J Periodont Res 2016; 51: 95–102 © 2015 John Wiley & Sons A/S.

All rights reserved Published by John Wiley & Sons Ltd

JOURNAL OF PERIODONTAL RESEARCH

doi:10.1111/jre.12287

D. C. Ferreira1, L. S. Goncß alves1,

Subgingival bacterial J. F. Siqueira Jr , F. L. Carmo2,
1

H. F. Santos2, M. Feres3,

community profiles in HIV- L. C. Figueiredo3, G. M. Soares3,

A. S. Rosado2, K. R. N. dos
Santos2, A. P. V. Colombo2

infected Brazilian adults with 1

Department of Endodontics and Molecular
Microbiology Laboratory, Esta
cio de Sa

University, Rio de Janeiro, Brazil, 2Institute of

chronic periodontitis Microbiology Prof. Paulo de Go
es, Federal

University of Rio de Janeiro, Rio de Janeiro,
Brazil and 3Dental Research Division,
Department of Periodontology, Guarulhos
University, Guarulhos, Brazil
Ferreira DC, Goncßalves LS, Siqueira JF Jr, Carmo FL, Santos HF, Feres M,
Figueiredo LC, Soares GM, Rosado AS, dos Santos KRN, Colombo APV.
Subgingival bacterial community profiles in HIV-infected Brazilian adults with
chronic periodontitis. J Periodont Res 2016; 51: 95–102. © 2015 John Wiley &
Sons A/S. Published by John Wiley & Sons Ltd

Background and Objective: To compare the subgingival microbial diversity

between non-HIV-infected and HIV-infected individuals with chronic periodon-
titis using denaturing gradient gel electrophoresis (DGGE).
Material and Methods: Thirty-two patients were selected: 11 were HIV-infected
and 21 were non-HIV-infected, and all had chronic periodontitis. Periodontal
measurements included probing depth, clinical attachment level, visible supra-
gingival biofilm and bleeding on probing. Subgingival biofilm samples were col-
lected from periodontal sites (50% with probing depth ≤ 4 mm and 50% with
probing depth ≥ 5 mm) and whole-genomic-amplified DNA was obtained. The
DNA samples were subjected to amplification of a 16S rRNA gene fragment
using universal bacterial primers, followed by DGGE analysis of the amplified
gene sequences.

Results: The non-HIV-infected group presented higher mean full-mouth visible

supragingival biofilm (p = 0.004), bleeding on probing (p = 0.006), probing
depth (p < 0.001) and clinical attachment level (p = 0.001) in comparison with
the HIV-infected group. DGGE analysis revealed 81 distinct bands from all 33
individuals. Banding profiles revealed a higher diversity of the bacterial commu-
nities in the subgingival biofilm of HIV-infected patients with chronic periodon-
titis. Moreover, cluster and principal component analyses demonstrated that the Lucio Souza Goncßalves, DDS, MSc, PhD,

cio de Sa
Faculty of Dentistry, Esta
University,
bacterial community profiles differed between these two conditions. High inter- Av. Alfredo Baltazar da Silveira, 580/cobertura,
individual and intra-individual variability in banding profiles were observed for Recreio, Rio de Janeiro, RJ, Brazil 22790-710
both groups. Tel: +55 21 24978988
Fax: +55 21 22566813
Conclusion: HIV-infected patients with chronic periodontitis present greater e-mail: luciogoncalves@yahoo.com.br

subgingival microbial diversity. In addition, the bacterial communities Key words: bacterial community profiling;
chronic periodontitis; denaturing gradient gel
associated with HIV-infected and non-HIV-infected individuals are different in
electrophoresis; HIV infections
structure.
Accepted for publication April 5, 2015

The periodontal microbiota is com- investigated the composition of the gens has been demonstrated in
posed of a highly complex bacterial periodontal microbiota in HIV-infected non-HIV-infected than in HIV-infected
multispecies community organized in individuals (3–23). Interestingly, a individuals (10,17). However, microor-
biofilms (1,2). Several studies have higher prevalence of periodontal patho- ganisms not commonly associated with
96 Ferreira et al.
periodontitis have frequently been
detected in subgingival sites of HIV-
infected patients, including Staphylo-
coccus epidermidis, Candida albicans,
Enterococcus faecalis, Clostridium diffi-
cile, Mycoplasma salivarium, Klebsiella
pneumoniae, Pseudomonas aeruginosa,
Acinetobacter baumannii and Ent-
amoeba gingivalis (3,16,17,19,22,23). In
particular, E. faecalis was more preva-
lent in the subgingival microbiota of
HIV-infected patients with reduced lev-
els of TCD4+ lymphocytes (< 200 cells/
mm3), suggesting that HIV-related
immunodeficiency can provide appro-
priate conditions for the colonization
and growth of opportunistic pathogens
in the oral microbiota (16).
Even though there are many studies
on the periodontal microbiota of
HIV-infected individuals, none have
consistently profiled the overall bacte-
rial community structures of the sub-
gingival biofilm of these patients in
comparison with the periodontal mic-
robiota of non-HIV-infected individu-
als. The concept of community as the
unit of pathogenicity has been pro-
posed for several endogenous human
infections and may certainly be appli-
cable to periodontal diseases (1,2,24).
This concept supports the notion that
disease is a result of the synergism of
microorganisms and the interaction of
their products in a multispecies con-
sortium (25). This piece of knowledge
in HIV-infected individuals may pro-
vide additional insight into the
ecology of periodontal diseases in this
population, leading ultimately to the
development of more effective
therapeutic interventions.
Several molecular methods have
been used to profile bacterial commu-
nities of the periodontal microbiota,
including pyrosequencing (26), termi-
nal-restriction fragment length poly-
morphism (27) and denaturing
gradient gel electrophoresis (DGGE)
(28,29). The latter technique has been
widely used in oral microbiology stud-
ies (30–33) and has the advantage of
providing a picture of the community
structure in the form of a pattern of
bands. DGGE allows the simulta-
neous analysis of multiple samples,
making it possible to compare the
diversity of different communities.
The aim of the current study was to
compare the subgingival microbial
diversity between non-HIVinfected
and HIV-infected individuals with
chronic periodontitis using DGGE.
Material and methods
Subject population
Thirty-two patients (11 HIV infected
and 21 non-HIV infected) were
selected for this study between 2010
and 2011. The HIV-infected group
comprised patients who were attend-
ing the University Hospital Clementi-
no Fraga Filho of the Federal
University of Rio de Janeiro for
HIV-infection monitoring and were
referred for dental treatment. The
non-HIV-infected patients were
selected from a pool of first-time
patients referred to the Periodontal
Clinic of Guarulhos University for
periodontal treatment. Of those,
patients with chronic periodontitis
(according to the clinical diagnosis
described in the inclusion criteria)
were selected during a period of a
year. The inclusion criteria for all
patients were as follows: diagnosis of
chronic periodontitis; at least four
sites with probing depth and/or clini-
cal attachment level ≥ 5 mm; and
bleeding on probing in different teeth.
Patients were > 20 years of age and
presented at least 15 teeth. Exclusion
criteria included: need of antibiotic
prophylaxis for dental procedures;
pregnancy; diabetes; autoimmune dis-
eases; necrotizing periodontal dis-
eases; having used antibiotics and/or
anti-inflammatory drugs in the last
6 mo; and periodontal therapy in the
last 6 mo. All subjects were informed
about the aims of the study and
signed an informed consent to partici-
pate. The study protocol was
approved by the Review Committee
for Human Subjects of the University
Hospital Clementino Fraga Filho.
Clinical evaluation
Patients were asked to complete a den-
tal and medical history questionnaire,
and data on gender, age and means of
HIV transmission were recorded. The
history of AIDS-defining opportunistic
infections, TCD4+ lymphocyte counts,
plasmatic HIV viral load and antiretro-
viral therapy were obtained from the
patients’ medical records. Oral exami-
nation included visual inspection of the
oral mucosa and periodontal evalua-
tion. Periodontal measurements were
recorded at six sites per tooth (disto-
buccal, buccal, mesiobuccal, distolin-
gual, lingual and mesiolingual) in all
teeth, excluding third molars, and
included probing depth, clinical attach-
ment level, bleeding on probing and vis-
ible supragingival biofilm. These
measurements were performed by one
calibrated examiner (intraclass correla-
tion coefficient of 0.96 for probing
depth and 0.97 for clinical attachment
level) using a conventional manual peri-
odontal probe (University of North
Carolina, Hu-Friedy, Chicago, IL).
After clinical examination, patients
received full-mouth scaling and root
planing under local anesthesia and
instructions for proper home-care pro-
cedures.
Microbiologic assessment
Sample collection— After removing
the supragingival biofilm with sterile
cotton pellets, subgingival biofilm
samples were collected from four to
nine periodontal sites (50% with
probing depth ≤ 4 mm and 50% with
probing depth ≥ 5 mm) of each indi-
vidual using sterile Gracey curettes
(Hu-Friedy), and the samples were
immediately placed in separate
microtubes containing 0.15 mL of
10 mM Tris–HCl, 1 mM EDTA, pH
7.6 (TE).
DNA extraction— Biofilm samples
were vortexed for 30 s and the micro-
bial suspensions were washed three
times with 100 lL of sterile Milli-Q
water. Bacteria were pelleted by cen-
trifugation at 2500 g, the pellet was
resuspended in 100 lL of Milli-Q
water and bacterial DNA was
extracted using the QIAamp DNA
Mini Kit (Qiagen, Valencia, CA,
USA), according to the manufac-
turer’s instructions. DNA extracts
were stored at
20°C until required
for further analysis.
 Subgingival microbial community in HIV infection 97

Multiple displacement amplification— (v/v) formamide], and increasing in ware (GelCompar II Software, version
DNA extracts from clinical samples the direction of electrophoresis. The 5.10; Applied Maths, Kortrijk, Bel-
were subjected to whole-genome DGGE gels were stained with SYBR gium). Dendrograms for diverse com-
amplification using the Illustra Gold (Invitrogen, S~ao Paulo, SP, Bra- parisons of DGGE banding patterns
GenomiPhi V2 DNA Amplification kit zil) and visualized using a Storm 860 were constructed with the unweighted
(GE Healthcare, Piscataway, NJ, Imaging System (GE Healthcare, pair group method using arithmetic
USA) according to the manufacturer0 s Munich, Germany). averages (UPGMA). A similarity level
instructions. In brief, 1 lL of DNA of 60% was arbitrarily considered for
template was added to 9 lL of sample Data analysis— Individual lanes of the cluster preview. The number of bands
buffer containing random hexamer DGGE gel images were straightened of each DGGE profile was calculated,
primers, denatured at 95°C for 3 min and aligned using the GelCompar soft- and significant differences between
in a thermocycler and then cooled to groups were sought using the Mann–
4°C. An aliquot of 1 lL of enzyme mix Whitney U-test. A matrix containing
Table 1. Immunological and HIV-related
containing the phi29 DNA polymerase features of the HIV-infected group (n = 11
the presence or absence of each indi-
and additional random hexamers was subjects) vidual band in the samples was gener-
mixed with 9 lL of reaction buffer ated and used for the detrended
containing deoxynucleotide triphos- Variables Value(mean  SD) correlation analysis (DCA), revealing
phates (dNTPs). This mixture was the linear distribution of data (34).
+
TCD4 lymphocytes a
618.0  296.5
added to the denatured sample to a TCD8+ lymphocytesa 979.4  459.5 Clustering of samples was made by the
final volume of 20 lL and then incu- TCD4+/TCD8+ 0.7  0.3 principal component analysis, using, as
bated at 30°C for 1.5 h. The enzyme HIV-infection 10.7  6.8 input, the presence or absence of data.
was then inactivated by incubation for exposure (years) Shannon diversity indices were calcu-
10 min at 65°C, and the amplified HAART 7.6  3.2 lated for each DGGE profile analyzed
exposure (years)
material was stored at
20°C. This using Canoco for Windows (Wagenin-
multiple displacement amplification a
Cells/mm3.HAART, highly active antiret- gen, the Netherlands) (35). Differences
step was used to improve the perfor- roviral therapy. on demographic and periodontal clini-
mance of the subsequent PCR assays.
Table 2. Demographic and periodontal clinical features of non HIV-infected and HIV-
PCR-DGGE assay— A 16S rRNA infected groups
gene fragment of the whole-genomic-
HIV positive HIV negative
amplified DNA extracts was amplified Variables (n = 11) (n = 21) p
using the universal bacterial primers
968f [50 -AAC GCG AAG AAC CTT Age 47.3  6.9 45.0  9.1 0.271
AC-30 ; containing a 40-bp GC clamp Gendera
(50 -CGC CCG CCG CGC GCG GCG Male 7 (63.6) 10 (47.6) 0.410
Female 4 (36.4) 11 (52.4)
GGC GGG GCG GGG GCA CGG
Periodontal parametersb
GGG G-30 ) added to its 50 -end] and Percentage of sites with:
1401r (50 -CGG TGT GTA CAA GAC Visible supragingival biofilm 58.0  15.1 75.8  13.3 0.004
CC-30 ). The presence of PCR products Bleeding on probing 34.8  21.4 63.4  27.9 0.006
was confirmed by electrophoresis in a Probing depth (mm) 2.7  0.4 3.7  0.5 < 0.001
1.5% agarose gel. The gel was stained Clinical attachment level (mm) 3.0  0.6 4.0  0.7 0.001
for 15 min with 0.5 lg/mL of ethidium Values are given as mean  SD or n (%).BOP, .
bromide and viewed under short-wave- a
Mann–Whitney U-test.
b
length ultraviolet light. A 100-bp DNA Fisher’s exact test.
ladder tool served as the molecular size
standard.
DGGE of the amplified gene
sequences was performed using the
Dcode Universal Mutation Detection
System (Bio-Rad Dcode, Richmond,
VA, USA) at 75 V and 60°C for 16 h
in 1 9 Tris-acetate-EDTA. The PCR
products (30 lL) were loaded onto
6% (wt/vol) polyacrylamide gels con-
taining a linear gradient, ranging from Fig. 1. Mean number of bands (A) and Shannon index (B) of bacterial communities of the
40% to 70%, of the denaturants urea subgingival biofilm of Control (non-HIV-infected) and HIV (HIV-infected subjects) with
and formamide [100% denaturant chronic periodontitis, determined by PCR-denaturing gradient gel electrophoresis (PCR-
corresponded to 7 M urea and 40% DGGE) analysis. (p < 0.001, Mann–Whitney U-test).
98 Ferreira et al.

Table 3. Frequency of denaturing gradient Table 3. (continued) under control in this group of sub-
gel electrophoresis (DGGE) bands jects. In addition, all subjects were
obtained from subgingival biofilm samples HIV HIV
positive negative
undergoing highly active antiretroviral
of 212 periodontal sites of the non-HIV-
[n = 11 [n = 21 therapy (HAART).
infected and HIV-infected groups
patients (66 patients (146
HIV HIV sites)] sites)] Clinical and demographic features
positive negative
of the sample population
[n = 11 [n = 21 Band N % n %
patients (66 patients (146 Information on demographic features
sites)] sites)] B53a 30 44.8 41 27.2
B54 21 31.3 35 23.2 and periodontal clinical parameters of
Band N % n % B55 27 40.3 36 23.8 both groups is presented in Table 2.
B56 12 17.9 21 13.9 No differences in gender and age were
B1 1 1.5 4 2.6 B57 27 40.3 23 15.2 found between groups. In contrast,
B2 3 4.5 12 7.9 B58 25 37.3 27 17.9 non-HIV-infected individuals pre-
B3 1 1.5 11 7.3 B59 24 35.8 32 21.2 sented significantly more visible
B4 5 7.5 20 13.2 B60 16 23.9 21 13.9
B5 9 13.4 24 15.9
supragingival biofilm, periodontal
B61 7 10.4 11 7.3
B6 9 13.4 31 20.5 B62 20 29.9 15 9.9 inflammation and tissue destruction
B7 16 23.9 11 7.3 B63 10 14.9 14 9.3 compared with HIV-infected individu-
B8 14 20.9 13 8.6 B64 3 4.5 13 8.6 als.
B9 19 28.4 14 9.3 B65 4 6.0 4 2.6
B10 14 29.9 6 4.0 B66 7 10.4 4 2.6
B11 20 29.9 12 7.9 B67 1 1.5 3 2.0 Microbiological data
B12 18 26.9 18 11.9 B68 2 3.0 4 2.6
B13 18 26.9 33 21.9 B69 4 6.0 1 0.7
Overall, DGGE analysis revealed 81
B14 30 44.8 23 15.2 B70 3 4.5 4 2.6 distinct bands from the 33 HIV- and
B15 27 40.3 31 20.5 B71 2 3.0 2 1.3 non-HIV-infected individuals. Among
B16 18 32.7 37 67.3 B72 3 4.5 2 1.3 the 212 subgingival sites analyzed,
B17a 29 43.3 47 31.1 B73 3 4.5 3 2.0 bands B17, B43–B48, B50, B52 and
B18 31 46.3 27 17.9 B74 6 9.0 4 2.6 B53 demonstrated the highest fre-
B19 13 19.4 22 14.6 B75 2 3.0 3 2.0
quency of detection (> 30%) (Table 3).
B20 22 32.8 28 18.5 B76 3 4.5 7 4.6
B21 15 22.4 17 11.3 B77 5 7.5 0 0.0 The mean percentage of bands
B22 13 19.4 6 4.0 B78 9 13.4 2 1.3 observed was significantly higher in the
B23 8 11.9 7 4.6 B79 5 7.5 0 0.0 HIV-infected group (mean  SD: 16.6
B24 9 13.4 10 6.6 B80 3 4.5 5 3.3  7.1%) than in the non-HIV-infected
B25 14 20.9 7 4.6 B81 4 6.0 1 0.7 group (mean  SD: 11.1  5.5%)
B26 8 11.9 13 8.6 a
The most frequently detected bands in all (p < 0.001, Mann–Whitney U-test)
B27 20 29.9 16 10.6
B28 17 25.4 10 6.6 samples (> 30%). (Fig. 1A). These results were also evi-
B29 13 19.4 28 18.5 Bold: difference of ≥ 20% in the frequency dent when the mean number of bands
B30 12 17.9 23 15.2 of bands between groups. was calculated using the Shannon
B31 21 31.3 21 13.9 diversity index (Fig. 1B), indicating a
B32 25 37.3 25 16.6 cal data between groups were exam- higher diversity of bacterial communi-
B33 13 19.4 33 21.9
ined with Fisher0 s exact and Mann– ties in subgingival biofilm samples
B34 19 28.4 25 16.6
Whitney U-tests, using the SPSS soft- from the HIV group compared with
B35 14 20.9 20 13.2
B36 25 37.3 30 19.9 ware program (SPSS Statistics v. 19.0; the control group.
B37 18 26.9 35 23.2 IBM Brazil, S~ao Paulo, SP, Brazil). High interindividual variability was
B38 20 29.9 34 22.5 Significance was established at 5% for also observed in terms of banding pat-
B39 27 40.3 29 19.2 all tests. terns (i.e. no two samples showed
B40 25 37.3 31 20.5 exactly the same bacterial community
B41 30 44.8 27 17.9 profile). Similar findings were observed
B42 27 40.3 32 21.2 Results
regarding intra-individual analyses
B43a 31 46.3 37 24.5
B44a 40 59.7 43 28.5 (data not shown). Comparative analy-
HIV-related characteristics
B45a 32 47.8 41 27.2 sis of the two data sets revealed that
B46a 32 47.8 57 37.7 Table 1 shows the immunological and the great majority of bands were pres-
B47a 33 49.3 44 29.1 HIV-associated data of the HIV- ent in both groups, and only two bands
B48a 33 49.3 36 23.8 infected group. All subjects had unde- (B77 and B79) were exclusively found
B49 24 35.8 37 24.5
tectable plasmatic HIV viral load in HIV-infected patients (both were
B50a 31 46.3 48 31.8
B51 27 40.3 30 19.9 (data not shown) and high mean lev- found in five sites) (Table 3).
B52a 18 26.9 49 32.5 els of TCD4+ lymphocytes (618 cells/ Cluster analysis revealed 33 different
mm3), suggesting that the disease was groups with similarity set at 60%

48.5%
HIV (+) (16 clusters)
6.0%
Fig. 2. Distribution of 33 clusters of different denaturing gradient gel electrophoresis
(DGGE) band profiles in non-HIV-infected [HIV (
)] and HIV-infected [HIV (+)] groups.
Fig. 3. Representative denaturing gradient gel electrophoresis (DGGE) profiles of PCR-
amplified bacterial 16S rRNA gene fragments from subgingival biofilm of eight non-HIV-
infected [HIV (
)] and 15 HIV-infected [HIV (+)] subjects.
(Fig. 2). Only two clusters (6%) had
samples from both groups; 48.5% and
45.5% of the clusters were exclusive to
HIV-infected and non-HIV-infected
groups, respectively. Figure 3 depicts
the community profile of represen-
tative subgingival clinical samples
obtained from eight non-HIV-infected
and 15 HIV-infected subjects.
Principal component analysis, based
on the PCR-DGGE banding profile,
showed a tendency of separation by
PC1 axis of the samples from the non-
HIV-infected subjects (Control) in
relation to those from HIV-infected
subjects (HIV) (Fig. 4). It can be
observed that most of the samples
from HIV-infected subjects were
located in the right of the figure and
Subgingival microbial community in HIV infection 99
in the subgingival biofilm of HIV-
infected patients compared with non-
45.5%
(15 clusters) HIV (–) HIV-infected individuals. There were
also different community profiles
between the groups, indicating the
possibility of some patterns related to
the HIV infection. Of interest, the
periodontal clinical profile was less
severe in HIV-infected patients than
in non-HIV-infected subjects, corrob-
Site Group orating data reported by other studies
with the same population (17,18). One
HIV (+)
hypothesis that could partially explain
HIV (+)
these data is that HIV-infected indi-
HIV (+)
viduals are normally more cautious
HIV (+)
about their general health and are
HIV (+)
exposed more to other oral/medical
HIV (+)
care professionals and other drugs,
HIV (+)
which might have an influence on the
HIV (+)
periodontal clinical parameters, such
HIV (+)
as the bleeding scores.
HIV (+)
Banding patterns of the subgingival
HIV (+)
microbiota in HIV- and non-HIV-in-
HIV (+)
fected individuals showed relative het-
HIV (+)
erogeneity. The fingerprints in the
HIV (+)
HIV-infected group revealed more
HIV (+)
DGGE bands than those in the non-
HIV (–)
HIV-infected (control) group. The
HIV (–)
number of DGGE bands may be inter-
HIV (–)
preted as the number of bacterial spe-
HIV (–)
cies in the community (36). Whereas
HIV (–)
some factors, such as as heteroduplex
HIV (–)
formation and discrepancies in the
HIV (–)
number of rrn operons in different spe-
HIV (–)
cies, may overestimate species richness,
other factors may underestimate it,
including sampling and sample homog-
enization biases, differential DNA
extraction, PCR biases and co-migra-
those from the non-HIV-infected sam- tion on DGGE gels (37–39).
ples in the left. In addition, samples The presence of a larger number of
from the non-HIV-infected subjects species composing the individual
were closer to each other compared subgingival communities in the HIV-
with samples from the HIV group. infected group suggests a more com-
This indicates that the bacterial com- plex microbiota, despite the lower
munity is more similar among systemi- level of periodontal destruction
cally healthy patients than among observed in these patients. In addi-
HIV-infected patients. tion, clustering analysis revealed that
the subgingival samples clustered
according to the clinical groups. This
Discussion
indicates that the bacterial communi-
The current study compared the sub- ties associated with HIV-infected and
gingival bacterial community struc- non-HIV-infected individuals were
tures of 11 HIV-infected patients with different in structure (species composi-
21 non-HIV-infected subjects (all with tion and abundance).
chronic periodontitis) using DGGE The high bacterial diversity of the
analysis. The results showed higher oral microbiota of HIV-infected indi-
diversity of the bacterial communities viduals has been previously reported
100 Ferreira et al.
0.8
PC 2
5.1
–0.6
–0.6 8.4
PC 1
Fig. 4. Score plot of principal component analysis based on PCR-denaturing gradient gel
electrophoresis (DGGE) profiles of the total bacterial community in control (non-HIV-
infected) and HIV (HIV-infected) individuals.
(17,40–42). Goncßalves et al. (17)
observed that HIV-infected patients
presented higher mean levels and prev-
alence of opportunist species involved
in nosocomial infections, such as
A. baumannii, E. faecalis and P. aeru-
ginosa, in the subgingival biofilm than
non-HIV-infected patients. Dang et al.
(40) used microarray to compare the
bacterial composition of the lingual
microbiome between 12 HIV-infected
patients (six untreated and six under
HAART) and nine uninfected con-
trols. The authors reported that the
administration of antiretroviral ther-
apy may change the profile of the oral
microbiota, and indicated that chronic
HIV infection may lead to substantial
disruptions in the community structure
of the lingual microbiota, even in the
absence of clinical oral manifestations.
Saxena et al. (41) utilized the DGGE
method and also observed a greater
diversity of the overall oral microbial
population of HIV-positive individuals
compared with HIV-negative controls.
The higher bacterial diversity and
the different community patterns in
the periodontal microbiota of HIV-
infected patients compared with non-
HIV-infected individuals may be
related to the HIV-infection condi-
tion, as well as to the long-term use
of HAART and several other drugs
SAMPLES
Control
HIV
1.0
that adversely affect the oral immu-
nity (43,44). For instance, it has been
suggested that Notch-1 signaling can
mediate epithelial differentiation in
oral mucosa through interaction with
TCD4+ lymphocytes (45,46). In this
way, it is possible that depletion of
TCD4+ lymphocytes from the oral
mucosa of HIV-infected individuals
may also lead to the impairment of
epithelial growth and, accordingly,
host–microbe dysbiosis (40). The
reduced capacity of the oral immune
response in this population has also
been demonstrated by decreased
expression of histatin-5 (a potent
antimycotic agent) (47), tumor necro-
sis factor-a, interleukin-6 (44) and
human b-defensin-2 (43). Several oral
tissues, including gingiva (50), can
express human b-defensin-2, which
has a function of chemoattractant for
dendritic cells and acts as part of the
innate antimicrobial defense (48,49).
Nittayananta et al. (43) demonstrated
that the levels of human b-defensin-2
protein in saliva of HIV-infected indi-
viduals was increased with short-term
use of HAART but decreased with
long-term use of the medication. This
change in oral mucosal immunity may
result in significant alterations in the
composition of oral bacteria, putting
HIV-infected patients at risk of
developing oral infections (43). Iwai
et al. (42) found distinct buccal and
airway bacterial communities in HIV-
infected patients with acute respira-
tory infections undergoing HAART
and receiving antimicrobial therapy
for pneumonia in comparison with
non-HIV-infected controls.
In summary, the results of this study
showed a high interindividual variabil-
ity in the bacterial composition of the
subgingival microbiota. This is in
agreement with the concept that
chronic periodontitis may have a heter-
ogeneous etiology and that periodontal
destruction is the result of an ecologi-
cal disruption of the microbiome, host
response and periodontal microenvi-
ronment (1,51). However, several
bands were shared by many of the
DGGE profiles, suggesting that these
species may be important keystone
pathogens (52). It was not possible to
identify these species after cutting out
the DGGE bands and sequencing them
because of the technical limitations, in
some cases, and the poor quality of
sequences obtained, even after cloning
(data not shown).
In conclusion, the microbiota
associated with periodontitis in HIV-
positive individuals was shown to dif-
fer significantly in comparison with
that of HIV-negative individuals.
Obtaining a better understanding of
the microbial composition of the peri-
odontal microbiota, and its complex
interaction with the host, has been a
great challenge for numerous investi-
gators over the years. Understanding
these mechanisms is even more chal-
lenging when many other factors are
present, as seen in HIV-infected
patients. In these individuals, an
unbalanced periodontal microbial
community as a result of local or sys-
temic conditions can be observed. The
complexity of this microbial commu-
nity needs to be further unraveled in
order to provide information regard-
ing the possible interaction with other
oral sites and the systemic repercus-
sion.
Acknowledgements
This study was supported by grants
from National Council for Scientific
 Subgingival microbial community in HIV infection 101

and Technological Development 12. Jabra-Rizk MA, Falkler WA Jr, gingival microorganisms in HIV-Infected
(CNPq), Brasilia, Brazil; Coordina- Enwonwu CO, Onwujekwe DI Jr, Merz brazilian adults: a multilevel analysis. J
WG, Meiller TF. Prevalence of yeast Periodontol 2014;85:697–705.
tion of Improvement of Higher Edu-
among children in Nigeria and the Uni- 24. Marsh PD. Microbial ecology of dental
cation Personnel (CAPES), Brasilia, ted States. Oral Microbiol Immunol plaque and its significance in health and
Brazil; and Foundation for Research 2001;16:383–385. disease. Adv Dent Res 1994;8:263–271.
Financial Support in the State of Rio 13. Tsang CS, Samaranayake LP. Predomi- 25. Siqueira JF Jr, R^ oßcas IN. Community as
de Janeiro (FAPERJ), Rio de Janeiro, nant cultivable subgingival microbiota of the unit of pathogenicity: an emerging
Brazil. healthy and HIV-infected ethnic Chinese. concept as to the microbial pathogenesis
APMIS 2001;109:117–126. of apical periodontitis. Oral Surg Oral
14. Paster BJ, Russell MK, Alpagot T et al. Med Oral Pathol Oral Radiol Endod
References Bacterial diversity in necrotizing ulcera- 2009;107:870–878.
tive periodontitis in HIV-positive sub- 26. Griffen AL, Beall CJ, Campbell JH et al.
1. Kuramitsu HK, He X, Lux R, Anderson
jects. Ann Periodontol 2002;7:8–16. Distinct and complex bacterial profiles in
MH, Shi W. Interspecies interactions
15. Patel M, Coogan M, Galpin JS. Peri- human periodontitis and health revealed
within oral microbial communities.
odontal pathogens in subgingival plaque by 16S pyrosequencing. ISME J
Microbiol Mol Biol Rev 2007;71:653–670.
of HIV-positive subjects with chronic 2012;6:1176–1185.
2. Marsh PD. Dental plaque: biological sig-
periodontitis. Oral Microbiol Immunol 27. Sakamoto M, Huang Y, Ohnishi M,
nificance of a biofilm and community
2003;18:199–201. Umeda M, Ishikawa I, Benno Y.
life-style. J Clin Periodontol 2005;32:7–
16. Goncßalves LS, Ferreira SM, Silva A Jr Changes in oral microbial profiles after
15.
et al. Association of TCD4 lymphocyte periodontal treatment as determined by
3. Zambon JJ, Reynolds HS, Genco RJ.
levels and subgingival microbiota of molecular analysis of 16S rRNA genes. J
Studies of the subgingival microflora in
chronic periodontitis in HIV-infected Med Microbiol 2004;53:563–571.
patients with acquired immunodeficiency
Brazilian under HAART. Oral Surg Oral 28. Fujimoto C, Maeda H, Kokeguchi S
syndrome. J Periodontol 1990;61:699–
Med Oral Pathol Oral Radiol Endod et al. Application of denaturing gradient
704.
2004;97:196–203. gel electrophoresis (DGGE) to the analy-
4. Murray PA, Winkler JR, Peros WJ,
17. Goncßalves LS, Ferreira SM, Souza CO, sis of microbial communities of subgingi-
French CK, Lippke JA. DNA probe
Souto R, Colombo AP. Clinical and val plaque. J Periodontal Res
detection of periodontal pathogens in
microbiological profiles of human immu- 2003;38:440–445.
HIV associated periodontal lesions. Oral
nodeficiency virus (HIV)-seropositives 29. Zijnge V, Welling GW, Degener JE, van
Microbiol Immunol 1991;6:34–40.
Brazilians undergoing highly active anti- Winkelhoff AJ, Abbas F, Harmsen HJ.
5. Lucht E, Heimdahj A, Nord CE. Peri-
retroviral therapy and HIV-seronegatives Denaturing gradient gel electrophoresis
odontal disease in HIV-infected patients
Brazilians with chronic periodontitis. J as a diagnostic tool in periodontal micro-
in relation to lymphocyte subsets and
Periodontol 2007;78:87–96. biology. J Clin Microbiol 2006;44:3628–
specific micro-organisms. J Clin Period-
18. Goncßalves LS, Souto R, Colombo AP. 3633.
ontol 1991;18:252–256.
Detection of Helicobacter pylori, Entero- 30. Siqueira JF Jr, R^ oßcas IN, Debelian GJ
6. Rams TE, Andriolo M Jr, Feik D, Abel
coccus faecalis, and Pseudomonas aeru- et al. Profiling of root canal bacterial
SN, McGivern TM, Slots J. Microbio-
ginosa in the subgingival biofilm of communities associated with chronic
logical study of HIV-related periodonti-
HIV-infected subjects undergoing HAART apical periodontitis from Brazilian and
tis. J Periodontol 1991;62:74–81.
with chronic periodontitis. Eur J Clin Norwegian subjects. J Endod
7. Moore LV, Moore WE, Riley C, Brooks
Microbiol Infect Dis 2009;28:1335–1342. 2008;34:1457–1461.
CN, Burmeister JA, Smibert RM. Peri-
19. Botero JE, Contreras A, Lafaurie G, 31. Pushalkar S, Ji X, Li Y et al. Compari-
odontal microflora of HIV positive sub-
Jaramillo A, Betancourt M, Arce RM. son of oral microbiota in tumor and
jects with gingivitis or adult periodontitis.
Occurrence of periodontopathic and non-tumor tissues of patients with oral
J Periodontol 1993;64:48–56.
superinfecting bacteria in chronic and squamous cell carcinoma. BMC Micro-
8. Cross DL, Smith GLF. Comparison of
aggressive periodontitis subjects in a biol 2012;12:144.
periodontal disease in HIV seropositive
Colombian population. J Periodontol 32. Al-Radha AS, Pal A, Pettemerides AP,
subjects and controls. (II) Microbiology,
2007;78:696–704. Jenkinson HF. Molecular analysis of
immunology and predictors of disease. J
20. Aas JA, Barbuto SM, Alpagot T, Olsen I, microbiota associated with peri-implant
Clin Periodontol 1995;22:569–577.
Dewhirst FE, Paster BJ. Subgingival pla- diseases. J Dent 2012;40:989–998.
9. Brady LJ, Walker C, Oxford GE, Stew-
que microbiota in HIV positive patients. J 33. Zhu X, Wang S, Gu Y et al. Possible
art C, Magnusson I, McArthur W. Oral
Clin Periodontol 2007;34:189–195. variation of the human oral bacterial
diseases, mycology and periodontal
21. Ramos MP, Ferreira SM, Silva-Boghos- community after wearing removable par-
microbiology of HIV-1- infected women.
sian CM et al. Necrotizing periodontal tial dentures by DGGE. World J Micro-
Oral Microbiol Immunol 1996;11:371–
diseases in HIV-infected Brazilian biol Biotechnol 2012;28:2229–2236.
380.
patients: a clinical and microbiologic 34. Andreote FD, Azevedo JL, Ara
ujo WL.
10. Tenenbaum H, Elkaim R, Cuisinier F,
descriptive study. Quintessence Int Assessing the diversity of bacterial com-
Dahan M, Zamanian P, Lang JM. Preva-
2012;43:71–82. munities associated with plants. Braz J
lence of six periodontal pathogens
22. Cembranelli SB, Souto FO, Ferreira- Microbiol 2009;40:417–432.
detected by DNA probe method in HIV 
Paim K et al. First evidence of genetic 35. Leps J, Smilauer P. Multivariate Analysis
vs non-HIV periodontitis. Oral Dis 1997;
intraspecific variability and occurrence of of Ecological Data Using CANOCO.
(suppl I): S153–S155.
Entamoeba gingivalis in HIV(+)/AIDS. Cambridge, United Kingdom: Cambridge
11. Scully C, Porter SR, Mutlu S, Epstein
PLoS ONE 2013;8:e82864. University Press, 1999:269 pp.
JB, Glover S, Kumar N. Periodontopath-
23. Pereira VT, Pavan P, Souza RC et al. 36. Muyzer G, de Waal EC, Uitterlinden
ic bacteria in English HIV-seropositive
The Association between detectable plas- AG. Profiling of complex microbial pop-
persons. AIDS Patient Care STDS
matic HIV viral load and different sub- ulations by denaturing gradient gel elec-
1999;13:369–374.
102 Ferreira et al.
trophoresis analysis of polymerase chain
reaction-amplified genes coding for 16S
rRNA. Appl Environ Microbiol 1993; 59:
695–700.
37. Gillan DC, Speksnijder AGCL, Zwart G,
de Ridder C. Genetic diversity of the bio-
film covering Montacuta ferruginosa
(Mollusca, Bivalvia) as evaluated by dena-
turing gradient gel electrophoresis analysis
and cloning of PCR-amplified gene frag-
ments coding for 16S rRNA. Appl Environ
Microbiol 1998;64:3464–3472.
38. G€obel UB. Phylogenetic amplification
for the detection of uncultured bacteria
and the analysis of complex microbiota.
J Microbiol Meth 1995;23:117–128.
39. Wintzingerode FV, G€ obel UB, Stacke-
brandt E. Determination of microbial
diversity in environmental samples: pit-
falls of PCR-based rRNA analysis.
FEMS Microbiol Rev 1997;21:213–229.
40. Dang AT, Cotton S, Sankaran-Walters S
et al. Evidence of an increased patho-
genic footprint in the lingual microbiome
of untreated HIV infected patients. BMC
Microbiol 2012;12:153.
41. Saxena D, Li Y, Yang L et al. Human
microbiome and HIV/AIDS. Curr HIV/
AIDS Rep 2012;9:44–51.
42. Iwai S, Fei M, Huang D et al. Oral and
airway microbiota in HIV-infected pneu-
monia patients. J Clin Microbiol
2012;50:2995–3002.
43. Nittayananta W, Kemapunmanus M,
Amornthatree K, Talungchit S, Sriplung
H. Oral human b-defensin 2 in HIV-
infected subjects with long-term use of
antiretroviral therapy. J Oral Pathol Med
2013a;42:53–60.
44. Nittayananta W, Amornthatree K,
Kemapunmanus M, Talungchit S, Sri-
plung H. Expression of oral cytokines in
HIV-infected subjects with long-term use
of antiretroviral therapy. Oral Dis
2013b;20:e57–e64.
45. Casey LM, Lan Y, Cho ES, Maltby
KM, Gridley T, Jiang R. Jag2-Notch1
signaling regulates oral epithelial differ-
entiation and palate development. Dev
Dyn 2006;235:1830–1844.
46. Dahan S, Rabinowitz KM, Martin AP,
Berin MC, Unkeless JC, Mayer L.
Notch-1 signaling regulates intestinal epi-
thelial barrier function, through interac-
tion with CD4 + T cells, in mice and
humans. Gastroenterology 2011;140:550–
559.
47. Torres SR, Garzino-Demo A, Meiller
TF, Meeks V, Jabra-Rizk MA. Salivary
histatin-5 and oral fungal colonisation in
HIV+ individuals. Mycoses 2009;52:11–
15.
48. Chertov O, Yang D, Howard OM,
Oppenheim JJ. Leukocyte granule pro-
teins mobilize innate host defenses and
adaptive immune responses. Immunol
Rev 2000;177:68–78.
49. Dale BA, Fredericks LP. Antimicrobial
peptides in the oral environment: expres-
sion and function in health and disease.
Curr Issues Mol Biol 2005;7:119–133.
50. Krisanaprakornkit S, Weinberg A,
Perez C, Dale B. Expression of the
peptide antibiotic human beta defensin
1 in cultured gingival epithelial cells
and gingival tissue. Infect Immun
1998;66:4222–4228.
51. Jenkinson HF, Lamont RJ. Oral micro-
bial communities in sickness and in
health. Trends Microbiol 2005;13:589–
595.
52. Hajishengallis G, Darveau RP, Curtis
MA. The keystone-pathogen hypothesis.
Nat Rev Microbiol 2012;10:717–725.