1: \section{Theoretical Model of DNA Packaging}
2:
3: Our theoretical model is based on earlier work~\citep{riemer78,
4: tzlil03, purohit03, purohit05}, which describes the packaging energy
5: as a function of the length of DNA in the capsid. We model the
6: {\lamphage} capsid as a sphere and its genome as a long semiflexible
7: rod. We
8: assume that the rod is wound into a cylindrically symmetric
9: spool~\citep{cerritelli97} with local hexagonal packing. The total
10: energy of the packaged DNA can then be approximated by a sum of
11: inter-axial repulsion energy and the bending energy of the rod:
12: %
13: \begin{equation}\label{packingenergy}
14: E = E_{\rm interaction} + E_{\rm bend} = \sqrt{3} F_0L(c^2 +
15: cd_s)\exp(-d_s/c) + \pi \kB T\xi
16: \int_{R_{\rm in}}^{R_{\rm out}} {N(r)\over r} dr \,,
17: \end{equation}
18: %
19: where $F_0$ and $c$ are empirically determined constants describing
20: the interaction between neighboring DNA double-helices, $\xi$ is the
21: persistence length of DNA, $L$ is the length of the DNA within the
22: capsid, $d_s$ is the inter-axial spacing, $\Rout$ and $\Rin$ are the
23: radius of the capsid and the inner radius of the DNA spool,
24: respectively, and $N(r)$ is the number of loops of DNA at a distance
25: $r$ from the spool axis. For the persistence length $\xi$ we use
26: 50~nm, though its value in Mg$^{2+}$ buffer may be~$\sim$10\%
27: smaller~\citep{hagerman1988}. The spacing between sequential bases
28: of DNA varies, depending on the base types, from 0.33 to
29: 0.34~$\nm$~\citep{olson1998}. To compute $L$ we disregard this
30: variation and use 0.34~$\nm$ times the number of base pairs within the
31: capsid. The inter-axial forces in buffers containing
32: $\rm Mg^{2+}$ have been measured~\citep{rau84}. Since the values
33: measured for 5~mM and 25~mM $\rm Mg^{2+}$ were not significantly
34: different, we assume that the forces at 10~mM (used in our experiments)
35: will be identical. A least-squares fit to the 5~mM and 25~mM data
36: in~\citep{rau84} gives $F_0 = \hbox{12,000}$~$\pN/\nm^{2}$ and $c =
37: 0.30~\nm$.
38: The radius of the phage capsid $\Rout$ is around $29$
39: nm~\citep{earnshaw77}. Once we know $d_s$, $\Rin$, and $N(r)$,
40: we can use Eq.~\ref{packingenergy} to calculate the
41: internal force
42: on the phage genome as a function of genome length inside the capsid,
43: providing an interpretation of the experimental results.
44:
45: We calculate the remaining variables as a function of $L$ according to
46: the following recipe, which
47: involves only simple geometrical considerations and elementary
48: calculus. The number of loops $N(r)$ in Eq.~\ref{packingenergy}
49: is given by $z(r)/d_s$, where $z(r) = \sqrt{\Rout^2-r^2}$ is the height of the capsid at
50: distance $r$ from the central axis of the DNA spool. The actual volume
51: $V(R_{\rm in}, R_{\rm out})$ occupied by the DNA spool can be related
52: to the genome length $L$ and the inter-axial spacing $d_s$ to get an
53: expression for $R_{\rm in}$ in terms of $d_s$, $R_{\rm out}$ and
54: $L$~\citep{purohit03}. This expression for $R_{\rm in}$ is substituted
55: into Eq.~\ref{packingenergy}, which can then be minimized with
56: respect to $d_s$ to give the equilibrium inter-axial spacing as a
57: function of the genome length $L$ inside the capsid. From $d_s$, $\Rin$,
58: and~$N(r)$,
59: Eq.~\ref{packingenergy} now gives us the total packing energy as
60: a function of genome length inside the capsid. The internal force
61: $F(L)$ acting on the genome is obtained by taking the derivative of
62: Eq.~\ref{packingenergy} with respect to $L$~\citep{purohit03}.
63:
64: The preceding construct is a parameter-free model that predicts the
65: ejection force from a {\lamphage} capsid. Experimental uncertainties
66: in the quantities quoted above should lead to errors of $10$--$50\%$ in
67: the magnitude of the force predicted, but the shape and relative
68: positions of the curves for different genome lengths should not be
69: strongly affected by these errors (these tests of the parameters are
70: not shown).
71:
72: DNA ejection in our experiment or \textit{in vivo}
73: is halted at the point where the internal force balances the
74: osmotic force. We have described above how to obtain the internal
75: \emph{force} (and hence, the external osmotic force because of the
76: equilibrium) acting on the genome. But the experimental variable is an
77: osmotic \emph{pressure}~\citep{tzlil03}. Thus we need to translate this
78: force into a pressure. The force $F(L)$ is given approximately by
79: $\Pi\cdot \pi \RDNA^2$, where $\Pi$ is the osmotic
80: pressure and $\RDNA$ is the effective radius of the DNA. We take
81: $\RDNA$ as 1.0~nm (bare DNA) plus 0.2~nm, half the PEG monomer length
82: found experimentally~\citep{abbot92,marsh04}.
83: This is an exact formula for the osmotic force on a very large area.
84: However, at the PEG concentrations used in our experiment, the
85: diameter of DNA is comparable to the correlation length
86: (mesh size) of PEG, $\sim$1--3~{\nm}~\cite[pp.78--80]{degennes79}, so
87: the formula for $F(L)$ is only approximately valid.
88: A scaling expression for the correction could be
89: used~\citep{devries01,castelnovo03,evilevitch04a},
90: but since this result is good only
91: up to a multiplicative constant, its importance is
92: unclear. Hence, to maintain clarity in our analysis we do not use the
93: correction. On the other hand, including this effect could result in
94: a better fit between theory and experiment.
95: