24
Biomolecules as polymers –at least two of the primary classes of biomolecules are polymers (proteins and DNA/ RNA, often saccharides as well) –physicists have developed a rich set of theories for polymers –How can we use them to gain new biological understanding? DNA packing into bacteria

Biomolecules as polymers - SimBac · of our chain (from Flory1:) = Na2 Therefore Lp = a (the Kuhn length) The random walk model we just developed is also known as the Freely

  • Upload
    others

  • View
    5

  • Download
    0

Embed Size (px)

Citation preview

  • Biomolecules as polymers

    –at least two of the primary classes of biomolecules are polymers (proteins and DNA/RNA, often saccharides as well)

    –physicists have developed a rich set of theories for polymers

    –How can we use them to gain new biological understanding? DNA packing

    into bacteria

  • Do atoms always matter?

    PBP1a

    For understanding the function of an enzyme, we need an atomic-detailed structural description

    The bacterial cell wall (PBP’s substrate) is often best described from a statistical (average) point of view

  • Random walksA simple model of a polymer is a random walk in space

    1D

    3D

    Fixed segments of equal length are connected at hinges, can go in any direction

    length of each segment is defined as the Kuhn length (a)

    PBoC 8.2 DNA as a random walkalso known as the Freely Jointed Chain (FJC) model

  • Measurable properties

    What do we want to measure for these random walks?

    “random” implies only average properties are useful, e.g., , , p(R,N)

    Apply the tools of statistical mechanics to calculate overall length (macrostate) from its underlying microstates

    counting # states (W) for each nr or nlPBoC 8.2.1

  • Probability distributions

    nr0 N

    P(nr;N)

    P (nr;N) =N !

    nr!(N � nr)!(1

    2)N

    What is P(R;N)?

    p(R;N) =2

    2πNe−R

    2/2Na2

    Binomial distribution

    Gaussian distributionR-Na Na

    P(R;N)

    R = (nr - nl)*a end-end distance

  • p(R;N) =2

    2πNe−R

    2/2Na2

    Normalization ( ) implies that

    ∫∞

    −∞

    p(R;N)dR = 1

    P (R;N) =1

    2πNa2e−R2/2Na2

    and in 3D: P (R;N) = (3

    2πNa2)3/2e−3R

    2/2Na2

    Distribution is Gaussian with mean = 0 and variance = Na2

    PBoC 8.2.1

    Probability distributions

  • Central limit theorem

    For a set of N random variables {xn} with finite mean and variance (e.g., the individual segments of a polymer), the sum X = x1+x2+...xN (e.g., the end-to-end distance) will tend towards a Gaussian distribution regardless of the distribution of xn

    P (R;N) =1

    2πNa2e−R2/2Na2

    sum of n dice throws

  • Persistence length (discrete)Persistence length (Lp) is defined microscopically by the correlation between the directions of successive segments of our chain (from Flory1:)

    = Na2

    Therefore Lp = a (the Kuhn length)

    The random walk model we just developed is also known as the Freely Jointed Chain model

    1Flory, Paul J. (1969). Statistical Mechanics of Chain Molecules. New York: Interscience Publishers

    Lp = limN→∞

    ⟨N∑

    j≥i

    a⃗i ·a⃗j

    a⟩

    Lp =⟨R2⟩ + Na2

    2Na

    http://openlibrary.org/b/OL5613440M/Statistical_mechanics_of_chain_molecules

  • Persistence length (continuous)Persistence length (Lp) is defined microscopically by the correlation between the directions of successive segments of our chain:

    Therefore Lp = a/2 (a is the Kuhn length)

    This is the start of the worm-like chain model

    h~t(s) · ~t(u)i = e�|s�u|/Lp

    ~R =

    Z L

    0ds ~t(s)

    h~R2i = hZ L

    0ds ~t(s) ·

    Z L

    0du ~t(u)i

    = 2

    Z L

    0ds

    Z L

    sdu e�(u�s)/Lp ⇡ 2LLp = Na2 = aL

    PBoC 8.2.1

    = 2

    Z L

    0ds

    Z L

    sdu e�(u�s)/Lp ⇡ 2LLp = Na2 = aL

    a = /Rmax in the continuous model

  • Radius of gyration

    hR2Gi =1

    N

    NX

    i=1

    h(~Ri � ~RCM)2iRadius of gyration (RG): average distance between links and center-of-mass

    ~RCM =1

    N

    NX

    i=1

    ~Riq

    hR2Gi =r

    LLp3

    E. coli genome is ~ 4.6x106 base pairsq

    hR2Gi ⇡ 5µm

    About 2x as big as seen in the picture, but it has some constraints (e.g., circular, so it has to return to the cell body often)

    Clearly strong forces are necessary to pack it in the cell!PBoC 8.2.2

  • Entropic force

    Recall that our free energy G = U - TS + PV (Gibbs)

    A change in energy (e.g., from applied tension) generates a restoring force, Frestore = -∂G/∂L

    Even though the chain does not store internal energy, it still can exert a force as an entropic spring

    PBoC 8.3.2

    F = kTL/Na2F = -T(∂S/∂L)

    U, T, P are constant for freely-jointed chain, making

  • Improvements to the FJC model

    FJC model (fixed segments)

    Worm-like chain (elastic, with an energetic bending cost)

    modified WLC, now with a stretching cost as well

    J. Hsin*, J. Strumpfer*, E. H. Lee, and K. Schulten. 


    “Molecular origin of the hierarchical elasticity of titin: simulation, experiment and theory.” Annual Review of Biophysics, 40:187-203 (2011).

  • Worm-like chain (WLC) modelOur polymer is now characterized by a unit tangent vector t(s), where s is the position along the chain

    How much energy does it cost to bend the chain?

    PBoC 10.2.1

    W (✏) =1

    2E✏2 =

    1

    2E

    ✓�L

    L0

    ◆2

    AAACQ3icbZBPaxNBGMZnq/2XVk316OXFUEgvYXdbsJdCqQoeeqhgmmI2DbOTd5OhszPLzLuFsOS7efELePML9NKDUrwKTtIVte0DAw/P732ZmSctlHQUht+CpUePl1dW19YbG5tPnj5rbj0/daa0ArvCKGPPUu5QSY1dkqTwrLDI81RhL714M+e9S7ROGv2RpgUOcj7WMpOCk4+GzU+9doKFk8roHTiAKkqMH4d4Bu/+5OfxAs BfAonCjNpQJW9REYfjmh0PQ5glVo4ntHMeD5utsBMuBPdNVJsWq3UybH5NRkaUOWoSijvXj8KCBhW3JIXCWSMpHRZcXPAx9r3VPEc3qBYdzGDbJyPIjPVHEyzSfzcqnjs3zVM/mXOauLtsHj7E+iVl+4NK6qIk1OL2oqxUQAbmhcJIWhSkpt5wYaV/K4gJt1yQr73hS4jufvm+OY070W4n/rDXOjyq61hjL9kr1mYRe80O2Xt2wrpMsM/sin1nP4IvwXVwE/y8HV0K6p0X7D8Fv34DmDetwA==

    E is Young’s modulus (units of force/area) ϵ is strain; W(ϵ) is strain energy density (energy/volume)

    neutral axis

    One way of visualizing the strain

  • Worm-like chain (WLC) modelPBoC 10.2.1

    θ = s/(R+z) = L0/R

    ΔL = s - L0 = (R+z)*L0/R - L0 = (z/R)*L0

    W (✏) =1

    2E

    ✓�L

    L0

    ◆2=

    1

    2E⇣ zR

    ⌘2

    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

    Eb =

    ZW (✏) dV = L0

    ZE

    2R2z2 dA

    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

    =EL02R2

    Zz2 dA =

    EIL02R2

    AAACKXicbVDLSsNAFJ34rPVVdelmsAiuShIF3QhVERRcVLEPaNoymU7aoZNMmJkINcTPceOvuFFQ1K0/4qTNQlsPDBzOOZe597gho1KZ5qcxMzs3v7CYW8ovr6yurRc2NmuSRwKTKuaMi4aLJGE0IFVFFSONUBDku4zU3cFZ6tfviJCUB7dqGJKWj3oB9ShGSkudQvkYxudXHRM6XMdgbN+07SSBDg0UvG/bD7EjfNhNTmCag5dwKtopFM2SOQKcJlZGiiBDpVN4dbocRz4JFGZIyqZlhqoVI6EoZiTJO5EkIcID1CNNTQPkE9mKR5cmcFcrXehxoZ/ecKT+noiRL+XQd3XSR6ovJ71U/M9rRso7asU0CCNFAjz+yIsYVBymtcEuFQQrNtQEYUH1rhD3kUBY6XLzugRr8uRpUrNL1n7Jvj4olk+zOnJgG+yAPWCBQ1AGF6ACqgCDR/AM3sC78WS8GB/G1zg6Y2QzW+APjO8f1k6j8Q==

    I is called the “geometric moment” (don’t have to calculate it!)

    For a circle? Eb =EIL02R2

    =EI(2⇡R)

    2R2=

    ⇡EI

    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

  • Persistence lengthThe WLC model represents a balance between internal energy (from the resistance to bending) and entropy

    How does Lp vary with temperature?

    PBoC 10.2.2

    R θ

    s Eb =EIL

    2R2=

    EIL

    2

    ✓✓2

    s2

    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

    Let L = s for a short segment

    =EI

    2s✓2

    AAACA3icbVDLSsNAFJ3UV62vqDvdDBbBVUmqoBuhKILuKtgHNLFMppN26OTBzI1QQsCNv+LGhSJu/Ql3/o3TNgttPTBwOOdc7tzjxYIrsKxvo7CwuLS8Ulwtra1vbG6Z2ztNFSWSsgaNRCTbHlFM8JA1gINg7VgyEniCtbzh5dhvPTCpeBTewShmbkD6Ifc5JaClrrl3jtOrG+xEOoTTqsoy7MCAAbmvds2yVbEmwPPEzkkZ5ah3zS+nF9EkYCFQQZTq2FYMbkokcCpYVnISxWJCh6TPOpqGJGDKTSc3ZPhQKz3sR1K/EPBE/T2RkkCpUeDpZEBgoGa9sfif10nAP3NTHsYJsJBOF/mJwBDhcSG4xyWjIEaaECq5/iumAyIJBV1bSZdgz548T5rVin1cqd6elGsXeR1FtI8O0BGy0SmqoWtURw1E0SN6Rq/ozXgyXox342MaLRj5zC76A+PzB04Jlq0=

    s = ✓RAAAB83icbVBNS8NAEN34WetX1aOXxSJ4KkkV9CIUvXisYj+gCWWznbRLN5uwOxFK6d/w4kERr/4Zb/4bt20O2vpg4PHeDDPzwlQKg6777aysrq1vbBa2its7u3v7pYPDpkkyzaHBE5nodsgMSKGggQIltFMNLA4ltMLh7dRvPYE2IlGPOEohiFlfiUhwhlbyDb2mPg4AGX3olspuxZ2BLhMvJ2WSo94tffm9hGcxKOSSGdPx3BSDMdMouIRJ0c8MpIwPWR86lioWgwnGs5sn9NQqPRol2pZCOlN/T4xZbMwoDm1nzHBgFr2p+J/XyTC6CsZCpRmC4vNFUSYpJnQaAO0JDRzlyBLGtbC3Uj5gmnG0MRVtCN7iy8ukWa1455Xq/UW5dpPHUSDH5IScEY9ckhq5I3XSIJyk5Jm8kjcnc16cd+dj3rri5DNH5A+czx+qBpDK

  • Persistence length

    Lp increases with increasing Young’s modulus (stiffer) Lp decreases with increasing temperature (more flexible)

    PBoC 10.2.2

    h~t(s) · ~t(u)i = e�|s�u|/Lp ! hcos(✓)i = e�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

    Assume t(u) = t(0) = z axishcos(✓)i = e�s/Lp

    AAACEnicbVA9SwNBEN2L3/ErammzGISkMN5FQRtBtLGwUDAxkIthbzNJluztHbtzQjjyG2z8KzYWitha2flv3HwUGn0w8Hhvhpl5QSyFQdf9cjIzs3PzC4tL2eWV1bX13MZm1USJ5lDhkYx0LWAGpFBQQYESarEGFgYSboPe+dC/vQdtRKRusB9DI2QdJdqCM7RSM1f0JVMdCdTnkSn42AVkRerrsXhC4S7dM/uXzXjQzOXdkjsC/Uu8CcmTCa6auU+/FfEkBIVcMmPqnhtjI2UaBZcwyPqJgZjxHutA3VLFQjCNdPTSgO5apUXbkbalkI7UnxMpC43ph4HtDBl2zbQ3FP/z6gm2jxupUHGCoPh4UTuRFCM6zIe2hAaOsm8J41rYWynvMs042hSzNgRv+uW/pFoueQel8vVh/vRsEsci2SY7pEA8ckROyQW5IhXCyQN5Ii/k1Xl0np03533cmnEmM1vkF5yPb3avnLM=

    hcos(✓)i ⇡ h1� ✓2

    2i = 1� 1

    2h✓2i

    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

    small angle approximation

    hEbi =EI

    2sh✓2i

    AAACJ3icbVDLSgMxFM34tr6qLt0Ei+CqzFRBN4ooBd1VsFXo1JJJb9vQTGZI7ghl6N+48VfcCCqiS//E9AXaeiBwOOdcbu4JYikMuu6XMzM7N7+wuLScWVldW9/Ibm5VTJRoDmUeyUjfBcyAFArKKFDCXayBhYGE26Bz0fdvH0AbEakb7MZQC1lLiabgDK1Uz576kqmWBFqsB9TXQ35C0+IV9SM7SNOC6fXoOOVjG5DdF8bRejbn5t0B6DTxRiRHRijVs69+I+JJCAq5ZMZUPTfGWso0Ci6hl/ETAzHjHdaCqqWKhWBq6eDOHt2zSoM2I22fQjpQf0+kLDSmGwY2GTJsm0mvL/7nVRNsHtdSoeIEQfHhomYiKUa0XxptCA0cZdcSxrWwf6W8zTTjaKvN2BK8yZOnSaWQ9w7yhevD3Nn5qI4lskN2yT7xyBE5I5ekRMqEk0fyTN7Iu/PkvDgfzucwOuOMZrbJHzjfP0/spQM=

    = 2

    ✓kT

    2

    ◆= kT

    AAACDXicbVC7SgNBFJ31GeNr1dJmMAqxCburoE0gaGMZIS/ILmF2MpsMmX0wc1cIS37Axl+xsVDE1t7Ov3GSbKGJBwYO55zLnXv8RHAFlvVtrKyurW9sFraK2zu7e/vmwWFLxamkrEljEcuOTxQTPGJN4CBYJ5GMhL5gbX90O/XbD0wqHkcNGCfMC8kg4gGnBLTUM0+r2MGuYAGUcTZqYDfWaexMsCv5YAjnuIpHjZ5ZsirWDHiZ2DkpoRz1nvnl9mOahiwCKohSXdtKwMuIBE4FmxTdVLGE0BEZsK6mEQmZ8rLZNRN8ppU+DmKpXwR4pv6eyEio1Dj0dTIkMFSL3lT8z+umEFx7GY+SFFhE54uCVGCI8bQa3OeSURBjTQiVXP8V0yGRhIIusKhLsBdPXiYtp2JfVJz7y1LtJq+jgI7RCSojG12hGrpDddREFD2iZ/SK3own48V4Nz7m0RUjnzlCf2B8/gCIdJlV

    2 DoF (one constrained by total length)

    e�s/Lp ⇡ 1� s/LpAAACBXicbVC7TsMwFHXKq5RXgBEGiwqJpSUpSDBWsDAwFIk+pDZEjuu2Vh3Hsh1EFXVh4VdYGECIlX9g429w0wzQcqQrHZ1zr+69JxCMKu0431ZuYXFpeSW/Wlhb39jcsrd3GiqKJSZ1HLFItgKkCKOc1DXVjLSEJCgMGGkGw8uJ37wnUtGI3+qRIF6I+pz2KEbaSL69T+6Skjq+9sUYdpAQMnqALizBVPLtolN2UsB54makCDLUfPur041wHBKuMUNKtV1HaC9BUlPMyLjQiRURCA9Rn7QN5SgkykvSL8bw0Chd2IukKa5hqv6eSFCo1CgMTGeI9EDNehPxP68d6965l1AuYk04ni7qxQzqCE4igV0qCdZsZAjCkppbIR4gibA2wRVMCO7sy/OkUSm7J+XKzWmxepHFkQd74AAcARecgSq4AjVQBxg8gmfwCt6sJ+vFerc+pq05K5vZBX9gff4ACu+W/Q==

    1� s/Lp = 1�1

    2h✓2i = 1� 1

    2

    ✓2skT

    EI

    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

    Lp =EI

    kTAAAB/nicbVDLSgMxFM3UV62vUXHlJlgEV2WmCroRiiIouKjQF7TDkEkzbWgmGZKMUIYBf8WNC0Xc+h3u/BvTdhbaeiBwOOdc7s0JYkaVdpxvq7C0vLK6VlwvbWxube/Yu3stJRKJSRMLJmQnQIowyklTU81IJ5YERQEj7WB0PfHbj0QqKnhDj2PiRWjAaUgx0kby7YN7P4aXML25gz1hgjAdNbLMt8tOxZkCLhI3J2WQo+7bX72+wElEuMYMKdV1nVh7KZKaYkayUi9RJEZ4hAakayhHEVFeOj0/g8dG6cNQSPO4hlP190SKIqXGUWCSEdJDNe9NxP+8bqLDCy+lPE404Xi2KEwY1AJOuoB9KgnWbGwIwpKaWyEeIomwNo2VTAnu/JcXSatacU8r1Yezcu0qr6MIDsEROAEuOAc1cAvqoAkwSMEzeAVv1pP1Yr1bH7Nowcpn9sEfWJ8/r6yUrg==

  • Force-extension relationship for WLC

    Ebend =EI

    2

    ∫ L0

    ds

    R(s)2

    partition function is summed over all possible curves (a path integral)

    applying a force adds a term to the energy: Eapp. = −Fz = −F

    ∫ L0

    tzds

    ⟨z⟩ =1

    Z(F )

    ∫Dt(s)z[e−(Ebend+Eapp(F ))/kT ]

    = kTd lnZ(F )

    dFPBoC 10.2.3

    =kTLp

    2

    ∫ L0

    |dt

    ds|ds2

    Z =

    ∫Dt(s)e−Ebend/kT =

    ∫Dt(s) exp(−

    Lp

    2

    ∫ L0

    |dt

    ds|ds)2

    Good luck calculating this!

    don’t assume R is constant

  • Force-extension relationshipA closed form solution for does not exist!

    Marko, J. F. and Siggia, E. D. (1995). Stretching DNA. Macromolecules, 28, 8759–8770.

    FLpkT

    z

    L+

    1

    4(1 − z/L)2−

    1

    4interpolation formula

    PBoC 10.2.3

    deviates by up to 10% from numerical solution (not shown)

  • Force spectroscopy

    DNA RNA protein

    • different force profiles are like molecular signatures

    PBoC 8.3.1

    • deviations from WLC imply a change in structure

  • Single-molecule techniques

    AFM optical tweezers

    magnetic tweezers

    pipette-based

    PBoC 8.3

  • Atomic-force microscopy

    Elias M Puchner, Hermann E Gaub, (2009) Force and function: probing proteins with AFM-based force spectroscopy, Current Opinion in Structural Biology, 19: 605-614.

    unfolding of titin kinase + Ig domains

    fits from WLC model

  • 20-ns SMD Simulation of fibrinogen, 1.06 million atoms

    A Blood Clot Red blood cells within a network of fibrin, composed of polymerized fibrinogen molecules.

    B. Lim, E. Lee, M. Sotomayor, and K. Schulten. Molecular basis of fibrin clot elasticity. Structure, 16:449-459, 2008.

    Atomic-force microscopy

  • Rico, Gonzalez, Casuso, Puig-Vidal, Scheuring. High-Speed Force Spectroscopy Unfolds Titin at the Velocity of Molecular Dynamics Simulations. Science, 342:741, 2013.

    High-speed AFM show agreement with (relatively) slow simulations

    Atomic-force microscopy

    .01 Å/ns

  • optical tweezers

    gradient induces lateral force

    gradient induces axial force

    unfocused laser focused laser

    PBoC 4.3.1