Portal motor velocity and internal force resisting viral DNA packaging in bacteriophage phi29

Abstract

During the assembly of many viruses, a powerful molecular motor compacts the genome into a preassembled capsid. Here, we present measurements of viral DNA packaging in bacteriophage phi29 using an improved optical tweezers method that allows DNA translocation to be measured from initiation to completion. This method allowed us to study the previously uncharacterized early stages of packaging and facilitated more accurate measurement of the length of DNA packaged. We measured the motor velocity versus load at near-zero filling and developed a ramped DNA stretching technique that allowed us to measure the velocity versus capsid filling at near-zero load. These measurements reveal that the motor can generate significantly higher velocities and forces than detected previously. Toward the end of packaging, the internal force resisting DNA confinement rises steeply, consistent with the trend predicted by many theoretical models. However, the force rises to a higher magnitude, particularly during the early stages of packaging, than predicted by models that assume coaxial inverse spooling of the DNA. This finding suggests that the DNA is not arranged in that conformation during the early stages of packaging and indicates that internal force is available to drive complete genome ejection in vitro. The maximum force exceeds 100 pN, which is about one-half that predicted to rupture the capsid shell.

FIGURE 1

Schematic illustration of the experimental method. Microspheres carrying prohead-ATPase complexes were captured in one optical trap (lower left sphere) and microspheres carrying DNA molecules were captured in a second trap (upper left sphere). To initiate DNA packaging, the microspheres were brought into near contact for ∼1 s (middle) and then separated to probe for DNA binding and translocation (right). The force acting on the DNA was recorded by measuring the deflection of the laser beam forming the top trap and the bottom trap was translated under computer control by use of an acousto-optic deflector.

FIGURE 2

Measurement of motor velocity at near-zero load. The DNA was held slack most of the time (F < 0.15 pN) and periodically stretched by separating the traps until the force reached 5–6 pN (lower graph) to accurately determine the trap separation (upper graph) and, from this, the DNA tether length.

FIGURE 3

Rate of packaging versus capsid filling. Data was recorded at near-zero load using the ramped stretching method (average over N = 24 complexes). Standard errors on the measurements are approximately equal to the size of the solid circles (∼3 bp/s).

FIGURE 4

(A) Rate of packaging versus total load acting on the motor. Each velocity was determined as an average of N = 6 to 34 complexes (mean N = 17, error bars indicate standard errors). Data for total loads <40 pN (circles) were recorded at 5–10% capsid filling, where the internal force is negligible. Data for total loads >40 pN (squares) were collected at 70–80% capsid filling, where both external and internal load contributed. The contribution of internal load for loads >40 pN was determined from the known filling using the velocity versus filling data (Fig. 3) and low-force velocity versus load data, as described in the text. The solid line is a fit of the data to the theoretical expression in Chemla et al. (34). (B) Average power generated by the motor versus load. Note that 1 zeptowatt = 10⁻²¹ watts ≅ 3 pN·bp/s.

FIGURE 5

Mean internal force versus capsid filling deduced from the results in Figs. 3 (rate versus filling) and 4 A (rate versus load). The inset plot is a magnified view showing the force increase during the early stages of packaging. The error bars were calculated as described in the text.