synergistic therapeutic potential of dexamethasone and l-arginine in lipopolysaccharide-induced...

Journal of Surgical Research 140, 99–108 (2007)

Synergistic Therapeutic Potential of Dexamethasone and L-arginine inLipopolysaccharide-Induced Septic Shock

Saurabh Chatterjee, M.Sc., Sudha Premachandran, Ph.D., Jyoti Shukla, M.Sc., and T. B. Poduval, Ph.D.1

Immunology and Hyperthermia Section, Radiation Biology and Health Sciences Division, Bhabha Atomic Research Centre,Trombay, Mumbai, India

Submitted for publication August 25, 2006

doi:10.1016/j.jss.2006.09.002

Background. Dexamethasone (DEX) is demonstratedto have anti-inflammatory properties and known toinduce hemodynamic improvement in sepsis and sep-tic shock. L-arginine (L-arg), a semi-essential aminoacid, depending on its metabolic pathway, becomesvery essential in stress situations such as heatstroke,burns, sepsis, trauma, and wound healing. The aim ofthis study was to evaluate the synergistic therapeuticeffect of DEX and L-arg in rescuing the mice fromexperimental septic shock induced by bacterial lipo-polysaccharide (LPS). The experiments were designedto delineate the molecular mechanisms responsiblefor the increased therapeutic benefit of the combina-tion therapy (CT) in LPS-induced septic shock.

Methods. Acute endotoxemia was induced in Swissmale mice by i.p. injection of LPS (18 mg kg�1) at 0 h.LPS-treated mice were divided into four groups. Thefirst group (DEX group) received DEX (2 mg kg�1) i.p.at �2 h of LPS. The second group (L-arg group) re-ceived L-arg i.p. at a dose of 120 mg/kg at �6 h of LPSinjection. The third group (CT group) received DEX(2 mg kg�1) at �2 h LPS followed by L-arg at �6 h of LPSinjection. The fourth group received saline in place ofL-arg or DEX (LPS group). A sham group was alsoincluded, where normal mice received saline in placeof LPS or L-arg or DEX. At �6 h, mice from sham group,LPS group, and DEX group were sacrificed at �24 h.Mice from sham group, DEX group, L-arg group, andCT group were sacrificed to examine various parame-ters associated with LPS endotoxemia.

Results. The CT with DEX followed by L-arg signifi-cantly increased the survival of mice injected with alethal dose of LPS. Monotherapy with either DEX orL-arg given at the same dose and time did not increase

1 To whom correspondence and reprint requests should be addressedat Immunology and Hyperthermia Section, Radiation Biology andHealth Sciences Division, Bhabha Atomic Research Centre, Trombay,
Mumbai-400 085, India. E-mail: [email protected].
99

the survival of the mice injected with LPS. DEX ad-ministration could significantly reduce the levels ofserum TNF-�, IL-1�, IFN-�, aspartate aminotransfer-ase (ASAT), alanine aminotransferase (ALAT), andnitrite. DEX also down-regulated the expression ofliver-inducible nitric oxide synthase (iNOS), and up-regulated the levels of serum anti-inflammatory cyto-kines like TGF-�1 and IL-4, hepatic and splenic argi-nase, in LPS-injected mice. The enhanced therapeuticeffect of CT correlated with reduced pathological symp-toms, decreased Th1 cytokines, increased TGF-�1 andarginase levels compared to the mice administeredwith either of the monotherapies. The CT group hadsignificantly increased expression of hepatic Hsp 70and reduced septic shock associated histopathology,in lung and liver, compared to the mice treated witheither DEX or L-arg.

Conclusions. The therapeutic combination therapywith DEX and L-arg, at the appropriate dose, time, andsequence of administration, changed the cytokine pro-file, in favor of reducing the inflammatory response.The significantly enhanced survival observed in theCT group was accompanied by an increased hepaticHsp 70, hepatic arginase, splenic arginase, and de-creased organ injury. This novel concept of combinedtherapy could form the basis of an effective therapeu-tic approach in the treatment of sepsis and septicshock. © 2007 Elsevier Inc. All rights reserved.

Key Words: shock; cytokine profile; iNOS; arginase;Hsp 70; aspartate amino transferase; alanine amino-transferase; multiple organ dysfunction; combinationtherapy.

INTRODUCTION

Septic shock is the systemic inflammatory responseto infection frequently associated with hypotension,
hypoperfusion, tissue injury, and multiple organ fail-
0022-4804/07 $32.00© 2007 Elsevier Inc. All rights reserved.

100 JOURNAL OF SURGICAL RESEARCH: VOL. 140, NO. 1, JUNE 1, 2007

ure. The release of inflammatory mediators and acti-vation of the coagulation cascade, followed by the acti-vation of monocytes/macrophages and neutrophils, iscentral to the pathogenesis of sepsis. The sequence ofevents of the acute phase response to heat stress hasa similarity to the events occurring in sepsis [1].L-arginine (L-arg), a semi-essential amino acid, depend-ing on its metabolic pathway, becomes very essential instress situations such as heatstroke, sepsis, trauma,and wound healing [2–4]. Arginine therapy has beensuggested to be harmful in grossly septic patients dueto intensified systemic inflammatory response and over-production of inducible nitric oxide synthase (iNOS)-induced nitric oxide (NO) [4, 5]. The results promptedthe need for an anti-inflammatory therapeutic strat-egy, aimed at attenuating the adverse inflammation inseptic shock, before the administration of L-arg in theLPS endotoxemia model.

Glucocorticoids accelerated shock reversal in earlyhyperdynamic septic shock and also reduced the ex-pression of NF-�B, plasma nitrite/nitrate and proin-flammatory cytokines, increasing the blood flow, com-pared with untreated septic host, contributing to theincreased survival in selected populations [6–10]. Anti-inflammatory therapies used early in patients withhyperinflammatory state may improve the likelihood ofsurvival, while it may worsen the outcome if appliedduring the hypoimmune phase [6]. The attenuation ofinflammatory response, used in the therapy of sepsis,was typically accompanied by evidence of impairmentof antimicrobial defenses [11]. Since the side effects ofhigh-dose corticosteroids including hypertension, hy-perglycemia, thromboembolism, infection, gastrointes-tinal dysfunction, and L-arg are known to reverse theseabnormalities, we hypothesized that the combinationtherapy of Dexamethasone (DEX) and L-arg would besynergistic in the therapeutic management of experi-mental septic shock [4]. Our studies and earlier studieshad indicated that T-helper 1 (Th1) cytokines TNF-�,IL-1�, IFN-� induce iNOS activity and the Th2 cyto-kine IL-4 and TGF-�1 induce arginase activity; thepattern of cytokine profile dictates the arginine path-way in the injured host [3, 4, 12]. Based on the abovefindings, we hypothesized that down-regulation of theinitial inflammatory response with DEX, followed byL-arg, which initiates a repair pathway, may act syn-ergistically in the rescue of mice from LPS endotox-emia. This new concept of synergistic interaction ofDEX and L-arg has been tested in LPS-induced septicshock model, regarding its therapeutic efficacy. Exper-iments were designed to delineate the biological basisof the therapeutic effect.

The studies selected a range of stress response indi-cators associated with critical illness, in particular,septic shock. In the present report, we studied the
stress response indicators like mortality, pro- and anti-
inflammatory cytokine profile, nitrite, iNOS, arginase,and Hsp 70 in various tissues of endotoxemic mice. Thestudies also looked at the organ injury associated withseptic shock, using histopathology and measurement ofserum aspartate aminotransferase (ASAT) and alanineaminotransferase (ALAT) activities. The therapeuticability of DEX and CT in the modulation of the aboveindicators of endotoxemia was looked at, in explainingthe possible mechanism of therapeutic protection of-fered by combination therapy (CT) in preventing thelipopolysaccharide (LPS)-induced death of mice. Thestudies looked at the conceptual framework for the devel-opment of a novel combination therapy in treating exper-imental septic shock.

METHODS

Mice. All of the experiments were carried out in accordance withBARC Animal Ethics Committee guidelines on experimental ani-mals. Swiss/Bh inbred male mice were acclimatized to room temper-ature at 23.1°C and a 12-h dark/light cycle for 1 week before the startof the experiment.

Materials. The following materials were used: LPS (serotypeEscherichia coli; 026:B6); DEX as sodium phosphate salt (WockhardtLaboratories, Mumbai, India); L-arg (Sigma Chemical Co., St. Louis,MO).

Murine endotoxin-induced shock model. Mice were injected at 0 hwith LPS (18 mg kg�1), at 11 a.m. (called the 0 h). LPS-treated micewere divided into four groups. The first group (DEX group) receivedDEX, at �2 h of LPS, at a dose of 2 mg kg�1. The second group (L-arggroup) received L-arg at a dose of 120 mg kg�1, at �6 h of LPSinjection. The third group (CT group) received DEX (2 mg kg�1) at�2 h LPS followed by L-arg (120 mg kg�1) at �6 h of LPS injection.The fourth group received saline in place of L-arg or DEX (LPSgroup). A sham group was also included, where normal mice receivedsaline in place of LPS or L-arg or DEX. LPS, DEX, and L-arg weredissolved in pyrogen-free saline and were administered intraperito-neally (i.p.). At �6 h, mice from sham group, LPS group, and DEXgroup were sacrificed and at �24 h, mice from sham group, DEXgroup, L-arg group, and CT group were sacrificed to examine variousparameters associated with LPS endotoxemia. Blood was collectedthrough retro-orbital puncture. Serum samples were stored at –80°Cuntil further use. For experiments in which mortality was the out-come parameter, separate sets of experiments were carried out andmice were observed twice per day for 7 days after the injection of LPSand the mortality was recorded. For assessing the visible patholog-ical symptoms, mice were observed continuously for the first 12 hafter the injection of LPS. DEX � 6 h group, DEX � 24 h group, LA� 24 h group, CT � 24 h group, LPS � 6 h group, sham � 6 h group,and sham � 24 h group are the subgroups of DEX, L-arg, CT, LPS,and sham group, respectively, indicating the time of sacrifice of themice, with respect to the administration of LPS.

Serum cytokines and NO assay. Levels of serum TNF-�, IL-1�,IFN-�, and IL-4 were measured by commercial ELISA kits (Opt EIAElisa kits; BD Biosciences Pharmingen, San Diego, CA). The methodof TGF-�1 estimation [2] in sera involved only the active form of thecytokine and acidification of the samples were carried out to includethe latent TGF-�1 content. Serum nitrite levels were estimatedaccording to the method described earlier [13].

Serum ALAT and ASAT measurements. Serum ALAT and ASATlevels in mice were estimated by a colorimetric kit (Ecoline R; MerckLimited, Mumbai, India) following manufacturer’s protocol. Briefly
50 �L of diluted serum was added to reaction solution (4:1). The

101CHATTERJEE ET AL.: L-ARGININE, DEXAMETHASONE FOR ENDOTOXEMIA

resultant reaction mixture was incubated at 37.4°C for 10 min andthe absorbance of the solution was read at 340 nm.

SDS electrophoresis and Western blots. Snap-frozen liver sectionswere homogenized and the homogenate was loaded (50 �g in 10 �l)and resolved onto a 10% SDS-PAGE. The proteins were electroblot-ted onto nitrocellulose paper. Western blotting was carried out byincubating the blots with monoclonal antibodies specific to iNOS(Calbiochem, Darmstadt, Germany), or Hsp 70 (Calbiochem) fol-lowed by incubation with rabbit anti-mouse antibody conjugatedwith alkaline phosphatase. The bands were visualized with 5-Bromo-4-Chloro-3-indolyphosphate/Nitrate tetrazolium as substrate. Thespecific bands were quantified using Labimage software.

Arginase activity measurements. Arginase activity was measuredby the method reported earlier [2]. One unit of enzyme activity wasdefined as the amount of enzyme that catalyzes the formation of1 �mol urea per min.

Preparation of tissues for light microscopy. Light microscopyanalysis of liver and lung were performed on tissue slices fixed inneutral buffered formalin, embedded in paraffin, sectioned at 5 �m,and stained with hematoxylin and eosin (H&E). Stained tissue sec-tions were examined microscopically by a scientist who was blindedto the experimental conditions of the animals. The extent, nature,and distribution of organ pathology were recorded. Sections werecarefully examined at higher magnification (�400). Polymorphonu-clear (PMN) cell infiltration in the liver and lung histopathology wasmonitored as per the methods reported [14, 15].

Statistical analyses. Data were pooled whenever appropriate tohave an increased sample size. Statistical significance of survivalwas assessed using the Fisher’s exact probability test. Levels ofcytokines, nitrite, arginase, and band densities of Western blots wereanalyzed using the using the Student’s t-test, followed by Bonferronicorrection. The results of PMN infiltration in the liver were analyzedby Mann–Whitney t-test. Significance was set at P � 0.05.

RESULTS

Therapeutic Effect of CT on Survival

Mice that received LPS exhibited symptoms of diar-rhea, piloerection, within 0.5 h of injection did not con-gregate for group warmth and became progressively le-thargic and unresponsive to stimulus when monitored at12 h. Mice administered LPS also had glazed eyes andwere tachypneic. Mice from the LPS group, DEX group,and L-arg group, when observed at �12 h of LPS for theirbehavioral patterns associated with septic shock, exhib-ited more severe responses as compared to the CT group.Results indicated that there was a significant increase inthe survival of mice (P � 0.05) in CT group compared tothe LPS group (Fig. 1). The survival in the DEX or L-arggroup was not statistically different from the LPS group.

Cytokine Levels in the Serum

To determine the molecular mechanism of the ther-apeutic benefit of the CT in LPS endotoxemia model,the levels of serum cytokines were measured at �6 hand �24 h of LPS. There was a significant increase inthe levels of TNF-�, IL-1�, and IFN-� in mice from theLPS group (P � 0.05) as compared to the sham group atboth the time points studied (Fig. 2A–C). The serumlevels of all pro-inflammatory cytokines in mice from
DEX � 6 h group were significantly lower (P � 0.05) as
compared to the LPS � 6 h group. The levels of IL-1�and IFN-� in the CT � 24 h group were reduced furthercompared to the either DEX � 24 h group or LA � 24 hgroup. The levels of TNF-� in the CT � 24 h group weresignificantly lower compared to the DEX � 24 h group.There was a significant increase in the levels of anti-inflammatory cytokines, namely, IL-4 and TGF-�1, inmice from the LPS � 6 h group compared to the sham� 6 h group (P � 0.05), which increased further in theDEX � 6 h group (P � 0.05) (Fig. 3A–B). The CT � 24 hgroup had elevated levels of TGF-�1 compared toeither the DEX � 24 h group or the LA � 24 h group.

Expression of Liver iNOS and Serum Nitrite Levels

The results demonstrated a significant increase inthe expression of iNOS protein in mice from the LPS �6 h group compared to the mice from sham � 6 h group(Fig. 4A). This increased iNOS expression was signifi-cantly reduced by the administration of DEX (P � 0.05).

There was a significant increase in the levels of ni-trite in mice from the LPS group (P � 0.05) as com-pared to the sham group at both the time points stud-ied (Fig. 4B). The serum levels of nitrite in mice fromthe DEX � 6 h group were significantly lower (P �0.05) as compared to the LPS � 6 h group.

Arginase Activity

Hepatic arginase activity decreased in mice fromLPS � 6 h group (P � 0.05) (Fig. 5A) when compared tothe sham � 6 h group. DEX � 6 h group had signifi-

FIG. 1. Synergistic therapeutic potential of CT in rescuing themice from LPS-induced endotoxemia. Cumulative survivals of vari-ous groups of mice indicated above were plotted as a function of time.The results represent pooled data from three experiments. The frac-tion of survivors has been indicated on the graph. The number on thex-axis refers to the hours post-LPS. *P � 0.05 versus either LPS orDEX or L-arg group.

cantly increased arginase activity (P � 0.05) compared

**P


to the LPS � 6 h group. Mice from the CT � 24 h grouphad increased hepatic arginase activity compared tomice from either the DEX � 24 h group or the L-arg �24 h group (P � 0.05). Splenic arginase activity de-creased in mice from the LPS � 6 h group (P � 0.05)(Fig. 5B) when compared to the sham � 6 h group.DEX � 6 h group had significantly increased splenicarginase activity (P � 0.05) compared to the LPS � 6 hgroup. Mice from CT � 24 h group had increased levels ofsplenic arginase activity compared to mice from eitherthe DEX � 24 h group or the LA � 24 h group (P � 0.05).

Hsp 70 Expression

Liver homogenate from mice from various groupswas prepared �6 h and �24 h of LPS administrationand subjected to SDS-PAGE and Western blotting. Theresults (Fig. 6) demonstrated a significant increase in

FIG. 2. (A–C) Effect of DEX or CT on serum Th1 cytokines in LPSwere monitored at �6 h and �24 h of LPS administration. The numbesham � 6 h group. **P � 0.05 versus respective LPS � 6 h group. *

the expression of Hsp 70 protein in mice from the

LPS � 6 h group compared to the mice from the sham �6 h group. This increased level of Hsp 70 expressionwas significantly reduced by the administration ofDEX (P � 0.05). However, mice from the CT � 24 hgroup had increased Hsp 70 expression when com-pared to either the DEX � 24 h group or the LA � 24 hgroup.

Serum ASAT and ALAT Levels

Mice from the LPS � 6 h group had significantlyelevated serum ASAT and ALAT levels, as compared tothe mice from sham � 6 h group (P � 0.05) (Table 1).The increased levels of both the enzymes thus observedwere reduced in mice from the DEX � 6 h group, whencompared to the LPS � 6 h group (P � 0.05).

Histopathology

The degree of multiple systemic injuries in acute

ated mice: Cytokine levels from various groups (five to six mice/group)n the x-axis refers to the hours post-LPS. *P � 0.05 versus respective

� 0.05 versus respective DEX � 24 h or LA � 24 h group.

-trer o

endotoxemia induced by i.p. administration of LPS was


further assessed by histopathological examinations ofliver and lung tissue sections. H&E-stained liver sec-tions from mice of the LPS � 6 h group showed signif-icant infiltration of granulocytes, disruption of cellularmorphology, and hepatocellular necrosis compared tothe sham � 6 h group. These symptoms were largelyattenuated in mice from CT � 24 h group (Fig. 7A–F).There was a significant decrease in the PMN infiltra-tion in the liver of mice from the DEX � 6 h group (P �

FIG. 3. (A–B) Effect of DEX or CT on serum anti-inflammatory cto six mice/group) were monitored at �6 h and �24 h of LPS admin0.05 versus respective sham � 6 h group. **P � 0.05 versus respectiv24 h group.

FIG. 4. (A–B) Effect of DEX on hepatic iNOS and serum nitrite inData shown are representative of one of the three independent expe(B) Serum nitrite levels. Serum nitrite levels from various groupsadministration. The number on the x-axis refers to the hours post-L
respective LPS �6 h group.
0.05) compared to the LPS � 6 h group(data notshown). At �24 h the CT group had significantly de-creased PMN infiltration compared to either theDEX � 24 h group or the LA � 24 h group (P � 0.05).H&E-stained sections of whole lungs were examinedmicroscopically. At a low magnification, sectionsfrom mice treated with LPS revealed edematous-interlobular septae, damage to the epithelial celllining, and infiltrates (Fig. 8A–F). At a higher mag-

kines in LPS-treated mice: Cytokine levels from various groups (fiveation. The number on the x-axis refers to the hours post-LPS. *P �PS � 6 h group. ***P � 0.05 versus respective DEX � 24 h or LA �

S-treated mice: (A) Western blot analysis of iNOS expression in liver.ents. The specific bands were quantified using Labimage software.

ve to six mice/group) were monitored at �6 h and �24 h of LPS. *P � 0.05 versus respective sham � 6 h group. **P � 0.05 versus

ytoistre L

LPrim(fiPS


nification the infiltrate consisted of neutrophils andwere seen filling alveolar and bronchial spaces, withintervening areas of uninvolved lung tissue. Thesepathological features are consistent with acute lunginjury. In marked contrast, there was a significantattenuation of the above pathological features inmice from the CT � 24 h group.

DISCUSSION

Our results convincingly demonstrate the ability of acombination approach to rescue the mice from LPS-

FIG. 5. (A–B) DEX and CT increase hepatic and splenic arginaseto six mice/group) were monitored at �6 h and �24 h of LPS admin0.05 versus respective sham group. **P � 0.05 versus respective LPSgroup.

FIG. 6. DEX decreases and CT increases hepatic Hsp 70 expres-sion in LPS-treated mice: Western blot analysis of Hsp 70 expressionin liver. Data shown are representative of one of the three indepen-dent experiments. The specific bands were quantified using Labim-age software. *P � 0.05 versus sham � 6 h group. **P � 0.05 versus
LPS � 6 h group. ***P � 0.05 versus DEX � 24 h group.
induced mortality. Experimental and clinical data in-dicate that severe stress, sepsis, and injury have all ofthe characteristics of an arginine-deficiency state, sug-gesting the need to replenish L-arg [2–4, 16]. The tim-ing and degree of any therapeutic approach are crucial;the amplification of a response that is primarily harm-ful at the onset of the illness may be beneficial ifdelivered at a later phase of the disease, or vice versa.It has been hypothesized that systemic inflammationmight be undesirably intensified by immune-enhancingnutrients like arginine in critically ill patients [5]. Thisconcept has been demonstrated in our model of heat-stroke, wherein a high dose of L-arg given at the onset

tivity in LPS-treated mice: Arginase levels from various groups (fiveation. The number on the x-axis refers to the hours post-LPS. *P �6 h group. ***P � 0.05 versus respective DEX � 24 h or LA � 24 h

TABLE 1

Serum ASAT and ALAT Levels

GroupASAT (IU/L)(mean � SE)

ALAT (IU/L)(mean � SE)

Sham � 6 h 174.57 � 2.83 82.25 � 7LPS � 6 h 282.6 � 16.2* 440.75 � 15.6*DEX � 6 h 167.5 � 5.69** 126.75 � 10.6**Sham � 24 h 168 � 4.6 118 � 12DEX � 24 h 178 � 12 162.75 � 20LA � 24 h 182 � 26 133.75 � 16CT � 24 h 167 � 21 140.5 � 18

Note. Effect of therapy on serum ASAT or serum ALAT activity inLPS-treated mice: ASAT and ALAT levels from various groups(5– 6 mice/group) were monitored at �6 h and �24 h of LPSadministration.

* P � 0.05 versus respective sham � 6 h group.

acistr

�

** P � 0.05 versus respective LPS � 6 h group.


of heatstroke increased the mortality, but the samedose of L-arg rescued the mice when administered at alater phase [2, 17].

Patients with septic shock have a biphasic immuno-logical response. The period immediately after an acuteinjury is characterized by increased expression of

FIG. 7. (A–F) Liver histopathology in LPS-induced septic shock athree independent experiments. Nine consecutive microscopic fields iwhich shows severe infiltration of granulocytes and altered cellulahepatocellular necrosis can be seen. (E) L-arg � 24 h group. (F) CT �is observed. Statistical analyses of the PMN infiltration in the liver:6 h group versus DEX � 6 h group; P � 0.05, CT � 24 h group versuis available online.)

FIG. 8. (A–F) Lung histopathology in LPS-induced septic shock athree independent experiments. (A) Sham group. (B) LPS � 6 h grouleukocytes, and damage of the epithelial lining. (C) DEX � 6 h grouprecruitment of leukocytes, and swelling of the epithelial lining. (E) L
lung edema, recruitment of leukocytes, and damage of the epithelial lin
proinflammatory cytokine expression leading to a laterperiod of generalized immunosuppression [6, 18] Thepro- and anti-inflammatory cytokine balance, whichshould mediate tissue repair and recovery, if uncon-trolled, can produce tissue injury on one spectrum andimmunosuppression on the other. Controlled inflam-

its amelioration by CT: Data shown are representative of one of thech slide were scored for PMN. (A) Sham group. (B) LPS � 6 h group,orphology. (C) DEX � 6 h group. (D) DEX � 24 h group, where

4 h group, where significant attenuation of symptoms of septicemia0.05, LPS � 6 h group versus sham � 6 h group; P � 0.05, LPS �

ther DEX � 24 h group or LA � 24 h group. (Color version of figure

its amelioration by CT. Data shown are representative of one of thehich shows acute lung injury including lung edema, recruitment of

) DEX � 24 h group, where acute lung injury including lung edema,24 h group. (F) CT � 24 h group, where significant attenuation of

ndn ear m

2P �s ei

ndp, w. (DA �

ing can be observed. (Color version of figure is available online.)


mation during a period after injury is essential fortissue repair and maintenance of immune competence.The increased magnitude and duration of the inflam-matory response are associated with delayed restora-tion of homeostasis and increased tissue injury leadingto multiple organ failure [18]. However, persistence ofthe hypo-responsiveness is associated with increasedrisk of infection and death. The control of immediateproinflammatory responses of the endotoxemic micecan initiate the recovery and homeostasis of the host.Anti-inflammatory therapies used early in patientswith a hyper-inflammatory state may improve the like-lihood of survival, while it may worsen the outcome ifapplied during the hypo-immune phase [6]. Glucocor-ticoids accelerated shock reversal in early hyperdy-namic septic shock and also reduced the proinflammatorycytokines [6–10]. Since DEX administration inhibitedNO formation, down-regulation of iNOS mRNA expres-sion, in a manner that is independent of endothelialNOS tetrahydrobiopterin and arginine transport, wehypothesized that DEX and L-arg may act synergisti-cally in rescuing the septic mice, if L-arg is adminis-tered after treating the endotoxemic mice with DEX[19, 20]. Selective blockade of iNOS enzyme, by theadministration of DEX, at a time when arginine trans-port and/or synthesis is elevated might promote theproduction of other bioactive compounds derived fromL-arg, which is favorable for the homeostasis of theendotoxemic mice [21]. It is well known that both IL-1�and TNF-� play a crucial role in the development ofsepsis [22, 23]. Available data support a causative roleof TNF-� and IL-1� acting synergistically in septicmyocardial depression, leading to the disruption of ho-meostasis of the cardiovascular system, leading to sep-tic shock [24]. Experimental studies indicate IFN-� asa mediator of the lethality induced by LPS and TNF-�[25]. In the present study, the increased survival ofmice in the CT � 24 h group correlated with the de-creased pro-inflammatory responses, suggesting therationale of combined therapy in rescuing the micefrom endotoxemia-induced death.

Our results indicated that DEX administration in-hibited an increase in iNOS and nitrite production,decreased the systemic inflammatory response as re-flected by the decreased levels of IL-1-�, TNF-�, andIFN-�, and increased anti-inflammatory cytokines likeTGF-�1 and IL-4 in mice from the DEX � 6 h group.The reduction in the acute inflammation initiated byDEX created a therapeutic window for L-arg to initiatea repair pathway and rescue the endotoxemic mice.Thus the combination treatment could be synergistic inthe recovery of the host from the pathophysiology as-sociated with the septic shock induced by LPS. Micefrom the LPS group, DEX group, and L-arg group,when observed at �12 h of LPS administration, had
significant piloerection, glazed eyes. They were pro-
gressively lethargic and did not congregate for groupwarmth. The symptoms were significantly amelioratedin the CT group, when monitored at �12 h of LPSadministration. Complete amelioration of the abovesymptoms were observed in mice from the CT group,compared to either the DEX � 24 h group or the LA �24 h group. We did not observe any therapeutic benefit inmice, when L-arg and DEX are administered either indi-vidually or together at �2 h of LPS administration,indicating the need to modify the host status, forachieving the therapeutic benefit with L-arg (data notpresented).

Mice treated with LPS exhibited significantly higherHsp 70 expression at 6 h post-LPS administration andDEX reduced the expression of Hsp 70 thus observed.Hsp 70, the most abundant of the heat shock proteins,has been postulated as responsible for the modulationof iNOS activation, in endotoxemia [26]. The up-regulation of Hsp has been demonstrated to inhibitNF-�B activation and cytokine production by LPS-induced Kupffer cells [27]. Mice from the DEX � 6 hgroup had significantly reduced levels of Hsp 70, com-pared to the LPS � 6 h group, indicating the deleteri-ous effect of DEX on LPS-induced endotoxemia. Ourearlier studies had indicated the ability of L-arg admin-istration, to increase the production of Hsp 70, in heat-stroke mice [2]. Mice from the CT � 24 h group hadsignificantly higher expression of Hsp 70, compared tothe DEX � 24 h group, which suggests the explicitrationale in using the conjunct therapy in improvingsurvival of mice from LPS-induced endotoxemia.

Urea cycle enzymes are down-regulated and iNOS isup-regulated in sepsis [28]. L-arg administration hasbeen demonstrated to increase arginase activity andthe down-regulation of iNOS expression in acute stressconditions [3, 4]. DEX administration at �2 h of LPSinjection resulted in iNOS inhibition and a subsequentup-regulation in arginase activity. A fall in the argi-nase activity following LPS administration in hepaticand splenic tissues at �6 h may be due to the fact ofincreased TNF-� and IL-1 � at the same time, whichalong with IFN-� not only increases iNOS activity butdecreases arginase activity in intact cells [29]. Ourearlier studies have suggested the importance ofTGF-� in the metabolism of L-arg through the arginasepathway in rescuing the mice from heatstroke [3]. Aconcomitant increase in IL-4 and TGF-�1 at �6 h ofLPS, by the administration of DEX, might divert theL-arg administered at �6 h of LPS, toward arginasepathway, which is favorable for the host homeostasisand might trigger an early repair phase where aninduction of arginase activity is of paramount impor-tance apart from inhibiting NO production. TGF-�1and IL-4 are known to activate the arginine–arginasepathway in acute stress conditions, which are favor-
able for wound healing [3, 4].


The logic of using DEX before the administration ofL-arg in endotoxemic mice is also supported by theobservation that both TGF-�1 and glucocorticoids havebeen shown to improve hemodynamics in shock andreduce the hypotension and degree of vascular hypo-reactivity [30].

In this study, a marked increase in serum ASAT andALAT concentrations in mice from the LPS � 6 h groupindicated severe hepatocellular damage. Incidentally asignificant increase in NO as well as pro-inflammatorycytokine surge is also evident at this time point. It iswell documented that release of NO is one of the pri-mary events of LPS-induced sepsis, which led to in-creased production of TNF-�. There is also evidencethat increased proinflammatory mediators and NO candirectly lead to hepatocellular damage as representedby the release of hepatocellular enzymes includingASAT [31]. Incidence of hepatocellular infiltration ofPMNs and necrosis are a common prognosis for multi-organ dysfunction associated with sepsis. Endotoxemicmice showed significant infiltration of granulocytes,disruption of cellular morphology, and hepatocellularnecrosis.

The alleviation of organ dysfunction in mice from theDEX � 6 h group was improved further by the admin-istration of L-arg. DEX treatment markedly improvedthe LPS-induced hepatic dysfunction, as shown by alarge reduction of these enzyme activities. The resultsof decreased ASAT and ALAT levels in mice from theDEX � 6 h group and a subsequent decrease in thelevels of IL-1�, TNF-�, and IFN-� in mice from CT 24-hgroup correlated with the prevention of hepatic dys-function, indicating the combined effect of DEX andL-arg in the homeostasis of the liver function, which isvery crucial in the recovery of mice from septic shock.LPS mice from the LPS � 6 h group showed acute lunginjury including lung edema, recruitment of leuko-cytes, and damage of the epithelial lining. Combinationtherapy ameliorated these lesions much better whencompared to the mice from either the LA � 24 h groupor the DEX � 24 h group. Taken together, a dualdosage of DEX and L-arg led to a significant reductionin the number and intensity of organ dysfunction andrescue of mice from LPS-endotoxemia.

Monotherapy directed against a specific marker ofsepsis is without success [11, 32]. The therapeutic pro-tocol should aim at targeting a plethora of biomarkersthat seem to play a crucial role in the development ofseptic shock. Our CT has the potential to target manyof the markers associated with septic shock. Arginine isa conditional indispensable amino acid with numerousphysiological and pharmacological activities, which in-clude the modulation of hypertension, thromboembolism,infection, and gastrointestinal dysfunction [4, 33, 34].Extension of the above scientific ideas can help in the
development of new and effective therapeutic ap-
proaches to treatment of sepsis and septic shock. Al-though the LPS-induced endotoxemia model cannot betransferred directly to the clinical situation, it providesa rational basis for the development of combinationaltherapy for human sepsis and septic shock.

ACKNOWLEDGMENTS

The authors sincerely acknowledge Narendra S. Sidnalkar andKashinath Munankar for useful technical assistance. The authorsalso sincerely acknowledge Dr. D. Datta, Health Physics Division ofour center, for advice in the statistical analyses.

REFERENCES

1. Kurahashi K, Kajikawa O, Sawa T, et al. Pathogenesis of septicshock in Pseudomonas aeruginosa pneumonia. J Clin Invest1999;104:743.

2. Chatterjee S, Premachandran S, Sharma D, et al. Therapeutictreatment with L-arginine rescues mice from heat stroke-induced death: physiological and molecular mechanisms. Shock2005;24:341.

3. Chatterjee S, Premachandran S, Bagewadikar RS, et al. Argi-nine metabolic pathways determine its therapeutic benefit inexperimental heatstroke: role of Th1/Th2 cytokine balance. Ni-tric Oxide (in press).

4. Tong BC, Barbul A. Cellular and physiological effects of argi-nine. Mini Rev Med Chem 2004;4:823.

5. Suchner U, Heyland DK, Peter K. Immune-modulatory actionsof arginine in the critically ill. Br J Nutr 2002;87:S121.

6. Hotchkiss SR, Karl EI. The pathophysiology and treatment ofsepsis. N Engl J Med 2003;348:138.

7. Mansart A, Bollaert PE, Seguin C, et al. Hemodynamic effectsof early versus late glucocorticosteroid administration in exper-imental septic shock. Shock 2003;19:38.

8. Oppert M, Schindler R, Husung C, et al. Low-dose hydrocor-tisone improves shock reversal and reduces cytokine levels inearly hyperdynamic septic shock. Crit Care Med 2005;33:2457.

9. Chang CK, Llanes S, Schumer W. Effect of dexamethasone onNF-�B activation, tumor necrosis factor formation, and glucosedyshomeostasis in septic rats. J Surg Res 1997;72:141.

10. Aboab J, Nardi O, Lipiner D, et al. Emerging drugs for thetreatment of sepsis. Exp Opin Emerg Drugs 2006;11:7.

11. Lorente JA, Marshall JC. Neutralization of tumor necrosis fac-tor in preclinical models of sepsis. Shock 2005;24:107.

12. Bronte V, Zanovello P. Regulation of immune responses byL-arginine metabolism. Nat Rev Immunol 2005;5:641.

13. Green LC, Wagner DDA, Glogowski J, et al. Analysis of nitrate,nitrite and [15N] nitrate in biological fluids. Anal Biochem1982;124:131.

14. Matesic LE, Niemitz EL, Maio AD, et al. Quantitative trait locimodulate neutrophil infiltration in the liver during LPS-induced inflammation. FASEB J 2000;14:2247.

15. Shalit I, Horev-Azaria L, Fabian I, et al. Immunomodulatoryand protective effects of moxifloxacin against Candidaalbicans-induced bronchopneumonia in mice injected withcyclophosphamide. Antimicrob Agents Chemother 2002;46:
2442.


16. Luiking YC, Poeze M, Dejong CH, et al. Sepsis: an argininedeficiency state? Crit Care Med 2004;32:2135.

17. Poduval TB, Chatterjee S, Sainis KB. Effect of nitric oxide onmortality of mice after whole body hyperthermia. Int J Hyper-thermia 2003;19:35.

18. Molina PE. Neurobiology of the stress response: contribution ofthe sympathetic nervous system to the neuroimmune axis intraumatic injury. Shock 2005;24:3.

19. Skimming JW, Nasiroglu O, Huang CJ, et al. Dexamethasonesuppresses iNOS yet induces GTPCH and CAT-2 mRNA ex-pression in rat lungs. Am J Physiol Lung Cell Mol Physiol2003;285:L484.

20. Radomski MW, Palmer RM, Moncada S. Glucocorticoids inhibitthe expression of an inducible, but not the constitutive, nitricoxide synthase in vascular endothelial cells. Proc Natl Acad SciUSA 1990;87:10043.

21. Lortie MJ, Ishizuka S, Schwartz D, et al. Bioactive products ofarginine in sepsis: tissue and plasma composition after LPS andiNOS blockade. Am J Physiol Cell Physiol 2000;278:C1191.

22. Fletcher JR, Collins JN, Graves ED 3rd, et al. Tumor necrosisfactor-induced mortality is reversed with cyclooxygenase inhi-bition. Ann Surg 1993;217:668, 674.

23. Laneuville O, Breuer DK, Dewitt DL, et al. Differential inhibi-tion of human prostaglandin endoperoxide H synthases-1 and-2 by nonsteroidal anti-inflammatory drugs. J Pharmacol ExpTher 1994;271:927.

24. Court O, Kumar A, Parrillo JE, et al. Clinical review: myocar-dial depression in sepsis and septic shock. Crypt Care 2002;6:
500.
25. Doherty GM, Lange JR, Langstein HN, et al. Evidence forIFN-gamma as a mediator of the lethality of endotoxin andtumor necrosis factor-alpha. J Immunol 1992;149:1666.

26. Lau SS, Griffin TM, Mestril R. Protection against endotoxemiaby HSP70 in rodent cardiomyocytes. Am J Physiol Heart CircPhysiol 2000;278:H1439.

27. Sun D, Chen D, Du B, et al. Heat shock response inhibitsNF-kappaB activation and cytokine production in murineKupffer cells. J Surg Res 2005;129:114.

28. Tabuchi S, Gotoh T, Miyanaka K, et al. Regulation of genes forinducible nitric oxide synthase and urea cycle enzymes in ratliver in endotoxin shock. Biochem Biophys Res Commun 2000;268:221.

29. Shearer JD, Richards JR, Mills CD, et al. Differential regula-tion of macrophage arginine metabolism: a proposed role inwound healing. Am J Physiol 1997;272:E181.

30. Titheradge MA. Nitric oxide in septic shock. Biochim BiophysActa 1999;1411:437.

31. Wang JH, Redmond HP, Wu QD, et al. Nitric oxide mediateshepatocyte injury. Am J Physiol 1998;275:G1117.

32. Mazolewski PJ, Barber A, Williams S, et al. Attenuating tumornecrosis factor alpha does not ameliorate other cytokine andperoxidase products during sepsis. Am J Surg 1999;178:556.

33. Sukhotnik I, Mogilner J, Krausz MM, et al. Oral argininereduces gut mucosal injury caused by lipopolysaccharide endo-toxemia in rat. J Surg Res 2004;122:256.

34. Cylwik D, Mogielnicki A, Kramkowski K, et al. Antithromboticeffect of L-arginine in hypertensive rats. J Physiol Pharmacol
2004;55:563.

synergistic therapeutic potential of dexamethasone and l-arginine in lipopolysaccharide-induced...

Documents