CA2334905A1 - Pattern recognition proteins with lectin homology from several animal species and method to use them for measure or modulate innate resistance against bacteria and other pathogenic agents - Google Patents

Abstract

Novel pattern recognition proteins and their use are described. The pattern recognition proteins are useful markers of innate resistance to bacterial infections in cattle and other different animal species. They exhibit homolo gy to polysaccharide-binding proteins and/or are identified as such.

Description

PCTlCA99/00544 Title: PATTERN RECOGNITION PROTEINS
FIFZ.17 OF THE INVENTION
The invention relates to novel pattern recognition proteins methods for determining if an animal has a high innate resistance phenotype; methods for identifying indicators of high innate resistance and methods for enhancing animal resistance.
BACKGROUND OF THE INVENTION
Bacterial infections have considerable economic consequences for the agriculture industry. Young animals, who haven't developed antibody-dependent immunity, are at the greatest risk for infection. Such naive animals require innate defences to mitigate diseases cause by bacterial infections. Consequently, there is a need to identify the proteins that contribute to innate resistance and to select animals that have a high innate resistance to disease. Diseases caused by endemic mucosal bacterial infections are presently dealt with by a combination of hygiene, medication and vaccination to reduce infection and by various dietary and environmental management practices to reduce the impact of stress factors that increase susceptibility.
Such management practises are cost burdens to production in economic market with narrow profit margins for producers.
SUMMARY OF THE INVENTION
The inventors have isolated and characterized bacterial surface oligosaccharide-pattern recognition proteins (hereinafter also referred to as "PRPs") that can bind and target commensal mucosal bacteria that cause respiratory and systemic infections of pigs, cattle, fish and poultry.
Host resistance is mediated by various pattern-recognition proteins (PRPs) in body fluids and on cells. The PRPs can intercept invading mucosal organisms by binding to particular surface oligosaccharides that are distinctly different from configurations of oligosaccharides in host glycoproteins. This "recognition" leads to secondary defences of phagocytic leukocytes which can engulf and destroy the invading bacteria. This facilitates the acquisition of specific immunoglobulin responses that may supplement resistance provided primarily by the PRPs. The resistance of the host is therefore defined by plasma and tissue concentrations of PRPs that can bind to the bacteria when they invade into these compartments. These PRPs are contributors to innate resistance and may participate in the development of antigen-specific immunity in naive animals. As a result, identifying animals with more effective PRPs may allow the selective breeding of high resistance animals. It also allows development of tools to determine if an animal has a high innate resistance to bacteria.
Accordingly, the present invention provides a method for determining if an animal has a high innate resistance phenotype comprising detecting a pattern recognition protein in, a sample from the animal. Some PRPs that function by targeting bacteria to tissues can be used to define tissue susceptability and infection.
Accordingly, the present invention further provides a method for determining if an animal has a high innate resistance phenotype comprising detecting a high innate resistance indicator in a sample from the animal.
The present invention also provides a method of identifying a high innate resistance indicator in an animal, the method comprising the steps of:
1. combining an acessible body fluid of an animal with the surface poly saccharides of an invasive pathogenic bacteria;
2. isolating bound proteins;

3. from the isolated proteins selecting a protein as an indicator, such indicator possessing at least two of the following traits:
a. present in low abundance in tissues;
b. present in high abundance in tissues;
I5 c. consumed coincident with disease;
and having stable expression, and minimal inducibility in response to one or more of common environmental, dietary, inflammatory or immunization experiences.
The present invention further provides a method of augmenting an animal's resistance to a pathogen comprising administering an effective amount of an appropriate indicator in an animal in need thereof.
Further, the present invention provides a method by which dietary or administered monosaccharides or polysaccharides can be used to modify, interactions between PRPs and target bacteria, preferably a PRP, or in combination of proteins, having either of the following traits:
a. present in low abundance in tissues or fluids of susceptible animals, and consumed coincident with disease;
b. present in susceptible tissues in which bacteria localize;
and having stable expression, and minimal inducibility in response to one or more of common environmental, dietary, inflammatory or immunization experiences, comprising administering a sufficient amount of said monosaccharides or polysaccharides to an aminal in need thereof.
Yet further, the present invention provides a method by which PRPs involved in targeting pathogenic bacteria to tissues can be reduced by administration of exogenous competing PRP ligands.
The present invention also provides a method by which PRP's involved in targeting pathogenic bacteria to tissue can be identified by administration of labelled exogenous competing PRP ligands.

Other features and advantages of the present invention will become apparent from the following detailed description. It should be understood, however, that the detailed description and the specific examples while indicating preferred embodiments of the invention are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description.
~R~F DESCRIPTION OF THE DRAWINGS
Figure 1 shows the autoradiographs of 2D SDS PAGE separated 35S-labeled proteins synthesized by primary hepatocyte cultures isolated from a control (A) and a LPS-treated (B) rainbow trout, (C) is 2D SDS PAGE of rainbow trout plasma proteins 48 hours after LPS-treatment.
Figure 2 shows the autoradiographs of 2D SDS PAGE separated 35S-labeled proteins synthesized by primary hepatocyte cultures isolated from form a control (A) and a LPS-treated (B) pig.
Figure 3 shows coomassie stain SDS-reducing PAGE of the various elutions of E. coli lipid free oligosaccharide (LFO) affinity chromatography. (A) Different elutions of E. coli LFO affinity chromatography where lane 1 is air sac fluid; lane 2 is unbound proteins; lane 3 is molecular markers; lane 4 is 40 mM EDTA; lane 5 is 300 mM
Mannose;
and lane 6 is 300 mM GIcNAc; (B) is rhamnan binding proteins of air sac surface fluid wherein lane 1 is molecular markers; lane 2 is 40 mM EDTA eluted; and lane 3 is 300 mM
rhamnose eluted; and (C) is air sac surface fluid proteins that were bound to E. coli LFO or LPS in a Ca++ dependent manner, where lane 1 is molecular markers; lane 2 is LPS bound, EDTA eluted; and lane 3 is LFP bound, EDTA eluted.
Figure 4 shows commassie stain SDS-reducing PAGE of various elutions of E.
coli lipid free oligosaccharide (LFO) affinity chromatography. (A) is various elutions as follows: lane 1 is plasma; lane 2 is salt wash; lane 3 is 40 mM EDTA; lane 4 is rhamnose elution; lane 5 is acid elution; lane 6 is air sac fluid; and lane 7 is MW
markers; and (B) is an immunoblot with rabbit antibodies to avian rhamnan binding lectin where lane 1 is air sac fluid; lane 2 is MW markers; lane 3 is 40 mM EDTA elution; lane 4 is rhamnose elution;
and lane 5 is acid elution.
DETAILED DESCRIPTION OF THE INVENTION
The present inventors have determined that certain proteins are measurable and saturable, and reflect genetically-determined innate resistance of vertebrate species to their host-adapted mucosal microbial flora. As such, these proteins provide markers indicative of the level of resistance in an animal which may, consequently be used in breeding programs to obtain animals with high resistance to pathogens.
Bacteria and other living cells are coated with complex polysaccharides, composed of various monosaccharides linked in linear or branched chains. The enormous number of permutations, combinations and linkage arrangements of the monosaccharides underlie the numerous known capsular and outer membrane-bound surface polysaccharides of bacteria. These differences can be defined structurally and antigenically by established procedures, and underlie the contributions of diverse antibodies to acquired immunity.
However, such acquired antibody-dependent resistance is under-developed in juveniles, but subsequently becomes inducible after exposure to the specifying antigen(s), so contributions of antibodies to polysaccharides, while important, cannot be reliably used to assay innate resistance.
The inventors have considered various polysaccharide binding proteins of vertebrates which are already known. For example, various C-lectins IHansen S, Holmskov U , Immunobiology 199:165, 1998], namely mannan-binding lectins (MBL) [Turner MW, Immunobiology. 199:327-39, 1998] pulmonary surfactant proteins (SP) (Crouch EC, Am J Respir Cell Mol Biol. 19:177, 1998], and S-lectins (mucosal galectins [Hughes RC
Biochem Soc Trans. 25:1194, 1997] have bacterial polysaccharide binding functions.
Humans with genetic loss-of-function mutations of the MBL gene are more susceptible to bacterial infections before antibody responsiveness to polysaccharides mature so there is evidence that this protein is an important part of the genetically determined innate defence of humans. Conglutinin (CG) is a C-lectin of ruminants, long known to be a host complement-binding and enhancing factor that increases complement-dependent opsonization of some bacteria [Holmskov U et al., Immunology 93:431, 1998].
However, it is unclear if and how it functions in direct recognition of bacterial surface polysaccharides of relevance to cattle.
Some pentraxins, namely C-reactive protein (CRP) and serum amyloid protein A (SAP) are long known proteins with microbial surface binding functions [Geu~urz H, et al. Curr Opin Immunol. 7:54, 1995]. There is remarkable variation in constitutive expression among species, and in many species, they can be highly inducible during the acute phase response to inflammatory and other stimuli. Mice expressing transgenic rabbit CRP are more resistant to bacteria [Xia D, et aI, J Immunol 155:2557,1995) and E. coli LPS [Xia D, et al. Proc Natl Acad Sci U S A. 94:2575, 1997); it apparently does this by mitigating the inflammatory response.
The only reported approach towards the use of microbial polysaccharide binding proteins as indicators of genetic resistance has been to assay bovine conglutinin [Holmskov U et al., Immunology 93:431, 1998]. They have limited evidence for a heritable component to blood CG levels in calves, and an inverse correlation with frequency of uncharacterized respiratory diseases of cattle, but there is no increased risk of pneumonia in calves with undetectable CG in serum [Holmskov U et aL, Immunology 93:431, 1998].
Conglutinin was examined because of its known carbohydrate-binding functions, and the authors propose that other known proteins such as MBL and bovine collectin 43 could also be monitored.
In the light of this literature and in the absence of polysaccharide-specific antibodies, the host must use other resistance mechanisms to recognize and dispose of their mucosal bacteria that sometimes invade the normally sterile vascular tissue compartments wherein harmful inflammatory responses develop when resistance is insufficient. This principle specifies that this recognition is achieved by internal expression of surface-polysaccharide binding proteins that bind to common components and arrays within the heterogeneous polysaccharides. Host mucosal adaptation and adherence is also specified to some extent by similar common structures, in accordance with well established evidence that relatively few among many polysaccharide forms (serotypes) are found in bacteria that can reside in a particular species. It follows that resistance mechanisms must match host adaptation mechanisms, so proteins involved would vary in amounts and activities among species. Moreover, it follows that polymorphism within a species, and acquired alterations in availability and/or function of these necessary defences specify the quality and limits of the resistant phenotype of an individual in relation to particular groups of mucosal bacteria.
This principle of surface-pattern matched host adaptation and resistance can be characterized by a "cup-and-saucer" analogy. The "permission" for mucosal occupancy is specified by surface polysaccharide patterns complementary to features of the "cup"
(i.e. the genetically determined mucosal attachment niche), whereas, "denial"
after invasion is specified by features of the "saucer"(i.e. the genetically determined bacterial binding proteins that contribute resistance). Thus, the resistance proteins can be isolated according to their ability to bind host-adapted polysaccharide patterns.
Bacteria that express surfaces that mimic host polysaccharides, such as sialic acid or hyaluronic acid rich surface polysaccharides are well known to be more virulent (i.e. harmful at lower dose). However, most mucosal-adapted bacteria can be effectively eliminated provided that the invading dose does not exceed the internal resistance mechanisms Rather than screen many potential proteins, in many species as is suggested by the work of Holmskov et al., in relation to uncharacterized disease, we have determined that certain of the PRPs are indicators of innate resistance. In this regard, we have developed a protocol by which to identify the most suitable measurable indicators of genetically determined resistance.
The present inventors have identified several novel PRPs that contribute to innate resistance especially in young animals that have not yet developed antibody-dependent immunity. The PRPs have been isolated from several sources including fish, pigs, cattle and turkeys.

Broadly stated, the present invention provides a method for determining if an animal has a high innate resistance phenotype comprising detecting a high innate resistance indicator in a sample from the animal. As used herein a "high innate resistance indicator" or "indicator" is understood to be a protein, preferably a PRP, or a combination of proteins, having either of the following traits:
a. present in low abundance in tissues or fluids of susceptible animals, and consumed coincident with disease;
b. present in susceptible tissues in which bacteria localize;
and having stable expression, and minimal inducibility in response to one or more of common environmental, dietary, inflammatory or immunization experiences.
Accordingly the present invention also provides a method of identifying a high innate resistance indicator in an animal, the method comprising the steps of:
1. combining an acessible body fluid or tissue of an animal with the surface poly saccharides of an invasive pathogenic bacteria;
2. isolating bound proteins;
3. from the isolated proteins selecting a protein, or protein combination as indicator based on it having at least two of the following traits:
a. present in low abundance in susceptible tissues;
b. present in tissues of fluid barriers to invasion;
c. consumed coincident with disease;
and having stable expression, and minimal inducibility in response to one or more of common environmental, dietary, inflammatory or immunization experiences.
PRPs that are unstable either because they degrade in samples, or because they are inducible or depleted consequent to environmentally based experiences measure a phenotype that is part genetically-determined (i.e. innate) and part acquired.
The innate contribution to the phenotype must be quantitatively distinguished from the acquired contribution. The genetic contribution of PRP indicators is more predictably measured if the acquired component is absent or minor. Accordingly, PRPs that are chemically stable and minimally induced in response to environmental influences are identified by establishing that their production is not substantially induced or reduced by inflammatory stimuli that increase expression of various known inducible acute phase proteins.
The term "high innate resistance phenotype" means that the animal is less susceptible to bacterial infections than an animal having lower amounts of functional PRPs required to eliminate invading bacteria or higher amounts of functional PRPs required for tissue localization of invading bacteria. The term "animal"
includes all members of the animal kingdom such as mammals, birds and fish.

With respect to the novel proteins of the present invention it will be appreciated that a protein of the invention may include various structural forms of the primary protein which retain biological activity.
The term "PRP" or "PRP protein" as used herein is intended to include analogs of PRP's, containing one or more amino acid substitutions, insertions, and/or deletions.
Amino acid substitutions may be of a conserved or non-conserved nature.
Conserved amino acid substitutions involve replacing one or more amino acids with amino acids of similar charge, size, and/or hydrophobicity characteristics. When only conserved substitutions are made the resulting analog should be functionally equivalent to a PRP. Non-conserved substitutions involve replacing one or more amino acids with one or more amino acids which possess dissimilar charge, size, and/or hydrophobicity characteristics.
One or more amino acid insertions may be introduced into the amino acid sequence of any PRP. Amino acid insertions may consist of single amino acid residues or sequential amino acids.
Deletions may consist of the removal of one or more amino acids, or discrete portions (e.g.amino acids) from a PRP amino acid sequence. The deleted amino acids may or may not be contiguous.
Also included in the expression "PRP" or "PRP protein" as used herein are homologs of a PRP. Such homologs are proteins whose amino acid sequences are comprised of the amino acid sequences of a PRP's regions from other sources that hybridize under stringent hybridization conditions {which conditions are known to those skilled in the art) with a probe used to obtain a PRP. It is anticipated that a protein comprising an amino acid sequence which is at least 72% preferably 75 to 90% similar, with the amino acid sequence of a PRl' will exhibit the PRP's activity.
As used herein the expression "PRP" or "PRP protein" also contemplates isoforms of a PRP protein. An isoform contains the same number and kinds of amino acids as a PRP, but the isoform has a different molecular structure. The isoforms contemplated by the present invention are those having the same properties as the protein of the invention as described herein. Also included in the expression are other PRP's that carry out the same binding a immune response function of the particular PRP.
As will be appreciated by those skilled in the art, the proteins of the present invention (including truncations, analogs, etc.) may be prepared using recombinant DNA
methods well known to those skilled in the art. In addition, the proteins of the invention including truncations, analogs, etc.) may also be prepared by chemical synthesis using techniques well known in the chemistry of proteins such as solid phase synthesis (Merrifield, 1964, J. Am. Chem. Assoc. 85:2149-2154) or synthesis in homogenous solution (Houbenweyl, 1987, Methods of Organic Chemistry, ed. E. Wansch, Vol. i5 I and II, Thieme, Stuttgart).

_g_ The inventors have isolated several PRPs from Salmonid fishes which are shown in Table 1. These include several novel multimeric ladderlectins (Table 1 A-E), two novel uncharacterized protein lectins (Table 1 H, I) and two known pentraxins with N-terminal sequence homology with trout PRP and SAP homologues (Table 1 F,G).
The PRPs have been demonstrated to bind surface components of Aeromonas salmoncida, a major pathogen in a genetically susceptible Atlantic salmon. In addition, the inventors have prepared polyclonal antiserum specific for ladderlectins for rainbow trout and Atlantic salmon and polyclonal antiserum specific for novel lectin 37.
Accordingly, the present invention provides a method of determining if a fish has a high resistant phenotype comprising detecting a pattern recognition protein according to Table 1 in a sample from a fish. In one embodiment, the pattern recognition protein is R-L37. The PRPs may be detected in a sample using the polyclonal antiserum prepared by the inventors, and by isolation by binding to bacterial surface components.
In a preferred embodiment, the present invention provides a method of determining if a fish has a high resistance to Aeromonas salmoncida comprising detecting a pattern recognition protein according to Table 1 in a sample from the fish.
Swine The inventors have isolated several forms of plasma ficolins with novel binding to the surfaces of Actinobacillus pleuropneumoniae and A. suis, which are important bacterial causes of pneumoniae and systemic infections of pigs (see Table 2, J).
The forms of ficolin identified include 3 subunit forms (molecular weights approx 38, 40 and 42 kilodaltons, each consisting of multiple isoforms differing in isoelectric point between pI 5 and 6). N-terminal aminoacid sequences of each subunit form of plasma ficolin are mixtures of ficolin alpha and ficolin beta (approximately 5-15%).
All forms of porcine plasma ficolins bind in an N-acetylglucosamine-dependent manner to various pathogenic serotypes in which N-acetylglucosamine {GlcNAc) and/or N-acetyl-galactosamine (GaINAc) are present in surface oligosaccharides.
Polyclonal antibodies raised against purified oligosaccharide-binding porcine plasma ficolins specifically identifies all plasma forms. These antibodies also demonstrate that ficolins are normally expressed in pulmonary fluids and alveolar macrophages, mostly as 40 and 42 kDa subunits, whereas ficolins are also expressed in intestinal crypt epithelial cells mostly as the 38 kDa form. Amounts of lung ficolins are normal in lungs of pigs that do not develop pneumonia after aerosol challenge infection with A. pleuropneumoniae, but are substantially and selectively depleted in regions of the lung that become pneumonic, indicating that loss of tissue ficolin correlates with the development of disease from infection. Ficolin levels can be measured in blood and tissues by ELISA and other immunoassays. Ficolins are high but variably expressed in plasma and tissues of piglets, and continue to be expressed throughout the first 6 months of age, but plasma levels are not substantially increased during the acute phase response to inflammation, because the liver expression does not change much, in comparison with known acute phase proteins which are adaptively induced.
Porcine plasma ficolins bind in a GIcNAc-dependent manner more effectively to some strains of Actinobacillus, depending on the GIcNAc and GaINAc composition of their surface polysaccharides. This predicts that some isolates are more likely to deplete ficolin in the lung, and thereby be more pathogenic. Accordingly, ficolin binding to Actinobacillus and other bacteria with GIcNAc and/or GaINAc in their surfaces can be ranked in potential virulence by ficolin-binding functions.
Porcine pulmonary fluids also contains a novel protein (approx 17 kDa subunits) that is recognized by antibodies to the N-terminal 12 aminoacid peptide of ASL-40 of turkeys (Table 3). Antibacterial functions of this protein are predicted based on its relationship with ASL-40, a rhamnose-binding and E. coli oligosaccharide-binding lectin to which it is related. It corresponds to a similar 17 kDa anti-ASL-40 reactive protein in air sac fluids and serum of turkeys.
The inventors have also prepared polyclonal antiserum to porcine ficolins.
Accordingly, the present invention provides a method of determining if a pig has a high resistance phenotype comprising detecting a pattern recognition protein according to Table 2 in a sample from the pig.
In a preferred embodiment, the present invention provides a method of determining if a swine has a high resistance to Actinobacillus pleuropneumoniae or Actinobacillus suis comprising detecting a pattern recognition protein according to Table 2 in a sample from the swine.
Bovine The inventors have isolated several PRPs from cattle as shown in Table 2 K,L.
The PRPs include known proteins with novel oligosaccharides binding function in cattle, namely two isoforms of serum amyloid P (SAP) in bovine plasma (Table 2, K,L).
These PRPs have been shown to bind surface components of Haemophilus somnus, responsible for significant respiratory and systemic disease in feeder cattle.
Accordingly, the present invention provides a method of determining if a bovine animal has a high resistance phenotype comprising detecting a PRP
protein according to Table 2, K,L, in a sample from the cattle.
In a preferred embodiment, the present invention provides a method of determining if a bovine has a high resistance to Haemophilus somnus comprising detecting a pattern recognition protein according to Table 2 K, L in a sample from the bovine.
Avian WO 99/bb321 PCT/CA99/00544 The inventors have also isolated several PRPs from turkeys which are shown in Table 3. The PRPs include a multimeric protein isolated from airsac fluid of young turkeys. This is designated ASL-40 in (Table 3, M). It has been shown to bind surface components of E. coli strain 02 responsible for significant endemic arisacculitis in young turkeys and chickens. ASL-40 is a large multimeric protein that binds to E.
coli oligosaccharides in a rhamnose- and calcium-dependent manner. Polyclonal antibodies raised against a synthetic peptide composed of the first 12 aminoacids of the N-terminus of purified ASL-40 subunits recognize ASL-40 in turkeys and chickens and cross react with another 17 kDa protein in air sac fluids and serum of turkeys. (Table 3).
The inventors have also isolated a novel E. coli binding protein from plasma of turkeys. This protein is composed of approximately 114 kDa subunits with a novel N-terminal acid sequence of PHYTTFDSRRYDFMGT. This protein binds to purified 02 oligosaccharide and intact bacteria in a calcium-dependent manner. This protein (designated PL-114) is a multimeric protien that is present in plasma and egg yolk of turkeys. It binds E. coli 02 oligosaccharide and L-rhamnose, and binds to surfaces of intact E. coli.
Accordingly, the present invention provides a method of determining an avian animal has a high resistance phenotype comprising detecting a PRP protein according to Table 3 in a sample from the avian animal. Preferably, the PRP is ASL-40 and PL-114.
In a preferred embodiment, the present invention provides a method of determining if an avian animal has a high resistance phenotype to E. coli infection comprising detecting ASL-40 in a sample from the avian animal.
Novel Methods Criteria-based protocol by which relevant proteins involved in such innate resistance can be discovered and measured are set out below.
Criteria a . Proteins bind common monosaccharide constituents of surface polysaccharides of invasive pathogenic bacteria adapted as mucosal commensals in a particular species.
The inventors have used as binding targets, purified polysaccharides from main pathogenic serotypes of the following organisms: Aeromonas salmonicida, a respiratory and systemic pathogen of salmonids, especially Atlantic salmon; ActinobacilIus pleuropneumoniae and A. suis, which are respiratory and systemic pathogen of pigs; E.
coli serotype 02, a respiratory and systemic pathogen of chickens and turkeys;
Haemophflus somnus, a respiratory and systemic pathogen of cattle.
b. Relevant proteins are active in accessible body fluids or tissue samples, wherein they are required to intercept invading organisms. The inventors have isolated active proteins from blood plasma from all species above, from respiratory fluids of pigs and turkeys, and from eggs of turkeys.

c. , Relevant proteins bind neutral monosaccharides that are commonly exposed on surfaces of many bacterial commensals. Monosaccharide dependence is established by eluting bound proteins with monosaccharides or oligosaccharides containing common functional groups, namely: N-acetamide-derivatives (of glucose, galactose, mannose and fucose) are eluted with N-acetyl-D-glucosamine, (or other monosacharide N-acetamides); rhamnose or mannose; Calcium chelators for C-lectins, and low pH for avid polyspecific binding.
Bacterial surface binding is established by demonstrating similar proteins bound and dissociated ~~ith specific competing monosaccharides from intact bacteria. The inventors have demonstrated the binding to polysaccharide and intact organisms, namely:
rhamnose- and calcium-dependent binding of turkey ASL-40 [Weebadda et al., FASEB J
13:639.5, 1999] and PL-114; Mannose- and N-acetyl-D-glucosamine- dependent binding of ladderlectins, (Hoover et al, Comp Biochem Physiol B T20:559. 1998];
N-acetyl-D-glucosamine-dependent binding of ficolins. (Delay J. DVSc thesis, University of Guelph 1999]; calcium-dependent binding of pentraxins of salmonids and cattle, and L37 of salmonids.
d. Relevant proteins must be assayable in representative samples.
Assays of amount are determined by specific antibody-based immunoassays;
polyclonal or monoclonal antibodies are prepared routinely (as for ASL-40, ficolins, ladderlectins, Lectin 37) or obtained where available.
Assays of function require prior preparative fractionation of those proteins with polysaccharide-binding activity. Potential competitors in samples and additives must be avoided.
Rainbow trout are genetically resistant to A. salmonicida and we demonstrated that they have more isoforms of ladderlectins than do sensitive Atlantic salmon. We showed that functions of ladderlectins differ between species. Atlantic salmon ladderlectin N-acetylglucosamine-dependent binding is 4 times less avid than in resistant rainbow trout. [Hoover et al, Comp Biochem Physiol B submitted].
e. Relevant proteins should be in low abundance in tissues and consumed coincident with disease.
Bacterial binding proteins that contribute most importantly to innate resistance will be more readily depleted or inhibited, and will be lost in areas where invading bacteria flourish. The inventors have demonstrated depletion of pulmonary ficolins in pneumonia caused by A. pleuropneumoniae [Delay ]. FASEB J
13:639.6, 1999;
DVSc thesis, University of Guelph 1999]
Some polysaccharide binding proteins that mediate binding of invading bacteria or their surface components are predicted to be inversely correlated with innate resistance. This alternative proposition relates to the potential role of some polysaccharide binding proteins in tissue localization of invading bacteria in ways that contribute to the pathogenesis of disease, rather than to elimination of bacteria. Such proteins would be expressed rather than depleted in tissues wherein bacteria are _ localized and cause disease.
f. Relevant proteins should be stably expressed and have minimal inducibility in response to common environmental, dietary, inflammatory or immunization experiences.
The inventors have demonstrated that plasma levels and hepatic synthesis of salmonid ladderlectins, lectin-37 and pentraxins are not induced by exposure to inflammatory stimuli that elicit an acute phase response (Simko E. DVSc thesis, University of Guelph, 1998; Simko et al, J. Fish Dis - I999, In Press] .
The inventors have demonstrated that plasma levels and hepatic synthesis of swine ficolins are not induced by exposure to inflammatory stimuli that elicit an acute phase response (Delay J. DVSc thesis, University of Guelph In preparation;
Seebaransingh et al,. FASEB J 13:643.19, 1999]
g. Relevant proteins should exhibit sufficient variability in constitutive expression to permit specification of high and low expression phenotypes. The inventors have demonstrated that plasma levels of swine ficolins are stably expressed (8-65 mg/L) in young swine (14 days - 6 months), and not induced by exposure to inflammatory stimuli that elicit an acute phase response. [Seebaransingh R. MSc thesis,In preparation]
As is apparent from the foregoing, the authors have developed methods for detection of PRP binding to bacteria. All of the above PRPs have been isolated using a novel chromatographic protocol based on monosaccharide competitive elution from sample proteins bound to purified bacterial oligosaccharides. This involves incubation of intact bacteria with fluids containing PRPs , and then specifically eluting them with competing monosaccharides or calcium chelators. Eluted lectins can then be identified by immunoassays with antibodies against the purified proteins. In an embodiment of this approach, rainbow trout and salmonid plasma ladderlectins bind intact Aeromonas salmonicida that is different between the sensitive (Atlantic salmon) and resistant species. Salmon ladderlectin is collagenase sensitive, and composed of fewer isoforms, which bind in a manner that is more easily competed by GlcNAc. This predicts that PRP
form and affinity correlates with differential susceptibility to bacteria to which these PRP bind.
The inventors have also developed a novel method for assay of functional isoforms and total amounts of several oligosaccharide binding proteins. They have defined a principle whereby functional forms of proteins can be captured in ELISA plates and quantified by labelling with primary anti-lectin antibodies they have prepared.
These approaches can be modified such that the ability of free monosaccharides and oligosaccharides to compete with binding can be specified and quantified.

The inventors have defined a means of quantifying absolute amounts of these proteins, based on competitive radioimmunoassay using preparations of IzSI-labelled purified PRP, and polyclonal lectin-specified antibodies for PRP.
Quantification of - immunoreactive forms in samples can be achieved by comparison of competition against standard curves built from purified authentic proteins. Purification of PRPs provides samples for limited proteolysis and further determination of N-terminal aminoacid sequences of fragments. From these degenerate oligonucleotide primers can be synthesized for PCR amplification of larger cDNA sequences that could be used to screen cDNA
libraries for clones of PRP genes for full sequencing of transdribed regions.
Methods of Detecting PRPs As mentioned previously, the identification of the above referenced PRPs provides a means of determining if an animal has a high resistance phenotype.
Accordingly, the present invention provides a method for determining if an animal has a high resistance phenotype to a bacteria comprising detecting a PRP in a sample from the animal. The PRPs may be detected either by detecting the PRP protein or detecting a nucleic acid molecule encoding a PRP protein.
Antibodies specifically reactive with the PRPs, including enzyme conjugates or labeled derivatives, may be used to detect the PRP in various samples such as tissue fluids and other samples. The antibodies may be used in any known immunoassays which rely on the binding interaction between an antigenic determinant of the PRP
and the antibodies. Examples of such assays are radioimrnunoassays, enzyme immunoassays (e.g.ELISA), immunofluorescence, immunoprecipitation, latex agglutination, hemagglutination, and histochemical tests. Thus, the antibodies may be used to detect and quantify the PRP protein in a sample. In an embodiment of the invention, the PRP
protein is detected by Western blots of samples from various animals.
As mentioned above, the inventors have prepared polyclonal antisera to some of the PRPs of the invention. The antisera can be used to detect the PRPs in a sample. In addition, antibodies specific for the PRP protein may be prepared using conventional methods. To prepare polyclonal antibodies, a mammal (such as a rabbit or mouse) may be immunized with the isoform, or a fragment or synthetic polypeptide specific to the isoform. The immunogenicity of the protein or protein fragment may be enhanced by adding an adjuvant to the protein and/or coupling the protein to an immunogenic carrier.
Adjuvants include Quil A, Freund's adjuvant (complete or incomplete), and tetanus toxin.
Immunogenic carriers include keyhole limpet hemocyanin (KLH) and bovine serum albumin (BSA). The mammal may be immunized several times. Routes of administration include intravenous, intraperitoneal and intramuscular injections. Sera or ascites fluid obtained from the immunized animal may be used as a source of polyclonal antibodies.

. To prepare monoclonal antibodies, lymphocytes may be harvested from a mammal immunized as described above. The lymphocytes may be fused with myeloma cells to prepare hybridoma cells secreting monoclonal antibodies. A hybridoma cell _ secreting an antibody with the appropriate affinity and avidity for the PRP
may be selected and cloned. The techniques for preparing monoclonal antibodies and for selecting clones are well described in the literature (Kohler and Milstein, 1975).
Other techniques may also be utilized to construct monoclonal antibodies (see William D. Huse et al., "Generation of a Large Combinational Library of the Immunoglobulin Repertoire in Phage Lambda;' Science 246:1275-1281, December 1989; see also L. Sastry et al., "Cloning of the Immunological Repertoire in Escherichia coli for Generation of Monoclonal Catalytic Antibodies: Construction of a Heavy Chain Variable Region-Specific cDNA Library," Proc. Natl. Acad. Sci. USA 86:5728-5732, August 1989;
see also Michelle Alting-Mees et al., "Monoclonal Antibody Expression Libraries: A
Rapid Alternative to Hybridomas;' Strategies in Molecular Biology 3:1-9, January 1990;
these references describe a commercial system available from Stratacyte, La Jolla, California, which enables the production of antibodies through recombinant techniques).
Similarly, binding partners may also be constructed utilizing recombinant DNA
techniques to incorporate the variable regions of a gene which encodes a specifically binding antibody.
Enzyme conjugates or labeled derivatives of the antibodies specific for the PRPs may be used in the methods of the invention. Generally, an antibody of the invention may be labelled or conjugated with a substance including various enzymes, biotin, fluorescent materials, luminescent materials, and radioactive materials.
Examples of suitable enzymes include horseradish peroxidase, alkaline phosphatase, [3-galactosidase, or acetylcholinesterase; examples of suitable fluorescent materials include umbellif erone, fluorescein, fluorescein isoth iocyanate, rh odamine, dichlorotriazinylamine fluorescein, dansyl chloride or phycoerythrin; an example of a luminescent material includes iuminoi; and examples of suitable radioactive material include radioactive iodine hz5~ 1131 or tritium.
Indirect methods may also be employed in which the primary antigen-antibody reaction is amplified by the introduction of a second antibody, having specificity for the antibody reactive against the isoform.
The PRP may be detected by screening for nucleic acid molecules (i.e. DNA) encoding the PRP in conventional hybridization methods.
In hybridization methods for detecting nucleic acid molecules encoding the PRP, suitable probes include those based on nucleic acid sequences encoding at least 6 sequential amino acid sequences from the PRP. The nucleic acid probe may be labelled with a radioactive substance including 3zP, 3H or 14C. The probe may also be linked to an antigen that is recognizable by a labelled antibody, a fluorescent compound, biotin, a luminescent compound, an enzyme, or an antibody that is specific for a labelled antigen.
An appropriate label may be selected having regard to the rate of hybridization and binding of the probe to the nucleotide to be detected and the amount of nucleotide available for hybridization. Labelled probes may be hybridized to nucleic acids on solid supports such as nitrocellulose filters or nylon membranes as generally described in Sambrook et a1.,1989.
A nucleic acid molecule encoding a PRP can also be isolated in a sample by selectively amplifying a nucleic acid encoding the isoform using the polymerase chain reaction (PCR) methods. It is possible to design synthetic oligonucleotide primers from the nucleotide sequence for use in PCR. A nucleic acid molecule encoding PRP
can be amplified in a sample using these oligonucleotide primers and standard PCR
amplification techniques (see for example, Innis et al, Academic Press, 1990 and U.S.
Patent 4,800,159).
Amplified products can be isolated and distinguished based on their respective sizes using techniques known in the art. For example, after amplification, the nucleic acid sample can be separated on an agarose gel and visualized, after staining with ethidium bromide, under ultra violet (UV) light. The nucleic acid molecule may be amplified to a desired level and a further extension reaction may be performed to incorporate nucleotide derivatives having detectable markers such as radioactive labelled or biotin labelled nucleoside triphosphates. The detectable markers may be analyzed by restriction and electrophoretic separation or other techniques known in the art.
The conditions which may be employed in methods using PCR are those which permit hybridization and amplification reactions to proceed in the presence of a nucleic acid molecule in a sample and appropriate complementary hybridization primers.
Conditions suitable for the polymerase chain reaction are generally known in the art.
Method of Producing Animals with High Innate Resistance The invention still further provides a method for producing animals which have a high innate resistance comprising selecting animals that express high levels of a pattern recognition protein; and breeding the selected animals. Animals that express high levels of a PRP can be identified using biochemical, immunological, and nucleic acid techniques as described herein.
Transgenic animals may also be prepared which produce high levels of a PRP.
The transgenic animals may be prepared using conventional techniques. For example, a recombinant molecule may be used to provide the gene encoding a PRP, or genes encoding molecules that regulate its expression, by homologous recombination. Such recombinant constructs may be introduced into cells such as embryonic stem cells, by a technique such as transfection, electroporation, injection, etc. Cells which show high levels of expression of a PRP may be identified for example by Southern Blotting, Northern Blotting, or by assaying for expression of a PRP using the methods described herein. Such cells may then be fused to embryonic stem cells to generate transgenic animals expressing high levels of the PRP protein. Germline transmission of the mutation may be achieved by, for example, aggregating the embryonic stem cells with early stage embryos, such as 8 cell embryos, transferring the resulting blastocysts into recipient females in vitro, and generating germline transmission of the resulting aggregation chimeras. Such a transgenic animal may be mated with animals having a similar phenotype i.e. producing high levels of the PRP protein, to produce animals having a higher innate resistance.
Compositions The PRP's identified by the methods described herein may be used for augmenting an animals response to a pathogen and accordingly may be used in the treatment of infections caused by pathogens. The PRP's identified by the methods described herein, including synthetic analogs thereof, are preferably used to treat infections caused by Aeromonas salmonicida, a respiratory and systemic pathogen of salmonids, especially Atlantic salmon;Actinobacillus pleuropneumoniae and A.
suis, which are respiratory and systemic pathogen of pigs; E. coli serotype 02, a respiratory and systemic pathogen of chickens and turkeys; and Haemophilus somnus, a respiratory and systemic pathogen of cattle. It will be appreciated that the substances may also be useful to treat infections caused by other members of the family Pseudomonadaceae (eg. P.
cepacia and P. pseudomallei).
The PRP's identified using the methods described herein may be formulated into pharmaceutical compositions for adminstration to subjects in a biologically compatible form suitable for administration in vivo. By "biologically compatible form suitable for administration in vivo" is meant a form of the substance to be administered in which any toxic effects are outweighed by the therapeutic effects. The substances may be administered to living organisms including humans, and animals. Administration of a therapeutically active amount of the pharmaceutical compositions of the present invention is defined as an amount effective, at dosages and for periods of time necessary to achieve the desired result. For example, a therapeutically active amount of a substance may vary according to factors such as the disease state, age, sex, and weight of the individual, and the ability of the PRP to elicit a desired response in the individual.
Dosage regima may be adjusted to provide the optimum therapeutic response. For example, several divided doses may be administered daily or the dose may be proportionally reduced as indicated by the exigencies of the therapeutic situation.
The PRl' composition may be administered in a convenient manner such as by injection (subcutaneous, intravenous, etc.), oral administration, inhalation, transdermal application, or rectal administration. Depending on the route of administration, the PRP

composition may be coated in a material to protect the composition from the action of enzymes, acids and other natural conditions which may inactivate the PRP.
The compositions described herein can be prepared by per se known methods for the preparation of pharmaceutically acceptable compositions which can be administered to subjects, such that an effective quantity of the active substance is combined in a mixture with a pharmaceutically acceptable vehicle. Suitable vehicles are described, for example, in Remington's Pharmaceutical Sciences (Remington's Pharmaceutical Sciences, Mack Publishing Company, Easton, Pa., USA 1985). On this basis, the compositions include, albeit not exclusively, solutions of the substances in association with one or more pharmaceutically acceptable vehicles or diluents, and contained in buffered solutions with a suitable pH and iso-osmotic with the physiological fluids.
The utility of the proteins and compositions of the invention may be confirmed in experimental model systems.
The following non-limiting examples are illustrative of the present invention:
EXAMPLES

Salmonid fishes:
Collection of trout plasma and serum Blood was collected from healthy adult rainbow trout raised at the University of Guelph as described previously. Serum was obtained from whole blood after coagulation for 18 hours at 4°C. Plasma was obtained from blood collected into 18 mM
sodium citrate (pH 6.0) and subsequently coagulated by incubation at 4°C with 50 mM
calcium chloride. Supernatants were collected by centrifugation at 4,500 x g for 15 min at 4°C, pooled and used fresh or stored at -20°C for later processing. In order to capture all potential LPS-binding proteins, including those for which the virulent LPS
might be inactive, most preparative procedures employed LPS from non-virulent strain ATCC
33658. However, similar profiles of LPS-binding proteins were obtained from virulent and non virulent LPS, and from fresh or frozen serum or plasma.
Purification of A. salmonicida LPS
Nonvirulent (ATCC 33658) and an autoagglutinating virulent (B18:94) (from Dr. V. Ostland, University of Guelph) A. salmonicida were grown at room temperature (RT) in tryptic Soya broth, (TSB), (Difco Laboratories, Detroit MI). When the bacteria were in the late logarithmic phase of growth, (approximately 24 hours for nonvirulent and 36 hours for virulent strains) they were harvested by centrifugation at 1,000 x g for 10 min. The bacterial pellet was washed 3 times with distilled water, lyophilized and then subjected to the Darveau and Hancock procedure for smooth (S)- and rough (R)-form LPS extraction. In brief, the dried pellet was resuspended in 10 mM Tris / 2 mM MgCl2 (pH 8.0) (Mg-Tris) containing deoxyribonuclease (DNase 1, Boehringer Mannheim, Laval, Quebec, Canada) and ribonuclease (IZNase 1 "A", Pharmacia Biotech Inc., Baie d'Urfe, Quebec, Canada). The mixture was then sonicated and incubated with additional RNase and DNase for 2 hours at 37°C. The cells were further lysed by treatment with 10 mM Tris containing 0.25 M EDTA and 5% SDS (pH 9.5) (EDTA-SDS-Tris) for 1 hour at 37°C. The mixture was then ultracentrifuged at 50,000 x g for 30 minutes at 15°C
and the resulting supernatant retained and incubated with pronase (Boehringer Mannheim) overnight at 37°C. LPS was precipitated by adding 2 volumes of 0.375 M MgClz in 95%
ethanol (Mg-ETOH), cooled to 0°C and then centrifuged at 12,000 x g for 15 minutes at 0°C. The 20 resulting pellet was resuspended in EDTA-SDS-Tris, sonicated, the pH
adjusted to 8.0 and then incubated for 30 minutes at 85°C. The pH was then adjusted to 9.5, and the mixture was treated once more with pronase overnight at 37°C. LPS was again precipitated with Mg-ETOH, resuspended in the 10 mM Tris (pH 8.0), sonicated, centrifuged at 150 x g for 5 minutes and the supernatant made to 25 ml with 10 mM Tris containing 25mM
MgCl2. A
gelatinous pellet of A. salmonicida (AS)-LPS was obtained after ultracentrifugation at 50,000 x g for 4 hours at 15°C. The gelatinous pellet was resuspended in distilled water and lyophilized. All incubations were with gentle shaking. The AS-LPS was recognized in a western blot by polyclonal rabbit antisera to whole A. salmonicida (from Dr. V.
Ostland, University of Guelph). The immunoblot of proteinase K-digested whole cell lysate and isolated AS-LPS revealed a major broad banding of 45 to 65 kDa and a minor faster migrating, low molecular weight banding below 14.3 kDa (not shown).
These results are consistent with molecular weights of A. salmonicida whole cell lysates and isolated LPS as seen previously in SDS/PAGE gels and immunoblots .
LPS-Affinity Chromatography The sera was applied to the AS-LPS-Toyopearl 650M or AS-LPS-Sepharose columns under either concentrated and dilute loading conditions. For concentrated loading conditions, 25 ml of the sera was dialyzed (l8hr, 4°C) against tris buffered saline (TBS) {50 mM Tris-HC1,150 mM NaCI, pH 7.8), brought to 5 mM with respect to CaCl2, filtered through Whatman 41 filter paper (Whatman, Maidstone, England) and applied to an AS-LPS-Toyopearl 650M or AS-LPS-Sepharose matrix equilibrated with TBS
containing 5 mM CaCl2 (Ca-TBS). For dilute loading conditions, 25 ml of sera was diluted to 125 ml with Ca-TBS, filtered with Whatman 41 filter paper, and directly applied to the column. Under both concentrated and dilute conditions the column was then washed with Ca-TBS until its absorbance at 280 nm was below 0.05 ABU, followed by a wash with Ca-TBS containing 500 mM NaCI (pH 7.8) until the absorbance at 280 nm was below 0.01 ABU.

The protein-loaded AS-LPS-Toyopearl 650M resin or AS-LPS-Sepharose matrices were eluted under various conditions, the sequence of which was varied according to objectives. Calcium- dependent binding proteins were eluted with TBS
containing 40 _ mM EDTA. Carbohydrate-dependent binding proteins were evaluated by sequential elution with 300 mM D-mannose, then 300 mM D-GIcNAc (Sigma-Aldrich Canada Ltd, Oakville, Ontario, Canada) both in Ca-TBS (pH 7.8), and finally with 40 mM
EDTA in TBS (pH 7.8). The protein fractions from each of the mannose, GIcNAc and EDTA
elutions were processed separately as follows; the eluent was concentrated and the buffer was exchanged with 25 mM Tris-HCl (pH 7.8) in Amicon 10 concentrators (10 kDa cutoff, Amicon, Beverly, MA), restored to 5 mM calcium, passed over the Sepharose 6B
matrix and the bound proteins eluted with the respective monosaccharide (300 mM
concentration in Ca-TBS (pH 7.8)). Similar methods were used for the Toyopearl 650M column.
All columns were washed in TBS (pH 5.0) and cleaned with 6.0 M guaninidine HCl after each run. All resulting eluates were concentrated and washed with 25 mM Tris-HCl (pH 7.5) as described above for later electrophoretic analysis.
Electrophoresis One-dimensional SDS/PAGE was performed using 15% gels following the method of Laemmli (Laemmli, U.K. (1970) as described previously [Hoover, G.H.
(1998).
To examine proteins under reducing conditions samples were heated for 5 min at 90-100°C
in the presence of 100 mM 2-mercaptoethanol. For 2-dimensional PAGE, samples were subjected to isoelectric focusing (IEF) in linear immobilized pH gradient (IPG) 110mm gel strips, pH 4-7 (Pharmacia Biotech Inc.), separated at 300 volts for 3 hours then 1000 volts for 28.6 hours in a Multiphor II horizontal system (Pharmacia Biotech Inc.).
In the second dimension (vertical) the IPG strips were subjected to gradient SDS/PAGE (9-16%) as previously described using a 0.5% w/v agarose stack and a constant current of 40 mA per gel. Gels were stained with Coomassie brilliant blue (Sigma-Aldrich Canada Ltd), destained and analyzed for pI according to horizontal distribution in the IPG
and subunit MW according to standard curves in comparison with protein standards for 1D
(New England BioLabs Inc., Beverly, MA) or 2D PAGE (Sigma-Aldrich Canada Ltd.).
N-Terminal Sequencing and Amino Acid Analysis.
N-terminal amino acid sequences and total amino acid compositions were performed by Dr. T.S. Chen, Biotechnology Service Centre, Department of Clinical Biochemistry, Banting Institute, University of Toronto. For N-terminal amino acid sequence analysis, LPS-affinity purified ladderlectins were separated by SDS/PAGE
under reducing conditions, electroblotted onto polyvinylidene difluoride (PVDF) membranes and stained with Coomassie Blue. N-terminal sequencing was performed on a Porton gas-phase microsequencer (Model 2090) with on-line identification of the phenylthiohydantoin (PTH) derivatives. Sequences were compared against all non-redundant protein and predicted cDNA peptide sequences (BLAST Sequence Similarity Searching, National Center for Biotechnology Information, Bethesda, MD.) and with published reports of plasma proteins of rainbow trout and Atlantic salmon.
Amino acid composition of ladderlectins on PVDF membranes (as above) was performed on a Pico-Tag system (Waters).
Western immunoblots Proteins in the isolated fractions from the Sepharose and Toyopearl columns) were detected by the peroxidase-antiperoxidase (PAP) staining method .
Briefly, PAGE
gels were electrotransferred to nitrocellulose membranes (Bio-Rad Laboratories (Canada) Ltd.) and incubated 3 hours with a 1:200 dilution of various primary rabbit antisera (below). This was followed by 1 hour incubations first with swine immunoglobulins to rabbit immunoglobulins (Z196, Dako Diagnostics Canada Ltd., Mississauga, Ont.
Canada) at a 1: 200 dilution and finally rabbit peroxidase anti-peroxidase (PAP), (Z0013, Dako Diagnostics Canada Ltd.) at a 1:200 dilution. All incubations were at room temperature.
Positive immunoreaction was detected with 4-chloro-1-naphthol (Sigma-Aldrich Canada Ltd.) and 30% hydrogen peroxide. Primary rabbit anti RT-LL and rabbit anti RT-L37 were respectively prepared by immunizing NZ rabbits with Coomassie blue stained 16 kDa or 37 kDA bands excised from reducing SDS/PAGE separations of AS-LPS
binding proteins. Immunization was performed by intradermal injection of 5-10 ~g of purified protein in a 0.5 ml solution of 0.38% Quil A (Langford Laboratories, Guelph, Ont.
Canada) every 3 weeks until high titre immunoblot reactive specific Ig was detected.
Serum collected from rabbits before immunization was used as negative control antisera.
Ladderlectin Binding to Intact A. salmonicda LPS-affinity purified ladderlectin preparations from both species were separately 125 I-labeled by the Iodobead method (Pierce) and purified from free label by G-25 Sepahadex gel filtration in TBS containing 300 mM a-D-glucose.
Nonvirulent (ATCC 33658) and autoagglutinating virulent (B18:94) A. salmonicida were grown at room temperature in TSB until they reached the late logarithmic phase of growth, (approximately 24 hours for nonvirulent and 36 hours for virulent strains). An 80 ml aliquot was removed and pelleted at 8000 x g for 5min, then resuspended/pelleted three times with 50 ml of TBS-Ca. Bacteria were then resuspended into 30 ml TBS-Ca containing 500,000 cpm of 125 I_labeled ladderlectin (RT-LL or SS-LL). The mixture was then allowed to incubate for 16 hrs at 4°C. A 200 ~1 aliquot was removed, 50 pl was counted for radioactivity and 150 ~l was centrifuged at 14,000 rpm to pellet the cells and 50 ltl of this supernatant was counted for radioactivity. The radioactivity of the supernatant plus the cells minus the radioactivity of the supernatant was used to measure total radioactivity associated with the cell pellet. The cell pellet was washed three times with TBS then resuspended into 5 ml of TBS-Ca. The bound 1251-labeled ladderlectins were back titrated from the bacteria by incrementally increasing concentrations of various monosaccharides. Briefly, 100 ~.l aliquots representing ~ 3000 cpm were incubated with increasiung concentrations of the rnonosaccharide. The cell pellet . was then removed by centrifugation and the supernatant was analyzed for the presence of the 1251-labeled ladderlectins released by monosaccharide competition.
Collagenase Digestion of 125 I_labeled Ladderlectins.
Approximately 30,000 cpm of purified lz5 I-labeled RT-LL and SS-LL was resuspended into 20 ~.1 of 25 mM Tris (pH 7.4) containing 50 mM NaCI and 10 mM
CaClz and incubated for 24 h at 37~ C with 10 units of Type I collagenase (Sigma).
As controls, ladderlectins were incubated under the same conditions without collagenase.
The incubation mixtures were then concentrated in Amicon 10 centrifugal concentrators and subjected to 1D SDS-PAGE under reducing and non-reducing conditions as described above.
The radioactive bands were visualized by autoradiography as described previously (LaMarre et al., BBA 1991).
RESULTS
Resistant rainbow trout (RT) have four main proteins that bind A. SaImonicide LPS {see Table 1). We have focused mainly on LPS Ca++-dependent ladderlectins (LL).
Novel ladderlectin forms that bind intact A. salmonicida and LPS in a Glc- and GIcNAc dependent manner have been purified from plasma of rainbow trout and Atlantic salmon (SS). These proteins in each species are composed of 16 kDa subunits covalently (-S-S) associated into multimers of various sizes that display as a ladder on non-reducing PAGE.
By 2D PAGE, purified ladderlectins exist as multiple isoforms. There are at least 3 rainbow trout forms and 2 salmon forms that may be products of N-terminal proteolysis (Table 1). The N-terminal sequences of forms that bind LPS are slightly different from that reported to bind Sepharose (Jensen et al., ref. 1. Table 1). Atlantic salmon have one prominent acidic and one minor LPS binding ladderlectins with related AA
sequences (Table 1) and different functions from those of rainbow trout. Rainbow trout ladderlectin is dissociated from pathogenic A. salmonicida by glucose in the physiological concentration range, conditions under which the Atlantic salmon form remains bound, suggesting that LL function is correlated positively with susceptibility of Atlantic salmon. Atlantic salmon LL binding to GIcNAc on A. salmonicida surfaces is less avid than is rainbow trout LL binding. Ladderlectins from each species bind to non-pathogenic A. saImonicida under these conditions. 125I-ladderlectin of rainbow trout but not Atlantic salmon is resistant to collagenase. Antibodies raised to the 16 kDa rainbow trout subunits label reticular histiocytes in the spleen, activated inflammatory macrophages and skin and pharyngeal surface epithelial cells. These various studies indicate that ladderlectin binding functions correlate with genetic differences in susceptibility to A. salmonicida and with bacterial pathogenicity.
In plasma from rainbow trout and Atlantic salmon, we have also found two pentraxins that bind in a calcium-dependent manner to A. salmonicida LPS. One of these is a protein composed of 24 kDa subunits similar to TCBP-2 (Murata et al., Ref. 2, Table 1), a C-Reactive protein homologue and the other composed of 34 kDa subunits related to trout Serum Amyloid P (Jensen et al., Ref. 3, Table 1). In addition, we identified two LPS-binding plasma proteins with novel N-terminal sequences with no close relationship to any know protein or predicted aminoacid sequence. One multimeric protein composed of 37 kDa subunits has been partially characterized and termed rainbow trout lectin 37 (termed RT-L37). Antibodies to RT-L37 recognizes similar proteins in Atlantic salmon by western blot, and these antibodies recognize colonies of A. salmonicida in tissue sections from rainbow trout and Atlantic salmon with furunculosis. However, the inventors have not yet fully purified RT-L37, so direct binding studies are still pending.
The two pentraxins and RT-L37 are produced by rainbow trout hepatocytes in primary culture, and none is inducible as part of the acute phase response to injected A .
salmonicida adjuvanted vaccines or LPS (Simko et al., submitted). The inventors submitted that salmonids do not exhibit a hepatic acute phase protein synthesis response to bacterial components, unlike mammals (Simko et al., submitted). It appears that these fish have a much broader repertoire of LPS-binding proteins than do the mammals we have examined (see below) but the significance of this observation is still unclear.
Rainbow trout are genetically resistant to Aeromonas salmonicida which causes local and systemic bacterial infections in susceptible salmonids.
Plasma from rainbow trout was examined for proteins with calcium-dependent binding to purified A.
salmonicida lipopolysaccharides (LPS) coupled to an acrylic chromatography matrix.
Four classes to proteins with calcium-dependent PS affinity were identified and partially characterized: (1) An array (80-220 kDa) of multimeric proteins composed of two related 16 kDa subunits with close N-terminal AA sequence homology with rainbow trout ladderlectin (RT-LL), but unrelated to mammalian proteins; (2) a multimeric (210 kDa) pentraxin composed of 34.5 kDa subunits with properties and N-terminal sequence identical to trout serum amyloid P (RT-SAP); (3) a protein composed of 24 kDa subunits with properties and N-terminal sequence identical to the trout C-reactive protein homolog (TCBP2); and (4) a group of multimeric (75-150 kDa) proteins composed of 37 kDa subunits (RT-37) of several isoforms with no close N-terminal homology with fish or mammalian proteins. LPS-binding of RT-LL was mannose dependent wherease RT-SAP
was GlcNAc dependent, and both also bound to Sepharose (galactan). These studies demonstrate that salmonids have a repertoire of oligomeric pentraxins and plasma lectins that might be involved in innate recognition of A. salmonicida and other arganisms with WO 99!66321 PCT/CA99/00544 related surface carbohydrates. RTL-37 and RTL-29 have no known related proteins, but the RT-LLs are related to rainbow trout ladderlectin sequence AAENRNQGPTG
(Jensen LE, et al A rainbow trout lectin with multimeric structure. Comp Biochem Physiol B
_ Biochem Mol Biol. 1997 Apr;116(4):385-90). P34 is related to several known serum amyloid P-like salmonid pentraxins. In addition, P34 is related to a C-reactive protein like known salmonid pentraxin.
Further, proteins SS-LL (Table 1) are proteins which have N-terminal amino acid relationship with rainbow trout ladderlectin sequence AAENRNQGPTG Qensen LE, et al A rainbow trout lectin with multimeric structure. Comp Biochem Physiol B
Biochem Mol Biol. 1997 Apr;116(4):385-90).
Non-Inducible Aspect In order to determine whether these PRP fit the criteria of being non-inducible we conducted experiments in salmonid hepatocytes. In both control and LPS-exposed hepatocytes, labeled proteins were detected in similar amounts in the regions of previously reported A. saImonicida LPS binding RT-L37 lectin (MW~37 kDA, ph5.5-6.1), SAP (MW~34 kDA, pI 5.3-5.6) and CRP (MW.-24 kDA, pI 5.8); however, there were no 35S_labeled proteins in the region of ladderlectin (MW~16 kDa ph5.3-5.75) (see Figure 1).

Swine Materials Chemicals were obtained from Fisher Scientific (Ottawa, Ontario, Canada), unless otherwise specified. Secondary antibodies and reagents used in western blotting and immunohistochemistry were supplied by Dako Corporation (Santa Barbara, CA, USA). Plasma samples for ficolin purification and for bacterial binding assays were obtained from normal 15-90 kg conventionally-reared castrated male or female Yorkshire-cross pigs. Plasma was collected into 0.11 mM sodium citrate and stored at -20°C until used; plasma-derived serum was obtained by coagulation with calcium chloride (40 mM). Tissue samples for immunohistochemical studies were also obtained from pigs aged 14 days (~5kg), 35-40 days (15-20 kg), 7 weeks (~ 20 kg), or 6 months (~90 kg).
Purification of plasma ficolin Isolates of A. suis (AHG96, 02) and A. pleuropneumoniae (4074, Ol) were cultured on chocolate agar, supplemented with NAD for A. pleuropneumoniae .
LPS was extracted by the method of Darveau and Hancock (1983) and characterized by silver-stained PAGE and western blots with anti-sera to LPS. Purified LPS (75 mg) was coupled to 5 g of epoxy-activated Sepharose 6B (Pharmacia) in 0.1 M NaOH.
Control columns were uncoupled epoxy-activated Sepharose 6B or plain Sepharose 6B.
Plasma ficolin was also purified by affinity chromatography on Toyopearl AF-Epoxy-matrix (TosoHaas, Montgomery, PA, USA) coupled with 150 mg N-acetylglucosamine (GlcNAc; Sigma-Aldrich, St Louis, MO). Columns were equilibrated with 50 mM
Tris-HCl (pH 7.8) containing 150 mM NaCI and 20 mM CaCl2. Chromatography was performed at 4°C, and elutions monitored by absorbance (280 nm), conductivity and pH.
Serum (10-30 mls) diluted 1:1 or 1:2 in Tris-HCl equilibration buffer (pH 7.8) was loaded onto 5 ml bed volume columns, then washed extensively with Tris-HCI.
Bound proteins were eluted from Sepharose columns with sequential Tris-HCl buffers (pH

7.8) containing NaCI (500 mM), EDTA (50 mM), aD-glucose (300 mM), or GlcNAc (300 mM). Finally, calcium and monosaccharide- resistant bound proteins were eluted at pH
4.5 with sodium acetate (NaOAc). Proteins bound to the GlcNAc-Toyopearl were eluted similarly except that the unproductive aD-glucose step was omitted. Eluent peaks were pooled in 10 kDa dialysis concentrators (Centricon 10 and Centriprep 10, Amicon Inc., Beverley, MA, USA). These were characterized by PAGE under reducing, non-reducing (SDS) or non-denaturing (native) conditions, and also by two-dimensional (2D)-PAGE and immunoblots under reducing conditions as described by Hoover et al. (1998}.
Anti-porcine ficolin antiserum was prepared by immunizing New Zealand White rabbits with the 38-42 PAGE region from fractions eluted by NaOAc (pH 4.5) from A. suis -LPS-epoxy-Sepharose. Bands were cut from several lanes, homogenized with an equal volume of 0.38% Quil A in water, and injected subcutaneously on three occasions at two week intervals. Immunoblots were performed with whole immune serum and sometimes with immune serum from which LPS-binding proteins and immunoglobulins had been removed, followed by horse radish peroxidase-conjugated goat anti-rabbit Ig as secondary antibody.
Ficolin characterization GIcNAc eluted three distinct reduced bands in the 38, 40 and 42 kDa regions from GlcNAc-Toyopearl. These were electroblotted onto PVDF membranes and each analysed by N-terminal amino acid (Edman) sequencing (NAPS Protein Sequencing and Peptide Mapping Laboratory, University of British Columbia, Vancouver, B.C., Canada).
The same 38-42 kDa region eluted by NaOAc (pH 4.5) from A. suis -LPS-epoxy-Sepharose were analysed together (University of Toronto Biotechnology Service Center, Toronto, Ontario, Canada). These amino acid sequences were compared (NCBI-Blast) with all primary and cDNA predicted sequences.
Plasma ficolin binding to Actinobacillus organisms Porcine Actinobacillus isolates used included reference strains A. suis (ATCC
15557, serotype Ol) and A. pleuropneumoniae (Shope 1) from ATCC. Characterized isolates from clinical cases originating from pigs in southwestern Ontario, Canada were obtained from the Huron Park and Guelph laboratories of the Animal Health Laboratory, University of Guelph, Guelph, Ontario, Canada. These included A.
pleuropneumoniae serotypes , Ol (VSB 5920), 05 ( VSB 6728, VSB 1104,) and 07 (VSB 1116, VSB
4990, CV13-261) and A. suis serotype 02 (H91-0380).
Bacteria were prepared from stocks stored at -70°C in 30%
glycerol. These were plated on blood agar, with 0.02% NAD for A. pleuropneumaniae isolates, and grown overnight at 37°C and 5% C02. Two colonies from each isolate were inoculated into 200 ml of brain-heart infusion (BHI) broth (Difco Laboratories, Detroit, MI,USA).
Serial dilutions made from samples of the resulting suspensions were replated to determine bacterial CFU. Bacterial pellets were harvested by centrifugation at 6000 x g for 15 min at 4°C and washed once with 200 ml of Tris wash buffer (50 mM Tris,150 mM NaCI, 20 mM
CaCl2, pH 7.8). Washed bacteria (approx 1011 were mixed with 30 ml of plasma-derived porcine serum (as used for ficolin purification above) and incubated overnight, with gentle agitation, at 4°C. These large proportions were used so that bound ficolins could be readily distinguished by Coomassie gels of bound proteins. Bacteria were then re-pelleted, washed once with 30 mls of Tris wash buffer, and bound proteins were sequentially eluted in 20 ml volumes of Tris buffer (pH 7.8) as for affinity matrices (see above), containing EDTA (50 mM), aD-glucose (300 mM}, or GIcNAc (300 mM}, and then sodium acetate (100 mM, pH 4.5). In several shortened runs, bacterial binding proteins were eluted with EDTA and GIcNAc only. Bacteria were washed once between each elution. Eluants were dialysed/concentrated (Centriprep 10) and reconstituted to a fixed serum-concentration factor (50x) then compared by reducing SDS-PAGE and anti-ficolin immunoblots. Residual bacterial pellets were lysed with sodium dodecyl sulfate (SDS) lysing buffer (2% SDS, 4% 2-mercaptoethanol, andl0% glycerol in 2M Tris-HCl buffer, pH
6.8), and also analyzed by reducing SDS-PAGE and immunoblots.
Tissue expression in porcine ficolins Immediately after euthanasia, the entire length of the mucosal surface of washed small and large intestine from one 10 kg and three 90 kg pigs were scraped with a #10 scalpel blade. Scrapings were resuspended in phosphate-buffered saline (PBS) supplemented with penicillin G (100 U/ml), streptomycin (0.2 mg/ml), and PMSF
(0.1 mM) (phenylmethanesulfonyl fluoride, Boehringer Mannheim, Laval, Quebec, Canada).
Homogenate preparations (9000 g, 30 minute supernatant) were obtained from scrapings after further disruption with a rotary tissue homogenizes (Tissumizer, Tekmar Co., Cinncinnati, OH, USA) and three freeze-thaw cycles (-70°C to 4°C). After the lungs were removed, supplemented PBS (30 mls) was infused into one mainstem bronchus, and immediately aspirated. Intestinal homogenates and lung lavage fluid were concentrated and subjected to reducing SDS-PAGE and immunoblots.
For immunohistochemistry, tissue samples were fixed in acetone (4°C), embedded in either low melting point (50 - 54°C) or standard (55-57°C) paraffin, and sections (5ltm) were adhered to charge-coated glass slides (Superfrost / Plus, Fisher Scientific, Ottawa, Ontario, Canada). The following incubations were then applied sequentially at room temperature: hydrogen peroxide (3%) to block endogenous peroxidase; bovine serum albumin (BSA, 4%) to block non-specific binding;
primary rabbit anti-ficolin polyclonal antiserum (1:100, 2 hrs), secondary goat anti-rabbit Ig (1: 40, 30 min); rabbit peroxidase-antiperoxidase conjugate (1: 200, 30 min), and 3,3-diaminobenzamine tetrahydrochloride with 30% hydrogen peroxide as chromogen.
Whole pre-immune rabbit serum and antibody diluents were used separately in place of the primary antiserum as negative controls. Sections were counter-stained with Ehrlich's hematoxylin and examined by light microscopy.
Hepatocyte origin and acute-phase inducibility of porcine ficolins was also examined. Hepatocytes were isolated by collagenase perfusion from liver lobes of young pigs (15-20 kg). Three were untreated and three were exposed in vivo to E.
coli LPS (at -28 and -4 hours with 200 wg/kg of E. coli 055:K5 LPS (Sigma-Aldrich, Oakville, Ontario, Canada), a regimen that induces hepatic synthesis and secretion of various porcine acute phase proteins including haptoglobin and major acute phase protein (MAP) of pigs (Seebaransingh et al., 1999). Isolated hepatocytes as attached monolayers were incubated for 3 hours in 35S-labelled methionine / cysteine (Tran 355, ICN
Biomedicals Inc., CA, USA) in methionine/cysteine-free IZPMI (Gibco Life Sciences, Grand Island, NY, USA), and then in serum-free modified Williams' E medium for 16 hours (Wollenberg et al., 1989). 35S-proteins were obtained from culture medium, dialysed (Centricon 10) to remove unincorporated label, subjected to 2D-PAGE (as above), and compared by autoradiography with kDa and pI purified porcine plasma ficoiins. In addition, labelled proteins were compared with immunoreactive forms of ficolin identified by western blots of 2D-PAGE separations of whole plasma of pigs from which hepatocytes were isolated, and from additional pigs 3 days after E, coli LPS treatment.
RESULTS
Ficolin isolation. GIcNAc eluted proteins from GlcNAc-Toyopearl consistently contained a major group of reduced subunits in the 38-42 kDa region, with a prominent ~81 kDa band, and lower concentrations of ~ 25 kDa, ~55 kDa, and ~63 kDa bands. The 38-42 kDa band could be resolved into a predominant broad ~40 kDa band, and two minor bands of ~38 and~42 kDa. These 38-42 kDa bands were not present in the earlier EDTA
elutions from GlcNAc- Toyopearl, whereas the other contaminating bands were present. The 38-42 kDa bands were also found in pH 4.5 (NaOAc) elutions from Epoxy activated-Sepharose coupled with LPS from A. suis or A. pleuropneumoniae but not in GlcNAc or other earlier elutions from these LPS matrices. The p38-42 kDa band was not eluted from LPS-Sepharose with GIcNAc or by other elution solutions, or by any elution from plain Sepharose 6B.

The 38-42 kDa bands from GIcNAc-Toyopearl were each subjected to NH3-terminal amino acid sequencing. This revealed that all three were homologous over 16 residues with porcine ficolins. More than 80% of each band was determined to be ficolin a, whereas between 13 and 18% porcine ficolin p, based on the proportion of alanine in the first sequencing cycle. The entire 38-42 kDa band from A.suis -LPS-Sepharose was homologous over 21 residues with the NH3-terminus of porcine ficolins. In this preparation, approximately 5% consisted of the ficolin (3. By 2D PAGE, at least 5 reduced isoforms with pIs ranging from 5.17-5.82 were identified in the 38-42 kDa region in ficolin preparations from GlcNAc-Toyopearl and A.suis -LPS-Sepharose. Similar pI forms of ficolin were detected by immunoblots prepared from whole porcine plasma subjected to this 2D-PAGE procedure (data not shown). The ficolin bands in all purified preparations analysed migrated in the 38-42 kDa region under non-denaturing SDS-PAGE, and around 800 kDa, below IgM and above human a2-macroglobulin standard under non-denaturing conditions 0720 kDa, LaMarre et al., 1991).
The minor bands were considered contaminants based on their abundance in whole plasma, and their N-terminal sequence homology with immunoglobulin heavy chains (55 kDa and 81 kDa, consistent with IgG/A and IgM respectively) and porcine albumin (63 kDa). The heterogeneous ~25 kDa contaminants were considered to be immunoglobulin light chains based on their more basic pI, their heterogeneity, and their proportional association with the identified heavy chains. The presence of IgM
in these preparations was further confirmed by non-denaturing electrophoretic mobility above immunoreactive ficolins (estimated 800 kDa) and human a2-macroglobulin (data not shown).
Binding of plasma ficolin to intact bacteria. In addition to various minor bands represented in washes from untreated bacterial control cultures, several prominent proteins corresponding to those in original plasma were observed by reducing SDS-PAGE
in plasma elutions with EDTA, glucose, GIcNAc, and NaOAc from intact Actinobacillus bacteria and there was considerable consistency in the specific proteins eluted with each solution and from each bacterial preparation (data not shown). Major bands at ~25 kDa and ~81 kDa (reduced) consistent with Ig subunits were present in EDTA, glucose, and GlcNAc elutions from all A. pleuropneumoniae and A. suis serotypes examined. A
major broad triplet band of typical ~38-42 kDa subunit with ficolin immunoreactivty was eluted by GIcNAc from A. pleuropneumoniae of serotypes 01 and 05. GlcNAc eluted fractions from serotype 05 A. pleuropneumoniae (3164) contained similar 38, 40 and 42 kDa bands and multiple isoforms (pIs 5.1-5.9), with anti-ficolin immunoreactivity corresponding to those in ficolins purified from A. pleuropneumoniae-LPS-Sepharose and GlcNAc-Toyopearl. However, only small amounts of ficolins were eluted with GlcNAc from two serotype 01 (Shope 1 and VSB 5920). Ficolin bands were not evident in Coomassie-blue stained SDS-PAGE preparations from any elutions from A. suis serotypes Ol and 02, but small amounts of immunoreactive ficolin was identified in all elutions from these bacteria. No ficolin was eluted from three serotype 07 A.
pleuropneumoniae (VSB
1116, VSB 4990, CV13-261). Lysates of pellets of all A. pleuropneumoniae strains investigated contained thin 40 kDa immunoreactive ficolin bands in similar amounts after all elutions were completed. This indicated that ficolin did not remain bound to those organisms from which EDTA, GIcNAc or low pH did not elute ficolins.
Tissue distribution of porcine ficolins. Antiserum raised in rabbits injected with LPS-Sepharose binding ficolin (38-42 kDa subunits) had a high titer {l: 2000) reactivity by western blot with the corresponding 3 ficolin subunit bands in plasma from normal pigs, and with these bands in purified ficolins from LPS-Sepharose and GlcNAc-Toyopearl. In plasma and serum samples, the rabbit peroxidase-antiperoxidase cross-reacted with porcine IgG and IgM in primary antibody controls, but no ficolin bands stained in the absence of primary anti-porcine ficolin antibodies. The ficolin triplet was present in similar amounts and proportions in serum and plasma samples from pigs of various ages, including germ-free, and sero-positive or sero-negative to Actinobacillus LPS
(data not shown). Intestinal and colonic mucosal homogenates contained a small ~38 kDa anti-ficolin reactive band, whereas the major .-40 kDa and minor 38 kDa bands, corresponding to those in plasma were evident in from lung washings.
Sections from all pigs revealed similar distribution of anti-ficolin-specific immunoreactive proteins in various tissues (Figure 3.6). Negative control incubations with pre-immune rabbit serum or antibody diluent alone in place of anti-p38-42 antiserum lacked immunoreactivity in the following ficolin-positive locations. Pulmonary alveolar macrophages had ficolin staining in a diffuse intracytoplasmic pattern. Few intravascular mononuclear cells within the lung had similar positive staining, but this was not observed for intravascular mononuclear cells in other tissues. There was positive staining of proteins located at the margin of individual alveolar cell surfaces at the air-tissue interface. Small intestinal cryptal epithelium consistently had strong positive cytoplasmic staining, but goblet cells and enterocytes of villar epithelium were negative.
While cryptal enterocyte staining was present as discrete cytoplasmic vesicles, no extracellular staining was detected in crypt lumens. Colonic crypt epithelial cells stained inconsistently and less intensely and than that identified in small intestine.
Staining of occasional small vesicles was observed in the cytoplasm of few hepatocytes. Small amounts of 35S-labelled proteins corresponding to purified ficolin isoforms (38 kDa, pI 5.2; 40 kDa, pI 5.4-5.6; and 42 kD a, pI 5.5-5.8) were visible by 2D-PAGE / autoradiography of secreted proteins from normal pig hepatocytes (data not shown). Similar bands in similar amounts were also visible after LPS
pretreatment.

Immunoreactive 38, 40, and 42 kDa ficolin bands in plasma were similar in reducing SDS-PAGE both before and 3 days after LPS treatment.
Non-Inducible Aspect Autoradiographs of 2D SDS PAGE separated 35S-labeled proteins synthesized by primary hepatocyte cultures isolated from form a control (A) and a LPS-treated (B) pig. While various known acute phase proteins are induced, ficolins are not.

Turkeys A novel protein (~40 kDa) termed airsac lectin 40 (ASL40) was discovered in airsac secretions. It binds to lipid-free oligosaccharide (LFO) of LPS from pathogenic 02 strains of E. coli responsible for airsacculitis in young turkeys and chickens. ASL-40 has been purified and it has a novel N terminal AA sequence related to a newly reported 2x70 kDa protein termed T-cell stimulating lectin identified in synovial fluid of humans with rheumatoid arthritis (see Table 3) but there are no other close homologies.
1251-labelled forms of ASL-40 bind to intact pathogenic E. coli in a Ca++-dependent manner.
The interesting possibility is that there is evidence for some conserved protein sequence in birds and humans, so it is likely that related proteins will be found in other animals.
Young pigs, broiler chickens and turkeys are all prone to polyserositis (inflammation of serosal cavities, joints and meninges) caused by various bacteria, whereas other mammals and fish are rarely affected.
Some virulent strains of E, coli such as strains 02 and 078 are important as major causative agents of airsacculitis in young turkeys and chickens. The purpose of present study was to examine avian airsac fluid for lectins that may involved in the innate recognition of inhaled virulent strains of E. coli. The approach to identifying these proteins is generally as described in detail for rainbow trout (Example 1) and swine (Example 2). Briefly, potential bacterial binding proteins were isolated by affinity to purified lipid-free oligosaccharide from virulent Escherichia coli 02 coupled to an acrylic matrix (Toyopearl 650). Airsac fluid was collected from 6-18 week old turkeys, loaded in 5 mM calcium buffer and then eluted with 40 mM EDTA and then 300 mM
N-acetylglucosamine (GlcNAc). From the numerous proteins present in the airsac fluid, only two monomeric proteins of 60-64 kDa and 40 kDa protein bound to E. coli oligosaccharide in a calcium- and GlcNAc-dependent manner (see Figures 3 and 4). Polyclonal antibodies to the mammalian pulmonary lectins (SP-A and SP-D) did not react with these or any other proteins in airsac fluids. These studies demonstrate that turkeys have two calcium-dependent respiratory tract lectins that bind E. coli oligosaccharides. These might be involved in local recognition of inhaled E. coli or other bacteria with related surface carbohydrates.

In addition, ASL-40 which is a rhamnose sub-binding protein is identified in Table 3. This protein is related to Human Synovial Stimulating Protein p205 with an N-terminus of DINGGGATLPQPLYQTA (Hair NAK, stuhlmuller, Hahn GR, et al. J.
Immunol 157:1773-1780, 1996). In addition, PL-114 is related to Human Immunoglobulin G
Fc-binding protein which has multiple repeat sequences of PHYTTFDGRRFDFMGT, but is of different size (Harada,N., Iijima,S., Kobayashi,K., Yoshida,T., Brown,W.R., Hibi,T., Oshima,A. and Morikawa,M. Human IgGFc binding protein (FcgammaBP) in colonic epithelial cells exhibits mucin-like structure J. Biol. Chem. 272 (24), 15232-15241 (1997)).
.EXAMPLE 4 Cattle The inventors identified 2 immunoreactive forms of SAP in low abundance in bovine plasma; these bind to glucose-rich lipid-free oligosaccharide (LFO) of H. somnus, a cause of septicemia and pneumonia in young feedlot cattle. These bacterial lectin properties represent a novel function for bovine SAP, but pentraxins of other species have antimicrobial functions.
Haemophilus somnus is a Gram negative inhabitant of mucosal surfaces in cattle and sometimes causes pneumonia and septicemia. To test the hypothesis that cattle have genetically determined but saturable resistance to H. somnus, the inventors looked for carbohydraft pattern recognition proteins in bovine plasma. The approach to identify these proteins as generally as described in detail for rainbow trout (Example 1) and swine (Example 2). Briefly lipo-oligosaccharides (LOS) and acid hydroysed lipid-free oligosaccharide (LFO) purified from HL. somnus was coupled to a non-glycan chromatography matrix (Toyopearl 650M) and each was used for calcium-dependent affinity purification of proteins that could be eluted with EDTA or monosaccharides present in LFO. Three major monomeric proteins (23.5, 25, 26.5 kDa) that bound to LFP but not LOS in a calcium- and carbohydraft-dependent manner had N-terminal amino acid sequence homology with bovine apoliprotein A-1. Two proteins (25.5 kDa major, 31 kDa minor) bound to LFP but not LOS in a calcium-dependent (EDTA-elutable) manner.
These had identical reduced and non-reduced PAGE migration as known forms of bovine serium amyloid P (SAP), and both were immunoreactive by immunoblots with anti-human SAP.
Immunoreactive SAP and corresponding 25.5 and 31 kDa PAGE bands were barely detectable in whole bovine plasma. Since SAP and other pentraxins have pattern recognition properties implicated in disease resistance in other mammals, these findings suggest that bovine SAP might be a saturable defence against invasive infections by H.
somnus.
While the present invention has been described with reference to what are presently considered to be the preferred examples, it is to be understood that the invention is not limited to the disclosed examples. To the contrary, the invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
All publications, patents and patent applications are herein incorporated by reference in their entirety to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated by reference in its entirety.

Salmonid LPS-binding proteins (Compared with Known Proteins in italics) Spe- PRP Targets N-terminal AA sequenceNovel cies Identity for A,B,C,D,E,H,I

RT p16- ASa-LPS AKGEDERVVAAENRNQGPT Ladderlectin-A
or major LFO GFFQ LLl RT p16 ASa-LPS EDERVVAAENRNQEP Ladderlectin-B
or minor LFO LL2 RT p16 ASa-LPS TGAK-----VVAA Ladderlectin-C
or minor LFO L~

RT Sepharose AAENRNQGPTG LadderlectinI

(Gal) RT p24 ASa-LPS ERPMRRSLVFP RT- CRPF F

ftT TCBP Sepharose EIZPMRRSLVFPCRP homolog 2 2 (Gal) RT p34 ASa-LPS DLQDLSGKVFV RT- SAP G

RT SAP Sepharose DLQDLSGKVFV SAP homolog3 (Gal) RT p37 ASa-LPS IQE(D/N) Novel RT-L37H

GHAEPGATTVLNEII,R

RT p29 ASa-LPS SLHPVDXNEVYQSMKXXQ Unknown I

NG

RT p34 A. salmonicidaDLQDLSGKVFV SAP

RT p24 A. salmonicidaERPMRRSLVFP TCBP2 SS p16- ASa-LPS TGAKGAEEGVVPAETRNQSP LadderlectinD

major TGSFQV

SS p16 ASa-LPS GAEEXVVPXXTRN LadderlectinE

minor SS-LL A. salmonicidaTGAKGAEEGWPAETRNQSP Novel TGSFQV

SS-LL A. salmonicidaAEEXXVXPXXTRN Novel Porcine and bovine LPS-binding proteins (Compared wifh Known Proteins in italics) SpeciesPRP Targets N-terminal AA Identity sequence PIG FicolinGIcNAc LDTCPEVKVVGLEGS Pig Ficolin4 a DKLSIL

PIG FicolinGIcNAc ADTCPEVKVVGLEGS Pig Ficolin4 p DKLTIL j3 PIG FC-a A. suis and LDTXPEVKVVGLEGS Ficolin-aJl A.

pleuropneumoniaeDKLSIL

PIG FC-(3 A. suis and ADTXPEVKVVGLEGS Ficolin-~3J2 A.

pleuropneumoniaeDKLSIL

Human FicolinGIcNAc ADTCPEVKVVGLEGS Hu-Ficolin5 a DKLTIL

Bovinep28 HSo-LPS or Blocked (Anti SAP K
LFP SAP

reactive) homolog Bovinep32 HSo-LPS or Blocked (Anti SAP L
LFP SAP

reactive) homolog BovineSAP-1 Yeast mannan, 25 Da SAP form Bovine 6 Sepharose (Gal) SAP major BovineSAP-2 Yeast mamman, 28 kDa SAP form Bovine 6 C3b SAP minor Turkey airsac LPS-binding proteins (Compared with known Proteins in italics) SpeciesPRP Targets N-terminal AA sequenceIdentity*
*

Turkey ASL-40 E. Coli DINGGGATLPQHLYLTPDY Novel M

rhamnose Turkey PL-114 E. Coli PHYTTFDSRRYDFMGT Novel N

rhamnose Human p270/05RheumatoidDINGGGATLPQPLYQTA Synovial7 factor T-cell stimulat-ing lectin

Claims (34)

WE CLAIM:

1. A method for determining if an animal has a high innate resistance phenotype comprising detecting a high innate resistance indicator in a sample from the animal.

2. The method according to claim 1 wherein the sample is a bodily fluid.

3. A method according to claim 1 or 2 wherein the animal is selected from the group consisting of human, bovine, avian, salmonid, and swine.

4. A method according to claim 3 wherein the animal is a salmonid and the indicator is selected from the group consisting of ladderlectin forms specified by N-terminal aminoacid sequences A,B,C,D,E,F,G, H, I as recited in Table 1.

5. A method according to claim 4 wherein the salmonid has a high resistance to Aeromonas salmoncida.

6. A method according to claim 3 wherein the animal is swine and the indicator is selected from the group consisting of bacterial surface binding forms of ficolins.

7. A method according to claim 6 wherein the ficolin is selected from the group consisting of forms specified by N-terminal amino acid sequence tags 4.alpha., 4.beta., Jl and J2 as recited in Table 2.

8. A method according to claim 6 or 7 wherein the swine has a high resistance to Actinobacillus pleuropneumoniae or Actinobacillus suis.

9. A method according to claim 3 wherein the animal is avian and the indicator is selected from the group consisting of ASL-40 or PL-114 forms specified by N-terminal aminoacid sequence tags M,N as recited in Table 3.

10. A method according to claim 9 wherein the avian is a turkey with a high resistance to E. coli.

11. A method according to claim 3 wherein the animal is bovine and the indicator is selected from the group consisting of SAP forms specified by N-terminal amino acid sequence tags K, L, G major and G minor, as recited in Table 3.

12. A method according to claim 11 wherein the bovine has a high resistance to Haemophilus suis.

13. An indicator having an N-terminal amino acid sequence tag of AKGEDERVVAAENRNQGPTGFFQ (Table 1-A).

14. An indicator having an N-terminal amino acid sequence tag of EDERVVAAENRNQEP.

15. An indicator having an N-terminal amino acid sequence tag of TGAK~
VVAA.

16. An indicator having an N-terminal amino acid sequence tag of SLHPVDXNEVYQSMKXXQNG.

17. An indicator having an N-terminal amino acid sequence tag of TGAKGAEEGVVPAETRNQSPTGSFQV.

18. An indicator having an N-terminal amino acid sequence tag of GAEEXVVPXXTRN.

19. An indicator having an N-terminal amino acid sequence tag of TGAKGAEEGVVPAETRNQSPTGSFQV.

20. An indicator having an N-terminal amino acid sequence tag of AEEXXVXPXXTRN.

21. An indicator having an N-terminal amino acid sequence tag of DINGGGATLPQHLYLTPDY.

22. An indicator having an N-terminal amino acid sequence tag of PHYTTFDSRRYDFMGT.

23. A method of identifying a high innate resistance indicator in an animal, the method comprising the steps of:
1. combining an acessible body fluid of an animal with the surface poly saccharides of an invasive pathogenic bacteria;
2. isolating bound proteins;

3. from the isolated proteins selecting a protein as an indicator, such indicator possessing at least two of the following traits:
a. present in low abundance in tissues;
b. present in high abundance in tissues;
c. consumed coincident with disease;
and having stable expression, and minimal inducibility in response to one or more of common environmental, dietary, inflammatory or immunization experiences.

24. A method according to claim 23 wherein the animal is selected from the group consisting of human, bovine, avian, salmonid, and swine.

25. A method according to claim 23 or 24 wherein the protein indicator binds monosaccharides selected from the group consisting of N-acetamide derivatives, rhamnose, and mannose.

26. A method according to claim 25 wherein the monosaccharide is an N-acetamide derivative selected from the group consisting of glucose, galactose, mannose and fucose.

27. A method of augmenting an animal's resistance to a pathogen comprising administering an effective amount of an appropriate indicator in an animal in need thereof.

28. The method of claim 27 wherein the animal is salmonid and the pathogen is Aeromonas salmonicida.

29. The method of claim 27 wherein the aminal is swine and the pathogen is selected from the group consisting of Actinobaccillus pleuropneumoniae and A.
suis.

30. The method of claim 28 wherein the animal is selected from the group of chickens and turkeys and the pathogen is E.coli.

31. The method of claim 28 wherein the animal is bovine and the pathogen is Haemophilius sinnus.

32. A method by which dietary or administered monosaccharides or polysaccharides can be used to modify, interactions between PRPs and target bacteria, preferably a PRP, or in combination of proteins, having either of the following traits:

a. present in low abundance in tissues or fluids of susceptible animals, and consumed coincident with disease;
b. present in susceptible tissues in which bacteria localize;
and having stable expression, and minimal inducibility in response to one or more of common environmental, dietary, inflammatory or immunization experiences, comprising administering a sufficient amount of said monosaccharides or polysaccharides to an aminal in need thereof.

33. A method by which PRPs involved in targeting pathogenic bacteria to tissues can be reduced by administration of exogenous competing PRP ligands.

34. A method by which PRP's involved in targeting pathogenic bacteria to tissue can be identified by administration of labelled exogenous competing PRP
ligands.