Lymphocyte subpopulations in peripheral blood of lambs experimentally infected with Pasteurella haemolytica

l~,terinaryMicrobiology, 27 ( 1 9 9 1 ) 159-168 Elsevier Science Publishers B.V., A m s t e r d a m

159

Lymphocyte subpopulations in peripheral blood of lambs experimentally infected with Pasteurella

haemolytica R. Sharma and Z. Woldehiwet* University qf Liverpool, Department of Veterinary Pathology, Veterinary Field Station, Leahurst, Neston, Wirral L64 7TE, UK (Accepted 18 October 1990

ABSTRACT Sharma, R. and Woldehiwet, Z., 1991. Lymphocyte subpopulations in peripheral blood of lambs experimentally infected with Pasteurella haemolytica. Vet. Microbiol., 27:159-168. The lymphocyte subpopulations in peripheral blood obtained from eleven lambs experimentally infected with Pasteurella haemolytica were compared with those obtained from eight control lambs by flow cytometry, using a panel of monoclonal antibodies against specific lymphocyte epitopes. Experimental infection with P. haernolytica was characterized by a transient but significant reduction in SBU-T1 + (CD5 + ) T cells and SBU-T4 + (CD4 + or helper) T lymphocytes ( P < 0.05 ) and a significant rise in lymphocytes which did not express the LCA p220 epitope and the pan T cell surface marker ( C D S - L C A p220 ) ("null"). The reductions in CD5 + and CD4 + lymphocytes occurred 24 h after experimental infection, returning to preinoculation levels 5 days post inoculation (DPI). Five to 9 days after experimental infection, there was a significant increase in the number of lymphocytes, which expresses the pan T cell surface marker (CD5 + ) but which were C D 4 - C D 8 . Lymphocyte transformation responses to the mitogen phytohaemagglutinin ( PHA ) were significantly reduced 24 h after experimental infection with P. haemolytica (P< 0.05 ).

INTRODUCTION

Pneumonic pasteurellosis of calves and lambs is a major cause of economic loss in many parts of the world (Gilmour, 1980; Alley, 1975). Pasteurella haemolytica is commonly associated with pneumonic pasteurellosis of sheep (Gilmour, 1980) and cattle (Allan, 1978 ). A number of reports has demonstrated that live P. haemolytica or culture filtrates of P. haemolytica are specifically cytotoxic for leucocytes of ruminants (Kaehler et al., 1980; Shewen and Wilkie, 1982; Sutherland, 1985). This cytotoxic activity has been associated with cell-free secretory products of logarithmic-phase cultures of P. haemolytica, designated as 'cytotoxin'. Shewen and Wilkie (1982) suggested that P. haemolytica cytotoxin may participate in the pathogenesis of pneu* A u t h o r for correspondence

0378-1135/91/$03.50

© 1991 - - Elsevier Science Publishers B.V.

I(;I)

I< ~ , I I k R % l \

\~I}.l

\\~Q III II1\\t

I

monic pasteurellosis either by impairing the cellular lung defcnces and subsequent immune responses or by the induction of an inflammatory reaction as a consequence of leucocyte lysis. In vitro experiments have shown that P. haemolylica culture filtrates lyse ovine neutrophils, lymphocytcs and alveolar macrophages (Sutherland, 1985 ) but the effects of t'. haemol)'Iica infection on lymphocyte subpopulations are not known. Recently, Mackay et al. (1987) produced and characterized a panel of monoclonal antibodies specific to surface antigens of lymphocyte subpopulations of sheep. The SBU-T1 (25-91) identifies the CD5 molecule and has been used as a pan T cell marker in sheep (Mackay et al., 1987). SBU-T4 (44-38) monoclonal recognizes the CD4 molecule (Maddox et al., 1985), which marks the functional subset of T cell responsible for helper/inducer activity (Reinherz and Schlossman, 1980: Mason et al., 1983 ). The SBU-T4 stains CD8 T cells, cortical and most medullary thymocytes and cells of the macrophage and Langerhans cell lineage ( Maddox et al., 1985 ). The SBU-T8 (38-65) monoclonal marks the CD8 epitope ofovine lymphocytes (Mackay et al., 1987). This epitope defines that population of T cells which exhibit cytotoxic or supressor function. The monoclonal LCA p220 (20-96) reacts with all B cells and null cell types being CD4 , CD8 , slgG and T19 (Mackay et al., 1987 ). The object of the present study was to investigate the effects of P. /memolytica infection on the distribution of lymphocyte subpopulations in peripheral blood of sheep using monoclonal antibodies which recognize surface antigens o f T and B lymphocytes (Mackay et al., 1987 ). MATERIALS A N D M E T H O D S

Lambs Nineteen 6-8-week-old conventionally reared Suffolk crossbred lambs were used. Eleven lambs were inoculated intranasally and intratracheally, with 5 ml log-phase culture of P. haernolytica containing 9 X 10v colony forming units (CFU) per ml. Two third of the inoculum was given intranasally and the remaining 1/3 was inoculated intratracheally by inserting a 20 G needle between tracheal rings. Five ml of sterile bacterial culture medium was injected into eight control lambs by the same route.

Bacterial inoculum A reference strain ofP. haemolytica biotype A serotype 1 was obtained from Dr. I.D. Aitken, Moredun Institute, Edinburgh. The 1-h log-phase cultures of P. haemolytica were prepared as described by Woldehiwet and Rowan ( 1988 ). Nasal swabs collected before and after infection were cultured onto blood agar plates, and P. haemolytica identified according to Cowan and Steel (1970).

LYMPHOCYTE SUBPOPULATIONS IN LAMBS INFECTED WITH P. HAEMOLY77CA

16 1

P. haemolytica antigen P. haemolytica cultures were grown in tryptose phosphate broth at 37°C. After overnight incubation, the bacteria were centrifuged at 3000 × g for 15 min, washed three times in phosphate-buffered saline (PBS) (pH 7.2), and resuspended in PBS to give a final concentration of 109 C F U / m l . The bacterial suspensions were heat-killed at 60°C for 45 min. After checking sterility on blood agar plates, serial two-fold dilutions of the heat-killed antigens were used to determine the o p t i m u m concentration to be used in lymphocyte transformation tests.

Collection of blood samples Peripheral blood samples collected in EDTA-coated tubes on 0, 1, 5, 9, 14, and 21 DPI were used to establish the total leucocyte counts using a Coulter Counter (Coulter Electronics Ltd, England) and differential leucocyte counts after staining blood smears with Giemsa. Heparinized samples collected on 0, 1, 5, 9, 14 and 21 DPI were used to establish the distribution of lymphocyte subpopulations and lymphocyte transformation (LT) responses. Mononuclear cells (MNC) were separated from diluted blood on Ficoll-Paque column (Pharmacia Fine Chemicals, Uppsala, Sweden). Cell viability was determined by the trypan blud exclusion technique.

Monoclonal antibodies Monoclonal antibodies were kindly provided by Dr. M.R. Brandon, University of Melbourne, Australia. The monoclonal antibodies used in the present study were 25-91, 44-38, 38-65, 20-96 which recognize the CD5, CD4, CD8 and LCA p220 antigens, respectively. The o p t i m u m dilution of each antibody was determined using checkerboard titration.

lmmunofluorescent staining Approximately l x 10 6 cells were stained with optimally diluted monoclonal antibodies and then with FITC-conjugated swine anti-mouse IgG (Nordic Immunobiological, Tilburg, The Netherlands) as previously described (Sharma et al., 1990). Stained cells were analysed by flow cytometry using a FACS 420 fluorescence activated cell sorter (Becton & Dickinson, Sunnyvale, California).

Identification of lymphocyte subpopulations The percentages of lymphocytes with epitopes to the various antibodies were transformed into absolute numbers using the total and differential counts of peripheral blood. CD5 + lymphocytes and LCA p220 + lymphocytes were regarded as T lymphocyte and B lymphocyte, respectively (Mackay et al., 1987 ). The lymphocytes which were neither CD5 + (T lymphocyte) nor LCA p220 +

1(12

R MIkt.~M\ \\1~/

\ \ ~ I I)1111\\1 i

( B lymphocyte) were calculated by subtracting the total number or CD5 + and LCA p220 + lymphocytes from the total lymphocyte counts and regarded as "null lymphocytes'" (Yang et al., 1988 ). T lymphocytes (CD5 +) which were CD4 CD8 were estimated by subtracting the total n u m b e r o f C D 4 + lymphocytes and CD8 + lymphocytes from the total n u m b e r o f C D 5 + lymphocytes as described earlier (Woidehiwet and Sharma, 1990 ).

Lymphoo,te tran,~lbrmation assays Lymphocyte transformation assays were performed on MNC obtained tYom seven infected and eight control lambs on 0, 1, 5, 9 and 14 DPI as described by Taylor et al. (1987) with slight modification. Briefly, peripheral blood mononuclear cells (10~/well) were dispensed in 100/tl volumes to the wells of 'V + bottomed microtitre plates containing 20/~1 of purified phytohaemagglutinin ( 4 0 / t g / m l ) (Wellcome reagents Ltd., Beckenham, Kent, England) or 20 ~1 of o p t i m u m concentration of heat inactivated P. haemoO'tica antigen or 20/tl of RPMI 1640 medium. After incubation for 3 days, cultures were pulsed overnight with 1 /tCi ~H-thymidine (Amersham International plc., Amersham, U K ) and harvested onto filtermats (Skatron AS, kier, Norway ) by multiple sample harvester (Skatron AS, Lier, Norway ). The radio-activity in the filters was counted in a scintillation counter (Packard Co., Pangbourne, Berks. England ). The degree of transformation was expressed as the ratio of ~H-thymidine incorporation in cultures stimulated with phytohaemagglutinin ( PHA ) or P. haernoO,tica antigen to that of the control cultures ( stimulation index, SI ).

Analysis o(dala The Student's l test was used to compare mean values obtained from lambs experimentally infected with P. haemolylica with those obtained from control uninfected lambs. The paired/-test was used to compare values obtained betore infection with values obtained after experimental infection. R ESU LTS

Lambs experimentally infected with P. haemolytica were dull and pyrexic lbr 2 days. Nasal discharges and coughings were observed in 42% and 33% of the lambs, respectively. P. haemolytica was isolated for 7 days in nasal swabs of all the lambs infected with P. haemolytica but not from control lambs. Experimental infection of lambs with P. haemolytica was characterized by a rapid change in the total leucocyte counts in peripheral blood. The total leucocyte counts rose significantly t¥om 9.3_+ 3.9× 10'~/1 before inoculation to 16.426.9)< 10'/124 h after inoculation ( P < 0 . 0 5 ) (Fig. 1 ). This rise was due to an increase in the n u m b e r ofneutrophils which lasted upto 9 DPI ( Fig.

163

LYMPHOCYTE SUBPOPULATIONS 1N LAMBS INFECTED WITH P. HAEMOLYTICA

20

P. haernolytica

20 1

-e- WBC •.e- Lymphocytes

18

Controls

18

-m- WBC -e- Lymphocytes -l- PMN

16 PMN

14

T

16 14

== 12

6

6'

4

4.

2

21

0

•

•

w

3

,

!

6

•

9

i

12

•

•

|

15

0-

•

18

211

•

i

•

i

3

•

i

6

Days Post Inoculation

•

i

9

Days

,

i

12

Post

•

15

f

•

18

!

21

Inoculation

Fig. 1. Total leucocyte counts (WBC), neutrophils (PMN) and lymphocytes in peripheral blood of lambs infected with Pasteurella haemolytica and control lambs (mean _+S.E. ). P.haemolytica I

LCA p220+ •4- CDS+ -m- CD4+

I

Controls

5

LCA ~20+ CD5+ 4=- CD4+ CD8+

4 o

3

.1=

f~ 1

•

0

i

3

.

.

6

.

.

;

.

1'2

i

1'5

Days Post Inoculation

18

•

i

21

0

i

•

i

3

•

i

6

•

i

9

Days Post

•

i

12

•

i

15

-

•

18

2'1

I n o c u l a t i o n

Fig. 2. Lymphocyte subpopulations in peripheral blood of lambs infected with Pasteurella haemolytica and control lambs (mean _+S.E. ).

1 ). There was an increase in the total number of circulating lymphocytes 24 h after infection but it was not significant ( P > 0.05). However, the proportions and numbers of lymphocytes subpopulations were drastically affected 24 h after experimental infection with P. haemolytica. Twenty-four h after infection, the number of CD5 + lymphocytes was significantly reduced from 2.44_+0.99)< 109/1 to 1.60+0.43)< 109/1 ( P < 0 . 0 5 ) (Fig. 2). The number of B lymphocytes was also reduced but not significantly ( P > 0.05 ). The number of lymphocytes which do not express the LCA p220 epitope and pan T cell surface marker (CD5) were increased significantly 24 h after infection and remained at significantly high levels until 5 DPI ( P < 0 . 0 5 ) (Fig. 3 ).

1(~4

R ~;II\RM\

',NI~/

~.~,~,11)1111\\1 I

P. haernolytica

Controls 6

~o

4-

-~

4

CD5 LCA p220

-~ ~

CD5 L C A p220 CD4 CD8

=o

1

o 5

10

15

20

Days Post Inoculation

[3

5

10

15

20

Days Post Inoculation

Fig. 3. CD5-L('Ap220 ( n u l l ) l y m p h o c y t e s a n d ( ' D 4 CD8 (TI9 +) lymphocx, lcsubpopulations in peripheral blood of lambs infected with l'a.steurella hacmo(vdua and control lambs {mean k S.E. ).

The number of C D 4 + T lymphocytes was reduced from 1 . 3 3 + 0 . 6 6 X 10'~/1 before infection to 0.82 +_0.21 X 10~/1 24 h after inoculation ( P < 0.05 ) (Fig. 2) but the number of CD8 + lymphocytes remained unaffected. The number of T lymphocytes which were neither helper nor cytotoxic/suppressor ( C D 4 - C D 8 - ) was significantly increased on 5 DPI and remained at high levels until 9 DPI ( P < 0 . 0 1 ) (Fig. 3). There were no significant changes in the total leucocyte counts, neutrophil counts and lymphocyte subsets in the peripheral blood of control lambs (Figs. 1,2 and 3).

Lymphocyte transjbrmation responses All the results are the mean of triplicate cultures. There were no significant differences in the uptake oftritiated thymidine between cultures from control and P. haernol.vtica-infected lambs when incubated without P H A or P. haemolytica antigen. There were animal to animal and day-to-day variations in the transformation responses of lymphocytes to P H A in both control and P. haemolytica-infected lambs. The variations observed in the control lambs were not statistically significant (Fig. 4) ( P > 0.05). However, the mean stimulation index in P H A stimulated cultures of P. haemolytica-infected lambs was significantly reduced from 36.19 + 6.97 before inoculation to 15.94 + 3.14 24 h after infection ( P < 0.05 ) (Fig. 4 ). The LT responses returned to normal levels 5 DPI. P. haemolytica antigen-induced transformations were not observed in any

L Y M P H O C Y T E S U B P O P U L A T I O N S IN LAMBS I N F E C T E D W I T H P. HAEMOLY77C,4

165

60.

50, 40' / ~ g

30.

_2 20'

:t

-,e- p. haemolytica 4- Controls

,

5

i

10

i

15

Days Post Inoculatoion

Fig. 4. Lymphocyte transformation responses of peripheral blood mononuclear cells to phytohaemagglutinin (PHA) in P. haemolytica infected (n = 7) and control lambs (n = 8). Each point represents mean stimulation index ( + S.E.).

of the MNC cultures obtained from P. haemolytica-infected lambs and control lambs on different days after inoculation. DISCUSSION

Very little is known about the effects of bacterial infections on the lymphocyte subpopulations in domestic animals due to the unavailability of suitable lymphocyte surface markers. The few published reports on this subject indicate that bacterial infections may affect the number of circulating T and B lymphocytes. For example, Yang et al. ( 1988 ) demonstrated that acute bacterial mastitis of cows due to Streptococcus, Staphylococcus and coliforms were accompanied by significant reductions in T and B lymphocytes in peripheral blood. Similarly, Vasilev and Kondourov (1984) reported significant reductions in T and B lymphocytes in the peripheral blood of cows vaccinated with live Brucella abortus. The most important findings of the present study were the rapid but transient reductions in CD5 ÷ and C D 4 + lymphocytes, a sustained increase in C D 4 - C D 8 - T lymphocytes and transient increases in LCA p 2 2 0 - C D 5 ( "null" ) lymphocytes. The data in the present study are not sufficient to speculate whether the

I(~0

I'~ ";11 \ R \ I ' ,

\\[)/

~3.~ ,I !)F H I \ \ [

[

reductions in CD5 ~ and CD4+ lymphocytes were due to direct eflects o f / '

/laemolytica cytotoxin or are due to alterations in immunoregulatory balance. It has been reported that the cytotoxin from logarithmic-phase cultures of 1'. haemo/ytica specifically kills ovine bronchoalveolar macrophages t Sutherland et al., 1983), ovine peripheral blood neutrophils and gastric lymphocytes ( Sutherland, 1985 ). The significance of the sustained increase in CD4 CD8 lymphocytes observed in the present study awaits elucidation of the role of these type oflymphocytes in infections. Lymphocytes with the pan T cell surface marker (CD5 +) but CD4--CD8 are found in high numbers in normal sheep ( Mackay and Mackay, 1988 ). These lymphocytes are believed to bear a molecule (T 19) different from any other molecule described in man, mouse or rat ( Mackay et al., 1986 ). The T 19 lymphocytes express the )'/fi T cell receptor (Mackay et al., 1990). It is reported that sheep ),/d T cells respond to alloantigen, nominal antigen and T cell mitogens in vitro (Mackay ct al,, 1990}. Janeway et al. ( 1988 ) suggested that T lymphocytes with this phenotype perform immunological surveillance on epithelial surfaces. The depressed lymphocyte response to PHA in lambs infected with P. haemol.vtica observed in the present study was similar to that reported in other bacterial infections. For example, lymphocytes from mice injected with Coo'nehacterium parvum showed depressed responsiveness to PHA (Scott, 1972 ). Bennett and Jasper ( 1977 ) demonstrated depressed lymphocyte response to PHA in calves infected with Mycoplasma bovis. Majury and Shewen (1990) recently demonstrated that P. haemolytica A1 culture supernatants were inhibitory to in vitro proliferative responses of bovine lymphocytes to the mitogens pokeweed and concanavalin A. This inhibitory activity was abrogated by preincubation of culture supernatant with polyclonal serum raised against recombinant cytotoxin. Further studies are required to examine whether the changes in lymphocyte subpopulations are attributable to P. haemolytica cytotoxin.

=XCKNOWLEDGEMENTS

The authors wish to thank Dr. M.R. Brandon, University of Melbourne, Australia, for providing the monoclonal antibodies and Mr. D. Spiller, Mr. M. Savage and Mrs, C.E. Savage for technical assistance. We thank Professor H.M. Warenius, Department of Radiation Oncology, University of Liverpool, Liverpool for allowing us to use the FACS facilities in his department. R. Sharma is a Commonwealth Scholar supported by the Association of Commonwealth Universities.

LYMPHOCYTE SUBPOPULATIONS IN LAMBS INFECTED WITH P. IIAEMOLY77C-I

167

REFERENCES Allan, E.M., 1978. Pulmonary bacterial flora of pneumonic and non-pneumonic calves. Curr. Top. Vet. Med., 3: 345-355. Alley, M.R., 1975. The bacterial flora of the respiratory tract of normal and pneumonic sheep. N. Z. Vet. J., 23:113-118. Bennett, R. and Jasper, D.E., 1977. Immunosuppression of humoral and cell-mediated responses in calves associated with inoculation of Mycoplasma bovis. Am. J. Vet. Res., 38: 1731-1738. Cowan, S.T. and Steel, K.J., 1970. Characters of gram-negative bacteria. Manual for the identification of medical bacteria, Cambridge University Press, London, pp. 61-82. Gilmour, N.J.L., 1980. Pasteurella haemolytica infections in sheep. Vet. Q., 2:191-198. Janeway, C.A., Jones, B. and Hayday, A., 1988. Specificity and function o f t cell bearing 7d receptors. Immonol. Today, 9: 73-76. Kaehler, K.L., Markam, R.J.F., Muscoplat, C.C. and Johnson, D.W., 1980. Evidence ofcytocidal effects of Pasteurella haemoO:tica on bovine peripheral blood mononuclear leukocytes. Am. J. Vet. Res., 41: 1690-1693. Mackay, C.R. and Mackay, I.R., 1988. Immunology and veterinary medicine Br. Vet. J., 145: 185-190. Mackay, C.R., Maddox, J.F. and Brandon, M.R., 1986. Three distinct subpopulations of sheep T lymphocytes. Eur. J. Immunol., 16: 19-25. Mackay, C.R., Maddox, J.F. and Brandon, M.R., 1987. Lymphocyte antigens of sheep: identification and characterization using a panel of monoclonal antibodies. Vet. Immunol. lmmunopathol., 17: 91-102. Mackay, C.R., McClure, S. and Hein, W.R., 1990. Phenotype, onlogeny and function of ~'/(~Tcells in sheep. Immunobiology (Suppl.), 4: 3. Maddox. J.F., Mackay, C.R. and Brandon, M.R., 1985. Surface antigens, SBU-T4 and SBU-T8, of sheep lymphocyte subsets defined by monoclonal antibodies. Immunology, 55: 739-749. Majury, A.L. and Shewen, P.E., 1989. Pasteurella haemolytica A1 culture supernatant and its efl'ect on bovine lymphocyte blastogenesis. Immunobiology (Suppl.), 4: 168. Mason. D.W., Arthur, R.P., Dallman, M.J., Green, J.R., Spickett, G.P. and Thomas, M.L., 1983. Functions of rat T lymphocyte subsets isolated by means of monoclonal antibodies. Immunol. Rev.. 74: 57-82. Reinherz, E.L. and Schlossman, S.F., 1980. The differentiation and function of human T lymphocytes. Cell, 19:821-827. Scott, M.T., 1972. Biological effects of the adjuvant Coo,nebacterium parvum. I. Inhibition of PHA, mixed lymphocyte and GVH reactivity. Cell. Immunol., 5:251-263. Sharma, R., Woldehiwet, Z., Spiller, D. and Warenius, H.M., 1990. Lymphocyte subpopulations in peripheral blood of lambs experimentally infected with bovine respiratory syncytial virus. Vet. Immunol. Immunopathol., 24:383-391. Shewen, P.E. and Wilkie, B.N., 1982. Cytotoxin of Pasteurella haemolytica acting on bovine leukocytes. Infect. Immun., 35:91-94. Sutherland, A.D., 1985. Effects ofPasteurella haemolytica cytotoxin on ovine peripheral blood leucocytes and lymphocytes obtained from gastric lymph. Vet. Microbiol., 10:431-438. Sutherland, A.D., Gray, E. and Wells, P.W., 1983. Cytotoxic effects of Pasteurella haemolytica cytotoxin on ovine bronchoalveolar macrophages in vitro. Vet. Microbiol., 8: 3-15. Taylor, G., Stott, E.J. and Thomas, L.H., 1987. Lymphocyte transformation responses of calves to respiratory syncytial virus. J. Med. Virol., 22: 333-344. Vasilev, A.A. and Kondourov, B.I., 1984. Number o f T and B lymphocytes and macrophages in

J (~

F ~,11 XRM

\ ,'~-!~ /

¢'.~ ~I ! q t tl\ ~. t

[he llli]l(. ()~'c()\~, '-, Illlllll.lllllcd ( L|,,~L|ill',;[ I',Ytlcuqh)'.;is ) b',, ~.~llWuin I'otllcn. \ ,.'I,.'MtlaT I\ :h MOSL'(~\\ No. I: 3 0 - 3 2 . \ V o l d e h i w c l , Z. and Ro',',an. [ . ( K , I'q8~4. ~-~fl',..'cts o f c n v i r , , ) n m c n t a l t C m p c r a l u l c . I-c]atj',,.' hulnv, jit,, and ,~accination on l'aslcurc/la haemo/rlica in lungs o t mi,,c..1. (,,m~p. [:'alll~).. q,~: 4 3 3

43~. WoldehiweL Z. and S h a m m . R., 1990. Altcralion in l~mphoc.,,Ic subpopulations in pcFiphcral blood of shccp persistently infeclcd \sith border disease virus. Vet. Micmbiol.. 22:153-1 ~II. Yang. T.J.. Malher. J.F. and Rabinovsky. E.I).. 1988. ('hangcs in subpopulations of I~ mphoc~tcs in peripheral blood, and supramammary and prescapular I:,mph nodes of c¢~s x~ilh mastilis and normal co~s. Vet. lmmunol. Imnlunopathol.. 18:27~,~-285