- Email: [email protected]
Bacterial infection (Legionella pneumophila ) stimulates fever, metabolic rate and brown adipose tissue activity in the Guinea pig
Life Sciences, Vol. 45, pp. 843-847 Printed in the U.S.A.
Maxwell Pergamon Macmillan
BACTERIAL INFECTION (LEGIONEI.!A PNEUIvK)PHILA) STI~dlATES RATE AND BROWN ADIPOSE TISSUE ACTIVITY IN THE GUINEA PIG.
FEVER,
METABOLIC
A.L. Coope rl, R.B. Fitzgeorge 2, A. Baskerville 2, R.A. Little 3 and N.J. Rothwell* IDepartment of Physiological Sciences and 3North Western Injury Research Centre, University of Manchester, MI3 9PT, 2PHLS Centre for Applied Microbiology & Research, Porton Down, Salisbury SP40JG, (Received in final form June 28, 1989) SUMMARY The objective of this study was to assess Whether bacterial infection stimulates oxygen consumption and brown adipose tissue (BAT) activity. Guinea pigs infected with Legionella pneumophila showed marked fever and a significant (33%) increase in resting oxygen const~ption (VOw), 24h after infection. At this time, food intake and body weight were normal and the i__nn vitro thermogenic activity of BAT taken from infected animals was elevated by 64% above that of control guinea pigs. VO~ and BAT activity fell to control values by 48h as infected animals became moribund and over this period food intake was markedly reduced. Bacterial infection is frequently associated with fever, which is generated by both increased heat production and reduced heat loss. Hypermetabolic responses to infection and injury may involve similar mechanisms to non-shivering thermogenesis (1,2,3). Studies on laboratory rodents have tended to concentrate on the use of bacterial endotoxins (eg 2), which do not mimic all aspects of the responses to infection, particularly since they are short-lived and are rapidly tolerated. We report here the results of a study using a pneumonia infection with Legionella pnetl~ophila in the guinea pig. The objective of the study was to determine whether bacterial infection which causes weight loss is associated with hypermetabolism and activation of BAT. METHODS
Female Dunkin-Hartley guinea-pigs weighing 250-300 g, were infected by inhalation of a small particle aerosol (<5 NM diameter) of Legionella pneumophila). Eight control animals and two separate groups of eight infected animals (treated on two consecutive days) were employed. The Corby Strain s e r o g r o u p 1 was g r o w n f o r 4 d a y s a t 37°C o n BCYE a g a r p l a t e s a n d s u s p e n d e d i n sterile w a t e r a n d t h e a e r o s o l was p r o d u c e d b y a C o l l i s o n s p r a y i n a H e n d e r s o n apparatus (4). This infection of guinea pigs results in fibrinopurulent b r o n c h o p n e u m o n i a a n d c a u s e s d e a t h w i t h i n 4 d a y s . The b a c t e r i a l counts were assessed in macerated tissue which were serially diluted in sterile water and s p r e a d o n t o BCYE a g a r p l a t e s a f t e r i n c u b a t i o n a t 37°C f o r 4 d a y s . Col~aies w e r e c o u n t e d a n d ntm~bers o f v i a b l e o r g a n i s m s c a l c u l a t e d .
Animals were housed in groups of eight at 24 ± I°C (light 0800-2000) in filter-topped cages. Body weight, food and water intake were recorded daily colonic temperature was measured between 0800 and I000 by insertion of a plastic coated thermocouple 5 cm beyond the reotum. Resting oxygen consumption (VO2) was measured in closed circuit calorimeters (5) at 24"C for 0024-3205/89 $3.00 + .00 Copyright (c) 1989 Maxwell Pergamon Macmillan plc
844
Infection and Metabolism
Vol. 45, No. 9, 1989
at least 2h on several occasions namely - 6, 24 and 48h after infection. Values were corrected for metabolic bod.v size (ml/min/kg0.Ts). On one occasion, 24h after infection, control and infected guinea pigs were injected with DL-propranolol (20mg/kg, subcut), and VO2 was measured for a further 2h. The actions of propranolol in rodents are very short-lived (<6hr) so the same animals were u s e d o n the following day. Separate groups of guinea pigs (n=6) were killed by cervical dislocation 24 and 48h after infection, together with controls. Interscapular BAT depots were removed and homogenized in 0.2 M sucrose. Mitochondria were p r e p e r e d b y differential centrifugation and tissue and mitochondrial protein contents were determined by using a dye-reagent method (Bio-Rad, Warlord, UK). The in vitro thermogenie activity of BAT was assessed from measurements of radiolabelled guanosine diphosphate (3H GDP, 2pM 10Ci/mmol, Amersham, UK) binding to isolated mitoehondria (6). Non-specific binding was determined at a ligand concentration of 200 pd~, and '4C sucrose was added as an extramitochondrial marker. This method provides an estimate of the activity of the proton conductance pathway which is considered to be responsible for thermogenesis in BAT (7). Data are presented as means ± SEM and statistical analysis Students' t-test for unmatched data, unless stated otherwise.
involved
RESULTS Control guinea pigs gained weight at a steady rate (Table I) and showed consistent (coefficient of variation < 10%), values for food and water intake (Table I), VO2 and body temperatures (Table If) throughout the experiment. No TABLE I Body Weight change, Food and Water Intake of Control and Infected animals.
Body w e i g h t c h a n g e 0-24h 24-48h
Food i n t a k e
(g) 0-24h 24-48h
Water intake (g) 0-24h 24-48h
Control
Infected
+6 ± 1 +9 ± 1
+8 ± 1 - 3 8 ± 2***
27.9 + 3.3 32.5 ± 2.9
30.5 ± 0.9 12.3 ± 1.9.**
79 ± 1 80 ± 1
68 ± 4 23 ± 4**
(g)
Mean values + SI~q, (n=8) *~ P<0.01, *~* P
Vol. 45, No. 9, 1989
Infection and Metabolism
845
{48h) became lethargic and exhibited hyperventilation. Marked fever (+ 1.4 °C), and hypermetabolism (33~ increase in VO2) were observed 24h after infection (Table II), but values returned to normal or below normal by 48h. Animals usually die by 3-4 days after infection. Injection of propranolol had no significant effect on VO2 of control guinea pigs and caused a small (7%, P<0.05, paired t-test) transient fall in VO2 in infected animals, but did not therefore abolish the difference between the two Eroups. TABLE II Effects of Legionella pne~nophila Infection on VO~ and Body Temperature. Control Colonic temperature ( ° C )
Infected (24h)
Infected (48h)
38.3±0.1
39.7±0.2.**
36.5±0.3*
V02 ( m l / m i n / k g ° . T s ) 12.5±0.3
16.7±0.2.**
10.8±0.6
Before After propranolol
15.7±0.4.**
12.3±0.5
Mean values ± S ~ , (n=8) * P<0.05, Z**P<0.001 vs control The mass of interscapular BAT was similar for all groups, but protein content was increased 48h after infection. Specific GDP binding was raised by over 60~, 24h after infection but declined to normal by 48h (Table III). TABLE III Effects of Legionella ~net~nophila Infection on BAT Mass, Protein Content and Mitoehondrial GDP Binding.
Interscapular BAT mass (g) Protein content (mE) (%) Specific GDP binding (pmol/mg protein)
Control
Infected (24h)
1.22±0.6
1.57±0.19
20±1 1.6±0.1
26±5 1.7±0.4
46±5
75±6**
Mean values +_ SEM, (n=6) *z P<0.01 vs control
Infected (48h )
1.57±0.23
33±4** 2.2±0.2** 45±10
846
Infection and Metabolism
Vol. 45, No. 9, 1989
DISCUSSION Legionella pnetm~oDhila infection induces similar pathological changes in the lungs of guinea pigs to those observed in h , ~ n patients (8) and appears to provide a useful and relevant model to stuc[v the metabolic responses to infection. ~ counts indicated a rapid increase in bacterial ntm~bers over 24h and a further increase 48h after inhalation. Body weight change and food and fluid intake were similar in control and infected animals in the infected group over the first 24h after infection. From 24-48h, hypophagia and weight loss were observed. In the absence of body composition analysis it is impossible to determine whether muscle wasting accompanied fat loss. VOw, body temperature and BAT activity, were all significantly elevated 24h post-infection. At this stage the animals did not appear to be hyperventilating and showed no other marked symptoms of infection although they were rather less motive than controls. The closely related changes in VO~ and BAT activity may not be causal, but strongly indicate that thermogenesis in brown fat is a major contributor to the hypermetabolic response. Interestingly, BAT protein content is preserved or even increased, at a time when total body weight, and therefore probably protein content, are declining. The falls in VOw, temperature and BAT activity by 48h presumably resulted at least in part from the decline in food intake. The rise in VOz probably eontributed significantly to loss of body weight, which was very much larger than could be expected from the decrease in food intake alone. It is ~a~_likely that the reduction in water intake contributed significantly to weight loss since we have found that very much greater water deprivation is required to induce weight loss. Thermogenesis in BAT is activated by the sympathetic nervous system and 8 adrenergic antagonists inhibit the thermogenic responses to endotoxin (2) as well as non-shivering and diet-induced thermogenesis (3). It is therefore surprising that propranolol caused only slight inhibition of the hypermetabolism in infected guinea pigs. This may be a reflection of the dose used, or due to hypoxia resulting from bronchooonstriction caused by propranolol which might have greater effect in animals with lung disease. It is also possible that inhibition of non-shivering thermogenesis by propranolol may have resulted in substitution by another form of heat production, such as shivering. Shivering motivity was not measured directly but none was observed. This study illustrates that BAT motivity is increased in animals infected with legionella pnetmlop~ila and the activity correlates with fever, oxygen consumption and weight loss; a causal relationship may therefore exist. ACKNOWLEDGI~ENTS We are grateful to Anthea Hardwick Angela Cooper is a SERC scholar.
for technical
assistance.
REFERENCES
I.
BLATrEIS, C.M.
2.
~ , M.M., M. COX, P.C. BATES, N.J. lc:~Oq"J~TL~.T,T-, M.J. STOCK, E.B. CADY and D.J. MILLWARD, Am. J. Physiol. 252, E581-E588, (1987). I%OTHWK~J., N.J. and M.J. STOCK, Bioscience Rep. 6, 3-18, (1986). BASKERVIIIZ, A., R.B. F I T Z ~ E , M. ~ , P. HAMBLETON, and P.J. DENNIS, Lancet ii, 1389 (1981). STOCK, M.J. J. Appl. Physiol. 39, 849-880, (1975).
3. 4. 5.
J. Appl. Physiol. 40, 29-34, (19761.
Vol. 45, No. 9, 1989
6.
Infection and Metabolism
847
7.
BROOKS, S.L., N.J. RUrHWRI,I. and M.J, STOCK, Quart. J. P h y s i o l . 67 259268, (1982). NICHOI,IR,D.G, B i o s c i . Pep. 3, 431-441, (1983).
8.
HAMBLEII)N, P., N.E. BAILEY,
R.B. F I T ~
J. Exp. Path. 66, 172-183, (1985).
and
A. ~ I L L E ,
A.
Br.
Maxwell Pergamon Macmillan
BACTERIAL INFECTION (LEGIONEI.!A PNEUIvK)PHILA) STI~dlATES RATE AND BROWN ADIPOSE TISSUE ACTIVITY IN THE GUINEA PIG.
FEVER,
METABOLIC
A.L. Coope rl, R.B. Fitzgeorge 2, A. Baskerville 2, R.A. Little 3 and N.J. Rothwell* IDepartment of Physiological Sciences and 3North Western Injury Research Centre, University of Manchester, MI3 9PT, 2PHLS Centre for Applied Microbiology & Research, Porton Down, Salisbury SP40JG, (Received in final form June 28, 1989) SUMMARY The objective of this study was to assess Whether bacterial infection stimulates oxygen consumption and brown adipose tissue (BAT) activity. Guinea pigs infected with Legionella pneumophila showed marked fever and a significant (33%) increase in resting oxygen const~ption (VOw), 24h after infection. At this time, food intake and body weight were normal and the i__nn vitro thermogenic activity of BAT taken from infected animals was elevated by 64% above that of control guinea pigs. VO~ and BAT activity fell to control values by 48h as infected animals became moribund and over this period food intake was markedly reduced. Bacterial infection is frequently associated with fever, which is generated by both increased heat production and reduced heat loss. Hypermetabolic responses to infection and injury may involve similar mechanisms to non-shivering thermogenesis (1,2,3). Studies on laboratory rodents have tended to concentrate on the use of bacterial endotoxins (eg 2), which do not mimic all aspects of the responses to infection, particularly since they are short-lived and are rapidly tolerated. We report here the results of a study using a pneumonia infection with Legionella pnetl~ophila in the guinea pig. The objective of the study was to determine whether bacterial infection which causes weight loss is associated with hypermetabolism and activation of BAT. METHODS
Female Dunkin-Hartley guinea-pigs weighing 250-300 g, were infected by inhalation of a small particle aerosol (<5 NM diameter) of Legionella pneumophila). Eight control animals and two separate groups of eight infected animals (treated on two consecutive days) were employed. The Corby Strain s e r o g r o u p 1 was g r o w n f o r 4 d a y s a t 37°C o n BCYE a g a r p l a t e s a n d s u s p e n d e d i n sterile w a t e r a n d t h e a e r o s o l was p r o d u c e d b y a C o l l i s o n s p r a y i n a H e n d e r s o n apparatus (4). This infection of guinea pigs results in fibrinopurulent b r o n c h o p n e u m o n i a a n d c a u s e s d e a t h w i t h i n 4 d a y s . The b a c t e r i a l counts were assessed in macerated tissue which were serially diluted in sterile water and s p r e a d o n t o BCYE a g a r p l a t e s a f t e r i n c u b a t i o n a t 37°C f o r 4 d a y s . Col~aies w e r e c o u n t e d a n d ntm~bers o f v i a b l e o r g a n i s m s c a l c u l a t e d .
Animals were housed in groups of eight at 24 ± I°C (light 0800-2000) in filter-topped cages. Body weight, food and water intake were recorded daily colonic temperature was measured between 0800 and I000 by insertion of a plastic coated thermocouple 5 cm beyond the reotum. Resting oxygen consumption (VO2) was measured in closed circuit calorimeters (5) at 24"C for 0024-3205/89 $3.00 + .00 Copyright (c) 1989 Maxwell Pergamon Macmillan plc
844
Infection and Metabolism
Vol. 45, No. 9, 1989
at least 2h on several occasions namely - 6, 24 and 48h after infection. Values were corrected for metabolic bod.v size (ml/min/kg0.Ts). On one occasion, 24h after infection, control and infected guinea pigs were injected with DL-propranolol (20mg/kg, subcut), and VO2 was measured for a further 2h. The actions of propranolol in rodents are very short-lived (<6hr) so the same animals were u s e d o n the following day. Separate groups of guinea pigs (n=6) were killed by cervical dislocation 24 and 48h after infection, together with controls. Interscapular BAT depots were removed and homogenized in 0.2 M sucrose. Mitochondria were p r e p e r e d b y differential centrifugation and tissue and mitochondrial protein contents were determined by using a dye-reagent method (Bio-Rad, Warlord, UK). The in vitro thermogenie activity of BAT was assessed from measurements of radiolabelled guanosine diphosphate (3H GDP, 2pM 10Ci/mmol, Amersham, UK) binding to isolated mitoehondria (6). Non-specific binding was determined at a ligand concentration of 200 pd~, and '4C sucrose was added as an extramitochondrial marker. This method provides an estimate of the activity of the proton conductance pathway which is considered to be responsible for thermogenesis in BAT (7). Data are presented as means ± SEM and statistical analysis Students' t-test for unmatched data, unless stated otherwise.
involved
RESULTS Control guinea pigs gained weight at a steady rate (Table I) and showed consistent (coefficient of variation < 10%), values for food and water intake (Table I), VO2 and body temperatures (Table If) throughout the experiment. No TABLE I Body Weight change, Food and Water Intake of Control and Infected animals.
Body w e i g h t c h a n g e 0-24h 24-48h
Food i n t a k e
(g) 0-24h 24-48h
Water intake (g) 0-24h 24-48h
Control
Infected
+6 ± 1 +9 ± 1
+8 ± 1 - 3 8 ± 2***
27.9 + 3.3 32.5 ± 2.9
30.5 ± 0.9 12.3 ± 1.9.**
79 ± 1 80 ± 1
68 ± 4 23 ± 4**
(g)
Mean values + SI~q, (n=8) *~ P<0.01, *~* P
Vol. 45, No. 9, 1989
Infection and Metabolism
845
{48h) became lethargic and exhibited hyperventilation. Marked fever (+ 1.4 °C), and hypermetabolism (33~ increase in VO2) were observed 24h after infection (Table II), but values returned to normal or below normal by 48h. Animals usually die by 3-4 days after infection. Injection of propranolol had no significant effect on VO2 of control guinea pigs and caused a small (7%, P<0.05, paired t-test) transient fall in VO2 in infected animals, but did not therefore abolish the difference between the two Eroups. TABLE II Effects of Legionella pne~nophila Infection on VO~ and Body Temperature. Control Colonic temperature ( ° C )
Infected (24h)
Infected (48h)
38.3±0.1
39.7±0.2.**
36.5±0.3*
V02 ( m l / m i n / k g ° . T s ) 12.5±0.3
16.7±0.2.**
10.8±0.6
Before After propranolol
15.7±0.4.**
12.3±0.5
Mean values ± S ~ , (n=8) * P<0.05, Z**P<0.001 vs control The mass of interscapular BAT was similar for all groups, but protein content was increased 48h after infection. Specific GDP binding was raised by over 60~, 24h after infection but declined to normal by 48h (Table III). TABLE III Effects of Legionella ~net~nophila Infection on BAT Mass, Protein Content and Mitoehondrial GDP Binding.
Interscapular BAT mass (g) Protein content (mE) (%) Specific GDP binding (pmol/mg protein)
Control
Infected (24h)
1.22±0.6
1.57±0.19
20±1 1.6±0.1
26±5 1.7±0.4
46±5
75±6**
Mean values +_ SEM, (n=6) *z P<0.01 vs control
Infected (48h )
1.57±0.23
33±4** 2.2±0.2** 45±10
846
Infection and Metabolism
Vol. 45, No. 9, 1989
DISCUSSION Legionella pnetm~oDhila infection induces similar pathological changes in the lungs of guinea pigs to those observed in h , ~ n patients (8) and appears to provide a useful and relevant model to stuc[v the metabolic responses to infection. ~ counts indicated a rapid increase in bacterial ntm~bers over 24h and a further increase 48h after inhalation. Body weight change and food and fluid intake were similar in control and infected animals in the infected group over the first 24h after infection. From 24-48h, hypophagia and weight loss were observed. In the absence of body composition analysis it is impossible to determine whether muscle wasting accompanied fat loss. VOw, body temperature and BAT activity, were all significantly elevated 24h post-infection. At this stage the animals did not appear to be hyperventilating and showed no other marked symptoms of infection although they were rather less motive than controls. The closely related changes in VO~ and BAT activity may not be causal, but strongly indicate that thermogenesis in brown fat is a major contributor to the hypermetabolic response. Interestingly, BAT protein content is preserved or even increased, at a time when total body weight, and therefore probably protein content, are declining. The falls in VOw, temperature and BAT activity by 48h presumably resulted at least in part from the decline in food intake. The rise in VOz probably eontributed significantly to loss of body weight, which was very much larger than could be expected from the decrease in food intake alone. It is ~a~_likely that the reduction in water intake contributed significantly to weight loss since we have found that very much greater water deprivation is required to induce weight loss. Thermogenesis in BAT is activated by the sympathetic nervous system and 8 adrenergic antagonists inhibit the thermogenic responses to endotoxin (2) as well as non-shivering and diet-induced thermogenesis (3). It is therefore surprising that propranolol caused only slight inhibition of the hypermetabolism in infected guinea pigs. This may be a reflection of the dose used, or due to hypoxia resulting from bronchooonstriction caused by propranolol which might have greater effect in animals with lung disease. It is also possible that inhibition of non-shivering thermogenesis by propranolol may have resulted in substitution by another form of heat production, such as shivering. Shivering motivity was not measured directly but none was observed. This study illustrates that BAT motivity is increased in animals infected with legionella pnetmlop~ila and the activity correlates with fever, oxygen consumption and weight loss; a causal relationship may therefore exist. ACKNOWLEDGI~ENTS We are grateful to Anthea Hardwick Angela Cooper is a SERC scholar.
for technical
assistance.
REFERENCES
I.
BLATrEIS, C.M.
2.
~ , M.M., M. COX, P.C. BATES, N.J. lc:~Oq"J~TL~.T,T-, M.J. STOCK, E.B. CADY and D.J. MILLWARD, Am. J. Physiol. 252, E581-E588, (1987). I%OTHWK~J., N.J. and M.J. STOCK, Bioscience Rep. 6, 3-18, (1986). BASKERVIIIZ, A., R.B. F I T Z ~ E , M. ~ , P. HAMBLETON, and P.J. DENNIS, Lancet ii, 1389 (1981). STOCK, M.J. J. Appl. Physiol. 39, 849-880, (1975).
3. 4. 5.
J. Appl. Physiol. 40, 29-34, (19761.
Vol. 45, No. 9, 1989
6.
Infection and Metabolism
847
7.
BROOKS, S.L., N.J. RUrHWRI,I. and M.J, STOCK, Quart. J. P h y s i o l . 67 259268, (1982). NICHOI,IR,D.G, B i o s c i . Pep. 3, 431-441, (1983).
8.
HAMBLEII)N, P., N.E. BAILEY,
R.B. F I T ~
J. Exp. Path. 66, 172-183, (1985).
and
A. ~ I L L E ,
A.
Br.