,||
,||
* Departments of Cellular and Molecular Biology and Pathogenic Bioagents, and
Clinical Medicine of the Faculty of Medicine of Ribeirão Preto, and
Faculty of Pharmaceutical Sciences of Ribeirão Preto of the University of São Paulo;
Medical School of the Ribeirão Preto University (UNAERP), Brazil; and
|| Blood Center Foundation of Ribeirão Preto, SP, Brazil
Correspondence: Enilza M. Espreafico, Av. Bandeirantes, 3900, Ribeirão Preto, SP, 14049-900, Brazil. E-mail: emesprea{at}fmrp.usp.br
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: Griscelli syndrome • centrosome • natural killer cells • Rab27a • dilute gene • endosomes
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Antibodies
Monoclonal anti-human cluster of differentiation (CD) markers
and antimajor histocompatibility complex (MHC) class II were purchased
from Becton Dickinson (San Jose, CA). Monoclonal anti-ß-tubulin,
anti-DLC8, and anti-LAMP-1 were obtained from Transduction Laboratories
(Lexington, KY). Monoclonal antibody (mAb) to 
-tubulin was purchased
from Sigma Chemical Co. (St. Louis, MO). We used two affinity-purified
rabbit polyclonal antibodies against chicken myosin-Va: the antibody
raised to the recombinant tail protein (32a clone) [11
]
for the first flow cytometry analysis in cell subpopulations and for
all the other analyses, an antibody raised against the medial-tail
region, which has been shown to be highly specific to a polypeptide of
190 kDa in whole brain and melanoma cell extracts and to give a
staining pattern equivalent to the anti-32a antibody in cells and
tissues (unpublished results). Secondary antibodies conjugated with
fluorescein isothiocyanate (FITC) or Texas Red were obtained from
Molecular Probes (Junction City, OR).
Cell isolation
PBMC were freshly isolated from whole heparinized blood by
Ficoll-Hypaque density-gradient technique (HYSTOPAQUE-1077; Sigma
Chemical Co.). Fresh, isolated cells were immunolabeled for flow
cytometry or confocal microscopy analyses, or they were taken into
culture as described below.
Immunolabeling and flow cytometry
Mononuclear cells from 19 healthy donors were used in these
assays. Cells were washed three times in phosphate-buffered saline
(PBS), pH 7.2, and resuspended in PBS at a density of 5 x
109 cells/L. Cell-surface labeling was performed by a
direct fluorescence technique, in aliquots of 100 µl cell suspension,
using 5 µl each of the phycoerythrin (PE)-conjugated mAb (Becton
Dickinson) against human CD3 (T lymphocyte), CD4 (T-helper
lymphocytes), CD8 (T-cytotoxic lymphocytes), CD16 [natural killer (NK)
cells], CD19 (B lymphocytes), CD56 (NK cells), and CD14 (monocytes),
respectively. Labeled mononuclear cells were washed three times with
PBS, followed by fixation and permeabilization in the same buffer
containing 2% paraformaldehyde and 0.01% saponin during 10 min at
37°C. After washing three times, they were blocked for 1 h in
PBS containing 2% bovine serum albumin (BSA) and 5% goat serum and
labeled by an indirect fluorescence technique with antimyosin-Va and a
FITC-conjugated anti-rabbit immunoglobulin G (IgG), both diluted at
10 µg/ml in blocking solution. Both incubations were performed for
1 h followed by four times washing with PBS. As a control for
nonspecific staining and to set the background level, cells were
labeled under the same conditions as specified above with PE- and
FITC-conjugated 12 antibody. Flow cytometry analyses were carried
out on FACSORT equipment (Becton Dickinson) using a linear amplifier
with a resolution of 1024 and the CELLQUEST analysis software.
Fluorescence intensity was expressed in arbitrary units (number of
channel-shifting) in a histogram plot on a 0–1023 linear scale for
10,000 cells counted in the lymphocyte gate and 15,000 in the monocyte
gate per blood sample analyzed. The results were expressed as the
median of fluorescence intensity, calculated from 10,000 or 15,000
cells, subtracted from the background fluorescence, for each cell
subpopulation from each one of the 19 blood samples analyzed, i.e.,
median of number of channel-shifting from isotype control.
Immunolabeling and confocal microscopy
Cells were plated on Cell-Tak (Becton Dickinson)-treated glass
coverslips inside 35-mm diameter Petri dishes. For the colabeling
studies with the CD markers, immunostaining was done as described above
for flow cytometry. For all the other colabeling assays, cells were
fixed and permeabilized as described above and then incubated with
blocking solution (2% BSA, 5% goat serum in PBS) for 1 h at room
temperature, followed by incubation with the primary antibodies (10
µg/ml) diluted in blocking solution, for 1 h at 37°C. Cells
were washed four times with PBS, incubated for 30 min at 37°C with
the secondary antibodies (10 µg/ml in blocking solution), FITC- or
Texas Red-conjugated, and then washed again with PBS, pH 8.3.
Coverslips were mounted in 1 mg/ml p-phenylenediamine in
90% glycerol and PBS and analyzed in a Leica confocal system, using
Leica TCS-NT software (Leica, Microsystems Heidelberg GmbH,
Germany) for image acquisition and processing.
Lymphocyte stimulation with PHA or OKT3 antibody
Freshly isolated PBMC from 18 healthy subjects were cultured in
RPMI-1640 medium (Sigma Chemical Co.) supplemented with L-glutamine (2
mM), 10% decomplemented fetal bovine serum, penicillin (100 units/ml),
and streptomycin (100 µG/ml) at a density of 2.5 x
109 cells/L in a flat-bottomed, 24-well plate (Costar,
Cambridge, MA) in a humidified atmosphere with 5% CO2 at
37°C. For lymphocyte stimulation, the culture medium was supplemented
with 10 µl OKT3 mAb to CD3 (Ortho Biotech Products, LP) or
with 2.5 µg/ml PHA (Sigma Chemical Co.). Cells were cultured for
24 h for OKT3 or during a period of 72 h for PHA stimulation.
Nonstimulated cells cultured for 72 h were tested for viability in
flow cytometry by the Annexin-V-FITC staining, using the APOPTEST-FITC
protocol (A-700) and propidium iodide staining (Nexins Research B.V.).
Viability above 95% was obtained.
Flow cytometry analysis in resting and activated T lymphocytes
To evaluate the effect of T-lymphocyte activation on myosin-Va
expression, we analyzed fluorescence intensity for myosin-Va staining
in CD3+ lymphocytes. Mononuclear cells were isolated from a total
number of 18 blood samples and were cultured as specified.
Immunolabeling procedures and fluorescein-activated cell sorter (FACS)
analysis for CD3 and myosin-Va staining were done as described above.
To confirm activation, cells were coimmunolabeled for CD3 and CD69 or
CD3 and HLA-DR by a direct fluorescence technique with mAb, conjugated
with FITC or PE. The antibody to HLA-DR was directed to a monomorphic
region. FACS analysis was done as explained above, and the results were
expressed as % of stained cells (above background levels) for CD69 and
HLA-DR and median of fluorescence intensity for myosin-Va staining.
Immunoblotting
Cells were cultured as described above for 72 h, without or
with the addition of PHA, in 25 cm2-vented flasks. After
this period, cells were washed in PBS and resuspended in 1 ml PBS, and
an aliquot of the cell suspension was taken for cell counting in a
Neübauer chamber in the presence of Trypan blue staining. (Near
100% viability was obtained in both cultures.) Cell suspension volume
was raised to obtain 1 x 106 cells/ml, 1 ml each cell
suspension was spun, and the pelleted cells were homogenized directly
in 100 µl sodium dodecyl sulfate-polyacrylamide gel electrophoresis
(SDS-PAGE) sample buffer (2x concentrated), boiled for 10 min, spun
briefly, ultrasonicated 3 x 5 s, and spun for 10 min. A
serial dilution of 2, 4, and 8 times was performed for both samples. A
volume of 10 µl from each dilution, containing the amount of total
protein, respectively, from 1, 0.5, 0.25, and 0.125 x
105 cells, was subjected to SDS-PAGE on 6–21% acrylamide
gels. Electrophoresed proteins from one gel were transfered to 0.22
µm nitrocellulose using a Hoeffer transfer apparatus for 12 h at
200 mA. An identical gel was subjected to protein staining with
SyproOrange (Molecular Probes) diluted at 1:5000 in 7.5% acetic acid
for 40 min under slow agitation protected from room light. Gel was
briefly washed (5 min) in 7.5% acetic acid and scanned in a Storm
instrument from Molecular Dynamics (Sunnyvale, CA), and the total
protein on each lane was analyzed based on the optical density (OD)
using the ImageQuant software. The results were expressed as % of OD,
relative to the sum of OD values (100%) for the eight lanes in the
gel, including the nonstimulated and PHA-stimulated cells. The analysis
indicated that equivalent amounts of total protein for stimulated and
nonstimulated cells were loaded on the gels. The nitrocellulose
membrane was subjected to immunodetection assay for myosin-Va and
-tubulin simultaneously as follows. The membrane was blocked in 5%
nonfat milk in Tris-buffered saline (TBS) with 0.2% Tween-20 (TBS-T)
for 1 h at room temperature and then was incubated with
affinity-purified polyclonal antibody to the medial tail of myosin-Va
diluted at 1 µg/ml and a mAb to -tubulin (clone GTU-88) 1:2000 in
TBS-T for 2 h at room temperature. After the wash, the membrane
was incubated for 1 h with goat anti-rabbit or goat anti-mouse
peroxidase-conjugated secondary antibodies (Promega, Madison, WI),
diluted at 1:2500 in blocking buffer, and developed using the enhanced
chemiluminescence (ECL) Western blotting detection reagent (Amersham
Pharmacia Biotech, Little Chalfont, UK) and Kodak BIOMAX Light-1 film,
according to the manufacturer’s directions. For myosin-Va detection,
the membrane exposed to the chemiluminescent substrate was allowed
exposure to film for 20 s, and for -tubulin detection, the
exposure was 1–2 s. OD for each band in the gel was measured by using
the ImageJ software, a Java port to NIH Image
(http://rsb.info.nih.gov/nih-image/), on a PC using the Linux
operational system. The results were expressed as % of OD relative to
the sum (100%) of the eight OD values obtained for each protein (four
dilutions from each of the cell samples).
RNA preparation and RT-PCR analysis
Cells were cultured for 72 h in 5 ml medium, without or
with the addition of PHA, in 25 cm2-vented culture flasks.
For total RNA extraction, cell suspensions were spun, and the cell
pellets were homogenized quickly in 1 ml Trizol (Life Technologies,
Gibco-BRL, Paisley, UK), according to the manufacturer’s directions,
except that precipitation was done overnight at -20°C. Dry RNA
pellets were dissolved in diethyl pyrocarbonate-treated water and
quantified by spectrophotometer analysis at 260/280 nm. For RT-PCR,
total RNA was treated with DNAse I (Promega) at 1 U/2 µg total RNA in
10 µl reaction volume and incubated for 30 min at 37°C, followed by
enzyme inactivation by the addition of 1 µl 20 mM
ethyleneglycol-bis(ß-aminoethylether)-N,N'-tetraacetic
acid (EGTA) and incubation for 15 min at 65°C. cDNA synthesis was
performed for three different amounts of total RNA (2.0, 0.2, and 0.04
µg) in 20 µl reaction with the Superscript II Reverse Transcriptase
(Life Technologies, Gibco-BRL), according to the manufacturer’s
instructions, using 4 µl 5x first-strand buffer, 1 µl 10 mM dNTP,
200 U Superscript II enzyme, 2 µl 0.1 M dithiothreitol (DTT), and 250
nG oligo dT primer (Life Technologies, Gibco-BRL). For PCR reactions, 1
µl each synthesized cDNA was used as template in a reaction volume of
25 µl containing 200 µM dNTPs, 1.5 mM MgCl2, 0.25 µM
each primer, and 1 U Taq DNA polymerase in the manufacturer’s
recommended buffer (Life Technologies, Gibco-BRL). The reaction was
allowed to denature for 4 min at 94°C followed by amplification. For
myosin-Va and ß-actin amplification, 34 cycles (1 min at 94°C, 1
min at 55°C, and 1 min at 72°C) were done, whereas 40 cycles of
amplification were done for myosin-Vb and myosin-Vc (1 min at 94°C, 1
min at 55°C, and 1 min at 72°C). In all cases, polymerase chain
reaction (PCR) was terminated by incubation for a further 10 min at
72°C. Amplification of ß-actin cDNA was done as control of mRNA
content. Human brain cDNA and HeLa cells cDNA were used as positive
controls and reference samples. The following forward (F) and reverse
(R) primers were used for amplification: F-agtctctgtgtcgttcattcg and
R-cctgtatgtaaactcacggta for myosin-Va; F-taactcatactgcctggaagc and
R-acttccttcttgctcacattg for myosin-Vb; F-agacagtgatgccaaggaga and
R-gagtgagagatgatgtggtc for myosin-Vc; and F-ggcatcgtgatggactccg and
R-gctggaaggtggacagcga for ß-actin. Equal volume (5 µl) of each PCR
product was loaded on a 1% agarose gel and electrophoresed in
TAE buffer (40 mM Tris-acetate, pH 8.0, 1 mM EDTA). Gels were
subjected to ethidium bromide staining and were imaged in a UV
transilluminator using a digital Kodak camera.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (14K): [in a new window] |
Figure 1. Flow cytometric analysis of fluorescence intensity for myosin-V
labeling in PBMC. Each of the data points plotted in this figure
represents the median value of fluorescence intensity from 10,000
lymphocytes or NK cells or 15,000 monocytes counted minus background
fluorescence of isotype control from each of the 19 healthy blood
donors. Fluorescence intensity was measured in arbitrary units
(channel-shifting) on a linear scale of 0–1023. Horizontal lines
represent the median value from the 19 data points. The different
symbols representing the data points correspond to the different cell
subsets indicated on the x-axis. The median of the
background fluorescence intensity for each cell subpopulation was 14.64
for lymphocytes (CD3, CD4, CD8, CD19) and NK cells (CD16, CD56) and
22.67 for monocytes. There is no statistically significant difference
between CD3+ lymphocytes and monocytes (P=0.8038), NK and
monocytes (P=0.0857), and CD19+ lymphocytes and monocytes
(P=0.5263). The median intensity for NK cells is higher when
compared with the median for all the lymphocyte subsets and monocytes
together (P=0.0251).
|
 View larger version (37K): [in a new window] |
Figure 2. Cellular distribution of myosin-V imaged by confocal microscopy in
PBMC. (A–C) 3-D Maximal projection images from multiple optical
sections. (A) CD markers; (B) myosin-V labeling; (C) superimposed
labeling of A and B images; (D) single optical section overlaid images
of myosin-V and CD-marker labeling. Note that myosin-V staining shows a
granular pattern throughout the cytoplasm with an accumulation in a
restricted perinuclear region. All cell subtypes are stained for
myosin-V in agreement with the flow cytometry analysis.
|
 View larger version (60K): [in a new window] |
Figure 3. Confocal microscopy analysis of the myosin-V colocalization with
microtubules (A); F-actin (B); endolisosomal marker lamp-1 (C); and the
8-kDa dynein light chain (DLC8; D). Myosin-V staining is shown in
green, and the staining for the other molecules is in red. Myosin-V was
detected with the medial-tail antibody, microtubules with a mAb to
ß-tubulin, and F-actin with rhodamine-conjugated phalloidin. Note
that a prominent myosin-V staining coincides with the centrosomal
region and the central microtubules array. Puncta of myosin-V staining
also colocalize with cortical actin-filament staining. Myosin-V
staining present in the cell center showed colocalization with lamp-1
and DLC8.
|
 View larger version (21K): [in a new window] |
Figure 4. Confocal microscopy analysis of the myosin-V colocalization with MHC
class II molecules. (A, B) Myosin-V staining; (C, D) MHC class II
staining; and (E, F) superimposed labeling. (A, C, E) Maximal
projection from multiple optical sections. (B, D, F) Single optical
sections. A striking centrosomal staining for myosin-V is particularly
noted in a single optical section through the cell (B). MHC class II
staining is concentrated on the cell surface (C, D) and is highly
colocalized with myosin-V punctate staining (E, F), suggesting that
myosin-V puncta are just subjacent to the plasma
membrane.
|
2.5 times
greater, by averaging the four dilutions) in samples from
PHA-stimulated cells when compared with the nonstimulated (NS) cells.
As an internal control of the relative amount of protein detected by
immunoblotting in resting or activated cells, we performed simultaneous
staining for -tubulin with a commercially available mAb. The
-tubulin staining was 1.5 times more intense in the stimulated
cells. Therefore, we observed a still greater increase in the myosin-V
staining compared with the -tubulin staining upon PHA stimulation.
Several cytoskeleton-associated proteins have been demonstrated to
increase upon T-lymphocyte activation [13
].
 View larger version (13K): [in a new window] |
Figure 5. Flow cytometric analysis of myosin-V fluorescence intensity in resting
and activated T lymphocytes. (A) Percentage of HLA-DR- and
CD69-positive T lymphocytes after 24 h in culture without
stimulation (NS) or with stimulation with OKT3 mAb. (B) Percentage of
HLA-DR- and CD69-positive T lymphocytes after 72 h in culture
without stimulation (NS) or with stimulation with PHA. The increase
observed in the number of T lymphcytes expressing CD69 and HLA-DR
antigens following OKT3 or PHA stimulation was statistically
significant (P=0.015 and P=0.002, respectively).
(C) Fluorescence intensity for myosin-Va labeling of CD3+
lymphocytes nonstimulated (NS) and stimulated with PHA or OKT3
antibody. Fluorescence intensity is expressed by the median
intensity—background subtracted—calculated from 10,000 cells analyzed
for each of the 17 blood samples for the PHA stimulation assay and 7
blood samples for the OKT3 stimulation assay. The symbols plotted
represent the data points for different groups as indicated on the
x-axis. Horizontal lines represent the median value from the
data points plotted. Myosin-Va labeling was done using
affinity-purified polyclonal antibody to the medial-tail domain of
chicken myosin-Va (the same antibody used for confocal microscopy and
Western blots). A significant increase is observed in myosin-Va
fluorescence intensity in T lymphocyte following PHA
(P=0.0002) or OKT3 antibody (P=0.0156)
stimulation.
|
 View larger version (47K): [in a new window] |
Figure 6. Immunoblot analysis of myosin-Va in mononuclear cells cultured in the
absence (NS) or presence of PHA stimulation. (A) SyproOrange-stained
gel containing total cell extracts from mononuclear cells cultured for
72 h in the absence (NS) or presence of PHA stimulus (PHA). The
gel was loaded with four dilutions of each cell extract on the basis of
cell number: 1 x 105 (lanes 1); 0.5 x
105 (lanes 2); 0.25 x 105 (lanes 4); and
0.125 x 105 (lanes 8) cells; molecular weight markers
(lanes M). (B and C) Immunoblots from a gel equivalent to the one shown
in A, probed with antibodies to the medial tail of myosin-Va (B) and
-tubulin (C). (D) Densitometric analysis for total protein,
myosin-Va, and -tubulin staining. The results are expressed as % of
OD relative to the sum of OD of the eight samples analyzed in each
case. Note that the increase observed for myosin-Va labeling following
T-lymphocyte activation is higher than the increase seen in total
protein and -tubulin labeling.
|
1.5 times in the case of donor 1 and 3–5
times in donor 2 in the PHA-stimulated cells as compared with
nonstimulated cells. These results were confirmed in additional and
independent RT-PCR assays from the same RNA samples (unpublished
results). Detection of PCR products for myosin-Vb and -Vc was possible
when amplification was done for 40 cycles (Fig. 7D
and 7F)
. It is
interesting that in these conditions, we detected a low amount of PCR
product for myosin-Vb (Fig. 7D)
and no product for myosin-Vc (Fig. 7F)
in the nonstimulated cells, whereas in PHA-stimulated cells, a great
increase in the amount of myosin-Vb PCR product was obtained (8
times, based on the densitometric analysis shown in Fig. 7E
). The same
result was obtained in additional RT-PCR experiments using the same RNA
samples from donor 1 as well as from donor 2 (unpublished results).
Myosin-Vc PCR product was only detected in the 2 µg RNA sample of the
PHA-activated cells (Fig. 7C)
. Therefore, these results indicate that
myosin-Va mRNA is by far the most abundant myosin-V mRNA in mononuclear
cells and reveal an increase in the amount of mRNA for the three
members of class V myosins following activation in T lymphocytes.
 View larger version (71K): [in a new window] |
Figure 7. RT-PCR from myosin-Va, myosin-Vb, and myosin-Vc mRNA in mononuclear
cells cultured in the absence (NS) or presence of PHA stimulation. (A,
D, F) Ethidium bromide-stained gels showing the RT-PCR products for
myosin-Va and ß-actin (A), myosin-Vb (D), and myosin-Vc (F),
amplified from total RNA extracted from nonstimulated (NS) and
PHA-stimulated mononuclear cells (PBMCs) from donors 1 and 2 for
myosin-Va, from donor 1 for myosin-Vb and –Vc, and from human brain
tissue and HeLa cells as indicated. The numbers below the lanes
indicate the amount (µg) of total RNA used in the
reverse-transcription reactions. PCR was done for 34 cycles for
myosin-Va and ß-actin and 40 cycles for myosin-Vb and myosin-Vc.
"T-"indicates PCR in the absence of template. (B, C, E)
Amplification ratio, calculated from the OD of the ethidium
bromide-stained PCR products for myosin-Va from donor 1 (B), donor 2
(C), and myosin-Vb (E), normalized to the amount of ß-actin PCR
product detected in gel. Amplification ratio for myosin-Va, donor 1,
was determined only based on the products in lanes 0.04 (NS, PHA),
which did not show saturation of the PCR amplification, and for
myosin-Vb, was determined based on the products in lanes 2 (NS and
PHA), because no product was detected in lane 0.2, NS. The ratios shown
here were confirmed in additional, independent experiments, using the
same RNA preparations.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
190 k on Western blots containing crude extracts
from melanocytes, whole brain, and leukocytes (unpublished results).
Also, we have determined by immunohistochemistry (ongoing studies) that
the medial-tail antibody exhibits a staining pattern similar to the one
obtained for the previously characterized 32a antibody
[11
], which has been used extensively in studies from
several laboratories. In addition, these antibodies exhibit a different
pattern of cellular and tissue distribution than that shown for an
antibody to myosin-Vb produced in our laboratory, which does not
cross-react with myosin-Va. Therefore, this implies that the staining
shown in leukocytes in the present work is not a result of recognition
of myosin-Vb. In fact, probing of the Western blot shown in Figure 6 with the antibody to myosin-Vb gave a much fainter signal than the one
detected with the antimedial tail (unpublished results). In addition,
the mRNA expression analyses shown here (Fig. 7)
also support the
supposition that the myosin-V staining presented here is specific to
myosin-Va, because myosin-Vb and -Vc appear to be expressed at much
lower levels than myosin-Va in blood mononuclear cells. In the present work, we demonstrate that the major cell subtypes of the PBMC, including cytotoxic and helper T lymphocytes, B lymphocytes, NK cells, and monocytes, do express myosin-V. The expression levels of myosin-V, as measured from the fluorescence intensity by flow cytometry, were shown to be basically equivalent among these cell types (Fig. 1) . A slightly higher fluorescence intensity median was found for NK cells but also a greater dispersion of the data points than that seen for the other cell types. This may be suggestive of the existence of a subset of NK cells that express higher levels of myosin-V (Fig. 1) .
A striking feature of the myosin-V staining in all mononuclear cells analyzed was the prominent staining at the centrosomal region (Figs. 2 and 3) . Espreafico et al. [27 ], Wu et al. [28 ], and Tsakraklides et al. [29 ] have previously shown the presence of myosin-V in the centrosome of many cell lines. Immune responses require the participation of the centrosome, especially in the cell polarization event that takes place in NK cells and T lymphocytes. Ménasché et al. [5 ] have shown that mutations in the RAB27a gene but not in the MYO5a gene lead to deficient degranulation of cytotoxic T lymphocytes that might be responsible for the immunodeficiency in GS2 patients. Cell polarization is decisive for an efficient cytotoxicity of target cells. This is a particularly interesting aspect to investigate given the recent demonstration that the myosin-V in Saccharomyces cerevisiae (Myo2p) is required for spindle microtubule orientation [16 ]. An explanation for the lack of cytotoxicity defect in T lymphocytes of GS1 patients is that additional paralogs of myosin-V could be substituting for the lack of myosin-Va. Some redundancy of function has already been pointed out between myosin-Va and -Vb in neuronal cells [30 ].
The punctate pattern of fluorescence staining for myosin-V, observed in all mononuclear cells, is suggestive of association with large protein complexes or membranous vesicles and organelles. The colocalization of myosin-V with the dynein light-chain DLC8 at the centrosomal region does not indicate by itself a direct association of these proteins, because the microtubule-based motor complex, dynein, is also enriched at the centrosome. It is an intriguing question as to why this light chain is shared by the two motor systems [31 32 33 ]. The centrosomal staining also markedly colocalizes with the endolysosomal marker lamp-1 (Fig. 3) in a region where endolysosomal vesicles of the late recycling compartments are known to be located. MHC class II are antigen-presenting, surface molecules that are transported preferentially and recycled via endolysosomal vesicles. We show here a striking colocalization of these molecules with myosin-V in the cell periphery (Fig. 4) . These data suggest that myosin-V could be involved in the pathway of MHC class II expression or recycling. To start addressing this hypothesis, we asked whether activation of T lymphocytes, which induces MHC class II expression, would be associated with changes in the expression of myosin-V. It is interesting that we found that a significant increase in the amount of myosin-V protein occurs following PHA or OKT3 activation of T lymphocytes (Figs. 5 and 6) . The variation in the myosin-V fluorescence intensity, which was higher in the stimulated lymphocytes than in resting cells among the 17 blood samples analyzed, might reflect differences in the activation response of T cells from different healthy individuals (Fig. 5) . Pastural et al. [4 ] has recently shown evidence from immunoprecipitation assays that the amount of myosin-V protein is augmented in human cells with mutation in the RAB27a gene. In light of the present results, it will be interesting to investigate in future studies a possible association of myosin-V with the uncontrolled cellular activation and proliferation that characterizes the immune defects in GS patients.
The possibility of functional redundancy among the class V myosin members and the demonstration that activation of T lymphocytes leads to an increase in the amount of myosin-V protein prompted us to extend our expression analysis to the level of mRNA and to include the three members of the class V myosins (a, b, and c). Our results indicate that mRNA for the three members of class V myosins is expressed in the PBMC and that myosin-Va mRNA is the most abundantly expressed, followed by myosin-Vb and -Vc, respectively (Fig. 7) . In addition, this analysis suggests that there is an increase in the amount of myosin-Va mRNA following PHA stimulation, which was shown to vary from 1.5 times in one donor to 3–5 times in another donor. This variation is compatible with the variation in myosin-V fluorescence intensity detected in stimulated lymphocytes among different individuals (Fig. 5) . It is interesting that despite the low levels of myosin-Vb and -Vc mRNA in PBMC, a pronounced increase was observed following mitogen activation, especially for myosin-Vb mRNA. Also at the level of protein expression, we observed an increase of the amount of myosin-Vb by probing the Western blot shown in Figure 6 with the antibody to myosin-Vb (unpublished results). It has recently been shown that myosin-Vb plays an important role in the expression of the IgA receptor to the plasma membrane in polarized and nonpolarized epithelial cells [34 ]. It is possible that myosin-Vb might also be required in similar processes, which are intense during cellular activation, in the immune response.
The myosin-V localization analysis shown here also indicated that many of the cytoplasmic particles stained for myosin-V in many of the cells analyzed on a given coverslip (Figs. 3 and 4 and unpublished results) do not colocalize with any of the markers used. It will be interesting to determine whether this important pool of myosin-V corresponds to lytic granules in cytotoxic T lymphocytes. The myosin-V staining that we show here is strikingly similar to the staining described by Haddad et al. [35 ] for granzyme B and Rab27a. Recently, it has been demonstrated that Rab27a is required to recruit myosin-Va to the melanosome surface, thereby allowing melanosomes to be anchored at the cell periphery [36 37 38 ]. Conversely, lymphocytes from dilute mice [35 ], as shown for GS1 lymphocytes, do not exhibit any defect in degranulation and cell cytotoxicity. It is remarkable that the RAB27a mutation leads to accumulation of lytic granules in cytotoxic T lymphocytes that fail to pass through the actin cytoskeleton and dock at the plasma membrane of the immunological synapse [39 ]. Together, these findings have led Stinchcombe et al. [39 ] to propose the hypothesis that class V myosins might be acting redundantly in lymphocytes, or residual amounts of functional myosin-Va would be present in the dilute or GS1 cells that have been analyzed. The pattern of localization of myosin-V shown here strengthens the idea that this myosin may play key roles in the motility/docking of vesicles at the plasma membrane and perhaps in the polarization toward the immunological synapse. In addition, the evidence that the expression of the three members of class V myosin is associated with the functional status of T lymphocytes supports the hypothesis that these myosins play relevant roles in the immune response, which might overlap to some extent, thus explaining the lack of immune defects associated with mutations in the MYO5a gene.
Received April 1, 2001; revised October 4, 2001; accepted October 5, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. M. Cebulla, D. M. Miller, Y. Zhang, B. M. Rahill, P. Zimmerman, J. M. Robinson, and D. D. Sedmak Human Cytomegalovirus Disrupts Constitutive MHC Class II Expression J. Immunol., July 1, 2002; 169(1): 167 - 176. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
