INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

AGED HUMANS OFTEN HAVE difficulty mounting adequate fever responses when they are infected (3, 20). The same is true for aged rodents. There are many reports of blunted fever in old animals after peripheral injections of endotoxin [lipopolysaccharide (LPS)] or proinflammatory cytokines (e.g., Refs. 4, 9, and 19). Most studies investigating the effects of age on thermoregulation have focused primarily on physiological aspects of thermal homeostasis. Indeed, aged organisms may have deficits in shivering (1) and nonshivering (2, 8) thermogenesis when they are in a cold environment. Although heat production and heat loss mechanisms have not been measured in old animals after LPS, it is quite possible that these peripheral thermoregulatory mechanisms are inadequate to raise body temperature (T_b). This is unlikely, because we have recently shown that after central injections of interleukin (IL)-1 and PGE₂ at an ambient temperature (T_a) of 23°C, old rats get as high a fever as do young rats (12, 15).

Thermoregulatory behavior can play a significant role in fever production and, indeed, may be crucial in individuals whose capacity for physiological thermoregulation is limited. For example, in infants, who have little insulation and very high heat loss, heat production mechanisms are inadequate to raise T_b. Satinoff et al. (14) found that infant rabbits did not develop fever after endotoxin injection if they were kept in incubators. However, they developed adultlike fevers when they were allowed to behaviorally regulate by choosing very warm positions in a thermally graded alleyway. Their data demonstrate that the drive toward fever was present in the infants, but could only be expressed behaviorally. Similar results have been observed in many ectothermic species with no capacity at all for physiological thermoregulation (see Ref. 11 for review).

In the present paper, we hypothesized that old rats that do not become febrile after LPS injection at room temperature might resemble infant rabbits. That is, the old rats might develop a fever if they were allowed to thermoregulate behaviorally. Indeed, this was the case, and we report here that old rats that do not develop fevers after peripheral LPS injections when they are kept at T_a 23°C develop fevers by selecting a warm T_a in a thermally graded alleyway.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals and surgery. Subjects were young (3-5 mo) and old (24-29 mo) male and female Long-Evans rats (n = 21). Animals were maintained in their home cages at T_a 23 ± 1°C on a 12:12-h light-dark cycle (lights on at 7 AM, i.e., 3 h before the LPS injection). Food and water were available ad libitum. Young rats were anesthetized with a solution of ketamine-HCL (87 mg/kg body wt) and xylazine (13 mg/kg body wt) and implanted intraperitoneally with a battery-operated biotelemetry device (model VM, Mini-Mitter, Sunriver, OR). Old rats were anesthetized with 80% of this dose and similarly implanted with transmitters.

Experimental design. At least 1 wk postoperatively, animals were placed in a thermally graded alleyway. At 10 AM on day 3, one-half the rats were injected intraperitoneally with LPS (50 µg/kg) or an equivalent volume of saline. They were placed in their home cage at 23°C for the 3 h immediately following injections, after which they returned to the gradient. The injection procedure was repeated on day 6 in the thermal gradient, with rats that had been injected with LPS given saline, and vice versa. All rats were left in the gradient for 2 more days. Thus each rat served as its own control. Other groups of young and old rats were injected with the same dose of LPS or saline and left in their home cages.

Thermal gradient. Two identical thermal gradients were used. Each consisted of an aluminum floor and wall (75 cm long × 12 cm wide × 15 cm high) with a hinged Plexiglas top. The floor was extended 30.5 cm beyond the walls on each end for heating and cooling. Heating tape was wrapped on one end and the other end was bathed in a circulating solution of chilled ethylene glycol. The gradient floor temperatures varied from 10 to 40°C. Position sensors and thermocouples were placed at 8-cm intervals along the floor of the gradient. The temperature gradient was linear, and no animals chose to stay at either the coldest or hottest gradient positions.

LPS. Purified lyophilized phenol extract of Escherichia coli endotoxin (0111:B4, Sigma, St. Louis, MO, Catalog No. L-2630) was dissolved in sterile saline (American Pharmaceutical Partners, Los Angeles, CA) aliquoted, and frozen at 20°C. LPS was injected intraperitoneally at a dose of 50 µg/kg body wt. Injection of an equivalent volume of sterile saline was used as a control.

Data collection and analysis. T_b of rats in their home cages were monitored continuously using a peripheral processor (Datacol III System) connected to a personal computer. Temperature-dependent transmitter pulse frequencies were converted to T_b by the Datacol system. The data were stored on hard disk every 5 min. When animals were in the thermal gradient, T_b data were collected in the same manner via a loop antenna placed in the gradient and connected to the Datacol receiver. While animals were in the thermal gradient, selected T_a (T_sel) was recorded every minute via a second data acquisition system (Dasylab, Amherst, NH) and stored on disk. Thirty-minute averages of T_b and T_sel were used in the analysis. Repeated-measures ANOVA was used to test for significant main effects. Student's t-test or Bonferroni t-test was used to assess post hoc differences between the groups. Significance level was set at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

T_b response to LPS. Figure 1 shows the time course of T_b changes in young and old rats before and after LPS injection. Because there were no significant sex differences, males and females are grouped together within each age group. Preinjection T_b (measured at time 0) were similar in both groups both before the saline injection (young, 37.5 ± 0.2°C; old, 37.7 ± 0.2°C; Fig. 1A) and before the LPS injection (young, 37.5 ± 0.2; old, 37.4 ± 0.2; Fig. 1B). Note the stress-induced hyperthermia immediately after the injection, which is significantly higher in young animals (P < 0.001).

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Body temperature (Tb) changes before and after injection of saline (Sal; A) or lipopolysaccharide (LPS; B) in young and old rats. TG, thermal gradient; HC, home cage. Stress-induced hyperthermia in home cage is higher in young rats (P < 0.001)

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Tb of old (of, old female; om, old male) and young (yf, young female; ym, young male) rats injected with LPS (50 µg/kg) and left in their home cages. None of the old rats developed fevers, whereas all the young ones did.

T_sel in the thermal gradient. The old animals selected significantly warmer floor temperatures than did the young animals after either saline or LPS injection (Table 1, P < 0.001).

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Selected temperature (Tsel) of young and old rats from 3 h to 9 h postinjection. In neither age group was any Tsel time point different from time 0 baseline for that age. At all time points, old rats selected higher gradient positions than did young rats after either saline (A) or LPS (B).

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Tsel of young (A) and old (B) rats after saline or LPS. In young rats the difference was not significant. In old rats it was P < 0.05. Note the increasing Tsel of the young rats in the 3 h after lights on.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

These data demonstrate that at T_a 23°C young rats injected with LPS developed febrile responses after 3 h; old rats similarly treated either had blunted fever or no fever. However, 3 h postinjection, when the old rats were placed in a thermal gradient, they chose warm positions and then became febrile. Our data suggest that 1) after LPS, the drive toward fever is present in old rats but 2) the old animals could not reach the higher T_b at a T_a of 23°C. At the higher T_a, however, the old rats reached a peak fever higher than that of the young rats (Fig. 1B) at the time when the febrile responses of the young rats were already abating.

The delayed defervescence in the old rats agrees with the observation of Sapolsky et al. (13) that once a response is elicited in old rats, it may take longer to return to baseline. For instance, when 24-mo-old rats were injected intravenously with ACTH, the corticosterone response was dampened and recovery was slower (16). The amount of tumor necrosis factor (TNF)- produced after challenge with endotoxin was 20 times higher in old mice than in young ones in the presence of elevated corticosterone levels (10).

Although the old rats chose warmer positions in the gradient after LPS than after saline, the young rats did not. Although there was a tendency toward higher T_sel after LPS, this was not significant. Sugimoto et al. (17) reported similar results in young rats. LPS produced similar fevers whether it was given during the light or dark phase, but the rats only selected warmer T_a during the dark phase. C. J. Gordon (unpublished data) has seen similar effects in a small sample of young rats; given an injection of LPS (50 µg/kg) in the early part of the light cycle, they do not choose a warmer T_a than do saline-injected controls until light off. Briese (6) reported that there was a significant correlation between the magnitude of the T_b rise in fever after a large dose of LPS (40 mg/rat) and the T_sel of male rats. However, neither the age of the rats, nor their initial T_b, nor the light cycle was reported. The injections were given at 1730. This could have been near the beginning of the dark cycle, but there is no way of knowing. The same problem occurs in work by Bodurka et al. (5). The rats were injected between 8 and 9 h, and were "exposed to seasonal illumination conditions."

With the exception of the latter two studies, in which the methods are unclear, the results in young rats from three different laboratories with three different protocols are puzzling. Apparently in the light, young rats do not use behavior in a thermal gradient to augment their autonomic responses to achieve fever. This is puzzling because no work is required to choose to stay in a warm position in the gradient. Clearly, more comprehensive studies such as that of Sugimoto et al. (17) need to be done.

Although young animals develop high fevers in response to endotoxin injection, old rats develop low fevers, no fevers, or become hypothermic. The underlying mechanism involved in this observation remains to be elucidated. There are several steps involved in the initiation of fever, and age may impact any or all of these processes.

The pathway leading to fever production is initiated when the infectious agent stimulates the immune system to produce cytokines, some of which (e.g., IL-1, IL-6, and tumor necrosis factor-) are proinflammatory and elicit fever. Proinflammatory cytokines cause the synthesis and release of PGE₂ in the brain, a final common pathway in fever induction. Elevated prostaglandin levels lead to an increased thermal set point and a resulting increase in thermogenesis, decrease in heat loss, and elevated T_b.

It is unlikely that blunted fever in old individuals is due to attenuated cytokine levels. LPS-induced cytokine levels actually increase with age, both in plasma and in brain. For instance, plasma and cerebrospinal fluid levels of TNF- were significantly higher in aged rats and mice (7, 9). Old rodents also showed high levels of IL-6 compared with young ones (9, 18). It would appear that the endotoxin-induced pyrogenic signal is present in old rats. However, if the initial pyrogenic signal is present, then why do old rats not develop high fevers? One explanation may be that in an older population, the pyrogenic signal stimulated by the infection may not reach the thermoregulatory control centers in the hypothalamus and as a result there will be no alteration in the thermal set point and thus no fever. Alternatively, in the event that the pyrogenic signal does reach the brain, it is possible that the signal is not transduced into an elevated set point or that the thermoeffectors are not capable of increasing T_b in response to the altered set point.

We have recently tested several of these possibilities. Results from our laboratory have demonstrated that a pyrogenic signal (i.e., PGE₂ or IL-1), when administered centrally, is appropriately transduced into an elevation in thermal set point in old rats resulting in high fever responses (12, 15). In addition, the febrile temperatures attained in the old rats were equivalent to or higher than those seen in the young febrile rats. The data indicate that 1) a centrally administered pyrogenic signal is appropriately transduced into an elevated set point, 2) the increase in set point leads to the stimulation of thermoeffectors, and 3) thermoeffectors in old rats are entirely capable of increasing T_b in response to the stimulus. Therefore, it appears that old rats do retain the capacity to respond to central administration of cytokines with fever.

Whereas there are an increasing number of studies investigating potential age-related effects on autonomic aspects of fever production, particularly activation of brown adipose tissue (8), few studies have focused on behavioral aspects of temperature regulation. Here, we demonstrate that in the case of old rats, behavioral selection of T_a also may have a significant impact on fever production. Elucidating the underlying mechanisms involved in the age-related reduction in fever during infection may provide not only a more appropriate and timely diagnosis of infection and augment survival in the elderly, but it will also contribute to our basic and fundamental understanding of the aging process.