WO2008149103A1 - Methods of assessing and treating multiple-organ failure - Google Patents

Abstract

The present invention provides medicaments for the prevention and/or treatment of multiple-organ failure as well as methods for determining the multiple- organ failure status of a patient.

Description

METHODS OF ASSESSING AND TREATING MULTIPLE-ORGAN

FAILURE FIELD OF THE INVENTION

The present invention provides medicaments for the prevention and/or treatment of multiple-organ failure as well as methods for determining the multiple- organ failure status of a patient. BACKGROUND

Multiple-organ failure, for which there is no effective prevention or treatment, is a major cause of death and can stem from conditions such as severe acute pancreatitis although the pathogenesis of multiple-organ failure is poorly understood.

Severe acute pancreatitis (AP) is characterised by acute inflammation in the pancreas, usually due to excessive alcohol consumption or the passage of gallstones (1). Most attacks are self-limiting, but on 20% of occasions AP causes multiple-organ failure (MOF) with a mortality of 30% to 50% (1, 2). How the initiating sterile event of pancreatic acinar cell injury erupts into an overwhelming systemic inflammatory response remains unknown (3, 4).

In patients with pancreatitis-induced MOF, the organ most frequently injured is the lung followed by the kidney and coagulation system, usually in a cumulative pattern (2, 7). It has been shown that other non-microbial triggers of multiple organ failure, for example trauma, hemorrhagic shock and severe burns, result in gut injury and the production of one or more factors carried in the humoral component of post- nodal mesenteric lymph and capable of mediating distant organ injury (8-10). In AP the first capillary bed encountered by mesenteric lymph draining from the injured gut is the lung via the thoracic duct, superior vena cava and pulmonary arteries. To date, the identity of these lymph-borne mediators has remained elusive (11). The mesenteric lymphatic system and gut-associated lymphoid tissue (GALT), are secondary lymphoid tissues which are strategically placed to marshal the interface between host and environment in the gastrointestinal tract (12). In health, immature dendritic cells (DCs) constitutively migrate from the periphery to draining lymph nodes (LNs) in the mesenteric lymphatic system, constantly trafficking debris from apoptotic and effete intestinal epithelial cells (13, 14) to maintain peripheral tolerance (15). In situations of acute inflammation, innate immune activation (16) results in chemokine-mediated DC recruitment to sites of injury and subsequent migration to the relevant draining nodes (17). DC recruitment and mobilisation to mesenteric lymph nodes (MLNs) is rapid, and has been shown in airway mucosa to occur within one hour of bronchial antigen administration (17).

In the MLN, DCs regulate the activity of the effector arms of innate and adaptive immunity through several mechanisms, including the production of cytokines, especially IL- 12 p75, IL-27, IFNγ, IL-4, IL-5, IL-IO and TGFβ, and through the differential expression of co-stimulatory molecules, for example OX40L (18). Other small soluble molecules are involved, in particular the immunoregulatory catabolites of tryptophan formed via the kynurenine pathway, controlled by the inducible rate-limiting enzyme indoleamine-2,3-dioxygenase (IDO) (19-22). Kynurenines, which are utilised by murine placenta to prevent foetal rejection by the elimination of maternal T cells (22), are important for the induction and maintenance of peripheral tolerance by DCs (21), and have been recently shown to participate in suppurative granuloma formation (23). In the formation of kynurenines, IDO catalyses the conversion of tryptophan to N-formylkynurenine, which is converted to kynurenine and then 3-hydroxykynurenine by the specific, constitutively active enzyme kynurenine 3-monooxygenase (24). Downstream metabolites of kynurenine are also generated without increased IDO activity if kynurenine is introduced directly as a substrate, with regulatory effects on the elimination and suppression of autoreactive T cell clones (24).

The present inventors have discovered that, experimental AP results in the presence of one or more cytotoxic factors transported in mesenteric lymph that are capable of causing lung injury, endothelial and gut epithelial cell death, neutrophil respiratory burst activation and decreased erythrocyte deformability. Furthermore, the hierarchical clustering of lymph-derived mass spectra generated by surface-enhanced laser desorption ionisation time-of-flight mass spectrometry (SELDI-TOF MS) predicted the onset of lymph cytotoxicity and its resolution to an inert state in survivors of AP. In particular, the inventors have shown that the lymph contained elevated concentrations of kynurenine and 3-hydroxykynurenine.

Moreover, in human patients with varying severities of pancreatitis-associated organ failure, it was shown that elevated peripheral plasma kynurenine concentrations correlated in real-time with the Acute Physiology and Chronic Health Evaluation II

Score (APACHE II), Multiple Organ Dysfunction Score (MODS), and preceded a requirement for artificial organ support in the intensive care unit.

The object of the invention is therefore to develop medicaments for use in treating MOF and methods for assessing a subject's MOF status. SUMMARY OF THE INVENTION

As stated, the present invention is based upon the observation that, following a multiple-organ failure (MOF) initiating event, certain compounds carried in biological fluids play an important role in the development or exacerbation of MOF. These proteinaceous compounds may be detected in biological fluids. The precise proteinaceous content of any given biological fluid may vary depending on the patient's MOF status - i.e. whether the patient has recently been exposed to a MOF initiating event and is likely to develop MOF, is currently suffering from or experiencing MOF, has recovered/is recovering from MOF or has not been exposed to a MOF initiating event.

The knowledge that the presence, absence or increased/decreased expression of certain proteins is indicative of MOF status provides an opportunity for the development of therapeutics for treating and/or preventing MOF.

The present inventors have discovered that, among others, the proteinaceous compounds involved in the development of MOF include components of the kynurenine pathway.

Thus, in a first aspect of the present invention there is provided a use of a compound capable of modulating the kynurenine pathway for the preparation of a medicament for the treatment and/or prevention of multiple-organ failure (MOF). In a second aspect, there is provided a method of treating MOF, said method comprising the step of administering a therapeutically effective amount of a compound capable of modulating the kynurenine pathway.

In a third aspect, there is provided the use of a compound capable of modulating the kynurenine pathway for treating MOF. The kynurenine pathway is responsible for the catabolism of tryptophan which is first converted to N-formylkynurenine by either of the enzymes tryptophan 2,3- dioxygenase or indolamine-2,3-dioxygenase (IDO). Tryptophan 2,3-dioxygenase is predominantly expressed in the liver while IDO occurs in many extrahepatic tissues, dendritic cells and macrophages. N-formylkynurenine is in turn converted to kynurenine and then to 3 -hydroxy kynurenine by the specific and constitutively active enzyme kynurenine 3-monoxygenase.

The present invention is principally concerned with MOF having a "sterile" aetiology. Sterile initiators of MOF (referred to hereinafter as "sterile events") may broadly be defined as non-microbial initiators and may include, for example, diseases and/or pathological conditions, trauma, injury, surgery and haemorrhagic shock.

Typically, MOF stems from an uncontrolled and inappropriate inflammatory response and may affect organs somewhat remote from the initiating event. As such, MOF may involve the lung, kidney, liver, cardiovascular, haematopoietic and other organ systems.

By way of example, injuries such as severe burns or diseases such as severe acute pancreatitis (AP) can result in MOF. Thus, in one embodiment, the present invention may be considered to relate to pancreatitis-associated MOF.

The components of the kynurenine pathway identified as involved in MOF and having cytotoxic activity may include an end product of the kynurenine pathway or kynurenine pathway intermediates.

As such, in order to serve as an effective treatment or prophylactic, the medicaments described herein may comprise one or more compounds capable of modulating the kynurenine pathway. Preferably, the medicament comprises one or more compounds capable of inhibiting or preventing the production of certain kynurenine pathway components.

Conveniently, compounds of the present invention may modulate an enzyme or enzymes of the kynurenine pathway. For example, a compound of the present invention may inhibit a particular enzyme involved in tryptophan catabolism. Enzyme inhibitors are well known to one of ordinary skill in the art. Enzyme inhibitors may be classified as, for example, specific, non-specific, reversible, nonreversible, competitive and non-competitive. In the present invention, any compound capable of inhibiting any enzyme or enzymes of the kynurenine pathway may potentially be used in the preparation of a medicament for the treatment and/or prevention of MOF.

By "inhibition" it is meant that, when compared to an enzyme which has not been contacted and/or exposed to a compound which might inhibit its activity, the rate at which an enzyme catalyses the production of a particular compound from a particular substrate is reduced.

Typically the compound capable of modulating the kynurenine pathway, may be a specific or non-specific enzyme inhibitor. It should be understood that the term "non-specific" refers to a compound, the effects of which are not restricted to a particular enzyme, or class of enzyme, but which generally affects the activity of substantially all enzymes. In contrast, the term "specific inhibitor" may be taken to refer to a compound which may inhibit a particular enzyme but which has no effect on the activity of another enzyme.

Typically, non-specific inhibitors may function as a result of the fact they denature by means of a change in temperature, pH or the like. However, non-specific inhibitors may also irreversibly bind to and block the active site of an enzyme thus preventing it from functioning to catalyse the production of a particular compound. Specific enzyme inhibitors may function by means of binding to a particular enzyme and causing either the active site to be blocked or the enzyme to undergo a conformational change such that it can no longer receive and catalyse the conversion of a particular substrate. The term "competitive inhibitors" may be taken to comprise compounds which are capable of occupying the active site of a particular enzyme resulting in the exclusion of the correct substrate. In this way the rate at which a particular enzyme is able to catalyse the production of a particular compound is reduced. Such compounds may closely resemble the particular substrate of the enzyme.

In the present case, compounds for use in modulating the kynurenine pathway, may also include competitive inhibitors of each of the enzymes involved in the kynurenine pathway. Generally, such compounds may structurally resemble the native (natural/correct) substrate of the enzyme. Alternatively, a competitive inhibitor may comprise a peptide which, for example, binds to the active site or substrate binding domain of an enzyme to block substrate binding.

Conveniently, said compounds may be small organic molecules capable of modulating the function of an enzyme or enzymes of the kynurenine pathway. Typically the small organic molecule may be an analogue of any of the substrates of the enzymes utilised in such pathways. Conveniently the substrate analogue may be modified in such a way so as to either inhibit and/or prevent the progression of a particular enzymatic reaction in a particular pathway.

Additionally, or alternatively, a compound for use in modulating the kynurenine pathway, for example a small organic enzyme substrate analogue, may interact with a particular enzyme or enzymes which in turn catalyse the production of a product which is incapable of being utilised by another enzyme of the kynurenine pathway. In this way further enzymatic reactions will not be possible and the pathway will be unable to progress.

The use of a modified substrate analogue may directly or indirectly prevent the production of a particular component of the kynurenine pathway. The term "non-competitive inhibitor" may be taken to include those compounds which affect the rate at which an enzyme catalyses the production of a particular compound. In contrast to the competitive inhibitors, non-competitive inhibitors may interact with an enzyme at a position other that the active site of an enzyme. For example a non-competitive inhibitor may bind to an enzyme and induce a conformational change in the enzyme such that the binding site of the enzyme may no longer receive a substrate.

A small organic molecule may modulate an enzyme of the kynurenine pathway and thus the production of one or more of the kynurenine pathway products. For example, by specifically inhibiting the activity of the enzymes Tryptophan 2,3- dioxygenase and/or indolamine-2,3-dioxygenase, it may be possible to reduce or eliminate the production of N-formylkynurenine. This would have the effect of eliminating or substantially reducing the amount of N-formylkynurenine available for conversion into kynurenine. This may, in turn, reduce the production of those components involved in MOF. Similarly, by inhibiting the activity of the enzyme kynurenine 3-monooxygenase, the amount of 3-hydroxykynurenine produced may be eliminated or significantly reduced.

Thus, in one embodiment, the present invention provides the use of an inhibitor of an enzyme involved in the kynurenine pathway for the preparation of a medicament for the treatment and/or prevention of MOF.

Suitable inhibitor compounds include 1-methyltryptophan (an IDO inhibitor) and 3,4-dichlorobenzoylalanine (PNU156561 : a K3M0 inhibitor). A more comprehensive list of kynurenine pathway modulators which may have therapeutic potential is provided below. Tryptophan 2,3-Dioxygenase inhibitors

• 3-(2-pyridylethenyl)indoles, in particular (E)-3-[2-(4'-pyridyl)-vinyl]- lH-indole, the corresponding 6-fluoro derivative and (E)-6-fluoro-3- [2-(3'-pyridyl)vinyl]-lH-indole. IDO inhibitors

• β-carboline compounds including norharman and 3-butyl-β-carboline

• Phytoalexins, such as brassilexin

• 1 -methyltryptophan

• 3-amino-2 -naphthoic acid • (-)-Epigallocatechin gallate kynurenine 3-monooxygenase inhibitors

• isoster of kynurenine, for example nicotinoylalanine

• m-nitrobenzoylalanine

• 3,4-dichlorobenzoylalanine (PNUl 56561) • (1S,2S) isomer of 2-(3,4-dichlorobenzoyl)cyclopropane-l-carboxylic acid (UPF 648)

As such, in a further embodiment, the present invention provides the use of one or more of the abovementioned kynurenine pathway modulators for the preparation of a medicament for the treatment and/or prevention of MOF. In addition to the above, compounds for use in modulating the kynurenine pathway may include oligonucleotide sequences. Such compounds may modulate the expression and/or activity of enzymes of the kynurenine pathway.

Conveniently, such oligonucleotide sequences may be complementary to the genes (or parts thereof) which encode the enzymes involved in the kynurenine pathway. Generally, the oligonucleotide sequences may be sequences of nucleic acid, either DNA or RNA which interfere with the functional sequences encoding enzymes of the kynurenine pathway. Cloning and characterization of the Tribolium castaneum eye-color genes encoding tryptophan oxygenase and kynurenine 3-monooxygenase Lorenzen MD et al. Genetics. 2002 Jan;160(l):225-34. Accordingly, the oligonucleotide sequences may be sections of messenger or ribosomal RNA (rRNA) which interfere with the RNA transcripts of the functional enzyme genes. Such oligonucleotides are referred to in the art as interfering RNA (iRNA), short interfering RNA (siRNA), short hairpin RNA (shRNA), or antisense oligonucleotides. Algorithms such as BIOPREDs/ may be used to computationally predict siRNA sequences that have an optimal knockdown effect for a given gene. One of skill in the art would be familiar with the use of the abovementioned techniques however the following references provide additional information: A Guide to Gene Silencing (ed. Hannon). Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, USA, 2003. In addition to the abovementioned components of the kynurenine pathway, a number of other compounds are involved in the development of MOF. Furthermore, the inventors have discovered that the components comprising certain biological fluids vary depending upon MOF status, and that by analysing these components, it is possible assess or determine the subject's MOF status. In this way, it may be possible to distinguish between patients who are unlikely to develop organ failure, those who are at the height of their illness, those who have been critically ill but are in the resolution phase and those who are not critically ill at the time of assessment but are likely to develop organ failure in the immediate future. Accordingly, the present invention also relates to methods for analysing the MOF status of a subject so to ensure that the most appropriate treatment is given without delay.

Thus, in a fourth aspect, there is provided a method for analysing the multiple organ failure status of a subject, said method comprising the steps of:

(a) providing a sample from a subject; and

(b) identifying a component profile of the sample; wherein the component profile of the sample is indicative of the MOF status of the subject. It is to be understood that the term "component" encompasses those molecules or compounds which comprise a biological fluid and may include, for example, proteins, peptides, amino acids, other small organic molecules and/or nucleic acids. Thus "identifying the component profile" may also be considered to mean assessing or analysing the constituents of the sample and encompasses identifying protein- expression profiles.

By way of example, one of skill in the art may analyse a sample for the presence, absence and/or level of various proteins. Thus, in one embodiment, there is provided a method for analysing the multiple-organ failure status of a subject, said method comprising the steps of: (a) providing a sample from a subject; and

(b) identifying a protein-expression profile of the sample; wherein the protein expression profile of the sample is indicative of the MOF status of the subject.

It is to be understood that the terms "biological fluid" and "sample" encompass fluids obtained from the human or animal body and include, for example, whole blood, plasma, serum, urine, lymph and saliva. Preferably the sample used is whole blood, plasma or lymph.

When identifying a component (or protein-expression) profile of a sample, the intention is to assess the level of component/protein expression (generally and/or specifically), the presence or absence of certain components/proteins and/or the presence or absence of any component/protein isomers or altered forms. A protein profile may also be referred to as a sample's "proteome".

One of skill in the art is aware of the techniques available to analyse and/or identify the component profile, and in particular the protein profile, of any of the abovementioned samples. Typically, techniques such as 2D gel electrophoresis and/or mass spectrometry may be used. Preferably, and in one embodiment, the method of determining the MOF status of a subject comprises analysing a sample, and in particular the protein content of a sample, by mass spectrometry. Fathman, C. G., Soares, L., Chan, S. M. and Utz, P. J.: 2005, 'An array of possibilities for the study of autoimmunity.' Nature 435, 605-11. Issaq, H. J.: 2003, 'Application of separation technologies to proteomics research.' Adv Protein Chem 65, 249-69. Issaq, H. J., Conrads, T. P., Prieto, D. A., Tirumalai, R. and Veenstra, T. D.: 2003, 'SELDI-TOF MS for diagnostic proteomics.' Anal Chem 75, 148A-155A.

More preferably, the sample may be analysed by matrix-assisted laser desorption/ionization (MALDI-TOF) or surface-enhanced laser desorption/ionization (SELDI-TOF) mass spectrometry.

Additionally or alternatively, the sample may be analysed by chromatography techniques such as, for example High Performance Liquid Chromatography (HPLC) or the like. When analysing a sample by mass spectrometry, the "read-out", for example the component or protein profile, is represented as a series of resolved peaks which reveal the level of expression and molecular weights of the various components of the sample (i.e., in the case of a protein profile, the level of expression and molecular weights of the proteins comprising the protein profile). One of skill in the art will appreciate that it is possible to adjust the detection parameters used such that only compounds falling within a particular range of molecular weights are included in the read-out. In the present case, it may be desirable to investigate the profile of those compounds, for example proteins, having a molecular weight of about 100 to 200000 Da, preferably 1000 to 180000, more preferably 2000 to 170000 Da and still more preferably, 3000 to 150000 Da. One of skill in the art will appreciate that by adjusting the abovementioned detection parameters to include more or less compounds in the profile, it may be possible to vary the profile generated. In one embodiment, prior to analysis by mass spectrometry, the compounds (for example proteins) comprising the sample may be subjected to gas chromatography-MS (GC-MS) to separate the compounds in the gas phase.

It should be understood that while mass spectrometry techniques allow the general analysis of the content of a particular sample, other techniques may be required to permit the identification of specific components. For example, in order to identify specific changes in the proteome of a sample, it may be necessary to use tandem MS techniques for example MS-MS, and/or separate/resolve the proteins by liquid chromatography or 2D gel electrophoresis, subject certain resolved proteins to a digestion protocol (for example tryptic digestion) and further MS analysis (for example electrospray MS analysis). One of skill in the art will appreciate that the specific proteins may be identified by subsequent database interrogation. Advantageously, in order to determine the MOF status of a subject, one of skill in the art may compare the component (for example, protein) profile of the sample, with the component profile of a reference sample. A "reference sample" may be considered as a sample derived from a subject of known MOF status - i.e. a reference subject. Additionally, or alternatively, the results obtained may be compared with a number of reference samples, each having been obtained from a subject of known MOF status. Preferably, the reference sample should be derived from the same biological fluid.

The MOF status of the reference subject may be determined by currently available clinical means. Most simply, organ failure may be inferred from a clinical need for organ support, for example renal replacement therapy (for example by haemodialysis) or ventilatory support for respiratory failure, or vasopressor support for failure to maintain peripheral vascular resistance. Certain combinatorial scoring systems, for example the multiple organ dysfunction score (Marshall, J. C, Cook, D. J., Christou, N. V., Bernard, G. R., Sprung, C. L. and Sibbald, W. J.: 1995, 'Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome.' Crit Care Med 23, 1638-52.) or the Acute Pyysiology and Chronic Health Evaluation score (Knaus, W. A., Draper, E. A., Wagner, D. P. and Zimmerman, J. E.: 1985, 'APACHE II: a severity of disease classification system.¹ Crit Care Med 13, 818-29.) By way of example, a sample to be representative of a subject in the resolution phase may be derived from a patient known to have recovered from MOF. A reference sample from a subject currently suffering from or experiencing MOF, may be derived from a subject clinically diagnosed as having MOF (as above). Resolution of organ failure might be regarded as recovery to the point where that particular organ system functions adequately without iatrogenic support. A sample to be representative of a subject who has not been exposed to an MOF initiating event may be obtained from a healthy individual. In order to obtain a sample, representative of a subject who, at the time the sample is taken, is not critically ill but may go on to develop MOF, it may be possible to take regular biological fluid samples, for example blood samples, until such time as MOF develops or begins to resolve. In this way, once MOF has been confirmed, it would be possible to select a sample representative of a subject about to develop MOF. Other methods for assessing MOF status and which are particularly useful for identifying those not critically ill but who may go on to develop MOF include, for example, clinical assessment by an experienced physician, serial markers of pulse, blood pressure, arterial blood gas analysis and urine output, serum biomarkers, for example procalcitonin and/or C-reactive protein.

Accordingly, in one embodiment, the present invention provides a method of determining the multiple-organ failure status of a subject, said method comprising the steps of: (a) providing a sample from a subject;

(b) identifying a component profile of the sample; and

(c) comparing the component profile of the sample with the component profile of a reference sample; wherein the component profile of the sample is indicative of the MOF status of the subject.

Preferably, the component profile is a protein-expression profile.

Typically, a number of reference samples - each derived from subjects of known MOF status, are analysed for their component or protein profile and subjected to a hierarchical clustering system. One of skill in the art will appreciate that the hierarchical clustering may be achieved by using an algorithm such as that described by Eisen et al. (29). By clustering the various reference samples, it is possible to associate particular component or protein profiles (or proteomes) with particular MOF statuses. For example, when the component or protein profile of a particular sample is analysed, it may be compared with the clustered reference data, and the subject can be appropriately identified as having a particular MOF status. One of skill in the art will also appreciate that component or protein profiles obtained from samples derived from different reference subjects will display considerable variation. One of skill in the art will appreciate that by increasing the number of reference samples and subjecting the reference data to a hierarchical clustering protocol, it may be possible to provide an accurate assessment of a subject's MOF status.

It is important to understand that when comparing the protein profile of a sample provided by a subject of unknown MOF status, the identity of the specific proteins present may not be not important. In order to assign an MOF status to that subject, it is simply necessary to compare the component or protein profile signature of the sample with one or more reference sample(s). Having previously allocated a particular MOF status to each reference sample(s) and by comparing the component or protein profile of the sample with those of the reference samples, it may be possible to assess the MOF status of the subject.

In one embodiment the method may be substantially automated such that a computer assesses the component or protein profile of the sample and automatically allocates the most appropriate MOF status. For example, the computer may be preloaded with the component or protein profiles of one or more reference samples. As stated, the reference samples may have been subjected to a hierarchical clustering protocol such that particular component or protein profiles are associated with particular MOF statuses. Upon being imputed with the component or protein profile of the sample, the computer may assign that sample a particular MOF status. The computer may simply compare the component or protein profile of the sample with the reference samples to find a close match, or, using a hierarchical clustering algorithm (such as that described above), allocate the component or protein profile of the sample to a particular cluster of reference samples.

As such, and in a fifth aspect, the present invention provides a data set, comprising representative component or protein profile data, wherein the data represents the component or protein profiles of biological fluids obtained from subjects know to have one of the aforementioned MOF statuses. Preferably, the component or protein profile data is generated by mass spectrometry and represents the levels and molecular weights of the components or proteins comprising the protein profile.

Advantageously, the data set may be subjected to a hierarchical clustering protocol such that particular component or protein profiles are associated with particular MOF statuses.

In a sixth aspect, there is provided a computer pre-loaded with the abovementioned data set.

As previously stated, a sterile MOF initiating event may lead to the generation of compounds which induce or exacerbate MOF. These compounds are carried in biological fluids such as blood and lymph to various organs and/or organ systems and may induce or exacerbate MOF. The compounds have an effect upon the cytotoxicity of the fluid and the cytotoxicity of the biological fluid is generally associated with the MOF status of the patient. Biological fluids obtained from subjects suffering from or experiencing MOF, may exhibit cytotoxic properties - i.e. when certain cells, for example endothelial cells such as Human umbilical vein endothelial cells (HUVEC) or epithelial cells such as HCTl 16 colonic epithelial cells, are brought into contact with the biological fluid, the cells die. The same fluid obtained from a patient who has not been exposed to a sterile MOF initiating event (i.e. a control patient) may not be cytotoxic (i.e. inert). Biological fluids obtained from subjects recovering from MOF may exhibit low/no cytotoxicity. Table 1 shows how the level of cytotoxicity of the sample may be equated to the MOF status of the subject.

Table 1 : Correlation between biological fluid cytotoxicity and MOF status. Cytotoxicity of the biological fluid MOF status Cytotoxic Critically ill

Post-cytotoxic Resolution phase

Pre-cytotoxic Likely to develop MOF

Inert Unlikely to develop MOF

It is to be understood that the term "cytotoxic" indicates that, when brought into contact with certain cells, for example endothelial and/or epithelial cells, the biological fluid causes the cells to die. An "inert" biological fluid is one which exhibits no cytotoxic effect. The term "post-cytotoxic" refers to biological fluids derived from subjects who are critically ill but are beginning to convalesce (i.e. in the resolution phase). Subjects having biological fluid identified as "pre-cytotoxic", although not critically ill at the time the sample is taken, are likely to develop MOF. By way of example and, in the case of a plasma sample, the method would require the analysis of the component or protein profile. Preferably, the component or protein profile of the plasma sample is analysed by SELDI-TOF mass spectrometry. The resulting profile may be compared with one or more reference plasma sample(s) derived from a subject or subjects of known MOF status. Preferably, the collection of reference samples will have been subjected to a hierarchical clustering protocol so as to associate a particular component or protein profile with particular MOF status. In this way, it is possible to assess a subject's MOF status and administer the most appropriate treatment as soon as possible.

By way of a further example, the method may comprise analysing a sample to identify a level of a kynurenine pathway component, for example kynurenine and/or 3-hydroxykynurenine. In addition to techniques such as mass spectrometry, the sample may be analysed by, for example, High Performance Liquid Chromatography (HPLC) or the like. The identified level of the kynurenine pathway component is indicative of a subject's MOF status and, in one embodiment, the identified level of a kynurenine pathway component or components may be compared or analysed together with one or more reference sample(s) or standards derived from a subject or subjects of known MOF status. A reference "standard" may be considered a particular level of a compound known to be associated with a particular disease, condition or subject status. In view of the above, the present inventors have now provided a method of determining whether or not a person who, although not critically ill at the time the sample is taken, is likely to develop MOF. As such, it is now possible to intervene before MOF occurs and offer an appropriate therapy.

While the above described methods require no specific knowledge of the components or proteins which comprise the profile, in addition to the kynurenine pathway components described above, the present inventors have identified certain other specific proteins involved in the development or exacerbation of MOF. These proteins are:

(i) transferrin (ii) haptoglobin; (iii) αl -protease inhibitor; and (iv) native apoAl.

More specifically, the inventors ascertained that, as cytotoxicity develops, the levels of haptoglobin and transferrin decrease, the level of αl -protease inhibitor increases and the native apoAl is altered to a number of distinct isoforms with variable pis.

Thus in a seventh aspect, the present invention provides a method for determining the MOF status of a subject, said method comprising the steps of:

(a) obtaining a sample from a subject; (b) identifying a level of transferrin, haptoglobin and αl -protease inhibitor and the isoforms of apoAl; and

(c) comparing the results obtained with those of a reference sample; wherein a decrease in the level of transferrin and haptoglobin, an increase in the level of αl -protease inhibitor and the presence of altered isoforms of apoAl, is indicative of MOF .

In addition, the method may include the further step of identifying a level of one or more of the kynurenine pathway components.

Altered isoforms of apoAl are identified as those having a pi value different to that attributed to native apoAl (pI5.5). Specifically, the altered isoforms may have pi values of about 3.2, 4.7 and 5.0.

While it is possible to use any of the abovementioned detection techniques to determine the level of kynurenine pathway components, transferrin, haptoglobin and αl -protease inhibitor, immunological techniques such as ELISA may also be used.

For example, an agent, for example an antibody, known to specifically bind to these proteins may be used to detect their presence in a sample derived from a subject. The identification of specific proteins which are involved in the development of cytotoxicity in biological fluids, presents a further opportunity for intervention with therapeutics aimed at treating or preventing MOF.

As such, in an eighth aspect, the present invention provides the use of one or more compounds capable of modulating the kynurenine pathway and/or the level of transferrin, haptoglobin and/or αl -protease inhibitor, for the treatment and/or prevention of MOF.

In a ninth aspect, there is provided a method of treating MOF, said method comprising the step of administering a therapeutically effective amount of one or more compounds capable of modulating the kynurenine pathway and/or the level of transferrin, haptoglobin and/or αl -protease inhibitor.

In a tenth aspect, there is provided the use of one or more compounds capable of modulating the kynurenine pathway and/or the level of transferrin, haptoglobin and/or αl -protease inhibitor, for treating MOF. In an eleventh aspect, there is provided the use of a compound capable of modulating the kynurenine pathway and/or compounds capable of modulating the level of transferrin, haptoglobin and/or αl -protease inhibitor for;

(i) treating or preventing MOF; or

(ii) for the preparation of a medicament for administration to a subject with MOF or a subject at risk of developing MOF; wherein the administrative pattern of the compound(s) or medicament comprises the steps of:

(a) providing a sample from a patient; and

(b) identifying a component profile of the sample; wherein the component profile of the sample is indicative of the MOF status of the subject and wherein subjects identified as having MOF or at risk of developing MOF are administered said compound(s) or medicament.

Preferably, the component profile is a protein-expression profile. The present invention will now be further described by way of example and with reference to the figures, which show:

Figure 1: Cytotoxicity of post-nodal mesenteric lymph. For (A), (B) and (C), the cell- free supernatant of mesenteric lymph collected from rats with experimental AP (A, n=6) and sham-operated controls (•, n=4) was assayed. (A) Viability of HUVECs

incubated with mesenteric lymph at 10% v/v, assessed by the MTT assay. (B) Neutrophil respiratory burst activation by mesenteric lymph at 5% v/v, measured by flow cytometry using dihydrorhodamine-123 fluorescence. (C) Erythrocyte deformability measured by laser ektacytometry following incubation of whole blood with 10% v/v mesenteric lymph (H=AP; D=sham; ND=not done). (D) Lung capillary

permeability to Evan's blue dye labelled-albumin after selective mesenteric lymph duct ligation (■), compared to sham ligation (D) (n=6 per group, mean ± 2xSEM).

Figure 2: AP lymph induces necrosis of HCTl 16 colonic epithelial cells. (A) Light microscopy (10Ox original magnification) of HCTl 16 cells incubated for 18 hours with 10% v/v FCS or sham-operated lymph or AP lymph (as marked) (B) Viability of cultures from Figure 2A measured by MTT assay. (C) Western blot of culture lysates from Fig. 2A probed with polyclonal antibody to PARP. Lanes: FCS; sham-operated lymph; AP lymph (as marked). Intact PARP migrates at an estimated MW of 116kDa. Cleaved PARP is seen at an estimated MW of 85kDa in lane 3. A prominent band of approximately 4OkDa is noted in AP lymph-treated cells, consistent with necrosis. (D) DNA fragmentation analysis. Lanes, as marked on figure: positive control using eHindIII restriction enzyme; FCS; sham-operated lymph; AP lymph. Data are representative of three separate experiments.

Figure 3: (A) Prediction of impending toxicity and confirmation of resolution by hierarchical clustering of SELDI-TOF mass spectra (anion exchange, pH 9) based on the Eisen algorithm (29). A binary value for each sample was allocated according to cytotoxicity to HUVECs. Lymph samples cluster into four groups: inert throughout (sham-operated animals); impending cytotoxicity (AP early in disease course); actual cytotoxicity (AP at height of disease); resolution to inertia (survivors of AP). Red (y) and blue (x) arrows indicate m/z species displayed in detail in Figure 3, B. (B) Representative sections of mass spectra used in the hierarchical clustering analysis in Figure 3A. Time course for individual rats undergoing either sham laparotomy (left) or experimental AP (right). (C) Scan of 2D gel electrophoresis of lymph collected from a single rat with AP at 1 hour (non-cytotoxic) and 6 hours (cytotoxic). Green circles: unaltered spots, for reference; Yellow circles: decreased expression at 6 hours. Red circles: increased expression at 6 hours. Peptide matches of tryptic digests of resolved spots using the MASCOT search engine are presented in Table 2. Figure 4: Established experimental AP is associated with CDl Ic⁺ cell migration from gut lamina propria to MLNs with nodal architectural disruption. (A) Gut, sham- operated; (B) MLN, sham-operated; (C) Gut, AP; (D) MLN, AP; (E) MLN, sham, showing normal architecture; (F) MLN in AP showing architectural disruption of MLN medullary cords with relative cortical preservation. Immunohistochemistry of gut and MLN sampled 18 hours after the induction of AP or sham operation was performed using anti-CDl lc mAb at 1:100, visualised with diaminobenzoate (brown) under light microscopy at 20Ox (A to D) or 4Ox magnification (E and F). Figure 5: Serum cytokine and chemokine profile in response to intra-arterial LPS administration in established experimental AP. (A) TNFα; (B) IFNγ; (C) IL- lβ; (D) IL-6; (E) CINC-2α; (F) IL-IO. A minimum of 6 rats was used for each combination of

experimental conditions. (H=AP; El =sham). Cytokines were measured by multiplex

ELISA.

Figure 6: Tryptophan catabolites mediate lymph cytotoxicity in experimental AP and plasma concentrations correlate with the extent of distant organ injury in human patients with AP. (A) Concentrations of kynurenine and (B) 3-hydroxykynurenine in mesenteric lymph samples from rats with AP (A, n=6) and sham-operated control rats (•, n=4) were

measured by HPLC. (C) Viability (MTT assay) of HCTl 16 cells incubated for 18 hours with combinations of tryptophan, kynurenine and 3-hydroxykynurenine at the concentrations shown. (D) Dose-response of 3-hydroxykynurenine cytotoxicity towards HCTl 16 cells in culture. For (C) and (D), bars represent the mean±2SEM of three experiments. (E and F) Kynurenine concentrations in peripheral blood plasma in a cohort of 34 patients with AP on admission (day 0), day 1 and day 7 after hospitalisation. 19 patients had a severe attack defined by the Atlanta criteria. 10 developed organ failure necessitating invasive organ support in the intensive care unit. Kynurenine concentrations are plotted against the APACHE II score (E) and multiple organ dysfunction score (F) of each patient at the time of blood sampling. (G and H) Peripheral plasma kynurenine concentrations plotted against the eventual requirement for haemodialysis (G) and intubation and mechanical ventilation (H).

Figure 7: proposed paradigm of the contribution of the kynurenine pathway to multiple-organ failure in severe acute pancreatitis. (1) An immunological "state of alarm" is triggered by tissue distraction in the pancreas. (2) Gut DC and macrophage activation occurs through danger signals, pattern recognition receptors and scavenger receptors, especially by self-derived molecules e.g. haptoglobin, oxLDL via SR-A and CD36 family. (3) Gut DCs migrate via afferent lymphatics to MLN. (4) In the MLN a contradictory combination of pro-inflammatory signals and lack of cognate microbial antigen, increased self or altered-self antigen and the presence of potentially autoreactive T cells leads to a necessary but potentially excessive suppression of immunity to self. Cytotoxic by-products of the induction of anergy or suppression, including cytotoxic kynurenine catabolites, overspill into the efferent mesenteric lymphatics (5) Cytotoxic mesenteric lymph enters the central venous circulation and arrives first at the lungs. Direct toxicity to endothelium with capillary leakage occurs. Erythrocyte injury affects the micro-circulation. Neutrophils are activated and effect further organ damage. Materials and methods. Reagents. All reagents were purchased from Sigma-Aldrich Company Ltd. (Gillingham, Dorset, UK) unless otherwise stated.

Animals. All experiments were performed with institutional ethical approval at each site (Department of Health, Social Services and Public Safety, Northern Ireland; New Jersey Medical School Institutional Animal Care and Use Committee; University of Helsinki). Specific pathogen-free adult male Sprague-Dawley rats weighing 250- 350g, (LSU, Northern Ireland; and Charles River Laboratories, Wilmington, MA), were multiply housed, acclimatised for >7 days with environmental enrichment, at 25°C with a 12 hour light/dark cycle. Rats had free access to water and standard diet (RM3, Special Diet Services, Witham, Essex; Teklad 22/5 Rodent Diet W-8640; Harlan Teklad, Madison, WI). Rats were not fasted preoperatively. Experimental AP. AP was induced in rats by the method of Schmidt et al (40). Briefly, under general anaesthesia (intraperitoneal ketamine 2.5 mg/kg and xylazine 1.2 mg/kg, post-operative buprenorphine 20 μg/kg i.v. 6 hourly), carotid artery and jugular vein catheters were placed, and at laparotomy a 10 minute, pressure-controlled retrograde biliopancreatic duct infusion (10 mM glycodeoxycholic acid, 500 mM glycyl-glycine, 4 mM CaCl₂, pH 8.0) was given at 8.4 mL/kg/hour. I.v. caerulein in 150 mM NaCl was given at 5 μg/kg/hour for 6 hours. Animals were resuscitated with i.v. 150 mM NaCl at 4 mL/kg/hr for 12 hours. Sham-operated rats underwent identical anaesthesia, surgery and resuscitation without glycodeoxycholic acid infusion and replacement of caerulein with 150 mM NaCl. AP with organ dysfunction was confirmed by sickness behaviour, pancreatic histology, elevated serum amylase >1000 IU, elevated haematocrit and metabolic acidosis on serial arterial blood gas analysis, as previously described (41). Mesenteric lymph duct ligation and lymph collection. Mesenteric lymph duct ligation and lymph collection was performed as previously described (10). Briefly, the mesenteric lymph duct was either divided between ligatures, or cannulated with heparin-flushed polypropylene tubing passing externally through the scruff. Continuously collected lymph was aliquoted hourly into sterile, heparin-rinsed polypropylene tubes onto ice. The hourly lymph volume was replaced with 150 mM NaCl i.v.. Lymph was centrifuged immediately (60Og, 4°C, 3 minutes) and the supernatant stored at -80°C for <12 months.

Intra-arterial LPS administration. In the relevant groups, E. coli 0111:B4 LPS at zero, 6 or 30 μg/kg in 150 mM NaCl was delivered intra-arterially 18 hours after surgery. LPS activity was confirmed by the limulus amoebocyte lysate assay (Pyrochrome® LAL Assay; Quadratech, Epsom, Surrey, UK) with a 6-point, duplicate standard curve (zero to 2.0 endotoxin units/mL).

Lung permeability to Evan's blue dye. 1 mL of a 1 :1 mixture of 4% BSA in PBS, and Evan's blue dye (20 mg/mL) was injected intravenously. After 5 minutes, 1 mL of arterial blood was drawn and the plasma separated (60Og, 10 minutes, 4⁰C). Under anaesthesia, bronchoalveolar lavage was performed using PBS, three times. Lavage fluid was centrifuged (60Og, 10 minutes, 4°C) and the concentration of Evan's blue dye measured by spectrophotometric absorbance at 620 nm and the percentage of dye relative to that in arterial plasma was calculated. Cytotoxicity assays. HUVECs (Clonetics, San Diego, USA), or HCTl 16 cells (European Collection of Animal Cell Cultures, Salisbury, Wilts., UK) were seeded at 2 x 10⁴ cells per well and grown for 24 hours in DMEM (95% air, 5% CO₂, 100% humidity). Lymph or FCS was added at 10% v/v, or an equivalent volume of DMEM was used as a negative control. After 18 hours, cell viability was measured using the TOXl (MTT) assay kit, quantified by spectrophotometric absorbance at 570 nm (T- Max, Molecular Devices, Wokingham, Berks., UK).

Neutrophil respiratory burst activation. Neutrophil respiratory burst activation was measured as previously described (9). Briefly, rat neutrophils were prepared from whole blood by erythrocyte lysis in 0.84 mM NH₄Cl and incubated (1 hour) with 5% v/v lymph (or HBSS as a negative control) and dihydrorhodamine-123 (Molecular Probes, Oregon, USA), prior to stimulation with 270 ng/mL PMA. The respiratory oxidative burst was quantified using a FACSscan flow cytometer and Cellquest software (Beckton Dickinson, San Jose, CA, USA). Daily calibration was performed using Calibrate beads (Pharmingen, USA). Neutrophils were identified using forward and side scatter analysis. Each sample was run in the acquisition mode until a granulocyte activation gate was established, when only gated events were acquired and viewed in the FLl histogram without marker settings. 5000 events were collected in all experiments and expressed as mean fluorescence intensity. Erythrocyte deformability. Lymph (10% v/v) was incubated at 37⁰C for 4 hours with fresh whole rat blood. 15 μL of this mixture (containing approximately 10⁷ erythrocytes) was suspended in 3 mL of 5% (0.14mM) polyvinylpyrridoline in PBS, isosmolar, with a final viscosity of 31 Pa. Ektacytometry was performed on a laser- assisted optical rotational cytometer (RR Mechatronics, Hoorn, Netherlands). The elongation index, E.I., was calculated as (x - y)(x + y)^'1 , where x and y represent the maximum longitudinal and transverse erythrocyte diameters.

Bacterial culture, pH, endotoxin assay, heat treatment and protease assay of lymph. Lymph aliquots were incubated on blood agar for 18 hours at 37°C prior to examination by eye. Endotoxin was assayed using the Limulus amoebocyte lysate assay (Pyrochrome® LAL Assay; Quadratech, Epsom, Surrey, UK). Heating lymph for 10 minutes in a block heater at 90°C prior to MTT assay was performed to evaluate the effect on cytotoxicity. Lymph pH was measured on solid state i-STAT® ^CG8⁺ cartridges (i-STAT Corporation, East Windsor, NJ, USA). TNF-α was measured by ELISA (Quantikine™, R&D Systems, Minneapolis, MN, USA). Protease activity was measured using the EnzCheck™ Protease Activity kit, (E-6638, Molecular Probes, Eugene, OR, USA).

Western blotting. Cells were washed with PBS and lysed directly into SDS-PAGE loading buffer. Soluble protein (30μg) was resolved by SDS-PAGE and transferred to nitrocellulose membrane. For all experiments, equal protein loading was confirmed by staining with Ponceau S. Membranes were probed with PARP antibody #9542 at 1 : 10000 (New England Biolabs, Hitchin, Herts., UK). Reactions were visualised with a secondary HRP-conjugated antibody (Dako, Ely, UK) and Luminol™-enhanced chemiluminescence (Santa Cruz, California, US).

DNA fragmentation analysis. Cells were washed with PBS, scraped into lysis buffer

(5OmM Tris-HCl, pH 8.0, 1OmM EDTA, 0.5% SDS, 0.5 mg/ml proteinase K) and incubated at 55⁰C overnight. Cell debris was pelleted (13,000g, 10 minutes) and the supernatants were incubated with RNAse A (0.1 mg/ml) at 37°C for 30 minutes.

Nucleic acids were precipitated with 0.1 volume 3 M sodium acetate and 3 volumes cold ethanol followed by centrifugation (13,000g, 15 minutes). Samples were resuspended in 1OmM Tris-HCl, pH 7.5, ImM EDTA, and visualised on a 1.5% agarose gel containing ethidium bromide.

SELDI-TOF MS. No specific sample preparation was performed for lymph. ProteinChips™ (Ciphergen, Fremont, CA, USA) employing anion exchange (QlO) with elution at pH 9 and pH 6, weak cation exchange (CMlO) with elution at pH 7 and pH 4, immobilised metal affinity capture (IMAC30) using Cu²⁺ and Ca²⁺ bait, and a hydrophobic (H50) surface, were screened. QlO and IMAC30 Cu²⁺ were used for comprehensive analysis. Arrays were equilibrated with 150μL binding buffer (5OmM Tris HCl pH 9 for QlO; 10OmM sodium phosphate, 50OmM sodium chloride, pH 7 for IMAC Cu²⁺) for 5 minutes, twice, with shaking. Samples were diluted 10-fold in binding buffer and incubated on the array for 30 minutes prior to washing off non- interacting proteins with binding buffer. Energy-absorbing matrix (sinapinic acid 100 μg in 200μL acetonitrile, 200μL 1% trifluoroacetic acid) was applied twice. Chips were read on a Ciphergen™ PBSII SELDI-TOF mass spectrometer, using ProteinChip™ Reader and Ciphergen Express software. The spot protocol was: high mass 100,000 Da, optimised from 2,000 to 100,000 Da; laser intensity 220, adjusted between 180 and 240 as necessary; focus by optimisation centre; SELDI acquisition set from position 20 to position 80, delta 2, transients 5; with 2 warming shots not included in the analysis. At each profiling session, calibration was performed with protein standards: hirudin 7.0 kDa, cytochrome c 12.2 kDa, myoglobin 17.0 kDa, carbonic anhydrase 29.0 kDa, yeast enolase 46.7 kDa, albumin 66.4 kDa, and bovine IgG 147.3 kDa (Ciphergen, Fremont, CA, USA).

Two-dimensional gel electrophoresis and protein identification. 20 μL of lymph was incubated at 20⁰C for 1 hour in 350 μL lysis solution (8M urea, 2M thiourea, 4% CHAPS, 4OmM Tris, 10 mg/mL dithiothreitol, 0.027% IPG Buffer pH 3-10 (GE Healthcare Ltd., Little Chalfont, Bucks., UK)). After centrifugation (600Og, 2 minutes), isoelectric focusing of the lysed sample was performed using an Immobiline DryStrip pH 3-10 (GE Healthcare Ltd., Little Chalfont, Bucks., UK), at settings: phase 1: 300V, 2mA, 5 W, 0.01 hour; phase 2: increasing to 350OV, 2mA, 5W, over 1.5 hour; phase 3: 3,500V, 2mA, 5W, 26.5 hours. The second dimension was run by equilibrating strips for 15 min at 20⁰C in 50 mM Tris pH 8.8, 6M Urea, 30% glycerol, 2% SDS, 0.002% bromophenol blue, 10 mg/mL dithiothreitol, followed by 15 min with dithiothreitol replaced by 25 mg/mL iodoacetamide. Using a vertical pre- prepared 12.5% acrylamide gel, second dimension electrophoresis was performed at 30 mA for 4 hours at 2O⁰C. Gels were fixed and stained with colloidal Coomassie Blue for 4 hours prior to destaining in 25% ethanol and scanning. Sequencing grade trypsin (Promega, Inc., Southampton, UK) with overnight incubation at 3O⁰C was used for in-gel digestion of spots visually selected on the basis of differential expression, followed by peptide extraction (5% v/v trifluoroacetic acid, 50% v/v acetonitrile 3 x 6 min with sonication at 2O⁰C), evaporation, rehydration, and desalting using Cl 8 reversed phase resin (ZipTip, Millipore (UK) Ltd., Watford, Herts., UK). Extracted peptides were analysed by electrospray tandem mass spectrometry using a ThermoQuest LCQ Deca mass spectrometer, in nanospray mode, and MS/MS data collected using the Xcalibur software suite (Thermo Fisher Scientific Inc., Waltham, MA, USA). MS/MS ion search was performed using the MASCOT search service (Matrix Science, www.matrixscience.com). Immunohistochemistry. 5μm sections of formalin-fixed paraffin-embedded sections were dewaxed in xylene and subject to antigen-retrieval in 20 raM Tris, 20 mM EDTA, 10 mM sodium citrate, pH 7.8 for 2 minutes in a domestic pressure cooker, followed by recovery in PBS. Sections were stained on a Ventana NexES autostainer (Ventana Inc., Tucson, AZ, USA) using anti-CDl lc mAb (NCL-L-CDl lc-563, Novocastra, Newcastle-upon-Tyne, UK) at 1:100 dilution, visualised with diaminobenzoate conjugate (DAKO Ltd, Ely, UK), counterstained with hematoxylin. Searchlight™ multiplex ELISA. Serum samples were outsourced to Pierce Biotechnology, Inc., Rockford, IL, USA for Searchlight™ Rat Cytokine/Chemokine Array multiplex analysis by ELISA using streptavidin/biotin-HRP conjugates visulaised with Supersignal™ Luminol Enhancer. 10 μL of serum, at 1 :5 dilution, was analysed by technicians blind to the sampling conditions. Luminescent images were digitally recorded within 10 minutes and analysed using Array Vision™ software (Imaging Research, Inc., GE Healthcare BioSciences Corp., Piscataway, NJ., USA). Kynurenine assay. This was a development of the HPLC method proposed by Yong and Lau (42). 20 μL aliquots of lymph or serum were added to 50 μL 1OmM NaPO₄, pH 3.0. Proteins were removed by adding trichloroacetic acid (10% v/v) at O⁰C followed by centrifugation (600Og, 2 minutes). 50 μL supernatant aliquots were separated on a Cl 8 reverse phase column (Hi-Pore RP-318 4.6 x 250 mm, BioRad, Richmond, CA, USA) using a linear 0-10% methanol gradient in 1OmM phosphate buffer pH 3.0 (Bo-₂=0%, B_i2=100%, Bi_2-i_{2 5}=100%, B_14-22=0% (subscripts are time in minutes, percentages are concentrations of 10% methanol in 1OmM phosphate buffer, pH 3.0; flow rate: 1 mL/min). An SCL6B machine controlled two LC 6A pumps and an SPD6A detector recording absorbance at 254 nm (Shimadzu UK Ltd., Milton Keynes, Northants., UK). Kynurenine and 3-hydroxykynurenine standards (0-50 μM in 10% TCA in 1OmM phosphate buffer, pH3.0) were used to allow the estimation of eluted peak areas from lymph/serum samples by CR-4AX Chromatopac software after baseline correction. The identity of selected HPLC peaks was confirmed by electrospray-MS. Human patient plasma samples. With institutional ethical approval and fully informed consent, peripheral blood was sampled from 34 patients with a diagnosis of AP based on a history of epigastric pain, vomiting and an elevated serum amylase of greater than 3 times normal. Patients were included if they had presented to hospital within 72 hours of the onset of symptoms. Peripheral venous blood was sampled on admission (day 0), 24 hours later (day 1) and 1 week later (day 7) and plasma stored at -8O⁰C for <12 months. The APACHE II score and the MODS score at the time of venepuncture were recorded (43, 44). Organ failure requiring mechanical ventilation or haemodialysis was noted. Severity of AP was defined according to the Atlanta criteria (45). Statistical analysis Experimental group sizes were prospectively calculated to detect a difference in dependent variables of 2 points on the logarithmic scale with statistical significance, α, set at 0.05, where Z_α/2=1.96 and a power, l-β=0.80. Continuous variables were tested for normality using the Kolmogorov-Smirnov test. Data which did not conform to the normal distribution were analysed by non-parametric multiple group comparison by the Kruskal-Wallis test. Data which conformed to the normal distribution were subjected to ANOVA with post-hoc Dunnett's T3 test (equal variances not assumed). A modification of the Eisen algorithm (29) for analysis of large data sets (embedded in Ciphergen Biomarker Patterns™ software (Ciphergen Inc., Fremont, CA, USA)) was used for analysis of SELDI-TOF mass spectra. Results

The humoral component of mesenteric lymph in rats with AP contains factor(s) which cause extrapancreatic organ injury. In light of the central role that gut injury and inflammation might play in the development of MOF in AP (5, 26), and that mesenteric lymph contains cytotoxic factor(s) during other sterile initiators of MOF (11), we explored whether the humoral component of mesenteric lymph caused extrapancreatic injury during experimental AP. We collected mesenteric lymph continuously for 22 hours through an indwelling selective mesenteric lymph duct cannula from six rats with experimental AP, using mesenteric lymph from four sham- operated rats as corresponding negative controls. The cell-free supernatant of mesenteric lymph became cytotoxic to HUVECs (quantitated by the methylthiazoletetrazolium (MTT) cell viability assay) from 3 hours after the induction of AP and reverted back to an inert state by 21 hours in survivors. Lymph from sham- operated rats remained inert throughout (Figure 1, A) (AP vs sham, P<0.001 by oneway ANOVA, post-hoc Dunnett's T3). Given that neutrophils are probably the main effectors of lung injury during acute inflammation (9), we tested whether AP lymph activated neutrophils. We showed that lymph from rats with AP, but not that from sham-operated controls, activated the respiratory burst when freshly isolated rat neutrophils were incubated with lymph at 5% v/v (AP vs sham, P<0.001 by one-way ANOVA, post-hoc Dunnett's T3), (Figure 1, B). As critical illness results in the emergence of a population of erythrocytes with decreased deformability (27), we determined whether AP lymph could damage normal erythrocytes. Incubation for 4 hours of AP lymph at 10% v/v with fresh whole rat blood decreased erythrocyte deformability under controlled conditions of shear stress measured by laser ektacytometry. Mesenteric lymph from sham-operated rats was inert throughout (Figure 1, C) (AP vs sham, P<0.001 by one-way ANOVA, post- hoc Dunnett's T3).

To explore whether mesenteric lymph caused lung injury in a whole animal model of AP, we interrupted the flow of mesenteric lymph into the circulation by selective mesenteric lymph duct ligation. We observed that duct ligation prevented lung injury in AP, measured by capillary permeability to Evan's blue dye-labelled albumin (Figure 1, D) (AP no MDL vs AP + MDL and sham ± MDL, P<0.001 by one-way ANOVA, post-hoc Dunnett's T3). Taken together, these results suggest that the humoral component of mesenteric lymph contains one or more factors which cause distant organ injury in AP.

AP lymph induces necrosis rather than apoptosis of HCTl 16 gut epithelial cells. In light of the importance of gut injury during AP (5, 26), and having observed AP lymph cytotoxicity to lung and endothelial cells in this study, we explored whether AP lymph damaged gut epithelial cells. Lymph aliquots identified as cytotoxic to HUVECs caused death of HCTl 16 colonic epithelial cells in vitro (Figure 2, A and B). We showed that there was minimal 85kDa cleavage product of poly(ADP-ribose) polymerase (PARP), the final common pathway of caspase- mediated apoptosis, and observed the presence of a smaller band (approximately 4OkDa) recognised by the antibody (Figure 2, C). This appearance has previously been shown by others to be consistent with necrosis (28). In addition, DNA fragmentation analysis failed to show any DNA laddering in AP lymph-treated cells (Figure 2, D), suggesting non-apoptotic cell death.

Hierarchical clustering of SELDI-TOF mass spectra predicts the onset and resolution of mesenteric lymph cytotoxicity. To determine the extent of alterations to the lymph humoral proteome during the development of cytotoxicity, we used high-throughput SELDI-TOF MS. We screened surface conditions of anion exchange (QlO) at elution pH 9 and pH 6, weak cation exchange (CMlO) at pH 7 and pH 4, immobilised metal affinity capture (IMAC30) using Cu²⁺ and Ca²⁺ bait, and a hydrophobic (H50) surface. Optimum MS resolution was achieved using QlO chromatography washed at pH 9, and IMAC30 Cu²⁺ (data not shown). We used these surface conditions, followed by hierarchical clustering of resolved peaks using an algorithm based on that described by Eisen et al. (29)), with binary grouping according to cytotoxicity to HUVECs to analyse all lymph samples (Figure 3, A). We observed separate clustering of 1) cytotoxic lymph from rats with AP, 2) inert lymph from controls, 3) inert lymph from survivor AP rats (which had previously generated cytotoxic lymph), and 4) non-cytotoxic lymph from rats with AP which were to produce cytotoxic lymph at later time points (Figure 3, A). 110 peaks were significantly altered between cytotoxic and inert lymph samples (P<0.05 by cluster analysis). Representative time course spectra from a rat with AP and a sham operated control are presented in Figure 3, B, to illustrate the disappearance over time of a peak with approximate m/z ratio 8739 Da and appearance of a peak with approximate m/z ratio 8670 Da (marked in Figure 3, A, by red and blue arrows respectively), flanked by two peaks which remain constant, acting as internal controls.

In order to identify specific changes in the proteome as lymph became cytotoxic, we used two-dimensional gel electrophoresis followed by tryptic digests of resolved spots, electrospray MS analysis and database searching using the Matrix Science MASCOT search engine (Figure 3, C). We observed that as lymph became cytotoxic, transferrin and haptoglobin were decreased while αl -protease inhibitor was increased and native apoAl (pi 5.5) was altered to three distinct iso forms with more acidic pis of 3.2, 4.7 and 5.0 (Table 2).

Table 2. Proteins expressed in significantly altered concentrations in cytotoxic AP lymph. Cytotoxic and non-cytotoxic lymph samples from the same rat were subjected to 2D-P AGE. Resolved spots were excised, digested with trypsin and subjected to electrospray MS. The MASCOT™ service was used to identify the most likely candidates for the individual proteins. Mowse values above 50 are taken as confirmation of identity.

Protein identified Relative change in AP Mowse score

Transferrin - 240

Haptoglobin - 74 αl -protease inhibitor + 60 apoAl, pI 5.5 - 159 apoAl, pi 3.2 + 69 apoAl, pI 4.7 + 147 apoAl, pI 5.0 + 160

CDlIc⁺ DCs migrate from the gut to draining mesenteric lymph nodes during experimental AP with disruption of nodal architecture. Given that MLNs are key sites of immunoregulation, and that increased DC migration from the gut lamina propria to draining MLNs occurs during systemic inflammation (13, 30), we asked whether DC migration occurred synchronously with the development of AP lymph cytotoxicity. In sham-operated controls, CDl Ic⁺ DCs were frequent in ileum lamina propria (Figure 4, A) and scarce in the corresponding MLN (Figure 4, B). Experimental AP (18 hours after induction) resulted in fewer CDl Ic⁺ cells in the lamina propria (Figure 4, C), and increased CDl Ic⁺ cells in the corresponding MLN (Figure 4, D). In AP, but not in controls, disruption of MLN architecture was evident in the medullary and paracortical areas of MLN with cortical sparing (Figure 4, E and F). These observations suggest that gut DC migration to MLN occurs during AP, and implicates the MLN as the site of generation of the cytotoxic factor(s) found in post-nodal mesenteric lymph. Experimental AP causes increased production of IL-10 in response to intraarterial LPS. Severe AP is characterised by destruction of pancreatic tissue, resulting in large amounts of self or altered self antigen coincident with abundant danger signals (3, 16). Given that mucosal DCs maintain peripheral tolerance partly through the action of anti-inflammatory cytokines (15), we measured the cytokine response to an intra-arterial injection of LPS in established AP eighteen hours after the induction of AP or sham operation. The results, presented in Figure 5, show that, in contrast to the dose- and time- dependent increase in serum cytokine concentrations seen in sham-operated controls, rats with established AP showed markedly diminished production of TNFα, IL- lβ, IFNγ, and IL-6, unchanged production of CINC2α (although a trend towards increased CINC2α production was observed), and increased serum IL-10 production in response to LPS, most marked at an LPS dose of 6 μg/kg (PO.002, one-way ANOVA, post-hoc Dunnetts T3, AP vs sham for each set of conditions). This pattern of cytokine production fits with a regulatory or suppressive expressed phenotype. Kynurenine pathway intermediates contribute to MOF in severe AP in rats and humans. Having found that cytotoxicity of mesenteric lymph in AP resides in the soluble cell-free compartment, and that the cytotoxic factor(s) damage several cell types, we wished identify the effector agent(s). First, we established that endotoxin, bacterial contamination, pH, total protease activity and TNFα content could not account for the cytotoxicity of mesenteric lymph. This conclusion follows the observation that some inert lymph samples contained elevated levels of endotoxin, TNFα or protease activity, while some cytotoxic samples showed no evidence of endotoxin, TNFα or protease activity. No viable bacteria were detected in any sample. Heating of mesenteric lymph from AP rats for 10 minutes at 90⁰C did not abrogate cytotoxicity.

Having observed in this study that AP induces DC migration and an antiinflammatory cytokine response to LPS, coupled with the work of others showing that DCs regulate unwanted immune responses through kynurenine pathway intermediates (19-22), we measured kynurenine and 3-hydroxykynurenine concentrations in mesenteric lymph using HPLC. Increased concentrations of kynurenine and 3- hydroxykynurenine were observed in the same time-frame as developing and resolving lymph cytotoxicity (Figure 6, A and B). To confirm whether either substance could cause cellular injury, we added kynurenine and 3- hydroxykynureneine to HCTl 16 cells in culture. 3-hydroxykynurenine, but not kynurenine, caused near total cell death (Figure 6, C) in a dose-dependent manner (Figure 6, D).

As kynurenine pathway intermediates were present in cytotoxic mesenteric lymph, and caused cellular injury in vitro, we measured the concentrations of these mediators in the peripheral blood plasma of 34 human patients with AP on admission to hospital, at 24 hours and 7 days after admission, using HPLC and with institutional ethical approval. 19 patients had severe AP defined by the Atlanta criteria, and 10 developed organ failure requiring mechanical ventilation or haemodialysis in the intensive care unit. Plasma kynurenine concentrations were elevated in patients with APACHE II scores >4 at the time of venepuncture, rising proportionately with deteriorating MOF (Spearman's rank correlation, R sq. 0.511, P<0.001) (Figure 6, E). Similarly, plasma kynurenine concentration was elevated in patients with a MODS score >3 at the time of venepuncture, with a proportional increase with worsening organ dysfunction (Spearmans's rank correlation, R sq. 0.523, P<0.001) (Figure 6, F). In patients who required mechanical ventilation or haemodialysis, the plasma concentration of kynurenine was higher at all time points, including on admission (P<0.0001, Kruskal-Wallis test), (Figure 6, G and H). These data, from the rat model and corroborated in human plasma, suggest that kynurenine pathway intermediates contribute to MOF in severe AP. Figure 8 is a confocal microscope image of mesenteric lymph node from a rat with acute pancreatitis. It shows that the first (and rate-limiting) enzyme of the kynurenine pathway (indoleamine-2,3,dioxygenase, IDO) is present and expected to be upregulated in CDl Ic cells in the time frame of acute pancreatitis. This supports the hypothesis of enzymes in the kynurenine pathway being targets for intervention in acute organ failure. Discussion

MOF is the key determinant of mortality in patients with conditions such as severe AP. In one study, those patients in whom organ failure persisted beyond 48 hours experienced a mortality of 37%, compared to those with transient organ failure or no organ failure, in whom mortality was less than 3% (2). In a population-based study of death from AP, one third of all deaths from acute pancreatitis occurred before the patient reached hospital, and, where known, the median duration of symptoms was less than 24 hours (31 ). MOF in acute pancreatitis is therefore the main determinant of outcome and an early event. The observation that cytotoxicity of the humoral component of mesenteric lymph developed at three hours after the induction of experimental AP in rats, and that in survivors, lymph reverted back to an inert state by 22 hours after the induction of AP is consistent with the timing of development of MOF. Using selective mesenteric lymph duct ligation, we also showed that mesenteric lymph was capable of causing lung injury within this time period. Gut-associated lymphoid tissue (GALT) and the mesenteric lymphatics are physiologically and functionally distinct from other secondary lymphoid tissues. The evolutionary advantage of the development of GALT is to provide a framework in which adaptive immune cells (mainly T and B cells) may co-localise with antigen- presenting cells, mostly DCs. Afferent lymphatics deliver antigen and facilitate the migration of DCs to MLNs, draining into the marginal sinus immediately below the LN capsule. We demonstrated that experimental AP was associated with increased numbers of CDl Ic⁺ cells in MLNs and decreased numbers of CDl Ic⁺ cells in gut lamina propria, suggesting DC migration in AP.

The MLN is separated into an outer cortex and an inner paracortex, the functional unit of which, and likely site of suppression or elimination of autoreactive T cell clones, is the paracortical cord, regarded as the space in which DCs encounter and signal the activation and proliferation of rare antigen-specific circulating T cells (for a review see (12)). In this study, architectural disruption of MLN in AP was observed towards the efferent regions within MLNs, at the putative sites of DC-T cell interaction, with relative cortical sparing. Post-nodal mesenteric lymphatics coalesce in a racemose network in the root of the midgut mesentery before passing cephalad to join the central venous system at the origin of the left brachiocephalic vein. The first capillary bed which returning lymph encounters is therefore in the lung, notably the most commonly injured extrapancreatic organ during AP. In one study, of 40 patients who died from multiple organ failure, 35 (88%) had pulmonary injury, compared to 14 (35%) with renal dysfunction (32). We have shown that experimental AP causes lung injury, manifested by increased capillary permeability to Evan's blue dye-labelled albumin, and that this may be prevented by interrupting the flow of lymph into the circulation. Lung injury in AP is mediated at least in part by neutrophils (9), and the humoral component of cytotoxic AP lymph activated rat neutrophils. Although we did not establish whether lymph-induced lung injury occurs directly, or through the action of neutrophils, we have shown that there is potential for lymph to act through either or both mechanisms. Gut epithelial cell injury is a feature of AP, and the cytotoxic activity of AP lymph to HCTl 16 cells suggests that recirculating mesenteric lymph may propagate gut injury. The effect of cytotoxic lymph on erythrocyte deformability implies that there is a non-specific damaging effect of lymph which does not require de novo gene transcription. Reduced erythrocyte deformability with evidence of oxidative stress has been shown previously in AP (33, 34). 3-hydroxykynurenine, as a candidate cytotoxic factor proposed in this study, causes cell injury by oxidative damage (24).

The presence of altered apoAl in cytotoxic mesenteric lymph suggests that native LDL is altered during experimental AP. Altered LDL, especially oxidised and acetylated LDL, acts as a powerful activator of innate immune responses, mediated through scavenger receptor A (SR-A) and the CD36 family of SRs, SR-BI and SRBII (35). Unlike the LDL receptor, which internalises intact lipoprotein particles, SR-BI facilitates the selective transfer of cholesteryl esters from the hydrophobic core, but not the apolipoprotein at the surface, which is rejected back into the circulation (36). This mechanism potentially explains the altered apoAl isoforms present in cytotoxic lymph. Cytotoxic lymph was also depleted of haptoglobin. Haptoglobin scavenges free hemoglobin, preventing free hemoglobin from causing oxidative damage to tissues, and the haptoglobin-hemoglobin complex, bound by the scavenger receptor CD 163 on macrophages, is a further powerful initiator of innate immune responses and potential stimulant of DC maturation (35, 37). DC-mediated immunoregulation depends on the microenvironment within the

MLN, in particular the relative concentrations of cytokines, especially IL-IO. High concentrations of IL-IO signal immature monocyte-derived DCs to induce a CD25⁺CD4⁺ regulatory T cell phenotype, characterised by the secretion of further IL- 10 and capable of limiting the activation of other CD4⁺ and CD8⁺ T cells by mature DCs (38). The diminished production of TNFα, IFNγ, IL- lβ and IL-6, the unchanged CINC2α chemokine response, enhanced IL-10 production and absent IL-4 seen in established experimental AP in response to intra-arterial LPS in this study suggest that DCs orchestrate a suppressive or regulatory phenotype in AP. Munn et al. have shown that an important dimension of DC control of T-cell polarisation is through the upregulation of IDO and generation of downstream kynurenine pathway metabolites (21). Mesenteric lymph cytotoxicity correlated with increased kynurenine and 3- hydroxykynurenine concentrations and cytotoxicity of lymph to HCTl 16 cells in vitro was reproduced by exogenous 3-hydroxykynurenine, but not kynurenine. These data suggest that 3-hydroxykynurenine contributes to mesenteric lymph cytotoxicity during AP in rats. In human patients with pancreatitis-associated MOF, we observed that organ dysfunction was associated in real-time with increased kynurenine concentrations in peripheral plasma and elevated kynurenine concentrations were observed in patients who required mechanical ventilation or haemodialysis in the intensive care unit. We have noted that peripheral plasma kynurenine concentrations in humans were lower than those observed in mesenteric lymph in rats, and it is possible that this observation reflects a dilutional effect of distributing lymph-borne mediators throughout the circulating volume. Importantly, the concentration of cytotoxic mediators carried in lymph will be much higher in blood entering the pulmonary circulation than in blood which has passed through the capillary beds of the lungs and peripheries. The low concentration in the peripheral circulation prevented adequate quantitation of 3-hydroxykynurenine in humans, as the elution time was too close to that of the trichloroacetic acid used during sample preparation. However, kynurenine 3-monooxygenase, which catalyses the formation of 3-hydroxykynurenine from kynurenine, is constitutively active and others have shown that introduction of kynurenine directly as a substrate leads to the formation of 3-hydroxykynurenine and other immunoregulatory metabolites in the absence of signals required for the upregulation of IDO (24). Therefore, the measurement of kynurenine in peripheral plasma gives a qualitative, but not accurately quantitive assessment of the amount of 3-hydroxykynurenine present.

It is highly unlikely that a single soluble mediator could account for all the injurious effects of cytotoxic mesenteric lymph or the complexity of multiple-organ failure in AP. The correlation of alterations in least 100 protein and peptide species with the onset and resolution of lymph toxicity is not surprising. The data generated by high- throughput SELDI-TOF MS and hierarchical clustering demonstrates the ability of the bioinformatic algorithm to cluster separately, 1) inert sham-operated lymph samples, 2) lymph samples from rats which were cytotoxic, 3) lymph samples from rats which had previously been cytotoxic and reverted to an inert state, and 4) lymph samples from rats which had not yet developed cytotoxicity but proceeded to become cytotoxic. The clinical correlate is the potential to distinguish between patients who are, 1 ) unlikely ever to develop organ failure, 2) at the height of their illness, 3) have been critically ill but are in the resolution phase, and 4) not critically ill at the time of assessment but are likely to develop organ failure in the immediate future. If borne out by a larger study, this predictive capability would be of great value in the early management of severe AP in humans.

Taken together, the data presented in our study support a novel paradigm for the development of multiple-organ failure in severe AP, presented in Figure 7. We propose that an initiating event of local inflammation and tissue injury in the pancreas triggers a immunological state-of-alarm, mediated by proinflammatory cytokines and other danger signals (39), which leads to DC migration to MLNs. A contradictory combination of proinflammatory signals and lack of cognate microbial antigen, with perhaps increased self or altered-self antigen, leads to a necessary but potentially excessive suppression of immunity. The cytotoxic by-products of the induction of this anergy/suppression, including kynurenine pathway intermediates, overspill into efferent mesenteric lymphatics and pass to the lungs and beyond. They mediate extrapancreatic organ injury, both directly or through the action of effector intermediaries, including damaged erythrocytes and activated neutrophils. This paradigm leads to the proposal that inhibition of IDO and of kynurenine-3- monooxygenase presents an attractive therapeutic strategy for preventing MOF in AP, particularly in patients who present without disseminated symptoms and are identified

by mass spectroscopic analysis as being at imminent risk of developing MOF.

References

1. Steinberg, W., and Tenner, S. 1994. Acute pancreatitis. N EnglJ Med 330:1 198-1210.

2. Johnson, CD., and Abu-Hilal, M. 2004. Persistent organ failure during the first week as a marker of fatal outcome in acute pancreatitis. Gut 53 : 1340-

1344.

3. Beutler, B. 2004. Inferences, questions and possibilities in Toll-like receptor signalling. Nature 430:257-263.

4. Steer, M.L. 1999. Primary intracellular events in pancreatitis. In Acute pancreatitis: novel concepts in biology and therapy. M. Buchler, and P.

Malfertheiner, editors. Berlin: Blackwell Science.

5. Hynninen, M., Valtonen, M., Markkanen, H., Vaara, M., Kuusela, P., Jousela, I., Piilonen, A., and Takkunen, O. 2000. Intramucosal pH and endotoxin and cytokine release in severe acute pancreatitis. Shock 13:79-82. 6. Moore, F.A., Feliciano, D.V., Andrassy, R. J., McArdle, A.H., Booth, F. V.,

Morgenstein- Wagner, T.B., Kellum, J.M., Jr., Welling, R.E., and Moore, E.E. 1992. Early enteral feeding, compared with parenteral, reduces postoperative septic complications. The results of a meta-analysis. Ann Surg 216:172-183.

7. McKay, C. J., Evans, S., Sinclair, M., Carter, C.R., and Imrie, CW. 1999. High early mortality rate from acute pancreatitis in Scotland, 1984-1995. Br J Surg

86:1302-1305.

8. Adams, CA., Jr., Sambol, J.T., Xu, D.Z., Lu, Q., Granger, D.N., and Deitch, E. A. 2001. Hemorrhagic shock induced up-regulation of P-selectin expression is mediated by factors in mesenteric lymph and blunted by mesenteric lymph duct interruption. J Trauma 51 :625-631 ; discussion 631 -622. 9. Deitch, E. A., Feketeova, E., Adams, J.M., Forsythe, R.M., Xu, D.Z., Itagaki, K., and Redl, H. 2006. Lymph from a primate baboon trauma hemorrhagic shock model activates human neutrophils. Shock 25:460-463.

10. Magnotti, L. J., Upperman, J.S., Xu, D.Z., Lu, Q., and Deitch, E. A. 1998. Gut- derived mesenteric lymph but not portal blood increases endothelial cell permeability and promotes lung injury after hemorrhagic shock. Ann Surg 228:518-527.

11. Deitch, E. A., Xu, D., and Kaise, V.L. 2006. Role of the gut in the development of injury- and shock induced SIRS and MODS: the gut-lymph hypothesis, a review. Front Biosci 11 :520-528.

12. Fu, Y.X., and Chaplin, D.D. 1999. Development and maturation of secondary lymphoid tissues. Annu Rev Immunol 17:399-433.

13. Huang, F.P., Platt, N., Wykes, M., Major, J.R., Powell, TJ., Jenkins, CD., and MacPherson, G.G. 2000. A discrete subpopulation of dendritic cells transports apoptotic intestinal epithelial cells to T cell areas of mesenteric lymph nodes. J Exp Med 191:435-444.

14. Pugh, C.W., MacPherson, G.G., and Steer, H. W. 1983. Characterization of nonlymphoid cells derived from rat peripheral lymph. J Exp Med 157:1758- 1779. 15. Steinman, R.M., Hawiger, D., and Nussenzweig, M.C. 2003. Tolerogenic dendritic cells. Annu Rev Immunol 21 :685-711.

16. Sansonetti, P.J. 2006. The innate signaling of dangers and the dangers of innate signaling. Nat Immunol 7:1237-1242.

17. Mc William, A.S., Napoli, S., Marsh, A.M., Pemper, F. L., Nelson, D.J., Pimm, C.L., Stumbles, P.A., Wells, T.N., and Holt, P.G. 1996. Dendritic cells are recruited into the airway epithelium during the inflammatory response to a broad spectrum of stimuli. J Exp Med 184:2429-2432. 18. Mazzoni, A., and Segal, D.M. 2004. Controlling the Toll road to dendritic cell polarization. J Leukoc Biol 75:721-730. 19. Mellor, A.L., Keskin, D.B., Johnson, T., Chandler, P., and Munn, D.H. 2002. Cells expressing indoleamine 2,3-dioxygenase inhibit T cell responses. J Immunol 168:3771-3776.

20. Munn, D.H., Shafizadeh, E., Attwood, J.T., Bondarev, L, Pashine, A., and Mellor, A.L. 1999. Inhibition of T cell proliferation by macrophage tryptophan cateboMsm. J Exp Med 189:1363-1372.

21. Munn, D. H., Sharma, M.D., Lee, J.R., Jhaver, K.G., Johnson, T. S., Keskin, D.B., Marshall, B., Chandler, P., Antonia, S.J., Burgess, R., et al. 2002. Potential regulatory function of human dendritic cells expressing indoleamine 2,3-dioxygenase. Science 297:1867-1870. 22. Munn, D.H., Zhou, M., Attwood, J.T., Bondarev, I., Conway, S. J., Marshall, B., Brown, C, and Mellor, A.L. 1998. Prevention of allogeneic fetal rejection by tryptophan catabolism. Science 281 :1191-1193.

23. Popov, A., Abdullah, Z., Wickenhauser, C, Saric, T., Driesen, J., Hanisch, F. G., Domann, E., Raven, EX., Dehus, O., Hermann, C, et al. 2006. Indoleamine 2,3-dioxygenase-expressing dendritic cells form suppurative granulomas following Listeria monocytogenes infection. J CHn Invest 116:3160-3170.

24. Belladonna, M.L., Grohmann, U., Guidetti, P., Volpi, C, Bianchi, R., Fioretti, M.C., Schwarcz, R., Fallarino, F., and Puccetti, P. 2006. Kynurenine pathway enzymes in dendritic cells initiate tolerogenesis in the absence of functional IDO. J Immunol 177:130-137.

25. Terness, P., Bauer, T.M., Rose, L., Dufter, C, Watzlik, A., Simon, H., and Opelz, G. 2002. Inhibition of allogeneic T cell proliferation by indoleamine 2,3-dioxygenase-expressing dendritic cells: mediation of suppression by tryptophan metabolites. J Exp Med 196:447-457.

26. Rahman, S.H., Ammori, B.J., Holmfield, J., Larvin, M., and McMahon, MJ. 2003. Intestinal hypoperfusion contributes to gut barrier failure in severe acute pancreatitis. J Gastrointest Surg 7:26-35; discussion 35-26. 27. Condon, MR., Kim, J.Ε., Deitch, E. A., Machiedo, G. W., and Spolarics, Z. 2003. Appearance of an erythrocyte population with decreased deformability and hemoglobin content following sepsis. Am J Physiol Heart Circ Physiol 284:H2177-2184.

28. Shah, G.M., Shah, R.G., and Poirier, G.G. 1996. Different cleavage pattern for poly(ADP-ribose) polymerase during necrosis and apoptosis in HL-60 cells.

Biochem Biophys Res Commun 229:838-844.

29. Eisen, M.B., Spellman, P.T., Brown, P.O., and Botstein, D. 1998. Cluster analysis and display of genome- wide expression patterns. Proc Natl Acad Sci USA 95:14863-14868. 30. MacPherson, G.G., Jenkins, CD., Stein, M. J., and Edwards, C. 1995.

Endotoxin-mediated dendritic cell release from the intestine. Characterization of released dendritic cells and TNF dependence. J Immunol 154:1317-1322. 31. Andersson, R., and Andren-Sandberg, A. 2003. Fatal acute pancreatitis.

Characteristics of patients never reaching hospital. Pancreatology 3:64-66. 32. Johnson, CD., Kingsnorth, A.N., Imrie, C. W., McMahon, M.J., Neoptolemos, J.P., McKay, C, Toh, S.K., Skaife, P., Leeder, P.C., Wilson, P., et al. 2001. Double blind, randomised, placebo controlled study of a platelet activating factor antagonist, lexipafant, in the treatment and prevention of organ failure in predicted severe acute pancreatitis. Gut 48:62-69.

33. Berezina, T.L., Zaets, S. B., Mole, DJ., Spolarics, Z., Deitch, E.A., and Machiedo, G. W. 2005. Mesenteric lymph duct ligation decreases red blood cell alterations caused by acute pancreatitis. Am J Surg 190:800-804.

34. Chmiel, B., Grabowska Bochenek, R., Piskorska, D., Skorupa, A., Cierpka, L., and Kusmierski, S. 2002. Red blood cells deformability and oxidative stress in acute pancreatitis. CHn Hemorheol Microcirc 27:155-162.

35. Taylor, P.R., Martinez-Pomares, L., Stacey, M., Lin, H.H., Brown, G.D., and Gordon, S. 2005. Macrophage receptors and immune recognition. Annu Rev Immunol 23:901-944. 36. Krieger, M. 2001. Scavenger receptor class B type I is a multiligand HDL receptor that influences diverse physiologic systems. J CHn Invest 108:793- 797.

37. Kristiansen, M., Graversen, J.H., Jacobsen, C, Sonne, O., Hoffman, H.J., Law, S. K., and Moestrup, S. K. 2001. Identification of the haemoglobin scavenger receptor. Nature 409:198-201.

38. Apostolou, L, Sarukhan, A., Klein, L., and von Boehmer, H. 2002. Origin of regulatory T cells with known specificity for antigen. Nat Immunol 3:756-763.

39. Matzinger, P. 1994. Tolerance, danger, and the extended family. Annu Rev Immunol 12:991-1045. 40. Schmidt, J., Rattner, D. W., Lewandrowski, K., Compton, CC, Mandavilli, U., Knoefel, W.T., and Warshaw, A.L. 1992. A better model of acute pancreatitis for evaluating therapy. Ann Surg 215:44-56.

41. Mole, D.J., Taylor, M.A., McFerran, N.V., and Diamond, T. 2005. The isolated perfused liver response to a 'second hit¹ of portal endotoxin during severe acute pancreatitis. Pancreatology 5:475-485.

42. Yong, S., and Lau, S. 1979. Rapid separation of tryptophan, kynurenines, and indoles using reversed-phase high-performance liquid chromatography. J Chromatogr 175:343-346. 43. Knaus, W.A., Draper, E.A., Wagner, D.P., and Zimmerman, J.E. 1985. APACHE II: a severity of disease classification system. Crit Care Med 13:818-829.

44. Marshall, J.C., Cook, D.J., Christou, N. V., Bernard, G.R., Sprung, CL., and Sibbald, W.J. 1995. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med 23: 1638-1652.

45. Bradley, E.L., 3rd. 1993. A clinically based classification system for acute pancreatitis. Summary of the International Symposium on Acute Pancreatitis, Atlanta, Ga, September 11 through 13, 1992. Arch Surg 128:586-590.

Claims

Claims

1. A compound capable of modulating the kynurenine pathway for use in treating and/or preventing multiple-organ failure (MOF).

2. Use of a compound capable of modulating the kynurenine pathway for the preparation of a medicament for the treatment and/or prevention of multiple-organ failure (MOF).

3. The compound or use of claims 1 or 2, wherein the MOF has a sterile aetiology.

4. The compound or use of claims 1, 2 or 3, wherein the MOF is associated with pancreatitis

5. The compound or use of claims 1-4, wherein the compound capable of modulating the kynurenine pathway inhibits or prevents the production of certain kynurenine pathway components.

6. The compound or use of claims 1-5, wherein the compound capable of modulating the kynurenine pathway is an inhibitor of an enzyme involved in the kynurenine pathway.

7. The compound or use of claim 6, wherein the compound capable of modulating the kynurenine pathway inhibits an enzyme involved in tryptophan catabolism.

8. The compound or use of any preceding claim, wherein the compound capable of modulating the kynurenine pathway is a small organic molecule or nucleic acid.

9. The compound or use of any preceding claim, selected from the group consisting of:

(i) 1 -methyltryptophan;

(ii) 3,4-dichlorobenzoylalanine;

(iii) Tryptophan 2,3-Dioxygenase inhibitors;

(iv) IDO inhibitors; and (v) kynurenine 3 -monooxygenase inhibitors.

10. A method for analysing the multiple organ failure status of a subject, said method comprising the steps of:

(c) providing a sample from a subject; and (d) identifying a component profile of the sample; wherein the component profile of the sample is indicative of the MOF status of the subject.

1 1. The method of claim 10, wherein the component is a protein, peptide, amino acid, small organic molecules and/or nucleic acid.

12. A method for analysing the multiple-organ failure status of a subject, said method comprising the steps of:

(c) providing a sample from a subject; and (d) identifying a protein-expression profile of the sample; wherein the protein expression profile of the sample is indicative of the MOF status of the subject.

13. The methods of claims 10 and 11 wherein the component profile or protein- expression profile is determined by 2D gel electrophoresis, mass spectrometry and/or

High Performance Liquid Chromatography (HPLC).

14. The method of claim 13, wherein the component profile or protein-expression profile is determined by matrix-assisted laser desorption/ionization (MALDI-TOF) or surface-enhanced laser desorption/ionization (SELDI-TOF) mass spectrometry.

15. The methods of claim 10-14, wherein the component profile or protein- expression profile of the sample is compared with the component profile or protein- expression profile of a reference sample.

16. The method of claim 15, wherein a number of reference samples - each derived from subjects of known MOF status, are analysed for their component or protein expression profiles and subjected to a hierarchical clustering system.

17. A data set, comprising representative component or protein expression profile data, wherein the data represents the component or protein expression profiles of samples provided by subjects of known MOF status.

18. A computer pre-loaded with the data set of claim 17.

19. A method for determining the MOF status of a subject, said method comprising the steps of:

(d) obtaining a sample from a subject;

(e) identifying a level of transferrin, haptoglobin and αl -protease inhibitor and the isoforms of apo A 1 ; and

(f) comparing the results obtained with those of a reference sample; wherein a decrease in the level of transferrin and haptoglobin, an increase in the level of αl -protease inhibitor and the presence of altered isoforms of apoAl, is indicative of MOF.

20. The method of claim 19, further including the step of identifying a level of one or more of the kynurenine pathway components.

21. Compounds capable of modulating the kynurenine pathway and/or the level of transferrin, haptoglobin and/or αl -protease inhibitor, for the treatment and/or prevention of MOF.

22. Use of one or more compounds capable of modulating the kynurenine pathway and/or the level of transferrin, haptoglobin and/or αl -protease inhibitor, for the preparation of a medicament for treating MOF.

23. The compound or use of claims 21 or 22, wherein the administrative pattern of the compound(s) or medicament comprises the steps of:

(c) providing a sample from a patient; and (d) identifying a component profile of the sample; wherein the component profile of the sample is indicative of the MOF statusubject and wherein subjects identified as having MOF or at risk of developinge administered said compound(s) or medicament.