Science - USA (2021-12-24)

Discussion
The homeostatic regulation of neutrophils is
critical for optimal function of the immune
system within the oral cavity. Both hypo- and
hyperrecruitment of neutrophils to the oral
mucosa are associated with inflammatory peri-
odontal bone loss. This requirement for finely
balanced neutrophil numbers and neutrophil
function in the maintenance of periodontal
health is revealed by the development of ag-
gressive forms of periodontitis in the setting
of both neutropenia and neutrophil accumu-
lation. Congenital disorders associated with
defects in neutrophil generation and life cycle,
such as the myelokathexis syndrome, severe
congenital neutropenia, and leukocyte adhe-
sion deficiency ( 18 , 19 , 22 , 23 ), lead to early-
onset periodontitis, whereas excess neutrophil
infiltration and activation has been associated
with non-Mendelian forms of periodontal dis-
ease ( 18 , 19 , 22 , 23 ).
The current study provides a clear demon-
stration that increased neutrophil recruitment
per se is insufficient for triggering immuno-
pathology within the oral cavity. Here, we show
that the engagement of neutrophils recruited to
the periodontal environment with locally depo-
sited fibrin is essential for the full execution of
key immunopathology-eliciting neutrophil ef-
fector functions, including NETosis. We found
that even large increases in the steady-state
levels of neutrophils within gingival tissues
because of compromised fibrinolysis failed to
elicit immunopathology in mice genetically
engineered to carry a mutant fibrin unable
to engage neutrophils via the major neutrophil
integrin,aMb 2. Mutations in theaMb 2 -binding
motif protect from fibrin(ogen)-associated im-
munopathology in models of autoimmune dis-
ease, including arthritis and multiple sclerosis,
because of the disruption of monocyte, macro-

phage, and microglial activation ( 24 – 26 ). How-

ever, mechanisms underlying fibrin-triggered

mucosal disease and fibrin-neutrophil–mediated

immunopathology have not been well under-

stood to date.

Neutrophil recruitment and activation, ROS

production, and NETosis have been docu-

mented in tissue biopsies of patients with peri-

odontitis ( 27 – 32 ). Exaggerated activation of

neutrophils, ROS production, and release of

NETs has been linked to a variety of inflam-

matory and autoimmune pathologies ( 33 – 37 )

and has been implicated in inflammatory bone

destruction in the setting of arthritis ( 38 ). How-

ever, how neutrophils become activated and

participate in periodontal immunopathology

is not fully understood. Here, we provide com-

pelling evidence that NETosis is partially

responsible for mediating oral mucosal im-

munopathology and periodontal bone loss

in the setting of defective fibrinolysis. Using

in vivo models of NET depletion (Plg−/−;Ela−/−

andPlg−/−–DNase I), we confirmed that NETosis

plays a significant role in fibrin-induced mu-

cosal immunopathology. However, rescue of

bone loss through NET inhibition was not com-

plete, as it was inPlg−/−;Fgg390-396A/390-396Amice,

which suggests additional roles for fibrin-

mediated immunopathology beyond NETosis.

We hypothesize that neutrophil activation

through the fibrinaMb 2 binding site is a phys-

iologically protective mechanism that is in

place to activate neutrophils at sites of injury

and inflammation and promotes microbial

clearance. Yet, in the setting of fibrin accu-

mulation as a result of defective fibrinolysis,

we reason that it becomes exaggerated and

leads to immunopathology. This mechanism

is particularly relevant to the oral mucosa,

where neutrophils continuously transmigrate

to perform homeostatic functions, even in the

setting of health ( 32 ). However, whether the

fibrin-neutrophil axis is responsible for muco-

sal immunopathology at other barriers (such

as the gastrointestinal tract and ocular sur-

face) inPlgdeficiency is not explored in our

study. It is well appreciated that mucosal dis-

ease inPlgdeficiency is mediated by fibrin

accumulation in all tissues ( 11 ), but it is not

clear whetheraMb 2 -binding by neutrophils

or other location-relevant myeloid subsets or

alternative fibrin-mediated inflammatory path-

ways are engaged to promote immunopath-

ology at different mucosal sites.

We demonstrated that the elimination of

theaMb 2 binding site on fibrin protects wild-

type mice from spontaneous age-related peri-

odontal disease, which indicates that activation

of neutrophil effector functions through fibrin

engagement could contribute to periodontal

disease in individuals with both compromised

and normal fibrinolytic function. Human ge-

netic evidence supports the role ofPLGin

common forms of periodontitis, in the absence

of a syndromic Mendelian disease. A recent

meta-analysis of genome-wide association

studies of periodontitis, including almost

15,000 adult participants of European ances-

try, has revealedPLGpolymorphisms as one of

the two major risk factors for aggressive peri-

odontitis ( 20 ). An earlier study has also high-

lighted polymorphisms downstream ofPLGas

a shared genetic risk locus between cardiovas-

cular disease and periodontitis ( 21 ). In agree-

ment with this previous work, we have now

independently confirmed thatPLGis a genetic

risk factor for periodontitis in a sizeable cohort

of adults of European ancestry in North America.

Furthermore, we find a strong association

ofPLGpolymorphisms with high levels of

the periodontal pathogenA. actinomycetem-

comitans(Aa), a microbe typically detected in

Silvaet al.,Science 374 , eabl5450 (2021) 24 December 2021 7 of 11

Fig. 6. Fibrin-neutrophil axis as
an initiator in common forms
of periodontitis.(A) Fraser-
Lendrum staining for fibrin(ogen)
(left) and immunohistochemistry
(IHC) staining of MPO (right)
in oral mucosal tissues from
patients with severe periodontitis
and healthy controls. Scale bars,
100 mm. (BandC) Regional
association plots ofPLG
(area ± 110-kb flankingPLG)
with severe disease (B) and
highAasubgingival colonization
(C) in the dental atherosclerosis
risk in communities study
(D-ARIC) population.

0.20.4

0.60.8

r^2

0.20.4

0.60.8

r^2

A

Position on chr6 (Mb)

161.05 161.15 161.25

Recombination rate (cM/Mb)

100

80

60

40

20

0

-log

(p-value) 10

10

8

6

4

2

0

rs2465836

Position on chr6 (Mb)

161.05 161.15 161.25

Recombination rate (cM/Mb)

100

80

60

40

20

0

-log

(p-value) 10

10

8

6

4

2

0

rs4252200

Healthy

Periodontitis

Fraser-Lendrum Stain: Fibrin

B C

LPA PLG LPA PLG

MPO

yh

tl

ae

H Periodontitis

RESEARCH | RESEARCH ARTICLE