Disruption of Four Kinesin Genes in Dictyostelium
© Nag et al; licensee BioMed Central Ltd. 2008
Received: 02 January 2008
Accepted: 22 April 2008
Published: 22 April 2008
Skip to main content
© Nag et al; licensee BioMed Central Ltd. 2008
Received: 02 January 2008
Accepted: 22 April 2008
Published: 22 April 2008
Kinesin and dynein are the two families of microtubule-based motors that drive much of the intracellular movements in eukaryotic cells. Using a gene knockout strategy, we address here the individual function(s) of four of the 13 kinesin proteins in Dictyostelium. The goal of our ongoing project is to establish a minimal motility proteome for this basal eukaryote, enabling us to contrast motor functions here with the often far more elaborate motor families in the metazoans.
We performed individual disruptions of the kinesin genes, kif4, kif8, kif10, and kif11. None of the motors encoded by these genes are essential for development or viability of Dictyostelium. Removal of Kif4 (kinesin-7; CENP-E family) significantly impairs the rate of cell growth and, when combined with a previously characterized dynein inhibition, results in dramatic defects in mitotic spindle assembly. Kif8 (kinesin-4; chromokinesin family) and Kif10 (kinesin-8; Kip3 family) appear to cooperate with dynein to organize the interphase radial microtubule array.
The results reported here extend the number of kinesin gene disruptions in Dictyostelium, to now total 10, among the 13 isoforms. None of these motors, individually, are required for short-term viability. In contrast, homologs of at least six of the 10 kinesins are considered essential in humans. Our work underscores the functional redundancy of motor isoforms in basal organisms while highlighting motor specificity in more complex metazoans. Since motor disruption in Dictyostelium can readily be combined with other motility insults and stresses, this organism offers an excellent system to investigate functional interactions among the kinesin motor family.
Dictyostelium discoideum is a compact amoeba that spends much of its natural existence crawling through the soil, searching for and ingesting bacteria. When food sources are exhausted, individual amoebae trigger a developmental program that initiates both inter and intracellular signaling, to aggregate ~100,000 amoebae and form a multicellular mass. Each cell within this mass undergoes multiple adhesions and conformational changes, forming a cooperative slug that can migrate to new areas. The slug undergoes further multicellular differentiation to form supportive stalk cells, a rudimentary immuno-like surveillance system, and regenerative spores that resist environmental stresses. This dualistic life cycle and its associated transitions (single cell to metazoan organism) have made Dictyostelium an attractive model in which to study cell motility, signal transduction, and a relatively simple developmental program (reviewed in , see also ).
Motility-wise, Dictyostelium behaves in a manner similar to that of many vertebrate cells (crawling, sensing, and engulfing targets, robust intracellular movements). Yet, this organism clearly retains a simplicity associated with its relatively small and compact genome, and exhibits features commonly seen in protozoa and fungi (for example, an intranuclear spindle for cell division). Characterization of the actin cytoskeleton in Dictyostelium has led to the identification of actin binding proteins, multiple myosin motors, and signaling cascades whose functions are conserved among eukaryotic cells. Preliminary characterization of the microtubule-associated network has revealed a level of complexity intermediate between some of the simple single-celled eukaryotes and metazoans. For example, the machinery in Dictyostelium that drives movement along microtubules contains 14 motors (13 kinesin ATPases, 1 dynein ATPase, [3, 4]); twice the number found in Saccharomyces cerevisiae , but less than a quarter of the number encoded in the human genome . Paradoxically, deletions of kinesins whose homologs are essential for vertebrate activities have produced relatively mild phenotypes in Dictyostelium. Are these results reflective of Dictyostelium's unique life cycle? Or do they reveal core functional redundancies and interactions that, like the actin system work, can be utilized to understand microtubule-based motor action in more complex systems?
MT Array Morphology in Interphase Cells
We have presented gene deletions for four of the 13 kinesin family members in Dictyostelium, and have described the effects of these deletions on cell growth and viability. Individually, none of the four gene products is essential for cell viability nor do the proteins play critical roles in this organism's ability to undergo chemotaxis or to develop upon starvation. The knockout strains do, however, show subtle defects suggesting that many of the key forms of intracellular motility essential for Dictyostelium biosynthesis and reproduction are supported by more than one motor protein.
In wild-type Dictyostelium cells, both plus end-directed microtubule pushing, and minus end-directed pulling forces are important for maintenance of centrosome position and the radial distribution of interphase microtubules [18, 21]. If minus end-directed dynein motility is impaired, a kinesin-like activity appears to dominate and push both the centrosome and microtubule array throughout the cytoplasm . Here we have identified two kinesins, kif8 (kinesin-4 family) and kif10 (kinesin-8 family), that appear to collaborate with dynein in this organization process. In other eukaryotic cells, kinesin-4 motors participate in a number of diverse activities . One subset of kinesin-4 family members (KIF4) function during mitotic events, with chromatin- and spindle-associated motors that organize bipolar microtubule assemblies and facilitate chromosome alignment . Other subsets of kinesin-4 motors (e.g., KIF21) appear to power interphase organelle transport in cultured cells such as fibroblasts and post-mitotic neurons [24, 25]. The single Dictyostelium kinesin-4 (kif8) is a divergent member of this family, the motor domain is most closely homologous with KIF4 subfamily, yet it contains carboxy-terminal WD-40 repeat motifs in the heavy chain tail that are characteristic of the KIF21 subfamily [3, 22]. The kinesin-8 family of motors (kif10 in Dictyostelium) is thought to mediate chromosome movements through a combination of translocation and microtubule depolymerization activities (recently reviewed in , see also [27, 28]. The S. cerevisiae isoform (Kip3) has previously been shown to cooperate with dynein in positioning mitotic spindles through cortically mediated force production and through control of microtubule length [27, 29, 30]. Deletions of kinesin-8 isoforms in Schizosaccharomyces pombe also suggest a combined force and length control mechanism that positions nuclei and spindles through microtubule-cortex interactions [31, 32]. In the absence of either kinesin-4 or kinesin-8 in Dictyostelium, we are unable to induce the distinctive centrosome movements via dynein motor overexpression. It is conceivable that Kif8 and Kif10 counterbalance dynein-mediated forces through force-production or anchoring activities at the cell cortex (e.g. kinesin-8) and via lateral microtubule-microtubule interactions (e.g. kinesin-4) that supply sufficient rigidity to allow plus end-directed motors to effectively push (and not simply bend) microtubules. In wild-type Dictyostelium, the balance between opposing dynein and kinesin motor activities serves to reinforce the centrosome position and help maintain the radial character of the interphase microtubule array as these cells crawl around and change shape.
Disruption of the kinesin-7 motor (CENP-E) in the mouse is embryonic lethal ; this motor is thought to be essential for the proper connection between kinetochores of condensed chromosomes and the mitotic spindle . In contrast, neither member of the kinesin-7 family in Dictyostelium (Kif4, Kif11) is essential for mitosis, although removal of Kif4, the isoform that is most homologous to the vertebrate kinetochore CENP-E greatly affects cell growth rate. Preliminary characterization of Kif4 suggests that this motor functions together with dynein in organizing spindle assembly during cell division. While the motor domain of Kif11 is homologous with the kinesin-7 family , this polypeptide is significantly shorter and expressed at a much higher level than other CENP-E-like proteins. Outside of a minor enhancement of stationary phase cell density, removal of this motor has no obvious effect on cell viability or function. Closer inspection of each kinesin, and of cells lacking their expression will be required before we can fully understand their individual function(s)
Our study here extends previous work from several laboratories that, taken together, have individually deleted 10 of the total 13 kinesins in Dictyostelium [7, 9, 10, 12–14]. All of these deletions have proven to generate cell lines that can survive over multiple generations of growth, indicating that none of these 10 kinesin motors is immediately required for cell viability. Although the Kif12 disruption (kinesin-6, MKLP) produced significant defects in cytokinesis, mutant cells were still able to undergo some form of division that allows strain propagation . The only, potentially essential, kinesin gene reported so far in Dictyostelium encodes one of the organelle transporter motors, kif3 (kinesin-1 family). Kif3 can be isolated biochemically and shown capable of powering microtubule gliding, but efforts by Röhlk et al,  and in our own lab (Nag, Tikhonenko, and Koonce, unpublished) have not yet yielded viable cells lacking this motor. The resiliency of Dictyostelium to motor disruptions is similar to systematic analyses of kinesin isoforms in S. cerevisiae, where all six kinesin-related motors (and one dynein isoform) can be individually deleted without loss of viability . The yeast work provided a major guiding principle, for it was the first to suggest that high degree of functional redundancy is present among kinesin family members, and that deletion of motor combinations is required to inhibit cell division. Although, to our knowledge, complete survey disruptions have not yet been reported in other simple eukaryotes, there are clear indications of motor redundancy in some cell models such as S. pombe , Aspergillus nidulans, and Ustilago maydis . The kinesins in Dictyostelium likewise possess overlapping functions.
Mitotic kinesin disruptions in simple eukaryotes vs metazoans.
Kinesin- 8 (Kip3)
Kif2A, 2B, MCAK
Analysis of the kinesin gene family in Dictyostelium suggests that a significant level of functional redundancy or overlap exists among the organism's motor activities. This result is similar to findings from functional analyses performed in basal organisms such as yeast and fungi, but it contrasts sharply with the roles of individual motors in metazoans. At first glance, most of the kinesins in Dictyostelium can be deleted individually without penalty to growth or viability. Yet, upon closer scrutiny or in cases where we impose under additional stresses, we can discern clear phenotypic changes in the cell that provide insight into motor function that may not be obvious in other organisms. Given its greater complement of motor isoforms, and its greater utility of microtubule function relative to other basal eukaryotes, Dictyostelium offers an interesting model in which to investigate functional interactions and the regulation of multiple motor proteins.
Kinesin gene sequences were obtained from the dictybase website (see Availability and requirements section). The following primer combinations were used to amplify kinesin gene fragments from AX2 cell genomic DNA; also listed are the downstream kinesin gene-specific primers used for screening recombinants:
Forward: 5'CGCAAGCTT AGCCACCAAGACCATTACTTGGACCA 3' (-501 to -476)
Reverse: 5'CGCGAGCTC TTAAACTACCACCAATTATTGCGTCATT 3' (+1318 to +1345)
Screen: 5'CATCATCATCCTCTTCACCACTACTATT 3' (+1501 to +1528)
Forward: 5'CGCGGATCC GGGTTGCATTAAGAGTTAGACCC 3' (+44 to +66)
Reverse: 5'CCCAAGCTT GAATCGGCAGGACTAACACATGC 3' (+ 1302 to +1324)
Screen: 5'GATTGGTTAATACACACCTAATTG 3' (+1381 to +1404)
Forward 5'CGCGGATCC TGATCAATATGCAACTCAAGAAGAAG 3' (+249 to +274)
Reverse 5'CCCAAGCTT GATCATTGTCATCATCATCATC 3' (+1408 to +1429)
Screen: 5'GTATCATTGATTCATCATTATCCCT 3' (+1501 to +1525)
Forward: 5'CGCGGATCC GAATGAACGAGAATATATCGGTTAGC 3' (-2 to +24)
Reverse: 5'CCCAAGCTT CCATTACCACTACCACTACCACCT 3' (+1497 to +1520)
Screen: 5'TGACTTGGTGAAACAAATGTTGATC 3' (+1532 to +1556)
+1 of the numbering scheme refers to the position A of the ATG start codon. Restriction enzyme sites were engineered into the ends of each primer (BamH 1, Hind III or Sac 1, shown in bold type) to facilitate cloning of the amplified DNA into a pUC19 host plasmid, and (in most cases) to excise the DNA construct for transformation. Each construct was sequenced to confirm the identity of the kinesin fragment. Native restrictions sites (Fig. 2) were used to excise and replace an internal fragment of the kinesin sequences (47–669 bp) with a 1.6-kb blasticidin resistance cassette (Bsr r ) (Sma I digest) from pLRBLP , obtained from the Dictyostelium Stock Center (see Availability and requirements section for URL). Final constructs were again sequenced to determine the orientation of the Bsr r cassette (diagramed in Fig. 2). The kif8 construct was designed to terminate message coding at S202; kif10 at N223; kif11 at S151; and kif4 at W45. In all cases, these disruptions occur upstream of the microtubule-binding domain of the motor.
Standard molecular biology procedures were followed for DNA isolation, manipulation, and blotting. RNA was isolated using the RNeasy kit from Qiagen, following the manufacturer's instructions. kif8, kif10, and kif11 blots were probed with 32P-labeled DNA. the kif4 Southern blot was performed using chemiluminescence procedures (ECL, Amersham Biosciences). All blots (Southern and Northern) were probed with the initial amplified genomic target corresponding to the relevant kinesin clone, as indicated above and in Figure 2A.
A calcium phosphate procedure was used to transform Dictyostelium AX-2 cells, with 15 μg of linearized DNA per near confluent 10-cm dish (107 cells) . Transformants were selected with 5 μg/ml blasticidin. Individual colonies were picked with a pipette into 24 well plates, and were screened by PCR for homologous recombination. Amplification of a 1.6-kb target with a primer internal to the Bsr r marker (5' GAATGGCAAGTTAGTCAAAACTACG 3') and a primer downstream of the recombination site (indicated above for each kinesin sequence) was used to initially identify positive recombinants. Cells from positive colonies were further purified by serial dilution, and were again confirmed by PCR with downstream and upstream primer combinations. For dynein disruptions, we introduced a motor domain expression plasmid (aa 1384–4725), into kinesin null cells by either a CaPO4 or an electroporation method . kif-/380 K expressing cells were selected with 10 μg/ml G-418 (geneticin, Sigma Chemical Co).
Cells were flattened on glass coverslips using an agarose sheet, fixed with formaldehyde, labeled with a tubulin antibody , and in some cases Hoechst 33342, as described in . Z-series of images were obtained on a DeltaVision light microscopy workstation and were deconvolved using softWoRx 2.5 (Applied Precision, Issaquah, WA). Maximum intensity projections were compiled using ImageJ (NIH); figures were assembled in Adobe Photoshop. For cell growth measurements, triplicate 100-ml cultures were seeded with 9 × 104 cells/ml, shaken at 200 rpm at RT, and counted with a hemocytometer every 24 hr. Growth curves were calculated and displayed with Microsoft Excel; error bars indicate standard deviation.
The Dictybase website: http://dictybase.org/
Dictyostelium Stock Center: http://dictybase.org/StockCenter/StockCenter.html
We are grateful to the efforts at http://dictybase.org/ to archive and annotate Dictyostelium sequence information, and to the Dictyostelium Stock Center Resource for plasmids. Drs. Alexey Khodjakov and Conly Rieder provided valuable discussion and assistance with the light microscopy. We appreciate the use of Wadsworth Center's Molecular Genetics Core for DNA sequencing. This work was supported in part by the NSF (MCB-0542731 to MPK).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.