Single-molecule force spectroscopy (SMFS) measurements are invaluable techniques which have gained increasing acceptability in recent times (Bustamante et al, 2000 Clausen-Schaumann et al, 2000 Lavery et al, 2002). Commonly used to quantify the forces between covalent bonds, nucleic acids and receptor-ligand pairs, SMFS seeks to determine the magnitude, in piconewtons, of forces which hold single molecules together (Albrecht et al, 2003 p.367 Moy, Florin  Gaub, 1994). The information obtained from the measurement of these forces can then be used to determine the functional characteristics of the molecules under study as well as conclusively map the inherent binding features. In brief, conventional SMFS entails the disruption of intramolecular or intermolecular bonds in single molecules using rupture force followed by measurement of this force using a cantilever spring. Meaurement using the cantilever spring is usually applied when the atomic force microscopy (AFM) technique is used. Measurement of the force can also be done using beads which are packed in magnetic or optical traps (Oesterhelt et al, 2000 p.143-6 Hugel et al, 2002).
In a bid to optimize the utility of this method, 2 different improvised SMFS techniques have been described. The first of these methods is the high-throughput single-molecule force spectroscopy (HT-SMFS) for membrane proteins which was described by Bosshart et al (2008). The second method, a Programmable Force Sensor technique, was described by Albrecht et al (2003).
This essay evaluates the 2 SMFS techniques in detail. Specifically, the principles of the 2 methods, their utility and uses are discussed. In particular, the 2 methods are compared and contrasted and their relative strenghts and weaknesses delineated. The conclusion is that the HT-SMFS is a more useful technique that the programmable force sensor method.
The conventional SMFS Method
The conventional SMFS technique is usually carried out in several stages. The first stage of SMFS involves the localization of the membranes which have the integral protein that is being studied. This is done through atomic force microscopy (AFM) imaging.  Thereafter, the AFM tip is inserted into the membrane protein then removed after some time. Strong attachment to the cantilever by the protein is resolved when the tip is removed. This resolution is termed as unfolding. Insertion and removal of the tip enables the computation of force-distance (FD) curves. The next stage entails the processing of data and this is done in 4 steps. First, the F-D curves which reprrsent the unfolding statistics are obtained by coarse filtering (Bosshart et al, 2008 p.208).
According to Bosshart et al (2008 p.208), the operator who is tasked with coarse filtring may ignore some groups of force spectra due to subjectivity. In order to prevent such subjective biases from interfering with the outcomes, automated collection and ssaving techniques are usually used. Secondly, the curves are classified and thridly they are aligned. The final step entails the assessment of the curves and this is done with reference to known polymer chain models. Bosshart et al (2008) describe a high throughput SMFS (HT-SMFS) technique which is an improved version of the conventional method. This HT-SFMS technique is described in the next section.
The High-Throughput Single-Molecule Force Spectroscopy (HT-SMFS) Method
First described by Bosshart et al (2008), this technique is used to study the mechanical features, unfolding pathways, energy landscapes and intramolecular and intermolecular forces of membrane proteins. To validate the utility of the technique, Bosshart et al (2008) utilized the proton pump of bacteriorhodopsin (BR) derived from the Haliobacterium salinarum and the L-arginineagmatine antiporter AdiC derived from E. Coli. The former was used as a control while the latter was used as the test substance. For AdiC, 2 types of forces were measuerd and these forces corresponded to detachment from the N terminal and detachment from the C terminal respectively. Since 2 different recombinant AdiC forms were utilized, it was possible to assign the forces to their corresponding termini. Since the N terminus is supposed to be nearly 2.6 times longer than the C terminus and the pH favoured the attachment of the postive N terminal to the engative silicon nitride tip, the possibility of the N terminal attaching to the cantilever is higher than that of the C terminus (Bosshart et al, 2008 p. 214).
Interactions between AdiC and its substrates was assessed by logging in data sets in the presence and absence of agmatine, L-arginine, and D-arginine. The net result was that  6 data sets were generated from nearly 400,000 F-D curves. The bacteriorhodopsin control was the exception as all the other recordings had 200 spectra while it had nearly 400 curves (Bosshart et al, 2008).
The experiment was done using a commercial AFM instrument from JPK Industries and silicon nitride cantilevers. For the cantilevers, the thermal noise method was used to ascertain the spring constant. Adsorption of the bacteriorhodopsin and AdiC particles was follwed by several rinses to get rid of the unattached particles and incubation in specially prepared buffer solutions. The buffer solutions were made up of 150 mM KC1,20 mM Tris-HCl pH 7.8 (BR) and 150 mM NaCl, 20 mM citric acid pH 5.0 (AdiC). The contact mode AFM was used to localize the AdiC-containing proteoliposomes. The topmost layers of collapsed esicles were removed using the AFM tip as neeed arose and spectroscopy performed using the NanoWizard II Ultra AFM instrument  (Bosshart et al, 2008 p.210).
A 150-300mm long point grid was layered on 2D bacteriorhodopsin or on tightly packed AdiC membranes. The density of the grid used was 0.125nm-1. A total of 10 consecutive measurements were done for every grid point and the AFM tip joined to the membrane proteins. The joining was accomplished by exerting a force of 1 nN for a time of 0.1-0.6 seconds on the cantilever 10 times and directed towards the membrane at each point of the grid. Retraction of the cantilever then followed and this was done with a velocity of 0.53 ms-1 for 0.25 seconds. Thereafter, the F-D curves that comprised of 4096 data points were saved and coarse filtering and data analysis carried out. Analysis of the curves was done using the retraction data only. During filtering, the positive spectral forces were indicative of pulling while negative ones were indicative of pushing. The main aim of filtering, as implied before, was to identify the unfolding events. A computer with a processing speed of 2.16GhZ and a random access memory (RAM) of 2 GB was used to filter the data using the IGOR Pro software (Bosshart et al, 2008 p.210).
Negative forces were ignored during the analysis of the data. To get rid of curves which had indefinite unfolding patterns and which had escaped the filtering process, Basset et al (2008) further carried out manual fine filtering followed by classification. This process was done with reference to the definitive AdiC patterns. Collection of data was not stopped until recordings of H 200 unfolding patterns of AdiC that could be classified were obtained (Bosshart et al, 2008 p.211).
As the results show, the N-Hisg-AdiC forces had no significant effect on the presence or absence of L arginine, agmatine and D arginine. With respect to the bacteriorhodopsin and AdiC antiporter, the obtained results are similar to the earlier outcomes reported by earlier investigators (Janovjak et al, 2003 Oesterhelt et al, 2003 Mueller et al, 2003 and Janovjak et al, 2004). This confirms that the technique is highly reliable.
A Programmable Force Sensor Technique
The programmable force sensor method of SMFS was first described by Albrecht et al (2003). Unlike HT-SFMS, this technique utilizes a differential format for measuring the forces inherent in the linkages of single molecules. Here, comparisons between the rupture force of the bond in the molecule of interest and that of the bond of a known reference molecule (the standard) are made. Another modification of this particular technique is that the cantilever spring is not used. In its place, the authors used a polymeric anchor and a reference molecule whose bond was known and which was attached to a fluorescent tag. This contrasts with the HT-SMFS method which utilizes displacement versus spring constants to meaure the forces (Albrecht et al, 2003 p. 367).
In their experiment, Albrecht et al (2003 p. 367) used a 20 bp DNA duplex as the test molecule and a 25 bp DNA duplex as the reference molecule. The polymeric anchor used was a 65 bp oligonucleotide while Cy5 was used to tag the reference molecule. The 2 surfaces were detached by stretching the polymeric anchor. Thereafter, the weakest of the bonds between the test molecule and the reference standard was ruptured by the gradual application of force. Breakage of the symmetry was as a result of the differences between the stability of the bonds (Albrecht et al, 2003.p.368).
As Bosshart et al (2008) show, the HT-SMFS is a faster technique which requires less time. In their experiments, entire data sets were collected in only a single day and this is a big improvement in terms of time utilization over the non-automated SMFS procedure. Another advantage of the HT-SMFS technique is that it is highly objective. The objectivity of the method is qualified by the observation that subjective biases are eliminated through the automation of the data collection and pre-processing stages. Additionally, the HT-SMFS is highly versatile in that it can be used with both crystalline and non-crystalline samples to produce reliable outcomes. The versatility of the technique is further enhanced by the observation by Bosshart et al (2008) to the effect that HT-SMFS can also be used to study cloned proteins, native and reconstituted membrane proteins and is adaptable to dynamic studies that investigate the enrgy landscape of such proteins.
Besides being verstaile, HT-SMFS is highly accurate. As stated earlier, Bosshart et al (2008) were able to validate the accuracy of this method using bacteriorhodopsin. The outcomes obtained conformed to earlier findings by Janovjak et al (2003), Oesterhelt et al (2000), Mueller et al (2002) and Janovjak et al (2004). This shows that the technique is highly accurate and is in contrast to the the conventional SMFS technique which has limited capability to resolve the forces. In addition to providing information about the topology of the molecules under study, HT-SMFS can also be used to reveal data about the mechanical stability of structural components.
Yet another advantage of this technique is that it allows all the F-D curves to be recorded and condenses the huge amount of data to a smaller number of force spectra. As a result, it is possible to carry out manual analysis of the data. Moreover, the HT-SMFS method is less laborious since it is semi-automated. According to Bosshart et al (2008), the technique requires that the operator be present for approximately 10 of the entire duration of data acquisition. This time is significantly less than the time required for the conventional SMFS methods  (Bosshart et al, 2008).
The HT-SMFS also overcomes the problem of low efficiency that is inherent in the conventional SMFS method. The conventional SMFS method cannot be used for large scale projects since the method has a low efficiency. This is because many F-D curves do not necessarily result in an unfolding event. This finding was confirmed by Kedrov et al (2004) who observed absolute protein unfolding in just about 3 of instances using bacterial sodiumproton antiporter. In contrast, the HT-SMFS is highly efficient. As described by Bosshart et al (2008), the HT-SMFS method allowed the collection of up to 40,000 F-D curves daily and this yield is much higher than that of the conventional SMFS method  (Bosshart et al, 2008).
In the conventional SMFS method, outcomes are encumbered by factors which have an impact on unfolding such as temperature and ions and this calls for the collection of large data sets hence the need for high throughput protocols or HT-SMFS (Jonovjak et al, 2003 Mueller et al, 2006).  Evidently, results from HT-SMFS are not affected by such considerations. Wheras there is a high possibility of experimental errors interfering with the outcomes due to continuous drifts of calibration measures in the conventional SMFS, HT-SMFS is, by and large, devoid of these errors  (Bosshart et al, 2008).
However, HT-SMFS does not completely overcome the weaknesses of the conventional SMFS technique. For one, complete automation is not possible. This is because AFM imaging must first be carried out prior to the SMFS procedures. Secondly, despite progress being made towards the formulation of pattern recognition and automated alignment algorithms, the collection and filtering of data remains sub optimal (Kuhn et al, 2005 Marsico et al, 2007 Dietz  Rife, 2007). As Bosshart et al (2008) assert, one F-D curve requires nearly 1 second to be recorded. Further, the manual logging in of an entire data set, which has about 200 spectra and where an efficiency of 1 is sought, is a burdensome affair. Additionally, the operator performs online filtering during the logging in step and this has been shown to cause a loss in force spectra thus resulting in a decline in the efficiency of data collection. The final drawback of the HT-SMFS is that it requires that the AFM be conducted for prolonged durations. J
Even though the HT-SMFS is seen to have superior characteristsics, the force sensor method also has a number of advantages which make it ideal than the conventional SMFS technique. First, the force sensor method is characterized by a relatively high symmetry. The net result is that there is a cencelling out of nearly all outside influences. Secondly, the force sensor method has a high precision. A third advantage of the force sensor technique is that it enables the identification of single base pair mismatches in nucleotides,  something which cannot be identified by conventional SMFS. Fourthly, the force sensor method is a more useful technique in certain applications where the differences between the test and references as opposed to absolut values is desirable. Fifthly, it allows many different measurements to be performed concurrently (Albrecht et al, 2003).
Importantly also, the force sensor method can be used to reliably discriminate between specific and non-specific protein interactions. Therefore, this technique is of high value in studies of protein arrays. Since the method discriminates between the different bindings and is characterized by mechanical stringency, it can be used in capture arrays to eliminate background noise and reduce the number of false positives thus enhancing the multiplexing capacities of the protein capture arrays. In sum, the force sensor technique enhances the specificity of protein biochips by eliminating cross-reactions and non-specific interactions. Other advantages of the force sensor method are that it is highly sensitive by virtuie of its high discriminatory power and it can be applied in highly specific parallel assays. The method can also be used to carry out highly accurate assays of single nucleotide polymorphisms (SNPs) (Albrecht et al, 2003). Unlike the conventional SMFS method, it is largely unaffected by prevailing environmental conditions such as ions and temperataure (Albrecht, 2003 p. 370).
The force sensor technique has several drawbacks though which make it a less ideal technique than the HT-SMFS. It largely fails to overcome most of the weaknesses of the conventional SMFS method highlighted earlier. Specifically, the force sensor technique is more labourious and time-consuming and may be affected by global conditions. In particular, these conditions necessitate the use of many controls. Besides, it is not possible to directly quantify the density of the fluorescent tag due to dissimilarity of the chemical and optical characteristics of the surfaces in use. This reduces the efficiency and yield (Albrecht et al, 2003).
Single-molecule force spectroscopy (SMFS) meaurements are indispensable tools in the study of membrane proteins. Whereas both the HT-SMFS and force sensor techniques have vastly improved the utility of the method, the former is seen to be better suited for such studies as it is less laborious, less time consuming, highly efficient, highly versatile, semi-automated and less subjective.


Post a Comment