Hindawi

Stem Cells International

Volume 2021, Article ID 5583421, 14 pages
https://doi.org/10.1155/2021/5583421

Research Article
Comparison of the Migration Potential through Microperforated
Membranes of CD146+ GMSC Population versus Heterogeneous
GMSC Population

1,2
Mohamed Al Bahrawy
1
Stony Brook University, NY, USA
2
Oral Medicine and Periodontology Department, Faculty of Dentistry-Ain Shams University, Cairo, Egypt

Correspondence should be addressed to Mohamed Al Bahrawy; bahrawy@asfd.asu.edu.eg

Received 25 June 2020; Accepted 31 August 2020; Published 12 March 2021

Academic Editor: Sangho Roh

Copyright © 2021 Mohamed Al Bahrawy. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.

Background. Guided tissue regeneration (GTR) is a powerful modality for periodontal regeneration, but it blocks the periosteum
and gingival stem cells (GMSCs), from supporting periodontal wound by the nutrients, growth factors, and regenerative cells.
The microperforated membrane considered a rewarding solution for this major drawback; GMSCs can migrate through a GTR
microperforated membrane toward a chemoattractant, with the blocking of other unfavorable epithelial cells and fibroblasts. In
the absence of a sole marker for MSC, a homogeneous population of GMSC is difficult to isolate; using CD146 as confirmatory
markers for MSC identification, testing the behaviour of such homogeneous population in migration dynamics was the question
to answer in this study. Materials and Methods. GMSCs from healthy crown lengthening tissue was isolated (n = 3), its stem cell
nature was confirmed, CD146 and CD271 markers were confirmatory markers to confirm homogenous stem cell population,
and magnetic sorting was used to isolate GMSC with CD146 markers. A homogenous CD146 population was compared to
heterogeneous GMSCs of origin; the population doubling time and MTT test of the two populations were compared. Migration
dynamics were examined in a transwell migration chamber through 8 μm perforated polycarbonic acid membrane, and 0.4 μm
and 3 μm perforated collagen-coated polytetrafluoroethylene membrane (PTFE) and 10% fetal bovine serum (FBS) were the
chemoattractants used in the lower compartment to induce cell migration, were incubated in a humidified environment for 24
hours, then migrated the cell in the lower compartment examined by a light and electron microscope. Results. GMSCs fulfilled
all the minimal criteria of stem cells and showed low signal 10% for CD146 on average and extremely low signal 2% for CD271
on average. Magnetic sorting optimized the signal of CD146 marker to 55%. GMSC CD146 population showed nonstatistically
significant shorter population doubling time. CD146 homogeneous population migrated cell numbers were statistically
significant compared to the heterogeneous population, through 0.4 μm and 3 μm perforated collagen membrane and 8 μm
perforated polycarbonate membrane. Scanning electron microscopy proved the migration of the cells. Conclusions. A subset of
the isolated GMSC showed a CD146 marker, which is considered a dependable confirmatory marker for the stem cells. In terms
of GMSC migration through the microperforated membrane, a homogeneous CD146 population migrates more statistically
significant than a heterogeneous GMSC population.

1. Introduction system that responds by continuous inflammatory cytokine

shower that affects the body homeostasis; with time,
Periodontitis is a chronic inflammatory bacterial infection, deregulated immune response eventually results, and a
where the oral flora organizes a biofilm subgingivally, which hyperresponsive immune reaction causes the destruction of
constitutes a continuous challenge to the host immune the tooth-supporting apparatus, leading to tooth loss.
2 Stem Cells International

Periodontitis is considered an irreversible degenerative dis- which was the reason for the improved clinical outcomes [7–
ease of the odontogenic supporting tissue; this throws light 9]. In 2018, Al Bahrawy et al.’s in vitro study concluded that
on the immune-mediated nature of periodontal disease [1]. macroperforation jeopardizes the GTR membrane occlusive
A historical debate did exist about the periosteum’s role in function and its mechanical properties; this study postulated
bone growth, repair, and regeneration. Two theories have been a new development of Gamal and Iacono’s concept by a
contrasting; one postulated that periosteum is an inert mem- microperforated membrane, a concept was proved, GMSCs
brane covering the bone with no exact role [2]; the other con- can migrate through the microperforation under chemoat-
sidered the periosteum as a functioning membrane with tractant influence, and the membrane was occlusive for cell
osteogenic potentials, responsible for regeneration and bone migration in the absence of a chemoattractant; hence, the
growth. A number of classical experiments created strong evi- GTR membrane could be a selective occlusive barrier allow-
dence that leads to modern literature where the essential role ing the migration of the desired cells while blocking others,
of periosteum for bone healing was understood. A classical under the influence of the right chemoattractant [4].
study when the periosteum surrounding fractures removed A cell to be considered stem cell must be multipotent, can
the result was the absence of callus in the fracture [3]. differentiate to other cell types than the tissue of origin, must
Guided tissue regeneration technique (GTR) was based be clonogenic, and has strong proliferation power. In fact,
on blocking the growth of unfavorable cells from invading only a fraction of the plastic adherent cells showed these
the periodontal wound, namely, the gingival epithelium and characteristics; this is explained by the heterogeneous popu-
connective tissue, but as collateral damage to this technique, lation of the isolated cells and attributed to the nature of tis-
were scalding the alveolar bone from its periosteum by ele- sue of origin [10]. Another issue to consider is the absence of
vating a full-thickness flap. Blocking the periodontal wound one specific marker that can identify stem cells from other
by a barrier membrane from the periosteum in fact excludes mature cells; many surface markers, for example, CD73,
the wound area from a powerful regenerative source which is CD90, CD105, CD146, or even neural crest markers; and
an essential source of blood and nutrient supply; besides that, intracytoplasmic markers like STRO-1, OCT-4, Nanog,
the periosteum is a niche of biologic mediators and progeni- Nestin, and Notch-1 that could be used to characterize stem
tor cells essential for the regeneration process [4]. cells [11, 12], taking into consideration that different stem
Not only periosteum but GTR also deprive the periodon- cells from different tissue origins show a different set of
tal wound from the gingival connective tissue, to block the markers, but as a minimal criterion, cells must show a high
rapidly proliferating fibroblasts which can invade the peri- signal of CD73, CD90, and CD105 together. The main draw-
odontal wound before the slowly proliferating periodontal back with these three markers was that they are expressed by
ligament and bone cells, but regrettably, a well-recognized fibroblast, although in weak signal [13]; besides, fibroblast
population of stem cells named the gingival mesenchymal morphology was identical to MSCs; both did plastic adher-
stem cells (GMSC) is blocked from the wound [5, 6] if ence and fibroblast dipotency, can differentiate to at least
GMSCs allowed migrating to the periodontal defect and two other cell lines, and made identification of fibroblast
induced to differentiate; using the suitable biological factors from MSC in vitro not an easy task; this urged the need for
into periodontal ligament cells and osteoblasts with a well- at least additional cell marker.
designed organized scaffold, biological factor release cascade, It was well described that MSC is located around blood
such as a system, would satisfy the real aim of GTR. Thus, the vessels; in 2008, Covas and colleagues compared MSC from
occlusive barrier membrane of the classical GTR is unfavor- different tissue origins to fibroblasts and pericytes, and they
able for periodontal regeneration [4]. concluded that 12 MSC populations were very similar to 4
In comparison to bone marrow stem cells, the gold fibroblasts and 2 pericyte populations phenotypically; the
standard, the first stem cell described, and the most stud- only difference was fibroblast weak signal of the CD146 sur-
ied, GMSC was superior in nearly every aspect, besides its face marker compared to MSC and pericytes; both showed a
ease of harvesting with very low morbidity; no scaring; strong signal of this marker in flow cytometry. Comparing
homogeneous population; high proliferation rate without the 3 cells genetically, they concluded that the gene expres-
the need for special growth factors; morphology stable sion pattern of MSC is similar to pericytes and stellate hepatic
within successive passages; reduced senility; and stable kar- cells, not fibroblast which showed the gene expression
yotype, maintains its telomerase activity to later passages, pattern of myofibroblasts and smooth muscle cells [14].
and shows very low tumorigenic potential [5]; this con- Another evidence of the pericyte and MSC similarity was
cluded that the gingiva is a very good source of stem cells proved in other studies, where the 3 minimal markers of
compared to the bone marrow, with functionally MSCs CD73, CD90, and CD105 have a vascular and perivas-
competent MSCs, that can be used with a wide range of cular distribution pattern in vivo [15, 16]; to confirm this
medical applications. assumption, other MSC-specific markers were examined,
Gamal and Iacono’s clinical study tested macroperfo- namely, CD146, NG2, Stro-1, and 3G5, which confirmed
rated GTR and posted improved clinical outcome, followed the previous results of the vascular and perivascular distribu-
by a series of studies in 2014 and 2016; they hypothesized tions of these markers [15, 17]. Connecting all of this data
that GTR membrane perforation allowed bone morphogenic together, we can conclude the close nature of MSC and
protein (BMP-2) and platelet-derived growth factor (PDGF- pericytes in vivo in contrast to fibroblast; this data built
BB), besides vascular endothelial cell growth factor (VEGF) strong evidence that CD146 which is essentially a pericyte
and other nutrients migrating freely through the membrane, marker would be a good candidate to confirm MSC; besides
Stem Cells International 3

(a) (b)

Figure 1: (a) Healthy gingival tissue specimen of discarded crown lengthening procedures. (b) Gingival connective tissue was meshed to
1 mm pieces using a surgical blade.

the other three fundamental markers, a cell population Louis, USA) and then in Collagenase IV (Fisher Scientific,
expressing them all with a high signal is a homogeneous Massachusetts, USA) for 40 minutes at 4°C; the resulted cell
MSC population. suspension was strained in 40 μm strain to remove the impu-
From all of what was mentioned, we conclude that rities, then centrifuged at 1200 rpm for 8 minutes. The
depriving the wound area from GMSC with its multipotent resultant single-cell suspension was inoculated in 10 cm cell
abilities was not a good idea because it is an important source culture dish, in alpha-minimal essential medium (alpha-
of regeneration. It has undenied the breakthrough the GTR MEM 1×, Gibco, Thermo Fisher Scientific, Massachusetts,
technique had achieved in the periodontal treatment in gen- USA), supplemented with 10% fetal bovine serum (FBS)
eral and in the regeneration concept, in particular, but it is (Hyclone, Thermo Fisher scientific, Massachusetts, USA)
now clear the GTR by its traditional occlusive membrane is and 50 U/ml penicillin G with 50 μg/ml streptomycin and
not the best practice for the regeneration procedure, and 2.5 μg/ml amphotericin B (Fungizone, Thermo Fisher
microperforation of this membrane is essential; besides, uti- scientific, Massachusetts, USA), at a concentration of 60
lizing a full system of specific chemoattractant in the peri- cells/cm2.The single-cell suspension plates were incubated
odontal wound side of the membrane for GMSCs will let in 37°C, 5% CO2 humidified incubators. Cells reached
this powerful cells invade the wound area, to achieve the confluence after approximately 28 days for the first passage,
optimum outcome of GTR technique, with organized chro- then subcultured in a P100 plate for the next passages, and
nologically activated cell differentiation induction biological reached confluence on average in 14 days.
factors.
2.2. Colony-Forming Unit. At passage five, cultured cells were
2. Materials and Methods detached using 0.05% trypsin/EDTA, cells were diluted in
alpha-MEM (Gibco, Thermo Fisher Scientific, Massachu-
2.1. Sample Selection. Gingival specimens are healthy gingival setts, USA) enriched with 10% FBS (Hyclone, Fisher Scien-
tissue of discarded crown lengthening procedures of outpa- tific, Massachusetts, USA) at a concentration of 103 cell/ml
tients who attended at Stony Brook University dental care in P10 dishes, and media were changed every 3 days and
clinics Figure 1(a). A parallel case-control experimental study examined under a microscope till typical fibroblast colonies
of two groups was designed. Four subjects accepted to partic- of 100 cells formed.
ipate in this study, all experiments were done in triplicate
(n = 3), and participants were informed about the nature of 2.3. Population Doubling Assay. Menicanin et al.’s protocol
the experiment and verbally accepted the use of their dis- was followed; briefly, GMSCs were seeded in 24-well plate
carded tissue in stem cell research. The ethical committee with a concentration of 5 × 103 cells/cm2; when 90% conflu-
of scientific research at School of Dentistry Ain Shams ence was reached, cells were detached using 0.05% trypsi-
University and Stony Brook University had approved this n/EDTA and then counted, cells were diluted and reseeded
study (IRB 575741). with the same concentration in another 24-well plate, and
The gingival epithelium was carefully scalded from the the same procedure was repeated for five passages. Cells were
specimen; the connective tissue was meshed to very small counted in each passage, and population doubling was calcu-
pieces using the surgical lancet Figure 1(b) then digested in lated using this formula: log2 final cell number/log2 seeding
2 mg/ml Dispase II overnight at 4°C (Sigma-Aldrich, St. cell number [18].
4 Stem Cells International

2.4. Flow Cytometry Assay. At the 5th passage, cell culture

was washed twice by PBS, then detached using 0.05% trypsi-
n/EDTA; detached cells were resuspended in 1% bovine
serum albumin as a blocking buffer for half an hour. Cells
were aliquoted with a concentration of 1 × 105 cells in two
test tubes, then 2 μg/ml of CD73 and its isotype control fluo-
rescein isothiocyanate- (FITC-) conjugated mouse monoclo-
nal antibodies in each tube, and then incubated for 30
minutes in 4°C (BD Pharmingen, San Jose, California, United
States). The same procedure was done with CD90 and its iso-
type using APC-conjugated mouse monoclonal antibodies
(BD Pharmingen, San Jose, California, United States), for
CD105 and its isotype using Alexa 555 gout anti-mouse
monoclonal antibodies (Dako, Agilent, Santa Clara, USA)
and finally, for CD146 and CD271 and their isotype PE-
conjugated mouse monoclonal antibodies(BD Pharmingen,
San Jose, California, United States). After incubation buffer
was aspirated, cells were washed twice by resuspension in
PBS and centrifugation at 1200 rpm for 8 minutes, and cells
were then transferred to a flow cytometry facility for the anal-
ysis of stem cell marker expression. Regarding the hemato-
poietic markers, namely, CD14, CD34, and CD45, the same
protocol was followed with no difference.
Figure 2: Ready-made osteogenic, adipogenic, and chondrogenic
2.5. In Vitro Differentiation Assay stem cell differentiation media (Gibco StemPro, Thermo Fisher
Scientific, Massachusetts, USA).
2.5.1. Osteogenic Differentiation. Cell suspension at passage 4
was seeded in six-well plates with a concentration of 8 × 103
washed twice by distilled water, then stained in dark with
cells/cm2 in a ready-made osteogenic induction medium
Alcian blue (10 mg in 60 ml ethanol with 40 ml acetic acid)
(Gibco StemPro, Thermo Fisher Scientific, Massachusetts,
overnight; the next day, the cell culture was destained by
USA), according to Gronthos et al.’s protocol; the medium
120 ml ethanol with 80 ml acetic acid for 20 minutes, finally
was changed every 3 days for 28 days in humidified incuba-
washed 2 times by PBS, and then checked under a micro-
tors ([19, 20]), the cell cultures were washed twice by PBS,
scope (Figure 2).
fixed by 4% paraformaldehyde for 1 hour, washed again twice
by distilled water, finally stained by 2% Alizarin Red for 45 2.5.4. MTT Assay. Detached cell culture of the fourth passage
minutes, finally washed 4 times by distilled water and 2 times was suspended in 500 μl alpha-MEM (Gibco, Thermo Fisher
by PBS, and checked under a microscope (Figure 2). Scientific, Massachusetts, USA) enriched with 10% FBS
(Hyclone, Fisher Scientific, Massachusetts, USA), poured in
2.5.2. Adipogenic Differentiation. Cell suspension at passage 4
spectrophotometer tube, and left in a humidified incubator
was seeded in six-well plates with a concentration of 8 × 103
(37°C, 5% CO2); the negative control was a medium-
cells/cm2 in a ready-made adipogenic induction medium
enriched tube without cells, within the same incubator. The
(Gibco StemPro, Thermo Fisher Scientific, Massachusetts,
next day, 100 μl of MTT stain was added, and tubes were
USA) according to Pittenger et al.’s protocol; the medium
incubated for another 4 hours; then, media were aspirated
was changed 2 times per week for 28 days [21]. After that
and 1000 μl of dimethyl sulfoxide (DMSO) was added, and
time, the cell cultures were washed twice with PBS, fixed by
tubes were analysed by a spectrophotometer at a 595 nm
4% paraformaldehyde for 1 hour, and washed again twice
wavelength.
by distilled water; the cell culture is washed by 60% isopropa-
nol for 5 minutes, then stained for 5 minutes by Oil Red O in 2.6. Cell Sorting
isopropanol (300 mg Oil Red in 100 ml isopropanol), washed
by tap water, and finally stained by hematoxylin for 1 minute, 2.6.1. Flow Cytometry Cell Sorting. The same protocol of cell
again washed by tap water, and checked under a phase- characterization was followed; the only difference was not to
contrast microscope (Figure 2). fix the cells; it has to be noted that the sorting procedure was
done as soon as possible after cell detachment, as the cells
2.5.3. Chondrogenic Differentiation. Cell suspension at pas- were suspended in serum-free media; finally, after sorting
sage 4 was seeded in six-well plates with a concentration of by the machine, cells were collected in media enriched with
8 × 103 cells/cm2 in a ready-made chondrogenic induction 20% FBS.
medium (Gibco StemPro Thermo Fisher Scientific, Massa-
chusetts, USA); the medium was changed 2 times per week 2.6.2. Magnetic Sorting. After detaching, cells were counted
for 28 days. After that time, the cell cultures were washed and suspended in 1 ml of buffer of the cell sorting kit (MACS
twice by PBS, fixed by 4% paraformaldehyde for 1 hour, cell separation, Miltenyi Biotech, USA). Cells were
Stem Cells International 5

Specimen_001-146 pos
Specimen_001-146 pos
105
105

104
FITC-A 104

FITC-A
P1
103 103

102
102
0
0
–55
–55 –40 0 102 103 104 105
–40 0 102 103 104 105 PE-A
PE-A
Specimen_001-146 pos
350
300 P2
250
Count

200
P3
150
100
50
0
101 101 103 104 105
PE-A
(a)
Specimen_001-146 pos iso
Specimen_001-146 pos iso
105
105

104
104
FITC-A

FITC-A

P1
103 103

102
102
0
0
–55
–55 –40 0 102 103 104 105
–40 0 102 103 104 105 PE-A
PE-A
Specimen_001-146 pos iso
100
P2
75
Count

50 P3

25

0
101 102 103 104 105
PE-A
(b)

Figure 3: Continued.
6 Stem Cells International

(c)

Figure 3: (a) The confirmatory flow cytometry graph of the magnetic sorted GMSC homogeneous CD146-positive population, which
optimized the signal to 55% purity. (b) The flow cytometry graphs of the negative control isotype, which showed no signal of the CD146
marker. (c) The magnetic sorting kit.

centrifuged at 300 × g for 10 minutes, the buffer was suc- The next day, the migration inserts were collected from
tioned, 20 μl Fc block was added, and 20 μl CD146 marker the plates, and media in the upper compartment aspirated.
was labelled with microbeads. Tubes were incubated in 4°C The inserts were washed two times in PBS. Using a cotton
for 14 minutes, washed in 1 ml buffer, and centrifuged at swab, the upper side of the membrane was scraped thor-
300 × g for 10 minutes. Cells were suspended in 500 μl oughly to remove all the cells still attached to the upper com-
MAC buffer solution. The magnetic sorting column was partment. Cells on the lower side of the membrane were fixed
primed by 500 μl MAC buffer solution, the cell suspension by 4% paraformaldehyde for 2 minutes; inserts were washed
was added in the sorting column, and the column was 2 times in PBS; cells were then permeabilized by 100% meth-
washed three times using 500 μl MAC buffer solution in each. anol and stained by crystal violet (1% in 80% ethyl alcohol,
Finally, we plunged out the cells from the sorting column Sigma Aldrich) and washed again two times in PBS; mem-
using serum-free alpha-MEM Figure 3. brane was examined under the microscope at 40x magnifica-
tion; migrated cells were counted in 5 different areas; and the
average was counted.
2.7. Migration Assay
2.7.2. Scanning Electron Microscope. Membranes were cut off
2.7.1. Microscopic Perforated Membranes. In the transwell the migration inserts, fixed by 4% paraformaldehyde for 2
chemotaxis migration chamber (Boyden chamber) was the minutes, and left to dry. Membranes were soaked in 50%,
test used to analyse the migration dynamics of GMSCs 70%, 80%, 90%, and 100% ethyl alcohol for 10 minutes each,
(Corning Life Sciences, New York, USA);2 types were used, finally frozen in minus 80°C overnight, then sent to an elec-
namely, 12 mm perforated collagen-coated polytetrafluor- tron microscope facility, were coated by gold, and examined
oethylene (PTFE) membrane with a pore size of 0.4 μm and by the electron microscope.
3 μm pores and a 6.5 mm perforated polycarbonate mem-
brane with a pore size of 8 μm. Cultured heterogeneous 2.7.3. Statistical Evaluation. All statistical analyses were done
GMSCs were the positive control group, and the homoge- using SPSS v20 program, IBM; the descriptive analysis was
neous CD146-positive sorted and expanded GMSCs were used to determine the distribution of the data, and graphs
the experiment group. Cells were detached using 0.05% tryp- were plotted; according to it, the Mann–Whitney U test
sin/EDTA, then suspended in serum-free alpha-MEM was the test of choice according to the data distribution, the
diluted to 1 × 104 concentration and added to the upper com- alpha significance of difference was set at p < 0:05, and all
partment of the chemotaxis chamber inserts. For both experiments were done in triplicate.
groups, the lower compartment of the chemotaxis chamber
received alpha-MEM with 10% fetal bovine serum as a che- 3. Results
moattractant (Hyclone, Fisher Scientific, Massachusetts,
USA), The migration chamber plates were incubated for 24 3.1. Colony-Forming Potential. Seeded GMSCs in P10 dish
hours in a humidified atmosphere (37°C, 5% CO2). showed typical distinctive fibroblast-like colonies of 50 to
Stem Cells International 7

(a) (b)

(c) (d)

Figure 4: (a) Representative image of CFU experiment showing stem cell colony-forming potential. (b) Representative image showing the cell
population doubling potential. (c) Representative image for the MTT essay showing black deposits in the experiment tube compared to the
control group. (d) Representative image showing cell differentiation potential; calcium deposition (upper), cartilage glycoprotein deposition
(middle), and fat droplet deposition (lower).

100 cells/colony after on average 14 days of culturing in vitro; 3.2. Population Doubling Potential. The two groups were
all experiments were done in triplicate (n = 3) for each group; similar in showing strong proliferation capability. The popu-
no significant difference was noted between the heteroge- lation doubling time was nonsignificantly less in the CD146-
neous GMSC group and CD146-positive homogeneous positive homogeneous group compared to the heterogeneous
GMSCs in shape, form, or number of cells in colonies group (p > 0:05; Mann–Whitney U test) (Figure 4(b)).
(p > 0:05; Mann–Whitney U test); the only difference noted
was that the homogeneous CD146-positive group reached 3.3. Cell Characterization. Both groups lacked the expression
100 cell colonies 1 day earlier on average compared to the of hematopoietic markers, namely, CD14, CD34, and CD45,
heterogeneous group (Figure 4(a)). and both groups could express the main MSC markers,
8 Stem Cells International

(a)

Figure 5: Continued.
Stem Cells International 9

(b)

Figure 5: (a) The CD271 flow cytometry graphs of 3 cell lines of the heterogeneous GMSC population, signal percentage of cell line A: 2%
(upper), cell line B: 1% (middle), and cell line C: 4% (lower). (b) The CD146 flow cytometry graphs of 3 cell lines of the heterogeneous GMSC
population, signal percentage of cell line A: 10% (upper), cell line B: 11% (middle), and cell line C: 17% (lower).
10 Stem Cells International

(a) (b)

Figure 6: (a) Migrated CD146-positive homogeneous GMSC in the lower compartment of 8 μm perforated polycarbonate membrane toward
fetal bovine serum as a chemoattractant; cells stained with crystal violet; 10,000 cells seeded in the upper compartment. (b) Migrated
heterogeneous GMSC in the lower compartment of 8 μm perforated polycarbonate membrane toward fetal bovine serum as a
chemoattractant; cells stained with crystal violet; 10,000 cells seeded in the upper compartment (40x magnification).

namely, CD73, CD90, and CD105 with a strong signal for the 3.7. Multilineage Differentiation Potential. GMSCs of both
three; another 2 markers tested the CD146 which showed a groups cultured in osteogenic induction media for 28 days
weak signal of 11% on average Figure 5(b) and CD271 which showed osteogenic differentiation capacity, which was
showed a very weak signal of 2% on an average Figure 5(a); proved by calcification stained by Alizarin Red stain. Cells
this was another reason to choose the CD146 as a confirma- of both groups cultured in chondrogenic induction media
tory marker for the gingival connective tissue stem cells, for 28 days showed chondrogenic differentiation capacity
where CD271-positive GMSCs were very rare in the gingival proved by cartilage glycoprotein deposits stained by Alcian
isolated stem cells. blue stain. Finally, cells of both groups cultured in adipogenic
induction media for 28 days showed lipid deposits proved by
3.4. Flow Cytometry Cell Sorting. Using flow cytometry cell lipid droplets stained by Oil Red stain. Control GMSCs
sorting module, the cells expressing the CD146 marker were cultured in alpha-MEM media with 10% FBS for 28 days
isolated successfully, and the isolated cells were attached to did not stain any deposits with three mentioned stains
plastic and started to show a fibroblast-like shape the next (Figure 4(d)).
day. Unfortunately, after 3 days, all the isolated cell dishes
showed bacterial contamination, in all the plates; the experi- 3.8. Transwell Migration Potential
ment was repeated three times with the same tragedy; this
3.8.1. Migration through Polycarbonate Membrane 8 μm Pore
forced us to resort to magnetic sorting.
Size. A significantly higher number of CD146-positive
homogeneous GMSCs migrated through the membrane
3.5. Magnetic Cell Sorting. Using a magnetic sorting method compared to the heterogeneous GMSC population toward
for cell isolation, CD146 homogeneous cell population was the 10% FBS chemoattractant, (the z-score is 2.50672. The
isolated successfully, cells showed plastic adherence after a p value is 0.01208. The result is significant at p < 0:05)
longer time than expected, and cells did not show (Figures 6(a) and 6(b), Table 1).
fibroblast-like morphology except after two to three days on
average; an explanation might be that the magnetic sorting 3.8.2. Migration through Collagen-Coated PTFE Membrane
antibodies hinder the MSC attachment to the plastic; after 0.4 and 3 μm Pore Size. A significantly higher number of
that, cells behaved normally and showed a colony-forming CD146-positive homogeneous GMSCs migrated through
unit after 12 to 13 days on average; another confirmatory the 0.4 (the z-score is 2.08893. The p value is 0.03662. The
flow cytometry assay was done to ensure the homogeneity result is significant at p < 0:05) and 3 μm pores (the z-score
of the CD146 cell population, which showed 55% signal is 2.50672. The p value is 0.01208. The result is significant
(Figures 3(a)–3(c). at p < 0:05) compared to the heterogeneous GMSCs toward
10% FBS in the lower compartment of the transwell migra-
3.6. Cell Metabolic Activity. This MTT test examined the met- tion chamber as a chemoattractant (Table 1); to be noted,
abolic activity of the cultured cells and hence its vitality. Both the migration of cells of both groups through 0.4 μm was
the experiment and the heterogeneous cell groups showed no nonsignificantly less than the migration through 3 μm pores;
significant difference between them regarding its metabolic both were statistically significantly less than migrated cells
activity (p > 0:05; Mann–Whitney U test) (Figure 4(c)). through the 8 μm pores.
Stem Cells International 11

Table 1: The average number of migrated cells counted in five, random fields at 40x light microscope magnification. For 8 microns, the z
-score was 2.50672. The p value was 0.01208. The result was significant at p < 0:05. For 3 microns, the z-score was 2.50672. The p value
was 0.01208. The result was significant at p < 0:05. For 0.4 microns, the z-score was 2.08893. The p value was 0.03662. The result was
significant at p < 0:05.

8 microns 3 microns 0.4 microns

Experiment
CD146+ Heterogeneous CD146+ Heterogeneous CD146+ Heterogeneous
1 202∗ 90 41∗ 25 3∗ 2
∗ ∗
2 149 50 40 22 3∗ 2
∗ ∗ ∗
3 229 45 33 26 2 1
4 223∗ 60 30∗ 24 3∗ 2
∗
A statistically significant difference of GMSC CD146 homogeneous population compared to the heterogeneous GMSC population.

3.9. Scanning Electron Microscope. No morphological differ- GMSC population, through 0.4 μm, 3 μm, and 8 μm pores
ence was noticed in the cell shape or form or migration pat- of microperforated membranes toward 10% FBS in alpha-
tern through the micropores between the two groups. Both MEM media as a chemoattractant; also, the homogeneous
groups’ cells looked to be flatter and spread over a larger area population showed nonsignificantly better proliferation
over the polycarbonate membrane compared to the collagen capacity than the heterogeneous GMSC population; this
membrane. On collagen, the cells of both groups looked more could prove the hypothesis that the homogeneous CD146
bulbous and extend strands all over the collagen meshwork population can show more proliferation potential and migra-
Figures 7(a)–7(d). tion chemodynamics through microperforated membranes
compared to the heterogeneous GMSC population; this
4. Discussion might be explained by a better migration potential of the
CD146-positive cells or the existence of a chemoattractant
The principle of GTR is to block the migration of the gingival factor in the serum more specific for the CD146-positive
epithelium along the cementum wall of the pocket, creating a cells; this result needs further investigation.
space for stabilization of the blood clot to allow the periodon- The isolated cells demonstrated all the criteria of the
tal ligament (PDL) cells to invade the blood clot for the aim International Society of Cellular Therapy of stem cells,
of periodontal tissue regeneration [22]. The GTR membrane namely, plastic adherence; the ability of colony forming;
is hence a physical barrier that has a biologic effect on the expression of immunophenotype markers CD105, CD73,
healing process of the PDL, affecting the differentiation and and CD90; lack the expression of hematopoietic markers
proliferation of the mesenchyme, and through clot protec- CD45, CD34, and CD14; and finally multipotent differentia-
tion during early stages of healing maintains space for the tion potential [6, 21, 23]. In comparison to the heterogeneous
growing periodontal tissue, to repopulate the wound area GMSC population, CD146-positive homogeneous GMSCs
with selective tissue populations. Hence, the GTR membrane were similar in every aspect, except for faster proliferation,
is considered a biomechanical membrane. and a significant number of cells migrated to the lower com-
In 2018, Al Bahrawy et al. [4] proved the concept that partment of the migration chamber during 24-hour period.
GMSCs can migrate through microperforated membranes In this study, 10% FBS was utilized as the chemoattrac-
of GTR in the presence of a suitable chemoattractant, while tant for both the homogeneous and the heterogeneous
in the absence of the chemoattractant, the membranes were GMSC populations to assess their migration potential
totally occlusive for cell migration; this can be a basis for a through microperforated membranes. In both groups, cells
new generation of the GTR technique, where a selective bar- were seeded in the upper compartment with a concentration
rier membrane was employed, which was occlusive for unde- of 10,000 cells; this concentration was chosen according to
sired cells in the gingival tissue, namely, the epithelium and our previous study which proved that adding a greater
connective tissue cells, and allowed the homing of GMSCs number of cells in the upper compartment made cell identi-
from the gingival tissue to the periodontal wound by utilizing fication and counting absolutely difficult in the lower com-
a suitable chemoattractant in the wound area [4]. partment [4].
In the present study, the proliferation and the migration GMSCs actively migrated irrespective to the effect of
potential of homogeneous CD146-positive GMSC popula- gravitational forces or fluid diffusion forces. In both groups,
tion were compared to a heterogeneous GMSC population the rate of cell migration in 24-hour intervals only varied
of origin; the reason for choosing the CD146 marker to iso- according to the sizes of the pores, where the highest migra-
late the GMSC population was the unique characteristics of tion rate was through the 8 μm pores and the least through
that marker, being not expressed by the fibroblasts and the 0.4 μm pores. GMSCs from both groups did not migrate
expressed by nearly all the MSC populations, which made it with statistical significance through the 0.4 μm and 3 μm
a very good candidate as a marker that insured the stemness pores within the same group, but with statistically significant
of the cell population. difference between groups in favor of the homogeneous pop-
In this study, homogeneous CD146-positive cell popula- ulation group; instead, there was a statistically significant dif-
tion migrated significantly more than the heterogeneous ference in the rate of cell migration through the 8 μm
12 Stem Cells International

WD = 8 mm 2 𝜇m File name = 060915-004.tif Signal A = SE2 Date: 9 Jun 2015 WD = 12 mm 2 𝜇m File name = 041716-070.tif Signal A = SE2 Date: 17 Apr 2016
Mag = 10.00 K X EHT = 5.00 kV Time: 11:15:07 Mag = 10.00 K X EHT = 2.50 kV Time: 16:11:32

(a) (b)

WD = 6 mm 2 𝜇m Signal A = SE2 Date: 5 May 2015 WD = 8 mm 2 𝜇m Signal A = SE2 Date: 5 May 2015
File name = 050515-062.tif Mag = 10.00 K X EHT = 5.00 kV Time: 15:00:18
Mag = 10.00 K X EHT = 5.00 kV Time: 14:40:30 File name = 050515-069.tif

(c) (d)

Figure 7: (a) Scanning electron microscope image of migrated GMSCs in the lower compartment of 8-micron pore perforated polycarbonate
membrane. (b) Scanning electron microscope image of migrated GMSCs in the lower compartment of 3-micron pore perforated collagen-
coated PTFE. (c) Scanning electron microscope image of 0.4-micron pore perforated collagen-coated PTFE showing a GMSC process
extending between collagen strands. (d) Scanning electron microscope image of 0.4-micron pore perforated collagen-coated PTFE
showing fully migrated GMSC.

compared to 3 μm and 0.4 μm within the same group and bonate and collagen substrates on the morphology of the
with a significant difference between the groups in favor of attached cells in Rasmussen et al.’s study [28, 29].
the homogeneous population group. This peculiar finding
suggests that the 8 μm pore size might have a selective 5. Conclusion
migratory effect on GMSCs according to its population
homogeneity. Homogeneous CD146-positive GMSC populations were
The SEM image analysis was indifferent in morphology more dynamically active in the migration through microper-
between the heterogeneous and the homogeneous groups; forated membranes and have shorter proliferation time,
both groups showed a fibroblast-like morphology. GMSCs where 8 μm perforation showed the highest number of
from both groups showed a flatter shape with longer pseudo- migrated cells compared to 0.4 and 3 μm pores. This would
podia over the polycarbonate membrane, compared to throw light on the importance of chemotaxis on homoge-
GMSCs migrated through collagen membrane which looked neous GMSC migration through the microperforated mem-
rougher with many extensions to collagen strands; the differ- brane, using a specific chemoattractant for the homing of
ence can be explained by the difference in the membrane specific GMSC population which would migrate more
roughness, the rougher collagen membrane, and the flat rapidly and proliferate better compared to a nonspecific
polycarbonate membrane [24–26]; this was consistent with chemoattractant which would attract less homogeneous or
previous researches which proved the effect of different sub- heterogeneous GMSCs to the periodontal wound, a pivotal
strates on the shape and morphology of the attached cells [27, development in the guided tissue regeneration technique.
28] and even specifically investigated the effect of polycar- Studying the effect of different chemotaxis factors on
Stem Cells International 13

different stem cell lines to choose the best chemoattractive potential of mesenchymal stem cells reveals a complex hierar-
factor is recommended, besides determining the best stem chy of lineage commitment,” Stem Cells, vol. 28, no. 4, pp. 788–
cell line within the GMSC heterogeneous population that 798, 2010.
can differentiate to multiple cells in the periodontal wound [11] F. Ferro, R. Spelat, A. P. Beltrami, D. Cesselli, and F. Curcio,
for optimum regeneration results. “Isolation and characterization of human dental pulp derived
stem cells by using media containing low human serum per-
centage as clinical grade substitutes for bovine serum,” PLoS
Data Availability One, vol. 7, no. 11, article e48945, 2012.
All the raw data of this study are available for whom is [12] C. Morsczeck, G. Schmalz, T. E. Reichert, F. Völlner, K. Galler,
and O. Driemel, “Somatic stem cells for regenerative den-
concerned.
tistry,” Clinical Oral Investigations, vol. 12, no. 2, pp. 113–
118, 2008.
Conflicts of Interest [13] K. E. Schwab, P. Hutchinson, and C. E. Gargett, “Identification
of surface markers for prospective isolation of human endo-
The author declares that there are no conflicts of interest. metrial stromal colony-forming cells,” Human Reproduction,
vol. 23, no. 4, pp. 934–943, 2008.
References [14] D. T. Covas, R. A. Panepucci, A. M. Fontes et al., “Multipotent
mesenchymal stromal cells obtained from diverse human tis-
[1] M. M. Al Bahrawy, K. K. A. Ghaffar, M. S. El-Mofty, and A. A. sues share functional properties and gene-expression profile
Rahman, “The mutual effect of hyperlipidemia and proinflam- with CD146+ perivascular cells and fibroblasts,” Experimental
matory cytokine related to periodontal infection,” Egyptian Hematology, vol. 36, no. 5, pp. 642–654, 2008.
Journal of Oral and Maxillofacial Surgery, vol. 2, no. 2, [15] M. Crisan, S. Yap, L. Casteilla et al., “A perivascular origin for
pp. 87–95, 2011. mesenchymal stem cells in multiple human organs,” Cell Stem
[2] S. L. Haas, “The importance of the periosteum and the endos- Cell, vol. 3, no. 3, pp. 301–313, 2008.
teum in the repair of transplanted bone,” Archives of Surgery, [16] C. Vaculik, C. Schuster, W. Bauer et al., “Human dermis har-
vol. 8, no. 2, pp. 535–556, 1924. bors distinct mesenchymal stromal cell subsets,” The Journal
[3] B. K. Hall and H. N. Jacobson, “The repair of fractured mem- of Investigative Dermatology, vol. 132, no. 3, pp. 563–574,
brane bones in the newly hatched chick,” Anatomical Record, 2012.
vol. 181, no. 1, pp. 55–69, 1975. [17] S. Shi and S. Gronthos, “Perivascular niche of postnatal mesen-
[4] M. Al Bahrawy, A. Gamal, K. A. Ghaffar, and V. Iacono, “In chymal stem cells in human bone marrow and dental pulp,”
vitro migration dynamics of gingival mesenchymal stem cells Journal of Bone and Mineral Research, vol. 18, no. 4,
through micro perforated membranes,” International Journal pp. 696–704, 2003.
of Dentistry and Oral Health, vol. 4, no. 5, 2018. [18] D. Menicanin, P. M. Bartold, A. C. W. Zannettino, and
[5] G. B. Tomar, R. K. Srivastava, N. Gupta et al., “Human S. Gronthos, “Identification of a common gene expression sig-
gingiva-derived mesenchymal stem cells are superior to bone nature associated with immature clonal mesenchymal cell pop-
marrow- derived mesenchymal stem cells for cell therapy in ulations derived from bone marrow and dental tissues,” Stem
regenerative medicine,” Biochemical and Biophysical Research Cells and Development, vol. 19, no. 10, pp. 1501–1510, 2010.
Communications, vol. 393, no. 3, pp. 377–383, 2010. [19] S. Gronthos, S. E. Graves, S. Ohta, and P. J. Simmons, “The
[6] Q. Zhang, S. Shi, Y. Liu et al., “Mesenchymal stem cells derived STRO-1+ fraction of adult human bone marrow contains the
from human gingiva are capable of immunomodulatory func- osteogenic precursors,” Blood, vol. 84, no. 12, pp. 4164–4173,
tions and ameliorate inflammation-related tissue destruction 1994.
in experimental colitis,” Journal of Immunology, vol. 183,
[20] S. Gronthos, A. C. W. Zannettino, S. J. Hay et al., “Molecular
no. 12, pp. 7787–7798, 2009.
and cellular characterisation of highly purified stromal stem
[7] A. Y. Gamal and V. J. Iacono, “Enhancing guided tissue regen- cells derived from human bone marrow,” Journal of Cell Sci-
eration of periodontal defects by using a novel perforated ence, vol. 116, no. 9, pp. 1827–1835, 2003.
barrier membrane,” Journal of Periodontology, vol. 84, no. 7,
[21] M. F. Pittenger, A. M. Mackay, S. C. Beck et al., “Multilineage
pp. 905–913, 2013.
potential of adult human mesenchymal stem cells,” Science,
[8] A. Y. Gamal, M. Aziz, M. H. Salama, and V. J. Iacono, “Gingi- vol. 284, no. 5411, pp. 143–147, 1999.
val crevicular fluid bone morphogenetic protein-2 release pro-
[22] A. H. Melcher, “On the repair potential of periodontal tissues,”
file following the use of modified perforated membrane
Journal of Periodontology, vol. 47, no. 5, pp. 256–260, 1976.
barriers in localized intrabony defects: a randomized clinical
trial,” Journal of the International Academy of Periodontology, [23] D. J. Prockop, “Marrow stromal cells as stem cells for nonhe-
vol. 16, no. 2, pp. 55–63, 2014. matopoietic tissues,” Science, vol. 276, no. 5309, pp. 71–74,
[9] A. Y. Gamal, K. A. Abdel-Ghaffar, and V. J. Iacono, “Gingival 1997.
crevicular fluid vascular endothelial cell growth factor and [24] H. J. Ronold, S. P. Lyngstadaas, and J. E. Ellingsen, “Analysing
platelet-derived growth factor-BB release profile following the optimal value for titanium implant roughness in bone
the use of perforated barrier membranes during treatment of attachment using a tensile test,” Biomaterials, vol. 24, no. 25,
intrabony defects: a randomized clinical trial,” Journal of Peri- pp. 4559–4564, 2003.
odontal Research, vol. 51, no. 3, pp. 407–416, 2016. [25] P. M. Brett, J. Harle, V. Salih et al., “Roughness response genes
[10] K. C. Russell, D. G. Phinney, M. R. Lacey, B. L. Barrilleaux, in osteoblasts,” Bone, vol. 35, no. 1, pp. 124–133, 2004.
K. E. Meyertholen, and K. C. O'Connor, “In vitro high- [26] L. Li, K. Crosby, M. Sawicki, L. L. Shaw, and Y. Wang, “Effects
capacity assay to quantify the clonal heterogeneity in trilineage of surface roughness of hydroxyapatite on cell attachment and
14 Stem Cells International

proliferation,” Journal of Biotechnology & Biomaterials, vol. 2,

p. 150, 2012.
[27] J. M. Payne, C. M. Cobb, J. W. Rapley, W. J. Killoy, and
P. Spencer, “Migration of human gingival fibroblasts over
guided tissue regeneration barrier materials,” Journal of Peri-
odontology, vol. 67, pp. 236–244, 1996.
[28] J. Zhao, L. Huang, R. Li et al., “Dynamic culture of a thermo-
sensitive collagen hydrogel as an extracellular matrix improves
the construction of tissue-engineered peripheral nerve,” Neu-
ral Regeneration Research, vol. 9, no. 14, pp. 1371–1378, 2014.
[29] C. H. Rasmussen, D. R. Petersen, J. B. Moeller, M. Hansson,
and M. Dufva, “Collagen type I improves the differentiation
of human embryonic stem cells towards definitive endoderm,”
PLoS One, vol. 10, article e0145389, 2015.