Introduction

Top Abstract Introduction Materials and methods Results Discussion References

The existence of a relationship between environmental, specifically stressful, stimuli and the immune system has been documented by anecdotal (Cannon, 1935) as well as epidemiological data. These latter indicate that acute stress impairs immune functions, worsens autoimmune syndromes, and increases susceptibility to infections and neoplasias (Ader et al., 1991). Stimulus-dependent alterations of immune functions are indicated by experimental data as well. The proliferative response of immunocompetent cells is reduced by inescapable, but increased by escapable, stress (Laudenslager et al., 1983; Shavit et al., 1984); according to the latter authors, these two forms of stress act via opioid and nonopioid mechanisms, respectively. Similarly, repeated shock enhances the phagocytic activity of polymorphonuclear lymphocytes (Shurin et al., 1994) and reduces proliferative response and natural killer activity (Sacerdote et al., 1994); however, these parameters are not modified by the application of a single stimulus (Sacerdote et al., 1994; Shurin et al., 1994). Stress has been described as activating the hypothalamo-pituitary-adrenal axis and as modifying the distribution of immunocompetent cell populations (Dhabar et al., 1996). At least some of the effects of stress on immune functions appear to be mediated by opioid peptides: shock increases opioid peptide concentration in the central nervous system (Madden et al., 1977), in blood (Madden et al., 1977), and within immunocompetent cells (Sacerdote et al., 1994). Stress-associated analgesia, which is partially reversed by naloxone (Akil et al., 1976), has therefore been attributed to the induced increase in opioid peptide concentration (Watkins and Mayer, 1987). Similarly, shock-induced immunosuppression appears to be mediated by the opioid peptides released under these conditions (Shavit et al., 1985).

Peptide-mediated neuroimmune interactions (for review see Carr and Serou, 1995; Roda et al., 1996; Plotnikoff et al., 1997) are a particularly complex topic: the mechanisms involved in the above-reported interactions are, therefore, far from being fully understand. Several factors contribute to this complexity: neuroimmune interactions are bidirectional, and plasma opioid peptides are synthesized not only by the adrenal medulla (Viveros et al., 1979) but also by immunocompetent cells (Buzzetti et al., 1989a) that release them in response to diverse stimuli (Heijnen et al., 1991). The complexity of neuroimmune interactions is enhanced by the existence of positive and negative effects (Roda et al., 1996). The existence of at least some of these effects can be rationalized under several hypotheses: according to Chukwuocha et al. (1994), the direction of the modulation depends on the degree of the immune response itself, so that met-enkephalin can suppress a strong response while increasing a weak response. Other effects can be attributed to the presence of different receptors: in immunocompetent cells, -endorphin binds both opioid and nonopioid receptors, inducing stimulatory effects (van den Bergh et al., 1993) and inhibitory effects (van den Bergh et al., 1993; Shaker et al., 1994), respectively. In addition, immune effects of the opioid peptides are both peripherally and centrally mediated. The existence of differences between in vivo and in vitro effects, such as those described by Bayer et al. (1992), has indeed been explained by the existence of both direct and indirect interactions: in the first case, the immunocompetent cell receptors are involved; in the second case, mechanisms such as the stimulation of the hypothalamo-pituitary axis (Buzzetti et al., 1989b), leading to the release of steroid hormones (Cupps and Fauci, 1982). Central neurons innervating immune-relevant tissues (Weber and Perth, 1989) can also be involved in these interactions.

Because modulation of blood immunocompetent cell activity is carried out by the opioid peptides present in this tissue, these peptides are exposed to hydrolysis by three different groups of enzymes: plasma soluble peptidases, peptidases present in the membrane of immunocompetent cells (Amoscato et al., 1989), and peptidases released by immunocompetent cells (Marini et al., 1992). The presence of these enzymes causes rapid hydrolysis of the opioid peptides, whose half-life in human plasma is approximately 5 min for enkephalins and 40 min for -endorphin.

The activity of plasma proteolytic enzymes is contrasted by inhibitors present in the tissue itself. These have mainly been studied regarding the enzymes that process plasma proteins, yet the low-molecular-weight inhibitors appear to contrast with the activity of the plasma enzymes active on opioid peptides (Roscetti et al., 1988), as well as the activity of the enzymes released by immunocompetent cells (Marini et al., 1992). The role of plasma proteolysis inhibitors in controlling peptide-mediated signaling has been studied at the whole-organism level: Chipkin and Coffin (1991) and Saksida et al. (1993) attributed the antinociceptive effect of proteolysis inhibitors to the decrease induced by these substances in the degradation of plasma opioid peptides. Katoh et al. (1992) found that nonhydrolyzable enkephalin analogs, or a combination of proteolysis inhibitors and opioid peptides, can modify the behavior of mice under stressful conditions. Moreover, the hydrolysis pattern of peripheral opioid peptides is altered by environmental conditions: Young et al. (1989) described increased degradation of -endorphin after chronic stress, whereas Schulteis and Martinez (1993) reported that external stimuli increase leu-enkephalin hydrolysis and modify the ratio between the hydrolysis by-products that are formed.

Taken together, the above data support the hypothesis that there exists a relationship between stressful events, the status of the immune system, and the degradation of plasma immunoactive peptides. In other words, it seems possible to hypothesize that the stress-induced differences in the status of the immune system may modify the enzyme degradation of neuropeptides. The data that follow attempt to verify this hypothesis, analyzing the possible relationship among stress, hydrolysis of plasma peptides, and the status of the immune system. The substrate used was leu-enkephalin, and the model was a homogeneous group of volunteers who had undergone a single traumatic event. The results obtained were compared with those obtained with age-matched control subjects. In the same subjects, several immune parameters were measured, and the possible relationship between modification of hydrolysis and modification of immune-related indexes was examined by analyzing the correlation between these two sets of data.

	Materials and Methods

Top Abstract Introduction Materials and methods Results Discussion References

Sample Population

The sample population that was studied consisted of 11 women (mean age, 80.0 years; S.D., 7.3 years) who had been hospitalized for proximal femur fracture. Before trauma, all patients were in apparently good health and did not present with any major syndrome. All were classified with regard to physical independence (Schoening and Iversen, 1968) and cerebral function (Folstein et al., 1975); in both cases, values were judged as normal for the age group. Analgesic treatment consisted of methadone (100 µg/kg) administered s.c. All patients underwent surgery under local anesthesia within 24 h of the accident. Postoperative treatment included thrombophylaxis with heparin (3000 IU/day), antibiotics (2 g of cefamandol twice; Eli Lilly, Indianapolis, IN), early physiotherapy, and enteral nutrition. Paracetamol (500 mg) or diclofenac (25 mg) was given as required. In the context of a parallel clinical study, five patients were treated with i.v. growth hormone (GH; 0.2 IU/kg/day; Pharmacia, Erlangen, Germany). Analyses performed to verify whether GH treatment affected substrate hydrolysis and the alteration of immune parameter did not indicate the existence of a relationship between these data: curves were identical within experimental error, and analyses of variance (ANOVAs) did not indicate a statistically significant relationship between these parameters (F = 1.293 to 0.011 and P = .271 to .917, respectively). Therefore, GH-treated and naïve subjects were treated as a homogeneous sample. The control population consisted of six women (mean age, 80.3 years; S.D., 13.4 years). Also, in this case, physical independence and cerebral function scores were normal for the age group. The administration of pharmacologically active substances, with the exception of cardioactive ones, was excluded for 5 days before the samples were taken.

Blood Collection

Blood referred to as t₀ was obtained immediately after hospitalization, simultaneous with analgesic treatment. Because hospitalization was limited to an average of 16 days, samples referred to as t₂₈ were obtained on an outpatient basis, 28 days after trauma. Once collected, blood was transferred to ice, and centrifuged (1000g for 10 min). Plasma was stored at 70°C until used.

Enzyme Assay

Enkephalin-degrading enzymes were quantified on the basis of the hydrolysis by-products formed. When necessary, samples were dialyzed at 4°C against 10 mM HEPES, 100 mM NaCl, 2.5 × 10⁵ M ZnCl₂, pH 7.2. Tritiated leu-enkephalin (5 µl corresponding to 3.5 × 10¹² moles) was added to 50 µl of each sample. Reactions were carried out at 37°C for 30 min and stopped with 5 µl of acetic acid (Marini et al., 1997). Intact leu-enkephalin and its labeled hydrolysis by-products were quantified by thin-layer chromatography as described below.

Chromatographic Procedures

Thin-Layer Chromatography. Thin-layer chromatography was performed on aluminum-backed Kieselgel 60 sheets (E. Merck, Darmstadt, Germany). The samples (4 µl), together with 2 × 10⁹ mol each of leu-enkephalin, Tyr, Tyr-Gly, Tyr-Gly-Gly, and Tyr-Gly-Gly-Phe used as internal standards, were applied to each spot. Chromatograms were developed at 50°C with 70:30:10:1.7 2-butyl alcohol/acetic acid/water/acetonitril and stained with 0.1% ninhydrin in acetone (Marini et al., 1997). Sheets were cut according to the standards, transferred to scintillation vials with the addition of 3 ml of 10% methanol in H₂O, and allowed to stand overnight at room temperature. After the addition of 3 ml of scintillation cocktail (Optiphase; Wallac Oy, Turku, Finland), they were counted for tritium. Samples were analyzed in triplicate, and the results were averaged.

Steric Exclusion Chromatography. Plasma enzymes were fractionated, as well as separated from the low-molecular-weight inhibitors, by a 5-µm 7.8 × 300-mm TSK G3000SW column (Toyo Soda Ltd., Tokyo, Japan) equilibrated in 10 mM HEPES, 100 mM NaCl, and 2.5 × 10⁵ M ZnCl₂ acetate, pH 7.2. Plasma (350 µl) was applied to the column, and eluted at 2.2 ml·cm²·min¹ with the same buffer. Absorbance at 280 nm was monitored, and 1-ml fractions were collected and assayed for enzyme activity as described above. In agreement with known data, the results obtained indicate the presence of all three classes of enkephalin-hydrolyzing enzymes: aminopeptidases, dipeptidylaminopeptidases, and dipeptidylcarboxypeptidases (e.g., Hambrook et al., 1976).

Immunophenotyping

Peripheral blood mononuclear cells were obtained by centrifuging whole venous blood drawn from the patients under investigation with use of a density gradient (Lymphoprep; Nyegard, Oslo, Norway). Peripheral blood mononuclear cells were resuspended at 10⁶ ·ml¹ in RPMI medium to which 2 × 10³ M glutamine, 100 mg·ml¹ kanamycin, and 10% human AB serum had been added. Cells were then cultured for 72 h in 96-flat-well plates (Costar, Cambridge, MA) in the presence of 1 µg/ml¹ purified anti-CD3 mitogenic monoclonal antibodies (CB3G; courtesy of Professor F. Malavasi, Turin, Italy). Tritiated thymidine (0.5 µCi/well) was then added to the cells, which were further incubated for 6 h, harvested, and counted for tritium to measure de novo DNA synthesis. The stimulation index was calculated as the ratio between the mean number of counts per minute measured in anti-CD3-stimulated cultures and that measured in unstimulated controls. Expression of leukocyte surface markers was evaluated with monoclonal antibodies on whole blood samples. Heparinized blood (50 µl) was incubated for 15 min at room temperature in the presence of optimal dilutions of fluorochome-labeled monoclonal antibodies (Becton-Dickinson, San Jose, CA) directed against CD14 (Leu-M3), CD45 (all isotypes), CD3, CD4, CD8, CD16, CD19, CD56, and HLA-DR (L243, specific for a monomorphic epitope). A red cell lysis solution was added to the blood, and incubation was prolonged for an additional 10 min. Cells were then washed twice and resuspended in paraformaldehyde 0.2% in 137 mM NaCl and 8 mM Na phosphate, pH 7.0 (phosphate-buffered saline). Specific fluorescence was measured with a fluorescent activated cell scan analysis (FACScan) fluorocytometer equipped with Lysis II software (Becton Dickinson). Granulocytes, monocytes, and lymphocytes were identified by their physical characteristics (FSC and SSC cytometric parameters), coupled with differential binding of CD14 and CD45 monoclonal antibodies (Spagnoli et al., 1995). This characterization allowed the analysis of individual "gated" leukocyte populations. Absolute counts of specific subsets were obtained by combining percent data measured by FACS analysis with the absolute leukocyte number obtained by automatic counting. Mean fluorescence intensity quantitatively estimates the binding of fluorochrome-labeled monoclonal antibodies to their molecular targets. This was evaluated on specific cell subsets from different samples after staining and fluorocytometer acquisition under identical conditions.

Data Analysis

Regression analyses were performed by a linear polynomial procedure of variable degree; related equation systems were solved using a modified Gram-Schmidt algorithm. Parametric statistics were calculated according to standard procedures. ANOVA was calculated by means of a commercial software package (Cohort Software, Minneapolis, MN), using one-way or two-way randomized complete blocks or completely randomized models. A homogeneity of variance test (Bartlett's test) was routinely performed before ANOVA.

Materials

Leu-enkephalin, Tyr-Gly, and Tyr-Gly-Gly were obtained from Bachem Feinchemikalien AG (Bubendorf, Switzerland); Tyr was obtained from Serva Feinbiochemica GmbH (Heidelberg, Germany). Tritiated leu-enkephalin was obtained by isotope exchange reaction followed by steric exclusion chromatography. All other materials were obtained through local suppliers and used without further purification.

	Results

Top Abstract Introduction Materials and methods Results Discussion References

Hydrolysis in Whole Plasma

Leu-Enkephalin Hydrolysis. The possible variations of leu-enkephalin hydrolysis associated with the sequence of events common to all patients (henceforth referred to as "trauma") were measured by incubating the tritiated pentapeptide in the presence of whole plasma obtained from normal control subjects and from traumatized patients; in the latter case, blood was drawn twice, immediately after hospitalization (t₀) and 28 days after the trauma (t₂₈). The relative amount of the intact peptide, and of its relevant hydrolysis by-products (Tyr, Tyr-Gly, and Tyr-Gly-Gly), was quantified as described in Enzyme Assay. The amount of substrate hydrolyzed was different under the three conditions tested (Fig. 1). Higher hydrolysis was measured in the control subjects: on average, 9.96 × 10¹³ mol·ml¹·min¹ (S.D. = 4.19 × 10¹⁴, CV = 4.2%), a value consistent with other data relative to this age group (Babst et al., 1998). The hydrolysis measured in plasma obtained from traumatized subjects was lower than that measured in the controls: 6.86 × 10¹³ mol·ml¹·min¹ (S.D. = 1.73 × 10¹³, CV = 25.2%) at t₀ and 8.05 × 10¹³ mol·ml¹·min¹ (S.D. = 1.73 × 10¹³, CV = 21.5%) at t₂₈. Therefore, the substrate hydrolyzed immediately after trauma (t₀) was 68.7% of that hydrolyzed in the control subjects, whereas that hydrolyzed after clinical recovery (t₂₈) was 80.6% of the substrate hydrolyzed in control plasma and 117.3% of that hydrolyzed at t₀ (Fig. 1, inset).

View larger version (37K): [in this window] [in a new window]   Fig. 1.   Leu-enkephalin hydrolysis and formation of its hydrolysis by-products in the presence of whole plasma obtained from control subjects and from traumatized individuals immediately after hospitalization (t0) and after 28 days (t28). Clear, light-gray, dark-gray, and black histogram bars indicate, respectively, leu-enkephalin degraded and Tyr, Tyr-Gly, and Tyr-Gly-Gly formed in the presence of control, t28, and t0 plasma; capped bars indicate standard error. Inset, the same data expressed as percent of the controls. Assay conditions as described in Materials and Methods.

Formation of Hydrolysis By-Products. The decrease in substrate concentration shown above is paralleled by the appearance of the three N-terminal hydrolysis by-products: Tyr, Tyr-Gly, and Tyr-Gly-Gly (Fig. 1). The formation of these peptides indicates the involvement of all three classes of enzymes known to hydrolyze leu-enkephalin: aminopeptidases, dipeptidylaminopeptidases, and dipeptidylcarboxypeptidases (Hambrook et al., 1976). The data shown in Fig. 1, inset, indicate that in t₀ plasma, the activity of aminopeptidases is 71% of the value measured in the controls, whereas in t₂₈ plasma, it is 86% of that value (121% of that measured in t₀ plasma). For dipeptidylaminopeptidases, these values are 73%, 79%, and 108%; for dipeptidylcarboxypeptidases, they are 40%, 60%, and 150%, respectively. Therefore, the above-reported decrease in substrate hydrolysis is supported by a decrease in the activity of all three classes of enkephalin-degrading enzymes, which, on a percentage basis, is particularly evident for dipeptidylcarboxypeptidases. As a result of these variations, the relative contribution to substrate hydrolysis of aminopeptidases, dipeptidylaminopeptidases, and dipeptidylcarboxypeptidases (65%, 25%, and 10% in the control subjects) is 67%, 27%, and 6% in t₀ plasma and 68%, 25%, and 7% in t₂₈ plasma. Therefore, although the trauma-induced reduction of dipeptidylcarboxypeptidase activity is the highest on a percentage basis, the low activity of these enzymes makes its impact on substrate hydrolysis moderate, and the variations in the relative activity of the three enzyme classes are likewise only moderate.

Hydrolysis in Fractionated Plasma

Leu-Enkephalin Hydrolysis. The modifications induced by trauma on the distribution of plasma enkephalin-degrading enzymes were analyzed in five donors, randomly chosen from the traumatized group, and in the whole control group. Plasma was fractionated as reported in Steric Exclusion Chromatography, and the relative activity of the three classes of enkephalin-degrading enzymes was analyzed as described in Enzyme Assay. Again, the data obtained indicate an increase in the amount of leu-enkephalin intact at reaction end (Fig. 2a) in traumatized compared with control individuals: in t₀ plasma, the average amount of substrate hydrolyzed by the fractionated enzymes is 64% of that measured in control plasma; in t₂₈ plasma, it is 84% of that measured in control plasma (131% of that measured in t₀ plasma). In t₀ plasma, the amount of substrate hydrolyzed decreases unevenly through the whole elution volume of the enzymatically active species: although the hydrolysis envelope is nearly symmetrical with respect to K' = 0.58 (approximately 75% of the reduction is measurable at K' lower, and only 25% at K' higher, than this value) (Fig. 2a, inset), causing the asymmetrical narrowing of the enkephalin envelope evident in the main panel of Fig. 2a. On the contrary, after recovery (i.e., in t₂₈ plasma), the amount of substrate hydrolyzed decreases at K' lower than 0.54, whereas it increases at K' higher than that value. Because the decrease and increase are nearly identical (Fig. 2a, inset), the activity of the enzymes eluted at lower K' (i.e., of higher apparent molecular weight) is lower, and recovers more slowly, than the activity of the enzymes eluted at lower apparent molecular weight.

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Hydrolysis of leu-enkephalin and formation of its hydrolysis by-products as a function of the elution volume of steric exclusion chromatography fractionated plasma. a, leu-enkephalin. b, Tyr. c, Tyr-Gly. d, Tyr-Gly-Gly. In all panels, gray lines represent controls, dotted black lines represent t28, solid black lines represent t0, and crosses indicate experimental points. Insets, difference between control and t28 plasma (thin lines) and between control and t0 plasma (thick lines). Chromatographic and assay conditions are as described in Materials and Methods.

Hydrolysis By-Products. In column-fractionated t₀ plasma, the activity of aminopeptidases (Fig. 2b), dipeptidylaminopeptidases (Fig. 2c), and dipeptidylcarboxypeptidases (Fig. 2d) is 64%, 62%, and 67% of the values measured in control plasma, respectively. In t₂₈ plasma, the activity of these enzyme classes is 84%, 93%, and 63% of the value measured in control plasma (131%, 150%, and 94% of that measured in t₀ plasma). These data indicate a quantitatively uniform reduction in the activity of the three enzyme classes immediately after trauma, followed by a less homogeneous recovery. Specifically, at t₂₈, the recovery of dipeptidylaminopeptidase activity appears to be nearly complete, whereas the activity of dipeptidylcarboxypeptidase is even lower than that measured at t₀.

Enzyme Distribution. The insets in Fig. 2, a and b, indicate that the reduction in the activity of aminopeptidases is almost inverse to the increase of intact leu-enkephalin. Specifically, the narrowing of the leu-enkephalin hydrolysis envelope observable in Fig. 2a is mainly determined by the absence in both t₀ and t₂₈ samples of the two shoulders of aminopeptidase activity evident in the controls at K' = 0.25 and K' = 0.75 (Fig. 2b); in addition, in both t₀ and t₂₈ plasma, the aminopeptidases eluted at lower K' undergo a reduction that is more noticeable than that of the aminopeptidases eluted at higher K'. Finally, the differences in the activity of aminopeptidases noticeable between t₀ and t₂₈ (Fig. 2b, inset) indicate a preferential recovery of the enzymes eluted at high K'.

Plasma Protein. Differences between control and traumatized plasma are also noticeable in terms of total protein (Fig. 3). These are to be found almost exclusively at low K' (less than approximately 0.47) because the elution volume of the substances centered at K' = 1.63 is not compatible with proteinaceous material. Measured as the integral of the absorbance at 280 nm across the whole elution volume of the column, protein is 71% of the control in t₀ plasma, whereas it is 86% of the control, or 122% of t₀, in t₂₈ plasma. These differences are more evident if the analysis is restricted to the material eluted between K' = 0.00 and K' = 0.47, that is, within the elution volume of the bulk of the enkephalin-degrading enzymes. Measured in this region, protein in t₀ plasma is 57% (S.D. = 46, CV = 114) of the control. In t₂₈ plasma, it is 86% (S.D. = 121, CV = 138) of the control, or 149% of the value measured at t₀. These data indicate a reduction in the protein (or, more precisely, of the absorbance at 280 nm) in the region between K' = 0.00 and K' = 0.47 in traumatized compared with control plasma, a reduction that recedes perceptibly with recovery. The statistical significance of the relationship between absorbance at K' between 0.00 and 0.47 measured at t₂₈ and that measured at t₀ was analyzed by ANOVA. The results obtained (F = 64.915, P = .0125, df = 5) indicate that the variations observed between the two sets of data are statistically significant. Finally, the hypothesis that the reduction in absorbance at low K' is related to the reduction in the activity of enkephalin-degrading enzymes was verified by calculating polynomial regressions of variable degree between these two sets of data. The results obtained (r = 0.828, 0.863, 0.907, and 0.999 for first to fourth degree) indicate a strong relationship between the two variables, together with deviations from linearity that may indicate the involvement of multiple factors.

Low-Molecular-Weight Inhibitors

Because steric exclusion chromatography fractionation separates the enzymes from the low-molecular-weight proteolysis inhibitors, a comparison of the data obtained in the presence of fractionated and whole plasma may allow the evaluation of the possible role of these substances. Specifically, in fractionated plasma (i.e., in the absence of inhibitors), the ratio between the substrate hydrolyzed by the three enzyme classes is slightly different from that seen in whole plasma (Table 1): under all three conditions tested, the activity of aminopeptidase is higher, that of dipeptidylaminopeptidase is lower, and that of dipeptidylcarboxypeptidase is higher in fractionated (Fig. 4) than in whole plasma (Fig. 1). These data can be interpreted as indicating inhibition of aminopeptidases (more evident in the controls) and of dipeptidylcarboxypeptidases (more evident at t₀), paired to null (or extremely low) inhibition of dipeptidylaminopeptidases. Because the differences measured are small, these data suggest a minor role of inhibitors in determining the trauma-associated reduction of hydrolysis. Consequently, this phenomenon appears to be principally caused by the per se reduction in enzyme activities described above.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Average absorbance of the steric exclusion chromatography separation of control (gray line), t28 (dotted line), and t0 (black line) human plasma. Chromatographic and assay conditions are as described in Materials and Methods.

View larger version (35K): [in this window] [in a new window]   Fig. 4.   Average hydrolysis of leu-enkephalin and formation of its hydrolysis by-products in the presence of enkephalin-active plasma enzymes fractionated by steric exclusion chromatography (Fig. 2). Groups are identified in the figure. Symbols and experimental conditions are as in Fig. 1.

Effect on Immunocompetent Cells

The possible effect of trauma on peripheral blood immunocompetent cells was measured as described in Immunophenotyping. Results obtained were analyzed by ANOVA, as described in Data Analysis. In the case of 6 of the 18 parameters measured, the probability of the null hypothesis was significantly low (P between 0.0003 and 0.033; Table 2). These were increase in absolute numbers of leukocytes, granulocytes, and monocytes; decrease in lymphocytes over total leukocytes; increase in lymphocytes B over total lymphocytes; and decrease in the HLA-DR surface expression in monocytes. The amount of NK lymphocytes over total lymphocytes was reduced by 23%, whereas the absolute number of activated T lymphocytes was reduced by 34%; in both cases, P was around .30. Finally, the variation of all other parameters measured was ±15% or less, associated with P  .44. 

The possible relationship among the six statistically significant immune parameters listed above and the reduction in hydrolysis was analyzed by calculating polynomial regressions of degree ranging from 1 to 4. Regression coefficients ranged from a minimum of 8.1 × 10⁴ to a maximum of 0.682, indicating the absence of any significant relationship between the indexes measured and the reduction of substrate hydrolysis. A similar analysis was performed on the above-listed immune parameters, taken in pairs. Except in the single case of leukocyte and granulocyte absolute counts (r = 0.826), calculated correlation coefficients ranged between 0.106 and 0.596. These values indicate an extremely limited correlation between each pair of indexes; thus, their quantitative variations can be taken to be practically independent of each other.

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

The data reported in Results indicate that the whole sequence of events indicated as trauma modifies the enzyme hydrolysis of plasma leu-enkephalin. Specifically, it reduces the amount of the peptide degraded and modifies the pattern of its hydrolysis by-products. The reduction in leu-enkephalin hydrolysis is mainly caused by a decrease in the activity of the enzymes involved, especially aminopeptidases, whereas the role of the low-molecular-weight inhibitors active on these enzymes is limited. The modified pattern of leu-enkephalin hydrolysis by-products can be related to the variations occurring between the rates of activity of the different enzymes capable of hydrolyzing the substrate. These variations, in turn, appear to be caused by trauma-associated shifts in the relative activity of the three classes of enkephalin-degrading enzymes, as well as by changes in the distribution of the active species within each class. Because of the technique used for enzyme separation (steric exclusion chromatography), this latter phenomenon is manifested as a reduction in the activity of the enzymes eluted at higher apparent molecular weight, which is more evident than that of the enzymes eluted at lower apparent molecular weight.

In addition to reducing the hydrolysis of leu-enkephalin and altering the ratio between its hydrolysis by-products, trauma modifies several immune parameters that can be related to stress. The low probability of the null hypothesis indicated by ANOVA of the trauma-associated alterations of substrate hydrolysis and of several of the immune indexes analyzed suggests that the variations induced in these two sets of data may be related to a common cause. Specifically, it is tempting to postulate that the stress associated with the sequence of events occurring before taking the blood samples may induce both sets of effects. This interpretation is, however, contrasted by the poor correlation found between decrease in hydrolysis and variations of immune parameters. The values of these correlation indexes indicate that the quantitative variation of one set of variables is not matched by a proportional variation of the other set. This datum cannot be assumed, however, as an indication of discordant variations of the two groups of variables; that is, the variations subsequent to trauma of both hydrolysis reduction and immune indexes are homogeneously positive or negative, and the probability that both sets are (independently) connected with the trauma is statistically significant. As mentioned above, the variations in the two sets of data may be connected by a direct causal relationship or may be linked indirectly. Finally, because the statistical probability indicates just a probability, the two sets of data may be totally unrelated. It seems, however, unreasonable to assume that modifications of metabolic parameters as complex as those examined here may depend on a single factor. In addition, the metabolic processes that relate the above-mentioned variations of enzyme activity to trauma are essentially unknown. Furthermore, in the case of the immune system, the existence of multiple regulatory factors is self-evident. Finally, the regression coefficients measured between the variations of the different immune parameters are just as low as those calculated between hydrolysis variation and immune indexes. At least to some degree, these results may be rationalized in terms of the nonhomogeneous status of the immune system in different donors, and the diphasic effect described for many neuropeptides in relationship to their role as immunomodulators, as detailed in the introduction.

In showing environment-induced modifications of peripheral opioid peptides, the data reported are in agreement with the few studies existing on this topic (Martinez and Weinberger, 1988; Young et al., 1989; Schulteis and Martinez, 1993). The differences between the data reported here and those referred to above, which were obtained with rat plasma, may be accounted for at least in part by theintraspecific differences in the degradation pattern of peripheral enkephalins (Bolacchi et al., 1996). In addition, differences in the stressful stimuli (e.g., chronic versus acute, single versus repeated) are known to influence some of their effects (e.g., Laudenslager et al., 1983; Sacerdote et al., 1994); therefore, the different experimental layouts may be responsible for the opposite direction of the effects reported. Another factor to be accounted for is the composition of the sample. The present study was conducted on a set of subjects who were particularly homogeneous regarding sex, age, and clinical history. This is both a drawback and an advantage: as reported just above, the effect of trauma varies in both intensity and direction with variations of the stressful stimuli. Therefore, homogeneity of both trauma and subsequent events should reduce stress-associated variability. On the minus side, the details of the specific mechanisms described, especially the reduced role of inhibitors, can be partially specific to the age (Babst et al., 1998), if not the sex (Marini et al., 1997), of the subjects under study.

In conclusion, it seems unlikely that communication between two major systems such as the nervous and immune systems would be anything but complex. Therefore, the lack of a clear relationship between different, although possibly related, effects such as those demonstrated above is to be expected. However, this implies that determination of the mutual dependence of these factors may require the accumulation of an impractically large body of data, at least until the mechanisms linking stress and the immune system and linking stress and enzyme activity are known in better detail. Finally, it seems worth noting that the lack of a definite tridimensional structure of relatively small peptides in polar solvent makes their hydrolysis relatively unspecific and that the enzymes that degrade enkephalins are also active on other low-molecular-weight peptides (Roscetti et al., 1988). Therefore, at least in principle, the data reported here, which were obtained using leu-enkephalin, may be extended to other blood peptides of suitable molecular mass. Likewise, it seems reasonable to hypothesize that the changes in the enzyme population induced by trauma, which are shown here as shifts in the hydrolysis pattern of leu-enkephalin, may also modify the relative hydrolysis of other peptides present in blood and involved in the transmission of information.