Possible in vivo tolerance of human polymorphonuclear neutrophil to low-grade exercise-induced endotoxaemia.

To address the question of whether translocation of bacterial lipopolysaccharide (LPS) into the blood could be involved in the process of exercise-induced polymorphonuclear neutrophil (PMN) activation, 12 healthy male subjects who took part in a sprint triathlon (1.5 km river swim, 40 km bicycle race, 10 km road race) were studied. While there was no detectable amount of endotoxin in the blood samples drawn at rest, exercise was followed by the appearance of circulating endotoxin molecules at the end of competition in four subjects, and after one and 24 h recovery in three and seven athletes, respectively. The concentrations of plasma granulocyte myeloperoxidase ([MPO]), were significantly higher immediately after exercise and one hour later-than baseline values (P<0.001). This variable returned to pre-race levels the day after exercise, despite the presence of detectable amounts of LPS, at that time, in seven athletes. The absence of significant correlation (r=0.26; P=0.383) and temporal association between [MPO] and plasma endotoxin levels led us to conclude that endotoxaemia was not involved in the process of exercise-induced PMN degranulation observed in our subjects.


Introduction
It is pre se ntly w e ll admitted that long-te rm stre nuous ex e rc ise is accompanied by transie nt e ndotox aemia 1-3 and activation of polymorphonuclear ne utrophils (PMN). 3,4 Bec ause endotox ins [lipopolysaccharide (LPS) molec ule s] are pote nt ac tivator s of PMN, 5 these findings raise the possibility that ex ercis e-induced e ndotox ae mia could be involved in the proce ss of PMN activation in ex e rcising subje cts . 6 To te st this hyp othesis, w e analyze d the changes in the plasma le ve ls of LPS ([LPS]) and granulocyte mye loperox idase (MPO) -taken as an in v ivo marke r of PMN activation -in 12 healthy male subje cts w ho took part in a sprint triathlon.

Subjects
Tw elve male re c re ational triathle te s aged 29± 1 (SEM) ye ars [mean body mass: 69± 2 (SEM) kg] w e re studie d. Ethical permission for the study w as obtaine d from the Committe e for Me dic al Ethic (Faculté de Médec ine, Univers ité de Liège). Written informe d consent w as obtain ed from all volunte ers . The y did not take any form of me dic ation during the month pre ce ding the study. The event include d a 1.5 km rive r sw im, a 40 km bike ride , and a 10 km road rac e.

Blood sampling and biochemical analyses
Tw o venous blood samples (10 ml) w ere c ollec te d in Vac utainers at the follow ing time points: at re st, 3-5 days be fore the event (base line); w ithin 5-15 min after the end of the rac e; after 1 and 24 h re cove ry. Blood for the me as ure ment of endotox in w as c ollec te d in p yroge n-fre e tube s (Endo tube s ET, Chromogenix AB, Mölndal, Sw e de n) c ontainin g sodium heparin (120 IU). Plate let-rich plasma w as pre pared by ce ntrifugation at 2200 g for 15 min at 4°C, and samples w ere store d at -20°C.
Blood use d for the measureme nt of MPO w as draw n on EDTA as antic oagulant. Plasma w as re move d afte r ce ntrifugation for 10 min at 2500 g and store d immediate ly at -70°C until analysis. Plasma MPO and endotox in conce ntration s w ere asse sse d ac cording to the radioimmunological method de sc ribe d by Pinc e mail e t a l. 7 and a chromogenic Lim u lu s amebocyte lysate assay (Coate st plasma-endotox in, Chromogenix AB; de te c tion limit of 5 pg/ml), re spe ctively. Plasma conce ntration s of MPO ([MPO]) and endotox in ([LPS]) w ere adjuste d for change s in p lasma volume during and after ex ercis e acc ording to the method desc ribed by Dill and Costill. 8

Statistical analysis
Changes in MPO values as a func tion of time w ere asse sse d by Friedman's te st. Assoc iations be tw e en endotox in le vels ranging from 5 to 23.3 pg/ml and corresp onding MPO values w ere investigate d using the Sp earman rank correlation coefficie nt. The leve l of statistic al signific ance w as set at P< 0.05.

Results
Time to c omple te the rac e averaged 150± 5.4 (SEM-) min. No trac e of circulating endotox in w as found at re st. This substance w as dete cte d in eight subjec ts, most frequently the day afte r the race (7 case s out of 12) ( Table 1). Imme diate ly after the race and one hour later, dete c table amounts of e ndotox in w ere pre se nt in four and three samples, re spe ctively. The plasma conc entrations of this c omp ound ranged be tw e en 5 and 23.3 pg/ml. The highest endotox in value w as dete cte d in one subjec t afte r 1 h re cove ry (Table 1).
Mean plasma MPO conce ntrations measured at the end of compe tition and afte r one hour re cove ry w e re significantly higher than baseline (P<0.001). Tw e ntyfour hours later plasma MPO le ve l had de cre ased to a level that did not differ signific antly from the preex e rc ise le ve l ( Table 1).
The re lationship be tw e en [MPO] and [LPS] w as not statistically signific ant (r= -0.261; P= 0.383), and the patte rns of change of the se variables as a func tion of time w ere fairly differe nt (Table 1). For ex amp le , in subject 2, [MPO] rose from a re sting value of 150 ng/ ml to 1200 ng/ml at the e nd of ex erc ise and one hour later. By the time the se MPO conce ntration s w ere re ached, circulating LPS le ve ls w e re equal to 14.5 and 5 pg/ml, re spe ctive ly. The day afte r the race , [LPS] re ached its highest value of 17.5 pg/ml w hile [MPO] had decre as ed to 380 ng/ml. A similar conclusion can be draw n by analyzing the data of subje ct 9. In this case , a marke d inc re ase of [MPO] from 140 to 1190 ng/ml w as obse rved despite the absenc e of any dete ctable amount of c irculating LPS.
The se findings led us to c onclude that endotoxae mia, at least w ithin the range of LPS c onc entration me asured in the pre se nt study, w as not involved in the proc ess of ex e rc ise -induc ed PMN activation observe d in our subjec ts.

Discussion
To our know ledge, the pre se nt study demonstrate d for the first time that strenuous ex ercis e lasting, on ave rage , less than three hours can initiat e the appearanc e of de te c table amounts of endotox in in the blood. Among the me chanisms thought to be involved in ex ercis e-induced e ndotox ae mia are; (1) re duced sp lanchnic blood flow, and (2) hyperthermia, both leading to damage to the inte stinal w all and subse que nt re lease of e ndotox in into the circulatory syste m. 1 ,2 Taken ove rall, the pre sent data demonstrate d that the magnitude of endotox ae mia w as gre ater in triathle te s than in subjects w ho took part in a Mediators of Inflammation · Vol 7 · 1998 marathon competition. 3 Furthermore , trace s of e ndotox in 5 pg/ml w e re dete cte d more fre quently in triathle te s (in 29 ve rs u s 12% of the blood samples draw n ac cording to the same protocol in triathlete s and marathon runners, re spec tive ly).
The absenc e of te mporal association and significant correlation be tw e en [MPO] and [LPS] strongly suggeste d that e ndotox ae mia w as not the underlying me chanism of ex ercise -induc ed PMN ac tivation. Furthermore , the re turn of MPO conc entration to baseline, show n by the time dete c table amounts of LPS molecules w ere found in se ven subje cts after 24 h re c overy, indicated that the pre se nce of this compound in the blood had no e ffec t, pe r s e , on the proce ss of PMN de granulation, at le as t in our subjec ts, and w ithin the range of LPS conc entration me as ure d. This appare nt unresponsive ne ss of PMN to LPS c ould be c onside re d as a further manifestation of the endotox in tolerance phe nomenon associate d w ith re pe ated e ndotox in challenge , 9 but the ex te nt to w hich its unde rlying mechanisms are similar to those re sponsible for the suppre ssion of the plasma tumor ne crosis factor re spons e to LPS by prior ex erc is e 1 0 (there by re flec ting the be ne fic ial e ffec t of trainin g on the host's re sistance to the detrimental e ffec ts of endotox ae mia) is an attrac tive hypothesis that should be ex pe rimentally verifie d.