the United States of America 94, 6291–6296. Schweitzer, M. H., Johnson, C., Zocco, T. G., Horner, J. R. & Starkey, J. R. (1997b). Preservation of biomolecules in cancellous
bone of Tyrannosaurus rex. Journal of Vertebrate Paleontology 17, 349–359. Shoshani, J., Lowenstein, J. M., Walz, D. A. & Goodman, M. (1985). Proboscidean origins of mastodon and mammoth demonstrated immunologically. Paleobiology 11, 429–437. Smith, A. J., Matthews, J. B., Wilson, C. & Sewell, H. F. (1985). Plasma proteins in human cortical bone: in vitro binding studies. Calcified Tissue International 37, 208–210. Smith, A. B. & Littlewood, D. T. J. (1994). Paleontological data and molecular phylogenetic analysis. Paleobiology 20, 259–273. Tobias, P. V. (1998). Some cumments on the case for early Pleistocene hominids in South-Eastern Spain. Human Evolution 13, 91–96. Tuross, N. (1989). Albumin preservation in the Taima-taima mastodon skeleton. Applied Geochemistry 4, 255–259. Tuross, N., Behrensmeyer, A. K., Eanes, E. D., Fisher, L. W. & Hare, P. E. (1989). Molecular preservation and crystallographic alterations in a weathering sequence of wildebeest bones. Applied Geochemistry 4, 261–270. Tuross, N. & Stathoplos, L. (1993). Ancient proteins in fossil bones. Methods in Enzymology 224, 121–129. Ulrich, M. M., Perizonius, W. R., Spoor, C. F., Sandberg, P. & Vermeer, C. (1987). Extraction of osteocalcin from fossil bones and teeth. Biochemical and Biophysical Research Communications 149, 712–719. Wilson, A. C., Carlson, S. S. & White, T. J. (1977). Biochemical evolution. Annual Review of Biochemistry 46, 573–639. Wyckoff, R. W. G. (1972). The Biochemistry of Animal Fossils. Bristol: Scientechnica. Zihlman, A. L. & Lowenstein, J. M. (1996). A spanish Olduvai? Current Anthropology 37, 695–697.