UEF//eRepository
DSpace https://erepo.uef.fi
Rinnakkaistallenteet Terveystieteiden tiedekunta
2015-10-01
Cell protrusions induced by hyaluronan synthase 3 (HAS3) resemble
mesothelial microvilli and share cytoskeletal features of filopodia
Koistinen, Ville
Elsevier BV
info:eu-repo/semantics/article
© Elsevier Inc
CC BY-NC-ND 4.0 http://creativecommons.org/licenses/by-nc-nd/4.0/
http://dx.doi.org/10.1016/j.yexcr.2015.06.016
https://erepo.uef.fi/handle/123456789/59
Downloaded from University of Eastern Finland's eRepository
Elsevier Editorial System(tm) for Experimental Cell Research Manuscript Draft
Manuscript Number:
Title: Cell protrusions induced by hyaluronan synthase 3 (HAS3) resemble mesothelial microvilli and share cytoskeletal features of filopodia
Article Type: SI: Cell Biol-High Resolution Keywords: Filopodia
Microvilli Hyaluronan Actin cytoskeleton Hyaluronan synthesis
Corresponding Author: Dr. Kirsi Rilla,
Corresponding Author's Institution: University of Eastern Finland First Author: Ville Koistinen
Order of Authors: Ville Koistinen; Riikka Kärnä; Arto Koistinen; Antti Arjonen; Markku Tammi; Kirsi Rilla
Editorial Office
Experimental Cell Research Editor-in-Chief
Stockholm Sweden
May 8, 2015 Dear Dr. Lendahl
Thank you for the invitation to submit a paper in the Special Issue Cell Biology at High Resolution of Experimental Cell Research. Attached, is our manuscript to be considered for publication in this issue entitled " Cell protrusions induced by hyaluronan synthase 3 (HAS3) resemble mesothelial microvilli and share cytoskeletal features of filopodia" by Ville Koistinen, Riikka Kärnä, Arto Koistinen, Antti Arjonen, Markku Tammi and Kirsi Rilla. The type of manuscript is regular article.
Our previous studies have shown that overexpression of enzymatically active GFP- HAS induces the growth of long, slender protrusions that share many features of both filopodia and microvilli (Rilla et al. 2005 JBC, Kultti et al. 2006, JBC). In this work the structural details, dynamics and regulation of these protrusions were studied by live cell confocal microscopy and electron microscopy. It was shown that hyaluronan synthesis is an important driving force for protrusion growth and maintenance.This study shows that the HAS3-induced protrusions share most cytoskeletal features with filopodia, but they do not require adherence to the substratum like traditional filopodia. The results also suggest that hyaluronan chain maintains HAS3 in plasma membrane.
All of the authors have directly participated in the planning, execution, or analysis of the study and resulting paper, and have read and approved the version submitted. We hope that this manuscript is acceptable for publication in The Experimental Cell Research.
Yours sincerely,
Kirsi Rilla, PhD
Institute of Biomedicine, Anatomy University of Eastern Finland P.O.B. 1627
70211 Kuopio Finland
Phone: +358 40 3553218 Fax +358 17 163032 Email: kirsi.rilla@uef.fi
*Cover Letter
Bullet points
Hyaluronan-dependent protrusions share cytoskeletal features of filopodia
Myo10 is localized to the tips of those protrusions
Hyaluronan retains active HAS on the plasma membrane
Actin-based long dorsal filopodia are extracellularly scaffolded by hyaluronan coat
Hyaluronan synthesis on the plasma membrane protrusions is a general cellular process
Highlights
1
8.5.2015 1
2
Cell protrusions induced by hyaluronan synthase 3 (HAS3) resemble mesothelial
3
microvilli and share cytoskeletal features of filopodia
4 5
Ville Koistinen1, Riikka Kärnä1, Arto Koistinen2, Antti Arjonen3, Markku Tammi1 and Kirsi Rilla1 6
7
1Institute of Biomedicine and 2SIB Labs, University of Eastern Finland, Kuopio, Finland and 8
3Turku Centre for Biotechnology, University of Turku, Turku, Finland.
9 10
Address correspondence to: Kirsi Rilla, Institute of Biomedicine, University of Eastern Finland, 11
P.O.B. 1627, FIN-70211 Kuopio, Finland, E-Mail: kirsi.rilla@uef.fi 12
13
ABBREVIATIONS 14
15
4-MU, 4-methylumbelliferone; CD44, hyaluronan receptor, cluster of differentiation 44; ECM, 16
extracellular matrix; ERM, ezrin-radixin-moiesin; GFP, green fluorescent protein; HAS, hyaluronan 17
synthase; Myo10, Myosin 10; SEM, scanning electron microscopy.
18 19
ABSTRACT 20
21
Previous studies have shown that overexpression of enzymatically active GFP-HAS induces the 22
growth of long, slender protrusions that share many features of both filopodia and microvilli. These 23
protrusions are dependent on continuing hyaluronan synthesis, and disrupt upon digestion of 24
hyaluronan by hyaluronidase. However, complete understanding of their nature is still missing.
25
This work shows that the protrusions on rat peritoneal surface are ultrastructurally 26
indistinguishable from those induced by GFP-HAS3 in MCF-7 cells. Analysis of the actin- 27
associated proteins villin, ezrin, espin, fascin, and Myo10 indicated that the HAS3-induced 28
protrusions share most cytoskeletal features with filopodia, but they do not require adherence to the 29
substratum like traditional filopodia.
30
GFP-HAS3 overexpression was found to markedly enhance filamentous actin in the protrusions 31
and their cortical basis. Analysis of the protrusion dynamics after enzymatic digestion of 32
hyaluronan revealed that while GFP-HAS3 escape from the protrusions and the protrusion collapse 33
takes place immediately, the complete retraction of the protrusions occurs more slowly. This 34
finding also suggests that hyaluronan chain maintains HAS3 in the plasma membrane.
35
*Manuscript
Click here to view linked References
2
The results of this work suggest that protrusions similar to those of HAS3 overexpressing cells in 1
vitro exist also in cells with active hyaluronan synthesis in vivo. These protrusions are similar to 2
common filopodia but are independent of substratum attachment due to the extracellular scaffolding 3
by the hyaluronan coat that accounts for the growth and maintenance of these structures, previously 4
associated to invasion, adhesion and multidrug resistance.
5 6
KEY WORDS 7
Microvilli, filopodia, hyaluronan, actin, hyaluronan synthesis 8
9
INTRODUCTION 10
Many vertebrate cells display slender plasma membrane protrusions like filopodia and 11
microvilli, which are thin (less than 200 nm in diameter), finger-like protrusions, supported by 12
parallel bundles of actin filaments. Cells utilize these protrusions for migration, attachment, 13
secretion, absorption [1] and sensing their environment [2] . 14
Formation and maintenance of protrusions depend on a bundle of actin filaments in the core of 15
these extensions [1] . To make a filopodium or microvillus, 10-40 filaments are usually nucleated in 16
the cortical actin network, and grown as a bundle towards the membrane [3] . Continuous bundling of 17
the actin filaments is crucial for growth of the filopodia and microvilli [4] , and involves proteins 18
such as villin [5] , fascin [6] and espin [7] . Binding of the actin filament bundle to the plasma 19
membrane is also required for the growth and maintenance of these membrane extensions. The ERM 20
family proteins, ezrin in particular [8] excecute this task. Activation of membrane associated 21
signaling by small GTPases of the Cdc42 family trigger filopodia formation, together with Src family 22
tyrosine kinases and p21-activated kinase (PAK) [9] . Myosin-X (Myo10) is an unconventional actin 23
motor protein regulating filopodial stability and localizing to the tips of the filopodia [10, 11] . 24
Most of the components in the actin polymerization apparatus listed above, when overexpressed 25
or experimentally activated, have been reported to increase the growth of filopodia or microvilli.
26
This suggests that in many cells there is a strong intrinsic potential to display these extensions, but 27
at the molecular level, mechanisms that induce the growth of filopodia and microvilli in vivo, are 28
still somewhat obscure.
29
Unlike microvilli, filopodia are typically adhered to a substratum or to another cell [1] . 30
Microvilli arise from luminal surfaces of intestinal and kidney epithelial cells and the apical 31
surfaces of many cultured cells [1] . The function of the typically short and regularly arranged 32
epithelial microvilli is to increase absorptive surface area, while mesothelial protrusions, often also 33
called microvilli, actively secrete extracellular matrix molecules [12] . The nomenclature of actin- 34
3
based protrusions is variable and often inconsistent, especially in cultured cells.
1
Hyaluronan is an abundant ECM glycosaminoglycan that is synthesized on the plasma 2
membrane by three isoenzymes of hyaluronan synthases (HAS1-3) [13] . Ongoing hyaluronan 3
synthesis is associated with formation of several different actin-based plasma membrane 4
protrusions, like filopodia [14, 15] , lamellipodia [16] , invadopodia [17] , retraction fibers and 5
surface blebs [18] and membrane ruffles [19] . Plasma membrane protrusions within a thick, 6
hyaluronan-containing glycocalyx are found on many normal cell types with high hyaluronan 7
secretion rate like oocytes [20] , synovial cells [21] , mesothelial cells [22] and chondrocytes [23] . 8
In fact, protrusions of cell types that secrete high amounts of hyaluronan, are covered by a regular 9
layer of hyaluronan, and act as a natural scaffold for traditional “hyaluronan coat” [24] . 10
We have previously shown that transfected, GFP-labeled hyaluronan synthases (GFP-HAS2 and 11
3) accumulate on the plasma membrane extensions of epidermal keratinocytes [25] . Furthermore, 12
the enzymatically active GFP-HAS not only seeks its way into cell surface protrusions, but actually 13
induces the formation of up to 25 µm long hyaluronan coated protrusions covering the apical 14
surface of all cell lines studied so far, including keratinocytes, MDCK cells, MCF-7 breast cancer 15
cells, and SKOV-3 ovarian cancer cells [26] . Additionally, protrusions are dependent on 16
continuous hyaluronan synthase activity [26] . However, their structure and function have been 17
partly unknown so far.
18
In this work the structural details, dynamics and regulation of these protrusions were studied by 19
live cell confocal microscopy and electron microscopy. It was shown that hyaluronan synthesis is 20
an important driving force for protrusion growth and maintenance. We suggest that hyaluronan 21
synthesis on plasma membrane protrusions is a general cellular process, involved in the 22
maintenance and function of normal tissues as well as in pathological states.
23 24
MATERIALS AND METHODS 25
26
Cell culture and treatments of the cultures 27
The human breast adenocarcinoma cell line MCF-7 was cultured in minimum essential medium 28
alpha (MEM, EuroClone, Pavia, Italy) supplemented with 5 % inactivated fetal bovine serum (FBS, 29
PAA Laboratories GMbH, Pasching, Austria), 2 mM glutamine (EuroClone) and 50 µg/ml 30
streptomycin sulfate and 50 U/ml penicillin (EuroClone). Myo10- silenced MDA-MB-231 human 31
breast adenocarcinoma cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM, 32
4500 mg/l glucose, EuroClone) supplemented with 1% nonessential amino acids (Sigma Aldrich, St.
33
Louis, MO), 2 mM glutamine, 10% fetal bovine serum and 0,33µg/ml puromycin.
34
4
Human mesothelial cells (LP-9) were cultured in MCDB 110 medium (Sigma) and medium 199 1
(Sigma) in a ratio of 1:1, supplemented with 15 % fetal bovine serum (PAA Laboratories), 2 mM 2
glutamine (EuroClone) and 50 g/ml streptomycin sulphate, 50 U/ml penicillin (EuroClone), 10 3
ng/ml epidermal growth factor (EGF, Sigma) and 0.05 g/ml hydrocortisone (Sigma). All cells were 4
passaged twice a week at a 1:5 split ratio using 0.05% trypsin (w/v) 0.02% EDTA (w/v) (Biochrom, 5
Berlin). Concentration of 4-methylumbelliferone was 1 mM (Sigma) and Streptomyces hyaluronidase 6
(Seikagaku, Kogyo, Tokyo, Japan) 1-10 TRU/ml.
7 8
Transfections 9
Subconfluent cell cultures grown on 8-well Ibidi chamber slides (Ibidi GmbH, Martinsried, 10
Germany) were transiently transfected using ExGen 500 transfection reagent (Fermentas, Helsinki, 11
Finland). Constructs used in the Has3 transfections were human Has3 cDNA in-frame with an N- 12
terminal GFP fusion protein in the pCIneo vector [27] and HAS alone in the pCIneo vector [28] . 13
The cells were examined 16-24 h after transfections with confocal imaging, either live for the GFP 14
signal only, or after fixation with 4 % paraformaldehyde for 1 h at room temperature for both 15
immunostainings and GFP. To visualize actin filaments, GFP-actin (Molecular Probes, Eugene, OR, 16
USA) and Lifeact (pLifeact-TagRFP, Ibidi GmbH, Martinsried, Germany) were utilized. Constructs 17
for WT mCherry-Myo10 and Pleckstrin homology (PH) domain deleted Myo10 (Myo10 WT, 18
Myo10∆PH123, Myo10∆PH2 and Myo10∆PH2pm) have been previously described [29] . The 19
stable, Myo10-silenced MDA-MB-231 cell line was prepared as previously described [30] . 20
21
Assay of hyaluronan 22
Subconfluent cell cultures on 24-well plates were used to measure the hyaluronan secretion. After 23
treatment or transient transfections, a fresh medium was changed (with 5% FBS), and cells were 24
cultured for 24 h before counting the cells and harvesting the media for the [31] sandwich type 25
hyaluronan ELSA assay as described previously [31] . Briefly, Maxisorp 96-well plates (Nunc, 26
Roskilde, Denmark) were coated with HABC (hyaluronan binding region of aggrecan and link 27
protein) to catch hyaluronan from the samples and standards (range 0 - 35 ng/ml). After washes, the 28
bound hyaluronan was detected using bHABC, horseradish peroxidase streptavidin (Vector 29
Laboratories, Inc., Burlingame, CA), and TMB substrate solution (0.5% 3,3’,5,5’- 30
tetramethylbenzidine in dimethyl sulfoxide from Sigma, diluted 1:50 with 0.1 M sodium acetate, 1.5 31
mM citric acid and 0.005% H2O2) for spectrophotometric quantitation at 450 nm. For staining of 32
pericellular hyaluronan coat in live cells, a fluorescently labeled hyaluronan binding complex 33
(fHABC) was used like described previously [24] . 34
5 1
Immunofluorescence stainings 2
For immunofluorescence stainings, cells were fixed with 4% paraformaldehyde in phosphate 3
buffer, pH 7.4 (PB) for 1 h, washed with PB, permeabilized for 20 min at room temperature with 4
0.1% Triton-X-100 in 1% BSA, and incubated overnight at 4°C with anti-CD44 monoclonal antibody 5
(Hermes 3, 1:100, a generous gift from Dr. Sirpa Jalkanen, Turku, Finland), villin (1:100, BD 6
Biosciences, Bedford, MA), ezrin (1:200, Thermo Scientific, Rockford, IL, USA), espin (1:10, a kind 7
gift from Dr. Jim Bartles, Chicago, IL), fascin (1:200, Dako, Glostrup, Denmark) or anti--tubulin 8
(1:500, Roche Diagnostics GmbH, Mannheim, Germany) in 1% BSA. After washing, the cells were 9
incubated for 1 h with secondary antibodies (1:1000, Vector Laboratories Inc., Burlingame, CA). For 10
the visualization of the actin filaments, the cells were washed, fixed, permeabilized and blocked as 11
described above and incubated with 5 µg/ml Alexa Fluor 594 Phalloidin (4 U/ml, Molecular 12
Probes) for 1 h.
13 14
Confocal imaging and FRAP analysis 15
The fluorescent images were obtained with Zeiss Axio Observer inverted microscope (40 x NA 16
1.3 or 63 x NA 1.4 oil –objective) equipped with Zeiss LSM 700 confocal module (Carl Zeiss 17
Microimaging GmbH, Jena, Germany). Image processing, like 3-dimensional rendering, FRAP 18
analysis of images and further modification was performed using ZEN 2012 software (Carl Zeiss 19
Microimaging GmbH), ImageJ 1.32 software (http://rsb.info.nih.gov/ij/) and Adobe Photoshop 20
elements 10. For fluorescence recovery after photobleaching (FRAP) analyses, bleach pulses were 21
performed with maximal laser intensity in regions of interest (ROIs) and serial images were collected 22
over a 5 min period. Background fluorescence and general bleaching during acquisitions were 23
measured, and the fluorescence intensity in the ROI was normalized to the measured values.
24 25
Transmission electron microscopy 26
Ultrastructural sample preparation was performed as described in [32] . Briefly, the rat tissue samples 27
were cut to approximately 1x1x1 mm3 pieces. Prior to fixation some of the samples were treated with 28
Streptomyces hyaluronidase (5-10 TRU for 30-60 min). Tissue samples and cells were rinsed with 29
0.1 M PB, pH 7.0 and fixed with 2% glutaraldehyde (Electron microscopy science, Hatfield, USA), 30
0.2 % tannic acid (Mallinckrodt, Inc., Paris, Kentucky), 20 mM EGTA in 0.1 M PB, pH 7.0 for 10 31
min in RT and 50 min on ice. Samples were rinsed 3x5 min with 0.1 M PB, pH 7.0 and post-fixed 32
with 1 % OsO4 (Electron microscopy science, Hatfield, USA), in 0.1 M PB, pH 6.0 for 1 h on ice.
33
After postfixation, samples were rinsed 3x5 min in ice cold water, then stained with 1% uranyl 34
6
acetate at 4°C overnight, dehydrated and embedded in LX-112 resin (Ladd Research industries, 1
Burlington, VT) and polymerized at 60°C for 48 h. 70 nm sections were stained with 1% uranyl 2
acetate and examined with a JEM-2100F transmission electron microscope (Jeol Ltd, Tokyo, Japan) 3
operated at 200 kV.
4 5
Correlative light and electron microscopy 6
In addition to conventional confocal microscopy and SEM we utilized correlative light and 7
electron microscopy (CLEM) as previously described [33] . Shortly, next day after transient 8
transfection the cells were fixed with 2% glutaraldehyde and fluorescent images were obtained with 9
Zeiss LSM 700 confocal module and an external DIC-capable transmitted-light channel. After 10
routine processing for scanning electron microscopy, including dehydration and coating with gold, 11
the cells were imaged with Zeiss Sigma HD|VP (Zeiss, Oberkochen, Germany) at 3 kV.
12 13
Statistical analysis 14
Statistical comparison was carried out using Mann-Whitney U-test to analyse the data from FRAP 15
experiments. The p-values less than 0.05 were considered statistically significant.
16 17 18
RESULTS 19
20
Ultrastructure of HAS3-induced and spontaneous mesothelial protrusions is similar 21
Because HAS3-induced protrusions share some features of both microvilli and filopodia, we 22
compared their morphology with similar structures in vivo. By utilizing transmission electron 23
microscopy, we compared the ultrastructure and actin cytoskeleton of protrusions from HAS3- 24
induced MCF-7 cells (Figure 1C and F) with protrusions from intact rat mesothelium (Figure 1B 25
and E) and microvilli of the small intestine brush border (Figure 1A and D). The ultrastructure of 26
protrusions from peritoneal mesothelium was similar with HAS3-induced protrusions (Figure 1).
27
The estimated density of actin filaments was lower in mesothelial and HAS3-induced protrusions as 28
compared to the brush border microvilli of enterocytes. Also the average length, density and 29
diameter of HAS3-induced protrusions corresponded to those of mesothelial cell protrusions in vivo 30
(Table 1), but differed from microvilli of enterocytes. Hyaluronidase digestion did not change the 31
ultrastructure or actin core of protrusions (Figure 1 F).
32
Cultured human mesothelial cells have been shown to spontaneously express long hyaluronan- 33
rich protrusions [24] . Because of the similar structure of mesothelial and HAS3-induced 34
7
protrusions, immunostainings of cytoskeleton-associated proteins in protrusions of cultured human 1
mesothelial cells (LP-9) and HAS3 overexpressing MCF-7 cells were compared. It was found that 2
proteins typical for microvilli, like villin and espin are not present. Instead, proteins typical for 3
filopodia, like ezrin and fascin, are located on protrusions of both cell types. The results of the 4
immunostainings are summarized in Table 2.
5 6
HAS3 overexpression drives actin to cell cortex and base of the protrusions 7
Utilizing phalloidin staining, previous studies on fixed cells have shown that HAS3-induced 8
protrusions are positive for actin filaments (Kultti 2006). By double transfections in live cells we 9
now studied the localization of actin in live MCF-7 cells, and the possible impact of GFP-HAS3 10
overexpression. A cell expressing GFP-HAS3 with a typical “hedgehog-like” morphology is shown 11
in Figure 2A and C. To confirm that HAS3 expression without the GFP-tag induces actin-based 12
protrusions, double transfection of HAS3 without label and GFP-actin was performed (Figure 2B 13
and D). Furthermore, to compare the distribution of filamentous actin in live cells, Lifeact was used 14
as a marker for filamentous actin. Double transfected cell with Lifeact together with GFP empty 15
vector and GFP-HAS3 are shown in (Figure 2E-H) and (Figure 2I-N), respectively. Additionally, 16
distribution of actin (Figure 2E) was changed in GFP-HAS3 expressing cells (Figure 2J), and 17
significant proportion of actin filaments accumulated in the protrusions. Higher magnification in 18
panels 1M and 1N shows that the base of the HAS3-induced protrusions is highly positive for actin, 19
but the staining is weaker towards the tips of protrusions (Figure 2L and N), while HAS3 signal is 20
high in all parts of the protrusions, and in fact strongest in the tip areas (Figure 2K and M).
21 22
Myo10 localizes to the tips of HAS3-induced protrusions but is not essential for their 23
maintenance 24
Next the possible role of Myo10, actin motor protein that typically accumulates on the tips of 25
filopodia (Kerber 2011) was studied. The mCherry-Myo10 was localized in the tips of tiny, 2-4 µm 26
long protrusions of MCF-7 cells (Figure 3A), but overexpression of Myo10 alone did not induce 27
such long protrusions as found in HAS3 overexpressing MCF-7 cells (Figure 3B). Furthermore, 28
Myo10 expression alone did not induce hyaluronan secretion of cells (Figure 3H). Double 29
transfection of GFP-HAS3 and mCherry-Myo10 revealed that Myo10 specifically accumulates on 30
the tips of GFP-HAS3-induced protrusions (Figure 3B and C). To find out if Myo10 is essential for 31
the growth of HAS3-induced protrusions, constructs with deleted pleckstrin homology domains 32
were utilized. As previously reported [29] , the filopodial localization of these mutated constructs 33
was decreased (Figure 3D, E, F). However, the expression of either Myo10∆PH123 (Figure 3D), 34
8
Myo10∆PH2 (Figure 3E) or Myo10∆PH2pm (Figure 3F) did not inhibit the formation of HAS3- 1
induced protrusions. Expression of these mutated Myo10 forms did not have any significant effect 2
on hyaluronan secretion, supporting the finding that the number and function of hyaluronan 3
secreting protrusions was unaffected (Figure 3H). Furthermore, MDA-MB231 cells with stable 4
Myo10 knockout didn’t show an impairment in their ability to form HAS3-induced protrusions 5
(Figure 3G). These results indicate that Myo10 is localized to, but is not essential for the formation 6
of HAS3-induced protrusions.
7 8
HAS3-induced protrusions are dependent on ongoing HAS activity and collapse when 9
hyaluronan coat is removed 10
Because the mCherry-Myo10 specifically accumulated on the tips of HAS3-induced protrusions, it 11
was further utilized to visualize the tips of protrusions after treatments shown to inhibit hyaluronan 12
synthase activity and the growth of protrusions [25, 26] . Digestion of hyaluronan coat with 13
Streptomyces hyaluronidase resulted in a disappearance of GFP-HAS3 signal from the plasma 14
membrane in 30 min (Figure 4A and B). However, as the higher magnification shows, mCherry- 15
Myo10 signal did not disappear (arrows in Figure 4D), suggesting that in fact the protrusions do not 16
disappear, even though their GFP-HAS3 is lost (Figure 4C and D). Instead, inhibition of HAS 17
activity with a 6 h treatment with 4-methylumbelliferone (Figure 4E and G) or a 6 h glucose 18
starvation (Figure 4F and H) resulted in GFP-HAS3 loss from the plasma membrane, and also 19
relocation of mCherry-Myo10 into cytoplasmic parts of cells (arrows in Figure 4G and H), which 20
suggests total disappearance of the protrusions.
21 22
HAS3-induced protrusions are dynamic structures with rapid growth and retraction 23
Next, the dynamics of HAS3-induced protrusions was studied with 3D confocal time lapse imaging.
24
We used mCherry-Myo10 in these experiments to visualize the tips of the GFP-HAS3-positive 25
protrusions. The 3D time lapse imaging revealed that many of the protrusions were dynamic with 26
rapid growth and retraction (arrows in Figure 5). However, most of the protrusions attached to the 27
substratum were relatively stable (asterisks in Figure 5). Interestingly, intrafilopodial movements of 28
bright Myo10 puncta was occasionally seen along the HAS3-positive protrusions (arrowhead in 29
Figure 5), like reported previously [34] . This suggests that Myo10 has a functional role in the 30
intrafilopodial traffic of HAS-induced protrusions.
31 32
Lateral motility of GFP-HAS3 on the plasma membrane of protrusions is restricted 33
Fluorescence recovery after photobleaching (FRAP) was utilized to study the dynamics of GFP- 34
9
HAS3 in the plasma membrane. After bleaching the GFP, recovery rate of the signal was slower on 1
the shaft of the protrusions, (Figure 6A) as compared to that of the plain plasma membrane (Figure 2
6B). A quantitative FRAP analysis (Figure 6C) of the shafts of GFP-HAS3 positive protrusions 3
showed a significantly slower recovery (half recovery 66.6 s ± 16.5 s) as compared to the plain 4
plasma membrane (half recovery 9.17 s ± 2.7 s). This indicates that turnover of GFP-HAS3 5
molecules is low on the plasma membrane of protrusions, suggesting that hyaluronan chains 6
attached to HAS3 during ongoing synthesis, or other factors restrict the endocytosis or lateral 7
motility of HAS3 in the plasma membrane of the protrusions.
8 9
Hyaluronan chains support the protrusions and retain HAS3 on the plasma membrane 10
To confirm the findings above, correlative light and electron microscopy was utilized. GFP- 11
HAS3 signal and ultrastructure of protrusions were correlated in untreated and Streptomyces 12
hyaluronidase digested cells expressing GFP-HAS3 (Figure 7). These experiments showed that the 13
GFP-HAS3 signal, normally on the plasma membrane and its protrusions (Figure 7A, E, G) 14
translocates into intracellular vesicular structures (arrows in Figure 7B, F, H) after removal of 15
hyaluronan with hyaluronidase digestion. However, SEM revealed that in fact hyaluronidase 16
treatment does not immediately destroy the protrusions, but results in their collapse on cell surface 17
(Figure 7D, J). A similar collapse of protrusions occurs during fixation and dehydration for SEM, 18
resulting in almost identical morphology of protrusions of both control (Figure 7C, I) and 19
hyaluronidase-treated cells in SEM images (Figure 7D, J).
20
Correlative fluorescence and electron microscopy indicated that the GFP-HAS3 fluorescence in 21
the protrusions of the control cells had disappeared in the hyaluronidase-treated cells (arrows in 22
Figure 7K, L), a finding consistent with the Myo10 data in Fig. 4C,D. GFP-HAS3 thus escapes very 23
quickly from the plasma membrane as a result of hyaluronan removal, and further supports the 24
hypothesis that hyaluronan chains retain GFP-HAS3 on the plasma membrane. However, a longer 25
(6 h) hyaluronidase treatment resulted in a total loss of the protrusions, apparently in the absence of 26
the supporting hyaluronan (data not shown).
27 28
Spontaneous filopodia are requirements for and consequence of active hyaluronan synthesis 29
Because the LP-9 human mesothelial cells are known for their high rate of hyaluronan synthesis and 30
high number of hyaluronan-positive protrusions [24] , these cells were utilized to test if the natural 31
protrusions are dependent on ongoing hyaluronan synthesis like those induced by HAS3 32
overexpression. These cells were studied with SEM for detailed visualization of plasma membrane 33
protrusions. Like seen in HAS3-overexpressing cells, hyaluronidase digestion did not result in an 34
10
immediate disappearance of the protrusions, and the structure of cells with both apical and lateral 1
protrusions (arrows) was similar with (Figure 8C, D) and without digestion (Figure 8A, B).
2
However, the treatment with 4-MU (Figure 8E, F) resulted in the disappearance of most of the 3
protrusions, while removal of the 4-MU block resulted in a rapid regrowth of the protrusions, 4
clearly detected after a 3 h chase (Figure 8G, H). The effect of the 4-MU block and its removal on 5
hyaluronan-positive protrusions was also shown in live cell hyaluronan staining. It was found that 6
the hyaluronan-coated protrusions in controls (Figure 9A) were absent after an overnight 4-MU 7
treatment (Figure 9B), simultaneously with a dramatic decline in the overall thickness of the 8
hyaluronan coat. The regrowth of the protrusions and their hyaluronan coat starts quickly after 9
removal of the block, as indicated by the image in Figure 9C, taken 3 h after 4-MU removal.
10 11 12
DISCUSSION 13
14
Hyaluronan drives the formation of actin-based protrusions 15
The results of this work support the hypothesis that long plasma membrane projections are both 16
requirements for and consequences of active hyaluronan production. Interestingly, cell types with 17
low HAS expression and hyaluronan secretion, like MDCK cells [27] and COS cells [10, 35] have 18
typically a smooth surface, with only tiny protrusions on their apical plasma membrane. However, 19
cell types with high hyaluronan synthesis rate, like smooth muscle cells, chondrosarcoma cells [24]
20
, fibroblasts [18, 36] mesenchymal stem cells [37] , growth factor-induced keratinocytes [38, 39] , 21
neuroblastoma cells [40] , esophageal squamous carcinoma cells [15] , and fibroblasts from Shar 22
Pei dogs with high HAS2 expression [41] , display large numbers of plasma membrane protrusions.
23
These findings indicate that hyaluronan synthesis associates with the formation of plasma 24
membrane protrusions and regulates the dynamics of the plasma membrane and the underlaying 25
actin network.
26 27
Hyaluronan acts as an extracellular cytoskeleton for HAS-induced dorsal filopodia 28
Relatively low density of the actin filaments and the length of the HAS3-induced protrusions 29
suggests that additional support is required for their maintenance. The highly concentrated, hydrated 30
layer of hyaluronan attached to HAS on the plasma membrane creates a swelling pressure that 31
promotes the outward membrane curvature needed to initiate the growth of the protrusions, while at 32
the same time the attached hyaluronan creates an extracellular scaffold that stiffens the extensions 33
[26] . 34
11
The correlative imaging revealed that most of the apical HAS-induced protrusions collapse, but 1
do not immediately disappear upon hyaluronidase digestion, unlike it first appeared when just the 2
confocal microscopy of GFP-HAS3 was examined [26] . Kultti et al. (2006) showed that actin 3
bundles are imperative for protrusions. It seems likely that the support of actin bundles inside the 4
protrusions makes the complete disappearance of the protrusions not as rapid as the loss of HAS3 5
from plasma membrane. Interestingly, fixation and dehydration for electron microscopy results in a 6
collapse of the apical protrusions, similar to that after hyaluronidase digestion. The dehydration 7
during SEM processing, removing the swelling pressure of hyaluronan, is probably analogous to the 8
hyaluronidase treatment, and further indicates that the hydrated hyaluronan coat is essential for the 9
maintenance of protrusions. Therefore, both intra- and extracellular cytoskeleton, actin bundles 10
inside and HA containing glycocalyx outside, is necessary for these long, stand-up extensions.
11 12
Nomenclature 13
Filopodia are defined as finger-like protrusions that must eventually be adhered in some manner 14
to substratum or another cell, while microvilli are not adhered [1] . HAS-induced protrusions are in 15
the grey area between these two types of protrusions, because they share many features typical for 16
filopodia, like structure, protein composition, and dynamics, but they are not dependent on 17
adherence. Indeed, the hyaluronan coat provides mechanical support for the dorsal, finger-like 18
protrusions. The length of HAS3-induced protrusions can reach 20 µm [26] or even more, while 19
the typical length of microvilli is less than 1-3 µm [42] . We suggest that cells with active 20
hyaluronan synthesis can grow very long apical filopodia.
21 22
Spontaneous protrusions of secreting epithelial cells 23
Mesothelial cells are secretory epithelial cells lining the pleural, pericardial and abdominal 24
cavities, and are known for their active hyaluronan synthesis. In standard culture conditions, the 25
human mesothelial cell line LP-9 produces hyaluronan at levels closely corresponding to those of 26
the GFP-HAS overexpressing cells [24] and displays long protrusions covered by a thick layer of 27
hyaluronan. SEM revealed that inhibition of hyaluronan synthesis with 4-methylumbelliferone 28
decreased the number of spontaneous mesothelial cell protrusions. After removal of the blocking 29
agent, hyaluronan synthesis and regrowth of the protrusions restarted simultaneously, as shown here 30
and before [24] by applying a fluorophore-tagged hyaluronan probe on live cells. These results 31
indicate that the growth of the natural protrusions are induced by, and dependent on, active 32
hyaluronan synthesis, like those induced by GFP-HAS-transfection on epithelial cells. This is in 33
line with the present structural analyses indicating that HAS3-induced protrusions shared all 34
12 features of the protrusions of mesothelial cells.
1 2
Functions of the HAS-induced protrusions 3
The functions of hyaluronan-dependent protrusions are beginning to be elucidated. In addition to 4
providing extra plasma membrane area for active hyaluronan synthesis [24] , they act as platforms 5
for shedding of extracellular vesicles [43, 44] . However, we expect that in the near future, other 6
specific functions of the hyaluronan-induced cell protrusions will be discovered. Typical functions 7
linked to filopodia include sensoring environment [2] , glucose uptake [45] and stimulation of 8
invasion and metastasis [30] . As recently reported, HAS3 overexpression and plasma membrane 9
protrusions regulate adhesion of cells [46] . Hyaluronan-positive filopodia are suggested to enhance 10
adhesion of cancer cells [15] and monocytes [18, 36, 47] , with potentially high importance in 11
inflammation and cancer. The hyaluronan receptor CD44 is abundantly expressed in filopodia [15, 12
36, 47] . Interestingly, upregulation of Has3 [48] together with other filopodia-associated genes 13
have been reported in breast carcinomas [49] . 14
15
Role of Myo10 in the formation HAS-induced filopodia 16
Myo10 is an unconventional actin motor protein that is reported to transport cargo-vesicles to the 17
tips of filopodia, as well as to initiate filopodia formation [50] . While mCherry-labeled Myo10 was 18
accumulated in the tips of HAS3-induced protrusions, permanent silencing [30] or expression of 19
mutant forms of Myo10 [29] did not inhibit the formation of HAS3-induced protrusions, indicating 20
that Myo10 is not essential for their formation. However, we cannot exclude the possibility that 21
Myo10 contributes to functions of HAS-induced protrusions, like shedding of extracellular vesicles.
22
The previously reported filopodia induced by Myo10 are short as compared to HAS3-induced 23
protrusions [10] , suggesting that additional factors, like hyaluronan coat is required to reach the 24
unusual length typical of the HAS3-induced dorsal filopodia.
25 26
Final conclusions 27
This work strengthens the role of filopodia as prime sites and specialized organelles for 28
hyaluronan synthesis, and the role of HAS activity in the regulation of filopodia formation. The 29
results of this work support the hypothesis that hyaluronan secretion in plasma membrane 30
protrusions is a universal mechanism, crucial for the creation of a hyaluronan-rich pericellular and 31
extracellular matrix and for the secretion of the viscous body fluids in body cavities like peritoneum 32
and synovial joints. Hyaluronan producing protrusions are both a requirement for, and consequence 33
of an active hyaluronan secretion, suggesting that hyaluronan contributes to the assembly of 34
13
protrusions. These protrusions may have a high biological importance, like regulation of invasion, 1
adhesion, microenvironmental sensing and modulation. Furthermore, they may produce most of the 2
total hyaluronan secreted by cells, and be responsible for the increased hyaluronan content of 3
tissues and body fluids in inflammation and cancer. They are indicators of active hyaluronan 4
production and malignant properties of cells, like invasion and multidrug resistance. Additionally, 5
they act as potential targets of therapy in pathological conditions associated with excess of 6
hyaluronan.
7 8
Properties of cell protrusions
Intestine Brush border
Peritoneal Mesothelium HYAL-
Peritoneal Mesothelium HYAL+
MCF-7 HYAL-
MCF-7 HAS3 dox- HYAL-
MCF-7 HAS3 dox+
HYAL-
MCF-7 HAS3 dox+
HYAL+
LP9 HYAL-
LP9 HYAL+
Diameter (m)
0.1050.03 n=214
0.1010.02 n=77
0.1070.01 n=46
0.1360.06 n=40
0.1440,02 n=15
0.1650.03 n=94
0.1490.02 n=77
0.1330,02 n=12
0.1340.03
Lenght (m)
0.900.31 n=13
3.090,02 n=67
3.991,14 n=138
n.d. n.d. n.d. n.d. n.d. n.d.
Number of actin filaments
247 n=201
93 n=77
93 n=46
72 n=35
73 n=15
83 n=94
83 n=77
134 n=12
142 n=10
Terminal web
+ - - - - - - - -
9
Table 1. Comparison of the morphology and properties of actin-based finger-like protrusions 10
of tissues and HAS-induced protrusions of cultured cells. TEM images were utilized for all 11
measurements. n.d. = not determined.
12 13 14
Protein HAS3-induced protrusions
Spontaneous HA- rich protrusions
actin + +
tubulin - -
CD44 + +
villin - -
ezrin + +
espin - -
fascin + +
15
Table 2. Cytoskeleton-associated proteins in HA secreting protrusions. Immunostainings were 16
utilized to compare the protrusions of HAS3-induced protrusions of MCF-7 cells and spontaneous 17
14
protrusions of mesothelial cells (LP-9) for expression of typical cytoskeleton-associated proteins.
1
Proteins typical for microvilli, like villin and espin are not present, instead proteins typical for 2
filopodia, like ezrin and fascin are located on both types of HA secreting protrusions.
3 4
LEGENDS TO THE FIGURES 5
6
Figure 1. Ultrastructure of actin-based protrusions of different tissues and cells. Cross-sections 7
(A, B, C, F) and longitudal sections (D, E) of microvilli from rat intestinal lumen (A, D), 8
protrusions from rat peritoneal mesothelium (B, E) and from cultured MCF-7 cell overexpressing 9
GFP-HAS3 without (C) and with (F) hyaluronidase digestion, imaged with transmission electron 10
microscopy. Inserts in panels A, B, C and F show high magnification images of cross sections of 11
the protrusions. Magnification bar 500 nm, in inserts 50 nm.
12 13
Figure 2. Overexpression of HAS3 has impact on the subcellular localization of filamentous 14
actin, which is accumulated to the basal areas of HAS3-induced extensions. 3D projections of 15
MCF-7 cells double-transfected with constructs expressing GFP-HAS3, HAS3, GFP-actin and 16
Lifeact (RFP-actin). For each combination, a vertical view and corresponding horizontal view are 17
shown. Panels M and N show a detailed view from panels I and J, respectively. Magnification bar 18
in L = 10 µm and in N = 5 µm.
19 20
Figure 3. Myo10 localizes to, but is not essential for the formation of HAS3-induced 21
protrusions. Confocal 3D images of MCF-7 cells transfected with (A) GFP empty vector and 22
mCherry-Myo10, (B) GFP-HAS3 and mCherry-Myo10, (C) a bigger magnification of an area of the 23
cell presented in B is shown. (D) Double transfection of GFP-HAS3 with Myo10∆PH123, 24
Myo10∆PH2 (E) or Myo10∆PH2pm (F) and Myo10 knockdown MDA-MB-231 cell transfected 25
with GFP-HAS3 (G). Hyaluronan secretion levels of MCF-7 cells transiently transfected with 26
HAS3 and Myo10 constructs in different combinations (H) Error bar = SD of three parallel 27
samples. Magnification bar = 20 µm.
28 29
Figure 4. Effect of inactivation of HAS3 on the protrusions and wild type Myo10 localization.
30
Confocal 3D images of MCF-7 cells transfected with GFP-HAS3 and mCherry-Myo10. The same 31
cell before (A) and after (B) hyaluronidase digestion (1 TRU/ml, 30 min), cell treated with 4-MU (1 32
mM for 6 h) is shown in (E) and a cell grown for 6 hours without glucose in (F). A selected area 33
with higher magnification from A and B is shown in C and D and high magnification area from E 34
15
and F are shown in G and H, respectively. Arrows indicate the Myo10-positive puncta.
1
Magnification bar = 20 µm.
2 3
Figure 5. HAS3-induced protrusions are dynamic structures with rapid growth and 4
retraction. Serial time lapse images of a section of a MCF-7 cell expressing GFP-HAS3 and 5
mCherry-Myo10. Arrows indicate the rapid changes in length, orientation, growth and retraction of 6
protrusions. Arrowheads show the movement of Myo10 along the body of filopodia. Some of the 7
protrusions, attached to the substratum, were more stable (asterisks). Magnification bar = 5 µm. See 8
supplemental movie 2.
9 10
Figure 6. FRAP analysis suggests that lateral movements of GFP-HAS3 on the plasma 11
membrane of protrusions are restricted. Representative series of images of GFP-HAS3 on the 12
shafts (A) of plasma membrane of protrusions and on the plain plasma membrane (B) before 13
bleaching (prebleach), immediately after the bleach pulse (bleach), and during recovery (60 s and 14
120 s) are shown. Columns in C represent the half-recovery times ± standard errors of the means 15
(SEM) (plain plasma membrane and shafts of protrusions, n = 8 in both groups). Significant 16
difference (p < 0.05) between groups is indicated by asterisks (**) in D. Magnification bar in A = 5 17
µm.
18 19
Figure 7. CLEM reveals that hyaluronan chains externally support the protrusions and retain 20
GFP-HAS3 on the plasma membrane. GFP-HAS3 expressing MCF-7 cells imaged by 3D 21
confocal microscopy in (A, B, E, F, G and H) and by SEM in (C, D, I and J). Untreated cultures are 22
shown in (A, C, E, G, I and K) and cultures treated with Streptomyces hyaluronidase are shown in 23
(B, D, F, H, J and I). A, B, E and F are surface projections from stacks of optical sections, G and H 24
show vertical views of cells to indicate dorsal GFP-HAS3-positive plasma membrane protrusions.
25
Surprisingly, SEM revealed that most of the HAS-induced protrusions collapse after enzymatic 26
hyaluronan digestion (D, J), but instead of total retraction and disappearance suggested by confocal 27
microscopy (B, F and H) remain lying on the cells. Asterisks in A and B indicate the cells that are 28
shown in higher magnifications in (E, G, I and K) and (F, H, J and L), respectively. Panels K and L 29
show merged images of those cells. Arrows in B, F and H indicate the GFP-HAS3 accumulation 30
into intracellular vesicles and in K and L indicate the presence (K) and absence (L) of the GFP- 31
HAS3 signal on the plasma membrane and its protrusions. Scale bars 20 µm in B and 10 µm in H.
32 33
Figure 8. Dorsal filopodia are requirements for and consequence of active hyaluronan 34
16
synthesis. Scanning electron microscopy shows the numerous finger-like plasma membrane 1
protrusions with variable length (arrows) of the non-transfected human mesothelial (LP-9) cells.
2
Untreated cells are shown in (A and B), cells treated with Streptomyces hyaluronidase (5 TRU/ml, 3
30 min) are shown in (C and D). In E and F, cell were treated with hyaluronan synthesis inhibitor 4- 4
MU (1 mM), that efficiently removes the protrusions in an overnight treatment (E, F). After 5
removal of the 4-MU block, the protrusions start to regrow after 1 hour (data not shown), and after 6
3 hours (G, H) are even more numerous than in control cells. Magnification bar 20 µm in G and 5 7
µm in H.
8 9
Figure 9. Hyaluronan secretion and growth of microvilli are reversibly inhibited by 4-MU.
10
The hyaluronan coated microvilli of LP-9 cells labelled with fHABC probe (red) (A) disappeared 11
after 24 h treatment with 1 mM 4-MU (B). After 4-MU was removed, the regrowth of microvilli 12
and hyaluronan synthesis start again quickly (3h) after removal of the block (C). Arrows indicate 13
hyaluronan-positive protrusions. Single optical confocal sections are shown in A-C. Magnification 14
bar 10 µm.
15 16
Supplementary movie 1. A time-lapse movie shows the dynamic movements of GFP-HAS3- 17
induced protrusions. Movie plays at 10 frames/s and shows GFP-HAS3 fluorescence signal as a 18
maximum projection from stack of images in a section of a MCF-7 cell during 10 min. The frames 19
were collected in every 20 seconds and the movie plays at 15 frames/s..
20 21
Supplementary movie 2. A time-lapse movie which shows GFP-HAS3 (green) and mCherry- 22
Myo10 (red) fluorescence signal as a maximum projection from stack of images in a section of a 23
MCF-7 cell and its protrusions during 11 min. The frames were collected in every 6 seconds and the 24
movie plays at 10 frames/s.
25 26 27
ACKNOWLEDGEMENTS 28
Special thanks are due to Eija Rahunen and Virpi Miettinen for expert technical assistance. This 29
work was supported by the Academy of Finland (grant #276426), Sigrid Juselius Foundation and 30
the Spearhead Funds from the University of Eastern Finland (Cancer Center of Eastern Finland).
31 32 33
REFERENCES 34
17
[1] E.S. Chhabra, H.N. Higgs, The many faces of actin: Matching assembly factors with cellular 1
structures, Nat. Cell Biol. 9 (2007) 1110-1121.
2
[2] C.A. Heckman, H.K. Plummer 3rd, Filopodia as sensors, Cell. Signal. 25 (2013) 2298-2311.
3
[3] M.D. Welch, R.D. Mullins, Cellular control of actin nucleation, Annu. Rev. Cell Dev. Biol. 18 4
(2002) 247-288.
5
[4] J.R. Bartles, Parallel actin bundles and their multiple actin-bundling proteins, Curr. Opin. Cell 6
Biol. 12 (2000) 72-78.
7
[5] E. Ferrary, M. Cohen-Tannoudji, G. Pehau-Arnaudet, A. Lapillonne, R. Athman, T. Ruiz, L.
8
Boulouha, F. El Marjou, A. Doye, J.J. Fontaine, C. Antony, C. Babinet, D. Louvard, F. Jaisser, S.
9
Robine, In vivo, villin is required for ca(2+)-dependent F-actin disruption in intestinal brush 10
borders, J. Cell Biol. 146 (1999) 819-830.
11
[6] J.C. Adams, Fascin protrusions in cell interactions, Trends Cardiovasc. Med. 14 (2004) 221- 12
226.
13
[7] P.A. Loomis, L. Zheng, G. Sekerkova, B. Changyaleket, E. Mugnaini, J.R. Bartles, Espin cross- 14
links cause the elongation of microvillus-type parallel actin bundles in vivo, J. Cell Biol. 163 (2003) 15
1045-1055.
16
[8] A.I. McClatchey, ERM proteins at a glance, J. Cell. Sci. 127 (2014) 3199-3204.
17
[9] E. Robles, S. Woo, T.M. Gomez, Src-dependent tyrosine phosphorylation at the tips of growth 18
cone filopodia promotes extension, J. Neurosci. 25 (2005) 7669-7681.
19
[10] A.B. Bohil, B.W. Robertson, R.E. Cheney, Myosin-X is a molecular motor that functions in 20
filopodia formation, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 12411-12416.
21
18
[11] J.S. Berg, B.H. Derfler, C.M. Pennisi, D.P. Corey, R.E. Cheney, Myosin-X, a novel myosin 1
with pleckstrin homology domains, associates with regions of dynamic actin, J. Cell. Sci. 113 Pt 19 2
(2000) 3439-3451.
3
[12] S. Yung, T.M. Chan, Hyaluronan--regulator and initiator of peritoneal inflammation and 4
remodeling, Int. J. Artif. Organs. 30 (2007) 477-483.
5
[13] P.H. Weigel, P.L. DeAngelis, Hyaluronan synthases: A decade-plus of novel 6
glycosyltransferases, J. Biol. Chem. 282 (2007) 36777-36781.
7
[14] D.H. Bernanke, R.R. Markwald, Effects of hyaluronic acid on cardiac cushion tissue cells in 8
collagen matrix cultures, Tex. Rep. Biol. Med. 39 (1979) 271-285.
9
[15] S. Twarock, M.I. Tammi, R.C. Savani, J.W. Fischer, Hyaluronan stabilizes focal adhesions, 10
filopodia, and the proliferative phenotype in esophageal squamous carcinoma cells, J. Biol. Chem.
11
285 (2010) 23276-23284.
12
[16] J. Bakkers, C. Kramer, J. Pothof, N.E. Quaedvlieg, H.P. Spaink, M. Hammerschmidt, Has2 is 13
required upstream of Rac1 to govern dorsal migration of lateral cells during zebrafish gastrulation, 14
Development. 131 (2004) 525-537.
15
[17] L.A. Gurski, X. Xu, L.N. Labrada, N.T. Nguyen, L. Xiao, K.L. van Golen, X. Jia, M.C.
16
Farach-Carson, Hyaluronan (HA) interacting proteins RHAMM and hyaluronidase impact prostate 17
cancer cell behavior and invadopodia formation in 3D HA-based hydrogels, PLoS One. 7 (2012) 18
e50075.
19
[18] S.P. Evanko, S. Potter-Perigo, P.Y. Johnson, T.N. Wight, Organization of hyaluronan and 20
versican in the extracellular matrix of human fibroblasts treated with the viral mimetic poly I:C, J.
21
Histochem. Cytochem. 57 (2009) 1041-1060.
22
19
[19] P. Bono, E. Cordero, K. Johnson, M. Borowsky, V. Ramesh, T. Jacks, R.O. Hynes, Layilin, a 1
cell surface hyaluronan receptor, interacts with merlin and radixin, Exp. Cell Res. 308 (2005) 177- 2
187.
3
[20] S. Makabe, T. Naguro, T. Stallone, Oocyte-follicle cell interactions during ovarian follicle 4
development, as seen by high resolution scanning and transmission electron microscopy in humans, 5
Microsc. Res. Tech. 69 (2006) 436-449.
6
[21] M. Lukoschek, K. Addicks, Histologic, morphometric study of the human synovial membrane, 7
Z. Orthop. Ihre. Grenzgeb. 129 (1991) 136-140.
8
[22] S.E. Mutsaers, The mesothelial cell, Int. J. Biochem. Cell Biol. 36 (2004) 9-16.
9
[23] J.E. Hale, R.E. Wuthier, The mechanism of matrix vesicle formation. studies on the 10
composition of chondrocyte microvilli and on the effects of microfilament-perturbing agents on 11
cellular vesiculation, J. Biol. Chem. 262 (1987) 1916-1925.
12
[24] K. Rilla, R. Tiihonen, A. Kultti, M. Tammi, R. Tammi, Pericellular hyaluronan coat visualized 13
in live cells with a fluorescent probe is scaffolded by plasma membrane protrusions, J. Histochem.
14
Cytochem. 56 (2008) 901-910.
15
[25] K. Rilla, H. Siiskonen, A.P. Spicer, J.M. Hyttinen, M.I. Tammi, R.H. Tammi, Plasma 16
membrane residence of hyaluronan synthase is coupled to its enzymatic activity, J. Biol. Chem. 280 17
(2005) 31890-31897.
18
[26] A. Kultti, K. Rilla, R. Tiihonen, A.P. Spicer, R.H. Tammi, M.I. Tammi, Hyaluronan synthesis 19
induces microvillus-like cell surface protrusions, J. Biol. Chem. 281 (2006) 15821-15828.
20
20
[27] K. Rilla, S. Pasonen-Seppänen, R. Kärnä, H.M. Karjalainen, K. Törrönen, V. Koistinen, M.I.
1
Tammi, R.H. Tammi, T. Teräväinen, A. Manninen, HAS3-induced accumulation of hyaluronan in 2
3D MDCK cultures results in mitotic spindle misorientation and disturbed organization of 3
epithelium, Histochem. Cell Biol. 137 (2012) 153-164.
4
[28] K. Törrönen, K. Nikunen, R. Kärnä, M. Tammi, R. Tammi, K. Rilla, Tissue distribution and 5
subcellular localization of hyaluronan synthase isoenzymes, Histochem. Cell Biol. 141 (2014) 17- 6
31.
7
[29] L. Plantard, A. Arjonen, J.G. Lock, G. Nurani, J. Ivaska, S. Stromblad, PtdIns(3,4,5)P(3) is a 8
regulator of myosin-X localization and filopodia formation, J. Cell. Sci. 123 (2010) 3525-3534.
9
[30] A. Arjonen, R. Kaukonen, E. Mattila, P. Rouhi, G. Högnäs, H. Sihto, B.W. Miller, J.P. Morton, 10
E. Bucher, P. Taimen, R. Virtakoivu, Y. Cao, O.J. Sansom, H. Joensuu, J. Ivaska, Mutant p53- 11
associated myosin-X upregulation promotes breast cancer invasion and metastasis, J. Clin. Invest.
12
124 (2014) 1069-1082.
13
[31] E.L. Hiltunen, M. Anttila, A. Kultti, K. Ropponen, J. Penttinen, M. Yliskoski, A.T. Kuronen, 14
M. Juhola, R. Tammi, M. Tammi, V.M. Kosma, Elevated hyaluronan concentration without 15
hyaluronidase activation in malignant epithelial ovarian tumors, Cancer Res. 62 (2002) 6410-6413.
16
[32] P.S. Hegan, V. Mermall, L.G. Tilney, M.S. Mooseker, Roles for drosophila melanogaster 17
myosin IB in maintenance of enterocyte brush-border structure and resistance to the bacterial 18
pathogen pseudomonas entomophila, Mol. Biol. Cell. 18 (2007) 4625-4636.
19
[33] K. Rilla, A. Koistinen, Correlative light and electron microscopy (CLEM) reveals the HAS3- 20
induced dorsal plasma membrane ruffles, Int J Cell Biol. in press (2015) . 21
21
[34] M.L. Kerber, D.T. Jacobs, L. Campagnola, B.D. Dunn, T. Yin, A.D. Sousa, O.A. Quintero, 1
R.E. Cheney, A novel form of motility in filopodia revealed by imaging myosin-X at the single- 2
molecule level, Curr. Biol. 19 (2009) 967-973.
3
[35] K. Rilla, S. Oikari, T.A. Jokela, J.M. Hyttinen, R. Kärnä, R.H. Tammi, M.I. Tammi, 4
Hyaluronan synthase 1 (HAS1) requires higher cellular UDP-GlcNAc concentration than HAS2 and 5
HAS3, J. Biol. Chem. 288 (2013) 5973-5983.
6
[36] S. Meran, J. Martin, D.D. Luo, R. Steadman, A. Phillips, Interleukin-1beta induces hyaluronan 7
and CD44-dependent cell protrusions that facilitate fibroblast-monocyte binding, Am. J. Pathol. 182 8
(2013) 2223-2240.
9
[37] C. Qu, K. Rilla, R. Tammi, M. Tammi, H. Kröger, M.J. Lammi, Extensive CD44-dependent 10
hyaluronan coats on human bone marrow-derived mesenchymal stem cells produced by hyaluronan 11
synthases HAS1, HAS2 and HAS3, Int. J. Biochem. Cell Biol. 48C (2014) 45-54.
12
[38] L. Barnes, F. Ino, F. Jaunin, J.H. Saurat, G. Kaya, Inhibition of putative hyalurosome platform 13
in keratinocytes as a mechanism for corticosteroid-induced epidermal atrophy, J. Invest. Dermatol.
14
133 (2013) 1017-1026.
15
[39] S. Karvinen, S. Pasonen-Seppänen, J.M. Hyttinen, J.P. Pienimäki, K. Törrönen, T.A. Jokela, 16
M.I. Tammi, R. Tammi, Keratinocyte growth factor stimulates migration and hyaluronan synthesis 17
in the epidermis by activation of keratinocyte hyaluronan synthases 2 and 3, J. Biol. Chem. 278 18
(2003) 49495-49504.
19
[40] A. Pusch, A. Boeckenhoff, T. Glaser, T. Kaminski, G. Kirfel, M. Hans, B. Steinfarz, D.
20
Swandulla, U. Kubitscheck, V. Gieselmann, O. Brustle, J. Kappler, CD44 and hyaluronan promote 21
22
invasive growth of B35 neuroblastoma cells into the brain, Biochim. Biophys. Acta. 1803 (2010) 1
261-274.
2
[41] M.J. Docampo, G. Zanna, D. Fondevila, J. Cabrera, C. Lopez-Iglesias, A. Carvalho, S. Cerrato, 3
L. Ferrer, A. Bassols, Increased HAS2-driven hyaluronic acid synthesis in shar-pei dogs with 4
hereditary cutaneous hyaluronosis (mucinosis), Vet. Dermatol. 22 (2011) 535-545.
5
[42] S.W. Crawley, M.S. Mooseker, M.J. Tyska, Shaping the intestinal brush border, J. Cell Biol.
6
207 (2014) 441-451.
7
[43] K. Rilla, H. Siiskonen, M. Tammi, R. Tammi, Hyaluronan-coated extracellular vesicles-a novel 8
link between hyaluronan and cancer, Adv. Cancer Res. 123 (2014) 121-148.
9
[44] K. Rilla, S. Pasonen-Seppänen, A.J. Deen, V.V. Koistinen, S. Wojciechowski, S. Oikari, R.
10
Kärnä, G. Bart, K. Törrönen, R.H. Tammi, M.I. Tammi, Hyaluronan production enhances shedding 11
of plasma membrane-derived microvesicles, Exp. Cell Res. 319 (2013) 2006-2018.
12
[45] K. Lange, Role of microvillar cell surfaces in the regulation of glucose uptake and organization 13
of energy metabolism, Am. J. Physiol. Cell. Physiol. 282 (2002) C1-26.
14
[46] A.J. Deen, K. Rilla, S. Oikari, R. Kärnä, G. Bart, J. Häyrinen, A.R. Bathina, A. Ropponen, K.
15
Makkonen, R.H. Tammi, M.I. Tammi, Rab10-mediated endocytosis of the hyaluronan synthase 16
HAS3 regulates hyaluronan synthesis and cell adhesion to collagen, J. Biol. Chem. 289 (2014) 17
8375-8389.
18
[47] T. Murai, M. Sato, H. Nishiyama, M. Suga, C. Sato, Ultrastructural analysis of nanogold- 19
labeled cell surface microvilli in liquid by atmospheric scanning electron microscopy and their 20
relevance in cell adhesion, Int. J. Mol. Sci. 14 (2013) 20809-20819.
21
23
[48] P. Auvinen, K. Rilla, R. Tumelius, M. Tammi, R. Sironen, Y. Soini, V.M. Kosma, A.
1
Mannermaa, J. Viikari, R. Tammi, Hyaluronan synthases (HAS1-3) in stromal and malignant cells 2
correlate with breast cancer grade and predict patient survival, Breast Cancer Res. Treat. 143 (2014) 3
277-286.
4
[49] A. Arjonen, R. Kaukonen, J. Ivaska, Filopodia and adhesion in cancer cell motility, Cell. Adh 5
Migr. 5 (2011) 421-430.
6
[50] M.L. Kerber, R.E. Cheney, Myosin-X: A MyTH-FERM myosin at the tips of filopodia, J. Cell.
7
Sci. 124 (2011) 3733-3741.
8
9
Figure 1
Click here to download high resolution image
Figure 2
Click here to download high resolution image
Figure 3
Click here to download high resolution image
Figure 4
Click here to download high resolution image
Figure 5
Click here to download high resolution image
Figure 6
Click here to download high resolution image
Figure 7
Click here to download high resolution image
Figure 8
Click here to download high resolution image
Figure 9
Click here to download high resolution image
Supplementary material for online publication only
Click here to download Supplementary material for online publication only: Movie 1.avi
Supplementary material for online publication only
Click here to download Supplementary material for online publication only: Movie 2.avi
Video Still
Click here to download high resolution image
Video Still
Click here to download high resolution image
