The growing use of legume proteins in human nutrition for their nutraceutical properties(Reference Sirtori, Eberini and Arnoldi1–Reference Arnoldi and Arnoldi4) has recently suggested detailed investigations on the possible clinical use of lupin proteins. Lupin beans are characterised by a lower content of antinutrients v. other legumes(Reference Muzquiz, Pedrosa, Cuadrado, Ayet, Burbano, Brenes, Jasman, Hill, Huisman and van der Poel5) and by an almost total absence of phyto-oestrogens(Reference Katagiri, Ibrahim and Tahara6, Reference Sirtori, Lovati, Manzoni, Castiglioni, Duranti, Magni, Morandi, D'Agostina and Arnoldi7). This last feature, on the one hand, may avoid potential problems that have been recently indicated for these hormone-like components(Reference Sirtori, Arnoldi and Johnson8); on the other, it allows a direct evaluation of the activity of ‘proteins’, independent of other components. Lupin protein isolates are nutritionally satisfactory(Reference Caligari, Chiesa, Johnson, Camisassi, Gilio, Marchesi, Parolini, Rubio and Sirtori9, Reference Sujak, Kotlarz and Strobel10) and have a neutral flavour(Reference D'Agostina, Antonioni, Resta, Arnoldi, Bez, Knauf and Wasche11), thus allowing the production of food items with optimal sensory characteristics(Reference Arnoldi, Resta, Brambilla, Boschin, D'Agostina, Sirtori and O'Kane12).
A previous study from our group(Reference Sirtori, Lovati, Manzoni, Castiglioni, Duranti, Magni, Morandi, D'Agostina and Arnoldi7) investigated the potential hypolipidaemic effect of a total protein extract from Lupinus albus. When given to rats fed a classical cholesterol–cholic acid regimen, lupin proteins significantly reduced both plasma cholesterol and TAG levels v. control animals. The cholesterol reduction appeared to be associated with a mechanism shared with soya proteins, i.e. a direct up regulatory activity on LDL receptors(Reference Lovati, Gianazza and Sirtori13–Reference Duranti, Lovati, Dani, Barbiroli, Scarafoni, Castiglioni, Ponzone and Morazzoni15).
Based on recent data by our group(Reference Castiglioni, Manzoni, D'Uva, Spiezie, Monteggia, Chiesa, Sirtori and Lovati16) and others(Reference Adams, Golden, Williams, Franke, Register and Kaplan17, Reference Walker, Adams, Franke and Register18) suggesting an anti-atherosclerotic effect of soya proteins, also characterised by a cholesterol-lowering activity(Reference Sirtori, Eberini and Arnoldi1, Reference Lovati, Gianazza and Sirtori13, Reference Duranti, Lovati, Dani, Barbiroli, Scarafoni, Castiglioni, Ponzone and Morazzoni15), the impact on atherosclerosis progression of a diet containing L. albus proteins was tested in a rabbit model of focal lipid-rich soft plaques, generated at the common carotid arteries(Reference Chiesa, Di Mario and Colombo19).
Materials and methods
Lupin protein preparation
A total protein isolate from L. albus seeds was manufactured by the Fraunhofer Gesellschaft, Fraunhofer-Institute (Freising, Germany), by an extraction and precipitation process followed by spray drying(Reference D'Agostina, Antonioni, Resta, Arnoldi, Bez, Knauf and Wasche11). The protein percentage was 91·21 % DM. A detailed description of the composition and a proteomic investigation of this protein isolate have been previously reported(Reference D'Agostina, Antonioni, Resta, Arnoldi, Bez, Knauf and Wasche11, Reference Wait, Gianazza, Brambilla, Eberini, Morandi, Arnoldi and Sirtori20).
Animals, diets and experimental protocols
Procedures involving animals and their care were conducted in compliance with national and European Union laws and policies.
Arterial plaque formation at the common carotid artery was induced in male New Zealand White rabbits by perivascular injury, as described previously(Reference Chiesa, Di Mario and Colombo19, Reference Chiesa, Monteggia and Marchesi21). Rabbits were divided into three groups of six rabbits, balanced for body weight. After surgery, rabbits were fed a 1 % cholesterol and 15 % SFA diet for 90 d, the protein source (20 % in each diet) being either casein (CAS), 50 % CAS+50 % total protein isolate from L. albus (CAS+LUP) or total protein isolate from L. albus (LUP).
Fasting blood samples were taken before and after 30, 60 and 90 d of dietary treatments for plasma total cholesterol, TAG and HDL-cholesterol measurements, by a Roche Diagnostics Cobas autoanalyser (Nutley, NJ, USA). After 90 d diet, animals were anaesthetised with xylazine/ketamine and then killed. Injured arteries were excised, embedded in optimal cutting temperature compound for cryosections under liquid N2 and stored at − 80°C, until analyses. Plaque volume was evaluated by measuring cross-sectional areas of the intima every 0·25 mm within the area of plaque accumulation using an image analysis system (ImageJ 1.37v; National Institutes of Health, Bethesda, MD, USA) interfaced to a Zeiss Axioscope microscope (Carl Zeiss, Oberkochen, Germany). Selected sections were incubated as described(Reference Chiesa, Di Mario and Colombo19) with mouse monoclonal antibodies directed against rabbit smooth muscle α-actin (HHF35; DAKO Corp., Glostrup, Denmark) and rabbit macrophages (RAM-11; DAKO Corp.). Sections were also stained with Oil red O to identify lipid accumulation within the plaque(Reference Chiesa, Di Mario and Colombo19, Reference Chiesa, Monteggia and Marchesi21). Quantification of the percentage of plaque area covered by lipids or macrophages (i.e. with positive staining for Oil red O or for the anti-RAM-11 antibody, respectively) was performed by using a Nikon Coolpix 950 digital camera interfaced with a Zeiss Axioscope microscope (Carl Zeiss), followed by computer-assisted planimetry.
Statistical analyses
Data are expressed as mean values and standard deviations. Group differences were tested for statistical significance by multivariate ANOVA (repeated measures), followed by the Tukey post hoc test; a value of P < 0·05 was considered statistically significant. The statistical analysis was performed using SYSTAT software (version 5.2; Systat Software, Inc., San Jose, CA, USA).
Results
Effect on plasma lipids of a total protein isolate from Lupinus albus in rabbits
The presence in each diet of 1 % cholesterol and 15 % coconut fat determined a marked increase in total cholesterol levels in all groups (Table 1). However, whereas in CAS and CAS+LUP animals total cholesterol concentrations increased progressively up to 60 d, reaching values close to 19 000 mg/l that were maintained until killing (90 d), in LUP-fed rabbits cholesterolaemia did not undergo marked variations after 30 d of treatment and reached a maximum of 12 500 mg/l at 90 d. As a consequence of the different responses to the dietary treatments, a significantly lower mean plasma cholesterolaemia was observed in LUP compared with CAS animals after both 60 and 90 d of experimental diet (Table 1). No significant variations were observed for HDL-cholesterol levels (Table 1).
a,b,c Mean values within a column with unlike superscript letters were significantly different (P < 0·05; ANOVA).
* For details of diets and procedures, see Materials and methods.
† Mean value was significantly different from that of the CAS-fed animals at 60 d (P < 0·05).
‡ Mean value was significantly different from that of the CAS-fed animals at 90 d (P < 0·05).
A marked increase in TAG levels was observed after each dietary treatment, with no significant differences among groups (Table 1). However, whereas in the CAS group a progressive rise of TAG levels was observed, in LUP-fed rabbits triacylglycerolaemia significantly increased only after 90 d dietary treatment.
Effect on atherosclerosis development
The diet based on lupin proteins significantly affected atherosclerosis development (Fig. 1); at the end of the dietary treatments, plaque volume was significantly reduced in LUP compared with CAS rabbits (P < 0·05) (Fig. 1 (a)). Rabbits fed CAS+LUP displayed an intermediate plaque size between CAS and LUP animals, but statistical significance v. either comparator was not reached.
As previously described in this animal model fed the same diet(Reference Castiglioni, Manzoni, D'Uva, Spiezie, Monteggia, Chiesa, Sirtori and Lovati16), CAS rabbits developed plaques characterised by a neointimal formation devoid of smooth muscle cells and mostly constituted by macrophages and extracellular lipids. Compared with CAS animals, atherosclerotic plaques of LUP rabbits displayed a reduced macrophage ( − 23·4 %) and lipid accumulation ( − 36·2 %), this latter reaching statistical significance (P < 0·05) (Fig. 1 (b)).
Discussion
In the present study, lupin proteins inhibited the cholesterol rise induced by a lipid-rich diet, confirming results previously obtained in rats(Reference Sirtori, Lovati, Manzoni, Castiglioni, Duranti, Magni, Morandi, D'Agostina and Arnoldi7). No significant differences were instead observed for triacylglycerolaemia among groups. It should be noted, however, that, differently from the CAS group, the increase of TAG levels in the LUP-fed rabbits occurred only at the end of the dietary treatment (90 d). It may be speculated that the long duration of the high-fat, high-cholesterol dietary challenge required for the development of atherosclerotic plaques may have overcome a possible hypotriacylglycerolaemic effect of lupin proteins in this model. Similarly, the aggressive dietary treatment required to induce atherosclerotic plaque formation may have hidden a possible hypolipidaemic effect played by lupin proteins in the CAS+LUP group. The main objective of the present study was, however, the investigation of the impact of lupin proteins on atherosclerosis progression. To investigate this issue, an appropriate animal model was selected, i.e. the rabbit that, differently from the rat, is highly susceptible to atherosclerosis development. In this animal model, perivascular manipulation at the common carotid arteries, followed by a hyperlipidaemic diet, induces the formation of focal plaques, mostly constituted by extracellular lipids and macrophages(Reference Chiesa, Di Mario and Colombo19), thus reflecting the main features of the human arterial plaques, defined as unstable, frequently associated with acute ischaemic events(Reference Virmani, Kolodgie, Burke, Farb and Schwartz22). This same rabbit model has proven to be sensitive to local interventions with recombinant apolipoproteins(Reference Chiesa, Monteggia and Marchesi21), as well as to dietary treatments(Reference Castiglioni, Manzoni, D'Uva, Spiezie, Monteggia, Chiesa, Sirtori and Lovati16).
The diet exclusively based on L. albus proteins clearly reduced atherosclerosis progression, possibly as a consequence of the observed hypocholesterolaemic activity. The reduction of circulating atherogenic lipoproteins by lupin proteins could easily explain the lower lipid accumulation within carotid plaques of the LUP compared with the CAS group. In the CAS+LUP group, where no hypocholesterolaemic effect was observed, plaque volume was not statistically different from that observed in the CAS rabbits. However, the CAS+LUP group displayed a trend toward a lower extent of atheromas compared with the CAS group, suggesting that the anti-atherogenic activity exerted by lupin proteins may be partially explained by other mechanisms, in addition to the observed reduction of cholesterolaemia. This issue will be the object of future investigations.
While there is evidence for a vascular protective effect of soya proteins(Reference Castiglioni, Manzoni, D'Uva, Spiezie, Monteggia, Chiesa, Sirtori and Lovati16–Reference Walker, Adams, Franke and Register18), no data are available on lupin proteins, except for those related to the hypolipidaemic activity(Reference Sirtori, Lovati, Manzoni, Castiglioni, Duranti, Magni, Morandi, D'Agostina and Arnoldi7, Reference Martins, Riottot, de Abreu, Viegas-Crespo, Lanca, Almeida, Freire and Bento23). Differently from soya, a relevant characteristic of lupin is the total absence of isoflavones(Reference Katagiri, Ibrahim and Tahara6, Reference Sirtori, Lovati, Manzoni, Castiglioni, Duranti, Magni, Morandi, D'Agostina and Arnoldi7), which permits us to conclude that the observed vascular protective properties should certainly be ascribed to the protein component.
When the nutraceutical properties of a food are due to the proteins, it becomes very important to assess their quality and integrity. A proteomic analysis of the protein isolate from L. albus used in the present study was recently reported(Reference Wait, Gianazza, Brambilla, Eberini, Morandi, Arnoldi and Sirtori20). The results show that the protein isolate had undergone limited damage, demonstrating that the manufacturing process had been very mild, different from the case of commercial soya protein isolates(Reference Gianazza, Eberini, Arnoldi, Wait and Sirtori24).
Interestingly, in a recent study, L. albus proteins were given as a beverage to subjects with moderate hypercholesterolaemia in an amount corresponding to a daily intake of approximately 35 g(Reference Naruszewicz, Nowicka, Klosiewicz-Latoszek, Arnoldi and Sirtori25). This dietary supplementation significantly reduced both cholesterolaemia and blood pressure, suggesting that the beneficial effects of lupin proteins observed in animal studies(Reference Sirtori, Lovati, Manzoni, Castiglioni, Duranti, Magni, Morandi, D'Agostina and Arnoldi7, Reference Pilvi, Jauhiainen, Cheng, Mervaala, Vapaatalo and Korpela26), including the present study, may occur also in humans. While these preliminary clinical data will require confirmation in appropriate controlled investigations, the results of the present study suggest the potential positive impact of lupin proteins on atherosclerosis prevention and treatment.
Acknowledgements
We are indebted to the Fraunhofer Gesellschaft, Fraunhofer-Institute (IVV) (Freising, Germany) which provided the protein isolate from L. albus.
This investigation was supported by a grant from the European Commission, Fifth Framework Programme, Quality of Life and Management of Living Resources Programme, Healthy Profood QLRT 2001-2235. There are no conflicts of interest. All authors contributed to the design of the study and/or analysis and interpretation of the data.