ASSEMBLING THE EUKARYOTIC FLAGELLUM: Another example of IC?

By Mike Gene (9/06/03)

In my previous essay [1] on the eukaryotic flagellum, I noted the inherent ambiguity associated with trying to score the dyneins and linkers as IC components. Higher resolution studies are required to better characterize this system in an IC relevant fashion with regard to these components. Nevertheless, given the flagellum is composed of over 200 proteins, the potential remains elsewhere for a less ambiguous scoring. As I noted, if only 10% of these proteins form an IC core that resists explanation by cooption and/or duplication, we have a serious challenge to the Darwinian explanation.

One promising arena for clearing up this ambiguity is the IFT system that has been proven essential for flagellar assembly. This machinery includes a kinesin II motor complex that moves material into the flagellum, a dynein complex that moves material out of the flagellum, and an IFT “raft” composed of 16 different proteins. The raft is what carries the cargo into and out of the flagellum. They can be isolated as two distinct complexes: A and B. Complex A contains IFT 144, IFT140, IFT139, IFT122A, IFT122B, and IFT43. Complex B contains IFT172, IFT88, IFT81, IFT80, IFT74/72, IFT57/55, IFT52, IFT46, IFT27, and IFT20. Complex B is thought to move material into the flagellum, while complex A is thought to move material out of the flagellum. [2]

Given that the machinery is found in protozoa, invertebrates, and vertebrates, Pazour et al. describe the IFT system as “an ancient and conserved mechanism.” [3] This degree of conservation fits well with the rest of the flagellum:


Although C. reinhardtii and mammals are separated by more than 10^9 years of evolution, C. reinhardtii flagella are amazingly similar in structure and function to mammalian cilia and flagella. For example, some of the flagellar proteins in C. reinhardtii show more than 75% identity and similarity to proteins with similar function in human sperm. [4]


While we are just beginning to characterize the IFT proteins, genetic work in algae have shown several of them to be essential for flagellar assembly, where knockouts completely eliminate the ability to construct flagella. Let us consider some of the better characterized proteins.



IFT88

Genetic work shows that this protein is essential for flagellar assembly [5]. Insertional mutants (“knock outs”) of IFT88 result in cells that “completely lack flagella” and


Electron microscopic analysis of these cells showed that the basal bodies were structurally normal but the flagella did not extend beyond the transition zone. In some cells, the membrane covering the flagellar tips was tightly apposed to the microtubules with no material between them and the membrane. In other cells, the flagellar stubs were slightly swollen and contained fragments of microtubules in random orientations. [5]


Some other features of the ift88 mutant are worth mentioning. First, growth and cell division were the same as that seen in wild type cells. This clearly indicates a flagellum-specific function. Secondly, IFT rafts were also not seen in the transition zone. This indicates that without IFT88, the rafts don’t assemble. Thirdly, another raft protein, IFT57, was present in much lower levels in the ift88 mutant (as shown by western blotting). One possible explanation for such reduced levels of IFT57 is that IFT88 might be needed to stabilize the tertiary structure of IFT57. Without IFT88, it unfolds and is picked up by the degradation machinery.

IFT88 from Chlamydomonas (green algae) is composed of 782 amino acids. It shows 42% sequence identity with its homolog in mice [3]. However, a BLAST search using Chlamydomonas hits many non-flagellar proteins. It is difficult to conclude any of these are true homologs. This is because IFT88 contains ten tetratricopeptide (TPR) motifs. The TPR motif is a commonly found domain that functions exceptionally well for scaffolding purposes and is thus involved in many protein-protein interactions. It is not unusual to find these motifs among multiprotein complexes, such as the peroxisomal import receptor. Given that there are only a few thousand different protein folds [6], we can’t really score these proteins as homologs based on shared motifs. But we can’t effectively rule this out either. A more detailed analysis is required, for example, to determine if protein similarities exist apart from the TPR motifs.

The TPR motifs of IFT88 are clustered at amino acid positions 200-300 and 450-650 (roughly). I thus used BLASTED with the algal sequence using amino acids 1-200, 300-450, and 650-782. None of these regions contained a common domain, yet there was decent sequence conservation. When compared to the human version of IFT88, the algal version shared 35% sequence identity and 50% similarity among the first 200 amino acids. Among amino acids 300-450, there was 33% identity and 56% similarity. The C-terminal region (650-782) was even more conserved, showing 56% identity and 72% similarity. When all three algal sections (lacking TPR sequence) were used to BLAST, they retrieved only IFT88 homologs from other other species (with an E value less that 10^-4). All of this suggests that apart from the TPR motifs, IFT88 contains almost 200 positions with amino acids that are identical between algae and humans and are specific to IFT88. This in turn suggests that common descent is not a well supported explanation for the origin of IFT88.


IFT52

IFT52 is another complex B protein that has been genetically characterized. Elimination of this gene product (through insertional mutagenesis) in Chlamydomonas results in the same phenotype as knockout of IFT88 – no assembled flagella:


The flagellar membrane closely abuts the transitional zone, indicating that the active movement of raft proteins and their cargo along the flagella is essential for the assembly of any flagellar proteins into the axoneme. [7]


IFT52 from Chlamydomonas (green algae) is composed of 454 amino acids. It shows 49% sequence identity with its homolog in rats [3]. Deane et al. note, “IFT52 homologs were not present in GenBank databases for organisms that lack flagella or cilia (e.g., Saccharomyces cerevisae and [/i]Arabidopsis[/i]).” [8] BLASTing with Chlamydomonas sequence returns only IFT52 homologs and ‘hypothetical proteins’ (from the sequenced eukaryotic genomes), which are also likely to be IFT52 from different species. Just as we saw with IFT88, IFT52 does not appear to have any identifiable homologs in noncilial tubulin-based transport systems. What makes this all even more significant is that IFT88 and IFT52 have been shown to physically interact with each other [9].



IFT20

Mutation in IFT20 also prevents formation of flagella [9]. IFT20 is one of the smallest components, being composed of only 135 amino acids. It’s a unique protein, with none of the recognized domains, and algal sequence shows 32% identity to sequence found in mouse. [3]

IFT20 has been shown to physically interact with IFT57 and one of the motor subunits of kinesin II [10]. In other words, IFT20 plays the essential role of linking the kinesin motor to the rest of the IFT raft. That IFT20 binds directly to the motor subunit rather than accessory subunits of kinesin indicates it is rather unique. Also worth mentioning is that the same research describing the IFT20-kinesin interactions [10] also found that the interaction between the IFT raft and kinesin is ATP dependent.

IFT20 thus represents a third essential component without a homolog in bacteria or noncilial tubulin-based transport systems.



IFT172

IFT172 is a massive protein containing 1746 amino acids from the nematode, C. elegans (known as OSM-1). The algal sequence shows 32% identity and 51% similarity to the OSM-1 gene [11]. Mutations in IFT172 also lead to complete loss of flagella [10]. The first 250 amino acids of this protein contain several WD repeats. A WD repeat is another common protein domain found in various eukaryotic proteins that forms a stable propeller-like platform. This then facilitates protein-protein interactions.

Since WD repeats are common (and thus not informative regarding common descent), I BLASTed with amino acid sequences 251-1746 from C. elegans. This 1500 amino acid sequence contained no conserved domains. Such a search did not retrieve any bacterial sequences or any noncilial tubulin-based transport systems. It did pick up several “hypothetical proteins” (with 30-40% identity) from sequenced eukaryotic genomes. However, one interesting homolog was LIM binding factor found in rats. LIM binding factor is a protein that interacts with LIM homeodomain transcription factors. These transcription factors play important roles in the development of the pituitary gland and neuron formation in rats. Thus, it would seem likely that this new function was acquired later in evolution (providing an example of how a front-loaded protein could be maintained). Oddly enough, IFT172 in Chlamydomonas is more similar to the rat LIM binding factor (41% identity, 60% similarity) than it is to the C. elegans homolog. [11]

To help clarify if the transcriptional activity of this protein is derived and ancestral, I used the C-terminal 1500 amino acids of C. elegans IFT172 to BLAST the genome of Arabidopsis thaliana. Since Arabidopsis does not make flagella, and does not have the IFT machinery, homologs would reflect either the transcriptional activity or some ancestral role with the cytoskeleton. The only sequences retrieved were a few fragments (ca. 100 amino acids) with around 20% identity and E values well above 10^-4. Put simply, no homologs were found. The same analysis was conducted with the yeast genome (also without flagella) and the same results were found. These results support the idea that the massive protein, IFT172, is the ancestral protein and without any obvious precursor.



IFT140

IFT140 is a complex A protein and genetic studies show it too is essential for flagellar assembly [2,10]. It is generally similar to IFT172, in that it is a very large protein (1360 amino acids) and the N-terminal region contains WD repeats. Conducting a search analysis similar to that done with IFT172 (but using the entire IFT140 sequence) elicits similar results – retrieving ‘hypothetical proteins’ among the sequenced eukaryotic genomes, but no homologs in Arabidopsis, yeast, eubacteria, or archebacteria.



Other Considerations

In addition to the IFT proteins mentioned above, mutations in IFT46 also result in the inability to assemble flagella [10]. However, I have been unable to track down its sequence. Furthermore, another IFT protein has been recently isolated – IFTA. The algal version of this protein shares 50% identity with its human homolog [12].

Yeast two hybrid analyses have been used to determine which IFT proteins directly interact [10]. Thus far, the following specific interactions have been established: IFT81-IFT27, IFT81-IFT74, and IFT46-IFTA. IFT88 interacts with IFT52 and also can form a triad, IFT88-IFT81-IFT74.

Of course, the IFT proteins most likely interact with the flagellar cargo. In fact, such interactions would have to be considered part of the IC system involving IFTs, as kinesin/dynein motors that simply moved IFT rafts would be a functionless and thus a tremendous waste of energy and material. Recent work with antibodies shows that the IFT rafts co-precipitate with outer and inner dynein arms and radial spokes [10].

The IFT system plays a special role. Assembling something as complicated as the eukaryotic machine requires complex machinery in turn. Rosenbaum et al. explain the problem:


Unlike most organelles, which are surrounded by cytoplasm, the flagellum protrudes from the cell surface extending tens or even hundreds of microns into the external medium. This elongated organelle must import all the macromolecules required for its assembly, maintenance, and function including >200 polypeptides that make up the microtubular axoneme (Dutcher, 1995), all the constituents of the flagellar membrane, as well as a prodigious amount of ATP to supply the thousands of dynein motors that drive flagellar motility. [13]


Cole et al. highlight the same problem[14]:


Assembly and maintenance of the eukaryotic flagellar axoneme presents some unique problems for the cell. . The organelle is composed of well over 200 different polypeptides that, after their synthesis in the cytoplasm, must rapidly find their way to the distal tip of the flagellum where most of the assembly of the microtubular axoneme takes place.


Rosenbaum et al. expand on this:


A dramatic example of the delivery of molecules into the flagellum is seen during flagellar regeneration in the biflagellate alga Chlamydomonas: flagella 10 µm long are assembled in ~1 h. As the organelle elongates, flagellar precursors must reach the site of assembly at the distal tip (Rosenbaum and Child, 1967; Johnson and Rosenbaum, 1992), which grows farther and farther away from the site of protein synthesis. The site of tubulin addition during flagellar assembly was identified by fusing cells with half-length flagella to cells containing epitope-tagged tubulin: all the tagged tubulin incorporated into the growing flagella at their distal tips. When cells with full-length flagella lacking radial spokes were fused to wild-type cells, radial spokes from the wild-type cytoplasm entered the spokeless flagella, assembled at the distal tips of the flagella, and gradually continued assembly toward the base (Johnson and Rosenbaum, 1992). Similar results were obtained with inner dynein arms (Piperno et al., 1996). Thus, there appears to be a mechanism for transporting axonemal precursors to the distal tip of the flagellum, whether or not it is elongating.


Also, when a temp-sens. mutant of kinesin-II was shifted to the non-permissive temperature, the fully formed flagella began to shrink and were eventually reabsorbed. With nothing new going in, the only thing happening was that stuff was coming out (due to turnover). The flagella are dynamic structures such that assembly and maintenance blur into one. All of this indicates the IFT process is not some luxury add-on, but instead is at the core of both flagellar assembly and existence.

A final interesting feature of IFT is described by Cole [3]:


Perhaps the most important role of IFT is not long distance travel within the flagellum but entry into and exit from the organelle itself. It is possible that IFT functions at the basal bodies to assist in the selection of proteins that will enter the flagellum. Immuno-electron microscopy has been used to show IFT proteins are docked onto the transition fibers that run between the basal bodies and the membranes (29). Since all proteins that enter the flagellum must pass by these fibers, it may be that the transition fibers act as a docking or staging area for IFT. Furthermore, this region may act as a filter which prevents nonflagellar proteins from entering the flagellum. If a protein or protein complex is not bound to the incoming anterograde train of IFT particles, then that protein will not enter the flagellum. If this hypothesis were true, then we could predict, that in the absence of anterograde IFT, there would not only be no flagellar assembly but there would also be no accumulation of material past the basal bodies and transition zone. This is exactly what is observed with the IFT mutants fla10 null, ift88, bld1, ift172, and ift140.


Immunofluorescent staining of IFT172 in Chlamydomonas shows this IFT protein localizes specifically to the region just beneath of the flagellum[14]. What’s more interesting is that these results come from the bld2 mutant which lacks basal bodies. While the existence of the basal body was determined to be necessary to localize the kinesin II motor, as can be seen from the above figure, it was not necessary to localize IFT172 (and thus, most likely, the IFT raft machinery). Rather than viewing the flagellum as simply an outgrowth of the basal body, these data suggest that the basal body is more akin to an IC component needed to form the flagellum in conjunction with the IFT machinery.



Summary

While there is still much more to be learned about the IFT machinery, a distinct IC picture is emerging. Genetic data clearly indicate IFT88, IFT52, IFT20, IFT172, IFT140, and IFT46 are required for flagellar synthesis. Mutations in these genes result in the serve “bald” phenotype but have no effect on cell growth or division, indicating a flagellar-specific function. These proteins show good sequence conservation, where protozoan and human sequences share 30-50% sequence identity. None of the proteins appear to have homologs among eubacteria, archaebacteria, eukaryotes lacking flagella, or noncilial tubulin-based transport systems.

Furthermore, the IFT proteins must specifically interact with two other classes. First, they must be able to interact with the kinesin and dynein motors. Secondly, they must be able to interact with the flagellar cargo – radial spokes, etc.

The IFT proteins appear quite special in light of their function. Their ability to screen out non-flagellar proteins is important, as mistakenly transporting cytoplasmic proteins into the flagellum would likely be deleterious. Yet when functioning as IFT rafts, the same proteins cannot bind too tightly to their cargo, otherwise this would hinder their ability to unload the cargo at the flagellar tip. Furthermore, they must be able to transport various different structural components (i.e., radial spokes and dynein arms). Thus, the IFT complex is specific enough to exclude cytoplasmic proteins, but not too specific such that it would be unable to transport the various different flagellar components into the flagellum.

References

1. http://www.idthink.net/biot/eflag/index.html
2. Cole, D. 2003. The intraflagellar transport machinery of Chlamydomonas reinhardtii. Traffic 4: 435-442.
3. Gregory J. Pazour, Sheila A. Baker, James A. Deane,Douglas G. Cole,Bethany L. Dickert,Joel L. Rosenbaum,George B. Witman,and Joseph C. Besharse. 2002. The intraflagellar transport protein, IFT88, is essential for vertebrate photoreceptor assembly and maintenance. JCB 157:103-113.
4. Silfow, CD and Lefebvre PA. 2001. Assembly and motility of eukaryotic cilia and flagella. Lessons from Chlamydomonas reinhardtii. Plant Physiology 127: 1500-1507.
5. Pazour GJ, Dickert BL, Vucica Y, Seeley ES, Rosenbaum JL, Witman GB, Cole DG. 2000. Chlamydomonas IFT88 and its mouse homologue, polycystic kidney disease gene tg737, are required for assembly of cilia and flagella. J Cell Biol. 151:709-18.
6. http://www.idthink.net/biot/denton/index.html
7. Brazelton WJ, Amundsen CD, Silflow CD, Lefebvre PA. 2001. The bld1 mutation identifies the Chlamydomonas osm-6 homolog as a gene required for flagellar assembly. Curr Biol. 11:1591-4.
8. Deane JA, Cole DG, Seeley ES, Diener DR, Rosenbaum JL. 2001. Localization of intraflagellar transport protein IFT52 identifies basal body transitional fibers as the docking site for IFT particles. Curr Biol. 2001 11:1586-90.
9. Meeting Report: Tenth International Conference on the Cell and Molecular Biology of Chlamydomonas. Protist, Vol. 153, 325–336, December 2002.
10. Baker SA, Freeman K, Luby-Phelps K, Pazour GJ, Besharse JC. IFT20 links kinesin II with a mammalian intraflagellar transport complex that is conserved in motile Flagella and sensory cilia. J Biol Chem. 2003 Jun 23
11. http://www.biology.duke.edu/chlamy/abstracts/millerm.html
12. http://www.biology.duke.edu/chlamy/abstracts/cole2.html
13. Joel L. Rosenbaum, Douglas G. Cole, and Dennis R. Diener. 1999. Intraflagellar Transport: The Eyes Have It. J. Cell Biol.144: 385-388.
14. Douglas G. Cole,Dennis R. Diener, Amy L. Himelblau, Peter L. Beech, Jason C. Fuster,and Joel L. Rosenbaum. 1998. Chlamydomonas Kinesin-II–dependent Intraflagellar Transport (IFT): IFT Particles Contain Proteins Required for Ciliary Assembly in Caenorhabditis elegans Sensory Neurons. The Journal of Cell Biology 141: 993–1008.

 

TeleoLogic