http://pen.sagepub.com/Nutrition

Journal of Parenteral and Enteral

http://pen.sagepub.com/content/19/1/69The online version of this article can be found at:

 DOI: 10.1177/014860719501900169

1995 19: 69JPEN J Parenter Enteral NutrLuca Gianotti, J. Wesley Alexander, Roberto Gennari, Tonyia Pyles and George F. Babcock

SepsisOral Glutamine Decreases Bacterial Translocation and Improves Survival in Experimental Gut-Origin

  

Published by:

http://www.sagepublications.com

On behalf of: 

  The American Society for Parenteral & Enteral Nutrition

can be found at:Journal of Parenteral and Enteral NutritionAdditional services and information for    

  http://pen.sagepub.com/cgi/alertsEmail Alerts:

 

http://pen.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

http://pen.sagepub.com/content/19/1/69.refs.htmlCitations:  

What is This? 

- Jan 1, 1995Version of Record >>

at NORTHWESTERN UNIV LIBRARY on September 3, 2014pen.sagepub.comDownloaded from at NORTHWESTERN UNIV LIBRARY on September 3, 2014pen.sagepub.comDownloaded from

69

Oral Glutamine Decreases Bacterial Translocation and ImprovesSurvival in Experimental Gut-Origin Sepsis

LUCA GIANOTTI, MD, ScD*; J. WESLEY ALEXANDER, MD, ScD; ROBERTO GENNARI, MD; TONYIA PYLES, BA;AND GEORGE F. BABCOCK, PHD

From the Department of Surgery, Transplantation Division, University of Cincinnati College of Medicine, Ohio

Correspondence and reprint requests to: J. Wesley Alexander, MD.ScD.University of Cincinnati College of Medicine, 231 Bethesda Avenue. Cincin-nati. OH 45267-0558.*Current address: University of Milano. IRCCS S. Raffaele. Chirugia 2. ViaOlgettina 60, 20132, Milano, Italy.

ABSTRACT. Background: Glutamine has been shown to be animportant dietary component for the maintenance of gutmetabolism. The purpose of this study was to assess the potentialbenefit of glutamine-enriched diets on experimental gut-derivedsepsis. Methods: BALB/c mice were fed either 2% glutamine-sup-plemented or 1% glycine-supplemented (near-isonitrogenouscontrol) AIN-76A diets. Control mice received either nonsupple-mented AIN-76A or regular Purina Rodent Laboratory MouseChow 5001 diets. After 10 days of feeding, the mice weretransfused with allogeneic blood (from C3H/HeJ mice), and thefeeding protocols were continued for an additional 5 days. Themice then underwent gavage with 1010 Escherichia coli labeledwith either indium-111 oxine or [14C]glucose followed immediatelyby a 20% burn injury. Some mice were observed 10 days postburnfor survival rates. Others were killed 4 hours after burn, and

the mesenteric lymph nodes, liver, and spleen were harvestedto determine radionuclide and bacterial colony counts. The

percentages of viable translocated E coli were also calculated.Results: Mice fed glutamine-enriched diets had a lower degreeof translocation (as measured by both radionuclide and bacterialcounts) to the tissues than did the other groups and had animprovement in the ability to kill translocated E coli (as measuredby the percentage of viable bacteria). Survival was significantlyhigher in the group fed 2% glutamine (81%) compared with thegroups fed 1% glycine (36%), AIN-76A (35%), and Purina RodentLaboratory Mouse Chow 5001 (36%) diets (p < .004). Conclusions:Glutamine-supplemented enteral diets may exert importantbenefits in preventing gut-origin sepsis after trauma (Journal ofParenteral and Enteral Nutrition 19:69-74, 1995)

Microbial translocation, defmed as the passage of bothviable and nonviable microbes and microbial productsfrom the gut to systemic organs,’ has been shown tobe a key factor in causing septic morbidity and mortalityin several different animal models of trauma.2.3 Althoughto date there is no reliable method to detect and

quantitate the degree of microbial translocation in

critically ill patients, several types of indirect evidencestrongly suggest that the intestine may serve as an

important reservoir for infections and sepsis observedafter severe trauma. 4,1 It is important to study potentialtherapeutic modalities that minimize this process becausethey may be relevant for the management of patientsin whom a failure of the gut-barrier function is suspectedor in whom it occurs. We have previously shown thatit is possible to achieve a significant reduction in

mortality secondary to gut-derived sepsis by decreasingthe magnitude of translocation,3,6 enhancing the immuneclearance of translocated bacteria, 7,1 or both.The following studies were designed to address the

hypothesis that prior oral intake of glutamine-enricheddiets could affect the gut barrier and improve survival

in animals undergoing blood transfusion, thermal injury,and oral bacterial challenge. Glutamine was testedbecause it is a preferred fuel for enterocytes, particularlyafter trauma, which may be an important predisposingfactor for translocation.9,10The degree of translocation was evaluated by using

Escherichia coli labeled with p&dquo;C]glucose, as previouslydescribed,’.&dquo; and by a different technique employing Ecoli labeled with indium-111 oxine. Thus it was also

possible to compare the reliability and suitability of thetwo labeling procedures.

MATERIALS AND METHODS

Animals and Animal Care

Adult female BALB/c mice (H-2~) (Charles River,Wilmington, MA) and female C3H/HeJ mice (H-2k)(Jackson Laboratory, Bar I-larbor, ME), weighing between20 and 22 g, were used for these experiments.

All mice were quarantined for 1 week to allow- for

adaptation to the new PI1Bironmental conditions and toexclude mice with i~rw~~~i~tiy disease. During this period,the mice were provided ad libitum access to food (PurinaRod(,tit laboratory Mouse Chow 5001. Purina Mills Inc,St Louis), l~i(~) and water until the onset of the

experiments. The experimental protocols were approvedby thf I’mvcr~try of (Ïlwinnati Animal Care and I’soCommune, and the %% ei-(- maintained in an AmericanAssociation for Accreditation of Laboratory Animal

at NORTHWESTERN UNIV LIBRARY on September 3, 2014pen.sagepub.comDownloaded from

70

Care-approved facility. In conducting the describedresearch, the investigators adhered to the Guide for theCare and Use of Laboratory Animals as set forth bythe Committee on the Care and Use of LaboratoryAnimals, National Research Council, and United StatesDepartment of Health and Human Services, NationalInstitutes of Health. The mice were killed by using anoverdose of methoxyflurane anesthesia and cervicaldislocation or by carbon dioxide inhalation accordingto the Guide for the Care and Use of Laboratory Animals.

Diet Preparation and Feeding Protocols

In these experiments, four different pelleted dietswere used. Two of these diets were purchasedcommercially: AIN-76A semipurified diet (ICN Biomedi-cals Inc, Costa Mesa, CA) and Purina Rodent LaboratoryMouse Chow 5001. The other two diets were preparedin our laboratory by modifying the AIN-76A semipurifieddiet. The first experimental diet had 2% glutamine added(nitrogen content, 9.52%); the second experimental diethad 1% glycine added (nitrogen content, 18.66%). The1% glycine was used to balance the nitrogen contentof the 2% glutamine so that the two diets were nearlyisonitrogenous. The AIN-76A diet contains 0.2% glu-tamine and 0.35% glycine, whereas the Purina Chowcontains 0.1% glutamine and 0.3% glycine. All percent-ages of nutritional components are expressed as

weight/weight. The detailed composition of the glu-tamine-enriched and glycine-enriched diets is shown inTable I. Pair feeding was not carried out because

preliminary experiments showed that during the 15 daysof feeding the amount of diet intake was comparableamong individual mice belonging to the different groups(range 2.8 to 3.2 g/d; mean, 3.07 ::!: 0.3 SEM). Fifteendays of feeding was chosen to allow sufficient time forincorporation of the nutrients into tissue compartments.BALB/c mice were assigned randomly to homogeneous

groups to be fed one of the four diets. All groups werefed the diet ad libitum for 10 days before bloodtransfusion and for an additional 5 days before

undergoing gavage with E coli and a full-thickness-flamebum (see following section). The same diets werecontinued postbum when appropriate.

TABLE IComposition of the pellet diets (3.86 kcal/g)

The AIN-76A diet contained all components except the added glutan-tineor glycine. All components were obtained from ICN Biomedicals Inc,Costa Mesa, CA

Blood Harvesting and Transfusion Procedures

Blood was obtained from C3H/HeJ mice (undermetho_xyflurane inhalation anesthesia) by cardiac punc-ture using aseptic technique. C3H/HeJ mice were usedas donors because preliminary experiments showed thattransfusion from this strain caused a significant increasein mortality in BALB/c mice subjected to burn injuryand gavage with bacteria. The blood was mixed with

citrate-phosphate-de~rose-adenine anticoagulant (Fen-wal Laboratories, Deerfield, IL) at a 7:1 ratio and storedat 4°C overnight. After 10 days of feeding, all groupsof BALB/c mice were transfused via a tail vein with0.2 mL of C3HlHeJ blood.

Preparation of 14 C-Radiolabeled E coli

A glucose-deficient minimal-nutrient broth was inocu-lated with E coli (Stock 53104; University of Minnesota,Minneapolis), and 500 VLCI of [1~C]glucose (Du Pont-NewEngland Nuclear, Boston) was added simultaneously.The culture was incubated for 18 hours at 37°C in a

shaking incubator, pelleted by centrifugation, and washedtwice in sterile saline to remove the unincorporatedisotope. The final pellet was resuspended in sterile saline,and the desired concentration of E coli of 1 X 1010

organisms per 0.2 mL was obtained using a standard Klettdensitometer (Klett Mfg Co, New York). The concentrationof viable E coli was confirmed by plating 10-fold dilutionsof the bacterial suspension on brain-heart-infusion agarplates. The radionuclide counts of the gavage suspensionaveraged 3.65 X 101/mL. This corresponded to 137 viableE coli per disintegrations per minute. The supernatantafter two washes had 3.89 x 104 dpm/mL.The same stock of E coli was also grown in brain-heart

infusion broth for 18 hours at 37°C without the additionof isotope. The culture was processed as above to obtainan identical fmal concentration of 1 X 1010 organismsper 0.2 mL.

Preparation of Indium-111 Oxine-Labeled E coli

Twenty-five milliliters of brain-heart infusion broth

(Becton Dickinson, Cockeysville, MD) was inoculatedwith E coli (Stock 53104) and incubated for 18 hoursat 37°C with shaking. The E coli were collected bycentrifugation, washed twice in sterile saline, and broughtto a desired fmal concentration of 1 X 101° organismsper 0.2 mL. Then 1 mCi of indium-111 oxine (Syncor,Cincinnati) was added to the bacterial suspension andincubated for 40 minutes at 37°C without shaking. Theculture was washed twice in sterile saline to removethe unincorporated isotope. The radionuclide counts ofthe gavage suspension were 4.09 x 108/mL. This

corresponded to 16 viable E coli per dpm. The

supernatant after two washes had 2.13 X 1<r dpm/mL.

Burn Model and Gavage Procedure

Four days after transfusion, all mice had the hairremoved entirely from their backs and flanks by clipping.The following morning, all mice underwent gavage with0.2 mL containing 1 x 101° of either 1~C or indium-111 loxine-E coli or with nonlabeled E coli, according to the

at NORTHWESTERN UNIV LIBRARY on September 3, 2014pen.sagepub.comDownloaded from

71

experimental design. The mice were then anesthetizedby methoxyflurane inhalation, immediately subjected toa 20% total body-surface area, full-thickness-flan1e burn.and resuscitated by intraperitoneal injection of 20 mLof saline per kilogram. The burn wound was left exposedthroughout the duration of the experiments.

Experimental DesignExperiment 1. Five days posttransfusion (15 days of

feeding on the selected diet), 40 BALB/c mice (n = 10,dietary group) underwent gavage with 1 x lOlo mC-labeledE coli and immediately thereafter underwent a 20% burninjury. Four hours later, the mice were killed, and theabdomen was prepped with 70% isopropyl alcohol. Theabdominal cavity was exposed with a midline laparotomy,and the mesenteric lymph nodes, liver, and spleen wereexcised using aseptic techniques. The organs were placedin sterile Petri dishes, weighed, and homogenized in 1mL of saline. One hundred microliters of each homo-

genate was plated on eosin-methylene blue plates (BectonDickinson), a selective medium for aerobic Gram-nega-tive bacteria. The plates were incubated in aerobicconditions at 37°C for 24 hours, and the number ofcolony-forming units (CFU), consistent with the mor-

phology of E coli, were counted. The remainder of thetissue homogenates were lyophilized overnight and

decolorized, as previously described. The radionuclidecounts, as measured by dpm, were determined using aliquid scintillation counter (Beckman Model LS 6000 TA;Beckman Instruments, Inc, Fullerton, CA). The CFU anddpm counts were adjusted to be expressed per gramof tissue, and the estimated percentage of viabletranslocated E coli was calculated using a formuladescribed previously’:

where K = the number of viable 1‘’C-labeled or indium-111 Ioxine-labeled E. coli in the gavage suspension corre-sponding to 1 dpm.

Experiment 2. Five days posttransfusion (15 days offeeding on the selected diet), four additional groups ofBALB/c mice (n = 10, dietary group) underwent gavagewith 1 x 10&dquo;’ indium-Ill oxine-E coli and a 20% buminjury, as described in experiment 1. Four hours

postbum, the mice were killed, and the mesenteric

lymph nodes, liver, and spleen were harvested andprocessed for bacterial culture, as in experiment 1. The

homogenates were placed directly into plastic tubes,and the residual radioactivity was measured using asolid crystal scintillation counter (5500 Counting System;Beckman Instruments, Inc).

Experiment 3. Eighty-five BALB/c mice were random-ized to be fed one of the four experimental diets. After10 days of feeding, the mice were transfused andcontinued on the feeding protocols for an additional 5days. The mice then underwent gavage with 1 x 1011E coli and received a 20’&dquo;c burn. as described above.Survival rates were recorded for 10 days postbum. Afterthermal injury, all groups w ere allowed ad libitum accessto their assigned diets and water.

Statistical Analysis

The differences among the means of the continuousvariables (CFL-. ~lpm. and percentage alive) of the fourgroups were analyzed using a one-way analysis ofvariance followed by the Duncan’s test. p B-allies < .05were considered significant. The sun ivaI rates of thefour groups were compared using the x’ test of

independence.RESULTS

The magnitude of translocation (as measured by dpmof tissue per gram) in mice that underwent gavage with14C-E coli is sown in Table II. Mice fed the

glutamine-enriched diet had significantly less transloca-tion to the mesenteric lymph nodes than did mice fedthe AIN-76A or glycine diets. The magnitude oftranslocation to the liver and spleen in the glutaminegroup was significantly lower than in all other groups.Consistent results were obtained when translocation was

quantitated using indium-111 oxine-E coli, but in this

study, the group receiving glutamine had less translo-cation to all tested organs than did the other groups(Table III).The number of viable bacteria recovered from the

tissues showed that the mice challenged with &dquo;C-E coliand fed the glutamine-enriched diet had fewer viablebacteria in the mesenteric lymph nodes than did theother groups and fewer viable bacteria in the liver andspleen than did the AIN-76A and glycine groups (TableII). Similar data were obtained using indium-111oxine-labeled E coli. The number of viable E coli weresignificantly lower in the glutamine group than in theother groups in all tissues examined (Table III).No difference in the estimated percentage of viable

translocated bacteria was observed among the four

groups of mice that underwent gavage with 14C-E coli,mth the exception ot the mice ted the A1N-’ltiA diet.This group had a higher percentage of viable organismsin the spleen than did the mice receiving the glutamine-enriched diet or the Purina Chow (Table II). When thepercentage of viable bacteria was estimated usingindium-Ill oxine-labeled E coli, the glutamine groupshowed a consistently greater ability to kill translocatedE coli in the mesenteric lymph nodes and in the spleenthan did the other groups (Table III).

Figure 1 depicts the survival rates at day 10 postburnof mice fed the four different diets. Survival was 35%

(7/20) in the AIN-76A group, 37% (8/22) in the glycinegroup, 40% (8/20) in the Purina Chow group, and 81%(17/21) in t he glutamine group ( p < .00~; X2 = 13.~3).

DISCUSSION

Bacterial translocation can be evaluated and quanti-tated hB ~eBeral methods. The 1111 )&dquo;’t frequently usedmethod has been the microbiW culturE&dquo; technique bywhich the number of viable enteric bacteria translocatedto extraintestinal organs is measured directly&dquo; However,the approach camot separate the role of gut-barrierfailure from the effect of host defense mechanisms. A

separate evaluation of these two functions allows abetter understanding of the patlmtfh~wiolugEv of translo-

at NORTHWESTERN UNIV LIBRARY on September 3, 2014pen.sagepub.comDownloaded from

72

TABIE n

Degree of translocation of HC-Escherichia coli (dpm/g), number of viable E coli (colony-forming units [CFUJ per gram), and calculated percentage oftranslocated E coli that survived in the fourgroups fed different diets

Values are shown as mean ± SEM. Numbers in parentheses denote number of animals that had positive cultures.* p < .05 vs glycine and AIN-76A.t p < .05 vs all.

,+. p < .05 vs glutamine and Purina Chow (analysis of variance and Duncan’s test).

TABLE III

Magnitude of translocation of indium-111 Escherichia coli (dpm/g), number of viable E coli (colony-forming units (CFZIJ per gram), and calculatedpercentage of translocated E coli that survived in the groups fed with four different diets

Values are shown as mean ± SEM. Number in parentheses denotes number of animals that had positive cultures.* p < .05 vs all (analysis of variance and Duncan’s test).

cation and the mechanisms of action of potentialtherapies capable of reducing the risk of gut-derivedsepsis. We have previously reported that the use of Ecoli labeled with [&dquo;Clglucose to quantitate translocationwas a suitable approach to estimate separately thefraction of translocated organisms that resist killing bythe host and the total load of viable and nonviablemicrobial products that cross the intestinal barrier

Therefore, the gut permeability to enteric bacteria andthe bactericidal mechanisms can be assessed inde-

pendently. By employing this method, it was possibleto evaluate the mechanism of action and the effect ofseveral treatments and insults on mucosal barrier andimmune clearance.3.6-8.12.13 Moreover, it was possible todemonstrate a significant correlation between the

magnitude of translocation shortly after burn injury (asmeasured by radionuclide counts in the blood) and

subsequent septic mortality3 as well as the degree ofhypermetabolic and catabolic responses (unpublisheddata). Despite these advantages, the labeling of bacteriawith [l4C]glucose has some inconveniences: (1) thebacterial suspension delivered by gavage intragastricallymay contain a fraction of free isotope because the

degree of incorporation of the isotope into the

microorganisms is not complete; (2) during the processof bacterial labeling, [1~C]glucose is used by all of thebiochemical pathways that use glucose, including thesynthesis of endotoxin of the cell wall. This means thata substantial portion of the radioactivity in the tissuescould be attributed to translocation of free endotoxinor other microbial products that incorporate glucosecarbon; (3) once bacteria are killed by the inununesystem, [14C]glucose may be liberated and partially reusedin the metabolism of the animal or eliminated as CO,;

at NORTHWESTERN UNIV LIBRARY on September 3, 2014pen.sagepub.comDownloaded from

73

and (4) to deternline the residual radioactivity.. the

organs and fluids of the animal need to be processedthrough lyophilization and decolorization procedures,which include several manipulations and dilutions ofthe samples that lead to a partial loss of precision ofmeasurement. Conversely, the use of indium-111 oxineto label E coli seems to have several advantages: (1)the degree of incorporation is consistently better thanthat of P4C]glucose inasmuch as both the quantity ofradioactivity in the supernatant and the number of viableE coli per dpm were almost 10-fold lower; (2) indium-111 loxine diffuses through the cell wall quickly and

efficiently&dquo; and is incorporated into cytoplasmic proteins.This assures measurement of translocation of bacteria

separately from endotoxin; (3) even when liberated,indium-111 oxine cannot be reused in any biochemical

pathway; and (4) the radioactivity can be detected

directly on the whole tissue sample without any furthermanipulation or dilution procedure, thus enhancing theaccuracy of measurement.The results on the magnitude of translocation sug-

gested that the two techniques of bacteria labeling werequalitatively equivalent in the evaluation of the intesti-nal-barrier function. In both experiments, the use of

glutamine-enriched diets significantly reduced the degreeof microbial translocation to the tissues being studied,compared with the other diets. Interestingly, theradionuclide counts were lower in the mice thatunderwent gavage with indium-111 oxine-E coli than inthose given 14 C-E cali, despite the fact that the dpm inthe two bacterial suspensions were comparable. Thisstrongly suggests that a fraction of the radioactivityfound in the tissues of the mice challenged with 14C-Ecali was attributable to translocation of free endotoxin,unincorporated isotope, or bacterial products that

incorporate glucose carbon. This hypothesis was sup-ported further when the intensity of translocation wasevaluated by the number of viable E coli recoveredfrom the organs. CFU counts in the tissues of mice

FiG 1. Percentage of survival at day 10 postburn of mice fed one of fourdifferent diets. ~p < .OM (x = 13 ~3) bs all.

that underwent gavage ~~s-itll =jC-E coli or indium-111oxine-E coli were similar. Thus the use of ’3C-E coliseems to reflect translocation of both bacteria andbacterial products, whereas the use of indium-Ill oxine-Ecoli seems to reflect translocation of intact baciena.The mice fed glutamine-enriched diets had lower

translocation rates than did the other groups, regardlessof the type of isotope used to label E coli. Theseobservations are consistent with the protective effectof glutamine on the intestinal-barrier function and themorphologic structure described by several otherauthors. Windmueller and Spaeth 15 showed that glutaminederived from the blood is a key respiratory substratefor the mucosa of the small intestine, which representsthe major site for the metabolism of this anino acid.Later, Fox et all’ reported that glutamine-enrichedelemental diets resulted in a significant improvement inanimal body weight, nitrogen retention, protein and DNAcontent, and mucosal weight of the jejunum and colonof rats treated with methotrexate. These findingscorrelated well with decreased incidence of translocation,bacteremia, and mortality. These data were confirmedby Klimberg et all ‘‘1g and Souba et al19 who describedthe beneficial effects of oral glutamine supplementationon intestinal morphometry and septic morbidity andmortality secondary to abdominal radiation-inducedtranslocation. In a different experimental model, Burkeand colleagues21 showed that, in noninjured mice,translocation induced by total parenteral nutrition couldbe reversed by intravenous administration of glutamine.The protective mechanism of action of glutamine wasattributed to the remarkable increase in secretoryimmunoglobulin-A production, which reduced the adher-ence of enteric bacteria to the gut epithelium. Consistentresults were reported by Alverdy et al2l using commer-cially available enteral diets enriched with glutamine.The lack of a protective effect of glutamine supplemen-tation on translocation and survival has been describedin mice challenged with systemic administration ofendotoxin22.23 or metronidazole. 2:3Our results are consistent with the hypothesis that

enteral formulations enriched with glutamine may exertan important protective role on the gut barrier afterthermal trauma. The models of trauma used to evaluatethe potential beneficial effects of glutamine may becrucial because glutamine concentration in the tissuesand blood declines markedly in eritiea! illness. whereasthe requirement for glutamine by the intestinal mucosais increased during catabolic states and injury. It IThe present data suggest that the protective rote of

glutamine in gut-derived sepsis is primarily a result ofthe preservation of intestinal-barrier function, whichallows fewer ~~iable bacteria and bacterial products tospread to systeiiiic organs. Furthermore, ill thi~- model.reduction of translocation showed a close relationshipwith improved survival. The results of the present studyalso demonstrated that glutamine improved the host

ability to kill translocated bacteria it,, lio%%-ii by the

percentage of organisms that remained :di;e. Inoue et

al:24 reported a significantly improved tolerance to Ecoli-induced peritonitis in rats fed glutamine. suggesting

at NORTHWESTERN UNIV LIBRARY on September 3, 2014pen.sagepub.comDownloaded from

74

a more efficient inunune response in this group. Theenhanced bacterial clearance observed in burned micefed glutamine-supplemented diets was evident in themesenteric lymph nodes and the spleen, which are

particularly rich in lymphoid cells. Lymphocytes and

macrophages are avid glutamine consumers and requirethis amino acid for optimal function and proliferation.25.26Moreover, glutamine depletion, which easily occurs fol-

lowing trauma, may exert an in1illunosuppressive effect27and inipair macrophage-mediated bactericidal activity. 28

In conclusion, glutamine-enriched diets significantlyenhanced survival in gut-derived sepsis by improvingboth the intestinal-barrier function and the clearance

(killing) of translocated organisms. The quantitation oftranslocation and bacterial killing using bacteria labeledwith indium-111 oxine seems to have more advantagesthan bacteria labeled with [14C]glucose.

ACKNOWLEDGMENTS

This study was supported by United States PublicHealth Service Grant AI 12936 and by the Shriners ofNorth America.

REFERENCES

1. Alexander JW, Gianotti L, Pyles T, et al: Distribution and survival ofEscherichia coli translocating from the intestine following thermalinjury. Ann Surg 213:558-566, 1991

2. Wilmore DW, Smith RJ, O’Dwyer ST, et al: The gut: A central organafter surgical stress. Surgery 104:917-923, 1988

3. Fukushima R, Gianotti L, Alexander JW, et al: The degree of bacterialtranslocation is a determinant factor for mortality after burn injuryand is improved by prostaglandin analogs. Ann Surg 216:438-445, 1992

4. Rush BF, Sori AJ, Murphy TF, et al: Endotoxemia and bacteremiaduring hemorrhagic shock. Ann Surg 207:549-554, 1988

5. Marshall JC, Christou NV, Horn R, et al: The microbiology of multipleorgan failure. Arch Surg 123:309-315, 1988

6. Gianotti L, Alexander JW, Pyles T, et al: Prostaglandin E, analoguesmisoprostal and enisoprost decrease microbial transolocation andmodulate the immune response. Circ Shock 40:243-249, 1993

7. Gianotti L, Alexander JW, Nelson JL, et al: The role of early enteralfeeding and acute starvation on post-burn bacterial translocation andhost defense. Crit Care Med 22:265-272, 1994

8. Gianotti L, Alexander JW, Pyles T, et al: Arginine-supplemented dietsimprove survival in gut-derived sepsis and peritonitis by modulatingbacterial clearance: The role of nitric oxide. Ann Surg 217:644-654, 1993

9. Souba WW, Smith RJ, Wilmore DW: Glutamine metabolism by theintestinal tract. JPEN 9:608-617, 1985

10. Souba WW, Klimberg VS, Plumley DA et al: The role of glutamine inmaintaining a healthy gut and supporting the metabolic response toinjury and infection. J Surg Res 48:383-391, 1990

11. Alexander JW. Boyce ST, Babcock GF. et al: The process of microbialtranslocation. Ann Surg 212:-196-512, 1990

12. Gianotti L, Pyles T, Alexander JW, et al: Impact of blood transfusionand burn injury on microbial translocation and bacterial survival.Transfusion 32:312 317, 1992

13. Gianotti L, Alexander JW, Fukushima R, et al: Reduction of bacterialtranslocation with oral fibroblast growth factor and sucralfate. Am JSurg 165:195-201, 1993

14. Thakur ML, Lavender JP, Arnot RN, et al: Indium-111-labeled autolo-gous leukocytes in man. J Nucl Med 18:1012-1019, 1977

15. Windmueller HG, Spaeth AE. Uptake and metabolism of plasma glu-tamine by the small intestine. J Biol Chem 249:5070-5079, 1974

16. Fox AD, Kripke SA, De Paula J, et al: Effect of a glutamine-supple-mented enteral diet on methotrexate-induced enterocolitis. JPEN

12:325-331,198817. Klimberg VS, Souba WW, Dolson DJ, et al: Prophylactic glutamine

protects the intestine from radiation injury. Cancer 66:62-68, 199018. Klimberg VS, Salloum RM, Kasper M, et al: Oral glutamine accelerated

healing of the small intestine and improves outcome after whole bodyabdominal radiation. Arch Surg 125:1040-1045, 1990

19. Souba WW, Klimberg VS, Hautamaki RD: Oral glutamine reducesbacterial translocation following abdominal radiation. J Surg Res48:1-5, 1990

20. Burke DJ, Alverdy JC, Aoys E, et al: Glutamine-supplemented totalparenteral nutrition improves gut immune function. Arch Surg124:1369-1399, 1989

21. Alverdy JC, Aoys E, Moss GS: Effect of commercially available chemi-cally defined liquid diets on the intestinal microflora and bacterialtranslocation from the gut. JPEN 14:1-6, 1990

22. Barber AE, Jones WG II, Minei JP, et al: Glutamine or fiber supplemen-tation of a defined formula diet: Impact on bacterial translocation,tissue composition, and response to endotoxin. JPEN 14:335-343,1990

23. Wells CL, Jechorek RP, Erlandsen SL, et al: The effect of dietaryglutamine and dietary RNA on ileal flora, ileal histology and bacterialtranslocation in mice. Nutrition 6:70-75, 1990

24. Inoue Y, Grant JP, Snyder PJ: Effect of glutamine-supplemented intra-venous nutrition on survival after Escherichia coli-induced peritonitis.JPEN 17:41-46, 1993

25. Ardawi MSM, Newsholme EA: Glutamine metabolism in lymphocytesof the rat. Biochem J 208:743-751, 1982

26. Taudow G, Wiart J, Panijel J: Influence of amino acid deficiency andtRNA aminoacylation on DNA synthesis and DNA polymerase activityduring secondary immune response in vitro. Mol Immunol 20:255-262,1983

27. Kaflkewitz D, Bendich A: Enzyme-induced asparagine and glutaminedepletion and immune system function. Am J Clin Nutr 37:1025-1031,1983

28. Wells CL, Maddaus MA, Simmons RL: Role of macrophages in thetranslocation of intestinal bacteria. Arch Surg 122:48-53, 1987

at NORTHWESTERN UNIV LIBRARY on September 3, 2014pen.sagepub.comDownloaded from