An inquiry into the selective protection of glycine during the radiolysis of glycine-alanine mixtures in aqueous solutions and its implications to the preservation of optically active

Page 1

Journal of Biological Physics 22: 87-100, 1996. @ 1996 Kluwer Academic Publishers. Printed in the Netherlands.

An Inquiry into the Selective Protection of Glycine during the Radiolysis of Glycine-Alanine Mixtures in Aqueous Solutions and its Implications to the Preservation of Optically Active Amino Acids in the Early Earth

RAFAEL NAVARRO-GONZkLEZ’j*, MITSUHIKO AKABOSH12, ALFRED0 ROMERO’ and CYRIL PONNAMPERUMAt3 ‘Institute de Ciencias Nucleares, Universidad National Autonoma de Mexico, Circuit0 Exterior, CiuaW Universitaria, Apartado Postal 70-543, Mexico D.F. 04510, Mexico; ’ Research Reactor Institute, Kyoto University, Kumatori, Osaka-590, Japan; ‘Laboratory of Chemical Evolution, Department of Chemistry and Biochemistry, University of Maryland, College Park, MD 20742, U.S.A. (*author for correspondence)

(Received: 12 October 1995; accepted in final form: 22 March 1996)

Abstract. Akaboshi et al. (1990) has found an unexpected protection of the achiral amino acid, glycine, towards ionizing radiation at the expense of the selective destruction of the chiral amino acids, alanine and aspartic acid. The present work examines the mechanism of this protection for the case of alanine. We have developed a computer model for the radiolysis of glycine, alanine and glycine-alanine mixtures in aqueous solution. It is established that this protection is due in part to the reaction of the o-radical of glycine with alanine to regenerate a more stable a-radical, according to the following reaction,

CH(NH,+)CO; + CH3CH(NH;)CO; + CH,(NH,+)CO; + CH&(NH;)CO;

The rate constant of this reaction was estimated to be 6 104MM-’ s-r. The implications for this selective protection of glycine are. considered for a hypothetical case in which there would be an enrichment of about 10% of L&mine in the primitive ocean and taking the glycine/alanine ratios obtained in CHs- and COZ- dominated atmospheres using electric discharge experiments. It is predicted that alanine would be rapidly destroyed and radioracemized in spite of the fact that the concentration of alanine is equal or significantly lower than that of glycine. Assuming that chiral amino acids were a prerequisite for the origin of life, it can be deduced that life could have appeared in a relatively short period of time unless there was a constant supply of optical amino acids from extraterrestrial sources.

Key words: Alanine, Amino acids, Chirality, Glycine, Ionizing radiation, Kinetics, Primitive hydrosphere, Racemization, Radiolysis, Radioracemization

1. Introduction

The effect of ionizing radiation on aqueous solutions of amino acids was extensively studied in the 1950s and 1960s because of a dual interest. On the one hand, these investigations were significant to acquire a basic understanding of the mechanism of

Page 2

88 RAFAEL NAVARRO-GONZALEZ ET AL.

radiolysis of water. On the other hand, since the amino acids are the building blocks of proteins, it was hoped that these investigations could lead to generalizations which enable one to extrapolate to the effects of ionizing radiation on larger molecules, e.g., proteins and enzymes, and even more complex systems, such as living cells. From these studies it was possible to elucidate the mechanism of formation of major products resulting from the radiolysis of amino acids in aqueous solutions. The principal radiolytic reactions of simple amino acids are reductive deamination and decarboxylation. Reductive deamination leads to the formation of NH3 and a carboxylic acid. Decarboxylation results in the formation of an amine with one less carbon atom than the original amino acid (for reviews see Urbain, 1977; Greenstein and Winitz, 1986; Spinks and Woods, 1990). The radiolysis of amino acids does not only lead to degradation products but also can bring about the synthesis of more complex amino acids than the starting one. For instance, the radiolysis of aqueous solutions of glycine leads to the formation, in low yield, of aspartic acid, hydroxyaspartic acid, iminodiacetic acid, serine, threonine, alanine, a-aminobutyric acid, diaminosuccinic acid and p-alanine (DraganiC et al., 1985).

Knowledge of the radiolytic behavior of amino acids is important in the context of chemical evolution since ionizing radiation might have potentially induced chem- ical transformations of this type of organic compounds either on the early Earth or in extraterrestrial environments. Vester, Ulbricht and coworkers have speculated that ionizing P-radiation could generate optically active organic molecules based on the asymmetry of P-decay (Vester et al., 1959; Ulbricht, 1959; Ulbricht and Vester, 1962). Several experiments have been devised to confirm experimental- ly the Vester-Ulbricht hypothesis by irradiating optically active amino acids (both solid and dissolved in water) with 9oSr,90Sr-9uY,90Y or 14C. Akaboshi et al. (1979) have found that the P-irradiation (from 9oY) of D- or L-alanine crystals leads to higher radical yields in D-alanine at low dose; however, analysis of the remaining amino acids in similar experiments have shown no selective destruction of either enantiomer at higher doses (Bonner, 1974; Bonner and Flores, 1975; Bonner et al., 1978). These negative results might be due, among other reasons, to the fact that the majority of the Bremsstrahlen intensity from these radionuclides is found at the low energy of the spectrum where the circular polarization of the photons is low (Bonner and Lemmon, 1978). To avoid this problem, Bonner and cowork- ers (1975, 1976, 1976/1977) have employed ‘artificial’ longitudinally polarized, monoenergetic electrons from a specially designed linear accelerator. This system is advantageous over natural ,&decay electrons in that either ‘natural’ antiparallel- spin (‘left-handed’) electrons or ‘unnatural’ parallel-spin (‘right-handed’) electrons can be generated at will, thus permitting the crucial test for the predicted reversal of any asymmetric effect by electrons of the opposite chirality. In these studies it was found that irradiation with left-handed electrons preferentially destroyed D-leucine in a crystalline D, L-leucine target while right-handed electrons prefer- entially destroyed the L-enantiomer. The magnitude of the enantiomeric excesses

Page 3

RADIOLYSIS OF GLYCINE-ALANINE MIXTURES IN AQUEOUS SOLUTIONS 89

produced in these experiments was small but reproducible, i.e., 0.6 - 1.4%. Nev- ertheless, Hodge et al. (1979) were unable to duplicate these results.

A phenomenon which may make questionable the inevitable efficacy of ionizing P-radiation has been reported by Bonner and Lemmon (1978). These authors found that ionizing radiation leads to racemization of optically active amino acids (both solid and dissolved in water) along with their well-known radiolysis. Aqueous solutions of the sodium salts of amino acids which underwent 53 - 66% radiolysis typically showed 5 - 11% racemization. The corresponding hydrochloride salts in aqueous solution, however, underwent little or no racemization. Radioracemization of the aqueous sodium salts of amino acids can be rationalized as resulting from the stabilization of the a-radical as the highly symmetrical resonance hybrid (I). Such a resonance-stabilized intermediate cannot arise from the corresponding a-radical from the cation (II) (Bonner and Lemmon, 1978).

I

0 . II

CH,&C-OH

l(lHzf

II Another potential problem for the efficiency of ionizing P-radiation for the origin

of optically active amino acids as well as for the preservation of these compounds formed by other abiotic mechanisms in the early Earth can be inferred from the work of Akaboshi et al. (1990), who found that radiolysis of aqueous glycine- alanine or glycine-aspartic acid solutions results in an unexpected protection of the achiral amino acid, glycine, towards ionizing radiation at the expense of the selective destruction of the chiral amino acids, alanine and aspartic acid. Since glycine is the most abundant amino acid formed in prebiotic experiments (DraganiC et al., 1977; Ferris et al., 1978; Schlesinger and Miller, 1983), it can be assumed that any enantiomeric excess of chiral amino acids could have been wiped out by ionizing radiations produced by the decay of radionuclides on the early Earth. The

Page 4

90 RAFAEL NAVARRO-GONZALEZ ET AL.

mechanism of this selective protection of glycine is not known. Akaboshi et al., (1990) suggested that this effect could be due to the presence of hydrophilic and hydrophobic bonding forces between the amino acids in aqueous solution. It could also be due to other effects, such as dissimilar reactivities among the amino acids, and/or different stabilities of the a-radical. The purpose of this work is to get an insight into the mechanism of this protection. A preliminary report has already been published elsewhere (Navarro-Gonzalez et al., 1994; Romero, 1995).

2. Computational Methodology

2.1. MECHANISM OF RADIOLYSIS OF WATER

The radiolysis of liquid water has been extensively studied (DraganiC and Draganic, 1971). The interaction of ionizing radiation with water produces ionization and excitation of the molecules which form highly reactive intermediates, such as hydrogen atom (H.), hydroxyl radicals (*OH) and hydrated electrons (eq), and stable molecules, such as molecular hydrogen and hydrogen peroxide (reaction 1):

H20-tH.+e&+.OH+H++Hz02+Hz (1)

We have previously developed a kinetic model that appropriately simulates the radiolysis of water (Navarro-GonzBlez et al., 1992). This model consists of a total of 48 equations that involved the reactions of hydrogen atom, hydroxyl radical and hydrated electron among themselves and with other species, such as water, molecular hydrogen, hydrogen peroxide and oxygen. We use this model in the present work to study the radiolysis of glycine-alanine mixed solutions.

For the case of dilute aqueous solutions of amino acids, we used the initial radiation chemical yields of the primary species for the radiolysis of water. For concentrated solutions, the yield of decomposition of water increases as a function of the reactivity of the amino acid towards hydroxyl radicals (DraganiC and Dra- ganic, 1973). In such cases the yields of the primary species were calculated from the work of DraganiC and DraganiC (1973).

2.2. RADIOLYSIS OF AMINO ACIDS

The radiolytic behavior of simple amino acids has been generalized from studies on the radiolysis of glycine and alanine. We have used the experimental data of Maxwell et al. (1954), Bonner and Lemmon (1978) and DraganiC et al. (1985) for the radiolysis of glycine, and Sharples et al. (1955) for the radiolysis of alanine, to construct a kinetic model for the radiolysis of aqueous solutions of such amino acids. In some of these studies, the irradiations were carried out in high concentra- tions (1 M). Consequently, direct action of ionizing radiations on the amino acids took placed in addition to the indirect action via the radiolysis of water. We have included in our kinetic model the direct action based on the work of Meshitsuka

Page 5

RADIOLYSIS OF GLYCINE-ALANINE MIXTURES IN AQUEOUS SOLUTIONS 91

et al. (1964) for the radiolysis of glycine crystals. It is estimated, based on the electron fraction of the water-amino acid system, that about 6.2% and 7.5% of the energy of ionizing radiations is deposited directly into glycine and alanine, respectively in 1 M solutions (Romero, 1995). At lower concentrations (< 0.1 M), the direct effect is negligible. These yields were derived by computer simulations using the experimental data of Maxwell et al. (1954) and Sharples et al. (1955).

2.3. COMPUTER PROGRAM

The chemical reactions from our kinetic model were translated into differential equations and solved numerically by Acuchem (Braun et al., 1988). This computer program solves a system of differential equations describing the temporal behavior of spatially homogeneous, isothermal, multicomponent chemical reactions. The version (1.4) used in this study can handle up to 200 reactions and 100 chemical species.

3. Experimental Procedure

3.1. CHEMICALS

All chemicals were of the highest purity commercially available and were used without further purification. Glycine and deuterated alanines were obtained from Sigma Chemical Co. Freshly prepared deionized-distilled water was used for prepa- ration and irradiation of samples. It was purified by flowing tap water through a Millipore system consisting of activated charcoal and anion<ation exchange cylin- ders, and finally distilled into a glass container.

3.2. SAMPLE PREPARATION AND IRRADIATION

Aqueous mixed solutions of 0.5 mM glycine and 0.5 mM deuterated alanine were prepared and sealed aerobically in polyethylene screw-cap bottles. The following systems werestudied: CH2(NHi)CO,/CHsCD(NHt)CO, andCH2(NHz)CO;/ CDsCD(NHt )CO, . The solutions were irradiated in presence of a small amount of dissolved oxygen (0.25 n&I), which is decomposed in the early stages of irradi- ation.

Irradiations were carried out at room temperature in a 6oCo unit located at the Department of Chemical and Nuclear Engineering of the University of Maryland with a dose rate of 23.8 kGy hr-’ . The irradiation dose varied from 0 to 72 kGy.

3.3. ANALYSIS

After irradiation the samples were stored in a freezer. Before analysis the irradiated solutions and blanks were freeze-dried and the residue containing the remaining fraction of amino acids was analyzed by mass spectrometry (VG Analytical, LTD

Page 6

92 RAFAEL NAVARRO-GONZALEZ ET AL.

model 7070E) using electron impact and chemical ionization techniques at 70 eV. We used the molecular ion of each amino acid to examine its degree of radiation destruction. In order to study the formation of deuterated glycine from the radiolysis of glycine-deuterated alanines, we used high-resolution mass spectrometry.

4. Results and Discussion

4.1. RADIOLYSIS OF GLYCINE AND ALANINE SOLUTIONS

We have constructed a model for the radiation-induced decomposition of aqueous solutions of glycine and alanine at dilute and concentrated solutions, taking into account that the principal radiolytic products are ammonia, amines, carboxylic acids, keto acids, aldehydes, carbon dioxide and molecular hydrogen, The complete set of equations used in our model may be found in Romero (1995). Table I summarizes the main reactions needed to model the indirect action of ionizing radiation on the amino acids. In some instances, the rates of these equations are known, such as for the reaction of hydrogen atom (Equation 3), hydroxyl radical (Equation 4) and superoxide radical (Equation 5) with glycine and alanine (Bielski et al., 1985; Buxton et al., 1988). For the case of hydrated electron (Equation 6), there are three possible reaction channels; however, only the overall rate is known (Buxton et al., 1988). Based on computer simulations using the experimental data of Maxwell et al. (1954) for glycine and Sharples et al. (1955) for alanine, we have derived the relative proportions for each channel. These values are reported in Table I (Equation 6). In the case of Equations from 7 to 18, their rates of reaction were not available in the literature. In such cases, we first estimated the magnitude of their rate constants by comparison with other known reactions. Finally these figures were slightly adjusted to adequately fit the experimental behavior for the radiolysis of glycine and alanine.

In addition to the 18 equations described in Table I, it was necessary to include 77 additional equations that take into account the reactivity of the principal products (CO, CO;?, NHs, CHJ, RCHzNHz, RCHO, RCHzC02H and RCOC02H) with the primary species of water radiolysis. This was necessary since glycine and alanine are less reactive than their products towards the primary species of water radiolysis, and therefore they take part in the mechanism of radiolysis of amino acids at later stages. Table II summarizes some of these reactions. The complete set of equations is given in Romero (1995).

Figures 1 and 2 show the formation of major products from the radiolysis of glycine and alanine at 1 M solution, respectively. The symbols are the experimental data reported by Maxwell et al. (1954) and Sharples et al. (1955). The lines are computed by our kinetic model. As can be seen, there is a reasonable fit for ammonia, amines, carboxylic acids, hydrogen and carbon dioxide. The fit is not as good, particularly at larger doses, for keto acids and aldehydes. The reason for this discrepancy is due to the fact that these products are very reactive towards the primary species of water radiolysis and they decompose by a complex radiolytic

Page 7

Table

I.

Equa

tions

co

nsid

ered

in

the

radi

olys

is of

am

ino

acid

s

No.

Equa

tion

Cons

tant

a G

ly (

R =

H)

Ala

(R =

CH

3)

1 2 3a

3b

RCH(

NH;)C

OzH

=

RCH(

NH$)

CO;

f H+

K

= 4.

46

x lO

-3

K =

4.57

x

lo-3

RCH(

NHz)

CO;

& RC

H(NH

s)CO

; +

H+

K =

1.65

x 1

0-r”

K =

2.04

x

10-l”

RC

H(NH

;)CO

; +

H- +

H2

+

CR(N

H;)C

O;

CHsC

H(NH

;)CO

; +

H.

+ Hz

+

CH

2CH

(NH

,f)C

O;

k =

7.7

x lo4

k

= 2.

2 x

lo5

k =

1.0

x lo4

4a

RCH(

NHz)

CO;

+ .O

H -+

H2

0 +

-CR(

NH;)C

O;

4b

CHsC

H(NH

t)CO

; +

.OH

-+ H

z0

+ CH

2CH(

NH,+

)CO

; 5

RCH(

NH;)C

O;

+ 0;

-+

CR

(NHT

)CO

; +

HO;

k =

1.7

x IO

’

k =

4.2

x 10

-l

k =

7.3

x 10

’ k

= 4.

0 x

lo6

k =

6.0

x 1O

-2

6a

RCH(

NH$)

CO;

+ eg

+

NH1

+ CH

(R)C

O;

k =

2.5

x lo6

k

= 5.

8 x

lo6

6b

RCH(

NHz)

CO;

+ e;

--t

H .

f RC

H(NH

z)CO

; k

= 4.

7 x

IO6

k =

1.9

x 10

’

6c

RCH(

NHz)

CO;

+ ei

%

RCH(

NHz)

CO.

+ 20

H-

k =

1.6

x lo6

k

= 1.

3 x

lo6

7 CR

(NH;

)CO

; +

CH(R

)CO

; +

RC(=

NH

,+)C

O;

+ RC

HKO

; k

= 8.

3 x

10’

k =

7.0

x lo7

8

-CR(

NH$)

CO;

-I- C

R(NH

;)CO

; -+

RC

(=

NH;)C

O;

$ RC

H(NH

;)CO

; k

= 9.

4 x

lo7

k =

8.0

x 10

’ 9

RC(=

NH

,+)C

O;

+ Hz

0 -+

NH?

+

RC(=

0)

CO;

k =

2.8

x lo@

k

= 6.

8 x

1O-6

10

RC(=

0)

CO;

+ H2

0 -+

CO

* +

RCHO

+

OH-

k

= 1.

9 x

1O-6

11

CH

(NH$

)CO

; +

Hz02

-+

CH

(=

NH,+

)CO

; +

Hz0

+ .O

H k

= 1.

0 x

lo4

12

CH(N

H$)C

O;

+ CH

2CO

; +

Aspa

rtic

acid

k

= 1.

7 x

10’

k =

4.6

x IO

-’

13

CH(N

H$)C

O;

+ CH

(NHz

)CO

; +

Diam

inos

uccin

ic ac

id

k =

6.0

x lo6

14

RC

H(NH

;)CO

. -+

CO

+

CH(R

)NH,

+ k

= 1.

0 x

10”

15

CH(R

)NH,

+ +

CH(R

)NH,

+ +

RCH2

NH,+

+

RCH

= NH

,+

k =

1.0

x lo9

k

= 1.

0 x

10”

k =

1.0

x 10

’

16

RCH

= NH

,+ +

H20

-+

RCH

O

+ NH

,+

17

2. C

HzCH

(NH;

)CO

; -+

CH2

=

C(NH

,+)C

O;

+ CH

sCH(

NH$)

CO;

18

CH2

= C(

NH:)C

O;

+ H2

0 +

NH

; +

CH3C

OCO

;

k =

6.5

x lo@

k

= 6.

5 x

lO-6

k

= 5.

0 x

10’

k =

7.0

x lO

-‘j

“Equ

ilibriu

m

cons

tant

s ar

e di

stin

guish

ed

by t

he s

ymbo

l K,

an

d we

re t

aken

fro

m D

ean

(198

5).

Rate

con

stan

ts (

k)

have

uni

ts o

f M

-‘s-l:

th

e va

lues

fro

m

Equa

tions

3

to 6

wer

e ta

ken

from

Bie

lski

et a

l. (1

985)

an

d Bu

xton

et

al.

(198

8).

In t

he c

ase

of E

quat

ions

7-

18,

thei

r co

nsta

nts

were

est

imat

ed

by c

ompa

rison

wi

th

othe

r kn

own

reac

tions

.

Page 8

94 RAFAEL NAVARRO-GONZALEZ ET AL.

Table II. Selected secondary equations considered in the radiolysis of amino acids

No. Equation Constant=

1 CO2 + Hz0 + HCO; + H+ K = 3.47 x lo-’ 2 HCO; + OH- *CO;-+H20 K =5.01 x 10-l’ 3 CO2 + .OH --t CO;+ H+ k = 1.0 x 106

4 CO2 + e&j + co; k = 7.7 x 10’ 5 HCO; + H. + CO;+ H20 k = 4.4 x IO4 6 HCO, + .OH --t CO,+ H20 k = 8.5 x lo6

7 NH: + e&, -+ H.+NHj k = 2.0 x lo6 8 2.NH2 --f N2H4 k = 2.2 x 10’ 9 .NH2 + .OH + NH20H k = 9.5 x 10’

aRate constants have units of M- ’ s- ’ , and were taken from Buxton et al. (1988) and references therein.

mechanism which was not possible to incorporate into our model due to the lack of available kinetic data.

4.2. RADIOLYSIS OF GLYCINE-ALANINE MIXTURES IN SOLUTIONS

Our experimental data confirmed the earlier observation of Akaboshi et al. (1990) of the selective protection of glycine in the radiolysis of glycine-alanine mixed solutions. Such protection of glycine could result from the reaction of the free- radical ’ CH(NHt)CO; with alanine to regenerate glycine and produce a radical of alanine according to reaction 2.

-CH(NH;)CO, + CHsCH(NH;)CO, +

CH:!(NHz)CO; + CH&NHz)CO, and/or CH&H(NHt)CO, (2)

This reaction was studied using two different types of deuterated alanines, CH$D(NH$)CO, and CD$.ZD(NHi)CO,, to establish if the alpha, the beta or both hydrogens were abstracted by the radical CH(NHz)CO, . The mass spec- tra data indicated that a negligible amount of deuterated glycine was formed using either deuterated alanines according to reaction 2. To get a better insight into the mechanism of this protection, we extended our computer model of the radiolysis of individual amino acids, glycine and alanine, to include the combined case.

To model the radiolysis of glycine-alanine mixed solutions, we would need to combine the reactions for each system; however, the number of reactions and chemical species exceed the maximum limit of the program. Therefore, we had to exclude as many reactions as possible to be able to model the radiolysis of mixed solutions of glycine and alanine. Figure 3 compares the experimental data of Akaboshi et al. (1990) (top-left panel) with three different computer simulations.

Page 9

RADIOLYSIS OF GLYCINE-ALANINE MIXTURES IN AQUEOUS SOLUTIONS 95

0.6

0.0

0.6

Dose kGY) Figure 1. Formation of principal products in the radiolysis of 1 M solutions of glycine. Symbols are the experimental data reported by Maxwell et al. (1954). Lines were derived by computer simulations.

As a first approximation we selected only the reactions of the radicals that origi- nate from glycine and alanine (Equations 1 through 18 in Table I). The results of this simulation are shown on the top-right panel in Figure 3. It can be seen that there is no special protection of glycine in the radiolysis of the mixed case. Fur- thermore, the destruction of the amino acids occurs at lower doses than observed experimentally. As a next approximation we included in our computer mode1 the secondary reactions of hydrogen atom, hydrated electron and hydroxyl radicals with two of the principal products of amino acids: carbon dioxide and ammonia (Table 11). The results of this simulation are shown in the left-bottom panel on Figure 3. In this case, the trends for the decomposition of the amino acids are more similar to the experimental results, except for glycine in the mixed solution, where it is not greatly protected. These results indicate that the products from the radiolysis of amino acids are much more reactive towards the primary species of water radiolysis than the amino acids themsekes. Finally, we have incorporated

Page 10

96 RAFAEL NAVARRO-GONZALEZ ET AL.

6.0

0.6

0.0

2.0

1 .o

0.0 0 7 14 0 7 14

Dose (kGy)

Figure 2. Formation of principal products in the radiolysis of 1 M solutions of alanine. Symbols are the experimental data reported by Sharples et al. (1954). Lines were derived by computer simulations.

the regeneration of glycine from the reaction (2) of the radical CH(NHz)CO, with alanine. The results of this simulation are shown in the right-bottom panel on Figure 3. In this case, glycine is protected but as a consequence it leads to an enhanced decomposition of alanine, which is not experimentally observed. From computer simulations, we have been able to derive an upper limit for reaction 2. The second-order rate constant was estimated to be 6 lo4 M-‘s-l . This value is low for a radical-molecule reaction. It could be argued that this value does not necessarily represent an upper limit of the rate constant because a similar reaction for regeneration of alanine could also take place, namely the reversal of reaction 2. Since the a-radical of glycine is located on a primary carbon and consequently it is less stable than the corresponding a-radical of alanine, which is located on a secondary carbon, it is estimated that the reversal of reaction 2 must be unimpor- tant. This is consistent with our experimental results in which a negligible amount of deuterated glycine formed from deuterated alanines.

Page 11

RADIOLYSIS OF GLYCINE-ALANINE MIXTURES IN AQUEOUS SOLUTIONS 97

0 20 40

20 40

Dose

0 10 20 I I - 100

7 10 2 z

3

I 1

(OkGYl

20 40

Figwe 3. Dose-response curves for the radiolysis of glycine (2) alanine (3) and the mixture of glycine (1) and alanine (4) at pH 6.0. Top-left panel: Experimental data of Akaboshi et al. (1990); top-right panel: Computer simulations considering equations in Table I; bottom-left panel: Computer simulations considering equations in Tables I and II; and bottom-right panel: Computer simulations considering equations in Tables I, II and reaction 2.

The mechanism of protection of glycine in the radiolysis of glycine-alanine mixed solutions is undoubtedly more complex. It is possible that the radical -CH(NH$)CO, could regenerate more glycine by another hydrogen atom abstrac- tion involving any product of the radiolysis of alanine.

5. Implications to the Preservation of Optically Active Amino Acids in the Early Earth

Glycine is among the most abundant amino acids formed in prebiotic experiments. For example, glycine and alanine are formed in similar yields in electric discharge experiments if the atmosphere is composed of CH4; however, glycine is almost exclusively formed from CO and CO2 model atmospheres (Schlesinger and Miller,

Page 12

98 RAFAEL NAVARRO-GONZALEZ ET AL.

1983). Based on the energy yields of HCN and HCHO in electric discharge exper- iments, Stribling and Miller (1987) have estimated a steady-state concentration of amino acids in the primitive hydrosphere of 3 x 10d4 M taking into account that the entire ocean would pass through submarine vents in 10 million years. Although there is no conclusive evidence of asymmetric prebiotic syntheses capable of pro- ducing any enantiomeric excesses at all on the primitive Earth, if we consider a hypothetical situation in which there would be an enrichment of about 10% of L-amino acids in the primitive ocean due to different mechanisms of asymmetric synthesis and amplification, we could crudely estimate the extend of radiation- induced destruction and radioracemization of amino acids in the early hydrosphere using our computer model. Of course the accuracy of this estimation depends on critical information that at present is not available, such as (1) evidence of prebi- otic syntheses capable of producing any enatiomeric excesses; (2) a demonstrable mechanism of asymmetric amplification; (3) a knowledge of the dose rate of ion- izing radiation of the early Earth; and (4) the correctness of the chemical model used.

It is normally difficult to assess the dose rate of ionizing radiation in the primi- tive hydrosphere. Radioactive disintegration of 40K,232 Th,235 U and 238U in a l-km column depth of lithosphere would have provided 11.7 J cm-*y-l four billion years ago (Miller and Urey, 1959). A significant percentage of the energy liberated would have been absorbed by the rocks themselves. If we assume that only 10% of this energy would have reached the primitive ocean, where it would be completely absorbed, and taking the volume of the ocean as 300 1 cm-* (Stribling and Miller, 1978), the dose rate of ionizing radiation in the primitive hydrosphere would have been 3.9 x 1 OP3 Gy y -’ four billion years ago for an ocean density of 1 g cmm3. Let us assume that the primitive ocean would have consisted of Hz0 and amino acids (total concentration: low4 M) with two different distributions: Ala/Gly = 1 and Ala/Gly = 0.01, for CI&- and COz-dominated atmospheres, respectively. Comput- er modeling suggests that alanine would be completely destroyed between 60,000 yrs (CO*-dominated atmosphere) and 90,000 yrs (CH4-dominated atmosphere) in spite of the fact that the concentration of alanine is significantly lower or equal than that of glycine. Complete radioracemization would occur within 30,000 yrs and 45,000 yrs, depending on the type of atmosphere used.

Other energy sources could enhance the rate of destruction and racemization of amino acids in the primitive hydrosphere. However, it is difficult to estimate the losses of amino acids by ultraviolet light, quench discharges, cavitation, etc. Undoubtedly, an additional mechanism for the racemization of amino acids in solution is through the ionization of their cr-hydrogens. For the case of alanine, its half-life of racemization is 11,000 years at 25” C in the pH range of 3-8 (Bada, 197 1). This value is comparable to the half-life of racemization of alanine induced by ionizing radiation in a CO*-dominated atmosphere. If we assume that chiral amino acids were a prerequisite for the origin of life, it can be deduced that life could have appeared in a relatively short period of time (less than 15,000 years)

Page 13

RADIOLYSIS OF GLYClNE?-ALANINE MIXTURES IN AQUEOUS SOLUTIONS 99

unless there was a constant supply of optical active amino acids from extraterrestrial sources.

Acknowledgements

This work was partially supported by grants of the National Council of Science and Technology of Mexico (CONACYT No. 1843-OE9211) and the National Autonomous University of Mexico (DGAPA No. IN100393) to one of us (R.N- G.).

Part of this work was carried out during a leave at the Laboratory of Chemical Evolution of the University of Maryland (R.N-G. and M.A).

References

Akaboshi, M., Noda, M., Kawai, H., Maki, H., and Kawamato, K.: Origins Life 9 (1979), 18 1. Akaboshi, M., Kawai, K., Maki, H., Ehrlich, W., and Honda, Y.: Origins Life Evol. Biosph. 20 (1990),

111. Bada, J.L.: Adv. Chem. Ser. 106 (1971), 309. Bielski, B.H.J., Cabelli, D.E., Arudi, R.L., and Ross, A.B.: J. Phys. Chem. Refi Data 14 (1985), 1041. Bonner, W.A.: J. Mol. Evol. 4 (1974). 23. Bonner, W.A. and Flores, J.J.: Origins Life 6 (1975), 187. Bonner, W.A., Van Dort, M.A., and Yearian, M.R.: Nature 258 (1975), 419. Bonner, W.A., Van Dort, M.A., and Yearian, M.R.: Nature 264 (1976), 198. Bonner,W.A.,VanDort,M.A.,Yearian,M.R.,Zeman,H.D.,andLi,G.C.:Isr.J.Chem. 15(1976/77),

89. Bomrer, W.A. and Lemmon, R.M.: Bioorganic Chem. 7 (1978). 175. Banner, W.A., Lemmon, R.M., and Noyes, H.P.: J. Org. Chem. 43 (1978). 522. Braun, W., Herron, J.T., and Kahaner, D.K.: Znt. J. Chem. Kin&. 20 (1988), 5 1. Buxton, G.V., Greenstock, C.L., Helman, W.P., and Ross, A.B.: J. Chem. Phys. Ref: Data 17 (1988),

513. Dean, J.A.: Lange’s Handbook ofchemistry, 13th edn., McGraw-Hill, New York, 1985. DragtiC, LG. and Draganic, Z.D.: The Radiation Chemistry of Water, Academic Press, New York,

1971. DraganiC, I.G. and Draganic, Z.D.: J. Phys. Chem. 77 (1973). 765. Draganic, Z., Draganic, I., Shimoyama, A., and Ponnamperuma, C.: Origins Life 8 (1977), 371. Dragattic. Z.D., NiketiC, V., and Vujosevic, S.I.: J. Mol. Evol. 22 (1985), 82. Ferris, J.P., Joshi, PC., Edelson, E.H., and Lawless, J.G.: J. Mol. Evol. 11(1978), 293. Greenstein, J.P. and Winitz, M.: The Chemistry of Amino Acids, Vol. 1, Robert E. Krieger Publ. Co.,

Malabar, FL, 1986, pp. 683. Hodge,L.A.,Dunning,F.B.,Walters,G.R.,White,R.H.,andSchroepferJr.,G.H.:Nature280(1979),

250. Maxwell, C.R., Peterson, D.C., and Sharpless, N.E.: Radiat. Res. 1(1954), 530. Meshitsuka, G., Shindo, K., Minegishi, A., Suguro, H., and Shinozaki, Y.: Bull. Chem. Sot. (Japan)

37 ( 1964), 2928. Miller, S.L. and Urey, H.C.: Science 130 (1959), 245. Navarro-Gonzalez, R., Ponnamperuma, C., and Kharma, R.K.: Origins Life Evol. Biosph. 21 (1992),

359. Navarro-Gonz&ez, R., Akaboshi, M., and Romero, A.: Viva Origin0 22 (1994), 14. Romero, A.: Computational Study of the Radioracemization of Amino Acids in Primitive Terrestrial

and Extraterrestrial Environments (in Spanish), B.Sc., Thesis, Facultad de Qufmica, UNAM, Mexico, 1995.

Schlesinger, G. and Miller, S.L.: J. Mol. Evol. 19 (1983), 376.

Page 14

100 RAFAEL NAVARROGONZ&Z ET AL.

Sharpless, N.E., Blair, A.E., and Maxwell, C.R.: Radiat. Res. 2 (1955). 135. Spinks, J.W.T. and Woods, R.J.: An Introduction to Radiation Chemistry, 3rd edn., Wiley, New York,

1990, pp. 339. Stribling, R. and Miller, S.L.: Origins Life 17 (1987), 261. Ulbricht, T.L.V.: Quart. Rev. 13 (1959), 48. Ulbricht. T.L.V. and Vester, F.: Tetrahedron 18 (1962), 629. Urbain, W.M.: in P.S., Elias and A.J., Cohen (eds.), Radiation Chemistry ofMajor Food Components,

Elsevier, Amsterdam, 1977, pp. 63. Vester, F., Ulbricht, T.L.V., and Krauch, H.: Naturwissenschafen 46 (1959), 68.