Protein Structure and Small Molecule Interactions Introduction

Protein-small molecule interactions and the structural changes associated with binding are extremely important. Such interactions are of key importance to the pharmaceutical industry, whose main concern is the mediation of protein function by small (drug) molecules. Protein-small molecule interactions have been investigated using optical sensor technologies, sometimes referred to as biosensors (Ramsey 1998). However, the majority of this work has been undertaken by immobilising the small molecule to the surface of the sensor and then measuring the binding of the receptor to the surface. There are several advantages to undertaking experimentation in which the receptor is bound to the sensor surface, including reduced consumption of receptor, enabling quantitative assessment of the amount of small molecule that binds to the receptor and ultimately allowing the high throughput screening of pharmaceutical "lead" compounds. There have been reports of the measurement of small-molecule binding using various SPR techniques for a limited number of systems (Davis and Wilson 2001; Frostell-Karlsson et al. 2000; Karlsson 1994; Karlsson et al. 2000). For the purposes of illustration, the model system, streptavidin - biotin, will be discussed here; this system has been reported in detail elsewhere (Swann et al. 2004).

Biotin Streptavidin

The surface immobilisation data shown in Fig. 8 shows that the layer thickness after immobilisation of the streptavidin is observed to increase by 6.63 nm. This is in good agreement with x-ray crystallographic structure data that suggests that the long axis of the apo-core-streptavidin molecule at pH 7.5 is around 6.8 nm (Weber et al. 1989) and the dimensions are consistent with the formation of a monolayer of immobilised protein on the sensor surface. The mass loading of the sensor surface enables calculation of the area per molecule surface coverage, which provides an additional indication of the likelihood of a monolayer having been formed. The calculated value of 4,350 A2 molecule-1 is greater than the theoretical value of 3,000 A2 molecule-1 for a fully saturated surface layer, which further suggests that a monolayer has been achieved. The detail of the experiment is shown in Fig. 9, the binding of D-biotin to streptavidin being clearly observed. Figure 9A shows the data, resolved into thickness and density for the binding event. On binding D-biotin, the streptavidin adlayer thickness decreases a small but measurable amount. This is to be expected as streptavidin is a relatively rigid globular protein. There is a commensurate increase in the density of the layer during the binding process. It is also

10

8 -

■D

0

HI

in

n .c

4

O-

2 -

n -

1

--

c

Transverse Magnetic Phase -Transverse Electric Phase

Streptavidin Film Thickness

Vifesh

Blolin Injection Sequences

21 18

15 e

4000

0T 1000 2000 3000 Streptavidin Injection Time (s)

Fig. 8. Immobilisation of streptavidin to a biotinylated sensor surface

5000

6000

21 18

15 e

4000

0T 1000 2000 3000 Streptavidin Injection Time (s)

Fig. 8. Immobilisation of streptavidin to a biotinylated sensor surface

5000

6000

Fig. 9. Detail of the binding of free D-biotin to the streptavidin surface. A Raw phase data. B Resolved thickness and density over the same time period shown in A (Swann et al. 2004)

possible to calculate the mass of protein immobilised on the surface, the consequent mass of D-biotin bound and from the two, the stoichiometry of the interaction. Assuming 100% activity for the immobilised protein, the ratio of D-biotin binding to streptavidin was 2.1:1 which is extremely close to the 2:1 stoichiometry which would be expected if 2 of the available binding sites had been taken up in the surface immobilisation step.

It is generally accepted that when streptavidin binds to a biotin function-alised surface, two (of the four available) binding sites are occupied by surface immobilised species leaving two available for subsequent interactions (Jung et al. 2000). On binding D-biotin it is generally accepted that strepta-

vidin contracts. Studies by (Weber et al. 1989) suggest that a reduction in adlayer thickness of between 0.1 and 0.4 nm when comparing free strep-tavidin to the tetrameric adduct might be expected, depending upon the orientation of the protein. Data obtained using DPI to measure changes in a layer of bound streptavidin as it binds free D-biotin at pH 7.4 (Fig. 9), are in good agreement with the x-ray crystallographic evidence.

It is interesting to note from Figs. 8 and 9 that there is an apparent loss of material after the surface has been challenged with D-biotin (measured as phase shifts in radians for both the transverse electric, TE, and transverse magnetic, TM polarisations) on returning the solution phase to phosphate-buffered saline. This approximates to a loss, during the wash phase, of approximately 8 fg mm-2 s-1. This process is not the reverse of the binding event. This might be attributed to the displacement of streptavidin from the surface as a consequence of the higher affinity of streptavidin for free D-biotin than for surface-bound D-biotin, as has been observed by Jung et al. (2000). The thickness and density profiles during the rinsing process support this hypothesis. The thickness during rinsing remains constant, whilst the density reduces very modestly implying that a small number of streptavidin molecules are removed during the rinse.

Was this article helpful?

0 0

Post a comment