WO2003035683A2 - Surfactant protein-d and atherosclerosis - Google Patents

Abstract

The invention describes a novel form of surfactant protein D (SP-D). It discloses a specific in vitro SP-D assay and a method to detect an increased risk for the development of atherosclerosis.

Description

SURFACTANT PROTEIN-D AND ATHEROSCLEROSIS

Technical Field

The present invention is in the field of biochemistry and medicine. The invention relates to a new form of surfactant protein-D, which is expressed in endothelial cells. The invention also relates to pharmaceutical compositions and the use of such compositions to prevent and treat the development of atherosclerosis and atherosclerosis-related diseases. The invention further comprises a method to detect surfactant protein-D in human plasma and serum samples. The invention further relates to a novel method to diagnose a person's susceptibility for having an increased risk for the development of atherosclerosis by measuring the amount of surfactant protein-D in plasma or serum samples. Finally, a further embodiment of the invention is an antibody sandwich ELISA kit to screen for persons with an increased risk for the development of atherosclerosis.

Prior Art

Atherosclerosis

Atherosclerosis is the most prevalent form of Arteriosclerosis, which is a common term for all diseases leading to arterial hardening.

Atherosclerosis is a complex inflammatory-fibroproliferative response to the retention of plasma- derived atherogenic lipoproteins in the arterial intima [Glass CK and Witztum JL (2001 ) Atherosclerosis. Cell 104: 503].

The disease is restricted to large and medium-sized arteries and results in immune activation and the progressive accumulation within the intima of smooth muscle cells (SMS), lipids, lipid laden macro- phages and connective tissues. The continued growth of the lesions encroaches on and destabilizes other layers of the arterial wall, narrows the lumen of the vessel and leads to life-threatening complications rupture or erosion of the plaque and superimposed thrombosis [Aldons JL (2000) Atherosclerosis. Nature 407: 223].

Atherosclerotic-related diseases are a leading cause of death in Western Europe and in the United States accounting for nearly 50 % of all deaths. Epidemiological studies have revealed numerous risk factors associated with atherosclerotic-related disease including family history, diabetes, obesity, gender, smoking, high blood pressure, and infection, respectively. But these factors do not fully account for the widespread prevalence of atherosclerotic-related diseases.

It is well known that the endothelium of the blood vessels has a central role in the progression of the disease and that high serum levels of low-density lipoprotein (LDL) predispose to cardiovascular disease. In particular, blood born monocytes adhere to the vascular endothelial layer and transmigrate through to the sub-endothelial space. Adjacent endothelial cells as well as activated macrophages in the subendothelial space at the same time modify LDL by oxidation. The oxidized LDL is taken up in large amounts by macrophages. Oxidized LDL is the prime candidate in the upregulation of the pro- inflammatory chemokine production by endothelial cells and macrophages. Thus, the accumulation of oxidized LDL in the intima of the vessel walls contribute significantly to monocyte recruitment and foam cell formation [Aldons JL (2000) Atherosclerosis. Nature 407: 223.]. Interaction between foam cells, endothelial cells and SMC lead to a state of chronic inflammation and may lead to proliferation of SMCs, migration of SMCs, and the formation of fibrous plaques.

Although the important clinical complications of atherosclerosis usually occur in the middle-aged or older people, the atherogenic process actually begins at childhood and early adult life. This time window provides the opportunity for the presymptomatic detection of the disease, the identification of persons having an increased risk for development of atherosclerosis, and the application of preventive methods.

Atherosclerosis and Diabetes

Prevention and treatment of long-term micro- and macro-vascular complications remain a critical problem in the management of patients with type 1 or type 2 diabetes mellitus. In the western world about 5 % of the population are affected by diabetes and the disease is the leading cause of new blindness in adults, of new cases of end stage renal disease, and of non-traumatic lower leg amputations. In addition, cardiovascular complications are now the leading cause of diabetes-related morbidity and mortality particularly among women and the elderly. In adult patients with diabetes, the risk of cardiovascular disease is three to five fold greater than in nondiabetics; up to three fourths of all deaths among diabetics can be directly attributed to accelerated atherosclerosis and cardiovascular disease.

Atherosclerosis and Assay Systems

Techniques to assess the onset and course of atherosclerosis are necessary for a subsequent therapy. One measure for the susceptibility to atherosclerosis is the assessment of coronary artery calcium [Arad Y et al. (1996) Predictive Value of Electron Beam Computed Tomography of the Coronary Arteries. Circulation 93: 1951]. But it is clear that more sensitive methods for the detection of the onset of the disease or even for the detection of a predisposition of atherosclerosis are needed.

WO 0013027 suggests that particular diagnostic carbohydrates, in particular heparin sulfate, may be used to determine an imbalance between endogenous heparin production and formation of atherosclerotic plaques. WO 0013027 also discloses a method for assessing risk for and monitoring the progress of development of atherosclerosis by determining the amount of endogenous heparin. This fluorophore assisted carbohydrate electrophoresis diagnosis can be used to diagnose alterations in levels of diagnostic carbohydrates, but WO 0013027 does not provide a method to detect a person's susceptibility for atherosclerosis by measuring a direct parameter of atherosclerosis.

A method for diagnosing a person's susceptibility for having an increased risk for the development of atherosclerosis is revealed in WO 0063430, which methods comprises the determination whether a person has a polymorphism in the signal peptide part of the human NPY (neuropeptide Y, a neurotransmitter widely present in the central and peripheral nervous system). NPY has multiple actions, e.g. control of body energy balance and cardiovascular function. The DNA sequence or the mutant signal peptide of NPY can be used for screening subjects on the polymorphism, whereby the polymorphism is indicative for an increased risk for the development of atherosclerosis. The determination of the polymorphism could be carried out by various molecular biology methods including e. g. restriction fragment length polymorphism (RFLP) or PCR-single stranded conformation polymorphism (SSCP), which methods require an experienced person conducting the diagnosis.

A new method for the diagnosis of cardiovascular disease is disclosed in US 6156500, which method is based on the evaluation of the expression and role of genes (hypothetical genes, endoperoxidase synthase II gene, and bcl-2 gene, respectively) that are differentially expressed in conditions that are physiologically relevant to the disease conditions. The method described permits the definition of disease pathways and the identification of targets in pathways that are useful for diagnosis. Human surfactant protein-D gene or other genes encoding collectins are not described.

Collectins

Collectins are oligomeric proteins composed of C-type lectin domains attached to collagen regions [Holmskov U (2000) Collectins and collectin receptors in innate immunity. Acta Pathologica, Microbio- logica et Immunologica Scandinavica 100: 1]. They play an important role in innate immunity by binding to carbohydrate structures and lipopolysaccharide (LPS) on the surface of pathogenic microorganisms. The binding initiates different effector mechanisms such as aggregation, hindrance of infection, activation of phagocytes, and initiation of phagocytosis. Six different collectins are known. Mannan- binding lectin (MBL) is a serum protein that after binding to microbial surfaces activated the complement system. Lung surfactant protein A and D (SP-A and SP-D) are primarily produced by epithelial cells in the lung but also by epithelial cells on other mucosal surfaces and in ducts of exocrine glands. CL-L1 is the most recent characterized collectin and it has only been found in the liver. Conglutinin and CL-43 are serum proteins only found in bovidae.

SP-D Functions and Implications in Inflammation and Phospholipid Homeostasis

SP-D (known SP-D protein accession numbers (NCBI database): XP_005776; NP_003010; P35247 and AAB59450; CAA46152) is predominantly assembled as dodecamer structures composed of four homotrimeric subunits. Nevertheless, multimers of up to 32 structural subunits are also found. The individual SP-D polypeptide chain comprising 355 amino acids has a molecular mass of 42-44 kDa as determined by SDS-PAGE analysis.

Until recently the expression of SP-D was thought to be restricted to the alveolar type II cell in the lung. But immunohistochemical localization of SP-D revealed that SP-D could be also found in epithelial cells in a variety of tissues [Madsen JA (2000) Localization of lung surfactant protein D on mucosal surfaces in human tissues. J. Immunol. 164: 5866].

SP-D is known to bind carbohydrate structures and LPS (lipopolysaccharide) on the surface of pathogenic microorganisms. The binding initiates the various collectin effector mechanisms such as aggregation, hindrance of infection, activation of phagocytes, and initiation of phagocytosis. SP-D shows a strong chemotactic effect on granulocytes and macrophages and the molecule binds directly to alveolar macrophages. SP-D immune modulation is suggested to be accomplished by the inhibition of T-cell proliferation and IL-2 production and through the direct binding and neutralization of allergens. Four allotypes are known for SP-D. The first is found at position 11 (Met/Thr), the second at position 102 (Pro/Ala), the third at position 160 (Ala/Thr) and the fourth at position 186 (Pro/Asp). It has been shown that allotypes with Thr at position 11 may be more susceptible to tuberculosis than allotypes with Met at position 11 [Floras J et al. (2001) Surfactant protein genetic marker alleles identify a subgroup of tuberculosis in a Mexican population. J. Infect. Dis. 182: 1473]. The allotypes have so far not been correlated to the serum levels of SP-D.

As a SP-D specific receptor, a new member of the SRCR superfamily (gp-340) has been identified [Holmskov U et al. (1997) Isolation and characterization of a new member of the scavenger receptor superfamily, glycoprotein-340 (gp-340), as a lung surfactant protein-D binding molecule. J. Biol. Chem. 272: 13743; Holmskov U (1999) Cloning of gp-340, a putative opsonin receptor for lung surfactant protein D. Proc. Natl. Acad. Sci. U.S.A. 96: 10794].

In addition to the well known immune and inflammation related properties of SP-D the molecule may play a very important role in the lipid homeostasis [Hawgood S and Poulain F (2001) The pulmonary collectins and surfactant metabolism. Annu. Rev. Physiol. 63: 495]. SP-D directly binds to phospholipids. The main implication of SP-D is demonstrated in gene targeted mice. The SP-D knockout mice, SP-D (-/-) mice, accumulate phospholipid in the alveolar space and alveolar macrophages are found in increased numbers with a larger fraction hereof being multinucleated and foamy in appearance [Botas C et al. (1998) Altered surfactant homeostasis and alveolar type II cell morphology in mice lacking surfactant protein D. Proc. Natl. Acad. Sci. U.S.A. 95: 11869; Korfhagen TR (1998) Surfactant protein-D regulates surfactant phospholipid homeostasis in vivo. J. Biol. Chem. 273: 28438]. These changes lead to chronic inflammation, emphysema and fibrosis in the lung of the SP-D knockout mice (SP-D (-/-) mice) possibly because of the increased activity of metalloproteinases in alveolar macrophages and increased hydrogen peroxide production as demonstrated in vitro [Wert SE (2001) Increased metalloproteinase activity, oxidant production, and emphysema in surfactant protein D gene- inactivated mice. Proc. Natl. Acad. Sci U.S.A. 97: 5972].

Finally, SP-D has recently been characterized as a potent endogenous inhibitor of phospholipid and LDL peroxidation and oxidative cellular injuries [Bridges JP (2000) Pulmonary Surfactant Proteins A and D Are Potent Endogenous Inhibitors of Lipid Peroxidation and Oxidative Cellular Injury J. Biol. Chem. 275: 38848]. It was suggested, that SP-D contributes to the protection of the lung surfactant phospholipids from oxidative stress due to atmospheric or supplemental oxygen, air pollution and lung inflammation.

Serum SP-D concentrations have been proposed as a useful prognostic marker in idiopathic pulmonary fibrosis [Takahashi H (2000) Serum surfactant proteins A and D as prognostic factors in idiopathic pulmonary fibrosis and their relationship to disease extent. Am. J. Respir. Crit. Care Med. 162: 1109] and progressive systemic sclerosis [Takahashi H (2000) Serum levels of surfactant proteins A and D are useful biomarkers for interstitial lung disease in patients with progressive systemic sclerosis. Am. J. Respir. Crit. Care Med. 162: 258], and serum SP-D concentrations change during the course of adult respiratory distress syndrome [Greene KE (1999) Serial changes in surfactant- associated proteins in lung and serum before and after onset of ARDS. Am. J. Respir. Crit. Care Med. 160: 1843] and clinical pneumonia [Leth-Larsen R et al. (2001) Serum concentration of surfactant protein D (SP-D) in community-acquired pneumonia: Submitted].

Description of the invention

The invention relates to a novel form of SP-D, the link between SP-D and atherosclerosis, the use of SP-D in the treatment and prevention of atherosclerosis and atherosclerosis-related diseases, and a method for diagnosing an increased risk for the development of atherosclerosis in humans.

Surprisingly, it has been found that the expression of SP-D is not restricted to epithelial cells. SP-D is also expressed in endothelial cells. In addition, the SP-D receptor protein, gp-340, is also present in endothelial cells.

In particular surprisingly, it has been shown that an SP-D polypeptide expressed in endothelial cells lacks 93 amino acids compared to the lung form of SP-D and therefore strongly differs in the amino acid sequence compared to the up to now known epithelial form of SP-D.

More surprisingly, by using a specific SP-D assay diverse SP-D levels in plasma samples derived from normal population were detected, and it was shown that the diverse SP-D levels are genetically based, indicating that there is a genetic determination for the development of atherosclerosis considering the function of SP-D especially as inhibitor of oxidation of LDL but also as an anti-inflammatory and antimicrobial molecule. A further aspect of this invention is the development of an in vitro assay for the determination of the concentration of SP-D in human plasma and serum samples. The method is easy to use and affordable for an ordinary research laboratory. For performing the assay, there is no necessity of a highly qualified person in molecular biology.

It is an advantage of this method that it is possible to detect of the onset of atherosclerosis, or even a predisposition of atherosclerosis by measuring one meaningful parameter, i.e. concentration of SP-D in human plasma and serum samples.

The assay has also been used for determining the SP-D concentrations in plasma samples derived from human twins. The results of this twin study indicate that there is a heritability related to SP-D concentrations in human plasma samples. It has also been investigated that serum SP-D correlates to cardiovascular risk factors, i.e. serum SP-D significantly positively correlates to smoking and to the cholesterol-to-HDL-C ratio and negatively correlates to BMI (Body Mass Index), waiste-to-hip ratio and hypertension.

The effects of SP-D in the modulation of vascular oxidative damage have been investigated. It has been shown that in a dose dependent manner native human SP-D inhibits plasma low density lipoprotein (LDL) oxidation.

By feeding C57BL/6 mice and corresponding SP-D knockout mice with a certain diet, it has been shown that the SP-D knockout mice developed higher levels of total plasma triglycerides, total plasma cholesterol and HDL-C (high density lipoprotein-cholesterol) than the C57BL/6 wildtype strain.

It has been investigated that human recombinant SP-D adminstered intravenously to female SP-D knock-out mice on the high-fat diet resulted in a reduction of plasma total cholesterol, LDL-C and HDL- C cholesterol compared to a control group.

Furthermore, it has been investigated that the total body weight is affected by the genotype of the mice and that SP-D knockout mice gained weight at significantly accelarated rate compared to the wildtype mice.

One subject of this invention refers to a nucleotide sequence selected from the group consisting of:

(a) the nucleotide sequence set out in Fig.1 or its complementary sequence;

(b) a nucleotide sequence which hybridize under stringent conditions to the nucleotide sequence defined in (a) or fragments thereof; and

(c) a nucleotide sequence which, but for the degeneracy of the genetic code, would hybridize to the nucleotide sequence defined in (a) and (b).

According to this invention, a nucleotide is a subunit or chain link in DNA or RNA composed of a sugar, a base, and at least one phosphate group, whereby the bases can be adenine (A), cytosine (C), guanine (G), and thymine (T) or Uracii (U). According to the genetic code, triplets of A, C, G, and T or U, respectively, encode the information for amino acids.

According to this invention, a complementary sequence refers to a nucleotide sequence, which is complementary to another nucleotide sequence strand; that is, wherever one has an adenine the other has a thymine (or an uracii), and wherever one has a guanine the other has a cytosine.

According to this invention, stringency of hybridization refers to the combination of factors, i.e. temparature, salt, and organic solvent concentration, that influence the ability of two polynucleotide strands to hybridize. At high stringency, only nearly perfectly complementary strands will hybridize. At reduced stringency, mismachted can be tolerated.

A further subject of this invention is a polypeptide, which is encoded by one of the nucleotide sequences as claimed in claim 1.

A further subject of this invention is a polypeptide expressed in endothelial cells having the amino acid sequence of Fig. 1.

A polypeptide according to this invention is a single protein chain which can be prepared either by isolating and purifying the protein chain from its natural environment, or by synthesizing the protein chain according methods known in the art, or with aid of an adequate recombinant expression vector system, comprising an adequate recombinant expression vector and an adequate prokaryotic or eukaryotic host cell or host organism. Polypeptides can be recombinantly expressed in vitro and in vivo. Eukaryotic host cells can be of any type comprising yeast cells, cell culture cells such as BHK-21 , CHO (various Chinese Hamster Ovary cell lines), 293 (human kidney carcinoma cell line), NIH 3T3 (murine cell lines), SP2/0 (hybridoma), or other eukaryotic cell lines of pharmaceutical or laboratory research interest, as well as primary cells such as ES cells (Embryonic Stem cells), and other mammalian stem or progenitor cells such as bone marrow progenitor cells. The techniques used of constructing an adequate vector, i.e. cloning the DNA element encoding said polypeptide into said vector, transforming/transfecting cells or organisms, selecting cells or organisms expressing said polypeptide, isolating and purifying said polypeptide are per se known to a person skilled in the art and are specified in Sambrook et al. [Sambrook et al., (1989) In: Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, second edition]. Transfection techniques, e. g. calcium- phosphate-transfection, transfection using DEAE (Diethylaminoethyl-dextran), lipofection, electroporation and infection, e.g. via a viral vector system, are also specified in Sambrook et al. [Sambrook et al. (1989) In: Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, second edition]. As an alternative, other methods not known to a person skilled in the art can be used. For selection of transfected cells, methods described in Sambrook et al. [Sambrook et al. (1989) in: Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, second edition], or other selection methods known to a person skilled in the art can be used. In the case of a bacterial host cell the transformation and selection of positive clones may be performed according to the techniques specified in Sambrook et al. [Sambrook et al. (1989) In: Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, second edition] or by electroporation, bacteriophage infection or other methods known to a person skilled in the art.

A further embodiment of this invention is a polypeptide which:

(a) has part or all of the primary structure of the polypeptide as disclosed herein and inhibits the formation of oxidized low-density lipoprotein;

(b) is a non-naturally occurring polypeptide, and

(c) is the product of procaryotic or eucaryotic expression of an exogenous DNA sequence.

Inhibition of peroxidation of LDL can be measured by the techniques specified by Bridges et al. [Bridges JP (2000) Pulmonary Surfactant Proteins A and D Are Potent Endogenous Inhibitors of Lipid Peroxidation and Oxidative Cellular Injury. J. Biol. Chem. 275: 38848].

This invention also refers to a polypeptide expressed in type II cells of the lung having the amino acid sequence of known SP-D proteins, i.e. the amino acid sequences disclosed by the accession numbers XP_005776; NP_003010; S18382; CAA46152, which polypeptide inhibits the formation of oxidized low- density lipoprotein. Said polypeptides are either homotrimers or dodekamers or higher order of oligomers composed of homotrimers.

A further embodiment of this invention relates to a pharmaceutical composition comprising an amount of the endothelial SP-D polypeptide as specified herein and a pharmaceutically acceptable carrier or excipient, which polypeptide is effective to treat a person to prevent the development of atherosclerosis, which person is diagnosed in having an increased risk for the development of atherosclerosis.

As a further embodiment of this invention there is provided a pharmaceutical composition comprising an amount of surfactant protein-D and a pharmaceutically acceptable carrier or excipient to treat a person to prevent the development of atherosclerosis, which person is diagnosed in having an increased risk for the development of atherosclerosis.

As a further embodiment of this invention there is provided a pharmaceutical composition comprising an amount of surfactant protein-D and a pharmaceutically acceptable carrier or excipient to treat a person which is diagnosed in having an increased risk for the development of atherosclerosis.

In an further embodiment of this invention, pharmaceutical compositions as disclosed before suitable for intra-venous, intra-muscular or apical application are preferred. In a further embodiment of this invention there is provided a pharmaceutical composition as disclosed before for the treatment of persons having diabetes.

In a pharmaceutical composition, the SP-D polypeptides according to this invention can be made available singly or in combination with one or more other pharmaceutically acceptable carrier or excipients, which are adjusted to the application form intended. In the present invention, a pharmaceutical carrier or excipient refers to any substance known to a person skilled in the art, which can influence the bioavailability, the effectiveness, and the maintenance of the SP-D polypeptide within a pharmaceutical composition. Pharmaceutical carriers and excipients are for example described in Bauer et al. [Bauer KH, Froemmig KH, Fuehrer C, Hilfsstoffe, (1999) p. 163 - 186. In: Bauer K. H., Froemmig K. H., Fuehrer C. (ed.) Lehrbuch der Pharmazeutischen Technologie. Wissenschaftliche verlagsgesellschaft mbH Stuttgart, Stuttgart].

Persons diagnosed in having an increased risk for the development of atherosclerosis are humans, which humans have been subjected to a sensitive method for the detection of a predisposition or the onset of the atherosclerosis and/or atherosclerosis related diseases, which method refers to a method known to a person skilled in the art. Such a method refers to noninvasive methods such as vascular ultrasonography, magnet resonance imaging, or electron-beam computed tomography. Blood tests for atherosclerosis, which measure certain risk factors, e.g. high level cholesterol, lipoprotein subtractions, and homocysteine, respectively, are preferred. Such a method also refers to tests detecting other specific markers, e.g. lipoprotein oxidation, or adhesion molecules specific for atherosclerosis. In addition, methods to detect specific markers of endothelial activation or inflammation, such as C-reactive protein and intercellular adhesion molecule 1 , are preferred. In particular one of the in vitro methods disclosed in WO 0063430, WO 0013027, and US 6156500 is preferred. It is particularly preferred that a sensitive method for the detection of a predisposition of atherosclerosis refers to the in vitro method specified herein.

A further embodiment of this invention relates to the use of SP-D or recombinant forms thereof for the manufacture of a pharmaceutical composition as disclosed before for the prevention or treatment of atherosclerosis. In particular, such a use may be preferred in persons diagnosed in having an increased risk for the development of atherosclerosis.

A further aspect of this invention refers to the diagnosis of biochemical or genetically parameters underlying said atherosclerotic-related diseases.

An increased risk for the development of atherosclerosis in a human is considered to refer to the existence of risk factors in said human, which risk factors are known to a person skilled in the art and which are associated with atherosclerotic-related disease including family history, diabetes, obesity, gender, smoking, high blood pressure, and infection, respectively. In connection with this invention, an increased risk for the development of atherosclerosis also refers to plasma or serum SP-D concentrations. An increased risk for the development of atherosclerosis in a human is considered at a detected serum or plasma SP-D concentration below 1104,6 ng/ml, which plasma concentration is the mean value of healthy humans detected by the in vitro method as specified herein (see Example 7). In particular, an increased risk for the development of atherosclerosis in a human refers to a detected SP-D concentration below 743 ng/ml. More particularly, an increased risk for the development of atherosclerosis in a human refers to a detected SP-D concentration below 370 ng/ml or to even lower concentrations, i. e. below 152,7 ng/ml.

In a further embodiment of this invention there is provided the use of surfactant protein-D or recombinant forms thereof as claimed before for the prevention or treatment of atherosclerosis-related disease selected from the group consisting of stroke, kidney failure, blindness, leg amputation and myocardial infarction.

In a further embodiment of this invention there is provided the use of surfactant protein-D or recombinant forms thereof for the manufacture of a pharmaceutical composition for the treatment of obesity.

In a further embodiment of this invention there is provided the use of surfactant protein-D or recombinant forms thereof for the manufacture of a pharmaceutical composition for the treatment of diabetes.

In a further embodiment of this invention there is provided a method for treating a person for the prevention of developing atherosclerosis, which method comprises administering to said person an effective amount of an agent counteracting the influence of a reduced surfactant protein-D plasma or serum concentration.

In a further embodiment of this invention there is provided a method for treating a person for the prevention of developing atherosclerosis, which method comprises administering to said person an effective amount of a surfactant protein-D or recombinant forms thereof counteracting the influence of a reduced surfactant protein-D plasma or serum concentration. In a variant of this embodiment, it is preferred that said method comprises administering to said person an effective amount of an endothelial surfactant protein-D polypeptide as claimed herein.

In a further embodiment of this invention there is provided a method for treating a person for the prevention of developing atherosclerosis, which method comprises subjecting the person to specific gene therapy aimed to repair the genetic basis for the reduced plasma surfactant protein-D concentration. Gene therapy, as used in this invention, refers to techniques described by Smith [Smith Cl (2000) Experiments of nature: primary immune defects deciphered and defeated. Immunol Rev. 178: 5 - 7].

In a further embodiment of this invention there is provided a method for treating atherosclerosis in a human, which comprises administering to a human in need of such treatment a therapeutically effective amount of SP-D or recombinant forms thereof.

In a further embodiment of this invention there is provided a method for treating an atherosclerosis- related disease in a human, which comprises administering to a human in need of such treatment a therapeutically effective amount of SP-D or recombinant forms thereof.

In a further embodiment of this invention there is provided a method for treating an atherosclerosis- related diseases as claimed before, whereby the atherosclerosis-related disease is selected from the group consisting of stroke, kidney failure, blindness, leg amputation and myocardial infarction.

In a further embodiment of this invention there is provided a method for treating or preventing a person developing obesity, which method comprises administering to said person an effective amount of an agent counteracting the influence of a reduced surfactant protein-D plasma or serum concentration.

In a further embodiment of this invention there is provided a method for treating or preventing a person developing obesity, which method comprises administering to said person an effective amount of surfactant protein-D counteracting the influence of a reduced surfactant protein-D plasma or serum concentration.

In a further embodiment of this invention there is provided an in vitro method for diagnosing an increased risk for the development of atherosclerosis by detecting surfactant protein-D in human plasma or serum samples comprising the steps obtaining a sample from a human, assaying the sample to determine the amount of SP-D in the sample, and relating the amount of SP-D in the sample to the clinical status of the human.

According to this invention a sample refers to a plasma or serum sample obtained from a human. A sample according to this invention can for example be obtained from humans by venous puncture.

In a further embodiment of this invention there is provided an in vitro method for diagnosing an increased risk for the development of atherosclerosis disclosed before whereby assaying the sample to determine the amount of SP-D in the sample comprises the use of at least one polyclonal antibody or monoclonal antibody against human surfactant protein-D.

In a further embodiment of this invention there is provided an in vitro method as disclosed before comprising the following steps: a) providing polyclonal or monoclonal antibodies against human surfactant protein-D, b) providing a microtiter plate coated with polyclonal or monoclonal antibodies against human surfactant protein-D, c) adding serum or plasma of samples, calibrator and control samples to a microtiter plate, d) providing a biotinylated polyclonal or monoclonal anti-human surfactant protein-D antibody, e) providing horse radish peroxidase conjugated streptavidin, and f) comparing the reaction which occurs as a result of steps a) to e) with results of calibrator and control samples.

Polyclonal and monoclonal antibodies can be generated according to the techniques described in Antibodies [Antibodies. A laboratory manual, Ed Harrow and David Lane; Cold Spring Harbor Laboratory, New York, 1988]. According to this invention, a microtiter plate refers to a microtiter plate for biochemical or cellular biology purposes known to a person skilled in the art. It refers to a 24-well plate, a 96-well plate, a 384-well plate, and 1536-well plate, respectively. A calibrator sample refers to a serum calibrator, which comprises a dilution series of a pool of sera from at least 2 healthy human subjects. In particular it refers to a dilution series of a pool of sera of at least 4 healthy subjects. It is particularly preferred that a calibrator sample refers to a dilution series of a pool of sera from at least 10 healthy subjects. In accordance to this invention, a control sample refers to sera from at least 2 donors with low and high serum concentration of SP-D, respectively. In particular, a control sample refers to sera from at least 4 donors with low and high serum concentration of SP-D, respectively. It is particularly preferred that a control sample refers to sera from at least 10 donors with low and high serum concentration of SP-D, respectively.

According to this invention, serum concentrations of SP-D can be detected by the techniques as specified herein. By using this technique, a low serum concentration of SP-D refers to less than 1104,6 ng/ml. In particular, it refers to less than 743 ng/ml. Particularly it is preferred that a low serum concentration of SP-D refers to less than 370 ng/ml. More particularly it is preferred that a low serum concentration of SP-D refers to less than 152,7 ng/ml.

According to this invention, a high serum concentration of SP-D refers to at least 1104,6 ng/ml. In particular, a high serum concentration of SP-D refers to at least 1304 ng/ml. It is particularly preferred that a high concentration of SP-D in serum samples refers to at least 2500 ng/ml. It is more particularly preferred that a high concentration of SP-D in serum samples refers to at least 5216 ng/ml.

In a further embodiment of this invention there is provided a method for diagnosing a person's susceptibility for having an increased risk for the development of atherosclerosis, said method comprises determining the concentration of surfactant protein-D in human plasma or serum samples as specified herein, and compare the concentration with the concentration determined in calibrator and control samples. In a further embodiment of this invention there is provided an antibody sandwich ELISA kit to screen persons for an increased risk for the development of atherosclerosis by detecting surfactant protein-D in human plasma and serum, the kit comprising a microtiter plate coated with polyclonal or monoclonal antibodies against human surfactant protein-D, a biotinylated polyclonal or monoclonal anti-human SP- D antibody, horse radish peroxidase conjugated streptavidin, and recombinant surfactant protein-D as an antigen standard.

In accordance to this invention, an antibody sandwich ELISA refers to an enzyme-linked immuno- sorbent assay according to Reen [Reen DJ (1994) Enzyme-linked immunosorbent assay. Methods. Mol. Biol. 32: 461] and to all other ELISA techniques known to a person skilled in the art.

The invention will now be particularly described by way of examples with reference to the given figures.

Fig. 1. Nucleotide sequence and deduced amino acid sequence of the endothelial form of SP-D. The potential glycosylation site is shown in box.

Fig 2. Western blot showing the reaction of the monoclonal antibody Hyb 246-4 (A and B) and the polyclonal antibody K477 (C and D) with SP-D in a pool of serum. A and C: unreduced proteins; B and D: reduced proteins.

Fig.3. Standard curve from serial dilutions of the pool of normal human serum used as calibrator in the SP-D ELISA. The working range was 16-520 ng/ml. The dotted line indicates the detection limit.

Fig. 4. SP-D concentrations in fractions from size chromatography on a Superose 6 column, (v) Normal human serum (200 μl prediluted 1 :2); (σ) human bronchoalveolar lavage fluid (200 μl); and (O) human amniotic fluid (200 μl). Note the different Y-scale for amniotic fluid. The elution positions of blue dextran (BD), IgM, IgG and human serum albumin (HSA) are shown.

Fig. 5. Dilution curves for human serum prediluted 1 :4 (v), human bronchoalveolar lavage fluid prediluted 1 :32 (σ), human amniotic fluid prediluted 1 :64 (O) and the human serum calibrator prediluted 1 :8 (λ) in the SP-D ELISA.

Fig. 6. Path model for genetic and environmental influences on serum concentrations of SP-D in twins. Influences are divided into additive genetic factors (A), genetic dominance factors (D), and shared (C) and non-shared (E) environmental factors. Additive genetic factors and genetic dominance factors are both perfectly correlated in monozygotic (MZ) twins, whereas in dizygotic (DZ) twins the correlation between additive genetic factors is set at 0.5 and the correlation between genetic dominance factors is set at 0.25.

Fig. 7. Correlation diagram for serum SP-D concentrations (μg/l) in MZ (C) and DZ twins (D). The line of identity is shown.

Fig. 8. Sequence alignment of the lung form of SP-D (ISP-D) and the endothelial form of SP-D (eSP-D).

Fig. 9. Immunohistochemical localization of SP-D in human vascular endothelial cells. SP-D immunoreactivity in endothelial cells in a (a) cross section of a lung arthery and type II cells, (b) cross section of a vein in the small intestine, (c) longitudinal section of a vein in the lung and clara cells (d) high endothelial venule in a tonsil, (e) longitudinal section of a vein in the skin, and (f) artherosclerotic lesion. Bars = 100 μm.

Fig. 10. SP-D expression in human endothelial cells, (a) Cross section of a contracted umbilical cord arthery immunostained with Hyb 245-1. Bar = 250 μm (b) Immunostaining of human umbilical cord endothelial cells with Hyb 245-1. Bar = 50 μm (c) Cytospin section of HUAEC #1 immunostained with anti CD31 antibodies. Bar = 50 μm (d) Cytospin section of HUAEC #1 immunostained with anti CD34 antibodies. Bar = 50 μm (e) Immunostaining of cytospin section of HUAEC #1 with Hyb 245-1. Bar = 50 μm. The insert shows a further enhanced endothelial cell, (f) RT-PCR analysis of SP-D mRNA expression in human umbilical cord arthery and vein tissue, HUAEC (AEC) and lung tissue show specific bands at 461 bp. (g) Upper panel shows increased SP-D mRNA expression (461 bp) in HUAEC positively correlating seeding density. Seeding densities were 100,000 cells/well (d=1.5cm), 200,000 cells/well, and 400,000 cells/well. The corresponding β-actin mRNA expression (199 bp) is shown in the lower panel, (h) Western blot analysis of HUAEC total protein and SP-D purified from amnion fluid (AF) in the reduced (R) and unreduced state (U).

Fig. 11. SP-D inhibition of oxidation. Human SP-D inhibition of copper induced LDL oxidation measured by TBARS. LDL was incubated with 10 μM Cu²+ in the presense of the indicated concentrations (μg/ml) of human SP-D, serum or C1q. Lipid peroxidation was quantified by the TBARS assay and expressed as a fold induction of lipid oxidation in the untreated LDL control. Data are means +/- SD, n = 3.

Fig. 12. Boxplots of plasma TBARS levels in SP-D-/- mice and in wildtype mice. N =18 SP-D-/- mice and n =19 wildtype mice.

Fig. 13. Annexin-V-FITC labelling of HUAEC's costimulated with LDL, oxLDL, camptothecin, IgG and mrSP-D. A total of 10,000 cells were counted in each case. All cells are showed in the histogram. The top lane shows annexin-V binding in endothelial cells incubated with treatment medium, 10 μg/ml LDL, or 5 μM camptothecin. The bottom lane shows annexin-V binding in endothelial cells incubated with vehicle (Cu²⁺/EDTA), 10 μg/ml oxLDL , 25 μg/ml oxLDL. ) without addition; ) 1 μg/ml mrSP-D; ) 1 μg/ml IgG. Fig. 14. mrSP-D dose response of Annexin-V-FITC labelling in oxLDL and vehicle treated HUAEC's. A total of 10,000 cells were counted in each case. All cells are showed in the histogram. The top lane shows annexin-V binding in endothelial cells incubated with 20 μg/ml oxLDL, or the corresponding Cu²7EDTA vehicle. The bottom lane shows annexin-V binding in endothelial cells incubated with 30 μg/ml oxLDL, or the corresponding Cu²⁺/EDTA vehicle. ) without addition; ) 0.1 μg/ml mrSP-D; ) 1 μg/ml mrSP-D; • • •) 10 μg/ml mrSP-D.

Fig. 15. The average concentrations of exogenous SP-D in mouse plasma following intravenous infusion of 90 μg human recombinant SP-D in SP-D-/- mice. The horizontal axis indicates the period of hours after SP-D infusion. The vertical axis indicates μg/ml plasma SP-D. Errorbars are 1 x standard deviation, n = 8 animals.

Fig. 16. Average bodyweights of female and male mice. Measurements for every third week are shown, a: female, b: male. The horizontal axis indicates the number of weeks after diet start. The vertical axis indicates grams of bodyweight. (x) wildtype mice fed on normal chow. (■) SP-D-/- mice fed on normal chow, (σ) wildtype mice fed on lipid-enriched diet. (♦) SP-D-/- mice fed on lipid-enriched diet.

Fig. 17. Concentrations of serum SP-D in 4 men following (a) fasting or (b) a high-fat meal. The horizontal axes indicates the period of hours after initiation of bloodsampling and the meal was ingested. The vertical axis indicates ng/ml serum SP-D. The symbols (^χ,B, A,*) each represent one specific person.

Fig. 18. Histogram showing the distribution of serum SP-D concentrations (ng/ml) in 1512 healthy Danish twins.

Fig. 19. Twin-twin correlations of SP-D (top panel) in ng/ml and ln(SP-D) (bottom panel, no unit) for monozygotic (MZ) and dizygotic (DZ) twins.

The following examples do not limit the scope of the invention.

Examples

Example 1 : In vitro assay for quantification of SP-D

Preparation of polyclonal F(ab^')? antibody against human SP-D

Polyclonal antibodies directed against recombinant neck-carbohydrate recognition domain (CRD) of human SP-D [Lausen M et al. (1999) Microfibril-associated protein 4 is present in lung washings and binds to the collagen region of lung surfactant protein D. J. Biol. Chem. 274: 32234] was raised in rabbits and the specificity of the antibody was confirmed by Western blotting. The IgG fraction from the selected antiserum (K477) was digested with pepsin (P-6887, Sigma-Aldrich, St. Louis, MO). Production of monoclonal antibodies

SP-D was purified from amniotic fluid as previously described [Strong P et al. (1998) A novel method of purifying lung surfactant proteins A and D from the lung lavage of alveolar proteinosis patients and from pooled amniotic fluid. J. Immunol. Methods 220: 139] and monoclonal antibodies were raised as previously described [Madsen J et al. (2000) Localization of lung surfactant protein-D on mucosal surfaces in human tissues. J. Immunol. 164: 5866]. Hyb 246-4 was selected for use in Enzyme-linked immunosorbent assay (ELISA) and biotinylated with biotin N-hydroxysuccinimide ester (H-1759, Sigma).

SDS-PAGE and Western blotting

SDS-PAGE and Western blotting was carried out as described in [Madsen J et al. (2000) Localization of lung surfactant protein D on mucosal surfaces in human tissues. J. Immunol. 164: 5866] on 8 - 25 % (w/v) polyacrylamide gradient gels.

ELISA technique

Polystyrene microwell plates (Maxisorp, Nunc, Roskilde, Denmark) were used, with 100 μl additions per well. The plates were washed four times between incubations and all washes and incubations were carried out with Tris-buffered saline (TBS)/ 0.05 % Tween 20 containing 5 mM CaCI₂, unless otherwise stated. The wells were coated with F(ab^')2 anti-human SP-D IgG (K477) at 1 μg/ml in 0.05 M sodium carbonate buffer, pH 9.6. After overnight incubation at 4°C, the plates were washed and left with washing buffer for 15 min at room temperature. Calibrator, controls and dilutions of serum samples were then added and incubated overnight at 4°C. This was followed by successive incubations with biotinylated monoclonal anti-human SP-D antibody 0.5 μg/ml (1 h at room temperature with shaking at 150 rpm), horse radish peroxidase-conjugated streptavidin (43-4323, Zymed, CA) diluted 1 :5000 (30 min) and o-phenylene diamine (Kem-En-Tec, Copenhagen, Denmark) 0.4 mg/ml in citrate-phosphate buffer, pH 5, containing 0.014 % H₂0₂ (15 min in the dark). The color reaction was stopped by adding 150 μl 1 M H₂S0₄ per well. Plates were read at 492 nm in a multichannel spectrophotometer.

SP-D standard and calibrator

A preparation of SP-D purified from amniotic fluid [Strong P et al. (1998) A novel method of purifying lung surfactant proteins A and D from the lung lavage of alveolar proteinosis patients and from pooled amniotic fluid. J. Immunol. Methods. 220: 139-149] was used as standard for the quantification of SP-D in sera. The extinction coefficient at 280 nm for SP-D was calculated from its deduced amino-acid se- quence [Lu J, Willis AC, Reid KB. (1992) Purification, characterization and cDNA cloning of human lung surfactant protein D. Biochem J. 284: 795-80] as 0.49 using the DNASTAR package (DNASTAR Inc., Madison, WI). The optical density at 280 nm of the standard solution was 0.106, giving a concentration of 0.216 mg/ml. By using this SP-D solution as standard, the SP-D concentration of a serum calibrator consisting of a pool of four healthy donor sera was assayed (100 U/ml corresponded to 2078 ng/ml).

Molecular size chromatography

Molecular size chromatography was performed on 200 μl samples of normal human serum, bronchoalveolar lavage fluid and amniotic fluid. The samples were applied to an analytical Superose 6 column connected to an FPLC system (Amersham Pharmacia Biotech, Uppsala, Sweden) using TBS, pH 7.4, containing 10 mM EDTA as eluant at a flow rate of 30 ml/h.

Results

Antibody specificity

The specificity of the antibodies against human SP-D (polyclonal K477 and monoclonal Hyb 246-4) used in the SP-D ELISA was analyzed by Western blotting of serum (Fig. 2). A pool of four sera was spiked with purified amniotic SP-D (15 μg/ml). The pooled serum was loaded at a dilution of 1 :160. Unreduced serum run without dithiothreitol gave a main protein band with a molecular mass of approximately 150 kDa, corresponding to a single subunit of three polypeptide chains [Leth-Larsen R (1999) Structural characterization of human and bovine lung surfactant protein D. Biochem. J. 343: 645]. Higher oligomers of the subunit were also revealed by K477. After reduction, only K477 detected the protein band corresponding to a single polypeptide chain of 43 kDa. The electrophoretic mobility of SP-D visualized by Western blotting agrees with earlier results obtained by mass spectrometry and protein sequence analysis [Leth-Larsen R et al. (1999) Structural characterization of human and bovine lung surfactant protein D. Biochem. J. 343: 645].

SP-D ELISA

The ELISA was set up with duplicate dilutions of the serum calibrator and two quality controls in each plate. The working range of the assay was between 16 and 520 ng/ml with the lower detection limit defined as the highest calibrator dilution (mean - 2 SD, n = 7) deviating from background values (mean + 2 SD, n = 7) (Fig. 3). The quality controls were sera from two donors with low (349 ng/ml) and high (2235 ng/ml) serum concentration of SP-D, respectively. The interassay coefficients of variation were 6.2 % and 9.2 % (n = 18) for low and high quality controls, respectively, and the intraassay coefficients of variation were 1.7 % for both quality controls (n = 7). All samples were tested in duplicates and accepted with a coefficient of variation of 5 %. The assay was free of interference with rheumatoid factors.

Fig. 4 shows the elution patterns from molecular size chromatography of normal serum, amniotic fluid and bronchoalveolar lavage quantified by the SP-D ELISA. Although the elution profiles were not the same for the three antigen sources, the titration curves revealed parallelism in the ELISA (Fig. 5).

Example 2: Heritability estimates for SP-D

A population of 26 monozygotic and 36 like-sexed dizygotic twin pairs aged 6 - 9 years was studied. Intra-class correlation showed significant heritability of SP-D serum concentrations. Biometrics model fitting estimated the heritability as 0.91 (95 % Cl: 0.83 - 0.95) for SP-D with additive genetic and non- shared environmental factors. The data indicate a very strong genetic dependence for the serum concentrations of SP-D.

Study population

The study population comprised 198 like-sexed twins aged 6 - 9 years, who had been enrolled in a study of cord-blood IgE and allergy at birth. Of the 198 twins, 124 individuals participated in the study of SP-D. Table 1 describes the study population in detail. None of the patients had fever or showed other clinical evidence of infection at the time of blood sampling. The study was conducted according to the Helsinki recommendations and approved by the Regional Committee for Research on Human Subjects in the Counties of Funen and Vejle and the corresponding committee of Aarhus County.

Table 1

Twin study population by zygosity: serum concentrations of surfactant protein-D (SP-D) (*p<0.01 ).

Measurement of serum SP-D concentrations

SP-D concentrations in serum were measured by ELISA as described in example 1. The ELISA was set up with a serum calibrator (dilution series of a pool of sera from 4 healthy human subjects) and two quality controls (sera from two donors with low (349 ng/ml) and high (2235 ng/ml) serum concentration of SP-D, respectively). The working range of the assay was 8-519 ng/ml. The interassay coefficients of variation for the two quality controls were 6.2 % and 9.2 % (n = 18) and the intraassay coefficients of variation were 1.7 % in each case (n = 7). The assay was not subject to interference from rheumatoid factors. All serum samples were diluted 10-fold in washing buffer and tested in duplicate. Analysis of data

The effects of sex and zygosity on SP-D concentrations were determined by analysis of variance. Proportions of variance attributable to genetic and environmental factors were then assessed from variance-covariance matrices using the structural equation model approach [Neale MC, Cardon LR (1992) Methodology for genetic studies of twins and families. Dordrecht. Kluwe Academic Publishers] with Mx as statistical software [Neale MC (1994) Statistical modeling. Richmond: Department of Psychiatry, University of Virginia].

The path diagram in Fig. 6 illustrates the univariate model for decomposing the variance of the SP-D concentrations. The total phenotypic variance can be resolved into two genetic and two environmental components. Additive genetic factors (A) are the effects of genes taken singly and added over multiple loci, whereas genetic dominance (D) represents genetic interaction (within loci). Shared environmental effects (C) are those shared by family members, and non-shared environmental effects (E) are those that are unique to each individual. The diagram indicates how each factor contributes to the covariance within an MZ or DZ twin pair. Additive genetic factors and genetic dominance are perfectly correlated in MZ twins, whereas DZ twins, like ordinary siblings, share only half of the additive genetic effects and one quarter of the genetic dominance effects. Shared environmental effects are assumed to be perfectly correlated in MZ and DZ twins. Lower case letters represent genetic and environmental loadings on the trait. The model assumes negligible effects of assortive mating, epistasis, genotype- environment interaction and/or correlation.

Model fitting was by maximum likelihood and the best fitting model was chosen on the following criteria:

- A non-significant p-value in the Chi-squared goodness of fit test;

- Minimizing the Akaike Information Criterion (AIC = P2 - 2 x df)

- No parameter could be eliminated from the model without a significant increase in the Chi-squared goodness of fit statistic.

This approach reflects a balance between goodness of fit and parsimony. Heritability was computed as genetic variance divided by total phenotypic variance. Genetic and environmental variances were derived from the best fitting model. Correlation analyses were visually depicted as diagrams.

Results

Analysis of variance was performed to examine the mean-level differences between sex and zygosity groups. Intraclass correlations for SP-D values were significantly higher in MZ twins than in DZ twins, indicating a substantial genetic contribution to SP-D variation (Table 1). The correlation diagrams for SP-D values in MZ and DZ twins are shown in Fig. 7. In MZ twins a very clear clustering is seen at the line of identity whereas the DZ twins show a scattered pattern. Table 2

Biometrics models for SP-D data (*Akaike Information Criterion (AIC) = χ² - 2 x df; **Best fitting model; A: Additive genetic factors; D: genetic dominance; C: shared environmental effects; E: non-shared environmental effects).

To find the most parsimonious explanation of the observed pattern of resemblance for SP-D, five biometric models were fitted to the normalized data (Table 2). For SP-D, the AE model gave the best fit, suggesting that additive genetic factors and non-shared environmental factors were important. Table 3 shows the genetic and environmental contributions to variation in SP-D for the best fitting model. The estimated heritability for SP-D was 0.91 (95 % Cl: 0.83 - 0.95). The analysis gave high heritability estimates for SP-D values, with acceptable 95 % Cls (Confidence Interval). The heritability was calculated as 0.91 , based on an AE model. Thus, for SP-D the environmental influence was non- shared and at a very limited level. The data show that serum concentrations of SP-D are largely determined genetically. Table 3

Genetic and environmental contributions to variation SP-D (95 % Cl in brackets) (*Variances are standardized).

Example 3: Cloning of surfactant protein D from arterial endothelial cells

Total RNA was purified from primary arterial endothelial cells (AEC) by means of the RNeasy mini kit (Qiagen, CA, USA). First-strand synthesis was performed with the Superscript II reverse transcriptase (Life Technologies, MD, USA) from 1 μg of total RNA with oligo-dT priming according to manufactures instructions. Shortly, total RNA was mixed with the oligo-dT primer and incubated at 65°C for 5 min. and then put on ice. The reaction was mixed with buffer and enzyme in a total volume of 20 μl and incubated at 42°C for 50 min followed by 70°C for 15 min. The cDNA reaction was diluted five times with water. PCR was performed in a volume of 30 μl containing 10 pmol of each primer (forward primer: Hu SP-D leader f1 : 5'-ATGCTGCTCTTCCTCCTCTCT-3' (SEQ ID No. 3) and reverse primer: Hu SP-D 3': 5'-TCAGAACTCGCAGACCA-3' (SEQ ID No. 4), 1 unit of Taq polymerase (Life Technologies), 100 μM of each dNTP and 1 x PCR buffer with 1.5 mM MgCI₂ (Life Technologies). The primer pair spans the translated region of human SP-D mRNA. The PCR was as following: 30 seconds at 94°C, 54 cycles of (94°C for 30 seconds, 60°C for 30 seconds and 72°C for 105 seconds), followed by 72°C for 7 minutes. The AEC SP-D band had a molecular size corresponding to 850 bp. The band was extracted (Qiaex II gel extraction kit, Qiagen) and sequenced (Big Dye Terminator Cycle Sequencing, PE Applied Biosystems, Warrington, UK).

Results

A fragment of 863 bp was amplified by RT-PCR applied to total RNA from an arterial endothelial cell line using primers spanning from the start methionine to the end of the CRD (Fig. 1). In parallel the expected 1125 bp product was amplified from lung mRNA. The sequence of the 863 bp endothelial cell product (Fig. 1) showed that the sequence corresponding to exon 3, 4 and first five Gly-Xaa-Yaa repeats of exon 5 was missing. So it seems as if the sequence jumps direct from the end of exon 2 into second half of exon 5 and them into the neck region. So the endothelial form of SP-D has only 28 Gly- Xaa- Yaa repeats as compared to 57 in the lung form of SP-D. One productive difference was found in the endothelial sequence at position 166 where a Ser to Gly substitution was found (Fig. 8). Example 4: Surfactant protein D is expressed by endothelial cells, inhibits oxidation of low density lipoprotein and protects endothelial cells from oxidative damage

SP-D expression has been detected in endothelial cells and in addition it was observed that human SP-D inhibited low density LDL oxidation in vitro. SP-D-/- mice had significantly higher plasma levels of oxidation than wildtype mice and a recombinant fragment of SP-D was able to inhibit oxLDL induction of endothelial cell apoptosis under serum free conditions.

Cell culture, RNA purification and RT-PCR

Primary human arterial endothelial cell cultures, HUAEC, were isolated from normal human umbilical cords by collagenase treatment. The cords were obtained from anonymous donors from the Department of Obstetric and Gynecology at Odense University Hospital. HUAEC cultures were seeded on gelatine coated tissue culture plastic and the cultures were maintained in EGM-2 medium (Clonetics) and kept at 37°C, 95% humidity, 5% C0₂. Cells were detached for passaging with trypsin- EDTA solution for endothelial cultures (Sigma) and used untill passage 7. Total RNA purification was performed by Rneasy Mini kit (Qiagen). Veins and artheries were dissected from normal human umbilical cords and stored in RNAIater (Ambion) until total RNA purification by Tri Reagent (Sigma- Aldrich). First-strand cDNA was generated (superscript II, Invitrogen) and PCR was performed with the following primers amplifying the neck/CRD region. SP-D f1 , 5'-ATGTTGCTTCTCTGAGG-3 (SEQ ID No. 5), SP-D r2, 5^'-TCAGAACTCGCAGACCACAAGA-3' (SEQ ID No. 6) (43 cycles), actin rew2 5'- AGTCCGCCTAGAAGCATTTG-3^' (SEQ ID No. 7), and actin forw, 5^'-ATGCAGAAGGAGATCACTGC- 3^' (SEQ ID No. 8) (24 cycles). Annealing temperature was 55°C. The reaction mixtures were run on a 1 % agarose gel and bands of the correct size were isolated for direct sequencing.

Western blotting and Immunohistochemistry

Western blotting of tissue homogenates or lysed cell culture was performed essentially as described in Madsen et al., 2000 [M adsen J, et al. (2000) Localization of lung surfactant protein D on mucosal surfaces in human tissues. J Immunol. 164: 5866] on 8-25% (w/v) polyacrylamide gradient gels. Primary anti SP-D antibodies were: mouse monoclonal antibody, Hyb245-2 or 246-4, 1 μg/ml or polyclonal rabbit K477 IgG 10 μg/ml. Secondary antibodies were alkaline-phosphatase-coupled rabbit anti-mouse IgG (D0314, Dako) or alkaline phosphatase-conjugated goat anti-rabbit IgG (A-8025, Sigma). Cell cytospins or tissue sections were mounted on coverslips and processed as described in Holmskov et al., 1999 [H olmskov U, et al. (1999) Cloning of gp-340, a putative opsonin receptor for lung surfactant protein D. Proc Natl Acad Sci U S A. 96: 10794]. The monoclonal antibodies directed against SP-D (Hyb245-1 , Hyb 245-2) were produced and characterized as described in Crouch et al. [Crouch E et al. (1993) Accumulation of surfactant protein D in human pulmonary alveolar proteinosis. Am J Pathol. 142: 241]. Normal human tissues were from the tissue bank at the Department of Pathology, Odense University Hospital. Monoclonal anti-CD34 and anti-CD31 antibodies (DAKO) were used as positive controls for endothelial cells.

Purification of human amnion fluid SP-D

Human SP-D was isolated from amniotic fluid by maltose-agarose (Sigma) affinity chromatography (Pharmacia Biotech) as described by Strong et al. [S trong P, et al. (1998) A novel method of purifying lung surfactant proteins A and D from the lung lavage of alveolar proteinosis patients and from pooled amniotic fluid. J Immunol Methods. 220: 139]. Expression and purification of murine recombinant SP-D (neck/CRD)

The sequence encoding the alpha-helical coiled-coil neck region and the carbohydrate recognition domain of SP-D was amplified by PCR using a TA-cloned cDNA template made from lung RNA. Primers used for amplification were 5'-CGTATCTACGTAGAGGTCAATGCTCTCAGGC-3' (SEQ ID No. 9) and 5'-CGTATCCCT-AGGTCAGAACTCGCAGATCACGA-3' (SEQ ID No. 10). The PCR product was ligated into the pPIC9K vector (Invitrogen). A purified linearized plasmid was electroporated into competent Pichia pastoris (GS115) and transformants carrying multiple inserts were isolated according to the manufacturer's manual.

Expression was carried out at 30°C for 4 days. A supernatent filtrate was mixed with SP-sepharose. The matrix was loaded onto a FPLC-column and washed with 400 ml buffer containing 50 mM citric acid, 0.5 mM EDTA and 0.01% Tween 20 (pH 3.5). Protein was eluted in a similar buffer gradually increasing the salt concentration to 1 M NaCI. The eluate was dialyzed into TBS, 5 mM CaCI₂, 0.01 % Tween 20 (pH 7.5) and loaded onto a 10 ml agarose-maltose FPLC-column (Sigma). Finally, the protein was eluted with TBS, 10 mM EDTA, 0.01% Tween 20.

The recombinant protein showed a reduced size of 17 kDa when analyzed by SDS-PAGE, and was recognized in Western blotting using a polyclonal antibody raised against native murine SP-D. Analyzed by gel filtration chromatography the protein showed a size of 45 kDa corresponding to a trimerised molecule.

Preparation of LDL and oxLDL

Blood samples were drawn from normal 12 hours fasting human donors. LDL was isolated from heparin-plasma by potassium-bromide (KBr) density gradient ultracentrifugation. The LDL fraction was defined as the yellow lipid band appearing in the KBr gradient after 24 hours centrifugation at 141 ,000xg (4°C) between the 1.019 g/ml and 1.063 g/ml KBr-density layers. The LDL was used for experiments within one month after preparation. LDL was aspired and incubated for 24 hours (37°C) with 10 μM CuS0₄. The copper induced oxidation was terminated by the addition of 20 μM EDTA. All procedures were performed using sterile technique. The protein concentration of the oxLDL preparation was measured using the Bio-Rad Protein Assay (Bio-Rad Laboratories). Lipid peroxidation was quantified by the measurement of TBARS (thiobarbituric acid reactive substances). Peroxidation of fatty acids results in the formation of malondealdehyde (MDA) which can be measured by the reaction with TBA to a defined chromogenic condensation product (TBA-MDA). The assay was performed essentially as described in Pyles et al. [Pyles LA, et al. (1993) Spectrophotometric measurement of plasma 2-thiobarbituric acid-reactive substances in the presence of hemoglobin and bilirubin interference. Proc Soc Exp Biol Med. 202: 407] with a 3-fold down-scalling of volumes. Each sample was assayed once whereas standards were assayed in triplicate. The MDA standard was produced by HCl hydrolysis of MDAbis-diethylacetal (Sigma) stopped by buffering the solution to pH 6.4. The final MDA concentration in molar was determined by 1/13700*OD245. The utilized oxLDL contained 0.29 nmol MDA/mg protein.

Assessment of LDL Oxidation by TBARS in vitro

Reaction mixtures composed of 150 μg/ml LDL, 10 μM CuS0₄, and various concentrations of human amnion fluid SP-D were prepared in 0.9% saline. The protein concentrations of LDL, SP-D, C1q, and in serum were estimated by OD₂₈₀. The mixtures were incubated at 37°C for 4 hours. TBARS were measured using the method described in Bridges et al., 2000 [Bridges JP, et al. (2000) Pulmonary surfactant proteins A and D are potent endogenous inhibitors of lipid peroxidation and oxidative cellular injury. J Biol Chem. 275: 38848]. C1q was a gift from Jens Christian Jensenius, Department of Medical Microbiology and Immunology, University of Aarhus, Denmark.

Animal studies, plasma TBARS and TAS-assay

Homozygous SP-D-/- mice were obtained from professor Samuel Hawgood, Department of Pediatrics, University of California. The corresponding C57BL/6NCrlBr background wildtype mice were obtained from Charles River Sweden. The National Animal Ethics Committee approved the study and all procedures.

In one experiment SP-D-/- and wildtype mice were separated into two groups and given either normal chow or a high fat enriched diet (Harlan TD 88051 , Harlan Netherlands B.V.) for 5 days. In the second experiment mice were fed on the high fat diet for 1 month. The mice received an overdose IP mebumal followed by exanguination by cardiac puncture and the blod was drawn into heparinized tubes. The plasma was separated and used for both TBARS and total antioxidant assay (TAS assay). Plasma TBARS assay was performed essentielly as described for oxLDL. Each sample was assayed once whereas standards were assayed in triplicate. The MDA standard was performed by HCl hydrolysis of MDAbis-diethylacetal (Sigma) stopped by buffering the solution to 6.4. The final MDA concentration in molar was determined by 1/13700*OD245. TAS was determined with the kit from Randox Laboratories GmbH, Germany, according to the manufacturers directions. Each sample was assayed once whereas standards were assayed in duplicate.

Endotoxin measurement

The amount of endotoxins present in the protein and lipid preparations was measured using the limulus amebocyte lysate QCL-1000 assay (Bio-whittaker). Protein and lipid preparations were accepted for the further use in live animals or cell culture if the final endotoxin concentration did not exceed 0.5 EU/ml.

Apoptosis assay and flow cytometry

HUAEC cultures were prepared and maintained as described above. For apoptosis assays 100,000 cells/sample were seeded in 1.5 cm gelatine coated wells in RPMI 1640 (Invitrogen) supplemeted with glutamine, penicillin/streptomycin and 10% Albumax I (Invitrogen). After over night attachment the cells were treated with either vehicle (containing Cu²⁺ and EDTA corresponding to the oxLDL solution), 10- 30 μg/ml oxLDL, 1 μg/ml or 10 μg/ml mouse recombinant neck-CRD SP-D (mrSP-D), 1 μg/ml human IgG (KABI), or 10 μM camptothecin (Sigma), alone or in combination. After approximately 24 hours the cells were very carefully trypsinized, washed and labelled with 5 μl propidium iodide (PI) staining solution and 5 μl Annexin-V-FITC (Pharmingen International) according to the manufacturers directions. The binding of FITC-labeled annexin-V and PI to endothelial cells was quantitated by flow cytometry using a FACSCalibur flow cytometer (Becton Dickinson) and CELLQuest software.

Statistics

Statistics were performed with Intercooled Stata version 7.0.

Results

SP-D localisation to vascular endothelial cells

A panel of 9 monoclonal antibodies directed against human SP-D and was used for immunohistochemical localization of SP-D. Ultimately one monoclonal antibody designated Hyb 245-1 was used for tissue localization. The antibody Hyb 245-1 reactivity could be found in epithelial cells in a variety of tissues (Madsen et al. J Immunol, 2000). Enhancement of antigen recognition was obtained by micro-wave treatment of the specimen. The improved immunohistochemical technique allowed detection of SP-D localization in vascular endothelial cells through out all examined tissues. Hyb 245-1 reacted with endothelial cells in artheries and veins in microvasculature as well as in macrovasculature. SP-D immunoreactivity was further detected in high endothelial venules of the tonsils (Fig. 9). SP-D expression by endothelial cells in vitro

Primary human umbilical arthery endothelial cell lines (HUAEC) were established to study endothelial cell SP-D expression. HUAEC cytospin preparations reacted with Hyb 245-1. The immunostaining was highly granular indicating that the intracellular localization of SP-D is vesicular. The primary endothelial cells co-stained for endothelial markers CD31 and CD34 (Fig. 10). In the 7th - 8th passage the endothelial cell cytospin preparations reacted with Hyb 245-1 although the immunoreactivity was significantly diminished in parallel with the endothelial cell marker CD31 (data not shown). The cultures sustained the expression of the endothelial cell marker CD34 throughout all the investigated passages. The loss of SP-D protein expression was suspected to be due to cell culture senescence.

Whole artheries were isolated from human umbilical cords. RT-PCR analysis amplifying the SP-D neck-CRD region and successive sequencing of the PCR product demonstrated the synthesis of SP-D polyadenylated RNA in the vascular tissue. The corresponding fragment could be amplified from HUAEC cultures. The SP-D mRNA rate of synthesis was accelerated by the implication of increasing seeding densities of the HUAEC culture. The PCR reaction was normalised to the β-actin mRNA expression (Fig. 10). RT-PCR analysis amplifying the full length SP-D transcript showed identically sized mRNA from HUAEC culture as from the lung (data not shown).

Western blotting of lysates prepared from HUAEC culture demonstrated that the monoclonal anti human SP-D antibody Hyb 245-2 recognised endothelial SP-D protein. The fully reduced endothelial protein showed slightly altered mobility in the SDS-PAGE compared to the 43 kD SP-D polypeptide monomers of SP-D protein purified from amnion fluid and no trimeric form was detected in the unreduced sample (Fig. 10). Corresponding Hyb 245-2 reactivity was observed in lysates of human umbilical artherial and vein tissue (data not shown).

Human SP-D inhibition of LDL oxidation

Rat SP-D and rat SP-A inhibition of copper induced LDL oxidation estimated by the accumulation of TBARS has previously been reported by Bridges et al. [Bridges JP, et al. (2000) Pulmonary surfactant proteins A and D are potent endogenous inhibitors of lipid peroxidation and oxidative cellular injury. J Biol Chem. 275: 38848]. In a similar experimental setup we demonstrated a dose dependent inhibition of copper induced LDL oxidation by human SP-D purified from amnion fluid. 20 μg/ml purified human SP-D fully supressed the development of TBARS in these experiments whereas 1 μg/ml showed no effect. Human serum was used as a positive control and gave the expected inhibition at 100 μg/ml. C1q required 500 μg/ml for inhibition to occur.

Thus, human SP-D appears to function as a high potent inhibitor of LDL oxidation (Fig. 11). Plasma oxidational status in SP-D deficient mice

The TBA reaction with MDA is a widely employed means of assessing lipid peroxidation in biological systems. Homozygous C57BL/6 SP-D-/- mice were obtained from professor Samuel Hawgood, University of California. One group of SP-D-/- mice and one group of C57BL/6 wildtype mice were put on a high fat diet for one month before exanguination.

The resulting plasma TBARS levels were correlated to an MDA standard. The TBARS levels in the plasma of SP-D-/- mice reached a mean of 1 ,09 (95% Cl 0.89-1.30) nmol/ml MDA whereas the TBARS level in wildtype mouse plasma was estimated to 0,78 (95% Cl 0.61-0.96) nmol/ml MDA. The TBARS levels in the two genotypes differed significantly with a p-value of 0,005 estimated by two-sample Wilcoxon rank-sum (Mann-Whitney) test. The plasma of wildtype mice contained 71 ,6% of oxidized groups compared to SP-D deficient plasma (Fig. 12).

The TBARS assay was performed in two independent experiments. A significant difference between TBARS plasma levels of the two genotypes was exclusively observed in mice on high fat diet. No significant differences were observed between genders. One plasma sample was in addition assayed with and without the addition of 1 μg/ml human recombinant SP-D. Addition of SP-D to the sample did not decrease the TBARS level (data not shown).

Plasma total antioxidant status in SP-D deficient mice

A commercially available kit from Randox Laboratories was used to assess total antioxidant status in the plasma of SP-D deficient and wildtype mice by its ability to supress the formation of radical cations. ABTS (2,2'-Azino-di-[3-ethylbenzthiazoline sulphonate]) is incubated with a peroxidase and H₂0₂ to produce the cation ABTS*⁺ and with plasma added. Antioxidants in plasma may inhibit the formation of the cation. The assay was performed in two independent experiments with mice on high fat diet. We found plasma TAS comparable to the levels reported for mice elsewhere, Bentzon et al. [Bentzon JF, et al. (2001) Red wine does not reduce mature atherosclerosis in apolipoprotein E- deficient mice. Circulation. 103: 1681]. No significant differences were found between the plasma TAS of SP-D deficient mice and wildtype mice (Table 4). Likewise, we found no significant differences in plasma TAS between genders. Plasma TAS was also estimated in mice on regular chow. The TAS levels were somewhat lower than in mice fed on the high fat diet. The wildtype mice showed 0,54 mM antioxidant in plasma (95% Cl 0.42-0.66 mM) and SP-D-/- mice showed 0,55 mM antioxidant in plasma (95% Cl 0.42-0.62 mM).

Table 4. Plasma TAS values in two independent experiments. The TAS values are means measured as mM antioxidant. Numbers in sqare brackets indicates the 95% confidens interval, n denominates the number of mice in each group. SP-D-/- Wildtype p-value

Experiment 1 5 75 0\88 δTΪ4

[0,67-0,85] [0,72-1 ,03] n=8 n=8

Experiment 2 0^3 5\83 0^49

[0,76-1 ,10] [0,55-1 ,10] n=18 n=7

Apoptosis assay and flow cytometry

OxLDL is previously reported to induce endothelial cell apoptosis [Dimmeler S, et al. (1997) Oxidized low-density lipoprotein induces apoptosis of human endothelial cells by activation of CPP32-like proteases. A mechanistic clue to the 'response to injury' hypothesis. Circulation. 95: 1760]. Apoptotic and necrotic cells expose phosphatidylserine (PS), which is normally present on the inner cell membrane leaflet, to the outer leaflet. The presence of PS on the cell surface allows annexin-V binding. PI stains primary necrotic cells or cells becoming necrotic after apotosis. Cells in the experiments were co-stained with Annexin-V-FITC and PI. A rather high background of PI fluorescent cells existed in all samples possibly due to trypsinization procedures of the attaching cells prior to the staining. PI staining was thus disregarded.

Fig. 13 shows Annexin-V-FITC staining of primary endothelial cells (passage 5) with apoptosis induced by either vehicle (Cu²⁺/EDTA control), LDL, oxLDL, or the positive control for induction of apoptosis camptothecin. Cells were coincubated with either mrSP-D or the negative control for SP-D treatment IgG.

Vehicle (PBS with Cu²⁺ and EDTA corresponding to the concentrations present in 25 μg/ml oxLDL) treatment induced 2 populations of annexin-V binding cells: One population showing low annexin-V staining with a mean FITC flourescence (FL1-H) of 15, which is suggested to represent the normal cells, and a high annexin-V binding population (37%) with a mean flourescence of 361 (Fig. 13a). The latter population is suggested to represent the level of apoptotic/necrotic cells inherent to the assay.

10 μg/ml oxLDL or 10 μg/ml LDL resulted in similar distributions of the two populations and the high annexin-V binding population included 40-41% of the cells. The effect of costimulation with 1 μg/ml mrSP-D was undetectable at this low level of stimulation (Fig. 13b and 13e). 25 μg/ml LDL induced a population of high annexin-V binding cells including 49% of the cells, which was reduced to 25% by costimulation with 1 μg/ml mrSP-D (data not shown).

Stimulation of the HUAEC's with either 25 μg/ml oxLDL or 5 μM camptothecin increased the high annexin-V binding population from 37% in the vehicle treated sample to 64% and 72%, respectively. The low annexin-V binding populations were reduced from 48% to 8% and 6%, respectively. Costimulation with mrSP-D resulted in an increase in the low annexin-V binding populations to 14% of the oxLDL treated cells (1.75 fold increase) and to 20% of the camptothecin treated cells (3.3 fold increase compared to vehicle treated). Costimulation with IgG and oxLDL resulted in a low annexin-V binding population including only 3% (Fig. 13c and 13f). This indicated reversal of the oxLDL induced apoptotic process specific to SP-D treatment.

Interestingly, the mean flourescence of the high annexin-V binding population was reduced from 361 to 221 by the mrSP-D treatment for the oxLDL treated sample and from 350 to 143 for the camptothecin treated sample by the mrSP-D treatment.

The effect of IgG coincubation with camptothecin or oxLDL treated cells was strikingly different from the effects of mrSP-D. IgG coincubation with camptothecin resulted in a depression of mean flourescence as seen for mrSP-D. In contrast, coincubation with oxLDL resulted in a similar distribution of annexin-V binding as with oxLDL alone demonstrating that the reduction in annexin-V binding was mrSP-D specific.

In a second experiment the dose effects of oxLDL and mrSP-D were studied. Primary endothelial cell cultures were prepared from the same donor cord and used in passage 5. The same batch of oxLDL was likewise used in both experiments. Treatment with 20 and 30 μg/ml oxLDL did not differentiate the annexin-V binding and resulted in similar distributions of annexin-V-binding. In contrast, the costimulation with mrSP-D differentiated the annexin-V binding induced by the two oxLDL doses. In these experiments the low annexin binding population of cells was increased from 4 % to 7 %, 19 %, and 24 % with the increasing (0.1 , 1 , and 10 μg/ml) mrSP-D doses in the 20 μg/ml oxLDL treated samples and to 6 %, 13 %, and 8 % in the 30 μg/ml oxLDL treated samples.

Coincubation with 20 μg/ml oxLDL and the increasing (0.1 , 1 , and 10 μg/ml) mrSP-D doses resulted in gradual decrease of mean fluorescense values in the high annexin-V binding population of 176, 93, and 75, respectively, whereas the coincubation between 30 μg/ml oxLDL and the increasing mrSP-D doses resulted in the higher mean fluorescence values of 204, 129, and 98, respectively (Fig. 14a and 14c).

The depression of mean fluorescence of apoptotic cells treated with mrSP-D could be the result of a reduction in PS sites available for annexin-V binding either due to competition between mrSP-D and annexin-V binding or due to SP-D inhibition of the apoptotic process. Yet, depression of the mean flourescense induced by camptothecin in the high annexin-V binding population was also observed with IgG (Fig. 13f) and depression was not observed in the high annexin-V binding population of cells in the vehicle treated samples costimulated with mrSP-D (Fig. 14b and 14d). Therefore, these studies suggest that mrSP-D inhibits the apoptotic process. Example 5: Surfactant Protein D modulation of mouse lipid homeostasis and body weight

Abbreviations

ABC1 , ATP-binding casette transporter-1 ; AUC, area under the curve; CETP, cholesteryl ester transfer protein; EDTA, ethylenediaminetetraacetic acid; ELISA, enzyme linked immunosorbent assay; EU, endotoxin units; GM-CSF, granulocyte-macrophage colony stimulating factor; HDL, high density lipoprotein; HDL-C, high density lipoprotein cholesterol; LDL, low density lipoprotein; LDL-C, low density lipoprotein cholesterol; SP-D-/-, SP-D gene knock out; SP-D, surfaktant protein D.

Animal studies

Homozygous SP-D-/- mice were obtained from professor Samuel Hawgood, San Francisco, California, USA. The corresponding C57BL/6NCrlBr wildtype mice were obtained from Charles River, Sweden. The study and all procedures were approved by the National Animal Ethics Committee.

The newborn mice had access to Altromin 1314 diet (Brogarden Aps, Demark). The lighting schedule was 12 hours light and 12 hours dark. Mice were fed and given water ad libitum.

Blood samples were drawn by puncture of the orbital venous plexus into EDTA-containing tubes from non-fasted animals unless otherwise is stated.

Mouse lipid-feeding experiment

Five weeks old wildtype and SP-D-/- mice were fed either a normal chow (Harlan 2018) or a high-fat diet (Harlan diet TD 88051) containing 15.8% (w/w) fat, 1.25% cholesterol (w/w), and 0.5% (w/w) sodium cholate. All animals were weighed weekly. A total of 164 animals were initially included in the experiment. During the time-span of the experiment 9 SP-D-/- mice and 21 wildtype mice died or were terminated due to failure to thrive, infections, fights etc.

Halflife of recombinant human SP-D in SP-D-/- mouse plasma

The halflife of the human recombinant SP-D (the neck and carbohydrate recognition domains) in SP-D- /- mouse plasma was estimated by enzyme linked immunosorbent assay (ELISA) technique. Eight 3-4 months old SP-D-/- mice were included (four females + four males). Each mouse recieved one bolus of 90μg/250 μl sterile isotonic sodium chloride solution injected in the tail-vein. The recombinant SP-D was kindly provided by Professor K. B. Reid, Oxford, England. The endotoxin level in the SP-D infusion was < 0.2 EU/ml measured by the Limulus Amebocyte Lysate QCL-1000 assay, Biowhittaker, Walkersville, USA. Plasma samples were frozen before the immunoassay was performed. Treatment of mice with human recombinant SP-D

5 weeks old SP-D-/- female mice were fed the high-fat diet for 5 days. One group of mice recieved 9 μg human recombinant SP-D/280 μl sterile isotonic sodium chloride by daily tail-vein infusions. The endotoxin level in the SP-D infusion was < 0.02 EU/ml. Another group received daily 9 μg human serum albumin/280 μl sodium chloride solution. The serum albumin was prepared under endotoxin free conditions for human infusion and kindly provided by Claus Koch, Statens serum Institute, Denmark. The 2 groups of mice were terminated by anaesthesia (100 mg/kg phenobarbital in the peritoneum) and exsanguination. Blood for lipid analysis was drawn from the right ventricle of the anestesized animals into EDTA containing tubes.

ELISA technique

The immunoassay was performed as described in example 1. Parrallel curves were obtained between the human recombinant SP-D and native human SP-D purified from amnion fluid. EDTA-plasma samples were diluted a minimum of 10 times in a Tris-buffer with 0.05% Tween 20 and 5 mM calcium chloride before testing.

Mouse plasma lipids

Lipids in mouse EDTA plasma was measured by a Cobas Mira instrument (Triolab A/S, Copenhagen, Denmark).

Triglycerides were measured with ABX Diagnostics Triglycerides 100. Total plasma cholesterol was measured with ABX Diagnostics Cholesterol reagents. LDL-C and HDL-C were measured with ABX Diagnostics LDL Direct and ABX Diagnostics HDL Direct reagents, respectively. All assays were performed according to the manufacturers directions (Triolab A/S, Copenhagen, Denmark). Plasma samples were used undiluted or diluted 2-3 times in 0.9% sodium chloride.

Human serum SP-D concentrations succeding a high-fat meal

Four non-smoking men aged 25-33 fasted for 12 hours before the ingestion of a high-fat meal. Following, conseccutive blood samples were drawn during a 5 hours period. A control experiment with no food intake was performed one month later. The high-fat diet comprised 6 slices of smoked bacon, 400 g egg, 50 g heavy cream, 2 teaspoons of butter, 1 slice of white bread, and 2 glasses of whole milk per subject. According to the table of Cholesterol Content of Common Foods by Pedriatrics online Medical college of Georgia the meal contained approximately 1800 mg cholesterol. The cholesterol amount is 6 times the maximum amount recommended by The American Hearth Association for daily consumption by healthy persons (300 mg/day). Serum SP-D was estimated by ELISA technique. Statistical analysis

The high-fat diet the mice recieved was hepatotoxic and all cholesterol readings from yellowish plasma were excluded due to false high readings in the colorimetric assays. Two very low cholesterol readings were excluded. Wilcoxon rank-sum tests were used to compare plasma lipid concentrations between experimental groups.

Analysis of variance and regression analysis were utilised to describe and compare mouse body weights. The utilised statistical software was Intercooled Stata version 7.0.

Human serum SP-D concentrations would apply to the normal distribution by taking the natural logaritm of the values. Paired t-tests were used to compare logarithmically transformed serum SP-D concentrations measured at different time-points in the same population of human subjects. For this purpose Microsoft Excel software was used.

Results

Plasma lipid levels in SP-D-/- and wildtype female mice on high-fat diet

This experiment was focused on the lipid metabolism in female mice on the high-fat diet. Such mice are previously described to develop lipid profiles which resemble the human lipid compositions in as much as they can develop relatively high total cholesterol and relatively low HDL-C. In contrast, male mice are less sensitive towards diet manipulations of their lipid homeostasis.

Time series of analysis of the plasma lipid concentrations induced by the high-fat diet in the female mice are shown in Table 5. Plasma lipids were recorded for 166 days. The studies included 22-34 mice in each group.

Table 5. Plasma lipid concentrations in female mice on high-fat diet. Triglyceride, total cholesterol, LDL-C, and HDL-C concentrations are measured in a series of days after diet start. Concentrations are given in mmol/l and are stated as means ± standard deviation (std.dev.). An asterics indicates a significant (p < 0.05) difference between genotypes. N = 22-34 mice. a) Triglycerides:

As previously described, the high-fat diet decreased plasma triglyceride and HDL-C levels in wildtype mice and increased total cholesterol and LDL-C levels compared to the normal chow (day 0).

The SP-D-/- genotype induced higher plasma triglyceride levels compared to wildtype mice. The significant difference was detected after 12 days on the diet and was sustained throughout the rest of the study. The female genotype difference in triglycerides was on average 0.12 mmol/l corresponding to 30% of the wildtype concentrations measured from day 12 to day 166 of the high-fat diet.

Total cholesterol and LDL-C concentrations were significantly increased in the SP-D-/- females in a transient period measured 5 days after the high-fat diet start. The SP-D-/- female cholesterol concentrations were increased by 0.93 mmol/l corresponding to 19% of the wildtype level and the SP- D-/- female LDL-C concentrations were increased 0.20 mmol/l corresponding to 16% of the wildtype level.

HDL-C concentrations were constitutively increased in the female SP-D-/- mice compared to wildtype mice. Apparently the HDL-C level was not altered from day 0 to day 5 by the high-fat diet (p = 0.13) in the SP-D-/- female mice whereas the levels decreased in the wildtypes (p= 0.005). The female genotype difference in HDL-C was on average 0.28 mmol/l corresponding to 29 % of the average wildtype level during the high-fat feeding period.

Plasma lipid levels in SP-D-/- and wildtype female mice on normal chow

Control experiments were set up including SP-D-/- mice and wildtype mice of both genders and on normal chow. The control experiments included 8-12 animals in each group. The lipids were examined on feeding day 5, 26, 54 and 82.

The control studies indicated that SP-D-/- females on normal chow showed significantly higher HDL-C levels compared to the wildtype females whereas the other lipid fractions were unaffected by the genotype (Table 6). The significant difference in HDL-C was found from day 5 after the transistion from one normal chow to another and untill day 54 when the mice were 13 weeks old. The average difference was 0.17 mmol/l corresponding to 12% of the average wildtype HDL-C level in the period.

Table 6. Plasma lipid concentrations in female mice on normal chow. Triglyceride, total cholesterol, LDL-C, and HDL-C concentrations are measured in a series of days after diet start. Concentrations are given in mmol/l and are stated as means + standard deviation (std.dev.). An asterics indicates a significant (p < 0.05) difference between genotypes. N = 10-11 mice. a) Triglycerides

Plasma lipid levels in SP-D-/- and wildtype male mice on high-fat diet

The triglyceride levels were significantly increased in the SP-D-/- male mice compared to wildtype male mice on high-fat diet. Whereas female triglyceride levels were relatively increased from the 12th day of diet the male triglyceride levels were not increased untill feeding day 82. Total cholesterol was relatively increased in SP-D-/- males compared to wildtype males on feeding day 5 in parrallel with the female mice and HDL-C was significantly increased in the SP-D-/- animals day but significance was only obtained on feeding 26 and day 54. The most prominant differences between female and male mouse plasma lipids on high-fat diet was the relatively high female total cholesterol compared to the male cholesterol. The total cholesterol was on average 8.02 mmol/l in females versus 6.66 mmol/l in males for SP-D-/- mice and 7.63 mmol/l versus 6.83 mmol/l for wildtype mice, correspondingly. In addition, female HDL-C levels lower were than the corresponding male levels. The HDL-C concentration was on average 1.22 mmol/l in females versus 1.97 mmol/l in males for SP-D-/- mice and 0.95 mmol/l versus 1.60 mmol/l for wildtype mice, correspondingly (data not shown).

Plasma lipid levels in SP-D-/- and wildtype male mice on normal chow

SP-D-/- male mice on normal chow developed a transient increase in triglyceride, total cholesterol and LDL-C concentrations compared to wildtype males (Table 7). The increases were measured day 5 after the transistion from one normal chow to another. Total cholesterol was relatively increased in SP- D-/- males compared to wildtype males throughout the rest of the study but significance was only obtained on day 82. This difference was 0.51 mmol/l which correponded to 18% of the wildtype level. SP-D-/- males mice showed a significant decrease in LDL-C levels compared to wildtype males measured feeding day 54. Due to the rather small difference (0.02 mmol/l) and the lack of concensus with the rest of the feeding study observations, this difference was disregarded. Overall differences between female and male mice on normal chow included relatively higher levels of triglyceride, total cholesterol and HDL-C concentrations in the males and relatively lower LDL-C concentrations.

Table 7. Plasma lipid concentrations in male mice on normal chow. Triglyceride, total cholesterol, LDL- C, and HDL-C concentrations are measured in a series of days after diet start. Concentrations are given in mmol/l and are stated as means ± standard deviation (std.dev.). An asterics indicates a significant (p < 0.05) difference between genotypes. N = 8-12 mice.

a) Triglycerides

Human Recombinant SP-D halflife in SP-D-/- mice

Eight mice received one bolus of 90 μg of human recombinant SP-D injected intravenously. Blood samples were drawn during a period of 30 hours. SP-D concentrations were measured by ELISA. The assay did not react to plasma from SP-D deficient mice. The obtained mean plasma concentrations are shown in Fig. 15. The initial mean concentration measured after 1 hour was approximately 8 μg/ml plasma and reflected that redistribution had occured at this time-point. The declining plasma ccoonncceennttrraattiioonn wwaass ddeessccrriibbfed by the equation (SP-D) ng/ml = 8145.9e^{"00238 x hours} coresponding to a mean halflife of 29.1 hours.

Kinetic measurements were performed in order to enable control of the plasma concentration of human recombinant SP-D injected into SP-D-/- mice. The plasma concentrations were measured on a daily basis in two mice after injection of 9 μg SP-D/day in 5 days. The concentrations oscillated around an average of 2513 ng SP-D/ml plasma in both animals (95% confindens interval; 1719 - 3306 ng/ml) (data not shown).

Treatment of SP-D-/- female mice on high-fat diet with human recombinant SP-D

The high-fat diet was fed to SP-D-/- female mice recieving 5 daily intravenous infusions of 9 μg human recombinant SP-D and to SP-D-/- female mice receiving 5 daily intravenous infusions of 9 μg human serum albumin. Boxplots of the measured plasma lipid concentrations are shown in Table 8.

Table 8. The average concentrations of plasma total cholesterol, LDL-C, and HDL-C measured in SP- D-/- females treated with either human serum albumin (HSA) or human recombinant SP-D (rSP-D). SP-D-/- mice were subjected to 5 days of high-fat diet and 5 daily intravenous infusions of 9 μg HSA or 9 μg rSP-D. Concentrations are given in mmol/l and are stated as means + standard deviation (std.dev.). Wilcoxon ranksum test determined the P-value. N = 10-24 mice.

a) Triglycerides

Mice recieving SP-D showed a significant decrease in plasma cholesterol levels relative to the albumin treated group. There was no significant alteration of triglyceride levels. The relative difference in total cholesterol, LDL-C and HDL-C between SP-D-/- mice recieving SP-D or albumin mice constituted 20 %, 22 % and 17 %, respectively.

In comparison, there was no significant difference in triglyceride concentrations whereas the relative difference in total cholesterol, LDL-C and HDL-C between untreated wildtype mice and untreated SP- D-/- mice constituted 19 %, 16 % and 28, respectively, on the fifth day of diet.

Mouse body-weights

The mouse body-weights were recorded for 24 weeks. Fig. 16 shows the average body-weights of female and male mice. Table 9 summarizes the weekly weight gains for the individual feeding groups. Female and male mice reacted oppositly to the high-fat diet. The high-fat diet increased the weight gain per week in female mice compared to the normal chow. In contrast, the male mice gained weight at a relatively slower rate on the high-fat diet compared to the normal chow. SP-D deficiens significantly elevated body-weights for both genders. Table 9. Weekly weight gains of mice: The table shows the mean weight gain (g/week) and 95 % confidens interval (g/weak) for mice in the feeding experiment separated in 8 different groups according to gender, genotype and diet, n = the total number of serial measurements, p-values are obtained from t-test comparing the means by genotype.

Group n mean [95 % confidens interval] p-value

Female, wildtype, normal chow 300 0.39 0.37 - 0.40 0.1973

Female, SP-D-/-, normal chow 275 0.37 0.36 - 0.39

Female, wildtype, high fat diet 275 0.46 0.43 - 0.49 0.0004

Female, SP-D-/-, high fat diet 250 0.53 0.51 - 0.56

Male, wildtype, normal chow 225 0.62 0.59 - 0.65 < 0.0001

Male, SP-D-/- normal chow 275 0.71 0.69 - 0.73

Male, wildtype, high fat diet 200 0.34 0.30 - 0.37 < 0.0001

Male, SP-D-/-, high fat diet 225 0.59 0.57 - 0.61

SP-D-/- female weight gain exceeded the wildtype by 0.07 g/week on the high-fat diet and the total body-weight was increased by 5.9 % after 24 diet weeks.

SP-D-/- male weight gain exceeded the wildtype by 0.09 g/week on the normal chow and by 0.25 g/week on the high-fat diet and the total body-weight was correspondingly increased by 16.4 % and 23.0 % after 24 weeks. Thus, SP-D deficient male mice fed the normal chow gained the heaviest body- weight whereas wildtype male mice fed the high-fat diet gained the leanest body-weight in this study.

Patoanatomical examination

Mouse lungs, liver and gallbladder were HE-stained and microscopically examined for patoanatomical changes after the diet experiment.

The SP-D-/- mice showed the previously described lung phenotype with the appearance of foam cells in the alveoli and increased airspace. Pulmonary foam cells were likewise detected in wildtype mice on the high-fat diet.

Liver anatomi appeared normal on the normal chow in both genotypes. On the high-fat diet, livers showed extensive infiltration with lymfocytes and fat. Fat droplets were microvesicular. The gall bladder epithelium appeared normal in the both genotypes on the normal diet, whereas hyperplasia and metaplasia appeared on the high-fat diet. All mice on the high-fat diet developed gallstones.

Human serum SP-D concentrations succeding a high-fat meal

Four persons were fed a cholesterol rich meal (1800 mg/person) after overnight fasting and the serum SP-D concentrations were then followed in a 5-hour period. As a control study serum SP-D was measured in the same persons in a 5-hour period without giving a meal. Fig. 17 shows the variation in serum SP-D concentrations. Serum SP-D was decreasing during the fasting period. The significant decrease (p = 0.008) amounted to 28 ng/ml/hour corresponding to 3.5 %/ml/hour. In contrast, transient depression of the SP-D serum concentration was observed after lipid ingestion. The transient decrease was detected between 90 minutes and 3 hours after the ingestion of the high-fat meal (p < 0.05). The maximal depression was detected after 1 hour and 30 minutes where serum SP-D fell from an average of 741 ng/ml to 543 ng/ml corresponding to 27 %.

In a separat study, variations in serum SP-D concentrations were measured with 30 minutes intervals and showed that the average serum SP-D did not decrease significantly after the intake of regular breakfast (data not shown).

Discussion

In this study, SP-D-/- mice and the corresponding wildtype C57BL/6N mice were fed either a normal chow or a high-fat diet and the plasma triglyceride, total cholesterol, LDL-C and HDL-C were measured in a time-course. The serial measurements were compared in groups separated by gender, diet, and genotype.

The high-fat diet increased total cholesterol and LDL-C levels and decreased triglyceride and HDL-C levels in the wildtype mice as previously observed [Srivastava RA, et al. (2001) Dietary cholate increases plasma levels of apolipoprotein B in mice by posttranscriptional mechanisms. Int J Biochem Cell Biol. 33: 1215, Srivastava RA, et al. (2000) Dietary cholic acid lowers plasma levels of mouse and human apolipoprotein A-l primarily via a transcriptional mechanism. Eur J Biochem. 267: 4272].

In general, the SP-D-/- mice showed either elevated or similar lipid levels compared to the wildtype.

The high-fat diet induced relatively higher triglyceride levels in female and male SP-D-/- mice compared to wildtype mice. The significant differences were recorded after 12 days in females and after 82 days in the males. Transistion from the normal chow to the high-fat diet lead to at faster increase in SP-D-/- mice than in wildtype mice in total cholesterol for both genders and in LDL-C for female mice. The relative difference between the actual levels was significant on the fifth diet day and was dissapeared on the twelfth diet day. SP-D-/- genotype specific levels of total cholesterol were likewise found in male mice on normal chow.

HDL-C was constitutively increased in SP-D-/- females on the high-fat diet. Significant HDL-C genotype differences were also recorded in SP-D-/- mice of both genders on the normal chow and in males on the high-fat diet. No other genotype differences were recorded in the females on normal chow.

As stated above, transition from normal chow to high-fat diet led to faster increasing levels of total cholesterol and LDL-C in SP-D-/- than in wildtype female mice. This difference was recorded on the fifth diet day. HDL-C levels were also relatively increased in SP-D-/- female mice on this day, whereas no difference was observed in triglyceride levels between genotypes. These "day 5 lipid profiles" were used as endpoints for intravenous treatment with SP-D. Daily infusions of human recombinant SP-D (the neck and carbohydrate recognition domains) were given to female SP-D-/- mice on the high-fat diet in order to test if the observed lipid derangement was SP-D dependent. The control group received human albumin.

The halflife of the injected SP-D was determined to be 29.1 hours and one daily infusion of 9 μg ensured that SP-D was present continously in the circulation. Spot sampling of the plasma concentration of SP-D showed an average of 2513 ng/ml.

The treatment of SP-D-/- mice with recombinant SP-D resulted in a lipoprotein profile similar to wildtype mice with significantly lowered total cholesterol, LDL-C and HDL-C after 5 days on high-fat diet as compared to albumin treatment. No significant alteration between SP-D and HSA treated groups were detected in the triglyceride levels.

The SP-D treatment affected total cholesterol and LDL-C levels relatively more than HDL-C levels.

The percentual alteration in total cholesterol between SP-D and albumin treated SP-D-/- animals were 1.3 fold larger than between the untreated wildtype and SP-D-/- mice (20 % versus 8 %). The corresponding alterations in LDL-C were 1.6 fold larger (22 % versus 14 %) wheras the SP-D induced alteration of HDL was smaller than the genotype difference (17 % versus 28 %).

The prolonged effect of the SP-D-/- genotype on lipid homeostasis in high-fat fed mice was the relative increase in both triglyceride and HDL-C levels. Risk factors for arterial wall damage include elevated levels of total cholesterol and triglyceride, high blood pressure and cigarret smoking. The observed effect of SP-D treatment on systemic lipid-homeostasis was a general lowering of plasma cholesterol levels. The effect of SP-D treatment was recorded on diet day 5 where cholesterol but not triglyceride levels were relatively disturbed in the SP-D-/- mice. Therefore, we speculate that triglyceride levels might be affected by SP-D treatment at another time point.

We also examined the total body-weight in the feeding study and observed that mouse body-weights were positively affected by the SP-D-/- genotype. The weight-gains were at least partly due to excessive visceral fat deposition.

SP-D-/- female mice gained weight at a 15 % faster rate compared to wildtype mice on the high-fat diet.

Male mice did not tolerate the high-fat diet as well as female mice and developed the highest body- weights on the normal chow. The rate of weight gain in male SP-D-/- mice was 15 % higher than the male wildtype rate on normal chow and 75 % on the high-fat diet.

Obesity is a powerfull predictor of cardiovascular disease (coronary heart disease). Thus, the observed SP-D deficient obese phenotypes may actually influence the patogenesis if present in humans.

To conclude this study, we fed a high-fat meal to humans and monitored serum SP-D in the postprandial state. We observed that the serum SP-D concentration was reduced transiently (1 hour 30 minutes - 3 hours) up to 27 % after the ingestion of the meal. The lipid binding ability of SP-D suggests a direct interaction with serum lipids resulting in a transient sequestration or consumption of the protein. Such interactions could potentially influence the cellular availability of the lipids for energy storage or expenditure.

Epidemiological studies have previously linked the innate immune system to lipidmetabolism and obesity in humans [Das UN (2001) Is obesity an inflammatory condition? Nutrition. 17: 953, Duncan BB, et al. (2001) Chronic activation of the innate immune system may underlie the metabolic syndrome. Sao Paulo Med J. 119: 122]. The hypothesis is that chronic activation of inflammation may cause metabolic disturbance and lead to obesity and cardiovascular disease. Given the general role of SP-D in depression of inflammation a role for SP-D could be in depressing the systemic inflammatory baseline or the postprandial cytokine production leading to a normolipidemic, lean and healthy phenotype. The most prominent effect of SP-D in our mouse studies seemed to be the influence on the body-weight. As human serum SP-D is largely determined by genetics (h² = 0.91) and varies more than 10 fold in a normal population [Husby S, et al. (2002) Heritability estimates for the constitutional levels of the collectins mannan-binding lectin and lung surfactant protein D. A study of unselected like-sexed mono- and dizygotic twins at the age of 6-9 years. Immunology. 106: 389] we speculate that "low SP- D" expressing individuals might be predisposed for an obese phenotype. Additional immunomodulatory, antiinflammatory, antioxidant and antiapoptotic effects of SP-D points to a general protective role in cardiovascular disease.

In summary, these studies have shown that SP-D deficiency induced increased plasma triglyceride and HDL-C levels. Transient increments in total cholesterol and LDL-C were induced in SP-D-/- mice and the administration of recombinant SP-D intravenously reversed the lipid derangement by the depression of total cholesterol, LDL-C and HDL-C levels. Lack of SP-D further lead to increased body- weight. The significantly increased weight gain was prominent in male mice. Serum SP-D was depressed transiently in humans after the ingestion of a high-fat meal indicating a role for SP-D in the postprandial lipid homeostasis.

Example 6: Heritability of surfactant protein D serum levels: relation to the metabolic syndrome among 1512 Danish twins

Morphological and metabolic variables were measured in a study including a total of 1512 healthy persons aged 18-67 years, and associations with serum SP-D examined. The cardiovascular risk factors considered were plasma glucose, serum insulin values, hypertension, body mass index (BMI), waiste-to-hip ratio, smoking, plasma cholesterol, high density lipoprotein cholesterol (HDL-C), very low density lipoprotein cholesterol (VLDL-C), triglyceride and cholesterol: HDL-C ratio. Heritability was estimated from variance-covariance analysis using the structural equation model approach.

Study population

A total of 1512 apparently healthy Danish twins from 18 to 67 years old were enrolled in the study including 783 women (37.5 ± 10.7 years) and 729 men (38.0 + 11.1 years). Table 10 describes the twin population in detail. The study was conducted according to the Helsinki recommendations and was approved by the Regional comittee for Research on Human Subjects (Review Board).

Table 10. The SP-D study population divided by zygosity.

MZ = monozygotic; DZ = dizygotic; OS = dizygotic twins with opposite sex; n total = the total number of individuals; n SP-D = the number of SP-D measurements.

Anthropometric measurements

Height, weight, waiste girth and hip girth were measured following the procedures recommended at the Airlie Conference, and BMI and waiste-to-hip ratio were calculated.

Measurements

Systolic blood pressure and diastolic bloodpressure were measured three times with the subject in the sitting position after resting for at least 5 minutes. Height, weight, fasting plasma glucose, serum insulin, serum SP-D, total plasma cholesterol, plasma HDL-C, plasma VLDL-C, and plasma triglyceride were measured in the morning after an overnight fast. Plasma glucose, serum insulin, and plasma lipids were measured by routine automated laboratory methods (Schousboe 2002).

Measurement of serum SP-D concentrations by ELISA technique

The immunoassay was performed as described in Leth-Larsen et al. Briefly, microtitre wells were coated with F(ab')₂ anti-human SP-D IgG (K477) prepared from rabbit anti-SP-D antibody, by 4°C overnight incubation at 1 μg/ml in bicarbonate buffer, pH 9.6. This incubation and all the following steps were carried out in a volume of 100 μl/well. Washes and incubations were carried out with Tris-buffered saline (TBS), 0.05 % (v/v) Tween-20 and 5mM CaCI₂ (assay buffer). The coated plates were washed and incubated with 200 μl assay buffer for 15 minutes at room temperature with rotary shaking. After washing, the plates were incubated overnight at 4°C with dilutions of serum, calibrator and control samples and washed. They were then incubated for 1 hour with 0.5 μg biotinylated monoclonal antibody anti-human SP-D (Hyb246-4) per ml assay buffer. This incubation and all the following steps were carried out at room temperature and with rotary shaking. After washing the plates were incubated for 30 minutes with horseradish peroxidase-labelled streptavidin (43-4323 Zymed, CA) diluted 1/1000. After a final wash, the bound enzyme was estimated by adding H₂0₂/orthophenyl diamin (Kem-En-Tec, Copenhagen, Denmark) substrate solution. The colour reaction was stopped after 15 minutes incubation in the dark by the addition of 150 μl of 1 M H₂S0₄. The absorbance was read at 492 nm using a multichannel spectrophotometer. The ELISA was set up with a serum calibrator and two quality controls. All twin pairs were analyzed within the same run.

Multiple regression analysis

All statistical analysis was performed using Intercooled Stata version 7. The model-fitting procedures assume a Gaussian distribution, which was not the case for the SP-D values. The SP-D values were logarithmically transformed before further analysis. The Gaussian distribution was verified by QQ-plot analysis.

Gender differences were determined by t-test analysis on transformed data. The measured or calculated variables entered the regression analyses in the following order: In (SP-D), gender, age, fasting plasma glucose (glucoseO), 30-minutes oral glucose tolerance test value (glucose30), 2-hours oral glucose tolerance test value (glucose120), fasting serum insulin (insulinO), 30-minutes oral glucose tolerance test insulin value (MnsulinSO), 2-hours oral glucose tolerance test insulin value (insulin120), BMI, waiste-to-hip ratio, hypertension, diabetic blood glucose profile, smoking, cholesterol, HDL-C, VLDL-C, triglyceride, cholesterol-to-HDL-C ratio. Smoking status was determined by a questionnaire with the two possible answers being yes or no. The binary variable hypertension was defined using guidelines from Harrison's Priciples of Internal Medicine to be a combination of a diastolic pressure > 90 mmHg and a systolic pressure > 140 mmHg. The binary variable diabetic blood glucose profile was defined as a fasting glucose value (glucose 0) > 7.8 mmol/l or a 2-hours oral glucose tolerance test value (glucose 120) > 11.1 mmol/l according to Danish medical guidelines. No subject was treated for hypertension or diabetes.

Relationships between serum SP-D levels and other parameters were analysed by multiple linear regression with a backward elimination procedure that was repeated until the Akaike information criterion was minimized. Introduction of the Stata cluster function controlling for twin relations did not alter the outcome of the analyses unless otherwise is stated.

Genetic analysis

Proportions of variance in serum SP-D attributable to genetic and environmental factors were assessed from variance-covariance matrices using the structural equation model approach with Mx as statistical software.

The total phenotypic variation can be decomposed into (A) additive genetic factors (C) shared environmental effects (E) non-shared environmental effects and (D) genetic dominance. The model assumes that shared environmental effects are perfectly correlated in MZ and DZ twins, negligible effects of assortative mating, epistasis, genotype-environment interaction and/or correlation.

Model fitting was by maximum likelihood and the best fitting model was based on the following criteria:

1 ) A non-significant P-value in the χ² goodness of fit test

2) Minimizing the Akaike Information Criterion (AIC = χ² - 2*d.f)

Heritability was computed as the genetic variance divided by the total phenotypic variance, which was derived from the best fitting model. Results

The population contained 1512 (783 women and 729 men) healthy twins between 18 and 67 years. Physical and metabolic characteristics of the subjects are shown in Table 11. Smoking status, hypertension, and diabetic blood glucose profile is summarized in Table 12.

Table 11. Physical and metabolic characteristics of women and men. Obs indicates the number of observations. Parameter values are expressed as means and standard deviations [std.dev.). Min and max states the minimum and maximum values of the parameter. An asterix indicates that the parameter is significantly different between women and men in the population.

a) Women

Variable Obs Mean Std. Dev. Min Max age (years) 783 37.5 10.7 18 67 height (cm) * 781 166.6 6.2 149 189 weight (kg) * 781 66.5 11.2 40.2 120.5

BMI * 781 24.0 3.8 16.1 43.7 hip girth (cm) 781 96.8 9.4 73 139 waiste girth (cm) * 781 78.6 9.6 58 119 waiste:hip ratio * 781 1.2 0.1 1.0 1.5 diastolic bp (mmHg) * 780 67.2 10.1 44 110 systolic bp (mmHg) * 780 113.5 13.6 82 195 glucose 0 (mmol/l) * 772 4.7 0.5 2.4 9.3 glucose 30 (mmol/l) * 774 8.3 1.5 3.6 16.9 glucose 120 (mmol/l) * 771 6.4 1.3 3.1 19.3 insulin 0 (pmol/l) * 755 38.2 19.4 6 150 insulin 30 (pmol/l) * 754 320.7 180.8 36 1684 insulin 120 (pmol/l) * 753 192.1 126.0 14 1041 cholesterol (mmol/l) 764 5.4 1.1 1.4 9.6

HDL-C (mmol/l) * 765 1.6 0.5 0.5 5.4

VLDL-C (mmol/l) * 763 0.5 0.5 0.0 12.2 triglyceride (mmol/l) * 765 1.2 0.7 0.4 14.5 cholesterol:HDL-C ratio * 764 3.5 1.1 0.5 11.5

SP-D (ng/ml) * 761 1048.8 609.8 152.7 4346.1 b) Men

Variable Obs Mean Std. Dev. Min Max age (years) 729 38.0 11.1 18 63 height (cm) 729 179.7 6.8 160 204 weight (kg) 729 80.3 10.8 55.5 125

BMI 729 24.9 3.1 17.7 40.4 hip girth (cm) 728 96.4 7.1 77 122 waiste girth (cm) 728 89.3 9.0 67 122 waiste:hip ratio 728 1.1 0.1 0.9 1.3 diastolic bp (mmHg) 729 70.1 10.5 42.7 112 systolic bp (mmHg) 729 120.4 14.0 78 204 glucose 0 (mmol/l) 721 4.9 0.6 2.8 13 glucose 30 (mmol/l) 721 8.7 1.6 4.5 18.7 glucose 120 (mmol/l) 720 5.9 1.6 2.5 23.6 insulin 0 (pmol/l) 715 36.3 19.6 7 182 insulin 30 (pmol/l) 715 301.2 193.0 40 1741 insulin 120 (pmol/l) 713 143.1 143.4 8 1992 cholesterol (mmol/l) 710 5.4 1.2 2.3 9.6

HDL-C (mmol/l) 712 1.4 0.4 0.4 5.4

VLDL-C (mmol/l) 702 0.6 0.4 0.1 8 triglyceride (mmol/l) 711 1.4 0.8 0.2 8.1 choletsrol:HDL-C ratio 710 4.1 1.2 1.1 8.8

SP-D (ng/ml) 715 1163.9 665.1 230.4 5216

Table 12. The number of individuals in the study population divided either by gender, smoking, hypertension, or diabetic blood glucose profile.

Serum SP-D concentrations and age correlations

The serum SP-D concentrations were within the range of 152.7 ng/ml and 5,216 ng/ml with a mean value of 1104.6 ng/ml (95 % Cl 1071.9 - 1137.2 ng/ml, SD = 639.6 ng/ml) and a median value of 913.1 ng/ml. The SP-D distribution was right-skrewed with a 10% percentile of 511.5 ng/ml and a 90 % percentile of 1867.7 ng/ml (Fig. 18).

The serum SP-D concentrations increased gradually with increasing age (Table 13). The increase in serum SP-D was steep between 20 and 50 years for both women and men and apparently levelling or decreasing after 50 years of age.

Table 13. Serum SP-D concentrations in different agegroups measured in women and men. Obs indicates the number of observations. SP-D values are expressed as means and standard deviation (Std.dev.).

a) Women Obs Mean (ng/ml) Std.dev. (ng/ml)

18-20 years 36 778.9 332.2

21-30 167 860.5 433.3

31-40 268 1018.2 564.4

41-50 170 1201.8 706.1

51-> 120 1243.4 715.5 b) Men Obs Mean (ng/ml) Std.dev. (ng/ml)

18-20 years 48 994.1 383.2

21-30 144 994.5 535.9

31-40 232 1156.7 637.1

41-50 155 1324.8 854.5

51 -> 136 1232.3 617.7

Pairwise correlations between parameters

All the measured and calculated parameters in the study were selected as risk factors for cardiovascular disease. Relations between these factors are extensive. Table 14 displays all the pairwise correlation coefficients between the variables in the dataset. Multiple regression analysis adjusts the specific covariate coefficients by keeping the other covariates constant, thus the outcomes of the regressions show the adjusted associations. Table 14. Covariance matrix of the covariates included in the regression model. Each covariate is given a number in the left colum. The numbers are repeated in the upper row. The diagonal with perfect covariances of 1.000 is omitted. Each significant (p < 0.05) covariance is shown.

1 2 3 4 5 6 7 8

1 male gender "

2 age -

3 insulinO -

4 BMI 0.1276 0.2582 0.4401 -

5 waiste: hip 0.6813 0.2009 0.1126 0.3897 -

6 hypertension 0.1478 0.1265 0.1765 0.0940 -

7 smoking -0.0734 -

8 cholesterohHDL 0.2247 0.1902 0.1666 0.2688 0.3336 0.1039 0.1127 -

Associations between serum SP-D concentrations and clinical and metabolic parameters in the total population.

The outcomes of multiple regression analyses on SP-D levels in the total population are shown in Table 15.

Table 15. Associations between cardiovascular risk factor variables and serum SP-D concentrations in the total study population. The regression coefficient of the variable in the population (Coef) and the standard error (Std. Err.) are shown. Obs indicates the total number of observations. F is the probability that all regression coefficients = 0. P is the p-value for testing that the specific regression coefficient = 0. Obs = 1462

F < 0.0001

R² = 0.1464 ln(SP-D) Coef. Std. Err. P gender 0.216046 0.035407 <0.001 age 0.012236 0.001223 O.001

BMI -0.015706 0.004120 O.001 waiste: hip -0.972323 0.228940 O.001 hypertension -0.173416 0.076121 0.023 smoking 0.184770 0.020872 <0.001 cholestero HDL 0.036988 0.011334 0.001 constant 7.324604 0.167343 <0.001

SP-D was positively correlated to male gender, age, smoking, and cholesterol-to-HDL-C ratio and negatively correlated to BMI, waiste-to-hip ratio and hypertension.

Effectively, this meant that the male population on average had a serum level of SP-D exceeding the serum level in women by 24 % when adjusted for the other parameters. Serum SP-D increased on average 1 %/year in the observed population. In accordance with previous observations that smoking induces serum SP-D this study demonstrated that smokers on average have 20 % higher serum SP-D than non-smokers. A positive alteration in SP-D of 4 % corresponded to an increase in the cholesterol- to-HDL ratio of one unit. A negative alteration in serum SP-D of 2 % corresponded to a one unit increase in BMI. Likewise, a negative alteration in serum SP-D of 62 % corresponded to a one unit increase in waiste-to-hip ratio and hypertensive individuals on average had 16 % lower serum SP-D than non-hypertensive.

Associations between serum SP-D concentrations and clinical and metabolic parameters in the old population.

The logarithmically transformed serum SP-D concentration was significantly (p < 0.0001) different between young subjects (< 50 years) and old subjects (> 50 years ) with a mean serum concentration in the old subjects of 1258.5 ng/ml and a mean serum concentration in the young subjects of 1067.7 ng/ml. The young population included 79% of the total population and the associations between serum SP-D and the variables in this study were essentially alike (Table 16). The old population only included 286 subjects (136 women and 150 men). In this small population regression analysis was corrected for twin relations. Several high significant associations to SP-D were found despite the limited sample size (Table 17). Age and BMI were no longer correlated significantly to serum SP-D in the old population. Smoking and waiste-to-hip ratio showed similar correlations to serum SP-D as were found for the total population. Smokers had on average 34 % higher serum SP-D than non-smokers and a negative alteration in serum SP-D of 24 % corresponded to a one unit increase in waiste-to-hip ratio. Two additional significant correlations were found in the old population. Fasting insulin was positively correlated whereas an overall diabetic blood glucose profile was negatively correlated. A positive alteration in SP-D of 1 % corresponded to an increase in the fasting insulin of 1 pmol/l and subjects with a diabetic blood glucose profile on average had 22 % lower serum SP-D compared to subjects with a normal glucose profile.

Table 16. Associations between cardiovascular risk factor variables and serum SP-D concentrations in the young (< 50 years) population. The regression coefficient of the variable in the population (Coef) and the standard error (Std. Err.) are shown. Obs indicates the total number of observations. F is the probability that all regression coefficients = 0. P is the p-value for testing that the specific regression coefficient = 0. Dia. glue. pro. = diabetic blood glucose profile.

Obs = 1179

F < 0.0001

R2 = 0.1510 ln(SP-D) Coef. Std. Err. P male gender 0.215168 0.038504 <0.001 age 0.014420 0.001772 O.001

BMI -0.017250 0.004598 <0.001 waiste:hip -0.884538 0.254987 <0.001 hypertens -0.196136 0.107586 NS dia. glue. pro. 0.172610 0.108284 NS smoking 0.166902 0.022070 O.001 triglyceride -0.034709 0.020154 NS cholesterokHDL 0.059488 0.013644 O.001 constant 7.182175 0.187291 O.001

Table 17. Associations between cardiovascular risk factor variables and serum SP-D concentrations in the old population. The regression coefficient of the variable in the population (Coe and the standard error (Std. Err.) are shown. Obs indicates the total number of observations. F is the probability that all regression coefficients = 0. P is the p-value for testing that the specific regression coefficient = 0. Dia. glue. pro. = diabetic blood glucose profile. Obs = 283

F < 0.0001

R² = 0.1335

Number of clusters = 142

Robust ln(SP-D) Coef. Std. Err. P male gender 0.229554 0.094567 0.016 insulinO 0.003350 0.001600 0.038

BMI -0.015684 0.012948 NS waiste:hip -1.442254 0.591068 0.016 dia. glue. pro. -0.248669 0.074962 0.001 smoking 0.294066 0.071743 0.000 constant 8.358868 0.407961 0.000

Associations between serum SP-D concentrations and clinical and metabolic parameters in women

The logarithmically transformed serum SP-D concentration was significantly (p = 0.0001) different between men and women with a mean serum concentration in women of 1048.8 ng/ml and a mean serum concentration in men of 1163.9 ng/ml. In addition, the majority of the measured or calculated parameters in this study varied significantly with gender (Table 12). Regression analysis performed on the separated female and male populations demonstrated the impact of the gender introducing a negative correlation between SP-D and plasma insulin in women and a positive correlation in men and the alteration of the levels of significance for the covariate regression coefficients.

Table 18 shows that age, hypertension, smoking and cholesterol-to-HDL ratio were correlated sinificantly to serum SP-D in women in a similar way as in the total population. Serum SP-D increased on average 1 %/year. Smokers had on average 16 % higher serum SP-D than non-smokers whereas hypertensive on average had 23 % lower serum SP-D than non-hypertensive. A positive alteration in serum SP-D of 6 % was correlated to an increase in the cholesterol-to-HDL-C ratio of one unit. As mentioned above, fasting insulin was negatively correlated. A negative alteration of serum SP-D of 0.3 % corresponded to an increase in fasting insulin of 1 pmol/l. 30-minutes oral glucose tolerance test insulin value (insulin30), BMI, and waiste-to-hip ratio covariate parameters were also included in the regression model although they did not correlate significantly to the ln(SP-D) variable.

Table 18. Associations between cardiovascular risk factor variables and serum SP-D concentrations in the female study population. The regression coefficient of the variable in the population (Coef) and the standard error (Std. Err.) are shown. Obs indicates the total number of observations. F is the probability that all regression coefficients = 0. P is the p-value for testing that the specific regression coefficient = 0. Obs = 747

F < 0.0001

R² = 0.1659 ln(SP-D) Coef. Std. Err. P age 0.012382 0.0017429 <0.001 insulinO -0.002883 0.0011174 0.010 insulin30 0.000155 0.0001087 NS

BMI -0.007979 0.0055641 NS waiste:hip -0.634915 0.3255538 NS hypertension -0.264319 0.1202144 0.028 smoking 0.151010 0.0246784 <0.000 cholesterohHDL 0.061853 0.0164939 <0.000 constant 6.849768 0.2513459 <0.000

Associations between serum SP-D concentrations and clinical and metabolic parameters in men

As shown in Table 19, the serum SP-D concentrations were positively correlated to age and fasting insulin and smoking in the male population and negatively correlated to BMI and waiste-to-hip ratio. Thus, serum SP-D increased on average 1 %/year. Smokers had on average 29 % higher serum SP-D than non-smokers and a positive chance of 1 % was correlated to an increase of 1 pmol/l fasting insulin. A negative alteration in serum SP-D of 3 % was correlated to a one unit increase in BMI and a negative alteration og 58 % corresponded to a one unit increase in the waiste-to-hip ratio.

Table 19. Associations between cardiovascular risk factor variables and serum SP-D concentrations in the male study population. The regression coefficient of the variable in the population (Coef) and the standard error [Std. Err.) are shown. Obs indicates the total number of observations. F is the probability that all regression coefficients = 0. P is the p-value for testing that the specific regression coefficient = 0.

Obs = 714

F < 0.0001

R² = 0.1352 ln(SP-D) Coef. Std. Err. P age 0.011954 0.001763 <0.001 insulinO 0.002689 0.001062 0.012

BMI -0.032576 0.007807 <0.001 waiste:hip -0.856535 0.326611 0.009 smoking 0.250781 0.038514 O.001 constant 7.893570 0.257549 <0.001

Intraclass correlations for serum SP-D levels

A significant proportion of the variance in SP-D (ln(SP-D) was accounted for by the ingoing covariates in the regression models for SP-D. The intraclass correlations were following analysed separatly for women and men and upon adjusted residual variances for ln(SP-D) after regression analysis in the two genders. Dizygotic twins of opposite gender were excluded from the analyses.

Intraclass correlations for ln(SP-D) residual levels were significantly higher in monozygotic twins than in dizygotic twins indicating a substantial genetic contribution to the variantion (Table 20). The correlation diagrams for SP-D and ln(SP-D) in monozygotic and dizygotic twins are given in Fig. 19. A clear clustering of SP-D and ln(SP-D) around til 45° identity line was seen for monozygotic twins whereas a scattered or circular pattern was seen for dizygotic twins for SP-D and ln(SP-D), respectively.

Table 20.

Zygosity N(pairs) Correlation P

Women

MZ 150 0.6915 <0.0001

DZ 156 0.3405

Men

MZ 146 0.8566 <0.0001

DZ 145 0.4447

Six biometric models were fittet to the normalised data (Table 21). The AE model gave the best fit by AIC for both women and men, indicating that additive genetic factors, and non-shared environmental factors were important. Model fit index

Model χ² d.f. P AIC

Women ACE 0.144 1 0.704 -1.856

CE 25.097 2 <0.001 21.097

AE* 0.144 2 0.931 -3.856

E 116.165 3 O.001 110.165

ADE 0.144 1 0.704 -1.856

DE 36.853 2 <0.001 32.853

Men ACE 0.027 1 0.868 -1.973

CE 62.502 2 O.001 58.502

AE* 0.125 2 0.940 -3.875

E 223.664 3 O.001 217.664

ADE 0.122 1 0.727 -1.878

DE 69.959 2 <0.001 69.959

Table 21. Biometric models for serum SP-D in women and men. d.f. = degrees of freedom. *Best fitting model by the Akaike information Criterion (AIC).

Table 22 shows the genetic and environmental factors contributing to variation in serum SP-D. The estimated heritability was 0.70 (95 % Cl 0.61-0.76) for female serum SP-D and 0.85 (95 % Cl 0.81-89) for male serum SP-D.

Variance due to

Additive genetic factors Non-shared environmental factors

Women 0.70 [0.61 - 0.76] 0.30 [0.24 - 0.39] Men 0.85 [0.81 - 0.89] 0.15 [0.11 - 0.19]

Table 22. Genetic and environmental contributions to variation in serum SP-D. Values in brackets denote 95% confidens interval.

Claims

Patent claims

1. A nucleotide sequence selected from the group consisting of: (a) the nucleotide sequence set out in Fig.1 or its complementary sequence; (b) a nucleotide sequence which hybridize under stringent conditions to the nucleotide sequence defined in (a) or fragments thereof; and (c) a nucleotide sequence which, but for the degeneracy of the genetic code, would hybridize to the nucleotide sequence defined in (a) and (b).

2. A polypeptide with SEQ ID No. 1 encoded by one of the nucleotide sequences as claimed in claim 1.

3. A polypeptide expressed in endothelial cells having the amino acid sequence of Fig 1 with SEQ ID No. 2.

4. A polypeptide which: (a) has part or all of the primary structure of the polypeptide as claimed in claim 3 and inhibits the formation of oxidized low-density lipoprotein; (b) is a non-naturally occurring polypeptide, and (c) is the product of procaryotic or eucaryotic expression of an exogenous DNA sequence.

5. A pharmaceutical composition comprising an amount of a polypeptide according to claims 2 to 4 and a pharmaceutically acceptable carrier or excipient.

6. A pharmaceutical composition comprising an amount of a polypeptide according to claims 2 to 4 and a pharmaceutically acceptable carrier or excipient, which polypeptide is effective to treat a person to prevent the development of atherosclerosis, which person is diagnosed in having an increased risk for the development of atherosclerosis.

7. A pharmaceutical composition comprising an amount of surfactant protein-D and/or any allelic forms and/or recombinant forms thereof and a pharmaceutically acceptable carrier or excipient.

8. A pharmaceutical composition comprising an amount of surfactant protein-D and/or any allelic forms and/or recombinant forms thereof and a pharmaceutically acceptable carrier or excipient to treat a person to prevent the development of atherosclerosis, which person is diagnosed in having an increased risk for the development of atherosclerosis.

9. A pharmaceutical composition comprising an amount of surfactant protein-D and/or any allelic forms and/or recombinant forms thereof and a pharmaceutically acceptable carrier or excipient to treat a person which is diagnosed in having an increased risk for the development of atherosclerosis.

10. A pharmaceutical composition as claimed in claims 5 to 9 for intra-venous, intra-muscular or apical application.

11. A pharmaceutical composition as claimed in claims 5 to 10 for the treatment of persons having diabetes.

12. Use of surfactant protein-D or any allelic or recombinant forms thereof, such as a polypeptide as claimed in claims 2 to 4, for the manufacturing of a pharmaceutical composition for the prevention or treatment of atherosclerosis.

13. Use of surfactant protein-D or any allelic or recombinant forms thereof, such as a polypeptide as claimed in claims 2 to 4, for the manufacturing of a pharmaceutical composition as claimed in claims 5 to 11 for the prevention or treatment of atherosclerosis.

14. Use as claimed in claims 12-13 of surfactant protein-D or any allelic or recombinant forms thereof, such as a polypeptide as claimed in claims 2 to 4, for the prevention or treatment of atherosclerosis- related disease selected from the group consisting of stroke, kidney failure, blindness, leg amputation and myocardial infarction.

15. Use of surfactant protein-D or any allelic or recombinant forms thereof, such as a polypeptide as claimed in claims 2 to 4, for the manufacture of a pharmaceutical composition for the treatment of obesity.

16. Use of surfactant protein-D or any allelic or recombinant forms thereof, such as a polypeptide as claimed in claims 2 to 4, for the manufacture of a pharmaceutical composition for the treatment of diabetes.

17. A method for treating a person for the prevention of developing atherosclerosis, which method comprises administering to said person an effective amount of an agent counteracting the influence of a reduced surfactant protein-D plasma or serum concentration.

18. A method for treating a person for the prevention of developing atherosclerosis, which method comprises administering to said person an effective amount of surfactant protein-D or any allelic or recombinant forms thereof, such as a polypeptide as claimed in claims 2 to 4, counteracting the influence of a reduced surfactant protein-D plasma or serum concentration.

19. A method for treating a person for the prevention of developing atherosclerosis, which method comprises subjecting the person to specific gene therapy aimed to repair the genetic basis for the reduced plasma surfactant protein-D concentration.

20. A method for treating atherosclerosis in a human which comprises administering to a human in need of such treatment a therapeutically effective amount of SP-D or any allelic or recombinant forms thereof, such as a polypeptide as claimed in claims 2 to 4.

21. A method for treating an atherosclerosis-related disease in a human which comprises administering to a human in need of such treatment a therapeutically effective amount of SP-D or any allelic or recombinant forms thereof, such as a polypeptide as claimed in claims 2 to 4.

22. A method for treating an atherosclerosis-related disease as claimed in claim 21 , whereby the atherosclerosis-related disease is selected from the group consisting of stroke, kidney failure, blindness, leg amputation and myocardial infarction.

23. A method for treating or preventing a person developing obesity, which method comprises administering to said person an effective amount of an agent counteracting the influence of a reduced surfactant protein-D plasma or serum concentration.

24. A method for treating or preventing a person developing obesity, which method comprises administering to said person an effective amount of surfactant protein-D or any allelic or recombinant forms thereof, such as a polypeptide as claimed in claims 2 to 4, counteracting the influence of a reduced surfactant protein-D plasma or serum concentration.

25. An in vitro method for diagnosing an increased risk for the development of atherosclerosis by detecting surfactant protein-D in human plasma or serum samples comprising the steps obtaining a sample from a human, assaying the sample to determine the amount of SP-D in the sample, and relating the amount of SP-D in the sample to the clinical status of the human.

26. An in vitro method as claimed in claim 25 whereby assaying the sample to determine the amount of SP-D in the sample comprises the use of at least one polyclonal or monoclonal antibody against human surfactant protein-D.

27. An in vitro method as claimed in claims 25 or 26 which method comprises the following steps of (a) providing polyclonal or monoclonal antibodies against human surfactant protein-D; (b) providing a microtiter plate coated with polyclonal or monoclonal antibodies against human surfactant protein-D; (c) adding serum or plasma of samples, calibrator and control samples to a microtiter plate; (d) providing a biotinylated polyclonal or monoclonal anti-human surfactant protein-D antibody; (e) providing horse radish peroxidase conjugated streptavidin; and (f) comparing the reaction which occurs as a result of steps (a) to (e) with results of calibrator and control samples.

28. A method for diagnosing a person's susceptibility for having an increased risk for the development of atherosclerosis, said method comprises determining the concentration of surfactant protein-D in human plasma or serum samples as claimed in claims 25, 26 or 27, and compare the concentration with the concentration determined in calibrator and control samples.

29. An antibody sandwich ELISA kit to screen persons for an increased risk for the development of atherosclerosis by detecting surfactant protein-D in human plasma and serum samples, the kit comprising a microtiter plate coated with polyclonal or monoclonal antibodies against human surfactant protein-D, a biotinylated polyclonal or monoclonal anti-human SP-D antibody, horse radish peroxidase conjugated streptavidin, and recombinant surfactant protein-D as an antigen standard.