Connecting biopharmas with the right preclinical cancer models and CRO partners


Posted by Richard, December 2017

Exploring the literature on SPG15 mechanisms


Back in July, one of our users told us about Maddi Thurgood, a 16-year-old girl diagnosed with SPG15, and asked if we knew of any relevant data or contacts. SPG15 is a neurodegenerative disorder within the category of Hereditary Spastic Paraplegias. In this modern era of genomics, individual members of this group are designated based upon which gene is disrupted to cause the disease (or at least which genomic region the search has been narrowed down to, when a specific gene within the region has yet to be implicated). In Maddi's case, the specific diagnosis of SPG15 corresponds to disruption of the ZFYVE26 gene - an identification made only within the past decade by Hanein et al. (2008)1. In the tradition of naming genes and proteins for the diseases that result from their malfunction (since the normal role of the protein is often not discovered until later), Hanein et al. proposed the name 'spastizin' (from 'SPASTicity due to the ZFYVE26 proteIN') for the protein encoded by ZFYVE26. Although the genetic cause of the disease is known precisely, there are as yet no drug treatments, so the current approach being pursued to treat Maddi is editing of the genome in her cells to correct the sequences of her ZFYVE26 genes.

In the absence of any direct hits for SPG15 or ZFYVE26 / spastizin on our platform, Craig made this initial request to our users. For me this raised the question of what types of human genomic data might be relevant. For gene editing, the relevant data would presumably be the proportion of targetted cells in which the correction was made successfully by a trialled technique and the nature of any off-target edits. I also wondered, however, what data would be useful to, for example, the folks at Healx, whose expertise is in 'intelligently matching drug treatments' to rare diseases, if they were working on SPG15? Is ZFYVE26 in some pathway such that its loss could be worked around by finding a drug to act on molecules downstream of it? Is it performing some function for which there a backup pathway that could be upregulated by drug treatment? At a more basic level, what does the spastizin protein that ZFYVE26 encodes do?

First Impressions of the Literature

Through reading journal articles about SPG15, I found that researchers have explored the effects of damage to the ZFYVE26 gene in cell lines and animal models and investigated where its normal protein is found within cells to try to determine precisely how it contributes to the functioning of a cell. At this point, however, I need to issue the disclaimer that (although I did once run a gel!) I am definitely not a wet-lab biologist. Thus I am not versed in the strengths, weaknesses and caveats of the various assays and protocols that have been deployed by different groups as evidence for or against spastizin having a particular role - not to mention the relevance of the different species and cell type contexts reported^1.

Another challenge in reading such articles is the different names that genes and their protein products are given over time, starting with effectively a catalogue number, then being renamed according to implication in some disease and perhaps finally being renamed again based on a discovered function that is disrupted in the disease. Genes may also be named based on their membership (usually detected by similarity in sequence) of a gene family in which another member has already been extensively characterised. ZFYVE26 was initially KIAA0321 in the Kazusa DNA Research Institute catalogue of genes. Its current official name is due to it being one (assigned the number 26) of 31 members of the 'Zinc fingers FYVE-type' or ZFYVE family of genes that have the ZFYVE 'domain' (a functional module that occurs in multiple proteins)^2. Before the exact gene involved in SPG15 was identified, the placeholder gene name 'SPG15' was used, so ZFYVE26 has now acquired this as an additional synonym. The protein has also been referred to as FYVE-CENT ('FYVE domain-containing centrosomal protein') by Sagona et al. (2010)2.

Suggested Roles

Eight (not all unrelated) roles have been proposed for spastizin. I have outlined them here in the chronological order of the first papers proposing them ^3.

Endosomal trafficking

Hanein et al. (2008)1 identified ZFYVE26 as the gene disrupted in SPG15^4. Among the molecular assays they performed to support and follow up this discovery were ones to find out where the spastizin protein it encodes is present within the components of a cell. The cells used were from a cell line (COS-7) derived from a monkey kidney, specifically from fibroblasts - a type of cell that produces connective tissue ^5. They found that ZFYVE26 'co-localised' with molecules that are 'markers' for 'endosomes' (see glossary below) because they occur primarily in endosome membranes and not in other cell components. They also noted that in proteins encoded by other members of ZFYVE26's gene family, the 'FYVE' domain is thought to enable them to bind to endosomes ^6. More specifically, this is thought to be the means by which they contribute to 'endosomal trafficking' of (other) proteins, i.e. transporting of proteins through the network of endosomes.


Sagona et al. presented evidence in 2010 2 that ZFYVE26 is involved in cytokinesis - the actual division of a cell after replication of its genome and separation of the copies. Interestingly in the context of the proposal by Hanein et al., among the interaction partners involved in the mechanism described is an 'endosomal sorting complex' (CHMP4B). In a 2011 follow-up paper 2b, they reported that a missense mutation (R1945Q) in ZFYVE26 (in the human breast cancer cell line, HCC-1954) is accompanied by cytokinesis arrest and thus cells with multiple nuclei. Using human T-lymphocytes (a type of white blood cell), they found that the protein Beclin 1 interacts with spastizin; in the breast cancer cell line with the R1945Q mutation this interaction did not occur. Significantly in the context of the cytokinesis role for spastizin, the localisation of Beclin 1 to the bridge between the separating daughter cells that they observed with control cells was greatly reduced in the cell line with the spastizin mutation.

Working with mouse oocytes, You et al. (2016)9 confirmed that knockdown of Beclin 1 resulted in impaired cytokinesis. Interestingly, they also found a converse localisation consequence result to that of Sagona et al. - the knockdown of Beclin 1 reduced recruitment of (the mouse version of) spastizin to the 'midbody' of the separating daughter cells.

DNA repair

Also in 2010, S-B³abicki et al. 3 reported their results of a screen for genes involved in (a specific type of) DNA repair. Out of their hit list of 61 genes, many were already known or suspected to be involved in DNA repair (supporting the efficacy of their assay). Among the others they focused on one that at the time had only a catalogue number - KIAA0415. They found hereditary spastic paraplegia patients, with no previously identified genomic cause of their condition, who had disruptive mutations in this gene that were not seen in many controlsAlong with other evidence, this led them to propose the renaming of KIAA0415 to SPG48 corresponding to a new category of HSP. The relevance to SPG15 (and SPG11) and what led them to a connection with HSP in the first place, was that both spastizin (associated with SPG15) and spatacsin (associated with SPG11) were among four interaction partners identified for the SPG48 protein (the other two being protein products of genes with catalogue numbers C20orf29 and DKFZp761E198). Although ZFYVE26 / spastizin was not a primary hit in the DNA repair screen, a follow-up experiment provided evidence for its involvement in this process.-A

Accessory protein of an Adaptor Protein Complex / Late endosomal trafficking

Hirst et al. (2011)4 started from the protein encoded by a gene with catalogue number C14orf108 and found that it was a binding partner of the DKFZp761E198 protein. They used sensitive searches to find sequence homology (similarity) evidence that these two proteins corresponded to two particular members (components) of each of the four previously known 'adaptor protein complexes'. To quote their definition, 'Adaptor protein (AP) complexes sort cargo into vesicles^8 for transport from one membrane compartment of the cell to another'. Using the interaction results from S-B³abicki et al. 3 above, they then identified the KIAA0415 / SPG48 and C20orf29 proteins as the remaining two components of what they named 'Adaptor Protein Complex 5' (AP-5). To account for the additional interactions, they proposed that spastizin (SPG15) and spatacsin (SPG11) are accessory proteins for AP-5, i.e. they interact with AP-5 to perform some functions but are not core components.-A

Working with HeLa cells (a human cervical cell line comonly used in research)^9, they demonstrated that AP-5 cyclically attaches to and detaches from the membranes of (probably) late endosomes and/or lysosomes (due to partial colocalisation withe LAMP1 marker protein). (See glossaries below for more details.) Based on these and other lines of evidence they suggested that AP-5 (in conjunction with spastizin and spatacsin) contribute to transport of reusable proteins out of late endosomes (by budding of veiscles) before they would otherwise be broken down by the fusion with lysosomes. This is a more specific form of the original proposal by Hanein et al. (2008)1.

In follow-up investigations (Hirst et al. (2013)4b), they found that the association of spastizin and spatacsin with AP-5 was present not only on membranes but also in the cytosol (the liquid inside cells) and that within this association there were always equal numbers not only of the individual AP-5 component proteins but also of spastizin and spatacsin - suggesting a very stable linkage. As already noted by Hanein et al. 1, the 'FYVE' domain that spastizin includes is used by other proteins containing it to bind to endosomes. They thus proposed that the particular role of spastizin is to act as the link attaching ('docking') the entire AP-5 complex to the endosome membrane and supported this by finding that removal ('knockdown') of spastizin resulted in AP-5 remaining only in the cytosol.

Khundadze et al. (2013)6 reported the generation of a mouse model for SPG15, in which the 'knockout' of the Zfyve26 gene ^10 reproduced many human SPG15 symptoms with onset at 12 months of (mouse) age. As in humans, there is no sign of a developmental disorder; when adjusted for relative lifespan, the onset of symptoms in mice is actually later than is typical in humans. Their localisation and functional assays of the Zfyve26 protein led them to concur with a role in the endosome-lysosome pathway.

Motor neuron axon development

Meanwhile, Martin et al. (2012)1c followed up from Hanein et al. (2008)1 and from localisation (both at the level of tissues and within neurons) investigations (in human and rat) of spastizin and spatacsin by Murmu et al. (2011)1b by generating zebrafish models of SPG15 and SPG11. They thus found that the zebrafish versions of both spastizin and spastacsin are important for proper development of the axons of spinal motor neurons.

Autophagy : fusion of autophagosomes with lysosomes

The interaction between Beclin 1 and spastizin, discovered by Sagona et al. (2011)2b in the context of cytokinesis (see earlier), prompted Vantaggiato et al. (2013)5 to investigate whether spastizin is involved in autophagy (recycling of cellular materials - see glossary). Their reasoning was that autophagy is among a variety of other processes in which Beclin 1 had already been implicated; in particular that it mediates the formation of complexes involved, respectively, in the formation and maturation of autophagosomes (structures packaging materials for recycling - see glossary). Within their definition of 'maturation of autophagosomes', the authors include the final step of the fusion of autophagosomes with lysosomes to form the autolysosomes that actually break down the material for recycling ^11.

Using lymphoblasts (white blood cells) and fibroblasts from four SPG15 patients with different spastizin mutations and from control subjects, they found that spastizin normally interacts not simply with Beclin 1 by itself but with a complex of the proteins Beclin 1, UVRAG and Rubicon that is involved in autophagosome maturation. The spastizin mutations in the SPG15 patient cells disrupted this interaction and resulted in accumulation of immature autophagosomes; the same effect was seen in in a human neural cell line (derived from a tumour) and normal mouse (primary hippocampal) neurons when artificially depleted of spastizin. Vantaggiato et al. (2014)5b reviewed these findings.

Renvoisé et al. (2014)7 investigated fibroblast cell lines derived from skin biopsies of two newly identified SPG15 siblings (along with corresponding cell lines from SPG11 patients in two other families). They found both enlarged lysosomes and subcellular structures containing material that would normally be broken down in lysosomes. They interpreted their results as supporting the proposal by Vantaggiato et al. (2013)5 of defective lysosomal fusion with autophagosomes (see glossary below) in SPG15, while allowing that the enlarged lysosomes could result from an 'endolysosomal trafficking defect' as proposed by Hirst et al. (2013)4b.

Axon transport of dense core vesicles

Kanagaraj et al. (2014)8 report evidence that in zebrafish oocytes (egg cells) spastizin controls secretory vesicle ^13 maturation. Although oocytes are a very different type of cell from neurons (and zebrafish are much more distantly related to humans than mice), they draw parallels with the secretory vesicles in neurons that release 'neurotrophic factors' ^14 into their synapses.

They suggest that the need to transport neurotrophic factors from the cell body, all the way along the axon, to the synapses explains why it is the neurons with the longest axons (motor neurons to the legs) that are the first to degenerate.

Autophagy : autophagic lysosome reformation

Returning to autophagy, Chang et al. (2014)7b found evidence that spastizin and spatacsin play a vital role in the production of new lysosomes from autolysosomes (the containers resulting from the fusion of autophagosomes with lysosomes). In HeLa cells, they found that spastizin location overlapped most with markers for lysosomes rather than for early or late endosomes, but that when the FYVE domain was removed or disrupted the spastizin was dispersed throughout the cytosol. Again making use of characteristic markers, they also found that both autophagosomes and enlarged autolysosomes accumulated in the cells when levels of spastizin or spatacsin were artificially reduced. (The accumulation of autophagosomes had previously been reported in fibroblasts from SPG15 and SPG11 patients 5.) Another effect of this spastizin or spatacsin depletion was a lack of 'lysosomal tubules' budding from the autolysomes after nutrient starvation. Tubule budding is seen in normal cells under such a condition as part of a process of lysosome regeneration that is an alternative to the standard pathway of lysosome generation.

Further experiments distinguished the role of spastizin and spactasin as being in the initiation of this 'autophagic lysosome reformation (ALR)' process rather than contributing to subsequent stages, such as stabilising the tubules as they extend or causing their separation from the autolysosome to begin forming new lysosomes. For example, depleting cells of a substance (PI4KB) known to regulate the separation ('scission') of the tubules resulted in unusually long tubules but when spastizin was depleted at the same time tubules were not seen at all - suggesting that the lack of spastizin was not simply causing tubules to separate as soon as they started to form and thus never ceome visible.

Chang et al. then moved on to consider normal ('feeding') conditions, in which lysosome reformation mainly occurs by the production of vesicles - as mediated by a protein called PI4KB. They found that depletion of spastizin (or spatacsin) resulted in a lack of free lysosomes (i.e. ones not fused to autophagosomes as autolysosomes) in this case as well. They also observed that spastazin (along with spatacsin) interacts with PIK4B (in a manner not involving spastazin's FYVE domain) but with no effect on the level or location of PIK4B, suggesting that instead PIK4B may be regulating the activity of spastazin.

Coherence and relevance to neurodegeneration in SPG15

Scientific research usually builds upon previous findings and I hope that my attempts above to summarise the relevant literature reflect that this is true of the quest to determine the function(s) of spastizin. Clearly, however, there have been a number of disparate proposals (or at least groups of proposals). Naturally, many of the authors making novel proposals for the role of spaztizin responded to previous proposals. In some cases they performed assays aimed at reproducing previously reported effects (though not as far as I am aware with the same set-ups) and failed to find them. In others they designed experiments to rule out the other roles - though again often in different contexts. Some authors accepted that previously proposed effects might be real but considered them to be indirect or secondary; others reinterpreted previously reported evidence. It does not appear to me, however, that the articles in chronological order reflect as yet convergence to a consensus. It is possible that this simply reflects a reality of multiple roles for spastizin (perhaps as different isoforms) or perhaps some of its effects actually are indirect.

Allowing for the moment that spastizin might have all of the roles proposed, what can be said about their relevance to neurodegeneration - the most urgent consequence of SPG15? Firstly, any putative effect on cell division seems unlikely to explain the loss of mature corticospinal neurons directly, since they no longer undergo such division. It is possible that the direct effect would be upon some type of supporting cell. For example, oligodendrocytes provide the insulating myelin sheath of corticospinal neurons and are known to divide^15. As far as I am aware, however, lack of myelination is not a reported feature of SPG15. Similarly, since (human) SPG15 patients appear to have normal motor function into their teens, it not obvious to me that the impairment of motor neuron axon development found in zebrafish with knockdown of the protein corresponding to spastizin is relevant to SPG15 neurodegeneration.

With regard to the others, one characteristic of SPG15 that intrigues me is the 'length-dependent axonopathy of corticospinal motor neurons' 7, meaning that its effects are first noticed in those neurons in the motor cortex of the brain with the longest axons - those connecting to the junction, near the bottom of the spine, with 'lower motor neurons' that carry signals onwards to the leg muscles. Ideally any proposed function of spastizin relevant to SPG15 neurodegeneration should incorporate an explanation of this feature, though it is possible that the relevant mechanism may become clear only at some time after other evidence becomes conclusive.

Impaired DNA repair seems plausible as a cause of cell degeneration. As I noted, above, however, S-B³abicki et al. 3 did not identify ZFYVE26 as a repair gene in the original screen - only the follow-up. A neuron with a longer axon would presumably need to transcribe some of its genes more frequently - both during its development and for subsequent maintenance. Does this reduce its tolerance to DNA repair relative to neurons with shorted axons? Is it worth comparing the state of genomes of such neurons in SPG15 and control model organisms?-A

My understanding of the interactions of endosomes, lysosomes, autophagosomes and autolysosomes is too superficial for me for say anything other that, among their roles, the experimental evidence seems to be inclining towards spastizin mutation impacting autophagy. Just as a neuron with a longer axon will have to produce more materials to support itself, it will have to recycle more materials. It is less clear to me, however, how the various processes scale relatively in this case. Does lysosome and autophagosome production scale up with neuron size or do a proportionately smaller number have a heavier workload that exposes impaired functioning? Referring to the axon-length dependence, Hirst et al. (2013)4b note that 'Late endosomes and lysosomes are found mostly in the neuronal cell body, but some are present in axons, where they are transported mainly in the retrograde direction', prompting them to suggest the possibility of mutated spastizin (or spactasin or AP-5 proteins) affecting such axonal trafficking (or less specifically 'impairing axonal maintenance'). As far as I am aware, this still 'remains to be determined'.

Although their proposal for the role of human spastizin (involvement in the transport of neurotrophic factors by dense-core vesicles from the cell body to the synapse) seems to require some more speculative connections from their direct evidence (in zebrafish oocytes), Kanagaraj et al. 8 are unique in citing axon-length depence in its support. I did not, however, follow the connection from vesicle maturation to vesicle transport.


While clarification of spastizin activity in human cells and especially neurons still seems to be required, my impression is that spastizin has more of a structural role than a signalling role. This would seem to limit the scope for bypassing it or working around it and reinforce the need to fix deleterious mutations in ZFYVE26 by genetic engineering.

As per my disclaimer, however, I am a novice in almost every field relevant to the literature I have been reviewing. Please let me know if I have conveyed any information incorrectly or with major omissions or biases or if my questions, comments or conclusion are naif. More importantly, if you have any pertinent information or data that Maddi's medical advisers might not be aware of, please let us know so it can contribute to a cure for patients such as Maddi. Thank you in advance. Note that Repositive is making donations to the SaveOurMaddi campaign for the first 10 leads on data, contacts or research related to SPG15.

Glossary of cell components and processes

Autophagy ( is recycling of cellular materials - either because they are damaged or because their components are needed more urgently for some other purpose, e.g. for energy during starvation.

Autophagosomes ( are spherical, membrane-bound structures within the cell that, during their formation, engulf damaged proteins or even excess or defective whole organelles like mitochondria for subsequent recycling when the autophagosome fuses with a lysosome (see below) and is then known as an 'autolysosome'.

Endosomes ( are 'sorting offices' for molecules within the cell. There are a number of types of endosomes and one type can change into another. For example, 'early endosomes' mature into 'late endosomes', which then fuse with lysosomes (see below) resulting in the breakdown of whatever cargo of molecules they contain at that stage. The different types can be distinguished biochemically by characteristic proteins embedded in their membranes, with replacement of these proteins between stages. Cargo molecules can be transported to endosomes from the cell membrane (either having previously been embedded in the cell membrane to serve some function or after being imported from outside the cell) or they can be proteins produced inside the cell and transported from a 'distribution centre' (the 'Golgi complex' - see below). Transport also occurs in the opposite direction in both these pathways, such that, for example, proteins arriving from the Golgi complex can be routed to the cell membrane for secretion. Endosomes can also fuse with each other (combining their cargos) or divide into new endosomes (splitting their cargos). They can also fuse with autophagosomes, resulting in 'amphisomes'.

Golgi complex ( - a 'distribution centre' that receives newly manufactured proteins and lipids (from the 'rough endoplasmic reticulum'), modifies some of them and packs them into vesicles for transport to the appropriate destimation (to endosomes and thence to lysosomes; to the cell membrane or to outside the cell).

Lysosomes ( are the cell's waste disposal system or recycling system. They are intracellular containers of an acidic soluton of many different enzymes that can break down all the main categories of biomolecules (sections of protein, nucleic acids (DNA and RNA), carbohydrates and lipids (including fats and pieces of cell membrane)) - a bit like tiny stomachs inside cells.

Glossary of the molecular players

ZFYVE26 - damaging variants in this gene cause SPG15

Spastizin - the protein encoded by the ZFYVE26 gene

Spatacsin - the protein encoded by the SPG11 gene, named for the corresponding disorder; its function is tightly coupled to that of spastazin such that the SPG11 paraplegia and SPG15 are considered clinically indistinguishable

AP-5 - Adaptor Protein Complex 5, composed of a set of four proteins; its exact function is yet to be determined but other family members (AP-1 to AP-4) sort cargo into vesicles. Spastizin (the protein damaged in SPG15) and also spatacsin (the protein damaged in SPG11) have been shown to be associated with AP-5 both on endosome (or perhaps lysosome) membranes and in the cytosol. See part A of this figure.

LAMP-1 (Lysosomal-associated membrane protein 1) - a protein primarily found in lysosome membranes.


1 Hanein S. et al. (2008) Identification of the SPG15 gene, encoding spastizin, as a frequent cause of complicated autosomal-recessive spastic paraplegia, including Kjellin syndrome. Am J Hum Genet. 82(4):992-1002 PMID: 18394578 PMCID: PMC2427184

1b Murmu R.P. et al. (2011)^7 Cellular distribution and subcellular localization of spatacsin and spastizin, two proteins involved in hereditary spastic paraplegia. Mol Cell Neurosci. 2011 Jul;47(3):191-202. PMID: 21545838

1c Martin E. et al. (2012)^7 Spatacsin and spastizin act in the same pathway required for proper spinal motor neuron axon outgrowth in zebrafish. Neurobiol Dis. 2012 Dec;48(3):299-308. PMID: 22801083

2 Sagona A. P. et al. (2010)^7 PtdIns(3)P controls cytokinesis through KIF13A-mediated recruitment of FYVE-CENT to the midbody. Nat Cell Biol. 12(4):362-71. PMID: 20208530

2b Sagona A. P. et al. (2011) A tumor-associated mutation of FYVE-CENT prevents its interaction with Beclin 1 and interferes with cytokinesis. PLoS One. 6(3):e17086. PMID: 21455500 PMCID: PMC3063775

3 S-B³abicki M. et al. (2010) A genome-scale DNA repair RNAi screen identifies SPG48 as a novel gene associated with hereditary spastic paraplegia. PLoS Biol. 8(6):e1000408. PMID: 20613862 PMCID: PMC2893954(]-A

4 Hirst J. et al. (2011) The fifth adaptor protein complex. PLoS Biol. 9(10):e1001170. PMID: 22022230 PMCID: PMC3191125

4b Hirst J. et al. (2013) Interaction between AP-5 and the hereditary spastic paraplegia proteins SPG11 and SPG15. Mol Biol Cell. 24(16):2558-69. PMID: 23825025 PMCID: PMC3744948

5 Vantaggiato C. et al. (2013) Defective autophagy in spastizin mutated patients with hereditary spastic paraparesis type 15. Brain. 136(Pt 10):3119-39. PMID: 24030950 PMCID: PMC3784282

5b Vantaggiato C., Clementi E. and Bassi M.T. (2014) ZFYVE26/SPASTIZIN: a close link between complicated hereditary spastic paraparesis and autophagy. Autophagy 10(2):374-5. PMID: 24284334 PMCID: PMC5396092

6 Khundadze M. et al. (2013) A hereditary spastic paraplegia mouse model supports a role of ZFYVE26/SPASTIZIN for the endolysosomal system. PLoS Genet. 9(12):e1003988. PMID: 24367272 PMCID: PMC3868532(]

7 Renvoisé B. et al. (2014) Lysosomal abnormalities in hereditary spastic paraplegia types SPG15 and SPG11. Ann Clin Transl Neurol. 1(6):379-389. PMID: 24999486 PMCID: PMC4078876

7b Chang J., Lee S. and Blackstone C. (2014) Spastic paraplegia proteins spastizin and spatacsin mediate autophagic lysosome reformation. J Clin Invest. 124(12):5249-62. PMID: 25365221 PMCID: PMC4348974

8 Kanagaraj P. et al. (2014) Souffle/Spastizin controls secretory vesicle maturation during zebrafish oogenesis. PLoS Genet. 10(6):e1004449. PMID: 24967841 PMCID: PMC4072560

9 You S. Y. et al. (2016) Beclin-1 knockdown shows abscission failure but not autophagy defect during oocyte meiotic maturation. Cell Cycle. 15(12):1611-9. PMID: 27149384 PMCID: PMC4934058

^1: I do know, however, that proteins can have multiple isoforms (different versions of the protein produced by reading the same gene in different ways, e.g. by including different 'exons' - pieces of genes) having different functions. The isoforms produced (or at least their proportions) may depend on cell type or upon the conditions within a type of cell. The current Ensembl entry for ZFYVE26 reports that it has 11 different 'transcripts' (ways of being read) - 7 at the highest level of support. While some of the articles reported on experiments that could (at least, in theory) distinguish the isoforms involved (e.g. because they used probe sequences of DNA, known as 'primers'), my understanding is that others (e.g. those based on antibody markers) either were unable to make the distinction or did not consider it to be significant.

^2: In case you were wondering what the 'FYVE' means, these are simply the initials of four proteins in which it was first found - Fab 1, YOTB, Vac 1 and EEA1. See

^3: The non-standard numbering system I have used for the citations is that the first paper cited from a group is placed in chronological order and then follow-up papers from the same group (even if a different first author) have the same number but have a letter suffix added.

^4: They first correlated occurrence or absence of SPG15 in members of a set of affected families with inheritance of sections of DNA (within the region of chromosome 14 to which other researchers had already narrowed down the search by similar means). This constrained the region of interest to 2.64 million DNA bases containing 28 genes. Of these they examined promising candidates (based on existing information about function, tissues expressed in and/or gene family membership). The fourth candidate was ZFYVE26 and they found that, unlike in the first three candidates, disruptive mutations in it correlated with SPG15.

^5: As in many places in this post, I am simplifying a bit. See COS cells for more details.

^6: The type of entity in the endosome membrane to which the FYVE domain can bind is a (phosporylated) lipid called PI(3)P. See Phosphatidylinositol 3-phosphate for many more details.

^7: This article is behind a paywall, so I have read only the abstract.

^8: membrane-bound sacs in a cell. See Vesicle(

^9: It was originally derived from a cervical tumour but it is used widely in many types of cell biology research due to its ease of culture. See HeLa.

^10: Human gene names are entirely capitalised, mouse gene names are given an initial capital letter and the gene names of other species are all lower-case (and all should be italicised - unlike the names of the proteins they encode). The corresponding names mean that the genes in the different species have very similar but not identical sequences and it is thus a reasonable initial hypothesis that the proteins they encode perform the same role(s).

^11: In full, 'the nascent autophagosomes fuse with several types of vesicles from the endosomal/lysosomal pathway including late endosomes and lysosomes to create a fully functional degradative compartment, the autolysosome'.

^12: The term 'endolysomal' seems to me to gloss diplomatically over whether the focus should be on late endosomes or lysosomes.

^13: 'Secretory vesicles' are containers within the cell that migrate to the surface of the cell and release their specialised contents to the outside of the cell in a controlled manner.

^14: 'Neurotrophic factors' are chemicals that sustain the synapses and the other neurons to which they connect.

^15: At least according to this abstract of a review paper: Norton W. T. (1996) Do oligodendrocytes divide? Neurochem Res. 21(4):495-503. PMID: 8734444

#Related Posts 20 days of searching for Spastic Paraplegia Gene 15

The search for Spastic Paraplegia Gene 15 begins

"Why Repositive" - Mission Control