вход по аккаунту


Outbreak of spontaneous staphylococcal arthritis and osteitis in mice.

код для вставкиСкачать
Staphylococcus aureus is the most common bacterial species found in association with nongonococcal
bacterial arthritis in humans. We present here the first
description of spontaneous bacterial arthritis and osteitis in mice. Clinically, the most obvious findings were
swelling and/or ankylosis of hindpaws and nodose
changes of the tail. The prevalence of arthritis and
osteitis ranged from 0% to >SO% of the mice studied,
depending on the mouse strain. The most prominent
histopathologic feature of the arthritis was hypertrophy
of the synovial tissue and destruction of cartilage and
underlying bone. Most of the S aureus-infected mice
displayed an identical phage type, which was also the
only S aureus phage type found in skin isolates from
From the Department of Rheumatology, the Department of
Clinical Bacteriology, Immunology and Virology, the Department of
Histology, and the Depaflment of Oral Radiology, University of
Goteborg, Goteborg, Sweden.
Supported by grants from the Goteborg Medical Society,
the Swedish Medical Society, The Swedish Association against
Rheumatism, King Gustaf V’s 80-Year-Fund, the Swedish National
Board for Technical Development, and the Swedish Medical Research Council.
Tomas Bremell, MD: Departments of Rheumatology and
Clinical Bacteriology, Immunology and Virology; Stefan Lange,
MD, PhD: Associate Professor. Department of Clinical Bacteriology, Immunology and Virology: Lena Svensson: Technical Assistant, Department of Clinical Bacteriology, Immunology and Virology; Eva Jennische, MD, PhD: Associate Professor, Department of
Histology; Kerstin Grondahl, DDS, PhD: Associate Professor,
Department of Oral Radiology; Hans Carlsten, MD, PhD: Departments of Rheumatology and Clinical Bacteriology, Immunology and
Virology; Andrzej Tarkowski. MD, PhD: Associate Professor,
Departments of Rheumatology and Clinical Bacteriology, Immunology and Virology.
Address reprint requests to Tomas Bremell, MD, Department of Clinical Immunology, Guldhedsgatan 10, S413 46 Goteborg,
Submitted for publication March 26, 1990; accepted in
revised form June 5 , 1990.
Arthritis and Rheumatism, Vol. 33, No. 11 (November 1990)
clinically healthy mice. However, a few S uureus isolates
were not typeable, indicating that an additional strain(s)
might cause bacterial arthritis in mice.
Bacterial arthritis is a common and serious
problem at major urban medical centers. Risk factors
include impaired host defense mechanisms caused by
factors such as immunosuppressive drugs and chronic
illnesses, as well as previous joint damage such as that
associated with prosthetic joint surgery and chronic
arthritis. These factors place rheumatoid arthritis patients at risk of developing bacterial arthritis. Staphylococcus aureus is the most common bacterium found
in nongonococcal bacterial arthritis in humans (I).
Infectious arthritis caused by S aureus is known to
cause severe destruction of cartilage and subchondral
bone within a short time (1).
There are only a few reports of spontaneously
occurring bacterial arthritis in animals. Boetner et al
reported Streptococcus suis infections in pigs, with
involvement of joints, as well as serous membranes
and meninges (2). Feinstein and Eld reported infection
of large limb joints with growth of Erysipelothrir
rhusiopathiae in 2 Sprague-Dawley rats (3). An adult
anorectic West African dwarf crocodile was examined
because of bilateral hind limb paresis, and Serratia
marcexens as well as Morganella morganii were
isolated from several organs, including the right coxofemoral joint (4). Kasari et a1 reported a beefmaster
cow with left hind leg lateral claw lameness due to
septic arthritis of the distal interphalangeal joint and
associated osteomyelitis of the second and third phalanges. Bacterial isolation from the joint revealed
growth of Actinomyces pyogenes and Fusobacterium
necrophorum ( 5 ) . In the present report, we describe a
spontaneous outbreak of staphylococcal arthritis in
Mice. We studied 6 different strains of male and
female mice, aged 4 months or more, that were housed
in 2 different rooms in the animal facility at the
Department of Clinical Bacteriology, Immunology and
Virology, University of Goteborg. The strains MRLlprllpr (n = 300), New Zealand white (NZW) (n = 30),
and New Zealand black (NZB) (n = 30) were originally
obtained from Harlan Olac Farm (Bicester, UK).
Crosses of (NZB x NZW)F, (NZB/W) (n = 40) and
(NZB/W x NZB)F, (NZB/W/B) mice (n = 54) were
performed at the animal facility. DBA/l mice (n = 100)
were originally obtained from Jackson Laboratories
(Bar Harbor, ME). Between 1 and 12 mice were
housed in each cage, under standard conditions of
temperature and light. They were fed standard laboratory chow and water ad libitum. The majority of the
NZB/W and NZB/W/B mice described here were
being used in experiments on the influence of sex
hormones on autoimmunity. The mice in those experiments were castrated and treated with regular subcutaneous injections of estradiol benzoate (Sigma, St.
Louis, MO) or testosterone deconate (Organon, Nijmegen, The Netherlands); control mice were treated
with plain olive oil. The NZB/W/B mice were castrated at the age of 2 months, and 1.5 months later,
they began receiving subcutaneous injections of either
testosterone deconate or olive oil every third week.
Bacteriologic examination. Samples for bacterial
analysis were obtained using charcoaled sticks, after
dissection of the talocrural and subtalar joints of the
hindpaw and radiocarpaljoint of the forepaw. Samples
from blood, lymph nodes, and tails were also obtained.
All the samples were initially transferred to chocolate
agar, Drigalski agar, chocolate agar with vancomycin
and bacitracin, enterococcosal agar (12205; BBL,
Cockeysville, MD), and blood agar containing 5%
human blood. All the bacterial samples were also
tested for coagulase activity. Since the only pathogen
found throughout was S aureus, subsequent bacterial
analyses were performed only on 5% blood agar.
Isolates from 70 arthritic and nonarthritic mice
were obtained from arthritic joints, nonarthritic joints,
normal and nodose tails, and blood. An isolate was
considered positive if more than 20 bacterial colonies
were present after 48 hours of incubation at 37°C.
Isolates were taken with wet charcoaled sticks from
the skin of the backs of 20 clinically healthy mice (10
MRL-lpr/lpr, 5 NZB/W, and 5 DBA/l) and transferred
to blood agar. Bacterial colonies were tested for
catalase and coagulase activity. Bacteriophage typing
was performed according to an internationally standardized technique (6).
Histopathologic examination. Histologic examination was performed after routine fixation, decalcification, plastic embedding, and staining with a mixture
of basic fuchsin/methylene blue/azure 11.
Radiographic examination. Radiographic examination of the tails of 7 mice, 4 with nodose tails and 3
with clinically normal tails, was performed using a
dental radiograph machine (Heliodent MD; Siemens,
Munich, FRG) at 60 kV, 7 mA, using Kodak Ektaspeed film (Eastman-Kodak, Rochester, NY). Radiographs permitted evaluation of both soft and hard
tissues. The hindpaws were also examined using the
same technique.
Epidemiologic patterns. This investigation was
initiated after repeated findings of spontaneously occurring joint swelling and nodose tails in several
mouse strains. The 54 NZB/W/B mice, which had been
castrated and treated with subcutaneous injections of
either testosterone deconate or olive oil, were systematically examined for the occurrence of joint swelling.
Approximately 2 weeks after the first injection, signs
of joint swelling were apparent in several mice. Six
weeks after the onset of arthritic symptoms, the mice
were again carefully evaluated. They were then killed,
and bacterial isolates from joints and tails were obtained. Nine of 28 testosterone-treated mice and 2 of
26 control mice treated with olive oil showed mild joint
swelling, contractures, or ankylosis in 1 or both hindpaws. Two of these 11 arthritic mice also had nodose
changes of the tail (Table 1). Eight testosteronetreated mice (4 of which also had signs of arthritis)
displayed sores on their backs, whereas those not
treated with testosterone did not exhibit any sores
(Table 1). Skin lesions and sores were observed at the
time of arthritic manifestations, whereas the occurrence of nodose tails was a late phenomenon.
The frequency of arthritis was estimated retrospectively in other mouse strains, and it varied greatly
between strains. The NZB/W mice displayed a high
prevalence of arthritic joints, with >50% of the animals clinically ill, whereas the MRL-lpr/lpr mice exhibited no signs of joint swelling at all. Arthritis
Table 1. Clinical signs and occurrence of bacterial growth in the entire population of 54 NZBIWIB mice and in selected NZBIW and DBAII
treatment (n)
No. with
Castrated, testosterone (28)
Castrated, olive oil (26)
Castrated, estradiol (3)
Castrated, olive oil (2)
No treatment (6)
DBA/I, no treatment (4)
* S aureus
Staphylococcus aureus; NA
No. with
sores and
No. with
No. with
growth of
S aureus
not analyzed.
occurred in 10-20% of the overall population of the
NZW, NZB, and D B N l mice. The dominant site of
arthritis was the hindpaws (Figure IA). Nodose swelling of the tail, indicative of osteitis and skin sores, was
also commonly seen in the arthritic mice (Figure 1B).
Moreover, other cutaneous phenomena, such as hair
loss and abscesses, were occasionally found.
Bacteriologic findings. Bacterial isolates were
obtained from 1 hindpaw of each of 52 NZB/W/B mice.
The arthritic hindpaw was chosen for the determination of bacterial isolates in cases of asymmetric joint
swelling. Joints from 2 arthritic NZB/W/B mice were
used for histologic examination, without bacterial
samples being taken. S aureus was obtained from the
hindpaws of 4 of 9 arthritic mice, whereas only 1 paw
from the 43 nonarthritic NZB/W/B mice tested showed
signs of S aureus infection (Table I).
Eleven NZB/W mice were analyzed for the
No. with
occurrence of bacterial growth in blood, joints, and
tails. Six of these mice displayed clinical signs of
arthritis or nodose tails. Of these 6 mice, 4 had positive
isolates with growth of S aureus. Another NZB/W
mouse with skin sores had a positive blood culture for
S aureus, whereas blood cultures of clinically healthy
NZBIW mice were negative (Table 1). One NZB
mouse tested showed growth of S aureus in the
affected tail and paw, while 1 NZW mouse had growth
of S aureus in the inflamed testicles.
Of 4 DBA/l mice with signs of arthritis or
nodose tails, 3 had positive isolates, all containing S
aureus (Table 1). Phage typing of S aureus isolates
from 6 NZB/W mice showed the identical phage type,
6147/53/54/75/77/85/88+, in all of them. Five NZB/W/B
mice with clinical signs of arthritis and growth of S
aitreus were analyzed with respect to bacteriophage
pattern. Four of them were not typeable, while 1 had
Figure 1. Arthritis of the hindpaw (A) and nodose tail (B),in a 6-month-old N Z B W mouse.
Figure 2. a, Micrograph showing a plastic-embedded sagittal section of an arthritic ankle joint from an NZBlW mouse. There is marked
destruction of the distal part of the tibia (T).The small joints of the foot appear only mildly affected (arrows). Insets b, c, and d are shown at
higher magnification in Figures 2b, c, and d, respectively. b, Hypertrophy and proliferation of the synovial tissue, partially invading and
replacing the affected area of the tibia. Remnants of bone and cartilage are shown at the right (arrow). c, Remnants of the tibia at the talocrural
joint. Most of the cartilage has been destroyed, and the joint space is filled with connective tissue and some inflammatory cells.
Polymorphonuclear cells are present in the vicinity of the cartilage destruction (arrows). d, Detail of granulation tissue, showing loose, highly
cellular, and well-vascularized areas mingled with more dense, organized connective tissue. Most of the cells are macrophage- or fibroblast-like.
(Bars = 100 pm.)
phage type 6/47/53/54/75/83Al85/+, which was almost
identical to the type found with the former analyses.
Seventeen of 20 skin isolates from clinically
healthy mice displayed growth of S aureus, and bacteriophage typing showed that all 17 isolates shared
the identical phage type. Moreover, the bacteriophage
pattern was identical to that of S aureus found in the
inflamed joints, tails, and blood. This phage type was
not found in rats or in laboratory staff members.
Histopathologic results. Swollen hindpaw joints
from 3 N Z B N mice and 2 NZB/W/B mice were
examined histopathologically. Analysis of clinically
severely swollen hind limbs showed hypertrophy and
proliferation of the synovial tissue, with occasional
granuloma formation (Figure 2). The majority of the
proliferating cells in synovial tissue were macrophageand fibroblast-like. However, in certain areas, considerable numbers of polymorphonuclear cells were observed. In contrast, only a few lymphocytes were
detected. The granuloma tissue was rich in polymorphonuclear cells, and focal gathering of cocci was
There was total destruction of both cartilage
and subchondral bone in certain areas of the hindpaws, and the presence of amorphous fibrotic tissue in
the joint space was noted (Figure 2). Some of the
surviving chondrocytes were pyknotic, whereas others appeared large, with signs of cell division. In other
joint samples, clinically characterized by mild joint
swelling and concomitant partial ankylosis, the histopathologic picture of destruction was less prominent.
In these sections, only low-grade synovitis with
patchy destruction of cartilage was seen. However,
even in these sections, the presence of fibrotic tissue
inside and outside the joint space was apparent. Histologic analysis of 3 nodose tails showed a uniform
picture of osteitis, with bone loss and invading fibrotic
Radiographic features. Radiographic examination revealed signs of soft tissue swelling, osteopenia,
and bone loss, with partial fragmentation of the bone
close to the joint space in the nodose tails of all 4
N Z B N mice studied (Figure 3). A total disappearance
of the joint space had occurred in some of those areas,
resulting in dislocation of the bones and a shortening
of the tail. Marked soft tissue swelling surrounded the
affectedjoints. Radiographs of the tails of the 3 control
mice did not show any abnormalities. The radiographic resolution was not high enough to permit
analysis of the hindpaws.
Figure 3. Radiograph of the tail of an NZBW mouse, showing
moderate (arrow)and advanced bony erosions. Note the marked
soft tissue swelling surrounding the affected areas.
This is the first description of spontaneously
occurring bacterial arthritis in mice. Spontaneous
staphylococcal arthritis in animals was previously
observed solely in birds, such as turkeys, broiler
breeders, and layers (7). The route by which the S
aureus entered the joints remains to be identified. At
least 4 potential routes of infection should be considered: through wounds, through subcutaneous injections, through contaminated injection solutions, or
through contamination during the castration procedure, The latter possibility is unlikely because of the
time lapse of 2 months between castration and the
occurrence of arthritis. Since arthritis occurred in both
testosterone-treated and olive oil-treated mice, contamination by subcutaneous injection also seems improbable. Moreover, bacterial growth was not detectable in the testosterone deconate, olive oil, or estradiol
solutions, and infectious arthritis was also found in
mice that did not receive subcutaneous injections (i.e.,
DBAl1, NZW, and NZB mice).
Interestingly, both skin wounds and arthritis
were more common in the testosterone-treated group
compared with controls. We believe that the administration of testosterone induced a more aggressive
behavior pattern in the mice, leading to increased
biting with subsequent wound formation and a hematogenic spread of bacteria to the hindpaw joints and
tail. The finding of S aureus in a blood sample from 1
mouse with skin wounds further supports the notion of
a hematogenic spread of the infection.
S aureus were the only bacteria identified in the
isolates. The phage types were almost identical when
the 6 NZB/W mice and the typeable NZB/W/B mouse
were compared. Since the bacteriophage typings were
performed on different occasions, small differences in
bacteriophage pattern may occur, even with the same
bacterial strain (6). We therefore consider all the
typeable S aureus strains to be identical. The fact that
all skin isolates proved to contain the same S aureus
strain indicates that this strain can colonize in the skin
of mice, and that this is indeed a prerequisite for the
further pathologic events observed. The frequent
finding of S aureus on the skin of clinically nonarthritic
MRL-lprllpr mice indicates that genetic host factor(s)
may predispose to the development of infectious arthritis. Notably, 4 S aureus isolates, obtained from
swollen joints, were not typeable, indicating that an
additional strain(s) might have induced bacterial arthritis in the mice.
The histopathologic features of the arthritic
joints and nodose tails varied in severity, but all
affected animals showed signs of synovial hypertrophy, cartilage destruction, and reactive fibrosis. A
similar picture has been described in human infectious
arthritis (8).
Studies are being conducted to further characterize the S aureus strain found in the inflamed joints.
Preliminary results indicate that the reinjection of this
strain into healthy mice induces a high frequency of
bacterial arthritis and tail osteitis. Efforts will be made
to establish a murine model of staphylococcal arthritis,
the most prevalent nongonococcal bacterial joint dis-
ease in humans, and to use such a model to study
pathogenic mechanisms governing the host-bacterial
relationship, as well as to evaluate potential treatment
Acknowledgments. We thank Drs. Lena Lind, Ake
Brandberg, Staffan Seeberg, and Bertil Kaiser, and Leif
Larsson and Karin Eriksson, for help with the bacteriophage
typing and for valuable discussions.
1 . Goldenberg DL, Reed JI: Bacterial arthritis. N Engl J
Med 312:764-771, 1985
2. Boetner AG, Binder M, Bille-Hansen V: Streptococcus
suis infections in Danish pigs and experimental infection
with streptococcus suis serotype 7. Acta Pathol Microbiol
Immunol Scand [B] 95:233-239, 1987
3. Feinstein RE, Eld K: Naturally occurring erysipelas in
rats. Lab Anim 23:256-260, 1989
4. Heard DJ, Jacobson ER, Clemmons RE, Campbell GA:
Bacteremia and septic arthritis in a West African dwarf
crocodile. J Am Vet Med Assoc 192:1453-1454, 1988
5. Kasari TR, Marquis H , Scanlan CM: Septic arthritis and
osteomyelitis in a bovine digit: a mixed infection of
Actinomyces pyogenes and Fusobacterium necrophorum. Cornell Vet 78:215-219, 1987
6. Williams REO, Rippon JE: Bacteriophage typing of
Staphylococcus aureus. J Hyg 50:320-353, 1952
7. Morris MP, Fletcher OJ: Diagnostic summary of 1986
turkey, broiler breeder, and layer necropsy cases at The
University of Georgia. Avian Dis 32:391-403, 1988
8. Keefer CS, Parker F Jr, Myers WK: Histologic changes
in the knee joint in various infections. Arch Pathol 18:
199-215, 1934
Без категории
Размер файла
739 Кб
spontaneous, osteitis, mice, arthritis, staphylococcus, outbreaks
Пожаловаться на содержимое документа