Impaired sympathetic nerve function in the inflamed rat intestine

GASTROENTEROLOGY 1991;100:675-682

Impaired Sympathetic Nerve Function in the Inflamed Rat Intestine M A R K G. SWAIN, PATRICIA A. BLENNERHASSETT, a n d STEPHEN M. COLLINS Intestinal Diseases Research Unit, McMaster University, Hamilton, Ontario, Canada

The effect of intestinal inflammation on norepinephrine release from the myenteric plexus in the Trichinella s p i r a l i s - i n f e c t e d rat was assessed. Longitudinal muscle-myenteric plexus preparations were preincubated with [3H]norepinephrine and release was evoked by electrical field stimulation and KCI administration. Preincubation of preparations with desipramine or pretreatment of rats with 6-hydroxydopamine significantly suppressed the uptake and evoked release of [3H]norepinephrine; electrical field stimulation but not KCl-evoked release of [3H]norepinephrine was sensitive to tetrodotoxin. These results confirm the presence of functioning sympathetic nerves in the preparations. T. spiralis infection was associated with significant suppression of both electrical field stimulation and KCIevoked release of [3H]norepinephrine on the sixth day postinfection, and the suppression persisted 100 days postinfection. No suppression of [3H]nor.epinephrine release was seen in the worm-free and noninflamed ileum of infected rats. Suppression of [3H]norepinephrine release from the jejunum of infected rats was attenuated by treatment with betamethasone (3.0 mg/kg SC daily). These results are consistent with the hypothesis that intestinal inflammation suppresses the release of norepinephrine from the myenteric plexus in the Trichinella-infected rat. 'ntestinal inflammation in humans is associated .with altered intestinal motility (1,2). The mechalnisms underlying the changes in motility are poorly understood but could reflect changes in smooth muscle, enteric intrinsic or extrinsic nerves, or the release of gut hormones. Inflammation in the gut is associated with structural damage to neurons in the myenteric plexus of patients with inflammatory bowel disease (3,4). Sympathetic nerves are known to play a modulatory role in the control of gastrointestinal motility, and sympa-

thetic nerve fibers are present in the myenteric plexus (5-7). Trich&ella spiralis is an intraepithelial nematode parasite that inhabits the proximal small intestine in rats. Primary infection with the parasite is associated with an inflammatory infiltrate in the mucosa and submucosa of the jejunum (8). The intestinal phase ends with expulsion of the worms from .the gut within 17 days. The enteric phase of the infection is associated with accelerated small bowel transit (9) and changes in intestinal myoelectric activity (10). The demonstration that changes in propulsive activity occur in extrinsically denervated jejunal segments of Trichinella-infected guinea pigs (11) suggests that the infection alters the properties of smooth muscle and/or intrinsic nerves. Recent studies have shown that jejunal longitudinal muscle from Trichinellainfected rats generates more active tension per unit of cross-sectional area than muscle from control rats (12). Our laboratory recently described changes in intrinsic cholinergic nerves in the jejunum of Trichinellainfected rats. Specifically, we demonstrated marked suppression of acetylcholine release from the jejunal myenteric plexus of infected rats (13). The extent to which this finding applies to the release of other neurotransmitters is not presently known. To evaluate this, we investigated the release of the sympathetic neurotransmitter norepinephrine (NE) from the myenteric plexus of Trichinella-infected rats. Because changes in host function may occur either as a result of a direct effect of the parasite on host tissue, or as a result of the inflammatory reaction of the host, we also examined [3H]NE release in a gut region remote from Abbreviations used in this paper: EFS, electrical field stimulaI tion; LM-MP, longitudinal muscle-myenteric plexus; MPO, myeloperoxidase; NE, norepinephrine;PI, postinfection; TTX, tetradotoxin. © 1991 by the American GastroenterologicalAssociation 0016-5085/91/$3.00

676 SWAINET AL.

the p r i m a r y habitat of the worm, a n d in the jejunum of rats in w h i c h the inflammatory response had been s u p p r e s s e d by corticosteroid treatment.

Materials and Methods

Infection of Rats by Trichinella spiralis Rats were infected as previously described by our laboratory (12). Briefly, male Sprague-Dawley rats (200250 g) were each infected with 7500 larvae suspended in 1 mL of 1% agar and administered by garage. Noninfected, age-matched rat~ were used as controls; these animals were not garaged.

Norepinephrine Release Experiments Rats were killed by a blow to the head. A 4-cm segment of jejunum was removed, starting at the ligament of Treitz, and placed over a glass rod. Longitudinal musclemyenteric plexus (LM-MP) preparations were obtained as described previously by this laboratory (13). The method for loading the tissue with [SH]NE was adapted from that described by Wu and Gaginella (14). Briefly, the LM-MP preparations were tied at each end and placed in 1 mL of buffer containing (in mmol/L) NaC1, 120.9; KC1, 5.9; CaC12, 2.5; MgC12, 1.2; NaHCO 3, 15.5; NaH=PO~, 1.2; glucose, 11.1; ethylenediaminetetraacetic acid (EDTA), 0.004; ascorbic acid, 0.11; and pargylline (a monoamine oxidase inhibitor), 0.03; bubbled with 95% carbon dioxide and 5% oxygen. [SH]Nor~pinephrine, 0.5 ~.mol/L (15 Ci/mmol), was then added and the preparation was incubated at 37°C for 40 minutes with oxygenation. The method used for superfusing the preparation has been described previously (13). Briefly, after the incubation the tissue was washed twice in 10 mL cold buffer (4°C), blotted dry, then suspended in a water-jacketed superfusion chamber maintained at 37°C. The tissue was superfused with the buffer at 37°C at a flow rate of I mL/min using a peristaltic pump, and fractions were collected every 2 minutes using an Altrorac 7000 fraction collector (L.K.B., S.carborough, Canada). Three milliliters of scintillation fluid (formula 963; Du Pont, Wilmington, DE) was added to the 2-mL fractions and the tritium content was measured using a Beckman liquid scintillation counter (model LS 5801; Beckman Instruments Inc., Fullerton, CA) at a counting efficiency of 35%. At the end of the experiment, the tissue preparations were blotted dry, weighed, and solubilized with l mL Pr0tosol tissue solubilizer (New England Nuclear, Boston, MA). The sample was then neutralized with 50 p,L glacial acetic acid, then 3 mL counting fluid was added and the samples were counted to determine the total residual tritium content of the tissue. The release of [3H]NE was stimulated by exposure of the tissue to 50 mmoEL KCI for 4 minutes or by electrical field stimulation (EFS) achieved by placing the tissue between two parallel silver electrodes (0.15 mm x 70 mm) and stimulating at 30 V for 0.5 milliseconds at a frequency of 10 Hz for 1 minute (S48 stimulator; Grass, Quincy, MA). Agents such as KC1 or tetrodotoxin (final concentration of 1 p,g/mL) were added to the superfustate buffer. In other

GASTROENTEROLOGYVol. 100, No. 3

experiments, the tissue was preincubated with desipramine at a final concentration of 10 ~mol/L for 10 minutes. [SH]Norepinephrine release was also examined in rats pretreated with 6-hydroxydopamine (50 mg/kg IV for 2 days) 1 week before the study. Betamethasone (3 mg/kg) was administered daily by SC injection for 6 days to infected or control rats. All experiments were performed on the sixth day post_infection (PI) except in time course studies, in which experiments were also performed on the 23rd, 40th, and 100th days PI. The evoked release of tritium was derived from the difference between the total amount of tritium released during stimulation periods and the spontaneous release of tritium. The latter was calculated as the average of the dpm released immediately before stimulation and the dpm just after the release had returned to a constant level following the stimulated release. The stimulated release of tritium was expressed as a percentage of the total amount of tritium present in the tissue at the time of stimulation (13,14).

Myeloperoxidase Assay The myeloperoxidase assay was performed as described by Wallace (15). Briefly, the mucosa was removed from a 2-cm segment of jejunum, starting at the ligament of Treitz, by scraping the mucosal surface with a glass slide applied with gentle pressure. The mucosa thus obtained was weighed, added to hexadecyltrimethylammoniumbromide buffer (14 mmo]/L),and homogenized on ice using a Polytron (Kinematica, Roxdale, Canada) for 15 seconds at a setting of 5. The sample was then vortexed and 1-mL a]iquots were decanted and centrifuged for 2 minutes at 12,000 rpm. An aliquot of 0.1 mL was then added to 2.9 mL of O-dianisidine solution (16.7 mg O-dianisidine dihydrochloride, 90 mL water, 10 mL 50 mmo]/Lpotassium phosphate buffer, pH 6.0, and 50 p.L 1% hydrogen peroxide). The samples were read in a Gilford spectrophotometer (Ciba-Corning, Richmond Hill, Canada). Myeloperoxidase activity was calculated as the slope of the absorbance divided by 0.0113, divided by the weight of mucosa] tissue in the preparation in milligrams,and was expressed as units of myeloperoxidaseactivity per milligramprotein (15).

Statistical Treatment of Data All studies involved at least three animals; results are expressed as mean _ SEM, and n indicates the number of rats used in a given experiment. When comparing two groups the Student's t test was used; an analysis of variance (ANOVA) was used for comparisons of more than two groups.'Statistical significance was achieved if the P value was

<0.05.

Materials Pargyline, 6-hydroxydopamine, ascorbic acid, EDTA, desipramine, tetrodotoxin, O-dianisidine dihydrochloride, and hexadecyltrimethylammonium bromide were obtained from Sigma Chemicals, St. Louis, MO; NaCI, CaCl 2, MgClz, NaHCOs, NaHzPO4, KI-IzPO4, K2HPO4, KCI., glucose, and

March 1991

SYMPATHETIC NERVE FUNCTION 677

acetic acid were obtained from Mallincourt, Paris, KY; tritiated NE (15 Ci/mmol) and protosol tissue solubilizer were obtained from New England Nuclear; formula 963 liquid scintillation fluid was obtained from Du Pont; and betamethasone was obtained from Schering Canada Inc., Pointe Claire, Quebec, Canada. Results

In control rats, the weights of LM-MP preparations ranged from 0.016 to 0.034 g (mean, 0.024 _+ 0.003 g) and the mean uptake of [3H]NE by preparations was 2.60 _+ 0.33 x 108 dpm, or 118.0 _+ 20.49 x 108 dpm/g. Preliminary experiments determined that the largest and most reproducible field stimulated [3H]NE release was obtained in tissues from control rats by stimulating at 30 V for 0.5 milliseconds at a frequency of 10 Hz for 1 minute. At these parameters the release of [3H]NE w a s 3.45% -+ 0.61%. Preliminary experiments also determined that the most reproducible release of [3H]NE induced by potassium depolarization occurred with 50 mmol/L KCI and was found to be 2.85% _+ 0.34%. We first examined the effect of tetrodotoxin (TTX), a neuronal sodium channel-blocking agent (16), on EFS- and KCl-induced release of [3H]NE (Figure 1). In the presence of TTX (1 p.g/mL), the release of [3H]NE evoked by EFS was suppressed by 89.4% (P < 0.05), whereas that evoked by KCI was unaltered. Desipramine is a selective noradrenergic neuronal uptake inhibitor (17). Preincubation of the jejunal LM-MP preparations with desipramine (10 mmol/L)

4.

_T_

A

,-.t '-r"

3-

u.

o LU U')

2-

UJ --I ILl UJ

1'

o

:> LU

~T EI~S

KCI

Figure 1. Effect of T r x (1 l~g/mL)and desipramine (10 nunol/L) on the release of3H inducedby EFS (30 V, 0.5 millisecond, 10 Hz for 1 minute) and KCI (50 mmoFL) in jejunal LM-MP preparations from uninfected control rats. l~, Experiments performed in the presence of T r x ; M, experiments involving desipramine. The evoked release of=H is expressed as a percentage of the total radioactivity present in the tissue, as described in Materials and Methods. Results shown are the mean +_ SEM from four separate experiments. *Significant difference from the respective control

(P < 0.05).

significantly inhibited the uptake of [3H]NE by the preparation; values for [3H]NE uptake fell by 83% from 2.60 +_ 0.33 x 10 s dpm for control to 4.30 _+ 0.67 x 10 s dpm for the desipramine-treated preparations (P < 0.05). As illustrated in Figure 1, desipramine pretreatment was also associated with a significant (77%) reduction in the fractional release of [3H]NE evoked by KC1 from the LM-MP prpparations. Treatment with 6-hydroxydopamine has been shown to deplete adrenergic nerves of NE stores (18). Chemical sympathectomy of rats with 6-hydroxydopamine was associated with a significant (80%) decrease in the uptake of [3H]NE into the preparations. Uptake was 2.60 +_ 0.33 x 10 e dpm for control rats and 5.30 _+ 0.97 x 10 s dpm for the preparations from 6-hydroxydopamine-treated rats (P < 0.05). In the 6-hydroxydopamine-treated animals release of [3H]NE evoked by KC1 was also significantly suppressed by 70% (P < 0.05) from 2.85% _+ 0.34% to 0.86% _+ 0.32%. The jejunal LM-MP preparations from rats infected 6 days earlier with T. spiralis were significantly heavier than those from controls and ranged from 0.061 to 0.085 g (mean, 0.075 +_ 0.005" g). The mean uptake of [3H]NE by LM-MP preparations from infected animals was 2.70 _+ 0.34 x 108 dpm, or 36.80 _+ 5.49 x 106 dpm/g. Infection of rats 6 days earlier with T. spiral& was associated with a significant reduction in [3H]NE release evoked by EFS and KC1; values for the evoked release by these stimuli were reduced by 59.7% and 73.3%, respectively (Figure 2), compared with controis (P < 0.05). As illustrated in Figure 2, the release of [3H]NE evoked by EFS in preparations from infected animals showed a pattern of TTX sensitivity similar to that exhibited by control animals. However, although TTX did not inhibit the KCl-induced response in control animals (Figure 1), it did significantly inhibit the response to KC1 in tissues from infected rats (Figure 2). We next examined whether the suppression of [3H]NE release was reversible 23, 40, and 100 days after rats had recovered from the infection by T. spiralis. As shown in Table 1, the amount of [3H]NE taken up by the preparations per gram weight increased over this period. However, as shown in Figure 3, the fraction of the [3H]NE released after stimulation by EFS or KC1 remained significantly suppressed relative to control for as long as 100 days PI (P < 0.05), when values were no different from those seen 6 days PI (P > 0.05). We next examined the relationship between changes m [3H]NE release and the presence of mucosal inflammation in Trichinella-infected rats and used myeloperoxidase activity to monitor the inflammatory response. As s h o w n in Figure 4, on the sixth day PI .

t

678

SWAIN ET AL.

GASTROENTEROLOGY Vol. 100, No. 3

,o

4-

"/r A

,<

I

3:

"I-

U.

±

o

o U.I

(n < ILl -I

W g) < W

-k

2.0'

T

a-

UJ n-

Q u,I v. O >

_T_

3.0'

U.

r~ W

1'

1.0'

o> W

ill

o.o O' EFS

KCI

Figure 2. The evoked release of 3H induced by EFS {30 V, 0.5 millisecond, 10 Hz for 1 minute) and KCI (50 mmol/L) from LM-MP preparations of rats infected 6 days earlier with T. spiralis. O, Experiments performed on control tissues: ~ , experiments performed on tissue from infected rats; B, experiments performed on tissue from infected rats in the presence of T r x (1 p.g/ml). The evoked release of 3H is expressed as a percentage of the total radioactivity present in the tissue, as described in Materials and Methods. Results shown are the mean ± SEM from five separate experiments from controls, five from infected rats, and three from infected rats plus TTX. *Significant difference from control; **significant difference between the presence and absence ofTFX (P < 0.05).

there was no significant increase in myeloperoxidase (MPO) activity in the mucosa of the ileum. Although the uptake of [3H]NE by ileal preparations was greater in infected animals than in controls, the weights of tissues from infected animals were greater (Table 1). However, the uptake of [3H]NE pez: gram weight of the preparations was similar between control and infected animals (Table 1). The release of [3H]NE evoked by EFS or KCI was no different between control animals and those infected 6 days earlier with T. spiralis, as shown in Figure 4. The release of [3H]NE evoked by EFS was 0.93% _ 0.11% in control rats and 1.18% _ 0.08% in rats on day 6 PI (P > 0.05). Table 1. Uptake of ffH]Norepinephrine by Longitudinal Muscle-Myenteric Plexus Preparations

Jejunum Day 0 Day 6 PI Day 23 PI Day 40 PI Day 100 PI Steroid-treated Day 0 Day 6 PI Ileum Day 0 Day 6 PI

[3H]NE uptake

Tissue weight

Normalized uptake

(1 X 106dpm)

(g)

(1 X 106dpm/g)

2.6 ± 0.3

1.7 ± 0.31 1.68 ± 0.17

0.024 0.075 0.069 0.040 0.019

0.003 0.005 0.010 0.002 0.002

116.0 ± 20.5 36.8 ± 5.5 41.6 ± 12.5 42.7 ± 5.0 69.4 ± 14.5

0.92 - 0.14 2.79 ± 0.33

0.011 ± 0.002 0.052 ± 0.003

61.0 ± 12.3 65.3 ± 10.3

0.65 ± 0.09 1.43 ± 0.12

0.Q23± 0.003 0.045 ± 0.003

28.5 ± 1.7 32.8 ± 5.6

2.7 ± 0.3 2.7 ± 0.4

± ± ± ± ±

6

(~

100

2'3 4 0 EFS

(~

6

i!i 23

40

100

KCL

Figure 3. Effect of 7". spiralis infection on EFS (30 V, 0.5 millisecond, 10 Hz for I minute)- and KCI (50 mmol/L)-evoked release of 3H from jejunal LM-MP preparations of control rats and rats infected 6, 23, 40, and 100 days earlier. The numbers on the horizontal axis reflect days PI. Control responses from noninfected rats are indicated by time 0 on the horizontal axis. The results are means ± SEM o f at least three experiments. *Significant difference from control (P < 0.05).

KCl-evoked release of [3H]NE was 0.71% _ 0.12% in controls and 0.97% _ 0.10% in infected rats (p > 0.05). Treatment of rats with betamethasone (3 mg/kg SC daily for 6 days) suppressed MPO activity in the jejunum of noninfected control rats and prevented the increase in MPO activity in Trichinella-infected rats (Figure 5). In betamethasone-treated rats, tissue weights and rates of [3H]NE uptake were higher in the infected tissues than in controls (Table 1), but the 3¢: ®

0}

U.I Z I U.

.=

2-

o

m (0

o< o

B

0.

. EFS

. KCI

Figure 4. A. Myeloperoxidase activity in mucosal preparations from the ileum of control ([7) and day 6 Trichinello-infected rats ([]). Myeloperoxidase activity is expressed in units per milligram of protein as described in Materials and Methods, and the results shown are the means ± SEM of six experiments. B. The evoked release of~H induced by EFS (30 V, 0.5 millisecond, 10 Hz for I minute) and KCI (50 retool/L) from LM-MP preparations of the ileum of rats infected 6 days earlier with T. spirolis. 0, Experiments performed on control tissues; [], experiments performed on tissue from infected rats. The results shown are the means _+ SEM of four experiments.

March 1991

SYMPATHETIC NERVE FUNCTION

679

4A C O

,< 3-

3I-

kl m

P

2-

>-

T

>

1-

LL O LU U) ,< UJ .J UJ n-

2

UJ

1

T

T

0 > UJ

6n" I UNTREATED

--r-

I,///--f~///i

Figure 5. Myeloperoxidase activity in mucosal preparations from the jejunum of control and day 6 Trichlnello-infected rats, in the presence or absence ofbetamethasone pretreatment {3.0 mg/kg SC daily}. [], Results from infected animals. The results are means + SEM of at least three experiments. *Significant difference from the untreated control (P < 0.05).

ratio of uptake to weight was nbt different between the two groups. Betamethasone'treatment abolished the decrease in [3H]NE release associated with T. spiralis infection 6 days earlier. The release of [3H]NE evoked by EFS was 1.74% ± 0.30% in steroid-treated controls and 2.15% ± 0.22% in infected rats (Figure 6). This contrasts with the 60% suppression of NE release observed in untreated infected rats (Figure 6). A similar profile was observed with KCl-induced release of [3H]NE, as shown in Figure 7. In steroid-treated control rats,

T

A v "I"

LL O UJ U) < UJ

UNTREATED

BETAME3~HASONE TREATED

BETAMETHASONE TREATED

Figure 7. Effect of betamethasone treatment on the evoked release of~H induced by KCi (50 mmol/L) from LM-MP preparations of control and Trichinella-infected rats. 17, Responses from noninfected rats; El, responses from infected rats. The evoked release of 3H is expressed as a percentage of the total radioactivity present in the tissue, as described in Materials and Methods. Results shown are the mean -+ SEM from four separate experiments in untreated rats and seven experiments in betamethasonetreated rats. *Significant difference from respective control

{P < o.o5}.

KCl-evoked release was 1.42% _ 0.20%; this value fell to only 1.21% ± 0.21% in steroid-treated infected rats (P > 0.05). Betamethasone treatment was also associated with a decrease in the release of [3H]NE from noninfected control rats, as shown in Figures 6 and 7. In the absence of steroid treatment, values for NE release were 3.45% ± 0.61% for EFS and 2.85% ± 0.34% for KC1. These values fell by about 50% in betamethasone-treated control rats (P < 0.05 for each data pair). Discussion

3'

2,

UJ n(3 UJ O > UJ UNTREATED

BETAMETHASONE TREATED

Figure 6. Effect of betamethasone treatment on the evoked release of 3H induced by EFS from LM-MP preparations of control and Trichinella-infected rats. [3, Responses from noninfected rats; ~}, responses from infected rats by the lmtched bars. The evoked release of 3H is expressed as a percentage of the total radioactivity present in the tissue, as described in Materials and Methods. Results shown are the mean -+ SEM from four separate experiments in untreated rats and seven experiments in betamethasone-treated rats. *Significant difference from respective control (P < 0.05).

The results of this study suggest that T. spiralis infection is not associated with impaired uptake of NE by myenteric plexus neurons in the inflamed jejunum. The uptake of [3H]NE was similar in the jejunum of control rats and of infected rats 6 days after infection. This interpretation is based on the absolute uptake of [3H]NE rather than the uptake of [3H]NE per unit weight of tissue. Because infection was associated with a twofold increase in the weight of the preparations, normalization for weight would have resulted in an apparent decrease in the uptake of the neurotransmitter by tissues from infected rats. Normalizing these data for tissue weight assumes that there is a proportionate increase in nerve mass in tissues from infected rats, and we have no information regarding this. Others have reported an increase in muscle mass in the jejunum of nematode-infected rats (19), and we have identified both hypertrophy and hyperplasia of longitudinal muscle in T. spiralis-infected rats (20).

680 SWAINET AL.

Therefore, it is likely that the twofold increase in weight of the preparations from infected rats observed in the present study is attributable primarily to an increase in muscle mass, and for this reason we elected not to normalize NE uptake to the weight of the tissue. These reservations do not apply to our results on the release of NE because they are expressed as fractions of the total amount of [3H]NE present in the preparation at the time. Our results are consistent with those of other studies that have characterized the release of NE from sympathetic nerves in the gut (14,21,22). Evidence in support of this is the demonstration that [3H]NE uptake is prevented by treatment with 6-hydroxydopamine, which produces a chemical sympathectomy (18), and by ]~rior exposure of the tissue to the selective uptake inhibitor desipramine (17). Evidence in support o~ a neural origin for the released [3H]NE derives not only from the substantial decrease in NE release after administration of 6-hydroxydopamine or desipramine, as observed in other studies (21,22), but also from the virtual abolition of EFS-evoked [3H]NE release in the presence of TTX. Because TTX is a neuronal sodium channel blocker, it would not be expected to alter neurotransmitter release induced by depolarization, and the resistence to TTX of KC1evoked NE release in control tissues is in keeping with this. However, we cannot explain the apparent ability of TTX to inhibit the response to KC1 in tissues from infected rats. The release of NE was unimpaired in the ileum, a region that contains no worms and no inflammatory infiltrate on day 6 PI (23,24). This finding indicates that the suppression of NE release is related either to the presence of the parasite in the gut lumen or to the inflammatory infiltrate in the mucosa and lamina propria. The localization of myenteric plexus changes to the jejunum in this model are in contrast to those fohnd in the epithelium and muscle in the nematodeinfected rat. Changes in muscle function described in the jejunum of the Trichinella-infected rat (12) were also found in gut segments excluded from the rest of the gut before the infection (24). Similarly, increased tension generation by gut muscle in the Nippostrongylus brasiliensis-infected rat was also found in vascular smooth muscle (19). In addition, changes in epithelial cell mitotic activity occur along an aboral gradient as far as the worm-free colon in the N. brasiliensis-infected rat (25). In light of the fact that changes in NE release were found only in the region of the maximum worm burden, we must consider the possibility that the changes were related to the presence of the worms. The finding by Sukhdeo and Mettrick (26) that the enhancement of glucose uptake found during infection by Trichinella could be transferred by adding

GASTROENTEROLOGYVol. 100, No. 3

intestinal washings to the intestinal perfusate of noninfected control rats raised the possibility that changes in host function might occur in response to factors elaborated by the parasite. It is established, for example, that nematode parasites produce active substances such as aliphatic amines (27), a variety of antigens (28), cholinesterases (29), and several acids (30) capable of altering function in the host. Although we demonstrated changes in NE release only in the jejunum, we believe that such changes are unlikely to be caused by parasite-derived factors in light of our results using corticosteroids. Betamethasone treatment was used to investigate the extent to which the observed changes in sympathetic nerve function were caused by the inflammatory response, and MPO activity was used to monitor the antiinflammatory effect of the steroid. The activity of this enzyme closely parallels the influx of myeloid cells (15,23,31-33) and has been positively correlated with other aspects of altered host responses in the nematode-infected rat (23,34). Betamethasone abolished the increase in MPO activity caused by the infection and prevented the suppression of NE release seen in untreated infected rats. Because we have shown that this administration of betamethasone under identical conditions does not redistribute the worms along the intestine on day 6 PI and in fact increases the worm number (24), these results indicate that the suppression of NE release occurs as a result of the inflammatory process. Similar conclusions have been drawn by others using this strategy to investigate intestinal transit changes in the Trichinellainfected mouse in vivo (35) and in vitro smooth muscle changes in the rat infected by N. brasiliensis (36) or T. spiralis (23). The mechanisms underlying the inflammation-induced changes in sympathetic nerve function in this model remain to be elucidated but may involve inflammatory mediators such as prostaglandins (20) and leukotrienes (37,38}, which have been shown in other systems to modulate nerve function. A relationship between sympathetic nerves and inflammation has been identified in other systems. In experimentally induced arthritis in the rat, depletion of catecholamines by guanethidine or reserpine attenuated the severity of joint injury {39}. In addition, arthritis was found to be more severe in spontaneously hypertensive rats that have increased sympathetic tone (40}. The impact of impaired sympathetic neurotransmission on the development of inflammatory changes in the Trichinella-infected rat intestine remains to be elucidated. With respect to inflammatory bowel disease, adrenergic nerves were found to be more prominent in rectal mucosal biopsy specimens obtained from patients with ulcerative colitis (41}. Changes included larger-diameter varicosities

March 1991

and increased intensity of fluorescence. However, the functional integrity of the adrenergic nerves in these patients was not examined. In addition, changes in the mucosa may not necessarily reflect alterations in the myenteric plexus. The implications of this and our previous study (13) are that an inflammatory process that is largely restricted to the mucosa and lamina propria and does not overtly infiltrate the myenteric plexus or deep muscle layer (8,9} can influence the release of neurotransmitters from deep enteric neurons. Because the enteric nervous system plays an important role in the organization and propagation of motor activity, it is possible that localized areas of mucosal inflammation can result in the alteration of gastrointestinal motility over more extensive segments. It is of interest that steroids attenuated MPO activity and NE release in uninfected rats. This may reflect a direct effect of the corticosteroid on nerve function. Alternatively, it is also possible that the changes could be caused by a corticosteroid effect on resident inflammatory or immune cells in the gut of control rats. If this is indeed the case, one might speculate that enteric nerve'function is subject to the influence of inflammatory cells normally present in the clinically uninflamed gut.

SYMPATHETIC NERVE FUNCTION 681

13.

14.

15.

16. 17.

18.

19.

20.

21.

22.

References 1. Kern F, Almy TP, Abbot FK, BogdonoffMD. The motility of the distal colon in non-specific ulcerative colitis. Gastroenterology 1951;19:492-502. 2. Snape WJ Jr, Matarrazo SA, Cohen S. Abnormal gastrocolonic response in patients with ulcerative colitis. Gut 1980;21:392398. 3. Dvorak. AM, Connell AB, Dickersin GR. Crohn's disease, a scanning electron microscopic study. Hum Patho11979;10:165177. 4. Storsleen KA, Kernohan JW, Bargen JA. The myenteric plexus in chronic ulcerative colitis. Surg Gynecol Obstet 1953;97:335343. 5. Llewellyn-Smith IJ, Furness JB, O'Brien PE, Costa M. Noradrenergic nerves in human small intestine: distribution and ultrastructure. Gastroenterology 1984;87:513-529. 6. Furness JB, Costa M. The adrenergic innervation of the gastrointestinal tract. Brgeb Physiol 1974:69:1-51. 7. Mirkin BL, Bonnycastle DD. A pharmacological and chemical study of humoral mediators in the sympathetic nervous system. Am J Physiol 1954;178:529-534. 8. Mogbel R. Helminth-induced intestinal inflammation. Trans R Soc Trop Med 1986;80:719-725. 9. Castro GA, Badial-Aceves F, Smith JW, Dudrick SJ, Weisbrodt NW. Altered small bowel propulsion in association with parasitism. Gastroenterology1976;71:620-625. 10. Palmer JM, Weisbrodt NM, Castro GA. Tmchinella spimlis: intestinal myoelectric activity during enteric infection in the rat. Exp Parasito11984;57:132-141. 11. Alizadeh H, Castro GA, Weems WA. Intrinsic jejunal propulsion in the guinea pig during parasitism with Trichinella spirolis. Gastroenterology 1987;93:784-790. 12. Vermillion DL, Collins SM. Increased responsiveness of jejunal

23. 24.

25.

26.

27.

28.

29.

30.

longitudinal muscle in Trichinella-infected rats. Am J Physiol 1988;254:G124-G127. Collins SM, Blennerhassett PA, Blennerhassett MG, Vermillion DL. Impaired acetylcholine release from the myenteric plexus of Trichinella-infected rats. Am J Physiol 1989;257:G898G903. Wu ZC, Gaginella TS. Release of ~H-norepinephrine from nerves in the rat colonic mucosa: effects of norepinephrine and prostoglandin E,. Am J Physiol 1981;241:G416-G421. Wallace ~L. Glucocorticoid-induced gastric mucosal damage: inhibition of leukotriene but not prostoglandin biosynthesis. Prostaglandins 1987;34:311-323. Kao CY. Pharmacology of tetrodotoxin and saxitoxin. Fed Proc 1972;31:1117-1123. Baldessarini RJ. Drugs and the treatment of psychiatric disorders. In: Goodman-Gilman A, Goodman LS, Gilman A, eds. Goodman and Gilmans: the pharmacological basis of therapeutics. 6th ed. New York: Macmillan, 1980:418-427. Malmfors T, Sachs C. Degeneration of adrenergic nerves produced by 6-hydroxydopamine. Eur J Pharmacol 1968;3:8992. Farmer SG, Brown JM, Pollock D. Increased responsiveness of intestinal and vascular smooth muscle to agonists in rats infected with Nippostrongylus brosiliensis. Arch Int Pharmacodyn Ther 263:217-227, 1983. Vignjevic P, Blennerhassett MG, Blennerhassett P, Vermillion DL, Collins SM. Hypertrophy and hyperplasia of smooth muscle in the inflamed intestine {abstr}..Gastroenterology 1989;96:A530. Wu ZC, Kisslinger SD, Gaginella TS. Functional evidence for the presence of cholinergic nerve endings in the colonic mucosa of the rat. J Pharmacol Exp Ther 1982;221:664--669. Wu ZC, Gaginella TS. Functional properties of noradrenergic nervous system in rat colonic mucosa: uptake of 3H-norepinephfine. Am J Physio11981;241:G137-G142. Smith IW, Castro GA. Relation of peroxidase activity in gut mucosa to inflammation. Am J Physio11978;234:R72-R79. Marzio L, Blennerhassett P, Chiverton S, Vermillion DL, Longer J, Collins SM. Altered smooth muscle function in worm-free gut regions of Trichinella-infected rats. Am J Physiol 1990;259: G306-G313. Symons LEA. Epithelial cell mitosis and morphology in the worm free regions of the intestine of the rat infected by Nippostrongylus brasiliensis. J Parasito11978;64:958-959. Sukhdeo MVK, Mettrick DF. Effect of size of Trichinella spirolis infection on glucose and ion transport in the rat intestine. J Parasito11984;70:499-506. Castro GA, Ferguson JD, Gordon CW. Amine excretionin the encysted larvae and adults of Trichinelln spirolis. Comp Binchem Physio11973;45A:819-828. Olson LJ, Richards B, Ewert A. Detection of metabolic antigens from nematode larvae by microculture-agar gel technique. Tex Rep Biol Med 1960;18:254-259. Yeates RA, Ogilvie BM. Nematode acetylcholinesterases. In: Van den Bossche H, ed. Biochemistry of parasites and their host relationships. New York: North Holland, 1976:307-310. Ferguson JD, Castro GA. Comparative metabolism in Trichinella spiralis larvae and adults. Am J Physiol 1973;225: 85-89.

31. Bradley PP, Priebat DA, Christensen RD, Rothstein G. Measurement of cutaneous inflammation: estimation of neutrophil content with an enzyme marker. J Inv Dermatol 1982;78:206209.

32. Castro GA, Roy SA, Stockdill RD. Trichinellu spiralis: peroxidose activity in isolated cells from the rat intestine. Exp Parasito11984;36:307-315. 33. Krawisz JE, Sharon P, Stenson WF. Quantitative assay for acute

682

34.

35. 36.

37.

38.

SWAIN ET AL.

intestinal inflammation based on myeloperoxidase activity. Gastroenterology 1984;87:1344--1350. Russell DA, Castro GA. Physiology of the gastrointestinal tract in the parasitized fiost. In: Johnson LR, ed. Physiology of the gastrointestinal tract. 2nd ed. New York: Raven, 1987;17491760. Sukhdeo MVK, Croll NA. Gut propulsion in mice infected with Trichinello spirolis. J Parasito11961;67:906-910. Farmer SG. The effect of betamethasone on altered responsiveness of isolated intestine from rats infected with Nippostrongylus brasiliensis (abstr). Br J Pharmacol 1962;76:192. Gazelius B, Panopoulos P, Odlander B, Claesson HE. Inhibition of inradental nerve excitability by leukotriene B4 and C4. Acta Physiol Scand 1984;120:141-143. Collins SM, Daniel EE, Fox JET. The effect of leukotriene D4 on gastrointestinal motility (abstr). Gastroenterology 1986;90: 1377A.

GASTROENTEROLOGY Vol. 100, No. 3

39. Levine JD, Moskowitz MA, Basbaum AI. The contribution of neurogenic inflammation in experimental arthritis. J Immunol 1985;135:843S-647S. 40. Levine JD, Dardick SJ, Roizen MP, Helms C, Basbaum AI. Contribution of sensory afferents and sympathetic efferents to joint injury in experimental arthritis. J Neurosci 1986;6:34233429. 41. Kyosola K, Penttila O, Salaspuro M. Rectal mucosal adrenergic innervation and enterochromaffin cells in ulcerative colitis and irritable colon. Scand J Gastroentero11977:12:363-367.

Received February 12, 1990. Accepted August 23, 1990. Address requests for reprints to: Stephen M. Collins, M.D., Intestinal Diseases Research Unit, Room 3N5C, McMaster University Medical Centre, Hamilton, Ontario LSN 3Z5, Canada.