Taking a strong ‘bottom-up’ approach to the

origins of life:

How far can prebiotically plausible molecules go?

Kepa Ruiz-Mirazo

CERN, Geneva20 May 2011

Dept. of Logic and Philosophy of Science (IAS Research Group)Biophysics Research Unit (CSIC-UPV/EHU)

«The idea that biological organization is fully determined by molecular structures is popular, seductive, potent and true up to a point -- yet

fundamentally wrong... It disregards the fact that the cell as a whole is required

to create the proper environment for self-assembly to proceed.»

[F. Harold 2003]

LIFE: SYSTEM PROPERTY ! → ORIGINS: SYSTEMIC APPROACH!!

CAPACITY FOR ‘SELF-CONSTRUCTION’

(metabolism)

POTENTIAL FOR ‘INDEFINITE GROWTH OF COMPLEXITY’

(‘Darwinian’ evolution)

A universal definition of ‘minimal life’

‘autonomy’ ‘open-ended evolution’

[Ruiz-Mirazo et al. 2004, Origs. Life Evol. Biosph.]

LIFE :

very complex molecules (DNA, RNA, proteins, sugars, lipids,...)

very complex organization (‘genetically-instructed’ cellular metabolisms)

[Reprinted (2010): Anthology -- CUP]

Universal biochemical features

DNAas the genetic material

Genetic code

Energy currenciesHomochirality

Cellular boundary Common coenzymes and metabolic

intermediaries

ATP Synthase

[INSERT: ‘grt video of atp synthase’]

Universal biophysical/biochemical features

SELF-ORGANIZATION & SELF-ASSEMBLY (!)

MAIN BOTTOM-UP PROBLEMS TO BE ADDRESSED:

-MINIMAL CONCENTRATION THRESHOLDS(BOUNDARIES: PROTOCELLULARITY)

- KINETIC CONTROL: ORIGIN AND DEVELOPMENT OF CATALYSIS(OLIGOMERIZATION, ANABOLIC AUTOCATALYSIS, ORGANOCATALYSIS!)

- PROCESS SYNCHRONIZATION + INTEGRATION SUPERSYSTEMS(e.g., chemical oscillations with division cycles)

-CONTROL ON MATTER AND ENERGY FLOW: (TRANSDUCTION + ENDERGONIC-EXERGONIC COUPLINGS)

between boundary (scaffolding)

and protometabolic reactions

‘co-evolution’

(Gánti 1975; 2002)

Infrabiological systems“Proliferating microsphere”

(Szathmáry et al., 2005 )

FORGET (temporarily) ABOUT COMPLEX BIOMOLECULES (e.g.: DNA, RNA, PROTEINS,…):

CAN WE DO INTERESTING ‘SYSTEMS CHEMISTRY’ WITH MUCH SIMPLER (and prebiotically plausible) MOLECULES ?

(e.g., fatty acids, amphiphiles/surfactants, alcohols, aminoacids, short peptides...)

‘basic autonomous’ systems

‘OLIGOMER (peptides)WORLD’

‘hereditary autonomous’ systems

‘ONE-POLYMER (RNA) WORLD’

minimal living systems(autonomy + open-ended evolution):

‘TWO/THREE-POLYMER WORLD’(RNA-protein/DNA-RNA-protein)

second major ‘bottleneck’: ‘template-replication’ mechanisms

third major bottleneck: phenotype-genotype decoupling(catalysis /// template activity)

‘translation’ mechanisms and genetic code

!

!

first major bottleneck: ‘proto-bioenergetic’ mechanisms!

INCREASE IN MOLECULAR AND ORGANIZATIONAL

COMPLEXITY

ORIGINS OF LIFE

FUNCTION

INFORMATION

[Ruiz-Mirazo et al. (2004) OLEB 34: 323-346 ]

If so, apart from self-organisation and self-assembly…

Are there any other “driving forces”?

…when proper Darwinian evolution (i.e., involving replication of modular templates)

is still on its way…

NATURE 2008 Jul 3; 454:122-5.

(Pre-biopolymer) scenario with:

• SELF-ASSEMBLING VESICLES made of fatty acids, amphiphiles/surfactants, alcohols, mixtures,... evidence from: (a) external sources [Deamer 1986, 1997; Dworkin et al. 2001]

(b) abiotic (Fischer-Tropsch) synthesis [Nooner et al. 1976; Allen & Ponnamperuma 1967; Rushdi & Simoneit 2001]

• SHORT PEPTIDE CHAINS (rudimentary channels/carriers and catalysts) made of: Ala, Gly, Asp, Glu, Ser, Val… evidence from: (a) external sources [Pizzarello et al. 2006; Bernstein et al. 2002]

(b) abiotic (Strecker, SIPF,… ) synthesis [Miller 1953; Rode 1999]

• VARIOUS ‘COENZYME-LIKE’ COMPOUNDS (e- carriers, pigments...)

• PAHs: PHOTOCHEMICALLY ACTIVE and MEMBRANE STABILIZING!

• PRIMITIVE ENERGY TRANSDUCTION MECHANISMS ? (‘chemical and chemiosmotic’ -- energy currency precursors)

BOTTOM-UP APROACH: MINIMAL ‘lipid-peptide’ CELL

+

PRODUCTION OFMOLEC. COMPLEXITY (e.g., POLYPEPTIDES)

DEVELOPMENT OFLIPIDIC

COMPARTMENTS

• avoid diffusion• adequate scaffolding to anchor

transduction mechanisms • catalytic effect (hydrophobic phase)

• osmotic control/regulation • accessibility of simple molecules

• constructive use of conc. gradients

Why postpone the appearance of compartments when they seem to be pivotal for the material-energetic

implementation of a complex reaction system ??(+later on: only makes integration problems worse!)

‘COMPARTIMENTALIST VIEW’: Morowitz, Deamer, Luisi, Szostak, Monnard...

[Skulachev, V.P. 1992; Harold F. M., 1986]

Luisi (2003) “Autopoiesis: a review and a reappraisal” Naturwissenschaften 90:49-59

‘Autopoietic vesicles’ (Pier Luigi Luisi)

OPEN QUESTIONS (difficulties):

- high cac?

- pH dependence?

- sensitivity to salts?

- capacity to hold gradients?

- fatty acids or isoprenoids?

MAIN SOLUTION to the problems:

continue experimental work

1) using MIXTURES! (of ffaa, alcohols, other surfactants, etc.)

2) trying different CHEMISTRIES (reaction networks)in those ‘messy’, heterogeneous conditions

1.- A flexible (object-oriented/C++) computational environment to simulate the dynamics of chemically reacting cellular systems

(e.g., minimal proto-metabolic cells, biological cells,…)

2.- Stochastic kinetics (Gillespie method): tool to explore all possible dynamic behaviours (including critical transitions at low population numbers,

role of noise, fluctuations,…)

3.- Realistic but not aiming to mimic nature (goal: to inform/complement in vitro models)3a. Not spatially-explicit, but vol/surf constraints calculated from molec. prop.3b. Water molecules not included, but buffering or osmotic effects into account

4.- Heterogeneous conditions (beyond the ‘well-stirred tank flow reactor’ hypothesis)4a. Various reactive domains/phases4b. Cell population dynamics4c. Transport processes between the different reactive domains

(‘molecular diffusion’, ‘gradient diffusion’, ‘aggregation process’,…)

General features of our simulation platform

[crucial point: coupling between transport and internal –or boundary– reaction proc.]

its time evolution is described in terms of the numbers of reactant molecules and requires solving the so-called Master Equation:

P(X, t | X0, t0) dt probability that the system will be in the state X at time t’[t , t+dt], if it was in X0 at

the beginning.

0 0

0 0 0 0

P , t , t = ( - ,t) , , t P , t , t - ( ,t)P , t , t

ta w a

X

X XX X X X X X X X X X X

1, 1 2, 2 , 1, 1 2, 2 ,... ...

1, 2,...

kN N N Nr S r S r S p S p S p S

R

a(X) =

rxxx

kR

r

j,jjj

j,

1ρ 1

N

jρ

N

jρ

1...1

X = ||x1, x2, x3… xN|| (( xj number of molecules of the jth species))k kinetic constant of the th reaction ; = NAV ( scale factor)

Given a homogeneous, well-stirred, chemically reacting system, where N species Sn (n=1,2…N) can be transformed according to R different elementary reactions:

Stochastic Kinetics (I)

that gives the rate of change of the probability density function P(X, t | X0, t0).

,

,1

1N

jj

N

j j j jjr

kx x x r

wa

X X

X

w(∆X |X, t) dt ≡ probability that the system, upon jumping away from state X at time t, will land in state X+∆X.

Where:

a(X, t)dt ≡ probability that the process will jump away from state X in the next infinitesimal time interval [t, t+dt], given X(t)=X.

G’s M: the stochastic time evolution of a reacting system can be seen as sequence of finite time intervals ti, where nothing occurs (dead times), followed by an infinitesimal time interval dt where one of the possible reactions takes place:

1 2 3 i i+1

dti = dead time

1 2 3 i i+1

i = reaction label (1 i R)

t =1 + 2 + … + i + i+1 + …

• This approach allows to study reacting systems with very low molecular populations, as well as micro-dispersed. It is also possible to generalize it to deal with heterogeneous cases (complex organization!).

• The Master Equation is very hard to solve analytically, but Gillespie introduced a Monte Carlo method to simulate the stochastic time evolution of a chemically reacting system.

1

1ln

1τ

na x

1

2

1

1

xxxxx

wanw

It can be shown that by drawing two random number n1 and n2 uniformly distributed in the range [0, 1] the following formulae can be used:

a) to stochastically simulate the dead time:

b) to draw the next reaction:

Stochastic Kinetics (II)

CSystemWater Environment

CSystemChemoton

XCMolecule

CReactorWater Pool

CReactorMembrane

TCMolecule

CFlux XCMolecule

OsmoticPressureBalance

YCMolecule CFlux YCMolecule TCMolecule

CGrowth

A1CMolecule

A2CMolecule A4CMolecule

A3CMolecule

CReaction CReaction

CReaction CReaction

CSystemWater EnvironmentCSystemAqueous Environment

CSystemChemotonCSystemChemoton

XCMolecule XCMolecule

CReactorWater PoolCReactorAqueous Pool

CReactorMembraneCReactorMembrane

TCMolecule TCMolecule

CFlux XCMolecule XCMolecule

OsmoticPressureBalance

YCMolecule YCMolecule CFlux YCMolecule YCMolecule TCMolecule TCMolecule

CGrowth

A1CMolecule A1CMolecule

A2CMolecule A2CMolecule A4CMolecule A4CMolecule

A3CMolecule A3CMolecule

CReaction CReaction

CReaction CReaction

A new object-oriented computational environment to simulate complex chemically reacting systems (minimal proto-metabolic cells)

The object-oriented paradigm allows a one-to-one correspondence between objects in the real world and the abstract objects -or classes- in the code. In this case, the C++ classes CSystem, CReactor, CMolecule CFlux and CReaction cooperate all together to perform the stochastic time evolution of a reacting system (minimal cell model) according to the master equation.

The hierarchical structure of this programming language gives the possibility to build up quite easily increasing levels of complexity in the system (e.g., from a single reactor, in which self-maintenance or growth dynamics can be analysed, to a population of them, in which competitive behaviour and selective evolution could arise).

A new object-oriented computational environment (II)

CSystem Water Environment

XCMolecule

YCMolecule

CSystem Chemoton

CSystem ChemotonCSystem Chemoton

CSystem Chemoton

Balance

BalanceBalance

Balance

CSystem Water EnvironmentCSystemAqueous Environment

XCMolecule XCMolecule

YCMolecule YCMolecule

CSystem ChemotonCSystem Chemoton

CSystem ChemotonCSystem ChemotonCSystem ChemotonCSystem Chemoton

CSystem ChemotonCSystem Chemoton

Balance

BalanceBalance

Balance

1) Realistic diffusion processes (passive transport) across the membrane, considering free flow of water

2) Conditions for division or an eventual ‘osmotic crisis’

so:

3 32 2

/ 236 2 2 9sphere sphere VV S S S V

Overall isotonic condition:

• If:

• When (or before):

3 236S V then: OSMOTIC BURST!

3 22 9S V then: DIVISION!

Our protocell model: main features/assumptions

ExternalInternalspeciesspecies

jiji

TotalA Core A Env

nnC

N V N V

1

2

membranespecies

i ii

S n

internalspecies

ii

Envexternalspecies

jj

nV V

n

23/ 36S V 31 1 2

0.9 1.386( = = 0.1)31 2

Initially spherical pre-division state

[Mavelli & Ruiz-Mirazo: Phil Trans. B (2007) ]

B

B

L LL

kLkL

kLkL B

B

L LLXkLkL

kLkL

kXL

B

B

L LL

Z

Z

Z

kLkL

kLkL

kZ

kZL

kZkZ

2Q

QkQ

Scheme 0

Scheme 2

Scheme 1

Scheme 3

L

W

L

A1

A3

A2A4

XA1

WL

B

B

X

k1

kW

k’1

k’3 k’2

k’4

kX

k3

k4

k2kLkL

kLkL

B

B

L LL

kLkL

kLkLB

B

L LLB

B

L LL

kLkL

kLkL

kLkL

kLkL B

B

L LLXkLkL

kLkL

kXL

B

B

L LLXkLkL

kLkL

kLkL

kLkL

kXL

B

B

L LL

Z

Z

Z

kLkL

kLkL

kZ

kZL

kZkZ

2Q

QkQ

B

B

L LL

Z

Z

Z

kLkL kLkL

kLkL kLkL

kZ

kZL

kZkZ

2Q

QkQ

Scheme 0

Scheme 2

Scheme 1

Scheme 3

L

W

L

A1

A3

A2A4

XA1

WL

B

B

X

k1

kW

k’1

k’3 k’2

k’4

kX

k3

k4

k2

Scheme 3

L

W

L

A1

A3

A2A4

XA1

WL

A1

A3

A2A4

XA1

WL

B

B

X

k1

kW

k’1

k’3 k’2

k’4

kX

k3

k4

k2kLkL

kLkL

[L]eq = 0.004 M ,[ ]L aq L L memk L S k n Eq. condition:, / 2L L memS n

2[ ] L

aqL L

kL

k

kL= 1.0 M-1t-1 ; kL = 0.001t-1

Minimal cell models [Mavelli & Ruiz-Mirazo: Phil Trans. B (2007) ]

‘Empty cell’ dynamics LLLE LC

LDL

BE

DL DL

DL

BC

Spherical membranes with different radius (R) in a pure water solution continuously exchanging lipids L with the internal core and the external environment. As a consequence, the volume fluctuates around the initial spherical value 4/3R3. In fact, these fluctuations bring small structures (R 30) to collapse due to an osmotic crisis.

Cell Volume vs time in a pure water solution

0.0E+00

1.0E+05

2.0E+05

3.0E+05

4.0E+05

5.0E+05

6.0E+05

0.0 100.0 200.0 300.0 400.0

time (a.u.)

Cell

Vo

lum

e n

m3

R = 50 nm

R = 45 nm

R = 40 nm

R = 30 nm

R = 25 nm

R = 35 nm

Stability of spherical membranes with different radius (R) in a pure water solution

8.000E-01

9.000E-01

1.000E+00

1.100E+00

1.200E+00

0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0

time (a.u.)

3.998E-03

3.999E-03

4.000E-03

4.001E-03

4.002E-03

0.0 100.0 200.0 300.0 400.0

time (a.u.)

[L] C

ore

Cell water pool lipid concentration against time for a 50 nm radius membrane: [L]C fluctuates around the equilibrium value 0.004M.

stability factor reported against time for an initial spherical membrane of 30 nm radius. fluctuates around 1. 0 (i.e., the value of a perfect sphere).

In the presence of an osmotic buffer B, the fluctuations of the core volume decrease as the buffer concentration increases and this enlarges size range for cell stability.

The average volume fluctuations are reported against buffer concentration

Osmotic buffer effect on the core volume fluctuations: [B]E=

[B]C=0.005M.

[B] = 0.05

5.8E+04

6.2E+04

6.6E+04

7.0E+04

7.4E+04

0.0 500.0 1000.0 1500.0 2000.0 2500.0 3000.0

Vo

lum

e n

m3

Osmotic Buffer effect on 25nm size Membrane

[B] = 0.005

5.8E+04

6.2E+04

6.6E+04

7.0E+04

7.4E+04

0.0 500.0 1000.0 1500.0 2000.0 2500.0 3000.0

Vo

lum

e n

m3

Osmotic buffer effect on membrane stability

0.0E+00

1.0E+03

2.0E+03

3.0E+03

4.0E+03

5.0E+03

1.0E-04 1.0E-03 1.0E-02 1.0E-01 1.0E+00

Buffer Concentration

Avera

ge V

olu

me F

luctu

ati

on

nm

3

Osmotic buffer effect on the core volume fluctuations: [B]E=

[B]C=0.05M.

aL = 0.5 nm2 DL, = 1.0M-1t-1 vL = 1.0 nm3 DL = 0.001M-1t-1

Reproducing real experimental data: swollen vs. deflated protocell competition dynamics

[Mavelli & Ruiz-Mirazo (2008): BIOCOMP’08 Proceedings]

Parameters Oleic Acid POPC

kin 7.6 103s-1M-1nm-2 7.6 103s-1M-1nm-2

kout 7.6 10-2 s-1 7.6 10-7 s-1

[L]Eq (=1) 6.667 M 2.857 10-10M

0.3 nm2 0.7 nm2

0.6 nm3 1.3 nm3

0.21 0.59 [Chen et al. (2004): Science 305]

Proto-Chemotons with different initial radius

0.8

1.0

1.2

1.4

0.0E+00 2.5E+03 5.0E+03 7.5E+03 1.0E+04

time (a.u.)

30 nm

45 nm

40 nm

60 nm

50 nm

Proto-Chemotons with different capability of throw out waste

0.8

1.0

1.2

1.4

0.0E+00 2.5E+03 5.0E+03 7.5E+03 1.0E+04

time (a.u.)

kW = 0.001

kW = 0.1

kW = 0.01

Dw

Dw

Dw

A critical size was found to overcome an eventual osmotic crisis. As much bigger is the size as higher are the stability and growth rate

The permeability coefficient to waste results to be a fundamental parameter to guarantee the stability of the cell.

[ Mavelli & Ruiz-Mirazo: Phil Trans. B (2007) ]

A ‘proliferating microsphere’? [Ganti T. 1975; 2002]

[Piedrafita et al. 2009 ECAL Proc. ]

Two-lipid membranes: FROM ‘SELF-ASSEMBLY’ TO ‘SELF-PRODUCTION’

[Ruiz-Mirazo et al. 2010 AEMB, Springer Series]

L

A1

A3

A2A

4

XA

1

W

L

B

k1

k

k’ 1

k’ 3 k’2

k’ 4

X

k3

k4

k 2k

L

kL

L

A1

A3

A2A

4

XA

1

W

L

B

k1

k

k’ 1

k’ 3 k’2

k’ 4

X

k3

k4

k 2

L

W

L

A4

L

A 1

A3

A 2

XA 1

W

P

B

B

X

k1

DW

k’1

k’3

k’

k’

DX

k3

k4

k2

k

24

5A5

k5

k’

P

P

PN

Laq L

L LaqL Laq

Laq Lk

kk

DW

P PaqPaq P

Pm

+

n+mP

(1, 2, 3,…N)P

n

kk

1

PPaq Pk

P Paqk

k

PN

1 1 1

P P P Paq aq

P P P Paq aq

L L L Laq aq

L L L Laq aq

k k

in outk k

k k

in outk k

P P P

L L L

Cell Metabolic Network

2,3 3,4 4,5 5,1

3,2 4,3 5,4 1,5

2,1

1,2

1.0

0.1

0.1

1.0

k k k k

k k k k

k

k

3

4

1.0 10

1.0 10

aq aq

aq aq

L L L L

P P P P

k k

k k

1,2

2,1

2,3

3,2

3,4

4,3

4,5

5,4

5,1

1,5

1 2

2 3

3 4

4 5

5 12

k

k

k

k

k

k

k

k

k

k

A X A

A A W

A A P

A A L

A A

Membrane Molecular Uptake and Release

Molecular Transport Processes

40

4

400

X

W

W

D

D

D

100 50P Pk k

Membrane Reactions

,

,

G

B

k n m

n m n mk n mP P P

/

/

0 if 20,

if 20

100 0.1

G BG B

G B

n mk n m

k n m

k k

20P

P

k

kP

( 1)

X

W

W

D

Env Core

D

Env Core

D

Env Core

X X

W W

W W

Modelling minimal self-(re-)producing

‘lipid-peptide’ cells (+ osmotic regulation)

[Ruiz-Mirazo & Mavelli (2007): ECAL Proceedings]

[Ruiz-Mirazo & Mavelli (2008): BioSystems 91(2)]

Chiang et al. 2003

www.dur.ac.uk/.../science/peptide_lipid/pl.html

Aromatic residues of actual membrane proteins:

typically located in the transition zone between the low dielectric lipid interior and the polar lipid head

groups

Thank you!