Variation in FtsZ measurement results: dependence on dissolved air and mechanical agitation.

It has been noted that there can be considerable variation between batches of FtsZ (Mukherjee and Lutkenhaus 1998; Dafforn 2008), and also that there can be considerable variation in results from one day to the next (Pacheco-Gomez 2008).

Both these effects were experience during the current series of experiments on FtsZ. Good results were obtained in the very first measurements, but variation from day to day was observed with there being some days when consistently poor results were obtained and where there were even difficulties obtaining stable baseline light scattering values prior to initiating protofilament formation by adding GTP (4.3). There was also significant difference between the behaviour of the two batches of FtsZ (4.1). Possible reasons for the difference are that batch A was refrigerated for approximately 5 days before freezing at –80 °C whereas batch B was frozen within a day. Measurement of protein concentration using BioRad assay (Bradford 1976) showed that batch B contained twice the quantity of protein contained in Batch A. A final potential difference is that it has been shown previously that this protocol results in a significant quantity of GDP co-purifying with the FtsZ (Mukherjee and Lutkenhaus 1998) and a significant quantity of nucleotide was observed coming off the ion-exchange column at similar salt concentrations to the FtsZ during purification. The choice of elutes will determine the relative GDP concentration, which is known to effect FtsZ polymerisation (Small and Addinall 2003).

Analysis by AUC of Batch A (4.2) indicates that it is largely present in a monomeric form. There is a small amount of FtsZ in a polymer state, with tetramers appearing to be a particularly stable configuration. It has been shown previously that tetramers are a particularly stable configuration when polymerisation is initiated by GTP or Ca²⁺ (Sossong, Brigham-Burke et al. 1999). Unfortunately no AUC measurements were made on batch B, which might have provided additional data to help understand the difference between the two batches.

Experience preparing FtsZ solution had already shown that FtsZ solution was sensitive to mechanical stress, in that vigorous pipetting alone was able to increase the amount of light scattering shown, and sonication for 20-30 seconds was sufficient to modify the solution such that light scattering measurements were off the scale of the fluorimeter (4.5). This appears not have been previously documented, but was a characteristic of all the FtsZ batches tested.

Implications for LD measurements

Consequently an important control for any LD measurement would be to monitor the effect of the mechanical shearing that is central to performing LD measurements. Light scattering measurements on FtsZ solution before and after LD measurements using both the conventional cell with a central quartz rod and the new nested capillary showed that the mechanical stress alone was able to induce a permanent increase in the measured light scattering that was at least 10 times greater than that induced by the addition of GTP (4.4)

The light scattering results suggest that mechanical agitation is able to induce the monomers into forming an aggregate that produces a strong light scattering result. However, the change in the state of the FtsZ solution does not however appear to produce any significant LD signal in the range 210 to 350 nm. Possible explanations include that the aggregation does not produce structures that align themselves with the cuvette laminar flow, perhaps because they are not based on FtsZ protofilaments, or that the protofilaments form complex structures with no overall alignment, such as those seen using electron microscopy (4.5) of the FtsZ solution where increased light scattering was induced by sonication.

Previous experiments have obtained LD measurements for FtsZ protofilaments (Marrington, Small et al. 2004), so it was necessary to determine how these results relate to each other. Earlier LD measurements were performed by inducing protofilament formation with GTP and then placing the solution into the Couette cell, by which time it is believed that all of the FtsZ will have been sequestered into protofilaments. The similarity between LD and light scattering results indicate that the protofilaments are mechanically strong enough to withstand the shear stresses of the cuvette flow.

There is significant evidence of dynamic interchange of FtsZ between free monomers and protofilaments. The evidence presented suggests that there are two potential destinations for free monomers within the cuvette cell in the presence of GTP: the first is to rejoin the protofilaments, and the second is to form the stable structures which do not have an LD signature. While there is a large relative concentration of FtsZ protofilaments this will increase the likely-hood of monomers rejoining protofilaments. It is only as the proportion of protofilaments drops as the supply of GTP runs out that the monomer concentration will increase, increasing the probability of forming the alternative stable structures.

The nested capillary cell has been developed to allow the addition of GTP to the monomers within the cuvette cell such that LD can be used to monitor the initial protofilament formation. These new data suggest that there would then be two alternative structures that the FtsZ can form in this situation: the GTP induced protofilaments and the mechanically induced stable structures. The competition between the two pathways may result in substantially different kinetics from those obtained previously, making it difficult to compare the results.

LD measurements of FtsZ

LD measurements using the new nested capillary cell (4.6) were able to reproduce the LD spectral obtained using the conventional cell and also results that have been obtained previously (Pacheco-Gómez 2008). It was also possible to produce LD kinetic results that closely mirror those obtained using light scattering. This provides a degree of validation of the use of the new nested capillary cell.

There was however a factor of two difference in the amplitude of the plateau in conditions that should have resulted in an identical trace. An inability to reproduce results would make comparative studies difficult using the new cell.

In the second run (Figure 18) the LD signal took over 60 seconds to reach half the plateau amplitude and then jumped to the full value after a very large spike in the LD signal. There is a significant problem introducing the inner capillary into the outer without trapping an air bubble because of the small gap between the two capillaries. This result could have arisen because of such an air bubble, initially impeding the mixing of the GTP with the FtsZ, until it finally moved, causing the large spike in the LD signal and allowing the GTP to mix fully.