Thermostability of Model Protocell Membranes Author(s): Sheref S. Mansy and Jack W.

Szostak Reviewed work(s): Source: Proceedings of the National Academy of Sciences of the United States of America, Vol. 105, No. 36 (Sep. 9, 2008), pp. 13351-13355 Published by: National Academy of Sciences Stable URL: http://www.jstor.org/stable/25464047 . Accessed: 16/03/2012 07:25
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

National Academy of Sciences is collaborating with JSTOR to digitize, preserve and extend access to Proceedings of the National Academy of Sciences of the United States of America.

http://www.jstor.org

Thermostability
Sheref S. Mansy*

of model
of Molecular

W. Szostak+ and Jack

protocell membranes
and Integrative Biology, Massachusetts General

Howard Hughes Medical Institute, Department Hospital, Boston, MA 02114 Edited by Gerald F. Joyce, The Scripps Research The

Biology and Center for Computational

Institute, La Jolla, CA, and approved are

July8, 2008 (received for review May 26, 2008) thermostable. This remarkably of environments that could be observation tolerated extends cells the and

of a self-replicating earliest cells may have consisted genetic a self-replicating within vesicle. membrane polymer encapsulated show that vesicles of simple Here, we composed single-chain and such as fatty acids, amphiphiles fatty alcohols, fatty-acid thermostable and retain internal RNA glycerol esters are extremely and DNA 100?C. The at temperatures oligonucleotides ranging strands of encapsulated double-stranded from 0?C DNA can be

opens up ways inwhich thermal fluctuations could be used to advantage by primitive cells. Results
Retention of DNA Oligonucleotides at High Temperatures. We ini

range

by early

to

re at high temperature while separated by denaturation being in primitive tained within vesicles, implying that strand separation could been mediated have fluctuations protocells by thermal without the loss of genetic material temperatures, complex charged cross fatty-acid-based membranes excursions may high temperature before the evolution of advanced of these membranes thermostability taneous replication of encapsulated of template-copying chemistry at separation and nutrient uptake molecules from the protocell. such as At elevated nucleotides

very rapidly, have facilitated membrane is consistent acids nucleic

that suggesting nutrient uptake The transporters.

tially examined the thermal stability of 2:1 myristoleic acid (MA)/monomyristolein (GMM) vesicles, because this composi tion leads to vesicles that tolerate a wide range of salt and pH conditions, including the presence of up to 4 mM Mg2+ (5, 6). We testedvesicle stability encapsulating a fluorescein-labeled by
10-mer

DNA

with the spon by the alternation with low temperature strand temperature. | vesicle | prebiotic

by size-exclusion chromatography (SEC), and then incu bating aliquots of the purified vesicles at differenttemperatures for 1 h. The fraction of the oligonucleotide thatwas released from the vesicles during the high-temperature incubation was
determined by a second round of size-exclusion chromatogra

oligodeoxynucleotide

(dAi0),

removing

unencapsulated

at high

origin of life | RNA world

| synthetic biology

We

have recentlydescribed a laboratorymodel of a simple compatibilityof the protocell components.We

tocell membranes reasonably are composed permeable of fatty acids to nucleoside and protocell that is useful for assessing the interactions related and

found that pro

molecules

place

and that efficient template directed copying reactions can take in the vesicle interior after the addition of external
activated nucleotides (1). These observations provide strong

phosphorimidazolides

support for theplausibility of the heterotrophic protocell model but immediatelybring to mind several additional questions.Most important, if a genetic polymer is copied inside membrane vesicles, how could the strands of the double-stranded product be separated? The possibility of thermal strand separation, as in
PCR, has seemed there might be environmental conditions that could more facilitate nucleotide efficient uptake, allowing replication or perhaps the utilization of more highly charged such substrates as nonactivated or short nucleoside nucleotides, polyphosphates,

phy,with free and encapsulated DNA measured by f luorimetry. Surprisingly,2:1 MA/GMM vesicles remained completely stable to 100?C,with no detectable release of the encapsulated oligo nucleotide (Fig. 1) or change in vesicle size (measured by dynamic light scattering). However, longer high-temperature incubation times did result in significant leakage (^20% re leased after 10 h at 100?C). To test whether this temperature stability is a general feature of vesicles composed of single-chain amphiphiles or is a unique characteristic of 2:1 MA/GMM vesicles, we examined vesicles made with a variety of amphiphiles and amphiphilemixtures by We using the same oligodeoxynucleotide retentionassay (Fig. 1). examined fourpure fatty acids at pH 8.5,which is approximately the pH at which half of the fatty acid is ionized and half is protonated, and which corresponds to the pH of maximum hydrogen bond donor to an adjacent ionized carboxylate (4, 8). Pure myristoleic acid (CI4:1) vesicles released small amounts of
after stability because every protonated carboxylate can act as a

of instability fatty-acid-basedvesicles (2, 3). A second question

is whether

problematic

in view

of the presumed

thermal

DNA

would seem likely oligonucleotides (3). Again, high temperatures to help, but thispossibility has not been explored because of the assumption that such conditions would disrupt vesicle integrity
and lead to the release of contents,

higher temperatures (^10-20% released after 1 h at 60-80?C). Palmitoleic acid (CI6:1) and oleic acid (C18.T) vesicles were completely stable to 90?C, but did show 30-35% leakage of entrapped DNA after 1 h at 100?C (Fig. IA). Thus, increasing chain length leads to increased thermal stabilityof the bilayer
membrane, entropic consistent with membrane release as stabilization

1 h at 50?C,

with

increasing

amounts

released

at

> CC I UJ X u

based on the instability of fatty-acid vesicles under several environmental conditions that do not affectphospholipid vesi
vesicles Also,

to the environment.

The

assumption

genetic materials, including of thermal instability

is

acids aremuch higher than forphospholipids (4), and dilution of

below this concentration such as leads to vesicle are dissolution. divalent cations out

cles. For

example,

the critical

aggregate

concentrations

for fatty

are buried in themembrane (9).We also examined linoleic acid (CI8:2) vesicles; this lipid leads tomore fluidmembranes with increased permeability to nucleotides at low temperatures (1). This C18:2 fattyacid behaved similarly to theC18:l oleic acid, with linoleic acid vesicles being completely stable to 90?C but
Author contributions:S.S.M. and J.W.S.designed research; S.S.M. performed research; S.S.M. and J.W.S.analyzed data; and S.S.M. and J.W.S. wrote the article. The authors declare no conflictof interest. Freelyavailable online through the PNAS open access option. *Presentaddress: Department of Chemistry and Biochemistry, of University Denver, Den ver,CO 80208. whom correpondence should be addressed. E-mail:szostak@molbio.mgh.harvard.edu. fTo ? 2008 byThe National Academy of Sciences of the USA

effect

of water

the hydrophobic

the through acyl chains

u 8 x Q. o

and fatty-acid glycerol esters fatty alcohols (7). as we show here, these environmental sensitivities do However, not extend to temperature, and certain vesicles fatty-acid-based PNAS

narrow posed solely of fattyacids are only stable within a fairly pH range (4), although stability to high pH can be conferred by
admixture with

to crystallize

lizing to fatty-acidvesicles, because the salts of fattyacids tend

of solution (5, 6). Furthermore, vesicles com

Mg2+

and Ca2+

very destabi

www.pnas.org/cgi/doi/10.1073/pnas.0805086105

September

9,2008

| vol.105

| no. 36

| 13351-13355

/\

100-n--a-a?4L-*

-*

t5

100-ji--

-i -

-*-?

> 1w-

60-

?405
0

\ vd 2

2020

\ V

\ \

V \

\ \ \ \

" 100

2 \\ 40" c \
5 20\

> 60i80~ \

"
"

"

^\\
\
40 60 (?C) 80 100

V-

OH-1-,-r?^?^=J???-L 40 60 80 temperature (?C)

OH-,-,-,-,?i?j? 0 20 temperature

5t?'-'-'-"r

I4 ^^\
1

\v N,

2-

? -

?
g ? 20OH-,-,-,- > 20 40

~-

iooi

. -l?

-x

\N
- V ? x5 \ *k ^ '

" \\
\
-V-L -

E ~

5ci-'-

?r

I --^^
? ^"^-^^

\\ 40-

1 4-?"

CN. I" 0 25 50 75 100

5 35_ ^^_
10

^^"\^
20 30

0-|-,-,-Ai-f60 80

3.0-)-1-1-1 100 0

time (h) temperature (?C)

time (min)

from vesicles of the indicated composition dA10 oligonucleotide Fig. 1. Thermostability of model protocell membranes. The leakage of a fluorescein-labeled was monitored as a function of time. (A) Influence ofthe acyl chain on vesicle stability. O, 2:1 decanoic acid/decanol; , ,oleic acid; MA; A, palmitoleic acid; , linoleic acid; , 2:1 MA/farnesol. (B) Influence of the head group on C14:1 vesicle stability. , ,2:1 MA/MA-OH; ,2:1 MA/GMM. Identical resultswere MA; and 2:1 oleic acid/monoolein as for 2:1 MA/GMM (C) Time-dependent given for 2:1 palmitoleic acid/monopalmitolein leakage from 2:1 MA/GMM vesicles at 100?C. (D) Influence ofthe head group on C10:0 vesicle stability. ,2:1 decanoic acid/decanol; #, 4:1:1 decanoic acid/decanol/monocaprin. (E) Time-dependent leakage from 4:1:1 decanoic acid/decanol/monocaprin model prebiotic vesicles at 100?C Solution conditions were 0.2 M sodium bicine, pH 8.5.

exhibiting ^20%
100?C. We are then examined stable

release of encapsulated DNA

the effect on vesicle incubation can interact stability 100?C,

after 1 h at
of a series of

were acid/decanol leak encapsulated capric were capable

somewhat

more

additives to pure MA. As noted above, MA/GMM

during prolonged at

(2:1) vesicles
presumably ionized car

oligonucleotide could be acid/monocaprin of retaining gel an

thermostable, at 50-60?C. observed

but Vesicles

started

to and as

of 2:1

by microscopy oligonucleotide, and

because of the ability of the glycerol headgroup to provide two

hydrogen boxylates bond donors that with the of MA. also

evidenced by dialysis and microscopy, but were not stable

enough to survive filtration crystallized monocaprin. explored monocaprin. 60-70?C

entrapped

alcohol (MA-OH), which can provide one hydrogen bond donor

stabilizes MA

Consistent

with

this

argument,

myristoleyl

reflect the large ratio of head group size to acyl chain length for
To decrease the of ternary mixtures head size, we average group and acid with decanol capric these in exhibited significantly periods of high temperatures

overnight

at room

temperature.

chromatography This

gradually may

instability

GMM.

per molecule, composed

To determine whether the thermostability of vesicles

of could be fatty acids longer-chain similarly of palmitoleic examined mixtures acid and oleic monoester derivatives. corresponding glycerol cases the addition Thus, of the glycerol im acid As

vesicles,

but

less effectively

than

creased thermostabilityand retained oligonucleotides for 1 h at

and could tolerate shorter

Interestingly,

we proved, with their expected, DNA

in both

monoester can

lipid conferred stabilityat 100?C, and no leakage of encapsulated

was observed (Fig. IB). vesicle thermostability

up to 100?C for 1min with no detectable loss of encapsulated oligonucleotide (Fig. IE). These experiments show that amphi phile mixtures of the kind thatmay have been present on the
early least earth could at high temperature. periods We wondered observed whether temperature or whether to multiple thermocycles rapid and for brief temperature result in vesicle disruption. To encapsulate nucleic acids and retain these at stability extends in large changes answer this question,

be enhanced either by strengthening head group interactions (e.g., adding GMM orM-OH to MA) or acyl chain interactions (e.g., increasing chain length from 14 to 16 to 18 carbons). In contrast, the highly branched isoprenoid alcohol farnesol de
creased We vesicle

interactions (Fig. IA).

examined are more

stability,

presumably of vesicles

by weakening composed

acyl

chain

we subjected MA/GMM (2:1) vesicles to 20 cycles of 2 min at 20?C and 2 min at 90?C, and we quantified the loss of entrapped
oligonucleotide. We observed. After then

previously reported (4) are only stable at >50 mM total amphi

these vesicles were phile concentration; size-exclusion chromatography (10), to measure unable to their ability temperatures. However, synthesis and too unstable dAi0 and retain therefore, at

than longer-chain unsaturated plausible prebiotically as Pure capric acid (C10:0, decanoic amphiphiles. acid) vesicles, to purify by we were elevated

(C10) saturated amphiphiles (Fig. ID) because thesemolecules

the stability

of short-chain

vesicles [decanoic acid (DA)/decanol (DOH)/glycerol monoester of decanoic acid (GMD)::4.T:1] to a similar thermocycling of 30 s at 25?C and 30 s at 90?C). In this case, regime (20 cycles 10-20% of the labeled DNA contents did leak out over 20 cycles,
to a loss of 0.5-1.0% of contents corresponding per cycle. The to multiple of vesicles is an important stability thermocycles characteristic because it demonstrates that vesicles of composed

no 20 cycles, loss examined the response

of nucleic

acid

was

of prebiotic-model

Type (FTT) chemistry is likely to yield a mixture of hydrocar

bons, alcohols, of fatty acids (13). Therefore, aldehydes, could form we acids in a typical examined mixtures (11, esters 12), and glycerol scenario lagoon" "drying of capric acid with some

by modified

Fischer-Tropsch

are envi of surviving simple single-chain amphiphiles capable ronments such as hydrothermal vents and hot that are springs for sites of early evolution candidates (14). DNA Strand Separation within Vesicles. We used an oligonucleotide

of these related compounds (Fig. ID). Vesicles of 2:1 capric

13352 | www.pnas.org/cgi/doi/10.1073/pnas.0805086105 Mansy and Szostak

FRET

pair strategy to demonstrate thatvesicle thermostability

A0;^

==

I-'-'-'-'-'-1

w '-'</

Intravesicular DNA strand melting. (A) Schematic representation of Fig. 2. the experimental setup. Black bars represent DNA molecules either modified with a fluorophore (D) or quencher (Q). White bars represent unlabeled strands of DNA. (B) Increase in fluorescence arising from intravesicular DNA strand melting and annealing. C, control; A, 2:1 MA/GMM; B, 4:1:1 decanoic acid/decanol/monocaprin; S, solution reaction (that is,not inside vesicles). The control reaction was ofthe same composition but not subjected to a thermocycle.

III

12

3 time (min)

B ?
?

I
4K
W Tt\ X. T\\

'?

'

_' -^=
X

'

'

\ \\X

within and

can be exploited forDNA

vesicles. ends, Complementary 3'

strand separation and reannealing

19-mer oligodeoxynucleotides and mixed were annealed with an

5
0

were labeled with a fluorophore and a quencher at their5' that

respectively,

\\\
5 10 i_i_i_i_i-1-1-1-1 ?^@>-?-@^-^^

\
15 20 25 time (min)

excess of unlabeled oligonucleotide duplex of exactly the same sequence (Fig. 2). Heating above theTm of the dsDNA (75?C)
leads to strand

random reannealing of labeled and unlabeled oligonucletides, toboth 2:1MA/GMM and 4:1:1 capric acid/decanol/monocaprin vesicles with a 1-min incubation at 90?C, followed by cooling to
20?C for reannealing. After ensure that only encapsulated compositions is diagnostic fluorescence and consistent a result Nucleotide gave rise of strand separation increase with was chromatography was monitored, both vesicle to the increased fluorescence signal that DNA and similar The reannealing. to control reactions and measured in solution as size-exclusion to thus yielding an increase in fluorescence. This assay was applied

separation,

and

subsequent

cooling

allowed

C^-k W 100_? O

X.

-Sk^-O?o X ^

O 60"U

2^ 20- \\
10

" 'I 40- \\

20 30 40 50

> \

\ \

^ \\ ? \i
60 70 80 90 100

strand melting complete of thermocycling within vesicles.

reannealing

temperature (?C)
nucleotide permeability. (A) Nucleotide mono Fig. 3. High-temperature phosphate permeability of 2:1 MA/GMM vesicles at 90?C T, AMP; V, AMP + ,CMP; O, CMP + MgCI2, ,GMP; <0, GMP + MgCI2, ,UMP; D, UMP + MgCI2,

at Elevated We have previ Temperatures. Permeability can nucleotides diffuse ously shown that activated spontaneously so that nucleo across the fatty-acid-based bilayer membranes, can enter vesicles to the outside and take part in tides added on the inside of reactions template-directed primer-extension the vesicle

we

concerted flipping of an amphiphile-solute complex as opposed to packing defects arising fromgel to liquid phase transitions,as seen fordimyristoyl phosphatidylcholine membranes (15).When
are stable to elevated realized that these vesicles tempera would be whether nucleotide tures, we wondered permeability at higher enhanced significantly temperatures. By encapsulating at high temperatures, radiolabeled fol nucleotides, incubating

(1). Nucleotide

permeation

appears

to operate

via

MgCI2. (B) Influence of 5' phosphates on 2:1 MA/GMM vesicle permeability at 90?C #, AMP; O, AMP + MgCI2; A, ADP; A, ADP + MgCI2; , ATP; D, ATP + MgCI2. (0 Oligomer permeability of 2:1 MA/GMM vesicles after a 1-h incubation at each indicated temperature. ,NAD; #, AAA; O, AAAA.

oside

(1).We further probed the influence of charge bymeasuring the permeation ofADP and ATP (Fig. 3B). In the absence ofMg2+, ADP crossed themembrane more slowly thanAMP (^20 min for complete equilibration of ADP vs. <10 min forAMP); thanAMP (complete equilibration in<3 min), consistent with the higher affinityof ADP forMg2+ (16) and consistent with previously observed trends at 23?C (1). ATP permeation was much slower (>1 h for completion) and could not be fit to a
curve. Simple exponential prebiotic model membranes more than previously robust clearly allowing appreciated, such as elevated that facilitate conditions, temperatures, single such as oligonucleotides. these are for the however, in the presence of Mg2+, ADP permeated more rapidly

monophosphates

crossed

the membrane

over

a 24-h period

> DC t s x u

were

lowed by SEC
able

to resolve retained and leaked nucleotides, we

rapid permeation of activated Because

to observe membranes

with uridine-5'-phosphorimidazolide
MA/GMM within low-temperature nonactivated slower than data nucleotide demonstrated

30 s at 90?C.

equilibrating across 2:1

our previous of the permeability was significantly to determine sought Unactivated of NMPs

nucleotides,

that

activated nucleoside

whether high temperatureswould also increase thepermeability

of unactivated are much less bears that permeable two negative monophosphates. than activated charges of the more NMPs instead polar because one. nonactivated with release the We

monophosphates we nucleotides (1),

v/> > X Q. o 00

U|

material

uptake of critical nutrientswithout the loss of larger entrapped Although the vesicle stability and nucleotide permeability
of membrane with PCR, we were unable compositions to successfully are clearly reconsti

phosphate observed nucleoside

the permeability

characteristics compatible

of encapsulated nucleotides from 100-nmMA/GMM vesicles AMP, GMP, CMP, UMP, reaching completion within 10min for and deoxyadenosine-5'-monophosphate (dAMP) (Fig. 3/4).Un
der the same solution conditions but at 23?C, <10% of nucle

monophosphates

dramatically

increased,

tutePCR within fatty-acidvesicles, most likelydue to the strong inhibitionof DNA polymerases by even low concentrations of fatty acids (17). PCR within phospholipid vesicles has been previously documented, but because of the impermeabilityof the
PNAS | September 9,2008 | vol.105 | no. 36 | 13353

Mansy and Szostak

membrane, formation Permeability

the nucleotides rather

had

than delivered

to be encapsulated during vesicle across the membrane (18). The permeation of mononucleo

is that mixtures more desirable pure

A thirdsignificantand surprisingoutcome of our experiments

of amphiphiles characteristics seem to form membranes have with than membranes studies that Previous composed shown that tolerant of the

membrane

side diphosphates and triphosphates,but not a 10-mer oligonu cleotide, suggested that short oligonucleotides might cross the
at elevated temperatures. Therefore, we measured

of Oligonucleotides.

addition of fattyalcohols and fatty-acidglycerol esters to pure presence of divalent cations (5), and which show increased permeability (1) to sugars and nucleotides. Our currentwork
shows that such mixtures form membranes with with enhanced branes. thermostability, Because prebiotic compared chemical pure fatty-acid reactions undoubtedly dramatically mem gen fatty acids lead to membranes are more to the

single,

amphiphiles.

the retentionof a series of oligomers within 100-nm MA/GMM (2:1) vesicles at 90?C. In the absence of Mg2+, dinucleoside diphosphates (NppN) required 4 min for 50% leakage from the vesicles, whereas trinucleotides (pNpNpN) leaked out more slowly, with 50% loss after 37 min. Tetranucleotides were completely retained after 1 h at 90?C (Fig. 3C). Thus, protocell replication schemes that invoke the sequential templated ligation of oligonucleotide substrates are only feasible for dinucleotides
and from used trinucleotides, the external to generate

mixtures and

erated complex mixtures of lipids, rather than a single lipid species, it isplausible thatvesicles assembled fromsuch prebiotic
might have permeability had greater ion tolerance, thermostability, than more model homogeneous laboratory the availability of a relatively For homoge

generation of substrate oligonucleotides of length>4

pool trapped molecules smaller than entrapped at 90?C. lost to the environment rapidly a of substrates a

are taken up that the substrates assuming environment. the intracellular Alternatively, for replication, tetramer would be

vesicles. This behavior is in striking contrast to the situationwith neous set of initialnucleotides is generally regarded as essential ochemical purity is important for RNA
presence of contaminating incorrect to the synthesis of useful oligonucleotides. genetic polymers, where

could be

while

enantiomers

replication, as the
leads to the

example,

stere

Discussion The high thermal stability of themodel protocell membranes we have studied has significantimplications for theorigin of that
cellular life. At the most with basic level, our results very primitive, in environments tuations. subsurface as well of early Thus, self-assembled large consideration protocell and/or frequent of appropriate suggest that even structures could survive temperature environments fluc for

strong inhibitionof furtherpolymerization (21).

Materials and Methods

Materials. Fatty acids, fatty alcohols, and the glycerol monoesters of fatty acids were obtained from Nu Chek Prep. All other chemicals were obtained from Sigma-Aldrich. Vesicle Preparation. Fatty-acid vesicles were prepared by oil dispersion in buffered solutions as previously described (22, 23). For vesicles composed of mixtures of amphiphiles, the oils were mixed before dispersion in aqueous

the origin of the first cells need not be limited to sheltered

surface and include locales, but can also reasonably to diurnal locations variations, temperature subject as locations near heat sources vents such as hydrothermal cells to tolerate temperature fluctuations is the possi

near-surface

and hot springs.Perhaps more interestingthan the simple ability

bility thatprimitivecells could take advantage of thermal cycling to facilitate both strand separation of replicated genomic tem plates and the rapid uptake of nutrientsduring high-temperature
excursions. For of

cycle," inwhich replication of the genetic material and growth

the vesicle compartment interrupted separation at low to moderate tem proceed that lead interludes by high-temperature and the influx of additional nucleotides. Cell shear forces, would by environmental of genetic molecules into daughter invoked in

example,

one

could

imagine

primitive

"cell

which the running buffer contained the same amphiphile composition as the vesicles at a concentration above their critical aggregate concentration. Ves icle size was measured by dynamic light scattering with a PDDLS/CoolBatch 90T (Precision Detectors).

solution. All vesicle preparations were extruded 11 times through 100-nm filters with an Avanti miniextruder (Avanti Polar pore-size polycarbonate Lipids). For the encapsulation of molecules, amphiphiles were resuspended in the presence of the encapsulant. Separation of entrapped and unencapsu lated material was by gel filtrationwith Sepharose 4B resin (Sigma-Aldrich) in

peratures, to strand

division, perhaps lead to random cells. Diurnal

triggered assortment

are fluctuations temperature frequently as a means scenarios of cyclic concentration prebiotic or as a means orative drying followed by redissolution, chemical transformations fluctuations even on within temperature could the early also evaporated drive periodic

by evap of driving materials. Such strand sepa

scintillation fluid (MP biomedicals) on a Beckman Coulter LS 6500 multipurpose scintillation counter. Isotopes were either 14Cor 3H and were obtained from Moravek Biochemicals and Radiochemicals. with UniverSol Vesicle Stability and Oligonucleotide Permeability. The assay was as described for nucleotide permeability except that oligonucleotides were used inplace of mononucleotides. sequences were poly(A), except for the Oligonucleotide General Hospital dimer, and were either synthesized at the Massachusetts DNA core facility or the Yale University Keck facility (New Haven, CT). The 10-mer had a covalently attached 5'-fluorescein. The dimer was 3H-labeled NAD (Moravek Biochemicals and Radiochemicals). Trimer and tetramer oligo nucleotides were 5'-labeled by using 32P-y-ATP and T4 polynucleotide kinase (New England Biolabs). Fluorescence was measured with a SpectraMAX Gemi niEM fluorescence plate reader (Molecular Devices).

Nucleotide Permeability. Nucleotide permeability measurements were made in0.2 M sodium bicine, pH 8.5. Radioactive nucleotides (0.1 mM) were encap sulated, and the vesicles were purified by gel filtration (Sepharose 4B). Vesicles were incubated at different temperatures in a Bio-Rad DNA Engine Peltier thermal cycler. Samples were then loaded on a gel filtration column, collected with a Gilson FC 203B fraction collector, and analyzed by scintillation counting

diurnal

ration and nutrient uptake in vesicles, but the length of the leakage of genetic polymers fromprimitive cells unless the cell
were composed possibility of attractive fairly long-chain for an environment fatty acids. that would cycle, earth, would lead to significant

membranes Another

subject primitive cells to large thermal fluctuations is provided by the possibility of convection cells near or within surface hot cyclingofDNA molecules between moderate and high temper atures, leading to amplification by PCR (19, 20). On the early
or hydrothermal springs cells have been shown, vents. Rayleigh-Bernard to allow in the laboratory, convection for the rapid

an cells might fluid such convection earth, subject adjacent a to reservoir of replicating containing protocells population sufficient for strand separa brief periods of high temperature, enhanced tion and nutrient by long and separated uptake random interludes at lower temperatures.

were ob Strand Melting Assay. Fluorescently labeled oligodeoxynucleotides 5'-ATGCGCCCGGCCTAGGGCC-3' tained from Integrated DNA Technologies; was synthesized with a 5' tetrachlorofluorescein and 5' modification, GGCCCTAGGCCGGGCGCAT-3'

was synthesized with a 3' black hole quencher-1 (BHQ-1) modification. MA/GMM (2:1) samples were incubated at on 20?C for 2 min, 90?C for 1min, and 20?C for 2 min before measurement SpectraMAX GeminiEM fluorescence plate reader (Molecular Devices) with excitation and emission at 518 and 539 nm, respectively.

13354

| www.pnas.org/cgi/doi/10.1073/pnas.0805086105

Mansy and Szostak

ACKNOWLEDGMENTS. We thank J.P. Schrum, R. J.Bruckner, M. M. Hanczyc, B. Seelig, I.A. Chen, S. Tobe, T. F. Zhu, Q. Dufton, J. Iwasa, F. P. Seebeck, A. Ricardo, A. J.Bell, and M. Sam for helpful discussions. This work was supported of 1. Mansy SS, etal. (2008) Template-directed synthesis a genetic polymer ina model protocell.Nature 454:122-125. membranes. Top and physicsof primitive 2. Deamer DW, Dworkin JP(2005) Chemistry CurrChem 259:1-27. Nature 409:387-390. 3. Szostak JW, BartelDP, Luisi PL (2001) Synthesizing life. 4. Hargreaves WR, Deamer DW (1978) Liposomes from ionic,single-chainamphiphiles. Biochemistry17:3759-3768. model protocellvesicles. 5. Chen IA, Salehi-Ashtiani Szostak JW(2005) RNA catalysis in K, J Am Chem Soc 127:13213-13219. of 6. Monnard PA,Apel CL, KanavariotiA, Deamer DW (2002) Influence ionic inorganic and polymerizationprocesses related to early formsof life: solutes on self-assembly for Implications a prebiotic aqueous medium. Astrobiology 2:139-152. 7. Apel CL, Deamer DW, Mautner MN (2002) Self-assembledvesiclesof monocarboxylic and forthe encapsulation of biopolymers. acids and alcohols: Conditions forstability Biochim BiophysActa 1559:1-9. 8. Haines TH (1983)Anionic lipid headgroups as a proton-conductingpathway along the surfaceof membranes: A hypothesis.Proc Natl Acad Sci USA 80:160-164. and Surface Forces (Academic, London). JN 9. Israelachvili, (1991) Intermolecular 10.Walde P, Wick R, Fresta Mangone A, LuisiPL (1994) Autopoietic self-reproduction M, Am Chem Soc 116:11649-11654. of fatty acid vesicles. J under hydrothermalcon 11. McCollom TM, Ritter Simoneit BR (1999) Lipid synthesis G, reactions.Orig LifeEvol Biosph 29:153-166. ditions by Fischer-Tropsch-type 12. RushdiAl, Simoneit BR (2001) Lipid formationby aqueous Fischer-Tropsch-type syn thesis over a temperature range of 100 to 400 degrees C. Orig Life Evol Biosph 31:103-118.

J.W.S. isan Investiga by NASA Exobiology Program Grant EXB02-0031-0018. tor ofthe Howard Hughes Medical Institute. S.S.M. was supported in part by the National Institutes of Health Award F32 GM07450601.

of 13. Apel CL,Deamer DW (2005) The formation glycerol monodecanoate bydeydration/ the condensation reaction: Increasing chemical complexity amphiphileson the early of earth. Orig LifeEvol Biosph 35:323-332. vents and associated gradient 14. Baross JA, HoffmanSE (1985) Submarine hydrothermal environments as sites for the origin and evolution of life.Orig Life Evol Biosph 15:327-345. 15. ChakrabartiAC, Breaker RR, JoyceGF, Deamer DW (1994) Production of RNA by a Mol Evol 39:555-559. polymerase proteinencapsulated within phospholipid vesicles. J constantsof 16. KhalilMM (2000) Complexation equilibria and determinationof stability ADP, and ATP) and salicyl binaryand ternarycomplexeswith ribonucleotides (AMP, hydroxamicacid as ligands.JChem Eng Data 45:70-74. 17.Weyant RS, Edmonds P, Swaminathan B (1990) Effect ionic of and nonionic detergents on the Taq polymerase. BioTechniques 9:308-309. Chem 18. Oberholzer T,AlbrizioM, LuisiPL (1995) Polymerasechain reaction in liposomes. Biol 2:677-682. 19. Braun D, Goddard NL, LibchaberA (2003) Exponential DNA replicationby laminar convection. PhysRev Lett 91:158103. 20. KrishnanM, Ugaz VM, BurnsMA (2002) PCR ina Rayleigh-Bernardconvection cell. Science 298:793. 21. Joyce Schwartz GF, AW,Miller SL,Orgel LE (1987)The case foran ancestral genetic system involving simpleanalogues of the nucleotides.ProcNatl Acad Sci USA 84:4398-4402. 22. Chen IA, acid vesicles.Biophys of Szostak JW(2004)A kineticstudy the growthof fatty J87:988-998. 23. Hanczyc MM, Fujikawa SM, Szostak JW (2003) Experimental models of primitative cellular compartments:Encapsulation,growth, and division.Science 302:618-622.

> CC I UJ X u

(/) u wr! > X a. O 2

Mansy and Szostak

PNAS

| September 9,2008

| vol.105

j no. 36

| 13355