In order to determine whether there is a discrete region of Myo1 necessary and sufficient for localisation, deletion constructs were created and transformed into wild type and myoIΔ strains. The number of rings at the bud neck as a percentage of the total number of bud necks was calculated for each construct in both wild-type and Δmyo1 strains and the results are shown in Fig 2, with fluorescent microscopy images of GFP labelled Myo1 and MLD shown below the localisation percentages. These results show that the minimum localisation domain (MLD) consists of residues 1044–1569. The coiled coil prediction plot shows that the MLD (indicated in Fig. 1) encompasses the region of low coiled coil forming probability, which includes 7 proline residues. Deletion of any additional parts of the MLD led to a lower percentage of localisation as compared to the full length MLD (Fig 2).

Since dimerisation and subsequently filament formation in many class II myosins is required for localisation we wanted to determine if the MLD forms oligomers. To this end, GST-MLD was purified after expression in E. coli. After removal of the GST tag, the MLD was applied to a Superose 6 gel filtration column. The MLD eluted at 49.4 +/- 0.4 minutes (n = 7) and the Stokes radius was obtained from a calibration curve of standards with known Stokes radii as shown in Fig. 3a; this resulted in a Stokes radius of 8.4 +/- 0.7 nm for the MLD (Fig. 3a and 3c). The void volume for the column was obtained using Dextran blue (2000 kDa, Stokes radius unknown), which gave an elution time of 32.2 minutes. Thyroglobulin, which has a molecular weight of 669 kDa and an elution time of 49.0 minutes, ran close to the MLD, showing that the calibration curve was well within the upper resolution limit of the column. Purified MLD was also applied to a preformed 6–20 % sucrose density gradient and was found in fraction 7.6 +/- 0.5 (n = 4). A sedimentation coefficient obtained from a calibration curve of known sedimentation coefficients, as shown in Fig. 3b, was 5.2 +/- 0.3 S (n = 4). Neither gel filtration nor sucrose density gradient centrifugation can be used in isolation to obtain the molecular weight of the MLD as the presumed elongated nature of the protein will result in a larger apparent Stokes radius and a smaller apparent sedimentation coefficient as compared with a globular protein. This makes it difficult to determine the true molecular weight and thus the oligomeric nature of the MLD. However, combining the Stokes radius and the sedimentation coefficient in Equation 1 (see Materials and Methods) allowed us to determine an experimental molecular weight of 177 kDa for the MLD. When this value is divided by the calculated molecular weight for a monomer (62 kDa), the oligomer is tentatively assigned to be a trimer (2.9 monomers).

We were concerned that the experimental values obtained were a product of expressing and purifying the MLD from E. coli. In order to be sure that our expressed protein was behaving in the same way as it would in yeast, the MLD with a C-terminal (myc)6 tag was transformed into yeast (RLY 1232). Yeast extracts were prepared by grinding frozen cells under liquid nitrogen using a pestle and mortar; the powder was resuspended in a small volume of buffer and centrifuged to remove cellular debris and unlysed cells. 250 μl of the supernatant was applied to the Superose 6 gel filtration column. Fractions were collected and western blots were carried out after trichloroacetic acid (TCA) precipitation to determine the elution position of MLD-(myc)6. As shown in Fig. 3c and 3d, MLD-(myc)6 eluted in the same fractions (48.2 +/- 0.4 minutes, n = 4) as the expressed bacterial protein (49.4 +/- 0.4 minutes) indicating that they are hydrodynamically similar and thus that the trimer value assigned to MLD was not an artefact of expression in E. coli.

We observed that deletion of the MLD from Myo1 resulted in complete loss of localisation (Fig. 2) but that deletion of the hinge region (residues 1162–1430) within the MLD only lowered the efficiency of localisation to the bud neck to 70 – 80 % of the full length molecule (Fig. 2) showing that the hinge region contributes to localisation but its loss does not abolish localisation. Since several other class II myosins contain kinks (where the coiled coil heptad repeat is interrupted) in their tail that are important for function, we wanted to examine the effect of removing the hinge region from Myo1 on CR contraction. A plasmid encoding Myo1Δhinge [Myo1(1–1161:1431–1928)-GFP-(myc)6] under the control of the MYO1 promoter was transformed into the heterozygous diploid myo1Δ strain (RLY621), and haploid strains straining expressing Myo1Δhinge as the sole Myo1 were obtained by sporulation and tetrad dissection.

We monitored contraction of the CR in wild-type and Myo1Δhinge strains. Cells were grown at the room temperature and arrested for 3 hrs with nocodazole. The cells were then released from arrest and contraction was monitored using time-lapse confocal microscopy with images taken every 30 s. Fig. 4 shows montages (A) and kymographs (B) of the wild-type and Myo1Δhinge ring as it goes through cytokinesis. Interestingly, Myo1Δhinge rings were unable to undergo contraction: the rings remained steadily at the bud neck until late in the cell cycle before gradually disappearing without reduction in diameter (n = 15). This was in contrast to the CR in wild-type cells, which contracted to a point before disappearing (n = 15).

Fluorescence recovery after photobleaching (FRAP) enables one to examine the dynamics of a fluorescently labelled molecular structure: fast recovery of the fluorescent signal indicates that there is a free pool of molecules that interchange rapidly with the molecules in the bleached area. We used FRAP to compare the in vivo dynamics of GFP-tagged full length Myo1, MLD and Myo1Δhinge [strains RLY1044, RLY1046 and RLY2139 (Table 2)]. Images were collected at 5 or 10 s intervals and analysed for fluorescence recovery. Table 1 shows the average percentage recovery and half time values. Myo1Δhinge appeared to be the most dynamic, recovering the greatest proportion of the bleached fluorescence and having the smallest recovery half time. This suggests that the contraction defect of the Myo1Δhinge could be due to a structural abnormality causing the mutant Myo1Δhinge molecules to be less well anchored within the CR. The MLD on the other hand appears to have a dynamic behaviour in between that of the full length molecule and Myo1Δhinge (Table 1) suggesting that the MLD is sufficient for stable association with the CR, but that other regions of Myo1 further stabilise the association.

Yeast cell culture and genetic techniques were carried out by methods described in 28. Yeast extract, peptone, dextrose (YPD) contained 2% glucose, 1% yeast extract and 2% Bactopeptone (Difco Laboratories, Detroit, MI). Synthetic complete (SC) media was prepared by the method described 29.

The plasmid expressing Myo1-GFP-(myc)6 (pNT50) was generated by excising MYO1 from a previously-described plasmid encoding Myo1-GFP (pLP8 30) with PstI and BamHI, and ligating it into PstI-BamHI-digested pRL304, a pRS315-based vector for C-terminal tagging with a GFP-(myc)6 fusion. To generate Myo1-GFP-(myc)6 in a pRS316 (URA3)-based vector, the 7 kb fragment from SalI-NotI digested pNT50 was ligated to SalI-NotI digested pRS316, creating pNT113. To generate the plasmid that expresses the C-terminal 884 amino acids of Myo1p under the control of the MYO1 promoter with a C-terminal GFP tag (pNT13), the sequence encoding the promoter region and the first 1043 amino acids was excised from pLP8, using PstI and BglII, and replaced with just the promoter region (generated by PCR and digested with PstI and BglII). The GFP tag of pNT13 was replaced with a GFP-(myc)6 fusion by ligating the 1 kb fragment generated by BamHI-NotI digest of pNT50 to BamHI-NotI digested pNT13, creating pNT76. To create the plasmid expressing Myo1(1044–1569), pNT13 was double digested with BglII and HaeIII yielding a 1.6 kb fragment encoding Myo1(1044–1569). pNT13 was digested with BamHI, blunted and digested with BglII, to yield a 7 kb fragment encoding the vector backbone, MYO1 promoter, and GFP. Ligation of both fragments produced an in-frame fusion of Myo1(1044–1569) and GFP (pNT43), which was then excised using PstI and BamHI, and ligated into PstI-BamHI-digested pRL304 to create Myo1(1044–1569)-GFP-(myc)6 (pNT49). The sequence encoding Myo1(1529–1928) was generated by PCR from yeast genomic DNA and digested with BglII and BamHI (restriction sites contained within the forward and reverse primers, respectively), and ligated to the 6.9 kb fragment produced by BglII-BamHI double digest of pNT13, to generate Myo1(1529–1928)-GFP (pNT40). To create the plasmid expressing Myo1(1–1043), pNT50 was double-digested with BglII and BamHI, blunted and self-ligated yielding an in-frame fusion of Myo1(1–1043) and GFP-(myc)6 (pNT51). The sequence encoding Myo1(1044–1253) was generated by digesting pNT43 with XbaI, blunting, and digesting with PstI, yielding a 0.9 kb fragment. This fragment was ligated to pNT49 that had been digested with BamHI, blunted, and digested with PstI, to create Myo1(1044–1253)-GFP-(myc)6 (pNT62). pNT43 was digested with XbaI, blunted and digested with BamHI to produce a 0.9 kb fragment encoding Myo1(1261–1569). This was ligated to pNT49 that had been digested with BglII, blunted, and digested with BamHI, to create Myo1(1261–1569)-GFP-(myc)6 (pNT63). A construct expressing Myo1 lacking the C-terminal 26 amino acids was generated by utilizing an existing EcoRI site, as described previously 27. Briefly, MYO1, cloned PstI-BamHI in the Bluescript vector (pLP9), was digested with EcoRI to yield a 6 kb fragment encoding Myo1(1–1902). This fragment was ligated to EcoRI-digested pRL164, a pRS313-based vector for C-terminal GFP tagging, to generate pRL221. This plasmid was double-digested with BglII and BamHI to generate a fragment encoding Myo1(1044–1902), and this was ligated to BglII-BamHI digested pNT50 to create Myo1(1–1902)-GFP-(myc)6 (pNT66). The sequence encoding Myo1(1044–1161) was generated by PCR from yeast genomic DNA and digested with BglII and BamHI (restriction sites contained within the forward and reverse primers, respectively) and ligated to the 6.9 kb fragment produced by BglII-BamHI double digest of pNT13 to generate Myo1(1044–1163)-GFP (pNT69). The GFP tag of pNT69 was replaced with a GFP-(myc)6 fusion by ligating the 1 kb fragment generated by BamHI-NotI digest of pRL304 to BamHI-NotI digested pNT69, creating pNT71. The sequences encoding Myo1(1160–1928), Myo1(1044–1430), Myo1(1044–1387) and Myo1(1044–1314) were generated by PCR from yeast genomic DNA, digested with BglII and BamHI (restriction sites contained within the forward and reverse primers, respectively) and ligated to the 7.3 kb fragment produced by BglII-BamHI double digest of pNT76, to generate pNT94, pNT110, pNT111 and pNT112, respectively. Similarly, the sequence encoding Myo1(864–1161) was generated by PCR from yeast genomic DNA, digested with BamHI (restriction site contained within the forward and reverse primers) and ligated to the 7.3 kb fragment produced by BglII-BamHI double digest of pNT76, to generate Myo1(864–1161)-GFP-(myc)6 (pNT96). To create a deletion of the non-helical sequences within Myo1-tail, pNT121 was constructed. Briefly, the sequence encoding Myo1(1431–1928) was generated by PCR from yeast genomic DNA, digested with BamHI (restriction site contained within the forward and reverse primers) and ligated to BamHI digested pNT71, to generate Myo1(1044–1160:1431–1928)-GFP-(myc)6. Similarly, the sequence encoding Myo1(1570–1928) was generated by PCR from yeast genomic DNA, digested with BglII and BamHI (restriction sites contained within the forward and reverse primers, respectively) and ligated to BglII-BamHI digested pNT50, to generate Myo1(1–1044:1670–1928)-GFP-(myc)6 (pNT120). To create a deletion of the non-helical region of the MLD in the full length Myo1, pIL11 was constructed. Briefly pNT121 was cut with PstI/BglII to yield a ~7000 bp fragment encoding Myo1(1044–1161:1439–1928). This fragment was ligated with a ~3300 bp fragment encoding Myo1 promoter -Myo1(1–1043) cut with PstI/BglII from pLP8. To create a plasmid (pIL10) encoding N-terminally tagged GST-MLD, a PCR product encoding Myo1(1044–1559) was cut with BamHI/EcoRI was ligated into pGEX6P1 that contains a precision protease cleavage site between the GST and MLD reading frames. To create an MLD construct C-terminally tagged with (myc)6 for expression in yeast, a fragment consisting of Myo1 promoter-Myo1(1044–1569) was cut from pNT49 using PstI and BamHI and ligated into pRL72 cut PstI/BamHI resulting in pIL13.

Exponentially growing cultures of cells expressing either Myo1-GFP or a deletion construct were assessed for bud neck localization using an Eclipse E800 microscope with a 100/1.40 oil differential-interference contrast objective (Nikon, Melville, NY). At least 100–200 live cells were analyzed, and data shown are averages from at least 3 independent experiments. Images were collected with a 0.5 s exposure to fluorescent light filtered through an EXHQ450/50 DM480 LP/BA465LP GFP filter set (Chroma, Brattleboro, VT) with a cooled RTE/CCD 782Y Interline camera (Princeton Instruments) using MetaMorph (Universal Imaging Corp., West Chester, PA). RLY2127 and 2128 fluorescence was weak and could not be collected using the method described above. Instead 20 z sections at 0.5 μm intervals were collected and deconvolved as described below for the CR movies. RLY1044 and RLY1046 were also measured in this way to ensure that counts were comparable with those obtained by the first method (Results for RLY1044: 79 %. Results for RLY1046: 75 %).

FRAP experiments were carried out using a 435 nm nitrogen pulsed Micropoint laser (Photonic Instruments Inc., St. Charles IL). Galvanometer based steering optics controlled with MetaMorph software were used to position the laser beam, which was focused into a diffraction-limited 0.7 μm spot using a Nikon 100× 1.4 N.A. objective lens. Images were acquired every 5 or 10 s with a Yokogawa spinning disc confocal microscope mounted on a Nikon TE2000U stand and a Hamamatsu ORCA ER cooled CCD. All measurements were done using MetaMorph and all data analysis was done using Excel (Microsoft). To analyse the data, the signal from the FRAPed bud neck and a control bud neck (to check for bleaching) were collected. Regression analysis to determine the half time to maximum was done using a one-phase exponential association function (Y = bottom + (top-bottom)·(1-exp [-k·x]), where k is the rate constant and t1/2 is 0.69/k in Prism (version 4.00; GraphPad).

For assaying CR contraction, cells were arrested in metaphase with nocodazole for 3 hrs at room temperature and then released from the arrest. The released cells were mounted on gelatine pads. Three dimensional (3D) images and movies of RLY1331 and RLY2139 were collected with the spinning disc confocal microscope described above, imaging 12 sections in the z-plane with a spacing of 0.5 μm, with exposures of 1 s. Each stack of z-series images was deconvolved and rendered in to 3D images and movies using MetaMorph (Universal Imaging Corp., West Chester, PA).

Myo1 minimum localisation domain (MLD) tagged N-terminally with GST (plasmid pIL10) was transformed into BL21 cells and grown in 1.5 L cultures at 37°C until an OD₆₀₀ of 1–1.2. The temperature was reduced to 24°C and IPTG was added to a final concentration of 50 μM before growing the cultures for a further 4 hrs. The cells were harvested and stored as pellets at -20°C. Pellets were lysed in 25 % sucrose, 50 mM Tris HCl, pH 7.5, 500 mM NaCl, 1 Roche inhibitor tablet per 50 ml solution, 2 mM DTT. The lysate was centrifuged at 35,000 rpm for 30 minutes and the supernatant added to 800 μl Glutathione Sepharose 4B (Amersham Biosciences) for 3 hrs at 4°C with constant agitation. The resin was washed with 50 ml buffer A+ Triton (500 mM NaCl, 50 mM Tris HCl pH 7.5, 1 mM DTT, 1 % Triton X-100) followed by 40 ml buffer A (500 mM NaCl, 50 mM Tris HCl pH 7.5, 1 mM DTT). The resin was resuspended in 300–500 μl buffer A and 30 μl of GST tagged precision protease at 0.6 mg/ml was added. Digestion was carried out overnight at 4°C and the supernatant was removed after low speed centrifugation to pellet the resin.

A Superose 6 column (BioRad) was equilibrated in buffer A. 250 μl of Glutathione sepharose purified MLD with the GST tag removed was added and 0.5 ml fractions were collected. Alternatively, 250 μl RLY2132 yeast grinding extract (see below) was run over the column. For SDS-PAGE analysis, 10 % TCA was added to each fraction, incubated on ice for 30 minutes and centrifuged for 15 minutes. The pellets were resuspended in 30 μl SDS-loading buffer and 5 μl 1 M non-pH'd Tris base. MLD-(myc)6 from yeast extract fractions run on the column was detected by Western blotting using the anti-myc monoclonal antibodies, 9E-10. The column was calibrated with globular proteins of known Stokes radii: Thyroglobulin – 669 kDa, 8.5 nm, Ferritin – 440 kDa, 6.1 nm, Catalase – 232 kDa, 5.2 nm, Aldolase 158 kDa, 4.8 nm. Dextran Blue with a molecular weight of approximately 2000 kDa was used to establish the void volume of the column.

5 ml 6–20 % sucrose gradients were prepared in ultra clear centrifuge tubes (Beckman) for the SWTi55 rotor (Beckman). The gradient was calibrated using protein standards of known sedimentation coefficients: catalase – 232 kDa, 11.2 S, aldolase – 158 kDa, 7.3 S, BSA – 66 kDa, 4.6 S, ovalbumin – 43 kDa, 3.6 S. The top 400 μl of the gradient were removed and replaced with 300 μl Glutathione sepharose 4B purified MLD, 5 μg of aldolase as an internal control, sucrose to a final concentration of 6 % and water to 400 μl. The gradient was centrifuged for 19 hrs at 40 k rpm at 4°C. At the end of the run, the gradient was fractionated from the top into 250 μl fractions. These were TCA precipitated as described above.

The native molecular weight of the expressed MLD was calculated using its Stokes radius measured by gel filtration and its sedimentation coefficient determined by sucrose density gradient centrifugation using the following equation 1 described in 31.

M .W = S ω 20 N .6 π . n .R s 1 − ν ¯ ρ MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciaacaGaaeqabaqabeGadaaakeaacqqGnbqtcqqGUaGlcqqGxbWvcqGH9aqpdaWcaaqaaiabdofatHGaciab=L8a3naaBaaaleaacqaIYaGmcqaIWaamaeqaaGqacOGae4Nta4KaeeOla4IaeGOnayJae8hWdaNaeiOla4IaeeOBa4MaeeOla4IaeeOuai1aaSbaaSqaaiabbohaZbqabaaakeaacqaIXaqmcqGHsislcuWF9oGBgaqeaiab=f8aYbaaaaa@462F@

where Sω₂₀ is the sedimentation coefficient, R_s is the Stokes radius, Avogadro's number, N is 6.02 × 10²³; viscosity coefficient, n is 1×10^-2 g s^-1 cm^-1; solution density, ρ is 1 g cm^-3 and partial specific volume, ν¯ MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciaacaGaaeqabaqabeGadaaakeaaiiGacuWF9oGBgaqeaaaa@2E83@ is 0.71 cm³ g^-1 (assumed to be that of an average soluble protein).

To prepare the yeast extracts used for gel filtration analysis, small pellets were extruded through a syringe into N₂ (l) creating cylindrical fragments that could then be ground by pestle and mortar in the presence of N₂ (l). The powdered yeast was thawed and resuspended in buffer A. Protease inhibitors (2 μg/ml Antipain, 2 μg/ml Leupeptin, 2 μg/ml Aprotinin, 2 μg/ml Pepstatin, 2 μg/ml Chymostatin) and 1 mM PMSF were added to both buffers. Lysis was checked using light microscopy. Thawed extracts were centrifuged at 50 K rpm in a Sorvall RP120AT rotor for 30 minutes at 4°C.