Design of Artificial Calcium-Calmodulin Regulation in Conventional Kinesin Adapted from Kinesin-Like Calmodulin-Binding Protein

Spring 2005 Senior Design Project

Jason T. Robosky

Department of Bioengineering, The Pennsylvania State University

Sponsored by Dr. William O. Hancock, Molecular Biomechanics Laboratory

Executive Summary

Kinesin motor proteins use the chemical energy of adenosine triphosphate (ATP) hydrolysis to transport cellular cargo along cytoskeletal microtubules (MTs.) Kinesin-like calmodulin-binding protein (KCBP) is one such motor that can be regulated by calcium-calmodulin (Ca-CaM) through its calmodulin-binding domain (CBD.) The goal of this design project was to use standard molecular biology techniques to fuse the CBD of KCBP directly to the N-terminus of a full-length conventional kinesin (HisKin) to confer artificial, reversible Ca-CaM regulation. Simple switching is ideal for molecular biosensor and sorting applications where these motors have an impact. This design was verified using sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE) to show that the mutant motor (CBD-HisKin) purified at full-length, motility assays to prove that the sliding velocity remained unchanged by the presence of the CBD under standard conditions, landing rate assays to quantify the extent of regulation under the influence of Ca-CaM, and ATPase assays to show how regulation affected catalytic rates in the presence of microtubules. Over the 14 weeks required to complete verification, economic costs totaled $4100.

Although complete regulation was not yet shown, this is thought to be due to high microtubule affinity by the conventional kinesin. Further verification through ATPase assays and alternative analyses may show the utility of this hybrid motor in molecular engineering applications. Alternative designs may also need to be considered. Despite an economically feasible approach costing only around $10/mg, the hybrid motor design documented here has not yet been shown to completely meet the need for which it was conceived.

Background and Significance

Kinesins are molecular motor proteins that transduce the chemical energy of ATP to mechanical work as they move intracellular cargo along MTs [1]. A schematic diagram of conventional kinesin is shown in Figure 1 [2], and its interaction with MTs is depicted in Figure 2 [1]. Although cellular regulation of proteins is essential to ensuring their proper function, the mechanisms of conventional kinesin inhibition are still unclear. KCBP, illustrated in Figure 3 [3], is a unique plant motor that has recently been isolated and shown to utilize the commonly employed Ca-CaM pathway for regulation through its C-terminal CBD [4]. Furthermore, this domain has been fused into several truncated motors to show that selective regulation can be transferred, but no full-length motor designs capable of cargo transport have yet been attempted [5].

In this design project, a hybrid, full-length motor protein was designed, constructed, and tested in which the CBD of KCBP was fused directly to the N-terminus of HisKin to confer artificial Ca-CaM regulation while native motor function was retained. The technological need for such simple switching is important for healing diseases of nerve transport, engineering efficient and effective drug delivery, and providing power to novel molecular transport nanodevices such as molecular sorting arrays or biosensors as shown in Figure 4 [6]. The Ca-CaM regulation conferred by KCBP’s CBD to the fast, processive motors needed for these applications is of high interest because it would essentially act as a simple, reversible on/off switch to enhance the efficiency and specificity of these devices. Portable, rapid, sensistive, reusable, and inexpensive sensors like these are becoming a necessity in this age of biological and chemical warfare for assessing the quality of small air, water, food, and soil samples.

Currently, conventional kinesins are not easily regulated in vitro. Since KCBP has been discovered and its CBD has been characterized, a feasible mechanism of simple Ca-CaM regulation has been envisioned. As such, the CBD lends a promising solution to the problem of conventional kinesin regulation by utilizing the tools of biotechnology. Although KCBP could potentially solve this problem on its own, the catalytic rate of this enzyme is insufficient for rapid transport applications. The issue of selective switching is one of the major features of molecular transport that must be considered before nano-scale diagnostic and therapeutic devices can be effectively produced and economically implemented.

                   Figure 1:  Conventional Kinesin Structure                                                    Figure 2:  Kinesin-MT Interactions

                        Figure 3:  KCBP Crystal Structure                                                                  Figure 4:  Kinesin Biosensor

Design Objectives

Design and characterization of this reversibly regulated, hybrid kinesin can be accomplished through three main objectives: 

• Design and create a kinesin construct containing the CBD of KCBP on its motor domain.
• Verify native activity of this enzyme through motility assays.
• Evaluate and quantify the extent of conferred regulation under the influence of Ca-CaM using landing rate and ATP hydrolysis measurements.

Following successful completion of each of these tasks, the motor prototype will be sufficiently characterized for use in transport devices. It should be noted that complete validation of this design cannot be totally accomplished since the devices for which this motor is intended are still in their infancy. However, the motility assays used during verification satisfactorily represent the conditions imposed by channel transport.

An effective protein design must satisfy the following design criteria: 

• The motor must be expressed and purified in full-length dimers:  mass must be at least 90% of the original motor.
• The hybrid motor must retain the native structure and chemistry of conventional kinesin:  sliding velocity and landing rate must be within 90% of the original motor under normal conditions.
• Calcium and calmodulin should regulate the chimera (mutant protein) together but not by themselves:  ionic strength of the regulatory solution should be kept below 100 mM salts.
• Insertion of the CBD must effectively regulate the motor in the presence of Ca-CaM:  landing rate and ATP hydrolysis rates must be reduced to less than 10% of the original motor under inhibitory conditions.

The major constraint placed on this design is that CBD insertion must not directly inhibit native functionality of substrate binding, catalysis, conformational changes, dimerization, and cargo binding respectively. Additionally, the design must be readily constructed using standard molecular biology techniques and recombinant deoxyribonucleic acid (DNA.) Moreover, design criteria verification during the review process must be readily performed using existing techniques. Cost, time, and resource limitations are automatically accounted for by these approaches since they involve the most efficient and feasible methods available.

Design Selection and Description

Several approaches were considered for this design. First and foremost, a design mimicking the natural structure of KCBP was proposed in which the CBD was appended directly to the kinesin motor head at the N-terminus (A.) Next, a design was examined in which the CBD was attached to the N-terminus by a 10 amino acid random sequence spacer (B.) Additionally, a design was considered where the CBD was inserted by a linker to the C-terminal tail of kinesin (C.) Since the ultimate goal is regulation through any means available, a final design was considered by which the motor is allosterically regulated (not at the active site) by enzymatic phosphorylation (D.) Each of these constructs involves full-length, dimeric Drosophila conventional kinesin heavy chain due to its superior motility properties, but other kinesins could be used as well. Schematics of each alternative are shown in Figure 5 as labeled above [2-3].

Figure 5:  Design Alternatives

Based upon an objective analysis of weighted decision criteria including vicinity to the MT binding domain, ease of reversible regulation, interference with native functionality, cost to construct, and regulation mode selectivity, design concept A was chosen for development since the likelihood of a successful design is enhanced by this placement strategy.

Alternatives for design verification were analyzed similarly against criteria including ease of evaluation, cost to analyze, interpretability of results, and feasibility of approach, the following methods were chosen:

• SDS-PAGE to analyze purification length

• Motility assays to assess native functionality

• Landing rate and ATPase assays to assess extent of regulation

Detailed Design

The CBD was cloned, amplified, and appended to the N-terminal motor domain of HisKin using standard molecular biology techniques to create the hybrid motor CBD-HisKin. Specifically from N- to C-terminus, this construct consists of the CBD (KCBP residues 1211-1261,) the motor domain of conventional kinesin (residues 1-326,) the neck linker (residues 326-338,) the coiled coil/tail domain (residues 338-975,) and a 6-Histidine (His) tag for purification. A linear domain map of the hybrid protein is shown in Figure 6 which corresponds directly to the design depicted in Figure 5A. The protein was then over-expressed in E. coli and purified on a histidine affinity chromatography column. It should be noted that the true CBD regulatory mechanism is unknown; however, this question is the subject of current research topics.

Figure 6:  CBD-HisKin Structure

SDS-PAGE was used to visualize the purified motors based on mass. In this assay, proteins are denatured and given a negative charge. They then migrate through a gel in the presence of an electric field. Afterward, the gel is stained and protein bands can be seen.

Motility assays were used to assess the native functionality of CBD-HisKin to show that the CBD does not interfere with motility when Ca-CaM is not bound. In this type of assay, a glass surface is covered with a biocompatible protein layer of casein to block non-specific protein-surface interactions. Next, motors are adsorbed to the surface by their tails leaving their heads free in solution at the surface. Finally, fluorescently labeled MTs are added and observed by fluorescence microscopy as they glide along the surface motors. Sliding velocity can then be measured. Since KCBP controls do not have tail domains, these motors were attached to the surface indirectly by an anti-his antibody. A concept diagram of this process is shown in Figure 7. During landing rate assays, the parameter measured is the frequency with which MTs land on surface motors and begin moving. These assays were also performed after Ca-CaM was added to observe regulation [7].

Figure 7:  Microtubule Gliding Motility Assay

ATPase assays were used to evaluate regulatory changes in CBD-HisKin from a kinetic perspective. Here, a steady-state enzyme-coupled assay was used in which ATP hydrolyzed by the motor leads to oxidation of nicotinamide adenine dinucleotide (NADH) as shown in Figure 8. This signal can be monitored spectrophotometrically. Enzymes and substrates shown in the diagram are phosphoenolpyruvate (PEP,) pyruvate kinase (PK,) and lactatic dehydrogenase (LDH) [8]. For the full-length HisKin and CBD-HisKin constructs, tail auto-inhibition was avoided by adsorbing the motor to silica beads as shown in Figure 9. The tails of these motors normally block MT binding in the absence of cargo [9].

                           Figure 8:  Enzyme-Coupled ATPase Assay                        Figure 9:  Kinesin Tail Auto-Inhibition

Major deliverables for this hybrid kinesin design were:

• Obtain or create purified KCBP and conventional kinesin protein constructs.
• Obtain or create plasmid (circular) DNA for each motor.
• Obtain or create a fused protein plasmid from the amplified CBD and kinesin.
• Obtain or create a bacterial culture of the chimera.
• Obtain or create a purified hybrid motor protein construct.
• Measure landing rates and sliding velocities of the chimera with respect to conventional kinesin under standard conditions.
• Measure solution ATPase and landing rates for the chimera with respect to conventional kinesin and KCBP under regulatory conditions.
• Quantify regulation for comparison to conventional kinesin and KCBP.
• Complete a technical poster presentation.
• Design and post a public project web page.
• Submit a complete written technical report.

Significant milestones for this project included:  

• Completing a working design for the hybrid kinesin.
• Obtaining a purified construct of the chimera.
• Successfully completing motility assays under standard conditions.
• Successfully measuring landing rate and ATPase rate under standard and regulatory conditions.

Although most significant equipment needed for this construction and analysis was already available (35 pieces,) approximately 70 biomaterials and biochemical reagents were required over the 14 week analysis period at a total material cost of just over $2000. Labor costs for this time period ran approximately $1400, and overhead was estimated at $700. Total economic costs came to roughly $4100. The final enzyme product could be produced with high purity at an approximate total cost of $10/mg without considering analysis making this an economically feasible design approach.

Design Evaluation

SDS-PAGE analysis showed the expected band at ~125 kD corresponding to full-length CBD-HisKin. According to this analysis as presented in Figure 10, the motor successfully purified at full-length indicating no genetic fusion complications.

Figure 10:  Motor Electrophoresis

The results of motility and landing rate assays are shown in Table 1 for KCBP, HisKin, and for various dilutions of CBD-HisKin. In addition to sliding velocity (SV,) landing rate (LR,) and release rate (RR) in motility solution (MS) and regulatory solution (RS,) the table also shows observations for microtubule surface density (σ) and a calculated parameter called the regulation factor (ρ.) The regulation factor, defined in Equation 1, is statistically similar to the odds ratio in that it describes how many times more likely the motor is to be regulated with than without Ca-CaM. Using t-tests to analyze results, sliding velocity and landing rate data show no statistically significant difference between HisKin and CBD-HisKin under standard conditions (P>0.50 and P>>0.50 respectively.) According to the regulation factor analysis, HisKin is unaffected by Ca-CaM while KCBP is highly regulated. CBD-HisKin lies between these values indicating incomplete regulation. Interestingly, however, sliding velocity of the chimera is significantly reduced (P=0.001) indicating some effect of Ca-CaM [10].

Results were not obtained reproducibly for ATPase assays. Although motors were successfully adsorbed to beads, problems with adequacy of signal, scattering of light, and poor system response. Most of these difficulties were successfully mitigated, but the motor stock was dilute and contained high concentrations of residual ADP thereby preventing adequate signal from being obtained. This roadblock also delayed project progress. Future work involves increasing motor concentration with ATP exchange followed by replication of ATPase assays.

Recommendations and Conclusions

Despite fulfilling 3 of the 4 design criteria, successful regulation was not yet shown for this design. Sliding velocity decreases in the presence of Ca-CaM provide hope for construct regulation, but ATPase assays must be completed before any judgment can be made. It is hypothesized that the reason for this lack of effective regulation comes from motor processivity, the ability to take many steps before dissociating from the MT. Processivity comes from high substrate affinity, and it may be that full-length processive motors are not easily regulated by this artificial system. Alternative techniques can also be used to evaluate regulation. If the design is shown to be completely unsatisfactory, alternative designs could also be implemented. Additionally, more should be learned about regulation in KCBP so that informed design considerations can be made. The final goal for measuring success of this design is through validation of the prototype in biosensor or drug delivery devices. Although these nanotransport applications are not currently available, this is the ultimate use for these motors. Without finalized conclusions regarding regulation of CBD-HisKin, the ultimate success of this design in meeting its need cannot yet be determined.

References

[1] J. Howard, “Movement of kinesin along microtubules,” Annu. Rev. Physiol., vol. 58, pp. 703-729, 1996.

[2] D.D. Hackney, “Kinetic cycles of myosin, kinesin, and dynein,” Annu. Rev. Physiol., vol. 58, pp. 731-750, 1996.

[3] M.V. Vinogradova, V.S. Reddy, A.S.N. Reddy, E.P. Sablin, and R.J. Fletterick, “Crystal structure of kinesin regulated by Ca²⁺-calmodulin,” J. Biol. Chem. vol. 279, pp. 23504-23509, 2004.

[4] B.E. Deavours, A.S.N. Reddy, and R.A. Walker, “Ca²⁺/calmodulin regulation of the Arabidopsis kinesin-like calmodulin-binding protein,” Cell Mot. Cytoskel., vol. 40, pp. 408-416, 1998.

[5] V.S. Reddy and A.S.N. Reddy, “The calmodulin-binding domain from a plant kinesin functions as a modular domain in conferring Ca²⁺-calmodulin regulation to animal plus- and minus-end kinesins,” J. Biol. Chem., vol. 277, pp. 48058-48065, 2002.

[6] L. Jia, S.G. Moorjani, T.N. Jackson, and W.O. Hancock, “Microscale transport and sorting by kinesin molecular motors,” Biomed. Microdevices, vol. 6, pp. 67-74, 2004.

[7] W. O. Hancock and J. Howard, “Processivity of the Motor Protein Kinesin Requires Two Heads”, J. Cell. Biol. vol. 140, pp. 1395-1405, 1998.

[8] D. D. Hackney and W. Jiang, “Assays for kinesin microtubule-stimulated ATPase activity,” Methods Mol. Biol. vol.164, pp. 65-71, 2001.

[9] D. L. Coy, W. O. Hancock, M. Wagenbach, and J. Howard, “Kinesin’s tail domain is an inhibitory regulator of the motor domain,” Nature Cell Biol. vol. 1, pp. 288-292, 1999. 

[10] O. J. Dunn and V. A. Clark, Basic Statistics:  A Primer for the Biomedical Sciences, New York:  John Wiley and Sons, 2004.