Exploring the role of Pex11p and Fis1p in Dnm1p localization
Interactions between Dnm1p, Pex11p and Fis1p
Marieke Mulder
University of Groningen June 2014
This report was written as a result of an 28 week research project of the master program of Molecular Biology and Biotechnology at the University of Groningen.
Research Group Molecular Cell Biology Supervisors
Dr. Chris P. Williams
Prof. Dr. Ida J. van der Klei
Index
ABSTRACT 3
INTRODUCTION 4
MATERIALS AND METHODS 8
STRAINS AND GROWTH CONDITIONS 8
MOLECULAR TECHNIQUES 9
PLASMID CONSTRUCTION 9
CELL FRACTIONATION 10
IMMUNOBLOTTING 10
SUCROSE DENSITY CENTRIFUGATION 10
CONFOCAL LASER SCANNING MICROSCOPY 10
DISTRIBUTION DNM1-GFP SPOTS 10
RESULT 11
PROTEIN LEVELS OF DNM1 11
DNM1P DOES NOT REQUIRE PEX11P OR FIS1P TO LOCALIZE TO THE PEROXISOME 11
ORGANELLES SHOW BINDING WITH DNM1 13
SUCROSE GRADIENT 13
MITO-DSRED SHOWS REMARKABLE MORPHOLOGY IN PEX11 15
DISCUSSION 16
FUTURE RECOMMENDATIONS 18
ACKNOWLEDGEMENT 19
REFERENCES 20
Abstract
The discovery of peroxisomes in eukaryotic cells in the 1950s lead to a whole new field of research. These versatile organelles, which have multiple functions, can form through fission of existing peroxisomes or arise de novo. Peroxisomal fission requires fission proteins such as the GTPase Dnm1p. Here we studied the localization of Dnm1p and its possible interactions with Fis1p and Pex11p, proteins with known roles in peroxisomal fission. Using confocal laser scanning microscopy, organelle fractionation and biochemical procedures we found that Dnm1p could localize to peroxisomes in absence of Pex11p and Fis1p. However deletion of FIS1 does effect the number of Dnm1-GFP spot in the cell suggesting that this protein plays a role in the release of the peroxisomal membrane.
Unexpectedly, we also observed that deletion of PEX11 disturbs the morphology of mitochondria.
Introduction
First discovered in 1954 by Rhodin[1], peroxisomes gained their name from Christian de Duve [2]. Peroxisomes are a core part of the eukaryotic cell and are considered as dynamic and multifunctional organelles. Peroxisomes contain a wide range of metabolic pathways which depends on organism, tissue and development stage. In yeast, peroxisomes house metabolic pathways that allows growth on different carbon sources such as oleic acid or methanol. Growth on such carbon sources causes the cell to increase the size and/or number of peroxisome.
To date 34 proteins, known as peroxins, are involved in peroxisomal biogenesis, including matrix and membrane protein import, division and inheritance[3]. Mutations in peroxisomal genes can lead to a set of peroxisomal biogenesis disorders such as Zelwegger syndrome[4]. Different malfunctions lead to different disorders and can even lead to death in childhood in severe cases.
New peroxisomes can form via 2 pathways; either de novo from the endoplasmatic reticulum or it forms through division by pre-existing peroxisomes through fission. When peroxisomes arise de novo, pre-peroxisomal vesicles bud from the ER and fuse with each other to form a mature peroxisomes[5]. In yeast it was observed that only in peroxisome deficient cells (lacking PEX3 or / and PEX19) do peroxisomes arise de novo [6].
Peroxisomes can also forms through fission, hereby an existing peroxisome divides (Figure1 [3]). The fission process can be divided in three stages; elongation, constriction and fission. Previous studies showed that the peroxisomal membrane protein Pex11p has a role in the early stages of fission, by causing changes in the peroxisomal shape which results in elongation [7][8][9]. Another very important group of proteins in the fission process are dynamin related proteins (DRPs). When elongated and constricted, peroxisomes recruit DRP’s which facilitates the membrane fission to form new peroxisomes. DRPs self-assemble around the fission site where the GTPase activity
Figure 1, An overview of the fission cycle. The fission cycle begins with membrane elongation mediated by Pex11p. Proteins associated with dynamin related proteins (DRPs) are recruited and the membranes become constricted. DRPs are recruited en mediates fission. Modified from the work of Smith & Aitchisun 2013[3]
causes fission. Recent studies show that fission protein 1 (Fis1), which was first found to play a role in mitochondrial fission, is now also involved in peroxisomal fission[9][10].
The exact role of Fis1p is still unclear, however it has been suggested that Fis1p recruits Dnm1p during fission.
The dynamin superfamily exist of a large number of GTPases that are involved in membrane remodeling. The GTPases that are responsible for fission and fusion in peroxisomes and in mitochondria belongs to the dynamin related protein (DRPs) family.
A number of domains can be found in both DRPs and the classical dynamin proteins, including, a GTPase domain, a Middle and a GTPase effector domain (GED) [11]. The Middle domain and the GED mediates the self-assembly and modulate the GTPase activity (Figure 2 A - B) [12][13][14]. In addition to these domains, DRPs also contain an additional domain, referred as Insert B. The function of Insert B was long unknown, however recent studies showed insight on its function. First the 3D-structure of Dnm1 (DRP in yeast) was determined in a GTPase bound state. Due to the 3D-structure and the genetics of Dnm1p, Mears et al proposed the idea that Insert B affects the interactions between the middle domain and the GED [15]. Further studies showed that mutations in Insert B disturbed the association between mitochondria and Dnm1p, and interactions between Dnm1p and Mdv1p, a protein required for mitochondrial fission. These finding indicate that Insert B has a role in the recruitment of Dnm1p to the fission site and where Mdv1p also plays a role [12].
Yeast contain multiple DRPs including Vps1p, Mgm1p and Dnm1p[16][17], where Dnm1p is the main DRP for mitochondrial or peroxisomal fission [18]. Mears et al showed that Dnm1p could mediate fission actively in mitochondria[15]. Cytosolic Dnm1p binds the fission site, however the mechanism for this is still unclear. Multiple Dnm1ps form a helix around the fission site and decreases the size of the helix, resulting in an increase of membrane curvature stress which leads ultimately in fission.
Figure 2, An overview off Dnm1p, Pex11p and Fis1p. (A) The structure of Dnm1p with 4 domains; GTPAse, Middle domain, Insert B. (InsB) and the GTPase effector domain (GED).
Modified from Bui and Shaw [13] (B) A folded structure of Dnm1p when cytosolic. Modified from Detmer and Chan [14] (C) The structure from pex11p with the amphipathic helix (AMP) and hydrophobic helices (MH) modified from Williams and Van der Klei [27]. (D) A folded structure of Pex11p. Modified from Opalinski[26]. (E) The structure of Fis1p with the tetratricopeptide (TPR) and the anchoring domain (M). Modified from Bui and Shaw [13]. (F) A folded structure of Fis1p.
Modified from Zhang and Chan [49]
Earlier we discussed the function Pex11p briefly. It was the first peroxin found to be involved in peroxisome proliferation and has been identified in different organisms.
Pex11p is part of the Pex11 family, containing not only Pex11p but also Pex11p-like proteins. In yeast the Pex11 family also contains Pex25p and Pex27p and the less related Pex11Cp [19] [20] [21] [22], however Pex11p was found to be the most important for fission. Mammals have 3 members in the Pex11 family: Pex11α, Pex11β and less related Pex11γ. Pex11α and Pex11β are homologous with yeast Pex11p and Pex11γ is homologous to Pex11Cp [23] [24] [25].
Pex11p is a peroxisomal membrane protein where both termini are directed to the cytosol [10] [26] Pex11p contains an amphipathic helix(a-helix) in the N-terminal and hydrophobic helices, which are suggested as transmembrane domains (Figu re 2 C –D) [7] [27]. Pex11p is responsible for remodeling the membrane during fission and deletion of PEX11 resulted in a smaller number and enlarged peroxisomes [19] [28] [29] [30].
Previous studies showed that the a-helix causes the remodeling of the peroxisomal membrane and is crucial for fission. Family member Pex25p also plays a role in peroxisome proliferation. It has been found that Pex25 has an role in de novo formation.
Cells lacking Pex11p and Pex25p result in cells containing one or no peroxisomes. Re- introduction of Pex25p and not Pex11p caused peroxisomes to arise de novo [21].
An important protein that has a role in fission is the adaptor protein Fis1p. Fis1p has been implicated in the recruitment of Dnm1p [8] [9] [31] [32]. It has been found that Fis1p can interact with Mdv1p and Mdv1p also interacts with Dnm1p [12] [32], where it has been suggested that Fis1p recruits Dnm1p via Mdv1p. However the result that Dnm1p can still localize to peroxisome in absence of Fis1p or Mdv1p[33], could indicate that there are additional proteins involved in the recruitment of Dnm1p. Shaw and Nunnari suggested that Fis1p has a second function in a later stage of fission [34].
Studies showed that deletion of both DNM1 and FIS 1 no longer recruited Mdv1p to the membrane, whereas deletion of only DNM1 localizes Mdv1p to the membrane [35]. They suggest that the interaction between Mdv1p and Fis1p, without the presence of Dnm1p, could indicate that there is a second role for Fis1p [34].
Fis1p is an transmembrane protein and was first found on the outer mitochondrial membrane, containing 2 domains: the membrane anchoring domain and the tetratricopeptide repeat (TPR) like domain (Figure 2 E – F). The TPR like domain in yeast consist of 6 α-helices, where the middle 4 α–helices form two TPR like motifs [36]. These two TPR like motifs in seems to form a concave binding surface in yeast. This is supported by the fact that in human Fis1p two TPR like motifs form the ligand binding site [37]. In yeast the end of the N-terminus (also referred N-terminal tail) is localized in to the concave. Deletion of the first 14-16 residues (the N-terminal tail) lead to an higher affinity for Dnm1p and a disturbed morphology of mitochondria, suggesting that the N-terminal tail regulates the concave binding surface [8] [32].
Dnm1p has been localized to mitochondria and peroxisomes in WT cells [18], however how Dnm1 is recruited is still unclear. Previously we showed that deletion of PEX11 or FIS1 resulted in enlarged Dnm1-GFP spots and immunoblots showed elevated levels of Dnm1-GFP when under control of the amino-oxidase promoter [38]. These data could suggest that Pex11p and Fis1p have a direct or indirect role in recycling Dnm1p although since these results were obtained using Dnm1p under an inducible promoter, care must be taken when interpreting them. To understand the role of Dnm1p better, we chose to place Dnm1-GFP under its own promoter. This construct was introduced into PEX11 and FIS1 deletion strains and in the PEX11/FIS1 double deletion strain, together with the peroxisomal marker DsRed-SKL or Mito-DsRed, to follow mitochondria. Using these strains, we followed the localization of Dnm1-GFP using microscopical and biochemical
approaches. Our results suggest that Dnm1-GFP is near peroxisomes in all strain checked, which could indicate that Dnm1p could localize to the peroxisomes, however association with peroxisomes needs to be established using biochemical approaches.
Deletion of FIS1 has an reducing effect on the number of Dnm1-GFP spots per cell. These result could suggest that Fis1p has an role in the recycling of Dnm1p. Unexpected results showed that deletion of PEX11 shows a disturbed mitochondria morphology when cells were grown on methanol but not on glucose, suggesting that the peroxisomal defects caused by deletion of PEX11 effects the morphology of mitochondria.
Materials and methods
Strains and growth conditions
Hansenula Polymorpha (H. Polymorpha) strains that were used in this study are listed in Table 1. Cells were grown on mineral media (MM) at 37 °C supplemented with 0.25 % glucose or with 0.5 % methanol as carbons source and 0.25 % ammonium sulfate or 0.25 % methylamine as nitrogen source [39]. When required, 60 μg/ml leucine was supplemented. Cells were pre-grown on MM containing glucose/ammonium sulfate. When cells reached an OD600 of ~ 2.0 they were transferred to MM containing methanol with ammonium sulfate / methylamine.
Growth on plates was on YPD (1 % peptone, 1% yeast extract and 1% glucose) media supplemented with 2% agar. When required, 100 μg/ml zeocin (Invitrogen) or 100 μg/ml nourseothricin (Werner Bioagents) was supplemented to the media.
Table 1, Hansenula Polymorpha strains used in this study
Strains Characteristics Reference
WT NCYC 495 leu1.1 (Saraya et
al.,2012)
Pex11p Pex11p deletion strain [30]
fis1 fis1 deletion strain [33]
Pex11pfis1 Pex11pfis1 double deletion strain [38]
WT dnm1-GFP WT with one copy integration of plasmid
pSNA01 [18]
Pex11p dnm1- GFP
Pex11p with one copy integration of plasmid
pSNA01 This study
fis1 dnm1-GFP fis1 with one copy integration of plasmid
pSNA01 This study
Pex11pfis1 dnm1-GFP
Pex11pfis1 with one copy integration of plasmid
pSNA01 This study
WT dnm1-GFP DsRed-SKL
WT dnm1-GFP with integration of plasmid
pSNA03 [18]
Pex11p dnm1- GFP DsRed-SKL
Pex11p dnm1-eGF with the integration of
plasmid pSNA03 This study
fis1 dnm1-GFP DsRed-SKL
fis1 dnm1-GFP with the integration of plasmid
pSNA03 This study
Pex11pfis1 dnm1-GFP DsRed-SKL
Pex11pfis1 dnm1-GFP with the integration of
pSNA03 This study
WT dnm1-GFP Mito-DsRed
WT dnm1-GFP with the integration of plasmid
pmm2 This study
Pex11p dnm1- GFP Mito-DsRed
Pex11p dnm1-GFP with the integration of
plasmid pmm2 This study
Molecular techniques
Plasmid used in this study are listed in Table 2 Standard recombinant DNA techniques and transformation of H. Polymorpha was performed as described earlier by Faber et al.
Transformed cells were grown on YPD plates as described before.
Insertions were checked with either colony PCR or by fluorescence microscopy. Colony PCR was performed with Phire Hot Start II DNA Polymerase (Fermentas) according to the manufacturer’s instructions. Primer that were used are listed in Table 3. Checking by fluorescence microscopy was done with the axioscope.
For cloning purpose Escherichia Coli DH5α (E. Coli) was used. Cells were grown at 37 °C in LB media supplemented with 100 μg/ml ampicillin. For agar plates 2 % agar was supplemented to the media.
Plasmid construction
For this study two plasmids were constructed: pmm1 and pmm2. For constructing pmm1 pVT100U Mito-DsRed was obtained as a kind gift from Prof. Neupert and digested with Hind III and XbaI and replaced the GFP-SKL in plasmid pHipZ18. Resulting in a plasmid containing Mito-DsRed under the alcohol dehydrogenase promoter (Padh) and Zeocin resistance marker.
For constructing pmm2, pmm1 was used. Pmm1 was digested with SacII and XbaI and inserted in plasmid pHipN4 between SacII and XbaI. This resulted in a plasmid containing Mito-DsRed under the pADH with a nourseothricin resistant marker. Both plasmids were linearized with munI for transformations in H. Polymorpha. Strains were checked by fluorescence microscopy for right insertion.
Table 2, Plasmids used in this study
Table 3, Primers used in this study
Plasmid Characteristics Reference
pSNA01 plasmid containing fusion gene between eGFP and C-terminal part of DNM1; zeoR, ampR
[18]
pSNA03 Plasmid containing PAOXDsRed-SKL, ampR, natR
[18]
pmm1 plasmid containing Mito-DsRed, ampR, zeoR This study pmm2 plasmid containing Mito-DsRed, ampR, natR This study pHipZ18-
GFP.SKL
Plasmid containing GFP.SKL under the PADH
promoter and zeoR [40]
pHipN4 Plasmid containing PAOX promoter and natR [41]
pVT100U MitoDsRed
Plasmid containing Mito-DsRed, ampR Prof. Neupert
Primer Sequence Reference
AOX int (F) CACCAGCGGATCTTCCTGG [38]
GFP int (R) GTGCAGATGAACTTCAGGGTCAGCTTG [38]
Dnm1 int (F) ATGGCACAAACGCTCTTGAC [38]
Cell Fractionation
Cell extracts were made as described by [42]. In brief, cells were treated with zymolase and broken with the potter homogenizer. This was centrifuged 2 times (3,000 g, 4 C) to obtain the post-nuclear supernatant (homogenate (H)). The homogenate was centrifuged at 30,000 g to obtain the organelle pellet (P) and the soluble fraction (supernatant (S)).
TCA extracts were made.
Immunoblotting
TCA extracts were prepared for SDS-PAGE and western blotting (WB) as previously described [43] [44].
Primary antibodies against Pex11p, GFP, AOX, DHAS, Pex3p and Pyc were used accordingly. Second antibody was a goat anti rabbit - horseradish peroxidase (HRP) or a goat anti rabbit - alkaline phosphatase (AP). Detection was with enhanced luminol-based chemiluminescent or with NBT/BCIP stock solution(Roche)
Sucrose density centrifugation
Organelles were separated based on their density as described in Douma et al [45]. In brief, the organelle pellet from the cell fractionation was dissolved in 40% sucrose in Buffer B (5mM MES (pH 5.5), 0.1 mM EDTA, 1 mM KCl) and 5-8 mg of protein was loaded on the sucrose gradient with an additional layer of 35% sucrose. The gradient was centrifuged at 18.000 rpm, 4 °C, SV288-rotor for 2,5 hours. Fractions of the gradient were harvest by a puncture in the bottom of the tube. The homogenate was also subjected to an sucrose gradient. Here homogenate (5-8 mg protein) was loaded on the sucrose gradient with an additional layer of homogenization buffer (5mM MES (pH 5.5), 0.1 mM EDTA, 1 mM KCl, 1.2 M Sorbitol) and subjected to the same centrifugation and sample harvesting as described before.
Confocal laser scanning microscopy
Fluorescence imaging was performed as described in Nagotu [33]. For imaging cells were grown on methanol for 16 hours after, images were taken after 10 and 16 hours.
Distribution Dnm1-GFP spots
To determine the distribution of Dnm1-GFP spots per cell in each strain, Dnm1-GFP spots in at least 100 cells, randomly chosen, were counted from each strain. Results were plotted in graph.
Result
Protein levels of Dnm1
To investigate the protein levels of Dnm1p in WT, pex11, fis1 and pex11/fis1 cells were grown on methanol and samples were taken after 10 and 16 hours. We show that WT has an comparable Dnm1-GFP protein level as fis1 cells and a lower level than pex11 or pex11/fis1 cells after 10h on methanol (Figure 3), unlike previous results [38]. pex11 and pex11/fis1 cells have an higher Dnm1p level after 10 hours, but this higher level can’t be seen after 16 hours. After 16 hours all strains seem to have an comparable Dnm1p level, suggesting that it could be that Pex11p has an role in Dnm1 in the early stage.
Dnm1p does not require Pex11p or Fis1p to localize to the peroxisome
Previously our group described that Dnm1 localizes between mitochondria and peroxisomes in H. Polymorpha WT cells [18]. Using the confocal laser scanning microscope images were taken from these strains containing a single copy of Dnm1-GFP and DsRed-SKL (Figure 4). Here we show that Dnm1-GFP spots in WT, pex11, fis1 and pex11/fis1 cells are sometimes near a peroxisome, however localization of the Dnm1-GFP is difficult to say based on these image alone. Therefore the function of Pex11p and Fis1p can-not be concluded based on these images.
However these images do lead to an unexpected observation; the amount of Dnm1-GFP spots per cell. When counted 100 cells from each strain, it became clear that there is a different distribution of Dnm1-GFP spots per strain. Where WT had 80 % of the cells with 1 or 2 (both ± 40%) Dnm1-GFP spots after 10 hours, fis1 has approximately 70 % of the cells with only one Dnm1-GFP spot and pex11/fis1 had in 30 % of the cells none spots (Figure 5A). After 16 hours 70% of the WT cell contain one peroxisome and the percentage of fis1 cells with no Dnm1-GFP spots increased (Figure 5B). This could suggest that Dnm1p builds upon deletion of FIS1 and has problems with the release of Dnm1p.
Figure 3, Protein levels of Dnm1-GFP. Immunoblots showing the Dnm1-GFP levels after 10 and 16 growth on methanol.
Figure 4, Images from WT, pex11, fis1 and pex11/fis1 with Dnm1-GFP and DsRed-SKL after 10 and 16 hours on methanol. Images show all strains at different time points where images were taken as described in materials en methods.
Figure 5, Number of Dnm1-GFP spots per cell after 10 and 16 hours. (A) Distribution of Dnm1-GFP spots per cell in WT, pex11, fis1 and pex11/fis1 cells after 10 hours. (B) Distribution after 16 hours.
Dnm1 is present in organellar pellets after fractionation
To investigate Dnm1 localization using biochemical approaches a fractionation was performed. Here the organelle pellet and the cytosolic supernatant was separated. Figure 6 shows the immunoblot results from the fractionation. These results show that Dnm1p can associate with the organelle pellet in all strains checked. Peroxisomal membrane proteins, peroxisomal matrix protein and cytosolic proteins were verified to see if the peroxisomes were intact in the organelle pellet. The peroxisomal matrix protein dihydroxyacetone synthase (DHAS) can been seen in the supernatant in all strains where the WT shows less DHAS in the supernatant compared to the other strains. The peroxisomal membranes proteins Pex3p and Pex11p were seen in the organelle pellet, however both can also be seen in the supernatant in fis1 and pex11/fis1 cells. The presence of DHAS in the supernatant indicates that some peroxisomes broke during the fractionation, which is substantiate in fis1 and pex11/fis1 cell where the peroxisomal membrane proteins can also be found in the supernatant. Although peroxisomes were broken during this fractionation, Dnm1-GFP is still presence in the pellet, indicating that Dnm1p associates with an organellar membrane.
Sucrose gradient
To determine if Dnm1p associate with peroxisomes, a sucrose gradient was performed on the homogenate and organelle pellet from WT cells. Both approaches were applied hence the homogenate does not undergo the resuspension after a 30,000 g centrifugation, which could result in disassociation with the organelle pellet . Note that only 4 mg of proteins was loaded on the organelle pellet whereas the homogenate was loaded with 8 mg of proteins. Both gradients show 2 protein peaks, one around fraction 5 and one around fraction 12-14 (Figure 7A). Peroxisomes are previously found around 52%
sucrose[45], which leads to believe that the peroxisomal fraction is around fraction 5. All fractions were used for immunoblots. The immunoblots from the homogenate gradient (Figure 7B) show that Dnm1-GFP is found in fraction 11 and 12 but not in fraction 5. The peroxisomal membrane protein Pex3 and the peroxisomal matrix protein AOX were found
Figure 6, Dnm1p localizes to the organelle pellet. Immunoblots from the fractionation show the homogenate (H) the organelle pellet (P) and the supernatant (S). Dnm1-GFP localizes to the organelle pellet in all strains.
Figure 7, Sucrose gradient of WT cells. (A) Graph showing the sucrose and protein
concentrations from the harvest fractions from the pellet gradient and the homogenate gradient.
(B) Immunoblots from the homogenate gradient where Dnm1-GFP, Pex3p and AOX were detected for all fractions and the homogenate. Suggesting that the peroxisome broke during fractionation.
(C) Immunoblots from the pellet fractionation where the same proteins were detected for all fractions, the homogenate and the supernatant after 30,000g centrifugation. Dnm1p is found in the peroxisomal fractions, however results from Pex3p and AOX lead to believe that peroxisomes broke.
Figure 8, WT and pex11 cells images with Dnm1-GFP and Mito-DsRed on methanol and glucose after 16 hours. Representative images were taken after 16 hours on methanol and on Hence the peroxisomal matrix is found outside the peroxisomal fraction lead us to concluding that peroxisomes broke during the fractionation. The pellet gradient (Figure 7C) shows a very unclear Dnm1-GFP band in fraction 12-16 and also in fraction 4-6. Here also Pex3p and AOX were detected outside the expected peroxisomal fraction, suggesting that here also peroxisomes broke during fractionation. Both gradients need to be repeated where the fractionation needs improving to obtain unbroken peroxisomes.
Mito-DsRed shows remarkable morphology in pex11
Figure 4 showed that Dnm1-GFP spot may localize to peroxisomes, however spots were also seen elsewhere. To see whether these spots could localize to mitochondria, a Mito- DsRed construct was introduced in WT and pex11 cells, to see whether Dnm1-GFP could localize to mitochondria. Results show that Dnm1-GFP spots are near the mitochondria when grown on methanol and glucose, suggesting a localization (Figure 8). Not all Dnm1- GFP spots were localized to the mitochondria, when grown on methanol, which could suggest that Dnm1-GFP localizes to both peroxisome and mitochondria. Remarkable are the mitochondria structures in pex11 cells after 16 hours on methanol. Whereas WT show tubular structures of mitochondria, pex11 cells show more dense mitochondria with a less tubular structure. This could suggest that deletion of PEX11 causes peroxisomal defects which effect the mitochondrial morphology.
Discussion
Previously we described that Dnm1-GFP under the amine oxidase promoter (PAMO) had a lower signal in WT when compared to pex11, fs1 or pex11/fis1 [38]. However, additional experiments suggested that the induction of Dnm1-GFP was incorrect in WT cells. Here we studied Dnm1-GFP under its endogenous promoter, avoiding over or under expressions of the protein. We found that protein levels from Dnm1-GFP in WT are comparable with fis1 cells and that pex11 and pex11/fis1 cells have an slight elevated level of Dnm1-GFP after 10 h grown on methanol. Previous studies show that deletion of Pex11p leads to a growth defect [28] [29] [30], therefore it could be suggested that the defect in growth leads to a different growth stage and could cause the different protein levels. Hence the defect of growth in deletion of PEX11, it could be that the protein levels of Dnm1p in cells deleted for PEX11 are not related to the interaction between Pex11p and Dnm1p but more related to the growth stage. Further studies where protein levels of Dnm1p is measured based on cell density could show if the elevated level is caused by growth stage.
Dnm1p has been localized to peroxisomes in WT and fis1 cells before [18]. Our study suggest that Dnm1 still co-localize to peroxisomes in absence of Pex11p and/or Fis1p (Figure 4), suggesting that these proteins are not required for Dnm1p binding to peroxisomes. However proper biochemical approaches are needed confirm these observations.
The amount of Dnm1-GFP spots between WT and fis1 and pex11/fis1 differs. Deletion of Fis1p seems to have a reduced number of Dnm1-GFP spots in the cell. Previous study confirms this phenomenon [33], were they also showed that Dnm1p is unable to detach from the mitochondrial or peroxisomal membrane when Fis1p or Mdv1p is deleted. This could indicate that Fis1p has a role in the recycling of Dnm1p during fission.
Results from the fractionation indicate that Dnm1p associates with the organelle pellet (Figure 6). But all strains show the peroxisomal matrix protein DHAS in the supernatant, suggesting that (some) peroxisomes broke during the process. Recent study showed that DHAS dimerizes prior to import[46], suggesting that the findings of DHAS in the supernatant could be due to dimerization in the cytosol instead of broken peroxisomes.
When comparing the supernatant and the homogenate, DHAS shows a stronger band in the supernatant for fis1, pex11/fis1 and pex11 cells than in WT cells. Considering that DHAS dimerizes prior to import, it could be suggested that deletion of FIS1 and/or PEX11 causes issues with the import of DHAS. To understand this connection better further studies are needed.
Observing Dnm1-GFP and MitoDsRed strains showed a disturbed morphology of mitochondria in deletion of PEX11 when grown on methanol only. This remarkable phenomenon, where deletion of PEX11 disturbs the morphology of mitochondria, has been seen before [47].There the mitochondrial structure of the heptocytes of in mice showed parallel crystals in the PEX11α deletion strain (after induction of proliferation of mitochondria an peroxisomes through a ciprofibrate diet). These cells also showed that mitochondria were more tightened around lipid vesicles than WT cells. These observation suggest that the peroxisomal defect, caused by deletion of PEX11, has an effect on mitochondria.
A recent study showed that Pex11p stimulates the activity of Dnm1p during peroxisomal fission [48]. The absence of PEX11 could reduce the velocity of Dnm1p’s GTPase activity on peroxisomes. The lower activity could than result in an slower fission of peroxisomes,
which may result in a lower concentration of cytosolic Dnm1p and could effect mitochdrial fission.
Another discussion point of these observations is the growth of pex11 cells. These cells have a slower growth rate than WT cells, where it was suggested that peroxisomal defects lead to a slower growth rate. However the disturbed morphology of mitochondria could also indicate that not only peroxisomal defect slows growth but that the disturbed mitochondria also influences the slow growth.
Overall it can be said that the relationship between Dnm1p, Fis1p and Pex11p is more difficult than anticipated. Although our results suggest that Fis1p may have an effect on recycling Dnm1p, further research is needed to understand the relations between these proteins.
Future recommendations
Many processes during peroxisomal fission are still unknown, also the roles and relationships between Dnm1p, Fis1p and Pex11p. To give further insight in these relationships further studies are needed.
Fluorescence microscope imaging of fixed cells containing Dnm1p- GFP, a peroxisome marker and a mitochondria marker could be helpful. Having peroxisomes and mitochondria both are marked in the cell, it can be seen if Dnm1p localizes to either peroxisomes or mitochondria, or that Dnm1p is somewhere else in the cell.
Also improvement of the fractionation could help. During fractionation cells and peroxisomes are vulnerable and need to be processed carefully. The process could be more done carefully. This may results in less broken cell, however it also leads to less broken peroxisomes. Less broken peroxisomes may not only lead to better Fractionation results, but also to better sucrose gradients. This biochemical approach can show, if processed well, if Dnm1p associates with the peroxisome and may give more insight in the role of Fis1p, Pex11p.
Deletion of PEX11 raises more questions when looked at the mitochondria. Does deletion of PEX11 cause peroxisomal defects that lead to a disturbed morphology of mitochondria or can it be completely different. To understand this observation better, more studies are needed.
Acknowledgement
I would specially like to thank Chris Williams for his enthusiastic and supportive guiding during this project. I also would like to thank Professor Ida van der Klei for giving me the opportunity to do my research project. Further I want to thank all members of the Molecular Cell Biology group for the support, the help and especially their kindness!
References
[1] Rhodin, „Correlation of ultrastructural organization and function in normal
experimentally changed convoluted tubule cells of the mouse kidney.,” Ph.D. thesis.
Stockholm, Aktiebolaget Godvil, 1954.
[2] C. De Duve en P. Baudhuin, „Peroxisomes (microbodies and related particles,”
Physiol Rev 46:323–357, 1966.
[3] J. Smith en J. Aitchison, „Peroxisomes take shape,” Nat Rev Mol Cell Biol 14:803- 817, 2013.
[4] R. Wanders en H. Waterham, „Peroxisomal disorders I: biochemistry and genetics of peroxisome biogenesis disorders,” Clin Genet 67:107-133., 2005.
[5] A. van der Zand, J. Gent, I. Braakman en H. Tabak, „Biochemically Distinct Vesicles from the Endoplasmic Reticulum Fuse to Form Peroxisomes,” Cell: 149(2):397-409, 2013.
[6] A. Motley en E. Hettema, „Yeast peroxisomes multiply by growth and division,” J Cell Biology178:399-410, 2007 .
[7] Ł. Opaliński, J. Kiel, C. Williams, M. Veenhuis en I. van der Klei, „Membrane curvature during peroxisome fission requires Pex11,” EMBO Journal 30(1):5-16, 2011b.
[8] R. Wells, L. Picton, S. Williams, F. Tan en R. Hill, „Direct binding of the dynamin-like GTPase, Dnm1, to mitochondrial dynamics protein Fis1 is negatively regulated by the Fis1 N-terminal arm.,” J Biol Chem. 282(46):33769-75, 2007.
[9] A. Koch, Y. Yoon, N. Bonekamp, M. McNiven en M. Schrader, „A role for Fis1 in both mitochondrial and peroxisomal fission in mammalian cells,” Mol Biol Cell.
16(11):5077-86, 2005.
[10] J. Koch en C. Brocard, „PEX11 proteins attract Mff and human Fis1 to coordinate peroxisomal fission,” J Cell Sci. 25:3813-26, 2012.
[11] A. Van der Bliek, „Functional diversity in the dynamin family.,” Trends Cell Biol
;9:96-102, 1999.
[12] H. Bui, M. Karren, D. Bhar en J. Shaw, „A novel motif in the yeast mitochondrial dynamin Dnm1 is essential for adaptor binding and membrane recruitment,” J Cell Biol. 199(4):613-22, 2012.
[13] H. Bui en J. Shaw, „Dynamin Assembly Strategies and Adaptor,” Curr Biol.
;23(19):R891-9, 2013.
[14] S. Detmer en D. Chan, „Functions and dysfunctions of mitochondrial dynamics,”
Nature Reviews Molecular Cell Biology 8, 870-879, 2007.
[15] J. Mears, L. Lackner, S. Fang, E. Ingerman, J. Nunnari en J. Hinshaw,
„Conformational changes in Dnm1 support a contractile mechanism for mitochondrial fission,” Nat Struct Mol Biol.18:20-6, 2011.
[16] H. Sesaki, S. Southard, M. Yaffe en R. Jensen, „Mgm1p, a Dynamin-related GTPase, Is Essential for Fusion of the Mitochondrial Outer Membrane,” Mol Biol
Cell.:14(6):2342-56, 2003.
[17] K. Kuravi, S. Nagotu, A. Krikken, K. Sjollema, M. Deckers, R. Erdmann, M. Veenhuis en I. van der Klei, „Dynamin-related proteins Vps1p and Dnm1p control peroxisome abundance in Saccharomyces cerevisiae,” J Cell Sci. ;119(Pt 19):3994-4001, 2006.
[18] S. Nagotu, R. Saraya, M. Otzen, M. Veenhuis en I. van der Klei, „Peroxisome proliferation in Hansenula polymorpha requires Dnm1p which mediates fission but not de novo formation,” Biochim Biophys Acta. 1783(5):760-9, 2008a.
[19] Y. Tam, J. Torres-Guzman, F. Vizeacoumar, J. Smith, M. Marelli, J. Aitchison en R.
Rachubinski, „Pex11-related proteins in peroxisome dynamics: a role for the novel peroxin Pex27p in controlling peroxisome size and number in Saccharomyces cerevisiae.,” Mol Biol Cell. 14:4089-102, 2003.
[20] H. Rottensteiner, K. Stein, E. Sonnenhol en R. Erdmann, „Conserved function of pex11p and the novel pex25p and pex27p in peroxisome biogenesis,” Mol Biol Cell.
14:4316-28, 2003.
[21] R. Saraya, A. Krikken, M. Veenhuis en I. van der Klei, „Peroxisome reintroduction in Hansenula polymorpha requires Pex25 and Rho1,” J Cell Biol. 193(5):885-900, 2011.
[22] J. Kiel, M. Veenhuis en I. van der Klei, „PEX genes in fungal genomes: common, rare or redundant,” Traffic. :7(10):1291-303., 2006.
[23] M. Schrader, B. Reuber, J. Morrell, G. Jimenez-Sanchez, C. Obie, T. Stroh, D. Valle, T. Schroer en S. Gould, „Expression of PEX11β Mediates Peroxisome Proliferation in the Absence of Extracellular Stimuli,” J Biol Chem. ;273(45):29607-14., 1998.
[24] A. Tanakaa, K. Okumotoa en Y. Fujikia, „cDNA cloning and characterization of the third isoform of human peroxin Pex11p,” Volume 300: Issue 4 Pages 819–823, 2003.
[25] I. Abe en Y. Fujiki, „cDNA cloning and characterization of a constitutively expressed isoform of the human peroxin Pex11p,” Biochem. Biophys. Res. Commun.: 252 , pp.
529–533, 1998.
[26] i. Ł. Opalińsk, M. Veenhuis en I. van der Klei, „Peroxisomes: Membrane events accompanying peroxisome proliferation,” Int J Biochem Cell Biol. ;43(6):847-51, 2011a.
[27] C. Williams en I. Van der Klei, „The functions of Pex11-family proteins in peroxisome biology”.unpublished.
[28] R. Erdmann en G. Blobel, „Giant peroxisomes in oleic acid-induced Saccharomyces cerevisiae lacking the peroxisomal membrane protein Pmp27p,” J Cell
Biol.128(4):509-23., 1995.
[29] P. Marshall, Y. Krimkevich, R. Lark, J. Dyer, M. Veenhuis en J. Goodman, „Pmp27 promotes peroxisomal proliferation,” J Cell Biol.129(2):345-55, 1995.
[30] A. Krikken, M. Veenhuis en I. van der Klei, „Hansenula polymorpha pex11 cells are affected in peroxisome retention.,” FEBS J. 276(5):1429-39, 2009.
[31] A. Mozdya, J. McCaffery en J. Shaw, „Dnm1p Gtpase-Mediated Mitochondrial Fission Is a Multi-Step Process Requiring the Novel Integral Membrane Component Fis1p,”
J. Cell Biol. 151:367–380, 2000.
[32] M. Suzuki, A. Neutzner, N. Tjandra en R. Youle, „Novel structure of the N terminus in yeast Fis1 correlates with a specialized function in mitochondrial fission,” J Biol Chem. 280(22):21444-52, 2005.
[33] S. Nagotu, A. Krikken, M. Otzen, J. Kiel, M. Veenhuis en I. van der Klei, „Peroxisome fission in Hansenula polymorpha requires Mdv1 and Fis1, two proteins also involved in mitochondrial fission,” Traffic. 9:1471-84, 2008b.
[34] J. Shaw en N. J, „Mitochondrial dynamics and division in budding yeast,” Volume 12, Issue 4 Pages 178–184, 2012.
[35] Q. Tieu en J. Nunnari, „Mdv1p Is a Wd Repeat Protein That Interacts with the Dynamin-Related Gtpase, Dnm1p, to Trigger Mitochondrial Division,” J Cell Biol.
;151(2):353-66., 2000.
[36] J. Tooley, V. Khangulov, J. Lees, J. Schlessman, M. Bewley, A. Heroux, J. Bosch en
H. RB, „The 1.75 Å resolution structure of fission protein Fis1 from Saccharomyces cerevisiaereveals elusive interactions of the autoinhibitory domain,” Acta Cryst. : F67, 1310-1315, 2011.
[37] M. Serasinghe, S. A. A. Smrcka en Y. Yoon, „Identification and Characterization of Unique Proline-rich Peptides Binding to the Mitochondrial Fission Protein hFis1,” J Biol Chem ;285(1):620-30, 2010.
[38] H. de Vries, „Novel insights into the relation between Dnm1p, Pex11p and Fis1p,”
Research report from the University of Groningen, 2013.
[39] L. Van Dijken, R. Otto en W. Harder, „Growth of Hansenula polymorpha in a methanol-limited chemostat.,” Archives of Microbiology 111:137-144, 1976.
[40] S. Manivannan, K. Knoops, A. Krikken, A. Kram, M. Veenhuis en I. Van der Klei,
„Yeast peroxisomes: De novo formation and maintenance. Pre-peroxisome formation in Hansenula Polymorpha pex19 cells does not require the ER,”
Manuscript, 2014.
[41] M. Cepińska, M. Veenhuis, I. Van der Klei en S. Nagotu, „Peroxisome Fission is Associated with Reorganization of Specific Membrane Proteins,” Traffic. ;12(7):925- 37, 2011.
[42] R. Baerends, F. Salomons, K. Faber, J. Kiel, I. Van der Klei en M. Veenhuis, „Deviant Pex3p levels affect normal peroxisome formation in Hansenula polymorpha: high steady-state levels of the protein fully abolish matrix protein import.,” Yeast 13(15):1437-48, 1997.
[43] M. McCammon, J. McNew, P. Willy en J. Goodman, „An internal region of the
peroxisomal membrane protein PMP47 is essential for sorting to peroxisomes,” J Cell Biol. 124:915-25, 1994.
[44] R. Baerends, K. Faber, A. Kram, J. Kiel, I. van der Klei en M. Veenhuis, „A stretch of positively charged amino acids at the N terminus of Hansenula polymorpha Pex3p is involved in incorporation of the protein into the peroxisomal membrane,” J Biol Chem. 275:9986-95, 2000.
[45] A. Douma, M. Veenhuis, W. de Koning, M. Evers en W. Harder, „Dihydroxyacetone synthase is localized in the peroxisomal matrix of methanol-grown Hansenula polymorpha,” Arch. Microbiol. 143: 237-243, 1985.
[46] M. Stewart, R. Esposito, J. Gowani en J. Goodman, „Alcohol oxidase and
dihydroxyacetone synthase, the abundant peroxisomal proteins of methylotrophic yeasts, assemble in different cellular compartments,” J Cell Sci. 114:2863-8, 2001.
[47] X. Li, E. Baumgart, G. Dong, J. Morrell, G. Jimenez-Sanchez, D. Valle, K. Smith en S. Gould, „PEX11alpha is required for peroxisome proliferation in response to 4- phenylbutyrate but is dispensable for peroxisome proliferator-activated receptor alpha-mediated peroxisome proliferation,” Mol Cell Biol. 22(23):8226-40., 2002 . [48] C. Williams, L. Opalinski, C. Landgraf, J. Costello, M. Schrader, A. Krikken, K.
Knoops, A. Kram, R. Volkmer en I. Van der Klei, „Pex11p functions as a GTPase activating protein for Dnm1p during peroxisomal fission,” submitted, 2014.
[49] Y. Zhang en D. Chan, „Structural basis for recruitment of mitochondrial fission complexes by Fis1,” Proc Natl Acad Sci U S A. ;104(47):18526-30, 2007.
