Glycine synthesis from nitrate and glyoxylate mediated by ferroan brucite: An integrated pathway for prebiotic amine synthesis - chimiak-et-al-2024-glycine-synthesis-from-nitrate-and-glyoxylate-mediated-by-ferroan-brucite-an-integrated-pathway-for.pdf

PNAS

2024 Vol. 121 No. 45 e2408248121

https://doi.org/10.1073/pnas.2408248121

1 of 7

RESEARCH ARTICLE

Significance

All known life requires amino

acids. Constraining the

environmental conditions

needed for amino acid synthesis,

therefore, can constrain the

environments in which life might

emerge. Previously, prebiotic

amino acid syntheses required

ammonia in the starting

conditions, which forms in a

reducing atmosphere or neutral

one with energetic inputs such as

lightning. This work

demonstrates the ability to

synthesize amino acids in an

environment with nitrate as the

initial nitrogen source, which

opens plausible environments for

life to include those with

oxidizing atmospheres. Here, we

synthesize glycine and

dicarboxylic amino acids from

nitrate, glyoxylate, and ferroan

brucite.

Author affiliations:

Department of Geological Sciences,

University of Colorado, Boulder, CO 80309; and

Depart

ment of Geological and Planetary Sciences, California

Institute of Technology, Pasadena, CA 91125

Author contributions: L.C. and A.S.T. designed research;

L.C. and E.H. performed research; L.C. and A.S.

contributed new reagents/analytic tools; L.C. analyzed

data; A.S. and A.S.T. aided with interpretation of results;

and L.C. wrote the paper.

The authors declare no competing interest.

This article is a PNAS Direct Submission.

This article is distributed under

Creative Commons

Attribution

NonCommercial

NoDerivatives License 4.0

(CC BY

ND)

To whom correspondence may be addressed. Email:

laura.chimiak@colorado.edu.

This article contains supporting information online at

https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.

2408248121/-

/DCSupplemental

Published October 28, 2024.

EARTH, ATMOSPHERIC, AND PLANETARY SCIENCES

Glycine synthesis from nitrate and glyoxylate mediated

by ferroan brucite: An integrated pathway for prebiotic

amine synthesis

L. Chimiak

, E. Hara

, A. Sessions

, and A.S. Templeton

Edited by Jonathan Lunine, Cornell University, Ithaca, NY; received April 24, 2024; accepted September 4, 2024

Amino acids are present in all known life, so identifying the environmental conditions

under which they can be synthesized constrains where life on Earth might have formed

and where life might be found on other planetary bodies. All known abiotic amino acid

syntheses require ammonia, which is only produced in reducing and neutral atmos-

pheres. Here, we demonstrate that the Fe

bearing hydroxide mineral ferroan brucite

[Fe

0.33

,Mg

0.67

(OH)

] can mediate the reaction of nitrate and glyoxylate to form glycine,

the simplest amino acid used in life. Up to 97% of this glycine was detected only after

acid digestion of the mineral, demonstrating that it had been strongly partitioned to

the mineral. The dicarboxylic amino acid 3

hydroxy aspartate was also detected, which

suggests that reactants underwent a mechanism that simultaneously produced mono

and dicarboxylic amino acids. Nitrate can be produced in both neutral and oxidizing

atmospheres, so reductive amination of nitrate and glyoxylate on a ferroan brucite surface

expands origins of life scenarios. First, it expands the environmental conditions in which

life’s precursors could form to include oxidizing atmospheres. Second, it demonstrates

the ability of ferroan brucite, an abundant, secondary mineral in serpentinizing systems

where olivine is partly hydrated, to mediate reductive amination. Finally, the results

demonstrate the need to consider mineral

bound products when analyzing samples for

abiotic amino acid synthesis.

origins of life | prebiotic chemistry | astrobiology

Amino acids are the building blocks of proteins, which drive life’s chemistry, so under-

standing the abiotic synthesis of amino acids can shed light on where life originated and

on what planets it might be found. Both

α-

amino acids and amines have been detected

in meteorite samples where they are postulated to form from abiotic Strecker synthesis

( 1 ). On early Earth, amino acids could have been produced via a myriad of reactions

including Strecker synthesis in early oceans or ponds and reductive amination in hydro-

thermal systems ( 2 – 4 ). Each of these pathways requires ammonia ( 5 , 6 ), so the environ-

ments from which life could emerge—based on current understandings of prebiotically

plausible amino acid syntheses—are limited to those which would have an ammonia

source. Ammonia production requires either a reducing atmosphere or water and an energy

input such as lightning in a neutral atmosphere ( 6 , 7 ). Early Earth and many astrobio-

logical targets such as Mars and Europa do not have reducing atmospheres ( 8 , 9 ) though

might have localized reducing environments. If a nitrogen source present in oxidizing

systems could produce aminated compounds, the environmental conditions under which

life could emerge would broaden.

Nitrate is an appealing alternative to ammonia as a N source for prebiotic syntheses: In

the absence of biology, nitrate is stable over long time periods, it is present in oxidizing and

neutral atmospheres, and it could be present on Europa, Mars, and early Earth ( 10 , 11 ).

Prior work has demonstrated the ability of iron (II) and iron (II–III) hydroxides to reduce

nitrate into ammonia; mixed iron (II–III) hydroxides can also perform reductive amination

on 2

keto acids with ammonia to form amino acids ( 3 , 12 ). However, attempts to link the

two processes of nitrate reduction and reductive amination have been unsuccessful ( 13 ).

Ferroan brucite [Fe

,Mg

(1−x)

(OH)

] is a notable mineral in prebiotic scenarios: It is

an abundant, metastable mineral formed from the hydration of primary silicates such

as olivine in serpentinizing systems ( 14 ), which were posited to exist on early Earth,

Mars, and moons including Europa ( 15 , 16 ). Although serpentinization produces local

anoxic and reducing conditions, it can occur on planets with globally oxidizing atmos-

pheres including modern Earth. Ferroan brucite with up to 33% Fe(II) currently occurs

in a modern terrestrial serpentinizing system, the Samail Ophiolite in Oman ( 17 ).

Ferroan brucite could reduce nitrate into ammonia by using Fe

as an electron donor,

and its Mg(OH)

component could aid in aldol condensation reactions and concentrate

Downloaded from https://www.pnas.org by "CALIFORNIA INST OF TECHNOLOGY, CALTECH LIBRARY SERVICES" on October 29, 2024 from IP address 24.24.204.12.

2 of 7

https://doi.org/10.1073/pnas.2408248121

pnas.org

organics ( 4 , 18 ), suggesting that this mineral might be an effec-

tive generator and reservoir of biorelevant compounds.

Glyoxylate, the simplest 2

keto acid, is a prime target to test

whether ferroan brucite can mediate reductive amination and to

assess the distribution of products that could be formed. Glyoxylate

has been demonstrated to undergo reductive amination into gly-

cine on prebiotically plausible minerals including green rust and

several transition metal sulfide minerals ( 3 , 4 ). Furthermore, gly-

oxylate’s propensity to polymerize enables it to form many biore-

levant compounds such as sugars, amino acids, and hydroxy acids,

which has led to the glyoxylate scenario ( 19 – 21 ). This scenario

posits that not only can glyoxylate form glycine and glycolic acid

through mineral

mediated reactions but that it also can polymerize

prior to these reactions to form the

α-

keto acid dihydroxyfumarate

(DHF), which then could be converted into tartaric acid and

hydroxy aspartic acid ( 21 ) as depicted in Scheme

1 .

Here, we demonstrate that nitrate reduction can be linked to

the reductive amination of glyoxylate to form glycine on a ferroan

brucite mineral surface. In addition to glycine production, ferroan

brucite mediates the formation of tartaric acid and 3

hydoxy

aspartic acid through reduction and reductive amination of DHF.

These results represent a successful experimental link of nitrate

reduction and reductive amination on a terrestrially occurring

mineral surface, which expands the environments in which amino

acid synthesis and possibly the emergence of life could occur to

include those with oxidizing atmospheres and/or a nitrate source.

Results and Discussion

Amino Acid Synthesis.

Glycine was produced in all experiments in

which glyoxylate reacted with nitrate or ammonia in the presence

of ferroan brucite (henceforth, “nitrate systems” and “ammonia

systems”). Positive identification was detected through comparison

to glycine standards on

NMR, LC

MS, and both GC

MS and

Orbitrap GC

MS (

SI Appendix

, Table S1 and

Fig. S1

;

Supporting

information

for further information). After reacting for two weeks

only the highly sensitive Orbitrap GC

MS detected the presence

of glycine in both nitrate systems’ aqueous fractions (

SI Appendix

Fig. S1

). The less sensitive LC

MS and NMR analyses detected

glycine only in the ammonia systems at this 2

wk time point

(Fig. 1). After 6 wk of reaction time, both ammonia and nitrate

systems had sufficient glycine production in both their aqueous

and hydrolyzed mineral fractions for NMR, LC

MS, and GC

detection (Figs. 1 and 2 and

SI Appendix

, Fig. S1

–

). At this

point, ammonia systems had a 52 ± 4% to 69 ± 15% glycine yield

relative to the initial 100 mM ammonia and 100 mM glyoxylate,

which was higher than the 0.1 ± 0.01% to 3.1 ± 0.2% glycine yield

relative to the initial 100 mM nitrate and 100 mM glyoxylate in

the nitrate systems (Fig. 3 and

SI Appendix

, Table S1 and

Fig. S1

Prior work demonstrates that lower Fe

:Fe

ratios increase the

rate of nitrate reduction (12), so the results here can be taken as a

lower bound of the amount of glycine that could be produced on

iron hydroxides.

The limiting reagent in these systems for the investigated reduc

tive amination mechanism was the Fe(II) in brucite, which pro-

vides electrons for the reduction reactions. As Fig.

4 demonstrates,

ammonia and nitrate systems, respectively, require 2 and 10 equiv

alents of Fe(II) per glycine produced.

The 62.5 mM Fe(II) in brucite therefore was the limiting reagent

in both systems. In the ammonia system, enough Fe(II) was present

to produce 31.25 mM of amino acids, and in the nitrate system, in

which nitrate was first reduced to ammonia, the maximum possible

yield of glycine would be 6.25 mM (

SI Appendix

for further details).

Relative to this, ammonia systems had a 167 ± 13% to 222 ± 50%

glycine yield and nitrate systems had a 1.6 ± 0.3% to 50.3 ± 1.2%

glycine yield (

SI Appendix

, Table S1

). Production of glycine from

glyoxylate and ammonia with no iron present has been previously

demonstrated and might result from aqueous chemistry with the

reductant, formate, a common contaminant in glyoxylate and preb-

iotically plausible compound (

SI Appendix

for further discussion).

Glycine was partitioned differently in the ammonia and nitrate

systems. At six weeks, the mineral fraction of the ammonia systems

had 17 to 27% of the total glycine produced while the mineral

fraction in the nitrate systems stored 71 to 97% of the total glycine

produced ( Figs.

2 and 3 and

SI Appendix

, Table

). Glycine was

not a contaminant: Experimental samples typically had greater

glycine concentrations than controls by an order of magnitude or

more ( Figs.

2 and 3 and

SI Appendix

, Table

S1 and

Fig. S1

). The

highest glycine abundance in a control or blank was in the

Mg(OH)

control with nitrate and glyoxylate; its glycine concen-

trations of 2.3 ± 0.4 × 10

−2

mM and below detection in the aque-

ous and mineral samples, respectively, were lower than any

measured sample (

SI Appendix

, Table

The nitrate and ammonia systems produced not only glycine

and glycolate but also tartaric acid and 3

hydroxy aspartic acid as

detected by NMR and by LC

MS and GC

MS in the case of the

amine compounds ( Fig.

2 and

SI Appendix

, Figs. S3–S10

Scheme

1 depicts the pathways by which these compounds would

form in an ammonia

containing system. Unlike with glycine, most

of the 3

hydroxy aspartic acid in both the nitrate and ammonia

systems were detected in the mineral fraction ( Fig.

2 ). At 6 wk,

NMR, LC

MS, and GC

MS measurements of samples from the

ammonia system also demonstrate the presence of iminodiacetic

acid, which implies that in this system, glycine reacts with glyox-

ylate, ( Figs.

2 and 4 and

SI Appendix

, Figs.

S3 and S10

) a down-

stream reaction known to occur in other reductive amination

systems whose first products are primary amines ( 22 ). LC

measurements also confirm iminodiacetic acid in the nitrate sys-

tems though in far lower abundance ( Fig.

2 ). As with glycine,

iminodiacetic acid has a higher abundance in the aqueous phase

than in the mineral phase in ammonia systems ( Fig.

2 ); however,

the nitrate systems show similar abundances in the two phases.

The success of these experiments in converting nitrate and glyox-

ylate into glycine when others such as Barge et

al. ( 13 ) were not

Scheme 1.

Simplified version of some of the first steps of the glyoxylate scenario.

Glyxolate can be reduced into glycolate, undergo reductive amination into glycine,

or polymerize into dihydroxyfumarate. Dihydroxyfumarate can subsequently

bereduced into tartarate or undergo reductive amination into 3

hydroxy

aspartate. All species resulting from polymerization are shown in plum.

Downloaded from https://www.pnas.org by "CALIFORNIA INST OF TECHNOLOGY, CALTECH LIBRARY SERVICES" on October 29, 2024 from IP address 24.24.204.12.

PNAS

2024 Vol. 121 No. 45 e2408248121

https://doi.org/10.1073/pnas.2408248121

3 of 7

successful likely results from three main factors. First, this work uses

samples with lower salt concentrations than prior work. The kinetics

of nitrate reduction and therefore ammonia available for reductive

amination decreases with increased salt concentration ( 12 ). Second,

this work uses more highly sensitive techniques to measure the pres-

ence of amino acids. Glycine was not apparent in samples from the

nitrate systems on NMR after 2 wk of reaction time and at 6 wk,

NMR analysis only showed of a minor peak. Detection on LC

analysis of samples from nitrate system samples was only made pos-

sible by concentrating initial samples by a factor of three. GC

was employed on the purified amine

fractions of all samples and

even then 2

wk samples were measured on a GC

Orbitrap

MS,

A

B

Fig. 1.

NMR of aqueous samples that have been treated with NaOH in the region where glycine’s

α-

H appears. All samples are diluted by the same amount,

so differences in sample intensity directly correlate to differences in glycine concentration in experiments. The structures in panel a are of glycine (g) and two

iminodiacetate (i), which are labeled in the spectra in panels a and b. All ammonia system samples are in green, nitrate system samples are in blue, magnesium

brucite controls are in plum, ferroan brucite controls are in orange, and a glycine standard is in red. Two

week samples are represented with dashed lines and

six

week samples are represented with solid lines. Panel (

) displays all

NMR spectra such that the full glycine peaks from the ammonia systems are visible.

Panel (

) displays the same data with the intensity axis zoomed in roughly 100 times such to better display the glycine peak from the nitrate systems and the

two

iminodiacetate peak in the ammonia system samples at 3.155 ppm. Full

NMR spectra for samples and standards are available in

SI Appendix

Downloaded from https://www.pnas.org by "CALIFORNIA INST OF TECHNOLOGY, CALTECH LIBRARY SERVICES" on October 29, 2024 from IP address 24.24.204.12.

4 of 7

https://doi.org/10.1073/pnas.2408248121

pnas.org

which is six orders of magnitude more sensitive than the NMR

detection limits that prior work employed ( 13 ). Finally, the majority

of glycine in samples in nitrate systems was associated with the min-

eral surface, which has not previously been analyzed in these systems.

If prior work produced glycine, it might have been in concentrations

too low to detect by the analytical methods or associated with the

mineral surface and therefore undetected.

Mineralogy.

Ammonia and nitrate systems were visually distinct

by 24 h (

SI Appendix

, Fig. S11

). Initially, all samples had clear

solution above the white brucite mineral; however, after 24 h,

the aqueous phase in the ammonia systems had developed a red

orange color and the bulk mineral fraction appeared gray. In

contrast, nitrate systems aqueous phase remained clear above a light

blue mineral component (

SI Appendix

, Fig. S11

). Mineralogical

analyses at 6 wk show that in all systems, the brucite constitutes the

predominant mineral phase (

SI Appendix

, Figs. S12–S14

), which

likely results from high degree of Mg(OH)

in all systems. New

components from the reaction also arise in ammonia and nitrate

systems. Specifically, ammonia systems develop a magnetite (Fe

)

component, and nitrate systems develop nitrate

substituted

pyroaurite [Mg

(OH)

(NO

) 4H

O], lepidocrocite

[Fe

O(OH)], and magnetite components (

SI Appendix

, Figs. S12

and S14

) as a result of Fe(II) oxidation and remineralization. The

nitrate systems also demonstrate mineral–nitrate and mineral–

organic interactions: Nitrate

substituted pyroaurite (

SI Appendix

Fig. 2.

MS of amine eluent at 6 wk with a 0.50 Da window around masses 150 Da, 134 Da, and 76 Da. Mineral samples are from hydrolyzed minerals, and

all other samples are from the aqueous phase. Fe,Mg(OH)

samples have 33% Fe and 67% Mg. When a spectrum has 1/10 in the upper right corner, the LC

trace has been scaled down by 1/10 (

i.e.

, the sample measurement intensity is divided by 10 at each data point). The 150 Da window has the 3

hydroxy aspartic

acid peak at 4.0 min. The 134 Da window has the iminodiacetic acid peak at 3.6 min and also includes aspartic acid (indicated by an asterisk at 6.0 min when

present). The 76 Da window has the glycine peak at 8.25 min. Chemical structures of the target analytes are depicted with the reference spectra.

Downloaded from https://www.pnas.org by "CALIFORNIA INST OF TECHNOLOGY, CALTECH LIBRARY SERVICES" on October 29, 2024 from IP address 24.24.204.12.

PNAS

2024 Vol. 121 No. 45 e2408248121

https://doi.org/10.1073/pnas.2408248121

5 of 7

Figs. S12 and S13

) forms through a mineral–nitrate interaction

and Raman spectra show that lepidocrocite phases have organic

and possible nitrate features (

SI Appendix

, Fig. S14

). The fact that

only nitrate systems have measurable mineral–nitrate and mineral–

organic interactions could impact the increased percentage of

glycine partitioned to the mineral fraction in nitrate systems. Three

phenomena could explain this result: 1) the mineral and/or organic

products in the nitrate systems have stronger sorption affinities

than those in the ammonia systems 2) the preferential scavenging

of iron from the brucite in the ammonia systems causing iron

bound organics to be in the aqueous phase (Fig. 4), 3) the increased

rate of reductive amination in the ammonia systems lowered the

pH to 7.81 and 7.89 (

SI Appendix

, Table S1

) which could lead

to mineral dissolution and with it the dissolution of the organic

mineral association.

Reductive Amination Mechanism.

In this study, nitrate and

glyoxylate produced glycine at conditions of approximately 1

atmosphere and 20 °C through reactions with ferroan brucite,

a common terrestrial mineral. While this study did not focus on

the synthetic mechanism, combining the results with prior studies

allows us to propose the following mechanism. Mineral

mediated

interactions differ between the ammonia and nitrate systems but

likely follow an overall scenario in which glyoxylate and ammonia

form an imine that is reduced to glycine using electrons provided

by ferroan brucite (Fig. 4 and

SI Appendix

for further discussion)

(Gomez

et al

, 2002). In the case of the nitrate system, this would

be preceded by the reduction of nitrate into ammonia as discussed

in Wang et al. (12) (Fig. 4

Several pieces of evidence support a proposed two

step mech-

anism in the nitrate system. First, the proposed mechanism would

delay glycine production: Nitrate reduction on Fe(OH)

does not

occur on short (<24 h) timescales at the temperatures and pressures

used here (1 atm and 20 °C) ( 12 , 23 ), so ammonia would need

time to form prior to glycine synthesis. At 2 wk, ammonia is

detected at millimolar concentrations while amino acids are

detected at micromolar levels in nitrate systems, and at 6 wk, this

relationship is reversed (

SI Appendix

, Table S1

). The rapidity of

reductive amination rates in the ammonia system demonstrates

that the use of nitrate as a nitrogen source causes this delay.

Second, XRD and FTIR analysis demonstrates that brucite

adsorbs and/or binds the carbon and nitrogen species (

SI Appendix

Figs. S12–S14

). Finally, glycine is almost fully partitioned into

the mineral fractions (

SI Appendix

, Table

), which would occur

in a process that preferentially has mineral

associated glycine pro

duction. A potential explanation for this partitioning is that min-

erals present only in the nitrate system (

e.g.

, nitrate

substituted

pyroaurite and/or lepidocrocite) bind more strongly to glycine

than do those in both systems (brucite and magnetite). However,

this hypothesis requires future investigation.

Prior to its reductive amination, some glyoxylate undergoes

aldol condensation, a process for which the rates increase in the

presence of hydroxides such as Mg(OH)

and related hydrotalcites

( 18 ). In the experiments here, condensation products undergo

reductive amination into 3

hydroxy aspartic acid, detected by

MS, almost exclusively in the mineral fraction in ammonia

systems ( Fig.

2 ). The amine site in glycine also equilibrates with

the carbonyl site in glyoxylate and is reduced to iminodiacetic acid

as an additional product. The proportion of these intermediates

and their ability to bind to an iron(II) site will determine the final

proportion of the products; systems with more ammonia will have

a higher imine abundance and therefore a higher concentration

of all amino acids.

The Emergence of Life.

Life is more likely to emerge in environments

that contain its building blocks including compounds such as lipids

and amino acids and energy sources for a protometabolism such

as redox active minerals. To this end, we must consider the abiotic

synthesis and preservation of these molecules and the physiochemical

conditions that aid these processes. The ability of ferroan brucite to

mediate the formation of glycine from nitrate and glyoxylate expands

environments that allow the abiotic synthesis of amino acids to

include those with oxidizing atmospheres. Nitrate could be produced

by photochemical reactions in a CO

atmosphere and would

then be stable in oxidizing conditions, resulting in up to micromolar

oceanic concentrations (10). When these nitrate

bearing fluids are

circulated into the subsurface into mafic or ultramafic rocks that had

produced minerals such as brucite during their hydration (e.g., during

serpentinization reactions), the nitrate could react with the brucite

and 2

keto acids such as glyoxylate to form amino acids (Fig. 4).

The rate of amino acid production from 2

keto acids in these

environments will depend on ferroan brucite’s ability to reduce

imine intermediates and on the concentrations of ammonia pro-

duced by nitrate conversion. Regarding brucite’s reduction capac-

ity, higher degrees of magnesium substitution will provide fewer

electrons for reduction and therefore decrease the rate of amino

and hydroxy acid production ( 24 ), so higher iron:magensium

ratios will favor amino acid synthesis. On the ferroan brucite sur-

face, 2

keto acids will either be reduced into hydroxy acids or will

react with ammonia and be reduced into amino acids, so higher

ammonia concentrations will increase amino acid production. To

this end, lower salinity and higher pH will be favored for amino

acid synthesis: Higher salinity decreases the rate for the conversion

of nitrate to ammonia ( 12 ) and high pH leads to a higher propor-

tion of the reactive ammonia species rather than ammonium ( 3 ).

In addition to being more terrestrially relevant than a pure fer-

roan endmember, the ferroan brucite Fe

(OH)

, with mixed

Fe(OH)

and Mg(OH)

surface functional groups rather than a

pure iron hydroxide mineral, provided two advantages for prebiotic

synthesis. First, the Mg(OH)

sites aid in the aldol condensation

of glyoxylate ( 18 ) without reducing it into unreactive glycolate ( 3 ).

Condensation enables the synthesis of dicarboxylic species using

simple precursors in the same area as glycine synthesis leading to

Fig. 3.

Glycine yield with respect to glyoxylate and to nitrogen source. Both

nitrogen and carbon sources start at 100 mM, so a 100% yield indicates that

both glyoxylate and nitrogen source were fully converted into 100 mM of

glycine. Nitrogen source is NO

−

for nitrate system samples (in dark blue for

Fe,Mg(OH)

samples and plum for Mg(OH)

samples) and NH

for ammonia

system samples (in green). Circles represent the glycine from hydrolyzed

mineral samples (mineral assoc); lines represent glycine in samples from the

aqueous phase.

Downloaded from https://www.pnas.org by "CALIFORNIA INST OF TECHNOLOGY, CALTECH LIBRARY SERVICES" on October 29, 2024 from IP address 24.24.204.12.

6 of 7

https://doi.org/10.1073/pnas.2408248121

pnas.org

a pool of diverse prebiotic compounds. Furthermore, brucite can

serve not only as a mineral to mediate the production of small

organic acids but also one to temporarily store it. Hydrolysis of

organic products is a common problem in prebiotic synthesis, and

mineral absorption is a potential solution ( 25 ). The Raman analysis

of washed mineral samples from nitrate systems here demonstrated

that there were grains on the mineral associated with carbon and

nitrogen. In these systems, amino acids are preferentially associated

with the mineral fraction, which points to a storage capability in

the experimental conditions (pH of 8 to 9 and low salt).

The scenario requires a nitrate source, glyoxylate source, water,

ferroan brucite, and a basic pH. Serpentinizing systems such as

subsurface aquifers and hydrothermal vents on early Earth could

meet these criteria ( 26 , 27 ). Extraterrestrial environments that expe

rienced serpentinization could have also been conducive for this

pathway to produce amino acids in the presence of nitrate

bearing

fluids. For example, on Mars both nitrate and amines have been

detected ( 11 , 28 ) in addition to evidence of past serpentinization

( 15 , 29 ) and a diverse suite of Fe(II)

Fe(III)

Mg minerals ( 25 ), so

reductive amination using nitrate could be a possible pathway by

which to form amines and amino acids. This pathway could also

occur on Europa, which is thought to have serpentinization of its

rocky core and nitrate availability in its ocean ( 8 ).

Conclusion

With ferroan brucite

mediated chemistry, amino acids and amines

can be synthesized using nitrate. This synthesis opens the con-

straints on early life to include environments with oxidizing

atmospheres. In the case of nitrate and glyoxylate reacting to form

glycine, the dicarboxylic acids, iminodiacetic acid, and 3

hydroxy

aspartate are also synthesized. These side products demonstrate

the ability to simultaneously produce mono

and dicarboxylic

amino acids from a single set of 2

keto acids and nitrate. As half

or more of all products were adsorbed to the mineral surface, the

environment would provide a manner by which to store these

amino acids together such that they could both be used to form

early proteins upon mineral dissolution and release.

Materials and Methods

Methods are depicted in

SI Appendix

, Fig. S1

and described in detail in SI. In

short, brucite samples with a composition of Fe

0.33

,Mg

0.67

(OH) were synthe-

sized in an anaerobic chamber with a 95%/5% N

headspace by precipitat-

ing FeCl

*4H

O and MgCl

*6H

O in a 1:2 ratio in a serum vial with KOH to a

pH of 10 (

SI Appendix

Materials and Methods

for a detailed description). All

water used was MilliQ 18.5

that was vigorously bubbled with N

for 15 min

and allowed to equilibrate with gas in the anaerobic chamber for at least 24 h

prior to use. The mineral precipitates were allowed to settle, and the overly-

ing solution was removed with a serological pipette and replaced with water

adjusted to pH 10 using 2 N KOH and 2 N HCl. This washing step was repeated

three times. Magnesium brucite was synthesized by following the same meth-

ods with a 0.75 M MgCl

*6H

O solution. The same day as the minerals were

synthesized, a 1.6 M glyoxylate solution, 1.6 M potassium nitrate solution, and

1.6 M ammonium chloride solution was made using glyoxylic acid monohydrate,

potassium nitrate, and ammonium chloride, respectively, and water with NaOH

added to make it pH 10. The glyoxylate solution was added at a 1:1 ratio with

a nitrogen source—either potassium nitrate or ammonium chloride solution.

A

B

Fig. 4.

Proposed mechanism for reductive amination in the (

) ammonia and (

) nitrate systems. In both schematics, the mineral surface is represented with

the deep blue color and aqueous solution is the overlying light blue panel. Dashed lines represent processes that use iron. Lines that have “RED” above them

depict reduction steps with the number of electrons used in parentheses next to it. Fe

is assumed to become Fe

as a result of each electron used in reduction.

Molecules with blue dashed lines connected to the mineral surface are thought to be bound during and after synthesis. The only intermediate imine depicted

is two

iminoacetate. Other imines are omitted for simplicity.

Downloaded from https://www.pnas.org by "CALIFORNIA INST OF TECHNOLOGY, CALTECH LIBRARY SERVICES" on October 29, 2024 from IP address 24.24.204.12.

PNAS

2024 Vol. 121 No. 45 e2408248121

https://doi.org/10.1073/pnas.2408248121

7 of 7

Glyoxylate

nitrate or glyoxylate

ammonia mixtures were then added to vials

containing mineral precipitates to achieve a final concentration of 100 mM of

each reactant in the solution. Controls for these experiments included ferroan

brucite with water and magnesium brucite with 100 mM glyoxylate and nitrate

from the same solutions as described above. The former provided a blank with

the same mineral reactant and the latter provided one with the same aqueous

reactants. All reaction vials were unstirred and left to sit for 24 h, after which

the pH was readjusted to

9.5 in NO

−

systems and 8.5 in NH

systems via the

addition of an anaerobic 2 N KOH solution. Samples were then left unstirred, in

the dark in a cabinet. Aqueous aliquots of each sample were taken at 2 wk and

6 wk and a mineral sample was taken at 6 wk. Subsamples of aqueous aliquots

were prepared for NMR analysis or GC

MS and LC

MS analysis. Samples for NMR

analyses were reacted with 1 N KOH to precipitate metals, centrifuged, decanted,

and reacted again until no metals precipitated. Those prepared for GC

MS and

MS were desalted on a Dowex 50 W

X8 100

200 mesh column with a 2 N

OH eluent. These sample eluents were then dried under N

and either worked

up in water for LC

MS and or

derivatized as an N

trifluoroacetic

methyl ester

for GC

MS analysis. Subsamples of mineral aliquots were dried under anoxic

conditions and analyzed by XRD or were hydrolyzed and desalted on a Dowex 50

X8 100

200 mesh column for LC

MS and GC

MS analysis as described above.

Data, Materials, and Software Availability.

All study data are included in the

article,

SI Appendix

, and the OSF Database (

https://osf.io/fexyj/

). The database

contains raw files for GC

MS, LC

MS, IC, XRD, and NMR in addition to spectra for

Raman, FTIR, and XRD (30).

ACKNOWLEDGMENTS.

This work was supported with funding from the Simons

Foundation Collaboration on the Origin of Life (SCOL) to Alexis Templeton. We

thank David VanderVelde at the Caltech Catalysis Center, John Eiler, and Victoria

Orphan for providing access to NMR, Orbitrap GC

MS, and the anaerobic chamber.

D. P. Glavin

et al.

, Primitive Meteorites and Asteroids (Elsevier, 2018), pp 205–271.

S. L. Miller, J. E. Van Trump, Origin of Life, Y. Wolman Ed. (Springer Netherlands, Dordrecht, 1981),

pp. 135–141.

L. M. Barge, E. Flores, M. M. Baum, D. G. VanderVelde, M. J. Russell, Redox and pH gradients

drive amino acid synthesis in iron oxyhydroxide mineral systems.

Proc. Natl. Acad. Sci. U.S.A.

116

4828–4833 (2019).

Y. Novikov, S. D. Copley, Reactivity landscape of pyruvate under simulated hydrothermal vent conditions.

Proc. Natl. Acad. Sci. U.S.A.

110

, 13283–13288 (2013).

N. Kitadai, S. Maruyama, Origins of building blocks of life: A review.

Geosci. Front.

9

, 1117–1153 (2017).

H. J. Cleaves, Prebiotic chemistry: What we know, what we don’t.

Evo. Edu. Outreach

5

, 342–360 (2012).

T. M. L. Wigley, P. Brimblecombe, Carbon dioxide, ammonia and the origin of life.

Nature

291

213–215 (1981).

K. P. Hand, C. F. Chyba, J. C. Priscu, R. W. Carlson, K. H. Nealson, in

Europa

, R.T. Pappalardo, W.B.

McKinnon, K.K. Khurana, Eds. (University of Arizona Press, 2017), pp. 589–630.

N. Noda

et al.

, Highly oxidizing aqueous environments on early mars inferred from scavenging

pattern of trace metals on manganese oxides.

J. Geophys. Res. Planets

124, 1282–1295 (2019).

10.

M. L. Wong, B. D. Charnay, P. Gao, Y. L. Yung, M. J. Russell, Nitrogen oxides in early earth’s

atmosphere as electron acceptors for life’s emergence.

Astrobiology

17

, 975–983 (2017).

11.

J. C. Stern

et al.

, Evidence for indigenous nitrogen in sedimentary and aeolian deposits from the

Curiosity rover investigations at Gale crater, Mars.

Proc. Natl. Acad. Sci. U.S.A.

112

, 4245–4250

(2015).

12.

X. Wang

et al.

, Abiotic reduction of nitrate to ammonium by iron (oxy)(hydr)oxides and its stable

isotope (

15N,

18O) dynamics.

Geochim. Cosmochim. Acta

347

, 28–41 (2023).

13.

L. M. Barge

et al.

, Prebiotic reactions in a Mars analog iron mineral system: Effects of nitrate, nitrite,

and ammonia on amino acid formation.

Geochim. Cosmochim. Acta

336

, 469–479 (2022).

14.

A. S. Templeton, E. T. Ellison, Formation and loss of metastable brucite: Does Fe(II)

bearing brucite

support microbial activity in serpentinizing ecosystems?

Philos. Trans. A Math. Phys. Eng. Sci.

378

20180423 (2020).

15.

S. D. Vance, M. Melwani Daswani, Serpentinite and the search for life beyond Earth.

Philos. Trans. A

Math. Phys. Eng. Sci.

378

, 20180421 (2020).

16.

M. Schulte, E. Shock, Coupled organic synthesis and mineral alteration on meteorite parent bodies.

Meteorit Planet Sci.

39

, 1577–1590 (2004).

17.

A. S. Templeton

et al.

, Accessing the subsurface biosphere within rocks undergoing active low

temperature serpentinization in the samail ophiolite (oman drilling project).

J. Geophys. Res.

Biogeosci.

126

, e2021JG006315 (2021).

18.

P. Kuśtrowski, D. Sułkowska, L. Chmielarz, R. Dziembaj, Aldol condensation of citral and acetone

over mesoporous catalysts obtained by thermal and chemical activation of magnesium–aluminum

hydrotalcite

like precursors.

Appl. Catal. A General

302

, 317–324 (2006).

19.

C. Butch

et al.

, Production of tartrates by cyanide

mediated dimerization of glyoxylate: A potential

abiotic pathway to the citric acid cycle.

J. Am. Chem. Soc.

135

, 13440–13445 (2013).

20.

V. N. Sagi, V. Punna, F. Hu, G. Meher, R. Krishnamurthy, Exploratory experiments on the chemistry of

the “glyoxylate scenario”: Formation of ketosugars from dihydroxyfumarate.

J. Am. Chem. Soc.

134

3577–3589 (2012).

21.

R. Krishnamurthy, C. L. Liotta, The potential of glyoxylate as a prebiotic source molecule and a

reactant in protometabolic pathways—The glyoxylose reaction.

Chem

9

, 784–797 (2023).

22.

A. Y. Sukhorukov, Catalytic reductive amination of aldehydes and ketones with nitro compounds:

New light on an old reaction.

Front. Chem.

8

, 215 (2020).

23.

C. J. Ottley, W. Davison, W. M. Edmunds, Chemical catalysis of nitrate reduction by iron (II).

Geochim.

Cosmochim. Acta

61

, 1819–1828 (1997).

24.

C. Huber, G. Wächtershäuser, Primordial reductive amination revisited.

Tetrahedron Lett.

44

1695–1697 (2003).

25.

J. T. T. Kloprogge, H. Hartman, Clays and the origin of life: The experiments.

Life (Basel)

12

, 259

(2022).

26.

M. Schulte, D. Blake, T. Hoehler, T. McCollom, Serpentinization and its implications for life on the

early Earth and Mars.

Astrobiology

6

, 364–376 (2006).

27.

K. R. Rempfert

et al.

, Subsurface biogeochemical cycling of nitrogen in the actively serpentinizing

Samail Ophiolite, Oman.

Front. Microbiol.

14

, 1139633 (2023).

28.

M. Millan

et al.

, Sedimentary organics in glen torridon, gale crater, mars: Results from the

SAM instrument suite and supporting laboratory analyses.

J. Geophys. Res. Planets

127

e2021JE007107 (2022).

29.

B. M. Tutolo, N. J. Tosca, Observational constraints on the process and products of Martian

serpentinization.

Sci. Adv.

9

, eadd8472 (2023).

30.

L. Chimiak, Glycine synthesis on ferroan brucite. Open Science Framework. https://osf.io/fexyj/.

Glycine synthesis on ferroan brucite. Deposited 3 October 2024.

Downloaded from https://www.pnas.org by "CALIFORNIA INST OF TECHNOLOGY, CALTECH LIBRARY SERVICES" on October 29, 2024 from IP address 24.24.204.12.