Generation of multinucleated osteoclast-like cells from canine bone marrow: Effects of canine distemper virus

Bone Vol. 17, No. 1 July 1995:47-55

ELSEVIER

Generation of Multinucleated Osteoclast-like Cells from Canine Bone Marrow: Effects of Canine Distemper Virus A. P. MEE, 1 C. MAY, 2 D. BENNETT, 2 and P. T. SHARPE 3 i Bone Disease Research Centre, Department of Medicine, Manchester Royal Infirmary, University of Manchester, Manchester M13 9WL, UK 2 Department of Veterinary Clinical Science, Small Animal Hospital, University of Liverpool, Liverpool L7 7EX, UK 3 Department of Craniofacial Development, UMDS, Guy's Hospital, London Bridge, London SE1 9RT, UK

bone marrow culture systems have been established to examine the biology of "osteoclast-like" cells in a variety of species, including mice, 31 humans, la cats, 32 and dogs. 5.29 The cells have been termed osteoclast-like on the basis of multinuclearity, tartrate resistant acid phosphatase (TRAP) activity, ultrastructural appearances, response to various osteotropic factors, and their ability to form resorption pits on dentine or bone slices. Paget's disease is a chronic focal skeletal disorder in which the primary defect appears to occur in the osteoclast. 15 Increased numbers of abnormally large osteoclasts cause excessive bone resorption and this is followed by irregular, uncontrolled bone formation. 15 The exact cause of the disease is still unknown, but evidence is growing to suggest that a member of the Paramyxovirus family might be responsible. Pagetic osteoclasts contain viral nucleocapsid-like inclusion bodies9'12'22"26 and immunocytochemical and molecular studies have implicated measles virus, 3"27"28"30 respiratory syncytial virus, 23"25"3° simian virus 5, 3 parainfluenza virus type 3, 3 and canine distemper virus (CDV). 7"1°'11 Recently, we have shown that CDV can infect bone cells in the natural host in vivo2° and that CDV might be involved in the canine bone disorder metaphyseal osteopathy (MO), in which there is also an increase in the number of osteoclasts. 2~ In view of the findings implicating CDV in both Paget's disease and MO, we have examined the effects of CDV on canine multinucleated osteoclast-like cell formation in vitro. The results show that CDV causes a dose-dependent increase in number and size of MNCs formed in culture, and hence support the hypothesis that the virus might be involved in the osteoclast abnormalities that are characteristic of Paget's disease and MO.

Recent evidence has implicated canine distemper virus (CDV) as a possible aetiologic agent in Paget's disease of bone and the canine bone disorder, metaphyseal osteopathy. We have therefore examined the effects of CDV on the formation of multinucleated osteoclast-like cells in cultures of canine bone marrow mononuclear cells. Marrow cells from a distemper-infected dog and from five uninfected dogs were cultured in the presence of lct,25-(OH) 2 vitamin D 3 and the number of tartrate resistant acid phosphatase positive multinucleated cells (MNCs) was determined. The presence of caicitonin (CT) receptors was confirmed by autoradiography with 12SI-labeled human CT. Cultures from the distemperinfected dog contained a higher level of MNCs than those from the normal dogs. The in vitro addition of CDV to the cultures from all the dogs produced a dose-dependent increase in the number of MNCs, and an increase in size of these cells in the cultures from the infected dog. Cells infected with CDV were hyperrespousive to lcq25-(OH) 2 vitamin D 3. The presence of the virus in the relevant samples was confirmed using molecular techniques. In situ hybridization studies also revealed a significant increase in the level of infection following in vitro addition of the virus to the culture from the distemper-infected dog, suggesting that further infection had taken place. Resorption pits were formed on bone slices, although the number of pits was not significantly altered by viral infection. These results support the hypothesis that CDV might be involved in the skeletal abnormalities seen in Paget's disease and metaphyseai osteopathy and suggest that canine marrow culture systems should prove useful to examine the disease processes further. (Bone 17:47-55;

Materials and Methods

1995)

Case Details

Key Words: Canine distemper virus; Osteoclast-like cells; Bone marrow cultures; lc~,25-dihydroxyvitamin D3; Paget's disease.

Samples were obtained from a 24-week-old German Shepherd dog with distemper, and from five uninfected (hence, referred to as normal) dogs: a 9-week-old Gordon Setter, a 16-week-old crossbreed, a 19-week-old greyhound, and two 6-month-old beagles. The German Shepherd dog was diagnosed as having distemper on the basis of clinical signs of respiratory, gastrointesfinal, and neurological disease. 2 The presence of distemper was later confirmed by in situ hybridization and polymerase chain reaction experiments (see below).

Introduction Osteoclasts are multinucleated cells (MNCs) that are responsible for bone resorption. They are formed by fusion of haematopoietic mononuclear precursors in the bone marrow. 17,19.24 While the exact identity of the precursor cell(s) remains a mystery,

Bone Marrow Cultures

Address for correspondence and reprints: Dr. Andrew P. Mee, Bone Disease Research Centre, UniversityDepartmentof Medicine, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK. © 1995 by Elsevier Science Inc.

Following euthanasia, the humeri, femora, and tibiae were excised and the epiphyses were removed. Bone marrow cells were 47

8756-3282/95/$9.50 SSDI 8756-3282(95)00140-9

48

Mee et al. Effects of canine distemper virus

flushed from the diaphyses with a 1.1% EDTA/0.9% NaC1 solution and then an equal volume of culture medium was added [Eagles minimum essential medium supplemented with 20% heat inactivated foetal calf serum, 2 mM glutamine, 100 U/mL penicillin, and 100 Ixg/mL streptomycin (GIBCO BRL, UK)]. The cell suspension was then layered on a ficoll-hypaque density gradient (Histopaque-1077, Sigma Chemical Company, UK) and centrifuged at 500g for 10 min at room temperature. The mononuclear cells were carefully removed and resuspended to l0 6 cells/mL in fresh culture medium. The cells were incubated in the presence or absence of varying concentrations (see Results section) of la,25-(OH) 2 vitamin D3 and the Onderstepoort strain of CDV in a humidified atmosphere of 5% CO 2 and 95% air at 37°C. Cells were grown in 24 well tissue culture plates (Falcon Multiwell, Becton Dickinson Labware, USA) (1 mL/well), 8 well tissue culture slides (Labtek chamber/slides, Miles Scientific, USA), or 25 cm2 culture flasks (Costar Corporation, USA). Cultures were replenished the following day, and then every three days, by replacing half the medium with fresh medium and hx,25-(OH)2 vitamin D3.

Demonstration of TRAP Activity After 8 days incubation, the cells were washed twice with Dulbecco's Ca 2+- and Mg2÷-free phosphate buffered saline (PBS) (GIBCO) and then fixed with acetone-formaldehyde (Sigma). The cells were then stained for the presence of TRAP activity using the acid phosphatase, leukocyte kit (Sigma) according to the manufacturer's instructions. Cells were counted in five high power fields (/>100 cells/ field), in duplicate samples from each dog, and the final results were expressed as an average percentage. Cells were classified as TRAP negative ( - ) mononuclear cells, TRAP positive ( + ) mononuclear cells, small TRAP + MNCs (3-5 nuclei), or large TRAP + MNCs (>5 nuclei). There were a few fibroblasts and T R A P - MNCs, but they were randomly distributed and were not present in all fields examined, so they were not counted. The size of the large MNCs was estimated in two ways. The nuclei in fifty cells in a random field were counted and the mean calculated for each sample. Also, the total surface area covered by the MNCs was compared by counting the number of squares on a grid that were occupied by MNCs in five high power fields and expressing this as an average percentage.

Demonstration of Calcitonin Receptors After 8 days incubation, the cells were washed twice with PBS (GIBCO) and then incubated with 0.5nM [125I] human CT (Amersham International, UK) in serum free culture medium conraining 0.1% BSA for 90 min at room temperature. Control samples were incubated with a 1000-fold excess of unlabeled CT (Sigma). The samples were then rinsed twice with PBS, and fixed and stained for the presence of TRAP (as above). The slides were dipped in Ilford K5 emulsion diluted 1:1 with distilled water, exposed for 14 days at 4°(2, developed as previously described, 25 and briefly counterstained with haematoxylin. Cells were counted as above and scored as C T - mononuclear cells, CT + mononuclear cells, C T - MNCs, or CT + MNCs. Only those cells that were strongly positive were scored as such.

Measurement of Bone Resorption Slices of devitalized bovine cortical bone (5 mm × 5 mm × 0.5 ram) were prepared using a low speed Isomet saw (Buehler

Bone Vol. 17, No. 1 July 1995:47-55 Ltd., IL, USA). Slices were added to the wells prior to the addition of cells. Following incubation for two weeks, the cells were removed by gently rubbing the surface. The slices were then briefly counterstained with toluidine blue and examined by conventional light microscopy for the presence of resorption pits.

In situ Hybridization Sense and antisense riboprobes were prepared to the CDV-N gene, using 35S-UTP, as described previously. 1° A probe to 13-actin served as a positive control. In situ hybridizations were carried out on the cells cultured in the multiwell slides using a modified method of that previously described by Mee et al. 2o Briefly, . after fixation with acetoneformaldehyde (Sigma), the cells were rehydrated and then permeabilized with 0.2M HC1 for 20 min at room temperature. Acetic anhydride (0.25%) was used to reduce nonspecific background. Control slides were treated with 100 Ixg/mL RNase for 30 min at 37°C. The hybridization reaction was carried out overnight at 50°C in 50% formamide and 0.3M NaC1, in a humidified container. Post-hybridization, the cells were washed at 50°C and treated with 20 ~g/ml RNase A to remove unbound probe. The slides were dipped in a 1:1 dilution of Ilford K5 emulsion and distilled water and exposed for seven days. The samples were then developed, counterstained with haematoxylin and eosin, and viewed by light- and dark-field microscopy. Cells were counted in five high power fields from each sample and the numbers of each cell type (mononuclear, small multinucleated, and large multinucleated) that showed positive hybridization were expressed as a percentage of the total number of each cell type.

Polymerase Chain Reaction All molecular biology enzymes and reagents were obtained from Boehringer Mannheim or Promega UK. All reagents were aliquoted and discarded following single usage, to prevent contamination.13 The methods used were a modification of those described previously.l~ Cells from the culture flasks were treated with 1X trypsin/EDTA (GIBCO) for 5 min at 37°C and then centrifuged at 1000g for 5 min. Total RNA was then extracted from each sample. Briefly, the pellet of cells was homogenized in a solution of 4.0M guanidinium thiocyanate, 0.5% sarcosine, 0.1M [3-mercaptoethanol, and 25ram sodium citrate. The RNA was then isolated in acidified phenol/chloroforrrgisoamyl alcohol, precipitated at - 2 0 ° C in an equal volume of isopropanol, and resuspended in ethanol and RNase free water. Single-strand cDNA was synthesized from approximately 20 Ixg of total RNA using 10 U of avian reverse transcriptase, 1 txg random hexamers, and 10 mM each of dATP, dCTP, dGTP, and dTTP at 42°C for 1 h. Another 10 U of reverse transcriptase was then added prior to incubation for another hour. The reaction was then terminated by heating to 95°C and cooling on ice. Specific primers to positions 450-980 and I231-1464bp of the CDV-nucleocapsid (N) gene were manufactured as described previously. ~1 The expected fragment sizes were 556 and 249bp, respectively. Aliquots (5 IzL) from the reverse transcription reaction were then amplified in 20 mM Tris-HC1 pH 8.0, 2.5 U Taq DNA polymerase, 1.5 mM MgC12, 10 pmol of each primer, and 100mM each of dATP, dCTP, dGTP, and dTrP, in a final volume of 100 I~L. Forty cycles of denaturation (94°C for 50 sec), annealing (55°C for 60 sec), and extension (72°C for 60 sec) were carried out in a Techne cycler. After a final extension

Bone Vol. 17, No. 1 July 1995:47-55

Mee et al. Effects of canine distemper virus

49

IP e

4 (a)

(d)

(b)

(c)

(e)

(g) Figure 1. Tartrate resistant acid phosphatase and calcitonin receptor positive multinucleated cell formation. Marrow mononuclear cells were cultured in the presence of lot,25-(OH)2 vitamin D 3 for eight days and were then stained for TRAP activity as described in the methods. TRAP activity is seen as a brown stain. (a) MNC showing diffuse granular staining pattern. Magnification x 125. (b) MNC showing more intense staining pattern. Magnification × 125. (c) Very large MNC formed by apparent fusion of several smaller MNCs. Magnification x60. (d) Light field and (e) dark field views of MNC showing TRAP staining and the presence of CT receptors (silver grains following autoradiography with 125I-labeled human CT). Magnification ×250. (f) Light field and (g) dark field views of MNC cultured in the presence of ~2~I-labeled human CT and a 1000 fold excess of CT (background level of silver grains following autoradiography). Magnification x 250. Counterstained with haematoxylin.

50

Mee et al. Effects of canine distemper virus

reaction at 72°C for 10 min, 30 p,L aliquots from each reaction were run in a 2.5% agarose gel containing ethidium bromide, in 1X Tris-borate-EDTA buffer. The DNA was visualized by ultraviolet light transillumination. Results All results from the uninfected dogs are represented as an average of the five dogs. Differences between samples were compared using the unpaired Student's t-test.

Multinucleated Cell Formation, TRAP Activity, and CT Receptors TRAP-positive MNCs were formed at all concentrations of lot,25-(OH)2 vitamin D 3 ( 1 0 - 7 - 1 0 - H M ) . The intensity of TRAP staining was variable from cell to cell. Some cells showed a diffuse granular staining pattern, while others were more intensely stained (Figures l a and b). Some very large MNCs were also formed (Figure lc). The cells were shown to possess CT receptors after autoradiography with 125I-labeled human CT [26 - 12% of the mononuclear cells were positive, and 68 +- 8% of the MNCs were positive at 10-8M lot,25-(OH) 2 vitamin D3] (Figures ld and e). Silver grains were not seen following the addition of excess unlabeled CT (Figures If and g). Maximum stimulation of MNC formation was seen, in both distemperinfected and normal samples, at 10-8M lot,25-(OH) 2 vitamin D 3 (Figure 2). Cultures from the distemper-infected dog contained a significantly higher percentage of MNCs (both small and large) than those from normal dogs (Figure 2). Even at low concentrations of lct,25-(OH) 2 vitamin 03, there were significantly (p = 0.001) more MNCs in the cultures from the distemper-infected dog than the number seen in the cultures from the normal dogs. Addition of the Onderstepoort strain of CDV (10105 pfu/mL) to the cultures caused a dose-dependent (up to l03 pfu/mL) significant increase in the percentage of MNCs formed, with a corresponding decrease in the number of mononuclear cells (Figure 3). At higher doses 004-105 pfu/mL) many of the cells were killed, and hence the relative decrease in MNCs seen at these concentrations. The dose-dependent effect differed slightly between the samples from infected and normal dogs: in the cultures from normal dogs, the increase was seen in both the small and large MNC populations (Figure 3a), whereas in the cultures from the distemper-infected dog, only the small MNCs were increased in number (Figure 3b). However, the average number of nuclei and the surface area of the large MNCs was increased in these cultures (Table 1 and Figure 4).

Bone Vol. 17, No. 1 July 1995:47-55 the original samples from the normal dogs (Figures 6c and d). However, positive hybridization was seen in the samples from these dogs that had been infected in vitro with the Onderstepoort strain of CDV (Figures 6e and f). No hybridization was seen in any of the samples that had been RNase treated (not shown), and positive hybridization was seen in all the samples with the probe to 13-actin (not shown). The average percentage positivity for each cell type in each sample is shown in Table 2. The original samples from the distemperinfected dog showed a similar level of infection to those from the normal dogs following in vitro infection. However, there was more variation in the levels of infection in the latter samples. Also, the number of positive mononuclear cells in the samples from the normal dogs was greater. The samples from the distemper-infected dog that were additionally infected with CDV in vitro showed the highest level of infection in all cell types. There was no obvious difference between the levels of infection in small and large MNCs in any of the samples studied.

Polymerase Chain Reaction Bands of the expected size for the CDV-N gene 1z,2~ were seen in the samples from the distemper-infected dog using both sets of primers (Figure 7). No bands were seen with either set of primers in the samples from the normal dogs (Figure 7). (a) 100 -

Mononuclear Cells TRAP+ Mononuclear Cells TRAP+ (_<5Nuclei) TRAP+ (>5 Nuclei)

80 m

Z

4o

o 0 Vitamin

-7 -8 -9 -10 -11 D Concentration (M x 10 x )

tb) 100 1

• • • []

Bone Resorption m

Resorption pits were formed in cultures containing 10 - 8 M lot,25-(OH)2 vitamin D3 (Figure $). However, there was only a small number of pits on each slice and, although the number of pits was slightly increased in the distemper-infected cultures, there was no statistically significant difference between normal and infected (either in vivo or in vitro) samples.

Z

40

o 0

In situ Hybridization Positive hybridization, using sense and antisense probes to the CDV-N gene, was seen in both the original and the in vitro CDV-infected cultures from the distemper-infected dog (Figures 6a and b). The levels of hybridization were similar with both the sense and antisense probes. No hybridization was seen in any of

Mononuclear Cells TRAP+ Mononuclear Cells TRAP+ (-<5Nuclei) TRAP+ (>5 Nuclei)

Vitamin

-7

-8

-9

D Concentration

-10

-11

(M x 10 x )

Figure 2. Graphs showing the effect of varying dose of 1~,25-(OH) 2 vitamin D3 on cell populations. Cell numbers are expressed as the average percentage of the total cells in five high power fields from each sample. Bars represent standard errors. (a) Normal dogs. (b) Distemperinfected dog. Note that even at low concentrations of ltx,25-(OH)2 vitamin D3, more MNCs were formed than the maximum seen in (a).

Bone Vol. 17, No. 1 July 1995:47-55

Mee et al. Effects of canine distemper virus (a)

(c)

60 50 40,

,,o

51

50-

I i MononuclearCeUs TRAP*MononuelearCells TRAP+(-<5Nuclei) TRAP+(>5Nuclei)

~ i MononuclearCells TRAP+MononucIearCells

i

TRAP+(>5Nuclei)

30"

E Z

z

20' 10' 0"

0

1

2

3

4

0

5

1

2

3

4

5

CDV Concentration (pfu x 10 x )

x CDV Concentration (pfu x 10 )

(b)

(dl

o..

TRAP+(>5Nuclei)

[] TRAP+(>5Nuclei)

40 ~

50

i i

t

~,~ 30 40

Z

Z

10

100

•

0•

0

1

2

3

4

20

5

0

CDV Concentration (pfu x 10 x )

1

2

3

4

5

CDV Concentration (pfu x 10 x )

Figure 3. Graphs showing the effect on multinucleated cell formation of the in vitro addition of CDV to the cultures. CDV was added to the cells on the first day of culture. The cells were cultured in the presence of 10-aM 1a,25-(OH)2 vitamin D 3. Cell numbers are expressed as the average percentage of the total cells in five high power fields from each sample. Bars represent standard errors. (a) and (b) Normal dogs. Note a dose dependent (up to l03 pfu/mL) increase in small and large MNCs. *p = 0.02, **p = 0.006, ***p = 0.001, tp = 0.05, t t p = 0.002. (c) and (d) Distemper-infected dog. Note a dose dependent (up to 103 pfu/mL) increase in the number of small MNCs. *p = 0.01, **p = 0.001, ***p = <0.001, • ***p = 0.005. Note also that the total number of MNCs in the cultures from the normal dogs infected with 10 3 pfu/mL CDV are similar (~65%) to the number in the original cultures from the distemper-infected dog. Discussion

M a r r o w m o n o n u c l e a r cells f r o m a dog with d i s t e m p e r a n d f r o m five n o r m a l dogs were cultured in v a r y i n g c o n c e h t r a t i o n s o f l a , 2 5 - ( O H ) 2 v i t a m i n D 3. Tartrate resistant acid p h o s p h a t a s e ( T R A P ) - p o s i t i v e M N C s c o n t a i n i n g C T receptors were f o r m e d at all concentrations o f let,25-(OH) 2 v i t a m i n D 3 u s e d . T h e variations in T R A P staining patterns o b s e r v e d were similar to those reported in p r e v i o u s m a r r o w culture e x p e r i m e n t s 5"36 a n d in isolated osteoclasts.1 T h e m a x i m u m n u m b e r o f M N C s w a s s e e n at a concentration o f 1 0 - S M let,25-(OH)2 v i t a m i n D3, and h e n c e all in vitro viral effects were e x a m i n e d at this concentration.

P r e v i o u s l y , w e h a v e s h o w n , u s i n g m o l e c u l a r t e c h n i q u e s , that canine d i s t e m p e r virus ( C D V ) c a n infect h u m a n a n d c a n i n e bone cells in v i v o , 7.10 ' •11•20' 21 s u g g e s t i n g that the virus m i g h t be involved in the p a t h o g e n e s i s o f h u m a n P a g e t ' s d i s e a s e a n d the canine b o n e disorder, m e t a p h y s e a l o s t e o p a t h y . W e report here the effect o f C D V ( f r o m both natural a n d in vitro infection) on m u l t i n u c l e a t e d osteoclast-like cell f o r m a t i o n in c a n i n e b o n e m a r row cultures.

Table 1. Effect of in vitro addition of CDV on the average number of nuclei in the large multinucleated cells. The nuclei were counted in fifty large multinucleated cells in a random field from each sample. Values represent the mean -+ standard deviation. *p values compare infected samples from the normal dogs with the basal number (7.8); tp values compare additionally infected samples from the naturally infected dog with the basal number (9.3); Sp values compare samples from infected and normal dogs CDV dose (pfu/mL) 103

S ample

0

10

102

104

Normal

7.8 - 1.9

9.6 -+ 3.8*

12.0 -+ 4.4**

16.1 +- 6.5**

10,4 _+ 3.7**

7.4 -+ 1.3

Infected

9.3 -+ 2.1

10.6 -+ 2.9t

15.1 +- 6 . 4 t t

20.5 - 9 . 0 t t

12,9 -+ 4 . 6 t t

7.8 -+ 1.Ttt

*p = 0.02; **p = <0.001; tp = 0.003; t t p = <0.001; Sp = <0.001; ~$p = 0.006; ~:~p = 0.007; $~t~c~p = 0.004.

105

52

Mee et al. Effects of canine distemper virus

Bone Vol. 17, No. 1 July 1995:47-55

60

I

k) ~ 20 ,< I0 0

. 0

. 1

.

. 2

3

4

5

CDV Concentration (pfu x 10 x ) Figure 4. Graph showing the effect on size of large multinucleated cells of the in vitro addition of CDV to the cultures from the distemperinfected dog. The cells were cultured in the presence of 10 .8 M la,25(OH)2 vitamin D 3. Note a dose-dependent increase in area of the large MNCs expressed as an average percentage. *p = <0.001. In situ hybridization and polymerase chain reaction studies confirmed the presence of CDV in the original cells (i.e., prior to the in vitro addition of virus) from the dog with clinical signs of distemper. Cells cultured from the five normal dogs were shown to be initially free of virus using the same techniques. Following in vitro addition of CDV to the cultures, the cells from all dogs were shown to have been infected by the virus. The in

(a)

situ hybridization results also provided a method of determining the number of infected cells in each sample. There was no obvious difference between the levels of infection in the original cultures from the distemper-infected dog and those from the normal dogs following in vitro addition of 103 pfu/mL CDV. This was reflected in the similar total numbers of MNCs (approximately 65%) seen in these samples. Following in vitro infection of the cultures from the distemper-infected dog, the number of positive cells (both mono- and multinucleated) was significantly increased, suggesting that additional virus had infected these cells. Owing to the sensitivity of the technique and hence, the possibility of contamination, the polymerase chain reaction was not used to confirm the presence of virus within the cells from the normal dogs following in vitro infection. However, the in situ hybridization findings were conclusive, and, hence, the effects seen are presumed to be the result of viral infection of the cells. Previous reports of canine osteoclast-like cells derived from marrow cultures have either failed to show bone resorption, 5 or shown that the cells could attach to bone spicules. 29 However, in the latter report, resorption was not convincingly demonstrated. In neither of these reports was lo~,25-(OH)2 vitamin D 3 added to the cultures. Bone resorption was assayed in our cultures using devitalized bovine cortical bone slices. A small number of resorption pits was seen with cells cultured in 10 - 8 M let,25(OH)2 vitamin D 3, suggesting that, while most of the ceils did

(b)

Figure 5. Formation of resorption pits. The cells were cultured on devitalized bovine cortical bone slices in the presence or absence of 10 -8 M la,25-(OH)2 vitamin D 3 for two weeks. The cells were removed and the slices were briefly counterstained with toluidine blue. (a) Resorption pit formed on bone slice. (b) Bone slice from control sample cultured without lct,25-(OH) 2 vitamin D 3. No resorption pits are visible. Magnification × 125.

Bone Vol. 17, No. 1 July 1995:47-55

Mee et al. Effects of canine distemper virus

(a)

(b)

(c)

(d)

(e)

(f)

53

Figure 6. In situ hybridization results. Probes used were to the nucleocapsid gene of CDV. Figures show results with the sense probe [similar results were obtained with the antisense probe (not shown)]. (a) Light field and (b) dark field views showing positive hybridization in the initial s~nple from the distemper-infected dog. Similar results were obtained following further addition of the Onderstepoort strain of CDV (103 pfu/mL) to the cultures from this dog (not shown), although more of the cells were positive (see Table 2). (c) Light field and (d) dark field views showing no hybridization (background level of silver grains) in the initial sample from the normal crossbreed dog. (e) Light field and (f) dark field views showing positive hybridization following addition of the Onderstepoort strain of CDV (103 pfu/mL) to the cultures from the crossbreed dog. Magnification x 125. Counterstained with haematoxylin and eosin. not resorb bone, at least some of the cells were behaving as authentic osteoclasts. It is, however, possible that a small number of preformed osteoclasts might have been present in the original marrow samples. Failure to resorb bone appears to be

one of the major problems with osteoclast-like cells cultured from marrow cells, and it has been suggested that the cells behave more like macrophage polykaryons (recently reviewed by Zaidi et al.33). Another possibility is that the limits of the culture

54

Mee et al. Effects of canine distemper virus

Bone Vol. 17, No. 1 July 1995:47-55

Table 2. Comparison of the numbers of distemper-positive cells in each culture. Cells were counted in five high power fields from each sample and the number of positive cells was expressed as an average percentage. TRAP staining was not carded out on these samples. The Onderstepoort strain of CDV (103 pfu/mL) was added to samples from both infected and normal dogs ( + CDV). Results are expressed as mean --- standard deviation, p values compare the additionally infected samples from the naturally infected dog with the original samples from this dog

Sample

Mononuclear ceils

Small multinucleated cells (3-5 nuclei)

Large multinucleated ceils (>5 nuclei)

Normal Normal + CDV Infected Infected + CDV

0 6.9 ± 1.8 3.7 --- 1.3 10.9 ± 1.9"

0 49.1 ± 5.7 47.4 ± 4.4 64.4 ± 5.3**

0 47.3 ± 7.4 46.4 • 3.5 66.0 --- 4.9*

*p < 0.001; **p = 0.001. system prevent resorption. The cells might be too immature to resorb bone (after two weeks, the cultures become rapidly overgrown with fibroblasts; hence, it was not possible to assay resorption after longer periods of time), or more likely, the presence of other factors, particularly those derived from osteoblasts, 6 may be necessary. The presence of CT receptors in these canine cells would suggest that they are osteoclast-like and that some other factor is needed to produce bone resorption. Despite these drawbacks, there was a slight increase in numbers of resorption pits following distemper infection, although, owing to the small numbers seen, this increase was not statistically significant. While the effects of CDV on bone resorption could not be satisfactorily determined in these cultures, there were obvious changes in the cell populations as a result of viral infection. As in previous reports, 5"1a'29"31"32 only cells with three or more nuclei were classified as multinucleated. There appeared to be a distinct population of small MNCs in all cultures (3-5 nuclei) and hence, these were counted as a separate group. Comparative cultures from the dog with natural distemper infection contained a significantly higher level of both small and large MNCs than those from normal dogs. Addition of virus to cultures from the normal dogs produced a dose-dependent significant increase in

M

1

2

3

4

5

6

Figure 7. Polymerase chain reaction results. Photograph of ethidium bromide stained agarose gel viewed by ultraviolet light trans-illumination. M-size marker (bp), pSP72 cut with HpalI. Lanes 1 and 2 are the original samples from the distemper-infected dog. Lanes 3 and 4 are from the same dog following addition of CDV to the cultures. Lane 5 and 6 are the original samples from the normal crossbreed dog. Expected band sizes for CDV were 249 (lanes 1, 3, and 5) and 556 (lanes 2, 4, and 6). Note bands confirming the presence of distemper in both samples from the distemper-infected dog (arrows). No bands are visible with either set of primers in the samples from the normal dog.

the MNC population. This increase was seen in both small and large MNCs, particularly in the former. Addition of virus to the cultures from the dog with distemper produced a further increase in the number of MNCs, although this increase was only seen in the small MNC population. However, the large MNCs did increase in size. The increase in number and size of cells was only seen up to a dose of 103 pfu/mL. At higher doses, cell clumping and death occurred, and hence, the relative number of MNCs was reduced. This suggests that the cells were incapable of supporting infection of high levels of virus. This effect was much less noticeable in the cultures from the distemper-infected dog, suggesting that previous exposure of the cells to the virus rendered them capable of surviving higher levels of additional infection. It is possible that breed and age differences might alter the numbers of MNCs in canine marrow cultures. However, the uninfected dogs used in this study ranged in age from nine weeks to six months, and were of several different breeds. Although none of the uninfected dogs were German Shepherds, it seems unlikely that the age and breed of the distemper-infected dog could account for the differences seen. Paget's disease is characterized by an increased number of abnormally large osteoclasts, t5 and increased numbers of osteoclasts are also seen in the early stages of metaphyseal osteopathy.2Z The in vitro findings detailed here are therefore supportive of the hypothesis that CDV might be involved in these diseases. The virus appears to enhance the production of MNCs, in a dose-dependent manner, by fusion of mononuclear precursors and also by fusion of several MNCs into one. The cultures from the dog with natural distemper infection were particularly interesting. In the original cultures from this dog, the cells appeared to be hyperresponsive to let,25-(OH) 2 vitamin D 3, as, even at low doses ( 1 0 - lo M) of lot,25-(OH) 2 vitamin D 3, the number of MNCs was greater than the maximum number seen in the cultures from normal dogs. Marrow cells in long-term cultures from Paget's patients have also been shown to be hyperresponsive to lot,25-(OH) 2 v i t a m i n D3,16 and CFU-GM (proposed osteoclast precursor) cells from these patients are similarly hyperresponsive. 8 It has been postulated that Paget s disease might be caused by repeated infection with the same, or a different, Paramyxovirus. 14 The fact that the cells from the distemper-infected dog were capable of supporting further, higher levels of CDV infection, and the significant increase in size seen in the MNCs from this dog following additional infection by CDV, would support this hypothesis, and this phenomenon merits further investigation. In summary, CDV caused a dose-dependent significant increase in multinucleated osteoclast-like cell formation in canine bone marrow cultures derived from normal dogs. Higher hum-

Bone Vol. 17, No. 1 July 1995:47-55 bers of MNCs were formed in comparable cultures from a dog with distemper. The addition of further virus to the cultures from this dog again caused a dose-dependent increase in the number of MNCs and also significantly increased the average size of the MNCs. These preliminary results therefore support the hypothesis that CDV is involved in both human Paget's disease and canine metaphyseal osteopathy, and it is hoped that the culture system described will prove useful as an in vitro model for examining this hypothesis further.

Acknowledgments: We would like to thank Dr. M. E. Hayes (Department of Medicine, University of Manchester, UK) for the vitamin D, Dr. W. S. K. Chalmers (Intervet, UK) for the CDV and the samples from the beagles, Dr. R. J. Cooper (Department of Pathological Sciences, University of Manchester, UK) for the loan of the Isomer saw, and Dr. S. L. Cosby (Queen's University, Belfast) for the CDV probe. This study was funded by the National Association for the Relief of Paget's Disease and the Salford Paget's Disease Appeal.

References 1. Addison, W. C. Enzyme histochomical properties of kitten osteoclasts in bone implant preparations. Histocbem J 10:64~656; 1978. 2. Appei, M. J. Canifie distemper vires. Appel, M. J., ed. Vires Infections of Vertebrates. Amsterdam: Elsevier; 1987;133-159. 3. Basle, M. F., Russell, W. C., Goswami, K. K. A., Rebel, A., Giraudon, P., Wild, F., and Filmon, R. Paramyxovinls antigens in osteoclasts from Paget's bone tissue detected by mononuclear antibodies. J Gen Virol 66:2103-2110; 1985. 4. Basle, M. F., Fournier, J. G., Rozenblatt, S., Rebel, A., and Bouteille, M. Measles virus RNA detected in Paget's disease bone tissue by in situ hybridisation. J Gen Virol 67:907-913; 1986. 5. Bird, M. C., Garside, G., and Jones, H. B. Multinucleated giant cells in primary cultures derived from canine bone marrow Evidence for formation of putative osteoclasts. Cell Tissue Res 268:17-30; 1992. 6. Burger, E. H., Van der Meet, J. W. M., Nijweide, P. J. Osteoclast formation from mononuelear phagocytes: role of hone-forming cells. J Cell Binl 99: 1901-1906; 1984. 7. Cartwright, E. J., Gordon, M. T., Freemont, A. J., Anderson, D. C., and S~, P. T. Paramyxovirnses and Paget's disease. J Med Viro140:133-141; 1993. 8. Demulder, A., Takahasi, S., Singer, F. R., Hosking, D. J., and Roodman, G. D. Abnormafities in osteoclast precursors and marrow a ~ r y cells in Paget's disease. Endocrinol 133:1978-1982; 1993. 9. Gharadi, G., Lo Cascio, V., and Bonueci, E. Fine stntcture of nuclei and cytoplasm of osteoclasts in Paget's disease of bone. Histopathol 4:63-74; 1980. 10. Gordon, M T., Anderson, D. C., Sharpe, P. T. Canine distemper virus lneo alised in bone cells of palients with Paget's disease. Bone 12:195-201; 1991. 11. Gocdon, M. T., Mee, A. P., Anderson, D. C., and Sbarpe, P. T. Canine distemper virus transcripts sequenced from pagetic hone. Bone Min 19:159174; 1992. 12. Howatson, A. F. and Fomasier, V. L. Microfdaments associated with Paget's disease of bone. Comparison with nucleocapsids of measles virus and respi* ratory syncytial virus. Intervirol 18:150-159; 1982. 13. Innis, M. A., Gulfand, D. H., Sninsky, J. J., and White, T. J., eds. PCR Protocols. A Guide to Methods and Applications. New York: Academic; 1990. 14. Kahn, A. J. The viral etiology of paget's disease of hone: A new perspective. Calcif Tis Int 47:127-129; 1990.

Mee et al. Effects of canine distemper virus

55

15. Kanis, J. A. Pathophysiology and Treatment of Paget's Disease. London: Martin Dunitz; 1991. 16. Kukita, A., Chenu, C., McManns, L. M., Mundy, G. R., and Roodman, G. D. Atypical multinucleated cells form in long-term marrow cultures from patients with Paget's disease. J Clin Invest 85:1280-1286; 1990. 17. Loutit, J. F., and Nisbet, N. W. The origin of ostenelasts. Immunobiul 161: 193-203; 1982. 18. MacDonald, B. R., Takahashi, N., McManns, L. M., Holahan, J., Mundy, G. R., and Roodman, G. D. Formation of multinucleated cells which respond to osteot~pic hormones in long-term human hone marrow cultures. Endocrinol 120:2326-2333; 1987. 19. Marks, S. C. The origin of osteoclasts: Evidence, clinical implications and investigative challenges of an extra-skeletal source. J Oral Pathol 12:226-256; 1983. 20. Mee, A. P., Webber, D. M., May, C., Bennett, D., Sharpe, P. T., and Anderson, D. C. Detection of canine distemper virus in bone cells in the metaphyses of distemper-infected dogs. J Bone Min Res 7:829-834; 1992. 21. Mee, A. P., Gordon, M. T., May, C., Bennett, D., Anderson, D. C., and Sharpe, P. T. Canine distemper vires transcripts detected in the bone cells of dogs with metaphyseal osteopathy. Bone 14:59-67; 1993. 22. Mills, B. G. and Singer, F. R. Nuclear inclusions in Paget's disease of hone. Science 194:201-202; 1976. 23. Mills, B. G., Singer, F. R., Weiner, L. P., and Hulst, P. A. Immunohistological demonstration of respiratory syncytial vires antigens in Paget's disease of hone. Prec Nat Acad Sci (USA) 78:i209-1213, 1981. 24. Osdoby, P., Krukowski, M., Oursler, M. J., and Salino-Hugg, T. The origin, development and regulation of osteoclasts. BioEssays 7:30-34; 1987. 25. Pringle, C. R., Wilkie, M. L., and Ellintt, R. M. A survey of respiratory syncyfial virus and paiainflnenza virus type 3 neutralising and immunoprecipitating antibodies in relation to Paget's disease. J MealVirol 17:377-386; 1985. 26. Rebel, A., Malkami, K., and Basle, M. Anomalies nucleaires des osteoclasts de la maladie osseuse de Paget. Nonv Press Med 3:1299-1301; 1974. 27. Rebel, A., Basle, M., Pouplard, A., Kouyoumdjian, S., Filmon, R., and Lepatezour, A. Viral antigens in osteoclasts from Paget's disease of hone. Lancet 2:344-346; 1980. 28. Reddy, S. V., Singer, F. R., and Roodman, G. D. Marrow mononuclear cells from patients with Paget's disease contain a mutated form of Measles virus (MV) nucleocapsid antigen. J Bone Min Res 9(Suppl 1):S152;1994. 29. Seed, T. M., Kaspar, L. V., Domaim, F., Niiro, G. K., and LeBuis, D. A. Developmental and radiobiologic characteristics of canine multinucleated, osteoclast-like cells generated in vitro from canine bone marrow. Scanning Microsc 2:1599-1611; 1988. 30. Singer, F. R. and Mills, B. G. Evidence for a viral aetiology of paget's disease of bone. Clin Orthop Rel Res 178:245-251; 1983. 31. Takahashi, N., Yamaha, H., Yoshiki, S., Roodman, D. G., Mundy, G. R., Jones, S. J., Boyde, A., and Suda, T. Osteoclast-llke cell formation and its regulation by osteotropic hormones in mouse bone marrow cultures. Endocrinol 122:1373-1382; 1988. 32. Testa, N. G., Allen, T. D., Lajtha, L. G., Onions, D., and Jarrett, O. Genecafion of osteoclasts in vitro. J Cell Sci 47:127-137; 1981. 33. Zaidi, M., Alam, A. S. M. T., Sbankar, V. S., Bax, B. E., Bax, C. M. R., Monnga, B. S., Bevis, P. J. R., Stevens, C., Blake, D. R., Pazianas, M., and Huang, C. L. H. Cellular biology of bone resorption. Binl Rev 68:197-264; 1993.

Date Received: November 16, 1994 Date Revised: March 7, 1995 Date Accepted: March 17, 1995