PERSPECTIVES IN PROTEIN AND AMINO ACID GEOCHEMISTRY
PERSPECTIVAS EN GEOQUIMICA DE PROTEINAS Y AMINOACIDOS
Torres1, T., Ortiz1, J.E., Llamas1, J., Canoira1, L, Lucini1, M., Garcia-Martinez1, M.J.,
Garcia de la Morena1, M.A. and Julia2, R.
' B i o m o l e c u l a r S t r a t i g r a p h y L a b o r a t o r y . M a d r i d S c h o o l of M i n e s . Rfos R o s a s 2 1 . E - 2 8 0 0 3 M a d r i d ( S p a i n ) . E-mail: m i i s e o @ m i n a s . u p m . e s
-Institut d e C i e n e e s d e la Terra " J a n m e A l m e r a " . C S I C . B a r c e l o n a .
Abstract
R e c e n t a d v a n c e s in p r o t e i n a n d a m i n o a c i d s a r e p r e s e n t e d . In spite of t h e interest o n u s i n g p r o t e i n a n d a m i n o a c i d s g e o chemistry for d a t i n g p u r p o s e s d e c r e a s e d for s o m e y e a r s , a b e t t e r u n d e r s t a n d i n g of e r r o r s o u r c e s a n d d i a g e n e s i s p r o c e s s e s a l l o wed to obtain a powerful tool for g e o l o g i c a l u s e . E r r o r s c a n be g r o u p e d in t h r e e different c l u s t e r s : a n a l y t i c a l error, s a m p l e depending e r r o r - i n l r a s h e l l . i n t r a g e n u s . i n t e r g e n u s a n d m i c r o e n v i r o n m e n t - a n d p a l a e o e n v i r o n m e n t d e p e n d i n g e r r o r - t h e r m a l history, s e d i m e n t g e o c h e m i s t r y , m o i s t u r e , d i a g e n e s i s . e t c . All t h e s e e r r o r s c a n b e c a l c u l a t e d a n d / o r e s t i m a t e d . A m i n o acid g e o chemistry study h a s b e e n f o c u s e d o n Q u a t e r n a r y d a t i n g : relative ( A m i n o s t r a t i g r a p h y ) a n d a b s o l u t e d a t i n g ( A m i n o c h r o n o l o g y ) of lacustrine, fluvial, m a r i n e d e p o s i t s a n d m a m m a l r e m a i n s h a v e b e e n o b t a i n e d . Protein p r e s e r v a t i o n h a s b e e n a n a l y z e d to ascertain fossil D N A p r e s e r v a t i o n p o t e n t i a l .
Key W o r d s : o r g a n i c g e o c h e m i s t r y , p r o t e i n s , a m i n o a c i d s , s t r a t i g r a p h y , g e o c h r o n o l o g y . Q u a t e r n a r y .
R e s u m e n
En este trabajo se p r e s e n t a n los avarices m a s r e c i e n t e s e n el u s o d e la g e o q u f m i c a d e protefnas y a m i n o a c i d o s . A p e s a r d e que el e m p l e o d e la g e o q u i ' m i c a d e a m i n o a c i d o s y protefnas para d a t a c i o n d i s m i n u y o d u r a n t e a l g u n o s a n o s . un m e j o r c o n o c i -miento de las fuentes d e e r r o r y de los p r o c e s o s d i a g e n e t i c o s le h a n p e r m i t i d o r e v e l a r s e c o m o u n a heiTamienta m u y valida p a r a su e m p l e o en G e o l o g f a . L o s e r r o r e s se p u e d e n a g r u p a r en tres g r u p o s d i f e r e n t e s : e r r o r analftico. e r r o r d e la m u e s t r a i n t e r concha, i n t r a g e n e r o , i n t e r g e n e r o y m i c r o a m b i e n t a l y e r r o r p a l e o a m b i e n t a l h i s t o r i a t e r m i c a . g e o q u f m i c a del s e d i m e n t o , h u m e dad. d i a g e n e s i s . e t c . E s t o s tres e r r o r e s se p u e d e n c a l c u l a r y / o e s t i m a r . El e s t u d i o d e la g e o q u f m i c a d e a m i n o a c i d o s se h a c e n -trado en la d a t a c i o n del C u a t e r n a r i o : se h a n o b t e n i d o d a t a c i o n e s r e l a t i v e s ( A m i n o s t r a t i g r a f f a ) y a b s o l u t a s ( A m i n o c r o n o l o g f a ) tanto de d e p o s i t o s l a c u s t r e s , f l u v i a l e s y m a r i n o s c o m o d e r e s t o s d e m a m f f e r o s . S e ha a n a l i z a d o la p r e s e r v a c i o n d e las protef nas para p r e d e c i r la p r e s e r v a c i o n d e D N A fosil.
Palabras c l a v e : g e o q u f m i c a o r g a n i c a . p r o t e i n a s . a m i n o a c i d o s . g e o c r o n o l o g f a . C u a t e r n a r i o .
Introduction
The a m i n o acid a n a l y s i s of fossils w a s initiated 5 0 years a g o ( A b e l s o n , 1954) a l t h o u g h it d i d n ' t b e c a m e w i d e s p r e a d until the last
3K\.& <i&\Vj 1970s w h e n t h e d e v e l o p m e n t WPTaster analytical methods occurred. Initial
works from these period c a m e from B a d a (1972. 1973), Bada and Prostch ( 1 9 7 3 ) . Hare (1969. 1971. 1974). Hare and Mitterer ( 1 9 6 6 ) , Hare and Hoering (1967). T h e initial c o n t r o versies about this method mostly a p p e a r e d because the behavior m e c h a n i s m of the r a c e m i -zation kinetic, which was supposed to be a first
II" Congreso Iberico de Geoqui'mica /IIP Congreso de Geoqui'mica de Espana hragoza-2001
H Logo, E. Arranz v C. Gale (Eds.)
)p. 155-174
'SBN: 84-9.10635-8-4
o r d e r c h e m i c a l reaction that a l l o w e d to obtain indirect and i m m e d i a t e d a t i n g s . w a s poorly k n o w n .
Recent works (Wehmiller. 1984, 1993, 1995 and W e h m i l l e r el al., 1992. 1995) have provided
e n o u g h criteria to evaluate the real possibilities of this method.
Amino acids and dating
In almost all living beings all a m i n o acids are L-amino acids, that is, the a m i n o group is placed at the "left s i d e " of the molecule. O r g a n i s m s incorporate amino acids as part of protein mole cules into their skeleton. In mineralized skeletal c o m p o n e n t s as shells, enamel teeth and bones, a m i n o acids are located in intracrystalline and intercrystalline positions, being the former less prone for leaching. A m i n o acids in collagen, a non-full mineralized structure, are less protected.
-After the o r g a n i s m ' s death the racemization process starts and Lamino acids c h a n g e into D -a m i n o -acids until the D/L r-atio equ-als 1. th-at is. w h e n the r a c e m i c e q u i l i b r i u m is r e a c h e d . Nevertheless, a m i n o acids with more than one C atom at their molecule, as isoleucine, can under go a process called epimerization. which con sists on the transformation of the L-enantiomer (Lisoleucine) into a different D e n a n t i o m e r ( D -/4/fo-isoleucine) not present in living beings. In this case the equilibrium is reached at a D-Allo/L-\\e ratio value of 1.3.
A n y w a y the raeemization/epimerization pro cess can be considered as a first order reaction kinetics which is temperature controlled. T h e equation of this process is:
L-enantiomer K,
: D-enantiomer
I n 1 + D/L
1 - D/L
C = (l + K ' ) KLt
In ( D / L ) : a m i n o a c i d e n a n t i o m e r s ratio. C : m e t h o d i n d u c e d r a c e m i z a t i o n t: l i m e K = K „ / K|
K . : r a c e m i z a t i o n reaction e q u i l i b r i u m c o n s t a n t
In the Biomolecular Stratigraphy Laboratory
of the Madrid School of Mines, eight a m i n o acid
pairs of e n a n t i o m e r s are usually determined: ala nine, valline, proline, D-A//r;-isoleucine and L-isoleucine. leucine, aspartic acid, glutamic acid and phenylalanine. In s o m e cases (dentine sam ples) L-hydroxyproline is also identified.
U p to now w e have used a Goodfriend ( 1 9 9 1 ) and G o o d f r i e n d and M e y e r (1991 )'s protocol for s a m p l e p r e p a r a t i o n method. For t h e r a c e m i z a t i o n r a t i o s m e a s u r e m e n t s a Hewlett Packard 5 8 9 0 ser. II G a s Chroma-t o g r a p h wiChroma-th a H P 6 8 9 0 a u Chroma-t o s a m p l e r and NPD d e t e c t o r has been e m p l o y e d . T h e main problem of this m e t h o d is the high g a s e s consumption and a c o m p l e x , time c o n s u m i n g , s a m p l e prepa ration m e t h o d . A l s o , the r e q u e s t e d sample weight is too large: 80 trig. N o w we have chan ged to a Hewlett Packard HP 1100 HPLC a n a l y z e r w i t h f l u o r e s c e n c e d e t e c t o r that d e c r e a s e s the s a m p l e p r e p a r a t i o n time and allows to use very small s a m p l e s (0.02 mg). T h e s a m p l e preparation m e t h o d is largely auto m a t i z e d a c c o r d i n g to Kaufman and M a n l e y ' s ( 1 9 9 8 ) p r o t o c o l .
Method constraints (Sources of error)
According to Murray-Wallace (1995) there are 6 factors (F) affecting the method reliability (Fig. 1): F l - A n a l y t i c a l error; F2-Intrafossil va riation: F3-intraspecific variation: F 4 - G e n u s effect; F5-Small scale environmental influence; F6-big scale environmental influence.
A n a l y t i c a l
S a m p l i n g e r r o r
F A C T O R 6
F A C T O R 2
"I'IMK " A "
F A C T O R 1
a n a l y t i c a l w i t h i n s p e c i e s error
C V 3 %
i n l r a - I o s s i l v a r i a t i o n
C V - 8 %
F A C T O R ;
\ \ i t h i n s p e c i e s
i n t r a - i ' o s s i l
v a r i a t i o n
C V ~ 1 2 %
F A C T O R 4
e f f e c t
F A C T O R 5
E f f e c t s o f n a t u r a l
v a r i a t i o n w i t h i n a d e p o s s i t d u e t o :
a: r e w o r k i n g ( b i o c e n o s i s v s .
t h a n a t o e e n o s i s )
b : l o c a l o r m i c r o - s c a l e v a r i a t i o n o f
d i a g e n e l i c c o n d i t i o n s c : d i f f e r e n t i a l w e a t h e r i n g
e r o s i o n a n d e x p o s u r e C V - ~ M ) / o d : c o n t a m i n a t i o n
C V :
S a m p l i n g e r r o r : s i n g l e d e p o s i t
B e t w e e n s i t e v a r i a t i o n d u e t o :
a: r e g i o n a l v a r i a t i o n l o f d i a g e n e t i c
1 c o n d i t i o n s
b : r e g i o n a l d i t f e r e n c e s I in w h e a t h e r i n g
e n v i r o n m e n t s
S a m p l i n g e r r o r : r e g i o n a l a m i n o s t r a t i g r a p h i c s t u d i e s
c v
;F i g u r e 1: S o u r c e s of e r r o r in a m i n o acid r a c e m i z a t i o n d a t i n g ( M u r r a y - W a l l a c e . 1995).
-F l : It is an easy work to face this problem which can be solved according to usual labora tory practice standards.
F2: S o m e authors (Wehmiller, 1993) have reported intrafossil variations on mollusk shells. This means that slight a m i n o acid racemization ratio differences, which appeared between sam ples taken from the u m b o and from the ventral border, are due to the different internal shell structure. To avoid the differences in our experi ments (Torres et al.. 1999a). Figure 2 and table
1, our sampling protocol r e c o m m e n d s to obtain
the samples from the same shell area ( u m b o ) . On gastropods, samples are obtained from the last whorl near the aperture or the parietal callus
if present. Recently a very interesting p h e n o m e non (Torres et al., 2000a) has been observed on m a m m a l teeth: a very high racemization ratio variation has been found on s a m p l e s taken across the tooth root. This can be explained as a diagenesis-linked process which we avoided by drilling through the central part of the whole
root, equidistant from the c e m e n t layer and the pulp cavity.
D/ L A s p i n s n m i n m a x CV K
Panopea 0 . 1 8 0 . 0 4 13 0 . 1 2 0 . 2 5 22 0 . 6 8
Ostrea 0 . 2 5 0 . 0 3 7 0 . 1 9 0 . 1 9 12 0 . 8 1
G l y c y m e r i s 0 . 2 4 0 . 0 3 6 0 . 2 0 0 . 2 0 13 0 . 8 3
Table 1: E l e m e n t a l statistics from intrashell a m i n o acid r a c e m i z a t i o n in s a m p l e s from a H o l o c e n e spit b a r in H u e l v a ( S p a i n ) (Torres et al.. 1999a). m = m e a n : s= s t a n d a r d d e v i a t i o n , n = n u m b e r of a n a l y s i s , m i n = m i n i m u m v a l u e , m a x = m a x i m u m v a l u e , C V = variation coefficient. r= c o r r e l a t i o n coefficient.
We have studied this F2 variation on recent material (Holocene) from a spit bar near Huelva, southwest of the Iberian Peninsula. (Torres et al. 1999) where a high variability of the aspartic acid racemization ratio was found between samples taken near the u m b o and the ventral border of dif
ferent mollusk shells: Panopaea sp, Glycymeris sp. and Ostrea sp., being noticeable that the hig hest racemization ratio values appeared in sam ples recovered from the ventral border of the shells. Obviously this study cannot be made on small shell representatives, or (such as) ostracods.
D / L A S P
F3: In spite of in most cases an absolute isoch-rony can be assumed for the whole sample set pic
ked from a single bed, a variation in the racemiza-tion ratios obtained from different shells can be expected. This fact makes necessary to analyze at least five samples from each bed. but in most cases the laboratory experience will act as procedure guideline to determine the highest admissible stan
dard deviation value and the requested minimum number of samples to be analyzed. W h e n "statisti cal representative samples" such as ostracods or small mollusks are analyzed the measured racemi-zation ratios are quite similar because of a great number of individuals needed (e.g. a standard
ostracod sample consists of two thousand valves), which makes the standard deviation smaller.
F4: The genus effect has been noticed since the method began to be applied and can be a very important source of error. In fact, different genus representatives can never be analyzed together. In figure 3 the variation of racemization/epimeriza-tion ratios of leucine, glutamic acid and D-Allo/L-Ile isoleucine of different pelecypod. gastropod and algae representatives appear. It is possible to notice that there are differences between the rea ched racemization ratios of the different genera but being generally greater between markedly dif ferent " g e n u s " , as " O p e r c u l a " and " C h a r a " are.
O "+-» CD
i_
C o •-+-> CO N " E CD O CO o •+-» CO N CD E C L L U 1.4 1.3 1.2 1.1 1.00.9
0.80.7
0.6 0.5 0.4 0.3 0.2 0.1 0.0 co o o •4—' If) o' A i \
\ \
I k
w / Jr
N T l •
x
1 X
-D-Allc
>/L-Ile
- m - D
I. Leu
—A-
D/L G lu
TO
0) -a o V, CDO
•g if) a.ro
CD O CD O "CD X CD co T 3 O CO if) o t z ro D . co co > i 5 Z3 -CD Q . O CO CO x : OLower Pleistocene (CCTB-2-2-182) sample
F i g u r e 3 : E p i m e r i / a t i o n / r a c e m i z a t i o n r a t i o s of i s o l e u c i n e . l e u c i n e a n d g l u t a m i c acid from different g e n u s r e p r e s e n t a t i v e s s a m pled at the s a m e strata ( L o w e r P l e i s t o c e n e ) in C u l l a r - B a / a B a s i n ( G r a n a d a . S p a i n ) ( T o r r e s . 1999).
T h e a f o r e m e n t i o n e d differences a p p e a r to be m o r e m a r k e d in D-A//o/L-Ile e p i m e r i z a t i o n ratios. Nevertheless the g e n u s effect d i m i n i s h e s when old s a m p l e s are analyzed and the m e a s ured racemization ratios d o not diverge in a
m a r k e d l y way as they d o in very y o u n g s a m ples. T h i s can be clearly o b s e r v e d in the D-Allo/L-Ue e p i m e r i z a t i o n h i s t o g r a m of different marine gastropod and p e l e c i p o d a g e n e r a from a raised b e a c h of C a b o de H u e r t a s (Alicante, East
-of Spain) dated ca. 120 ka (own u n p u b l i s h e d data).
F5: Small scale environmental variation. This factor comprises a wide variety of local environ m e n t a l a s p e c t s such a s : m o i s t u r e , m e t a l l i c cations, buffered environment, reworking, ero sion and exposure to external agents, burial depth and c o n t a m i n a t i o n .
Moisture has been demonstrated to be one of the most important factors affecting a m i n o acid racemization/epimerization: a total lack of mois ture can inhibit this process and differential per meability in fossil bearing bed loci, or differen tial leaking, in caves, can explain high ratio values variability. pH values of interstitial water can also explain s o m e differences.
Metallic cation presence, associated to detrital
or chemical mineral grains ( M g+ + and C u+ +) can
produce a m i n o acid chelation and. thus, a higher amino acid racemization rate can be produced.
Reworking has been used as a factor explai ning high racemization values variance and can affect strongly the method reliability in samples from recent deposits: reworking factors are of very different origin: bioturbation by burrowing organisms can displace fossils from a bed to ano ther making it older or younger. Anyway, a statis tical analysis of an adequate number of analyzed samples can allow us to discard spurious results. Erosion of former deposits and further deposition of fossil remains into modern ones either by wave or river action is a well known phenomenon. In any case this seems not to be very important in old deposits, but in recent ones can be an important source of error making necessary to analyze a big number of samples and to discard erratic results.
O Y A M B R E . D / L ( L E U + A S P + G L U )
> - 5 4 . 4 2
t
DC
5
5
W 2 2 . 7 9
1 C 2 C 3 C 4 C 7 C 1 0 C 5 G 1 1 G 8 G 9 G
OBSERVATIONS
F i g u r e 4: S i m i l a r i t y a n a l y s i s of r a c e m i z a t i o n r a t i o s of O y a m b r e old b e a c h of E e m a g e . C a r e p e l e c i p o d a (Cardiwn sp.) s a m p l e s : G a r e g a s t r o p o d a (Murex s p . ) s a m p l e s ( G a r z o n et al.. 1 9 9 6 ) .
A l s o , reworking as an error source can never be discarded. G a r z o n et al. (1996) described a very paradoxal case in O y a m b r e (Cantabria, Spain) raised beach. Two marked clusters of a m i n o acid racemization ratios were described (figure 4 ) . T h e h i g h e s t r a c e m i z a t i o n ratios values appeared on Cardium sp. shells while the lowest, on g a s t r o p o d (Murex sp.) s a m p l e s . However, according to our experience (Torres et al.. 2000b), gastropods show higher racemiza tion ratios than pelecypods. T h u s , this fact can be explained in terms of reworking although due to the well k n o w n fragility of Cardium sp. shells, w e finally c o n c l u d e that g a s t r o p o d s {Murex sp.) shells were introduced in the sand beach bed by digging.
Burial depth can seriously affect the final a m i n o acid analyses result. Very shallow buried fossils located in open air sites, w h i c h is very c o m m o n in marine terraces, can suffer erosion, recent contamination and, the most important, non h o m o g e n e o u s exposition to solar heating. A c c o r d i n g to this fact and to our own experien ces, raised beaches are the most difficult sites to be dated through a m i n o acid racemization analy sis being necessary to take s a m p l e s , w h e n possi ble, from t r e n c h e s as d e e p as p o s s i b l e . In Mediterranean old shoreline deposits, green and m a g e n t a patches are very frequent coating on fossils and sediments recovered from holes 10 c m depth, which are interpreted as r e m a i n s , even still living, of simple organisms that can prolife rate under extreme environmental conditions. T h e a m i n o acids of these organisms, either L or D a m i n o acids, can be a very important source of error.
In brief: small scale environment variation-linked uncertainty can be corrected by m e a n s of careful statistical analysis of analytical results which, in s o m e cases, will m a k e necessary to analyze a large set of samples. T h e full sedi-m e n t o l o g i c a l , t a p h o n o sedi-m i c a l and diagenetical understanding of the sample bearing bed seems to be fundamental.
F6: Big scale environmental influence lies on t w o main factors: thermal history and diagenetic conditions. Because racemization is a thermal controlled first order c h e m i c a l reaction, the regional thermal histories of the sampling sites play a very fundamental role in the final dating results. A s an simple e x a m p l e , the estimated m e t h o d range in Alaska an be estimated in > 4 M a while in N e w G u i n e a is > 200 ka. A c c o r d i n g
-to our labora-tory experience, important racemi zation ratio differences can be expected between s a m p l e s from the M e d i t e r r a n e a n B o r d e r , Atlantic Coast, Balearic and Canary Islands and Outback. It is necessary to work with local m o d e l s which can be exported from one zone to another. In figure 5 the calculated ages from dif ferent a m i n o acid racemization ratios in samples from G u a n c h e localities (Canary Islands) and raised beaches appear. In the lower part of the X Y plot the calculated age from specific algo rithms m a d e from the Canary Islands thermal history appear while in the upper part the calcu lated ages are represented when the Outback of the Iberian Peninsula are used. In this case very marked differences in calculated ages appear, being source of unacceptable w r o n g values. Not all a m i n o acid behave in the s a m e way with ther mal history differences and. therefore, it is diffi cult to export age calculation models from one area to another.
AGE ( C ' & U / T h ) ( x 1 0 Y e a r s )
PRIEGO MODEL CANARY ISLANDS MODEL
• D-/4/toL-l,e A LEU 1 ASP * PHE + GLU
F i g u r e 5 : C o m p a r i s o n of r e s u l t s after the a p p l i c a t i o n of a g e c a l c u l a t i o n a l g o r i t h m s of C e n t r a l S p a i n and the C a n a r y I s l a n d s ( d a s h e d lines) to s a m p l e s from the latter locality ( G a r c f a - A l o n s o et at., 1996).
S a m e difficulties will appear linked to diage-netical history of the sites: racemization rates and, even, protein preservation through time, can be used together if samples c o m e from different sedimentolodical and/or diagenetical environ ments. Caves with very h o m o g e n e o u s sedimen-tological and diagenetical conditions: stable ther mal history, constant moisture, buffered environ m e n t etc. h a s a l l o w e d the B i o m o l e c u l a r Stratigraphy Laboratory to date an important
n u m b e r of localities where tooth protein was still well preserved. However, a total lost of proteins has been observed in fossil bones at open air sites or in shallow lacustrine environments under arid climate.
Which amino acid can be used
O n e of the first questions to be solved is which a m i n o acid can be used for dating p u r p o ses. According to their racemization rates a m i n o acids can be described as "fast" or " s l o w " d e p e n d i n g on their activation energy values. The lower energy activation value appears in aspartic acid racemization process (cf. Torres et al.. 2001 and Goodfriend and Meyer, 1991) being the most adequate to date recent samples. In fact, (cf. Goodfriend, 1992; Goodfriend et al.. 1992) it was proposed for dating secular or even deca dal events.
Glutamic acid, phenylalanine, isoleucine, leu cine and proline with higher activation energy racemize with more paucity, being adequate for old samples. Furthermore, not all a m i n o acids show similar reliability. Isoleucine seems to be the most reliable a m i n o acid probably because D-A//<visoleucine is not present in the Nature and tardive contamination processes can be easier detected. A n y w a y we have analyzed the different a m i n o acid reliability in samples from raised beach deposits from C a b o de Huertas (Alicante. Spain) and Garrucha (Almeiia. Spain). Marine terraces constitute one of the most problematic deposits to be dated because a high number of source errors converge on them: shallow burial, carbonate (and a m i n o acid?) vertical migration to form caliches, differential (usually high) sun hea ting and contamination from the rizosphere. interstitial extremophyle organisms, g u a n o from sea birds, etc. To analyze the reliability of some a m i n o acids, Torres et al. (2000b), leucine, aspar tic acid and glutamic acid racemization values from samples of a pelecipod shell (Glycymeris sp) were plotted against D-A//r>isoleucine/L-Isoleucine epimerization values as a funcion of burial depth (figure 6). D / L Leu values were found to well correlate with D-Allo/L-l\e values for all burial depths. On the other hand, for both Glu and A s p . a good correlation with D-Allo/L-Ile values was found only for deep buried sam ples. This fact suggests that samples recovered from shallow or intermediate burial depth (less than 25 cm) m a y be diagenetically altered with respect to A s p and Glu.
-0 . 8
j A Shallow |
I x Intermediate
J • Deep
tP
3n cA
•
D - A l l o / L - l l e
01 ' 0.X
D - A l l o / L - l l e
0.4 On 0.8 D - A l l o / L - l l e
Figure 6: X Y p l o t s of D / L A s p . G l u a n d Leu v a l u e s v s . D -Allo/L-Ile v a l u e s in Glycymeris s p . s a m p l e s f r o m s o m e rai sed b e a c h e s of t h e C a b o d e H u e r t a s ( A l i c a n t e ) a n d Garrucha ( A l m e r f a ) ( T o r r e s el al.. 2 0 0 0 b ) .
Because a correlation b e t w e e n age and terra ce elevation can be established, it seems to be evident that a m i n o acids from samples taken from the higher terraces will show the higher racemization/epimerization values, a s s u m i n g a lack of n e o t e c t o n i c . T h e calculation of the
Spearman's rank correlation between the afore mentioned four a m i n o acids racemization ratios and their rank position (elevation) demostrate that the o r d e r of reliability is: D-Allo/L-lle>Leu>Glu>Asp. This m e a n s that in favorable environments, w h e r e h o m o g e n e o u s conditions
have been maintained through time, as lacustri ne deposits and caves, all the identified a m i n o acids can be used for dating purposes while in samples from less favorable paleoenvironments, as marine terraces shelters, the best results are
obtained through D-Allo/L-Ue and D / L Leu determination. Anyhow, our experience indica tes that the best way to use this method is to han dle the racemization ratios of the eight a m i n o acids that we usually identify and check through statistics the lack of contamination or even the
Tightness of the genus, or even material, identifi
cation. In Figure 7 the cluster analysis calculated using five D/L ratios measured on different materials, recovered from a fossiliferous bed in Venta Micena (Granada, Spain) appears. This analysis reveals two a m i n o acid racemization ratios clusters: fragments of gastropods (mostly L y m n a e a c i i d a e ) and o p e r c u l a (from Bithyniidae). In this case there is a Gf sample (fragment of gastropod), which was formerly " w r o n g " classified, that according to the cluster where it was grouped, must be identified as opercula (O). This cluster analysis also reveals the importance of the intergenus-linked error. T h e s e samples, which are almost one million
years old, still maintain the former differences
between the proteins, m a d e of different b o n d e d a m i n o acids, linked to their mineral structure (shell) building.
V E N T A M I C E N A D/L ( D - A / f o / L - l l e + L E U + A S P + G L U )
O O O O O O ' f G Gf G G f G G t G t G f
S A M P L E S
F i g u r e 7: C l u s t e r a n a l y s i s of the a m i n o acid r a c e m i z a t i o n ratios ( D - A l l o / L - l l e . D / L L e u . D / L A s p . D / L G l u ) from G a s t r o p o d a a n d O p e r c u l a r e m a i n s r e c o v e r e d from Venta M i c e n a ( G r a n a d a . S p a i n ) , a L o w e r P l e i s t o c e n e fossilife r o u s locality (Torres et al.. 1997).
A minostratigraphy
Taking into account the p r o b l e m s involved in the i m m e d i a t e application of a m i n o acid racemi zation rates in age calculation a first step could be the definition of an aminostratigraphical scale for local use. T h e s e scales must be based on a m i n o acid racemization rates of similar genus (mollusca. ostracoda...) and are related only to a limited g e o g r a p h i c a l realm w h o s e t h e r m a l (palaeoclimatological) history had seemed to be
-similar during Pleistocene times as well as with a similar s e d i m e n t o l o g i c a l e n v i r o n m e n t . We widely use aminostratigraphy where algorithms for age calculations are not available. Therefore, aminostratigraphy is a powerful tool for geolo gical correlation. Higher, lower or similar race mization ratios c o m p a r e d with the ones from the standard site indicate that the studied s a m p l e is older, y o u n g e r or isochronous, respectively.
Thus, aminostratigraphy consists of arran ging geological sites into a sequence on the basis of observed clusters of racemization ratios, esta blished using representatives of one s a m e zoolo gical genus. This method is an excellent tool for the correlation of sea level oscillation-linked deposits: marine and fluvial terraces indicative of warm and stable climate periods (Kaufman, 1992: Hearty et al, 1992; Miller and M a n g e r u d , 1985), and is especially useful for the determi nation of n e o t e c t o n i c s affecting P l e i s t o c e n e deposits ( D u m a s et al.. 1988). In s o m e cases, and mainly where almost continuous sedimenta tion has taken place, as in Cullar-Baza continen tal Basin (Granada, Spain), it is necessary to produce numerical-age results, the most c o m
mon being a g e - c a l i b r a t e d (l 4C or U - s e r i e s )
results ( D u m a s et al.. 1988; Goodfriend, 1987; Hearty et al, 1992; Wehmiller, 1993).
The analysis of Pleistocene raised beaches is one of the most satisfactory applications of ami nostratigraphy. We have employed this method in the relative dating of marine terraces of the C a b o de Huertas (Alicante) z o n e (Figure 8), and Garrucha (Almerfa). According to the racemiza tion ratio values it has been possible to establish both the aminostratigraphy of each of these zones and a whole aminostratigraphycal scale for the Mediterranean Border of the Iberian Peninsula, which probably is still incomplete. Our scale has been based on the D-A//c^Isoleucine/L-Isoleucine epimerization ratios of Glycymeris sp., the most common p e l e c y p o d found in the m a r i n e Pleistocene record of the Iberian Peninsula. The results of this study demonstrated that the present height above the sea level of the marine terraces can never be used as a stratigraphic argument and that neotectonics played a very important role rai sing, sinking and tilting terraces.
According to the published data, cf. Torres et al. (2000b) (Figure 9), five high-sea-level events (aminozones) can be distinguished:
1. Uppermost C a b o de Huertas terrace (FAR-A)
2. La Marina ( G A 5 - 1 . 2 ) and La Gurulla ( G A 2 )
3. C a b e z o de la Pel la ( G A 1 6 )
4. Tyrrenian U p p e r terraces: G a r r u c h a Castle ( G A 1 4 - 2 , G A 1 7 ) Puerto Rey ( G A 1 5 ) and upper C a b o de Huertas (FAR-C).
5. Tyrrenian lower terraces: G a r r u c h a Castle
( G A 1 4 - 1 ) and l o w e r m o s t C a b o de H u e r t a s ( F A R - D ) terrace.
A/l
1 2
0 9
0 8
0 7
-
0.60 . 5
0 4
0 3
0 2
-0 1
0 0
-M A L L O R C A F A R O D E H U E R T A S
Hearty (1967)
| + +
A/l
- 1 2
— 1.1
- 1 0
0 9
0 8
0 7
0.6
h 0 5
0 4
|—0 3
- 0 2
- 0 1
- 0 0
F i g u r e 9 : A m i n o s t r a t i g r a p h y of I b e r i a n M e d i t e r r a n e a n B o r d e r m a r i n e t e r r a c e s ( H e a r t y . 1 9 8 7 ' s a m i n o z o n e s are a d d e d ) (Torres et al.. 2 0 0 0 b ) .
T h e aminostratigraphy of Pleistocene bears of the Iberian Peninsula, a very important cha llenge for us, has been established too. In this case the analytical m e t h o d d e v e l o p m e n t was a " s w e a t i n g , blooding and t e a r i n g " process that needed m o r e than six years to be a c c o m p l i s h e d . T h e final success was reached after the adoption of a modified dialysis protocol published in M a r z i n (1990). T h e aim of this process is to a n a l y z e only h o m o g e n e o u s well p r e s e r v e d collagen proteins. Probably the good results obtained can be linked to the special g o o d envi ronmental conditions of caves w h e r e stable ther mal history, constant moisture presence in sedi
ments, buffered ( C a C 03) and low c o n t a m i n a t i o n
opportunities favored the consistent a c c o m p l i s hment.
A c c o r d i n g to the results o b t a i n e d (figure 10), it is p o s s i b l e to define an u n q u e s t i o n a b l e
-M i d d l e P l e i s t o c e n e Ursus deningeri V. R e i c h , a m i n o z o n e . w h i l e Ursus spelaeus R o s . H e i n . localities g r o u p e d in t w o very different s u b a m i n o z o n e s , w h e r e the y o u n g e r o n e (with l o w e r a s p a r t i c acid r a c e m i z a t i o n ratios) indica tes that the last Ursus spelaeus r e p r e s e n t a t i v e s w e r e coeval with b r o w n b e a r (Ursus arctos Lin.). T h i s a m i n o s t r a t i g r a p h i y c a l scale will be very useful for a r c h a e o l o g y and p a l e o n t o l o g y . A l s o c o n f i r m s t h e S i m a de l o s H u e s o s ( A t a p u e r c a , B u r g o s ) m a n and b e a r r e m a i n s a g e .
D/L ASP
0 . 4 5 - 1
0 . 4 0
0 . 3 5
0 . 3 0
<).2>H
0 . 2 0
0.15
0.10—1
0 . 0 5
0 . 0 0
I 1 U.daiingerl
1 I U. spelaeus
—' rn
a
D
J ,
T
r
~ ~ i — i — i — i — i — i — i— i i i i—i—i—r~
1 I TR A X I k Hi X X SS I T A A U LI 131 SI HI!
F i g u r e 10: A m i n o s t r a t i g r a p h y of Iberian Ursus deningeri and Ursus spelaeus localities (Torres et al.. 2 0 0 1 ) .
Aminochronology
T h e transformation of the a m i n o acids race mization or epimerization ratios into datings must be m a d e through the use of calibrated methods: that is the analyzed samples must be correlated with datings obtained from, mainly,
radioactive m e t h o d s (, 4C . U/Th or E S R ) . The
steps to do this were: m e t h o d validation, first age calculation ecuations definition, refining of calculation e q u a t i o n s and application to the Venta M i c e n a (Orce. G r a n a d a ) site where very old h u m a n remains were found in a very rich m a m m a l paleontological bearing bed.
After setting up the analytical method in our laboratory, we worked to obtain an external vali dation of the method, in order to control any kind of systematic error that might introduce a bias in the results. Three samples from an interlaboratory amino acid racemization exercise (Wehmiller.
1984) were also analyzed: ILC-A (Saxidomus sp. ca. 5 0 ka). ILC-B (Mercenaria sp. between 100 and 250 ka) and ILC-C (Mercenaria sp. ca 1.000 ka). These samples were analyzed by 11 laborato ries in an exercise of interlaboratory control. The lack of bias and the validity of the process fit pro vided by our laboratory were confirmed (Table 2).
D / L [ S A M P L E ( P O W D E R ) ]
A m i n o a c i d I L C - A I L C - B I L C - C
D - A l l o / L - I l e I L R 0 . 2 1 2 ± 0 . 0 7 2 0 . 5 4 ± 0 . 0 1 6 2 1.215 ± 0 . 0 3 0
L A B 0 . 1 8 0 0 . 6 5 0 1.245
D / L Pro ILR 0 . 2 7 8 ± 0 . 0 6 8 0 . 5 9 5 ± 0 . 2 1 0 0.81 ± 0 . 2 6
L A B 0 . 1 9 5 0 . 5 0 7 0 . 7 8 6
D / L L e u I L R 0 . 1 9 6 ± 0 . 0 4 2 0 . 4 9 7 ± 0 . 0 9 8 0 . 8 3 3 ± 0 . 0 8 6
L A B 0 . 1 8 2 0 . 4 4 4 0 . 8 4 9
D / L A s p I L R 0 . 3 7 8 ± 0 . 0 5 6 0 . 7 0 5 ± 0 . 0 5 6 0 . 8 9 4 ± 0 . 1 5 8
L A B 0 . 3 7 3 0 . 7 2 8 0 . 9 1 4
D / L P h e ILR 0 . 2 3 9 ± 0 . 0 4 0 0 . 5 8 3 ± 0 . 0 5 9 0 . 8 7 3 ± 0 . 1 7 8
L A B 0 . 2 2 0 0 . 6 0 8 0 . 8 8 5
D / L G l u I L R 0 . 2 0 3 ± 0 . 0 2 2 0 . 4 3 2 ± 0 . 0 3 4 0 . 8 4 9 ± 0 . 0 7 0
L A B 0 . 1 8 5 0 . 4 2 6 0 . 8 3 2
ILR Inter Laboratory Results (confidence intervals were calculated 2a. LAB Laboratory E.T.S.I. Minus.
T a b l e . 2 : C o m p a r i s o n of D / L ratios b e t w e e n the i n t e r - l a b o r a t o r y r e s u l t s e x e r c i s e and t h e r e s u l t s o b t a i n e d by the B i o m o l e c u l a r S t r a t i g r a p h y L a b o r a t o r y of M a d r i d S c h o o l of M i n e s (Torres et al.. 1997).
S a m p l e s of molluscs from U / T h m e t h o d dated Pleistocene fluvial travertine terraces in the Priego area (Torres et al., 1994), were collec ted, prepared and analysed. T h e age of the sam pled terraces ranged between 6 and 113 ka. In
order to calculate the induced r acem i zat ion during sample preparation, seven samples of alive gastropods were also collected, prepared and analysed. Likewise, D/L ratios of old sam ples. 100-250 ka and ca. 1000 ka, were used.
-S p u r i o u s results were o b s e r v e d during inter pretation of c h r o m a t o g r a p h i c analysis of the samples from Priego: a n o m a l o u s D/L ratios in samples from the s a m e stratigraphic level (tra vertine terrace). All the observed deviations appeared as an excess of D a m i n o acids and see m e d to be related with the o b s e r v e d algae-fungi mat covering s o m e of the s a m p l e d gastropod shells. S o m e bacteria have D a m i n o acids in their cellular walls (Leive and Davis, 1980) and algae and fungi mats are usual bacteria d w e
llers. By the time Venta M i c e n a was sampled, we had already modified our recovery process in order to reduce the d a n g e r of c o n t a m i n a t i o n : the samples were obtained from d e e p e r parts of the strata, w h e r e there is no influence by the a t m o s p h e r e and sunlight. In order to avoid the influence of the spurious values, a previous sin gle regression analysis of the D and L a m i n o a -cid pairs in the s a m p l e s w a s performed. Table 3 s h o w s the average ratios of racemization of Priego s a m p l e s .
A g e a v e r a g e ( U / T h ) D - A / f o / L - I l e D / L L e u D / L A s p D / L P h e D / L G l u
T o d a y 0 . 0 0 0 ± 0 . 0 0 0 0 . 0 0 8 ± 0 . 0 0 0 2 0.051 ± 0 . 0 0 8 0 . 0 1 9 ± 0 . 0 1 1 0 . 0 1 6 ± 0 . 0 1 2
6 . 0 0 0 0.041 ± 0 . 0 0 8 0 . 2 1 5 ± 0 . 0 5 1 0 . 0 2 9 ± 0 . 0 4 4 0 . 0 2 8 ± 0 . 0 2 2
12.500 0 . 1 4 9 ± 0 . 0 7 8 0.071 ± 0 . 0 1 8 0 . 2 5 7 ± 0 . 0 1 9 0 . 0 7 9 ± 0 . 0 0 9 0 . 0 7 0 ± 0 . 0 0 6
2 0 . 0 0 0 0 . 0 1 3 ± 0 . 0 0 9 0 . 3 4 6 ± 0 . 0 4 0 0.121 ± 0.031 0 . 1 0 5 ± 0 . 0 1 3
1 0 5 . 0 0 0 0 . 2 3 7 ± 0 . 0 1 6 0 . 4 2 3 ± 0 . 0 2 6 0 . 5 9 4 ± 0 . 0 2 6 0 . 4 7 9 ± 0 . 0 4 0 0 . 2 5 2 ± 0 . 0 3 1
T a b l e 3 : A v e r a g e D / L ratios of m o l l u s c s a m p l e s from U / T h - d a t e d t r a v e r t i n e t e r r a c e s in P r i e g o ( C u e n c a . C e n t r a l S p a i n ) (Torres
eta I., 1997).
Finally, prediction models were calculated using a set of 30 U/Th-dated samples from Priego: 21 freshwater gastropods, 6 terrestrial gastropods and 3 freshwater pelecipoda, as well as three samples of an inter-laboratory c o m p a r i son exercise (Wehmiller, 1994). T h e racemiza tion models are based on a first order reversible reaction (Bada and Protsch. 1973; S c h r o e d e r and Bada. 1976; Bada. 1985), the algorithms being based on a time square root (Vt) adjus tment (Goodfriend, 1987). This would appear to be justified since its use stabilizes the variance of the error that, in the case of time (t) adjus tment, b e c o m e s progressively greater with gro wing age samples.
The models obtained are as follows:
LEU vT= 1.17 + 1 1.38 x l n { L I + (D/L)J/[ 1
-(D/L)]}
(±0.62) (±0.88)
Correlation coefficient = 0.9774 (n = 33)
ASP V T = 2 . 1 7 + 10.02 x l n { [ l + ( D / L ) | / [ 1 -(D/L)]}
(±1.02) + (±0.85)
Correlation coefficient = 0.9740 (n = 32)
PHE v/T= ( ) . 9 9 + 10.26 x l n { [ l + ( D / L ) 1 / [ 1
-(D/L)]}
(±0.78) (±0.74)
Correlation coefficient = 0.9798 (n = 34)
G L U ^ 7 = 2.16 + 12.44 x ln{[l + (D/L)]/[l -(D/L)]}
(±0.65) (±0.82)
Correlation coefficient = 0.9816 (n = 36)
T h e algorithm selected for the isoleucine epi-merization m o d e l , following the adjustment of different models, was the one c o m m o n l y p r o p o sed by n u m e r o u s a u t h o r s ( M i t t e r e r , 1975; Goodfriend and Mitterer, 1988; Goodfriend, 1991). In this case the best fit was obtained through time (t) adjustment (c.c. 0.9872) rather than time square root (c.c. 0.9533). T h e model is as follows:
A/I t = 34.99 + 267.14 x ln{0.565/}0.565 -(A/I)/(l + A / l ) | }
(±25.8) + (±20.32)
Correlation coefficient = 0.9872 (n = 20 )
Although the final equilibrium state is affec ted only by temperature, for the D-Allo/L-Ue epimerization reaction, it was demonstrated the existence of a relationship, in Foraminifera, bet ween racemization percentages, age and the c u r r e n t m e a n a n n u a l t e m p e r a t u r e ( C M A T ) (Wehmiller, 1984). In our study, we have deter m i n e d the C M A T for the areas of Priego (Torres
-et al., 1994) and Reduena (Llamas -et al., 1995). w h i c h are c l i m a t o l o g i c a l l y similar, with a C M A T of 11-14 °C. Thus, it has been conside red that the same m o d e l s would be applicable. As regards the type of fossils under study, the initial results pointed to the existence of diffe rences between a m i n o acid kinetics d e p e n d i n g on the genera of mollusca analysed. T h e s e diffe rences were m o r e marked in the oldest samples analysed. In 10 samples of ancient freshwater gastropoda (Planorbis sp. and Radix sp.) from Priego (Torres et al., 1994) this effect was very marked for glutamic acid but negligible for leu cine. A poor correlation between the racemiza tion ratios of different a m i n o acids might be e x p l a i n e d in t e r m s of early d i a g e n e s i s (Goodfriend, 1991), e.g.: the loss of the most easily hydrolyzable a m i n o acid from the termi nal protein chains, aspartic acid, might be reflec ted in lower D/L ratios, this apparently being uncorrelated with the higher D/L ratios of the less easily hydrolyzable a m i n o acids, such as isoleucine.
10 individual samples from the Priego area were dated according the D/L leucine ratios, obtaining an average value of 7 3 3 ± 1 4 0 ka. For the P r i e g o area a C u r r e n t M e a n A n n u a l
Temperature ( C M A T ) between 11 and 14HC was
obtained. T h e s e values were coherent with those corresponding to this geographical location, and the C M A T was used according to the Wehmiller (1993) criteria. From our point of view, t e m p e rature might affect only the final D-Allo/L-Ue equilibrium stage. According to these results, other algorithm families were adjusted using only Planorbis sp. and Radix sp. remains, the adjusted m o d e l s being as follows:
L E U v T= 0 . 9 4 + 11.81 x l n { [1 + ( D / L ) ] / [ 1
-(D/L)]}
(±0.63) (±0.70)
Correlation coefficient = 0.9870 (n=32)
A S P vT= - 3.34 + 12.38 x ln{ [1 + (D/L)]/[l
- (D/L)]} (±2.31) (±2.24)
Correlation coefficient = 0.9050 (n=29)
P H E Vt = 0.48 + 15.64 x ln{ [1 + (D/L)]/}1 -(D/L)]}
(±2.88) (±3.71)
Correlation coefficient = 0.8304 (n=34)
G L U Vt = 0.33 + 21.93 x ln{ [1 + (D/L)]/[l -(D/L)]}
(±1.44) (±2.48)
Correlation coefficient = 0.9476 (n=37)
In this case the best model found for D -A//o/L-Ile epimerization was the square root of time (Vt) adjusted (c.c. 0.9495). T i m e (t) adjus tment had a lower correlation coefficient value (c.c. 0.9359).
A/I vT = 0.02 + 21.85 x ln{0.565/[0.565
-(A/I)/(l + A/I)]} (± 1.42) (± 2.31)
Correlation coefficient = 0.9495 (n=21)
Finally, at Venta Micena, only 21 analytical results were obtained from the 23 original sam ples, 2 being lost during the sample preparation process. T h e remaining samples could be clas sed into t w o groups: Opercula and Gastropoda. T h e former w e r e 2 - 3 m m sized, subcircular-ellipsoidal shaped dishes of aragonite, probably of Bithynia sp. T h r e e different genera were grouped into Gastropoda.
First, a cluster analysis was performed with G C analysis results, three different groups being identified. O n e of these groups, which differed widely from the other two. included four sam ples with very low racemization rates, this disa greeing with the other two. T h e s e results were interpreted as being due to the influence of con tamination, and were discarded. A second clus ter analysis of the 17 remaining samples, revea led the existence of t w o groups: Opercula. along with an unidentified fragmenta, and Gastropoda (2 Radix sp.. 1 Bulimus sp. and 6 unidentified gastropod fragmenta). In the second group a fragmenta sample presented very low racemiza tion ratios, with the exception only of D - Allo/L-Ile. This result has been interpreted as a diage-netic effect and was also discarded prior to per formance of the final cluster analysis. All the analyses were carried out on the basis of four a m i n o acids (D-Allo/L-\\e, leucine, aspartic acid and glutamic acid) since s o m e results for pheny lalanine were missing. W h e n the clustering was accomplished using complete results, phenylala nine included, the grouping obtained was the s a m e .
Final a g e c a l c u l a t i o n w a s a c c o m p l i s h e d according to the first set of models for 7 samples grouped into Opercula, and according to the
-ond set of m o d e l s for 9 samples included in the this giving rise to a slight difference with respect Gastropoda g r o u p (Table 4 ) . T h e unidentified to previously p u b l i s h e d data (Torres et al., fragmenta was included in the Opercula group, 1995a).
D-/4//r>/L-Ile D / L L e u D / L A s p D / L P h e D / L G l u G l o b a l
O p e r c u l a G a s t r o p o d a
7 7 8 ± 6 6 1511 ± 3 1 8
1 1 7 4 ± 149 1 111 ± 2 5 8
9 4 7 ± 18 591 ± 77
9 0 9 ± 7 8 8 0 9 ± 144
1062 ± 138 1012 ± 107
9 7 6 ± 6 5 k a 1009 ± 130 ka
T a b l e 4 : A g e c a l c u l a t i o n s of Venta M i c e n a s a m p l e s ( T o r r e s et al.. 1997).
Taking into a c c o u n t the fact that the results obtained are i n d e p e n d e n t m e a s u r e m e n t s of the same p a r a m e t e r over a time interval, it is possi ble to calculate a global dating for e a c h g r o u p (Table 4 ) ; it m a y be o b s e r v e d that age c a l c u l a tions for the Fragmenta g r o u p s h o w a h i g h e r scattering than those for the Opercula g r o u p . Prior to c o m p u t i n g a " g e n e r a l a v e r a g e d a t i n g " , the former global results from (Table 3), w e r e analyzed for each s a m p l e g r o u p , Opercula and Gastropoda. T h e test of the h o m o g e n e i t y of variances for the t w o g r o u p s s h o w s that both are different, a c c o r d i n g to the L e v e n e tests (p-value 0.002). T h e test for the equality of the means, taking into a c c o u n t the n o n - h o m o g e neity of v a r i a n c e , d e m o n s t r a t e s that the n o n -equability h y p o t h e s i s c a n n o t be rejected (p = 0.65).
To estimate the global mean from the m e a n s of the two sample g r o u p s , Opercula (1001 ka) and Gastropoda (933 ka), a weighted average has been calculated using the inverse of varian
ces (1043 and 4 2 1 2 k a2) as weighting factors.
The result obtained (Torres et al., 1997) was 9 8 3
ka. with a variance of 836 k a2 (983 ± 58 ka for
the 9 5 % confidence interval).
Recently (Torres et al., 2001) have obtained the age calculation algorithms for D/L aspartic acid ratio of Pleistocene bears (Figure 11). For this purpose we have analyzed cave bear sam ples from three different localities:
1. A bone sample from Eiros cave (EE) was
l 4C ( A M S ) dated (Grandal d ' A n g l a d e and Vidal
Romani, 1997) resulting 24.090 ± 4 4 0 a BP.
2. In the U.spelaeus bearing bed from La Lucia (LU) cave s o m e small stalactites appea red, and a thin flowstone 2-3 c m thick sealed the bone bearing bed. T h e two calcite samples were U/Th dated ca. 77 ka.
3. A combination of electron spin resonance (ESR) and U-series dating m e t h o d s on the
remains of the S i m a de los Huesos (BB) bear gave a probable date of 320 ± 4 ka Bischoff et al. 1997).
t = - 6 4 92 + 1191 (DO. Asp)
r= 0 9 9 . p= 0.00
0 . 4 5 :
0 •• • • • • •
-0 5 -0 1 -0 -0 1 5 -0 2 -0 -0 2 5 -0 3 -0 -0 3 5 -0 4 -0 -0
AG1: (Ky)
I
A / r,n., iini^i'it'.iiiii* (ncliinl jHli Sima los Hikso* ' Aiapittrca1
• Sues (iaieJ In "C (Ht>. U 111 (LUIoi U-series and t S k t'BBt
F i g u r e 11: X Y plot of a v e r a g e D / L a s p r a t i o s a n d r a d i o m e t r i c a g e of the t h r e e s a m p l e d b e a r l o c a l i t i e s .
First of all. we want to state that the use of a previous dialysis process allowed a noteworthy accuracy in our analytical results, that was not achieved in all the former analyses carried out on non-dialyzed samples. In the first sets of sam ples we obtained racemization ratios of total (both free and bound) a m i n o acids, that were too erratic to be seriously considered. T h e effect of L - h y d r o x y p r o l i n e p e a k w a s a l s o i m p o r t a n t b e c a u s e in s o m e a n a l y s e s ( t y p i c a l l y those e a r n e d out with an old Chirasil L-Val c o l u m n ) , it overlaps the L-asp peak, and bizarre aspartic acid racemization ratio values were obtained. In our opinion the use of dialysis and a newly pur chased Chirasil L-Val c o l u m n and N P D detector could be the key to success.
T h e next step was to establish a statistical relationship between the mean D/L aspartic acid racemization values and radiometric ages from the three radiometrically dated localities: Eiros (EE- U.spelaeus). La Lucia (LV-U.spelaeus)
-ond set of m o d e l s for 9 samples included in the this giving rise to a slight difference with respect Gastropoda g r o u p (Table 4). T h e unidentified to previously p u b l i s h e d data (Torres et al., fragmenta was included in the Opercula g r o u p , 1995a).
D-Allo/L-Ue D / L Leu D / L A s p D / L P h e D / L G l u G l o b a l
O p e r c u l a 7 7 8 ± 6 6 I 174 ± 149 9 4 7 ± 18 9 0 9 ± 7 8 1062 ± 138 9 7 6 ± 6 5 ka
G a s t r o p o d a 1511 ± 3 1 8 1111 ± 2 5 8 591 ± 77 8 0 9 ± 144 1012 ± 107 1009 ± 130 ka
T a b l e 4 : A g e c a l c u l a t i o n s of Venta M i c e n a s a m p l e s (Torres et al.. 1997).
Taking into a c c o u n t the fact that the results o b t a i n e d are i n d e p e n d e n t m e a s u r e m e n t s of the s a m e p a r a m e t e r over a time interval, it is possi ble to calculate a global d a t i n g for e a c h g r o u p (Table 4 ) : it m a y be o b s e r v e d that age c a l c u l a tions for the Fragmenta g r o u p s h o w a h i g h e r scattering than those for the Opercula g r o u p . Prior to c o m p u t i n g a " g e n e r a l average d a t i n g " , the former global results from (Table 3), were analyzed for each s a m p l e g r o u p . Opercula and Gastropoda. T h e test of the h o m o g e n e i t y of variances for the two g r o u p s s h o w s that both are different, a c c o r d i n g to the L e v e n e tests (p-value 0 . 0 0 2 ) . T h e test for the equality of the means, taking into account the n o n - h o m o g e neity of v a r i a n c e , d e m o n s t r a t e s that the non-equability h y p o t h e s i s c a n n o t be rejected (p = 0.65).
To estimate the global m e a n from the m e a n s of the two sample g r o u p s . Opercula (1001 ka) and Gastropoda (933 ka), a weighted average has been calculated using the inverse of varian
ces (1043 and 4 2 1 2 k a2) as weighting factors.
The result obtained (Torres et al., 1997) was 9 8 3
ka, with a variance of 836 k a2 (983 ± 58 ka for
the 9 5 % confidence interval).
Recently (Torres et al.. 2001) have obtained the age calculation algorithms for D/L aspartic acid ratio of Pleistocene bears (Figure 11). For this purpose we have analyzed cave bear sam ples from three different localities:
1. A bone sample from Eiros cave (EE) was
l 4C ( A M S ) dated (Grandal d ' A n g l a d e and Vidal
Romanf. 1997) resulting 24.090 ± 4 4 0 a BP.
2. In the U.spelaeus bearing bed from La Lucia (LU) cave s o m e small stalactites appea red, and a thin flowstone 2-3 c m thick sealed the bone bearing bed. T h e two calcite samples were U/Th dated ca. 77 ka.
3. A combination of electron spin resonance (ESR) and U-series datina m e t h o d s on the
remains of the S i m a de los Huesos (BB) bear gave a probable date of 320 ± 4 ka Bischoff et al.. 1997).
t = - 6 4 9 2 * 1191 (D/L Asp)
1= 0 9 9 . p= 0 0 0
1 5 0 2 0 0 2 5 0 3 0 0 3 5 0 4 0 0
A G H ( K y )
A I. r™, ii/m-Mc.miu I actual t EE E l m ,
LU U L u c t a
HH Sinw de Ins I [itesos i Alapiicica i
:sdak\Jlrt " C ( t h > . U Ih fl.T JI ur I
F i g u r e 11: X Y plot of a v e r a g e D / L a s p ratios and r a d i o m e t r i c a g e of the t h r e e s a m p l e d b e a r l o c a l i t i e s .
First of all, we want to state that the use of a previous dialysis process allowed a noteworthy accuracy in our analytical results, that was not achieved in all the former analyses carried out on non-dialyzed samples. In the first sets of sam ples w e obtained racemization ratios of total (both free and bound) a m i n o acids, that were too erratic to be seriously considered. T h e effect of L - h y d r o x y p r o l i n e peak w a s also i m p o r t a n t b e c a u s e in s o m e a n a l y s e s ( t y p i c a l l y those carried out with an old Chirasil L-Val c o l u m n ) , it overlaps the L-asp peak, and bizarre aspartic acid racemization ratio values were obtained. In our opinion the use of dialysis and a newly pur chased Chirasil L-Val c o l u m n and N P D detector could be the key to success.
T h e next step was to establish a statistical relationship b e t w e e n the mean D/L aspartic acid racemization values and radiometric ages from the three radiometrically dated localities: Eiros (EE- U.spelaeus), La Lucia (LU-U.spelaeus)
-y = 0 . 0 0 8 4 X + 1 . 1 1 3 7 R2= 0 . 7 8 1 5
5 0 0 1 0 0 0 1 5 0 0 2 0 0 0 t ( h )
glu 105°C
2 . 0
y = 0 . 0 0 0 9 x + 0 . 0 5 8 4 R2= 0 . 9 4 1 4
5 0 0 1 0 0 0 t ( h )
1 5 0 0 2 0 0 0
y = - 8 9 9 9 . 5 x + 1 8 . 9 9 2 R2= 0 . 9 9 8 2
Ea (glu)
0
c
y = - 1 4 3 6 4 x + 3 0 . 9 6 2 R2= 0 . 9 9 9 6
-5
- 1 0
. 1 5
-0 . -0 -0 2 6 -0 . -0 -0 2 7 -0 . -0 -0 2 8 -0 . -0 -0 2 9 -0 . -0 -0 3 -0
1/T
„ . I + D/L
F i g u r e 12: X Y plot of 2k. I - C + 1 n — — for aspartic and c l u t a m i c acids from the 6 5 ° C . 85 °C and 105 ° C kinetic
1 - D/L
e x p e r i m e n t s c a r r i e d o u t on a c t u a l Ursus americcmus P a l l a s d e n t i n e . X Y plot of In k, a g a i n s t 1 / T (K 1) for t h e e s t i m a t i o n of the
a c t i v a t i o n e n e r g i e s for a s p a r t i c and g l u t a m i c a c i d s (Torres et al.. 2 0 0 1 ) .
-and Sima de los Huesos (BB-U.cleningeri). A simple linear correlation model was used. T h e resulting equation is:
t = -64.92+1 191 (D/L Asp)
where t is time (in years) and D/L the mean aspartic acid racemization value. T h e correlation coefficient value is extremely high (r = 0.998) and significant (p = 0.002).
We have decided to adopt the m o r e conve nient easier to handle linear model because, according to many authors (Masters and Bada. 1977: Mitterer and Kriasakul, 1989), for lower racemization ratios, c o m p r i s e d between 0.00 and 0.40 (as in our case) a linear racemization behaviour can be assumed. This linear D/L A s p ratio against radiometric age relationship cannot be considered an exception since a similar beha viour has been published for a m i n o acids race mization in wool textiles ( C s a p o et al., 1998).
T h e radiometric ages and mean aspartic acid racemization values appear in a XY plot. Figure 11; the origin ordinate value (0.05) is the indu ced aspartic acid racemization ratio measured in modern U. americanus dentine collagen (Torres et al., 1999). which supports the correctness of the obtained correlation algorithm. We can con clude that there is g o o d consistency between all of t h e m . T h e age of U. spelaeus from La Lucia cave, d e d u c e d from U/Th dating of bear-bearing bed related s p e l e o t h e m s . was 77 ka. According to the average aspartic acid racemization ratio, it could be slightly older, a p p r o a c h i n g 9 0 ka.
A linear aspartic acid racemization trend was found w h e n average racemization ratio was regressed against radiometric ages of each loca lity. This m e a n s that the obtained mathematical equation can be confidently used for numeric age calculation. However, it is necessary to con sider that local taphonomical conditions could play a very important role either in intersample variation or in intrasample racemization ratio variation.
Kinetics
S o m e experiments of dentine a m i n o acids kinetics were m a d e to determine their activation energy and to c o m p a r e this results from the obtained from geological data. In our opinion it was necessary to verify the supposed o n e - w a y sense of the racemization process because s o m e
a u t h o r s ( K i m b e r et al.. 1986; K i m b e r and Griffin, 1987) described an "apparent kinetic reversal" in the > 1000 Dalton peptide fraction in bivalve shells (Ostrea sp.). The "racemization kinetics" must be taken, at least, as the result of two different processes: one reversible (racemi zation) and another irreversible (hydrolysis) affecting proteins and peptides. T h e existence of a peptide hydrolysis resistance in the peptide bonds w h e r e h y d r o p h o b i c a m i n o acids are lin ked to aspartic acid exists has been reported, and since most of the racemization takes place at the terminal position of the peptide chains, an appa rent racemization reversal can be produced. In our kinetic e x p e r i m e n t s . Figure 12. this effect was evidenced by taking into account that the
obtained kL value for an adjusted m o d e l :
2A
^
= c + l nr 7 ^ :
corresponds to a first order reversible kinetics
model ( F O K ) where kD/ k , = l and D/L is the
racemization ratio. T h e value obtained was k, = 0.0042 h~' for aspartic acid in samples from the first 576 h of experiment (excluding the " k i n e tics reversal" interval). In the experiment denti ne p o w d e r was mixed with quartz sand and water soaked. U n d e r nitrogen a t m o s p h e r e was heated in a controlled stove at 105 °C .
R a c e m i z a t i o n rates in other a m i n o acids (e.g., glutamic acid) are lower than in aspartic
acid. T h e calculated k( for glutamic acid it is
kL= 0.00045 h 1 for the 1704 h experiment.
Two additional kinetic e x p e r i m e n t s w e r e carried out heating the dentine samples at 65 °C and 85 °C for 1272 h. in test tubes with quartz sand in the oven (full moisture conditions): the apparent rate constant k, values are summarized
in Table 5. F r o m the plot of In kL versus T"1 ( K1)
following the Arrhenius relationships we have d e d u c e d that the activation energy for aspartic
acid is 17.88 kcal/mol ( R: = 0.9982) and for glu
tamic acid is 28.54 kcal/mol ( R2 = 0.996). No
"kinetics reversal" has been observed for the glutamic acid at any heating temperature, nor for the aspartic acid at 65 °C and 85 °C. In any case, the best method for a m i n o acid racemization dating is to analyze a set of different a m i n o acids making it possible, through similarity analysis, to determine apparent kinetics reversal and to correct it.
T h e kinetic experiment has also shown the influence of moisture, also observed in bones
-T e m p e r a t u r e ("C) T i m e ( h o u r s ) kL A s p ( h1) k, G l u ( h1)
6 5 1 2 7 2 2.5 • 1 0 -4 5.0 • 1 0 -6
8 5 1 2 7 2 1.0 • 1 0 -3 5.0 • 1 0 -5
105 1 7 0 4 - 4 . 5 • 1 0 -4
105 5 7 6 4.2 • 1 0 -3
-T a b l e 5 : A p p a r e n t rate k. c o n s t a n t s of a s p a r t i c acid a n d g l u t a m i c acid from b e a r d e n t i n e kinetic e x p e r i m e n t s (-Torres et al.. 2 0 0 1 ) .
(Hare, 1980) in the racemization rate of dentine a m i n o acids. Only taking into account the sam
ples that were heated up to 100 h, kL values
obtained range from k, = 0.0008 l r1 for the dry
samples to kL = 0.0170 h ' - 20 times greater- for
the samples heated in the stove (full moisture conditions). T h e value obtained for the heating block samples (moisture saturated atmosphere)
is an intermediate one: k( = 0.0070 h ', due to
the formation of a condensation water ring in the upper part of the test tube.
C U R R E N T B E A C H
C H L A M Y S Wm A N O M A
Protein and amino acid preservation
O n e of the most exciting recent possibilities of a m i n o acid analysis in fossils is to h e l p in the search of fossil D N A . T h e search of fossil D N A is a painful and expensive process that someti mes ends with the determination of recent D N A . in most cases of h u m a n contamination origin. T h e previous determination of a m i n o acids con tent, better if the presence of intact collagen molecules can be assured, can avoid to start wor king with injectable material.
Protein and a m i n o acid decay is a process which starts very soon after the organism death. We also have studied a m i n o acid preservation in marine shells (Torres et al.. 1995b). In this case we analyzed free and b o n d e d a m i n o acids toge ther. S a m p l e s c a m e from the Gulf of C a d i z at the S W coast of Spain. A - s a m p l e s from present day beach; B - samples from a spit bar 14-C dated 2 2 3 5 - 2 1 7 5 B P and C-samples from an offshore vibrocore dated ca. 10000 BP.
In Fig. 13 w e have represented the D + L aspartic acid p e a k s areas (:1000). T h e 3 5 0 (: 1000) area corresponds to a standard of 0.0025 m g of total aspartic acid content in 80 m g of total weight sample. In A - s a m p l e s there are large a m o u n t s of aspartic acid and no differences bet ween Chlamys sp. and Anomia sp. analyses has been found. T h e spit bar samples (B) show an important decrease in aspartic acid content, mostly in Glycymeris sp. and Cardium, sp. fewer
H O L O C E N E SPIT BAR
o a n 4ffl eoc
PANOPAEA
• M v C Y M E P I S C A R O U M
O F F S H O R E V I B R O C O R E " G O L C A 1 "
E ' : H I U O O E R M A T A G A S T R O P O D A \ | E S C A F O P O D A
P E I E C I P O C ' A
F i g u r e 1 3 : D + L a s p a r t i c a c i d c h r o m a t o g r a m p e a k a r e a s (: I ()()()) of m a r i n e faunal r e m a i n s from t h e G u l f of C a d i / , a r e a : c u r r e n t b e a c h , H o l o c e n e spit b a r ( 1 4 - C d a t e d in 2 2 3 5 - 2 1 7 5 B P ) a n d v i b r o c o r e s a m p l e s d a t e d ca. 1 4 . 0 0 0 y e a r o l d . T h e vertical t i p i n d i c a t e s an a s p a r t i c acid c o n c e n t r a t i o n of 0 . 0 0 2 5 m g in a s t a n d a r d 8 0 m g s a m p l e ( T o r r e s et al., 1 9 9 6 ) .
-in Panopaea sp. rema-ins. S a m p l e s from vibro-core (C), mostly gastropoda and e c h i n o d e r m a (sea urchins), have only a little amount of a m i n o acids w h e r e a s the highest values correspond to
Venus sp. shells.
On m a m m a l , fossil bear, dentine there is also a progressive collagen loss because of the colla gen" triple helix deterioration and the molecule break down. Notwithstanding the peculiar favo rable environmental conditions in caves (Fig.
14): stable thermal history allowed a surprisingly
through-time good preservation of collagen. We worked on bear teeth dentine amino acid analysis with a previous 3500 Dalton dialysis in the way proposed by Marzin (1990) and Torres et al. (1999b). After the chromatographic analysis, because of the high accuracy of the volume injec
tion reached with the H P 6 8 5 0 autosampler, it has been possible to calculate in a quantitative way the total amount of bonded a m i n o acids included in collagen molecules, since free a m i n o acids were eliminated after dialysis.
C U E V A M A Y O R
(Atapuerca, Burgos, Castilla-Leon)
C U E V A C U E T O D E L A L U C I A (Quintanilla, C a n t a b r i a )
C U E V A E I R O S ^ _ (Triacastela, Lugo, Galicia)
C U E V A D E A M U T X A T E (Aralar. N a v a r r a )
C U E V A L A P A S A D A (Guriezo, Cantabria)
C U E V A Sa I S A B E L
( R a n e r o , V i z c a y a , Pais Vasco)
C U E V A D E L E Z E T X I K I
( M o n d r a g d n , G u i p u z c o a , Pais Vasco)
C U E V A D E A R R I K R U T Z (Oflati, G u i p u z c o a , Pais Vasco)
C U E V A D E E K A I N ( D e b a , G u i p u z c o a , P a i s V a s c o )
C U E V A D E T R O S K A E T A (Ataun G u i p u z c o a , Pais Vasco)
C U E V A C O R O T R A C I T O (Telia, H u e s c a , A r a g o n )
C U E V A D E E L T O L L (Moia Barcelona, CataluPla)
C U E V A D E E L R E G U E R I L L O (Patones, Madrid)
Figure 14: G e o g r a p h i c a l s i t u a t i o n of s a m p l e d c a v e s .
Results appear in Fig. 15, where it is possible to observe a net differentiation b e t w e e n the oldest samples (Middle Pleistocene ca. 300 ka) from Ursus deningeri von Reichenau and the youngest ones of Ursus spelaeus R o s e n m u l l e r -Heinroth (Upper Pleistocene) and Ursus arctos Linneo (Holocene). This differentiation was pro duced because there is a dramatic decrease in the total amount of b o n d e d a m i n o acids ( A S P ) with older geological age. In fact other samples from Lower Pleistocene age bear (Ursus etruscus G. Cuvier) recovered from an open air site (Venta Micena, Granada) did not contained any b o n d e d amino acid whereas samples of gastropoda and ostracoda of the s a m e locality still contained
a m i n o acids a m o u n t s , large e n o u g h to be analy zed. This can be interpreted not only in terms of ageing of a no-all mineralized molecule but also because of paleoenvironmental influence becau se Venta M i c e n a fossil bearing consisted of cal-citic-dolomitic m u d s t o n e s deposited in a sha llow saline lacustrine environment under extre me hydrological stresses.
