AU5939399A - Sterol compositions from pulping soap - Google Patents

Description

I-i 47446DIV AWT:HM P/00/011 Regulation 3.2

AUSTRALIA

Patents Act 1990 COMPLETE SPECIFICATION FOR A STANDARD PATENT

ORIGINAL

4 Name of Applicant: THE UNIVERSITY OF BRITISH COLUMBIA Name of Inventors: Egon NOVAK James P KUTNEY Peter J JONES Address for Service: COLLISON CO.,117 King William Street, Adelaide, S.A. 5000 Invention Title: STEROL COMPOSITIONS FROM PULPING SOAP The following statement is a full description of this invention, including the best method of performing it known to us: Name of Inventors: Egon NOVAK James P KUTNEY Peter J JONES Address for Service: COLLISON CO.,117 King William Street, Adelaide, S.A. 5000 Invention Title: STEROL COMPOSITIONS FROM PULPING SOAP The following statement is a full description of this invention, including the best method of performing it known to us: WO 96/10033 PCT1CA95100555 STEROL COMPOSITIONS FROM PULPING SOA FIELD OF THE INVENTION This invention relates to the preparation and purification of sterol compositions from pulping soaps, to the actual compositions per se and to the use of these compositions and derivatives thereof as agents to prevent or treat primary and secondary dyslipidemias.

BACKGROUND OF THE INVENTION The direct cause of heart attack and angina is a degenerative process known as atherosclerosis. Atherosclerosis results from a number of integrated inherited (genetic) and environmental factors.- The interplay of these factors, of which diet in our civilization appears to be the most important, leads to the development of atherosclerosis. Growth of cholesterol filed ~o ~:atherosclerotic plaques ultimately cuts of blood supply to the heart muscle, or alternately to brain or legs, depending on the location of the plaque in the arterial tree.

One of the major risk factors for atherosclerosis that is potentially diodifiable is the level of blood cholesterol. A number of well *documented studies have shown that the blood cholesterol level is indeed an important predictor for the risk of heart attack and also for strokes. The relationship between blood concentration of cholesterol and risk of these disorders is continuous (spans across all levels of cholesterol) graded (the higher the level, the more likely the disease) with no apparent threshold (even by lowering so-called low levels, one can further decrease risk of the disease).

For examnple, in people over 40 years of age, a blood cholesterol level of mmol/L presents a risk of coronary artery disease three to four times that associated with levels below 5.0 mznol/L. The relationship becomes especially steep when the levels are above 5.2 mxnol/L. For instance, the death rate among men with cholesterol of levels 8.0 rnmol/L was almost six times that __jlli/ijirl___ii WO 96/10033 PCTICA95/00555 -2among men with levels of 4.0 mmol/L. These more recent findings are consistent with earlier studies.

Other large clinical trials have shown clearly that by lowering high cholesterol levels, one can reduce the risk of fatal and non-fatal myocardial infarctions, angina, changes in electrocardiograms and in coronary artery bypass surgery. The best known and the first of these trials was at Lipid Research Clinics at which Coronary Primary Prevention Trials showed that with every 1% reduction in total blood cholesterol level, there was a 2% reduction in the risk of coronary artery disease.

For any long term preventative therapy of hypercholesterolemia to be successful, it has to be commenced at a relatively early age and continue indefinitely. While a low-fat diet is the corner stone of such long term *:therapy, up to 60% of patients become non-compliant after six months. The difficulty in non-compliance is marked in many Western countries by a general 15 diet which is high in fat. The poor cholesterol profile of many patients is exacerbated by the prevalence of additional risk factors for cardiovascular disease such as high blood pressure, diabetes, obesity and smoking.

"Dietary modification as a therapy for atherosclerosis and other cardiovascular diseases has been refined significantly over the past 10 to 20 years. In particular, it has been recognized by researchers that plant sterols (phytosterols) are effective in lowering plasma cholesterol levels: Lees et al.

Atherosclerosis, 28 (1977) 325-338; Kudehodkar et al., Atherosclerosis, 23 (1976) 239; Day Artery, 18(3):125-132 (1991).

9 Phytosterols are sterol-like compounds synthesised in plants with no nutritional value to humans. In plants they are required for cell function in a manner similar to the way in which cholesterol is required in humans.

The average Western diet contains up to 360 mg of phytosterols per day.

Recently, these dietary plant sterols have received a great deal of attention because of their possible anti-cancer properties and their ability to decrease WO 96110033 PCTICA9500SSS -3cholesterol levels when fed to a number of mammalian species, including humans.

Chemically, phytosterols closely resemble cholesterol in structure. The major phytosterols are beta-sitosterol, campesterol and stigmasterol. Others include stigmastanol (beta-sitostanol), sitostanol, desmosterol, chalinasterol, poriferasterol, clionasterol and brassicasterol. The chemical structures of beta-sitosterol, canipesterol and stigmasterol are as follows: beta-sitosterol campesterol -stigmasterol WO 96/10033 PCT/CA95/00555 -4- The mechanism by which phytosterols lower blood cholesterol in animals is unclear, but it appears to involve the inhibition of cholesterol absorbtion from the proximum jejunum by competing with cholesterol at specific uptake sites. Research data has also suggested that some phytosterols are not absorbed at the proximal jejunum at all (sitostanol) and, when there is absorbtion (beta-sitosterol), it is in very limited quantities.

Based on these research findings, the use of phytosterols as a dietary supplement to reduce cholesterol absorbtion has been widely investigated. Lees et al., supra; Pollak, Pharmac. Ther., 31 (1985) 177-208; Raicht et al., Biochimica et Biophysica Acta, 388 (1975) 374-384.

In Lees et al., supra, a comparison was made between the effects of siterosterol preparations from two sources, soy sterols and tall oil sterols, on plasma cholesterol. Plant sterol preparations were found to be effective in treating patients with hypercholesterolemia. Pollak, supra, is a 15 survey paper of phytosterols and their effect on serum lipids. Raicht, supra, describes further the effect of beta-sitosterol on sterol balance and rate-limiting enzymes of sterol metabolism.

It is generally accepted that phytosterols offer a unique combination of long-term safety, efficacy, and versatility in human treatment.

20 The ongoing challenge with respect to phytosterols is in their isolation and purification from plant sources and in determining additional sources which are cost-effective, manageable on a large-scale and which exhibit Shypocholesteremic effects.

Traditionally, phytosterols have been isolated from sources such as corn oil, wheat germ oil, soya bean pitch and corn oil pitch. Similarly, tall oil pitch, which is obtained during the process of preparing paper from wood, particularly pine wood, has been used as a phytosterol source. Generally, in this process, wood chips are digested with caustic soda to produce a pulp or "soap". -The soap is then distilled to remove the volatile materials leaving a ~_~muP___lp~-uarra_~ii~~ WO 96/10033 PCTICA95/00555 "pitch" as the residue. It is from this pitch that researchers have isolated phytosterols.

There are some marked disadvantages to these traditional sources of phytosterols. The tall oil pitch is an extremely complex material comprising resins, fatty acids, oxidation products, esterified materials and phytosterols. Although the pitch is inexpensive in that it is the tailing left from various manufacturing processes, it is very difficult to recover high molecular weight sterols from it in good yields and at the high purities required for pharmaceutical uses.

United States Patent No. 3,840,570 to Jullan provides a process for preparing sterols from tall oil pitch by extraction in a water alcohol hydrocarbon mixture followed by saponification and subsequent purification.

The starting material in this process is tall oil pitch from which are extracted S phytosterols and various impurities. It is recognized that, in any tall oil pitch 15 purification process, the long-chain alcohol and acid impurities are particularly difficult to separate from the sterols (which are, themselves, high molecular weight alcohols).

Other researchers have addressed the issue of sterol purification from tall oil pitch: United States Patent No. 2,835,682 to Steiner and Fritz; 20 United States Patent No. 2,715,638 to Albrecht and Herrlinger; United States Patent No. 2,573,891 to Christenson. It is important to note that in each of these known purification processes, the starting material was tall oil pitch which has the recovery problems discussed above.

It is an object of the present intention to obviate or mitigate the above disadvantages.

SUMMARY OF THE INVENTION The present invention provides a process for purifying and preparing phytosterol compositions from pulping soap which comprises extracting from the pulping soap a creamy precipitate and purifying this WO 96/10033 PCT/CA95/00555 -6precipitate to form a unique phytosterol composition. More specifically, the creamy precipitate is extracted from the pulping soap using a solvent extraction procedure. The resultant composition is then purified from the creamy precipitate by crystallization.

The present invention also provides unique compositions which are effective in preventing or treating dyslipidemias and which comprise betasitosterol, campesterol and stigmastanol. The phytosterol compositions provided herein are significantly different from those found in plants, foods and oils. In particular, the provision of stigmastanol appears to enhance the efficacy. These compositions may additionally comprise various co-occurring compounds, which may or may not be phytosterols. In particular, these cooccurring compounds may include triterpenes, long chain alcohols and other alcohol-soluble organic compounds.

The present invention further provides the use of the 15 compositions described herein to prevent or treat primary and secondary dyslipidemias and atherosclerosis including coronary heart disease, peripheral vascular disease and strokes in humans and animals.

The unique compositions of the present invention have exhibited excellent results in lowering total (TC) and low density lipoprotein (LDL) 20 blood cholesterol. In addition, and quite surprisingly, the compositions of the present invention were found, in different animal species, to maintain or elevate plasma levels of high-density lipoprotein (HDL) blood cholesterol.

This feature of the present invention is critically important given the fact that research has shown that, irrespective of TC levels, as the plasma HDL level decreases, the risk of atherosclerosis increases. Phytosterols isolated from tall oil pitch, soybean and other sources have not, to the knowledge of the present inventors, exhibited this unique HDL effect.

Although it is known to produce some types of phytosterols from the pitch distilled from the soap of wood chip treatments, phytosterol compositions have not heretofore been produced from the pulping soap _LI WO 96/10033 PCT/CA95/00555 -7component of the wood chip treatment process. The tall oil pitch is significantly different in composition from the pulping soap. It is believed that the surprising effect of the compositions of the present invention is due, at least partially, to the use of the pulping soap as the starting material and to the unique separation process.

BRIEF REFERENCE TO THE DRAWINGS: Various aspects of the invention will be illustrated by the following non-limiting drawings wherein: Figure 1 is a gas-chromatography profile for one composition (hereinafter Forbes-2) within the scope of the present invention; Figure 2 is a representation of the profile in Figure 1 from to 45 minutes retention time; Figure 3 is a representation of the profile in Figure 1 from 22 to 27 minutes retention time; Figure 4 is an index of the gas-chromatography profile of Figure 1; Figure 5 is a gas-chromatography profile for another composition (hereinafter Forbes-3) within the scope of the present invention; 20 Figure 6 is a representation of the profile in Figure 5 from 32 to 48 minutes retention time; Figure 7 is an index of the gas-chromatography profile of Figure *Figure 8 represents a bar graph illustrating the effects of Forbes- 1 and Forbes-2 on TC concentrations in rats; Figure 9 represents a bar graph illustrating the effects of Forbes- 1 and Forbes-2 on LDL-cholesterol concentrations in rats; Figure 10 represents a bar graph illustrating the effects of Forbes-1 and Forbes-2 on HDL-cholesterol concentrations in rats; _li~l~C~ WO 96/10033 PCT/CA95/00555 9 9 9* *9 9 9 *r a *9 9 .9 4e**e *9 Figure 11 represents a bar graph illustrating the effects of Forbes-3 on serum TC in hamsters; Figure 12 represents a bar graph illustrating the effects of Forbes-3 on serum LDL-cholesterol in hamsters; Figure 13 represents a bar graph illustrating the effects of Forbes-3 on serum HDL-cholesterol in hamsters; Figure 14 represents a bar graph illustrating the effects of various dietary treatments on cholesterol levels in male and female hamsters; Figure 15 represents a bar graph illustrating the effects of various dietary treatments on plasma cholesterol levels in male hamsters; Figure 16 represents a bar graph illustrating the effects of various dietary treatments on plasma cholesterol levels in female hamsters; Figure 17 represents a bar graph illustrating the effects of various dietary treatments on plasma triglyceride levels in male and female 15 hamsters; Figure 18 represents a bar graph illustrating the effects of various dietary treatments on HDL/apo-B ratios in male and female hamsters; Figure 19 represents a bar graph illustrating the effects of dietary treatments on total cholesterol in hamster 45 day study; Figure 20 represents a bar graph illustrating the effects of dietary treatments on cholesterol correlation with sitostanol; Figure 21 represents a bar graph illustrating the effects of dietary treatments on HDL levels in hamsters over 45 days; Figure 22 represents a bar graph illustrating the effects of dietary treatments on Non-apoA/apoA ratios in hamsters over 45 days; and Figure 23 represents a bar graph illustrating the effects og dietary treatments on Non-apoA sterols in hamsters over 45 days.

-IE-L-IY~I ~_-_-_li~I~^I1IIICY-_LYIII~-IIYIIY-~III~ WO 96/10033 PCT/CA95/00555 -9- PREFERRED EMBODIMENTS OF THE INVENTION The process of the present invention comprises the steps of: obtaining or preparing the starting material, a plant-derived pulping soap; extracting from the soap a creamy precipitate using an appropriate solvent; and purifying from the creamy precipitate a phytosterol composition.

There are numerous possible sources of the plant-derived pulping soap. Generally, in a known process (the "Kraft" process) wood chips are treated with caustic soda to produce a soap. The wood chips may be derived from any hard wood or soft wood variety of tree including, but not limited to, fir, cedar, pine, spruce, oak, hemlock and poplar. Most preferably, the chips are derived from any Pacific Northwest American or 15 European forest variety of woods.

In the extraction phase, the soap is mixed with a ketone and water solution. A hydrocarbon solvent is used to extract the sterols. This step can be performed at temperatures generally from about 25"C to about 150"C, but most preferably from about 50C to about 100*C. Most preferably, this 20 extraction phase is continued over 15 to 24 hours. It is important to note that the use of alcohol is required nor suggested during the extraction phase. The extraction process of the present invention is conducted using a ketone-water-hydrocarbon solvent.

~The ketone is selected from the group having the general structure RCORI where R and R' are alkyl groups. Preferably the alkyl groups are C, C 6 groups. Most preferably, the ketone is 2-propanone (acetone). The hydrocarbon may be selected from the group comprising all

C

5 CIo hydocarbons. Most preferably, the hydrocarbon is hexane.

As depicted in Figure 1, the product of the extraction phase is a creamy precipitate or residue from which is purified the phytosterol WO 96/10033 PCT/CA95/00555 composition. This purification phase may be conducted by crystallization, chromatographic separation or by any other suitable procedures. Most preferably, the creamy precipitate is dissolved in alcohol, cooled slowly, then filtered and washed with cold alcohol. The residue is dried, and the resultant product is a phytosterol composition.

In a preferred form, the alcohol used in the purification phase is selected from the group having the general structures R-CHOHR, R-

CH

2 OH, and RCOH where R is a C, C 4 alkyl group. Most preferably, the alcohol is methanol. The cooling phase may be affected at temperatures from 10* Celsius to 0° Celsius, most preferably at 3 to 4° Celsius for 24 hours.

The phytosterol compositions resulting from the processes described herein may be incorporated directly into food supplements and vitamin formulations and into drugs for on-going and preventive treatment of atherosclerosis and its consequences, strokes, heart attacks and peripheral 15 vascular disease. In addition, it is contemplated within one embodiment of the present invention that the phytosterol compositions described herein be provided in the form of medications with suitable adjuvants or carriers. For example, these compositions may be incorporated or prescribed concurrently with selected lipid-lowering agents, to decrease the necessary dosage, and hence the toxicity, of these latter compounds.

The phytosterol compositions of the present invention have exhibited a marked ability to modify lipoproteins, even at lower phytosterol concentrations than in known formulations. More surprisingly, however, has been the effect of these compositions on increasing plasma levels of high density lipoproteins (HDL), an effect heretofore not associated with any other tall oil-derived phytosterol composition. It is believed that this unique effect may be due to the use of pulping soaps as the starting material or the provision of stigmastanol as an element of the composition.

In a preferred form, the compositions of the present invention comprise the following ratio of phytosterols: beta-siterosterol campesterol *C WO 96/10033 PCT/CA95/00555 -11and stigmastanol (0.2 More preferably, campesterol and stigmastanol together represent at least 50% of the total concentration of betasitosterol. In a most preferred form, the compositions of the present invention comprise the following ratio of phytosterols as compared to soybean-derived phytosterols: Ratio of Known Phytosterols Approximate B-Sitosterol Campesterol Stigmastanol Purity Soybean 1 '0.640 0.005 Forbes-1 91.0 1 0.354 0.414 Forbes-2 77.0 1 0.330 0.203 Forbes-3 90.0 1 0.268 0.299 The composition and purity of two other extracts within the scope of the present invention are as follows: 9.

9 Composition Approximate B-Sitosterol Campesterol Stigmastanol Purity Forbes-4 99.0 62.6 16.6 23.2 98.3 64.7 16.4 17.2 9 In every composition described herein, there may be additional compounds present which may or may not be phytosterols. For example, it has been found that campestanol, another phytosterol, may be present in a relatively small quantity. In addition, straight chain fatty alcohols, such as behenyl (C22) and lignoceryl alcohol (C24) may be present. In order to determine the nature of these co-occurring compounds, gas liquid chromatography analysis has been conducted on each of the most preferred compositions of the present invention.

c WO 96/10033 PCT/CA95/00555 -12- Gas chromatography running conditions for the phytosterols were: initial temperature 80*C which was held for 1 minute; ramp to 120"C at 20*C per minute, which was held for 7 minutes; ramp to 24*C at 20*C per minute which was held for 15 minutes; and ramp to 269*C at 20*C per minute which was held for 25 minutes. At the end of each run, the temperature was ramped to 320*C and held for a minimum of 5 minutes. The injection temperature was 300"C and the detector temperature was 320"C. The column flow rate was 1 ml per minute and the split vent flow rate was 4 ml minute.

The purge vent flow rate was 4.5 ml minute. The carrier gas was helium.

The results of the gas liquid chromatography analysis for two of the most preferred compositions of the present invention are depicted in Figures 1-7.

With respect to the Forbes-2 composition, the known sterols S appear in the 35 45 minute region in Figures 1 and 2. Beta-sitosterol is 15 indicated at peak 87; campesterol is indicated at peak 81 and stigmastanol is indicted at peak 84. Peaks 65, 66 and 77 in Figure 2 are co-occurring compounds which may exhibit hypocholesterolemic effects. It is possible, however, that these co-occurring compounds may have a synergistic effect on the actions of the known phytosterols in the compositions. Similarly, in 20 Figures 5 and 6, campesterol, stigmastanol and beta-sitosterol are represented at peaks 6, 7 and 8 respectively.

Another preferred composition within the scope of the present invention comprises the following components: Campesterol 14.1% Campestanol B-sitosterol 62.8% stigmastanol 16.9% for a total phytosterol concentration of 97.3%.

~L WO 96/10033 PCT/CA95/00555 -13- EXAMPLE 1 Extraction and Purification A batch of 3 kg of pulping soap was obtained from B.C.

Chemicals Inc. A mixture of 3 L of acetone and 1.5 L of water was prepared to which the soap was added. The mixture was extracted continuously with 4.5 L of hexane at 50*C for 24 hours using an 18 L evaporator. The resultant extraction product was then dried over sodium sulphate and allowed to evaporate. This produced 460 g of residue or creamy precipitate.

The creamy precipitate was warmed and stirred using a magnetic bar and 460ml of methanol was slowly added. The mixture was refluxed under stirring for 15 min. and cooled slowly for 3-5 hours. The mixture was refrigerated at 3-4°C overnight and then filtered and washed (twice) with 150 ml cold methanol. Finally, the mixture was maintained in a vacuum for 2 days yielding 100 g of mixture with a purity of 82% 82 g of phytosterols).

S* S EXAMPLE 2 Evaluation of the Effects of Phytosterol Compositions in

A

Rats Ninety male Wistar rats (80-100 g) were divided into 3 experimental modules: Forbes-1 composition; Forbes-2 composition and 20 soybean. The thirty rats within each module were further divided into dietary regimes as indicated in Table 2. The rats were kept on reverse lighting cycle, and fed for 10 days with a basal semi-purified diet (Table 1) supplemented with different amounts of cholesterol and phystosterol (Table 2).

•Within each of the 5 dietary groups, 2 rats were administered the Forbes-1 composition, 2 rats were administered the Forbes-2 composition and 2 rats were administered soybean-derived phytosterol (Sigma).

91C WO 96/10033 PCTICA95/00555 a .a a.

a a. a a a a. Table 1. Composition of experimental diet Ingredients Casein Cornstarch 21.5 Sucrose Fixed-oil* 18 Dl-methionine Mineral mixture 4.00 Vitamin mixture 1.00 Safflower and lard mixed in a 1:3 ratio.

Table 2. Dietary regimens Groups Sterols added to the basal diet Cholesterol Phytosterol 15 1 0 0 2 1 0 3 1 0.2 4 1 5 1 1 At the end of the feeding period, the rats were intraperitoneally injected with deuterium oxide (0.4 ml) and deprived of food and water for at least 2 hours. The rats were then anaesthetized with halothane. Blood samples were withdrawn from the heart. Samples of liver, small intestine and muscle were quickly removed, weighed, put in liquid nitrogen and stored at 800C until determination of cholesterol synthesis. Total cholesterol, LDL and HDL cholesterol were determined with a commercial kit (Biopacific Diagnostic Inc).

The results of the effects of the phytosterol compositions on total cholesterol, LDL and HDL are represented in Figures 8, 9 and respectively. The efficacy of the Forbes-1 and Forbes-2 is evident from the reduction in LDL-cholesterol shown in Figure 9 and in the increase in HDLcholesterol shown in Figure 10, particularly by Forbes-1. In Figure 8, the a a WO 96/10033 PCTICA95/00555 addition of cholesterol (dietary group 2) to the base diet (group 1) resulted in an increase in circulating cholesterol concentrations. Progressive addition of increasing levels of phytosterols (groups 3-5) resulted in a normalization of cholesterol levels in groups fed Forbes-2 and Forbes-1, but not soybean phytosterols, as determined by regression analysis. Figure 9 shows that Forbes-2 and Forbes-1 phytosterols possess better cholesterol lowering efficacy than the soybean phytosterols for LDL. Figure 10 demonstrates the greater HDL-raising ability of the preferred compositions of the present invention, particularly Forbes-1, compared to the soybean phytosterols.

EXAMPLE 3 Evaluation of Effects of Phytosterol Compositions in Hamsters The present study was to examine the effect of dietary phytosterol compositions of the present invention on the dietary cholesterol- 15 induced elevation of serum cholesterol concentrations in hamsters.

A total of 40 male hamsters (80-100g), housed individually in stainless mesh cages were fed rodent chow and acclimated for three days in an *6.

air conditions room (20-22 0 C, lights on 1700-0500). Hamsters were then divided into five groups of 8 animals each group, and fed for 34 days, a basal 20 semi-purified diet (Table 3) supplemented with different amounts of cholesterol and one of the phytosterol compositions of the present invention (Forbes 3) (Table 4).

WO 96/10033 PCr/CA95/00555 -16- Table 3. Composition of experimental diet Ingredients Casein Cornstarch Sucrose Corn oil Cellulose D l-methionine Mineral mixture Vitamin mixture Choline bitartrate Cholesterol by weight 28 36.3 4.00 1.00 0.2 0.025, 0.25 00 @0 00 S 0 0000 0* 0* 0* 00 I S 0 0 0*00 *0 00 0 00 0 000 0 0000 00 00 0 *000 00 00 0 000000 0 0 000000 0 00 0 0 000 0 000000 0 Table 4. Dietary regimens 15 Groups Cholesterol added to Phytosterol added to control diet control diet 1 0.025 None 2 0.25 None 3 0.25 0.25 4 0.25 5 0.25 At the end of the feeding period the animals were 25 intraperitoneally injected with deuterium oxide (0.4 ml), and deprived of food and water for at least 2 hours. The hamsters were then anaesthetized with halothane. Blood samples were withdrawn from the heart. Other tissue samples including liver, small intestine and muscle were quickly removed, weighed, put in liquid nitrogen and stored at -80 0 C until determination of cholesterol synthesis. Total cholesterol, HDL and LDL cholesterol were determined using a commercial kit. The results were statistically evaluated with ONEWAY analysis of variance procedure (SYSTAT).

Hamsters fed the high cholesterol diet had significantly higher serum total cholesterol and LDL cholesterol than did those fed the normal I II I WO 96/10033 PCT/CA95/00555 -17cholesterol (0.025%) diet. The supplementation of phytosterol at levels of and 1% remarkably abolished these increases induced by high cholesterol consumption (Figures 11 and 12). The LDL cholesterol concentration in group was lower compared to the levels in hamsters fed normal cholesterolcontaining diets (Figure 12). Furthermore, there was negative regression association of total cholesterol and LDL cholesterol to the level of phytosteroladded in diet (Figure 13).

Supplementation of phytosterol caused a slight increase in HDL, but without yielding a significant difference '(Figure 13).

EXAMPLE 4 Evaluation of Phytosterol Composition Effects in Hamsters 90 day trial Six groups of 20 hamsters (10 males, 10 females) were fed semi-purified diets containing 30% fat (polyunsaturated/saturated fat ratio 0.3) for 90 days. Diet 1 was cholesterol free. Diets 2-6 contained 0.25% (wt/wt) dietary cholesterol. Diets 3 and 4 contained Forbes phytosterols (greater than 90% purity) at 0.5 and respectively. Diets were made from S: primary ingredients every week. Fat, phytosterol and cholesterol levels were determined by gas liquid chromatography. All animals had free access to water and diets throughout the experimental period. Animals were weighed weekly. Food intake was also determined every day and averaged per week by weighing food cups before and after each 24h feeding period. After days feeding, animals were sacrificed using halothane and blood was collected for lipoprotein profile analyses. Circulating total, apoB containing particles 25 and HDL cholesterol and triglyceride levels were determined. Also, just prior to sacrifice, cholesterol absorption and synthesis rates were determined using 14C-cholesterol disappearance from the gut and deuterium incorporation into tissue cholesterol methods, respectively. Uptake of phytosterols into intestine and other tissues was also examined. In addition, samples of intestine and WO 96/10033 PCT/CA95/00555 -18liver were stored to be provided to Bio-Research Laboratories of Senneville, Quebec, for histopathological, carcinogenicity and enzyme function analyses.

Results: Results are shown in Figures 14-18. Groups in these Figures are often referred to by number, which correspond to those described in Experimental Design. Letters above bars in bar-graphs identify significant differences between groups. Where letters are given, bars sharing the same letter are not significantly different, while those with different letters are different at a statistical level of p<0.05.

Shown in Figure 14-16 are the circulating cholesterol data for male and female hamsters consuming the test diets over 90 days. Significant effects of sex were observed in total circulating cholesterol levels for animals consuming diet 2, the base diet with added. cholesterol only, and diet containing cholesterol soyabean phytosterols. Females, although no different from males in cholesterol levels on the basal diet (group exhibited a higher response to added cholesterol alone compared with males. Adding phytosterols negated this difference, with the exception of group Cholesterol level data broken down into sexes are shown in Figures 15 and 16. For males (Figure 15), addition of cholesterol alone to the basal diet resulted in a significant increase in total circulatory cholesterol level.

Addition of Forbes phytosterols at 0.5 percent resulted in a trend towards a decrease in cholesterol level, however, addition of Forbes phytosterols at 1% elicited a statistically significant reduction in cholesterol, to approximately the 25 same levels of the control group without added cholesterol. When soyabean phytosterols were added to the diet at 0.5 or there was no significant decrement in circulating total cholesterol level in males. For HDL, a differential effect of Forbes versus soyabean was observed, where Forbes feeding produced no change in the HDL levels of the group given cholesterol alone (group however, feeding soyabean resulted in a significant decline WO 96/10033 PCT/CA95/00555 -19in HDL values at both levels tested (groups 5 and There were no significant differences across apoB containing cholesterol particles, however, there was a trend towards lower levels with the feeding of cholesterol Forbes at compared with other groups.

Data for females are shown in Figure 16. For total and HDL cholesterol, there was a pronounced influence of adding dietary cholesterol alone. Addition of either type of phytosterol source at 1% resulted in significant and similar declines in total and HDL cholesterol concentrations.

ApoB containing particle levels were not influenced by diet.

Circulating triglyceride levels in hamsters consuming the test diets for 90 days are shown in Figure 17. There was an increase in circulating triglyceride levels in female animals given the basal diet with cholesterol and Forbes 0.5 compared to the basal diet alone, however, no other inter-group differences were observed in either sex. In males, there was no effect or trend of diet on triglyceride concentrations.

In summary form, the ranking of male results on the test diets are as follows: Table 20 Total Cholesterol LDL HDL Forbes 0.5% 4 2 1 Forbes 1.0% 1 1 2 Soya 0.5% 2 34 4 Soya 1.0% 3 4-3 3 The advantage of the Forbes compositions (those of the present invention) can be clearly seen. In addition, similar rankings were found for the HDL:ApoB ratio in males (Figure 18): Forbes 0.5% 2 Forbes 0.1% 1 WO 96/10033 PCT/CA95/00555 Soya Soya It was also found that the HDL:LDL ratio for the Forbes compositions were almost double those of B-sitosterol alone.

EXAMPLE 5 Effects of Phytosterol Composition on Rabbits In this study, two rabbits were assessed over 43 days with respect to the effects of one of the compositions of the present invention (Forbes 1% in diet) on their total cholesterol profiles. The results are as follows: Table 6: Rabbit Total Cholesterol Profiles (mg/dl) S.

S

S

Date 14/06/95 18/06/95 26/06/95 07/07/95 09/07/95 18/07/95 20 25/07/95 01/08/95 08/08/95 Rabbit A 95 129 Rabbit B 215 162 starting feeding with Forbes 1% 106 starting feeding with Forbes 1% 112 118 119 114 A decrease in total cholesterol can be seen in both Rabbits A and B 25 during the two weeks of phytosterol composition (Forbes administration.

This effect continued even after the discontinuation of Forbes The effects of the compositions of the present invention on total cholesterol lowering linger past the initial administration phase.

WO 96/10033 PCTICA95/00555 -21- EXAMPLE 6 Effects of Phytosterol Composition on Apo-E Deficient Mice Animals: Nineteen 5-week old male Apo-E deficient mice were purchased from Jackson Laboratory, USA. Animals were randomly divided into 2 groups, 9 animals in control group and 10 mice in experimental Forbes group. After 5 days as an adaptation period mice were bled from tale into capillary tubes and plasma was separated by centrifugation of blood. Mouse plasma lipids were estimated.

Diet: Low-fat, low-cholesterolmouse chow was purchased from Jamieson's Pet Food Distributors Ltd., Vancouver, B.C. Tall-oil derived phytosterols were extracted from tall-oil soap, using the process described in the present invention. The purity and percentage of each individual phytosterol in the mixture of final product were assessed by gas liquid chromatography. The final produce showed up to 95 in purity and contained 69% sitosterol, 15% campesterol and 16% stigmastenol. Mouse chow was ground to a fine powder state. To this powder 0.15% cholesterol (Sigma) was added and mixed well. A portion of this cholesterol- •supplemented diet was repelleted, dried and used for feeding control group of mice, and another portion of that was supplemented with 2% tall-oil 20 extracted phytosterols, repellet, dried and used for feeding of experimental group of mice.

S..Biochemical assays: Plasma total cholesterol and triglyceride were measured using enzymatic kit (Boehringer Mannheim) and HDLcholesterol was estimated by previously published precipitation method using polyethylene glucol 6000.

Body weight and food consumption: Mouse body weight and food consumption were measured weekly.

*t Table 7: Mouse mean body weight (only monthly measurements are reported) _I ~C l-.l~ WO 96/10033 PCT/CA95/00555 Date 20/06/95 24/07/95 22/08/95 12/09/95 p<0.

0 Forbes group 21.42 29.82 32.12 34.16* Control group 21.02 28.00 29.73 30.78 Table 8: Mouse mean weekly food consumption (only monthly measurements are reported) Date 26/06-03/07/95 01/08-08/08/95 15/08-22/08/95 05/09-12/09/95 p<0.05 Forbes group 19.61 32.62* 29.32 35.56 Control group 21.21 29.20 27.10 30.25

S

S S S. *5 *r S

S

S

*5 S Other findings: No side effects were observed with regard to new diets. All animals from both groups look normal, with normal habits including bowel habit. One mouse in the control group was found dead on 11/07/95. Since the mouse body was not kept, the autopsy was not performed. Another mouse from Forbes group was found dehydrated with body weight lost. The animal was sacrificed and the reason for its sickness was found to be due to malocclusion (teeth overgrowth).

Statistical analysis: Results were analyzed using t-test two-samples assuming equal variances.

Results: Results up to the present time are summarized in the following Tables.

SS

WO 96/10033 PCT/CA95/00555 -23- Mouse mean plasma total cholesterol level (mg/dl).

Table 9: Date 20/06/95 18/07/95 06/09/95 p<0.0001 Forbes group 606.28 1027.96* 1168.57* Control group 599.61 1622.56 1508.63 Table 10: Mouse mean plasma triglyceride level (mg/dl).

Date 20/06/95 18/07/95 06/09/95 Forbes group 110.97 210.17 224.48 Control group 120.31 143.71 152.25 Table 11: Mouse mean HDL-cholesterol level (mg/dl).

S

a a a a S a a. *n a Sa Date 08/07/95 p<0.001 Forbes group (9) 42.00* Control group (7) 18.29 It can be seen from the results that the Forbes composition group showed a significant decrease in total cholesterol, an insignificant increase in triglycerides, and a significant increase in HDL cholesterol EXAMPLE 7 Effect of Phytosterol Composition on Hamsters 45 day 25 Trial Fifty GS hamsters were accommodated for two weeks in an animal care facility before feeding them a semi-purified diet for 45 days.

They were divided into five groups fed 0.25% cholesterol along with one of four mixtures of plant sterols: soy bean, tall oil, pure sitostanol, and artificial WO 96/10033 PCT/CA95/00555 -24mixture representing tall oil phytosterols. The control group received 0.25% cholesterol only. Their food intake was monitored during the study period every three days. Their body weight was measured every week, and at the time of tissue collection. Three days before sacrificing the GS hamsters, they were lightly anaesthetized with diethyl-ether and injected intravenously through the jugular vein with 0.4 ml Intralipid containing 0.18 mg. "C-cholesterol.

Directly after the injection, the animals were fed by gavage, 0.6 ml of lipid mixture (coconut, olive, and safflower oils) containing 0.44 mg ')-cholesterol.

Then, the GS hamsters were kept in their wired cages for 72 hours, provided water and food ad libitum.

On the day of sacrificing, each GS hamster was injected i.p., with 1 ml deuterated water, and left for one hour before the killing. The animals were anaesthetized with diethyl ether, and blood samples were collected by cardio-puncture. Liver, gall bladder, small intestine, large intestine, and heart were collected, frozen in liquid nitrogen, and stored in the freezer at -80*C. Their carcasses and faeces were stored at -20°C for further total lipid and'sterol analysis.

Food Intake, Body and Liver Weight Measurements: GS hamsters food intake was measured every three days. The statistical analysis shows no significant difference among the five groups in their food consumption. The average daily intake in the five groups varied from 8.84 to 9.34g per day, p-value 0.4. The animals showed a significant increase in their body weight of about 25 to 40g during the study period, •25 p-value<0.05 (paired t-test). The final measurement of their body weight ranged from 112.5 to 154.3g. No statistical significance was noticed among the different treatment groups, p-value 0.43.

Liver weights varied significantly among the different divisions.

The sitostanol treated group showed the lowest liver weights as compared to the control and other groups fed 0.25% cholesterol and different phytosterols i ii-i- WO 96/10033 PCT/CA95/00555 respectively. In addition, the soy bean treated group presented similar significant difference to the sitostanol treated one when the data are statistically analyzed using Newman-Keuls test. In general, all groups fed plant sterols showed lower liver weight as compared to control fed 0.25% cholesterol, p-value 0.01. The tall oil sterols and the sitostanol fed GS hamsters had an average liver weight of 15% and 20% less than that of the control, respectively. The natural tall oil and the artificially prepared tall oil demonstrated similar values of liver weights suggesting that the missing compound in the soy bean plant sterol, sitostanol, play a major role in decreasing the sterol content in the liver.

Lipid Analysis: I. Total cholesterol: Total cholesterol values were determined by using a commercial enzymatic reagent kits on a VP autoanalyzer. Blood samples were measured twice, and the average of the two values was used in the final statistical analysis. One'way ANOVA, with Newman-Keuls, and Bonferroni methods were used for the different lipid analysis. Sitostanol decreased significantly total cholesterol level in GS hamsters plasma by 34% as compared to the 20 control. The mean value for control group was 226.9 mg/dl, and the one for sitostanol treated group was 151.2 mg/dl, p-value 0.007. Tall oil phytosterols, and the artificial tall oil mixtures showed similar decrease in plasma cholesterol 118.4 mg/dl and 186.1 mg/dl respectively.

However, tall oil phytosterols (Forbes) showed significant decrease in total 25 cholesterol value (175.2 mg/dl) when one off-scale sample value was excluded from the analysis, p-value <0.02 in the tall oil group as compared with the control (0.25% cholesterol only) group. The correlation between the presence of sitostanol and lower plasma level was significant, p-value <0.0001, and r=0.46.

WO 96/10033 PCT/CA95/0055S -26- II. HDL cholesterol: The apoA portion of the lipoproteins present in the HDL cholesterol did not show any significant changes in their values among the five different groups, p-value 0.18. A non significant decrease of 15% in the mean HDL value in the sitostanol treated group was observed. Nevertheless, this element did not affect the significant decrease in the total cholesterol which was of 34%. The ratio of the non apoA lipoproteins to the apoA (HDL) lipoproteins did not vary significantly among the groups. The confounding effect of the non-significant lower values of HDL in the sitostanol and tall oil groups contributed in creating a non significant result in the non apoA/apoA (HDL) values. In general, the different types of phytosterols did not vary the HDL cholesterol level in GS hamsters plasma.

III. LDL and/or non apoA sterols: Sitostanol was efficient in decreasing the non apoA sterols in the plasma. A 55% decrease in the non apoA lipoproteins was shown, p-value i* 0.02. Similrly to their effect on total cholesterol, tall oil and artificial mixture tall oil decreased the non apoA sterols by 21% respectively, again suggesting of a strong correlation existing between sitostanol content in the 20 phytosterols and their beneficial effects in decreasing cholesterol levels in the GS hamsters plasma. However, this decrease was not statistically significant due to the variability in the triglyceride values.

IV. Triglyceride: 25 When applying the one way ANOVA method on the TG values, they did not pass the normality test. GS hamsters were sacrificed in a non fasting condition (important status for the future cholesterol synthesis, kinetics, and absorption analysis). Because of such situation, different values were off scale. With ANOVA on Ranks, the TG levels showed no statistical difference ~1~1~1_1~ WO 96/10033 PCT/C95/00555 -27among the groups. Plant sterols did not affect the plasma TG levels in GS hamsters.

Table 12: Potency Rank Total Cholesterol HDL LDL TG Control C C C C Soybean B-sitostanol 4 1 3-4 1-4 Tall oil (Forbes) 2-3 2-2 2-3 1-4 Sigmastanol 1 4 1 1-4 Artificial tall oil 3-2 3-2 4-3-2 1-4 In conclusion, the tall oil soap-derived composition of the present invention exhibited the most favourable profile due to the increase in HDL cholesterol and decrease in total cholesterol. This HDL effect was not seen in the artificial tall oil composition. Similarly, the overall effect of sitostanol is not favourable due to the significant decrease in HDL.

Although it is not entirely clear, it does appear that, with respect to plant sterols, the relatively hydrophobic sterols inhibit more total cholesterol 2 0 absorption while the relatively hydrophilic sterols have more influence on the level of HDL. The phytosterol compositions of the present invention are unique in that both of these effects are preserved.

Claims (13)

1. A process for isolating a phytosterol composition from a plant-derived pulping soap said soap not being derived from a sulphate degradation process which comprises: in a first phase, mixing the pulping soap with a solvent mixture comprising a ketone selected from the group having the general structure RCOR 1 where R and R 1 are alkyl groups, an aliphatic hydrocarbon selected from C5-C10 hydrocarbons and water and having no alcohol, at a temperature generally from 250 to 1500 C to form a creamy precipitate; and in a second phase, purifying from the creamy precipitate a phytosterol composition.

2. The process of claim 1 wherein the phytosterol composition is purified from the creamy pr4cipitate by crystallization. C

3. The process of claim 1 wherein the pulping soap is produced from wood chips consisting of cedar, fir, pine, spruce, oak, hemlock and poplar.

4. A cholesterol-lowering composition comprising no more than 70% by 0 weight of beta-sitosterol, at least 10% by weight of campesterol and further including stigmastanol.

5. The composition of claim 4 in which campesterol and stigmastanol together comprise at least 50% of the concentration of beta-sitosterol.

6. The composition of claim 4 wherein the ratio of beta-sitosterol is Scampesterol is between 0.2 and 0.4 and stigmastanol is between 0.2 and

7. The composition of claim 4 wherein the ratio of beta-sitosterol is campesterol is 0.354 and stigmastanol is 0.414. L-=i -ii~ii-ii-iilll-LLliIIL1-~_illll-._iD~I~ -29-

8. The composition of claim 4 wherein the ratio of beta-sitosterol is campesterol is 0.330 and stigmastanol is 0.203.

9. The composition of claim 4 comprising a ratio of phytosterols as follows: 1.0 beta-sitosterol to 0.268 campesterol to 0.299 stigmastanol.

A composition prepared according to the process described in claim 1.

11. The use of the composition described in claim 4 to manufacture a medication to lower the plasma concentration of low-density lipoprotein- cholesterol.

12. A product therapeutically effective to prevent or treat primary and secondary dislipidemias and atherosclerosis comprising the composition defined in claim 4 ahd a pharmaceutically effective carrier therefor. 15

13. The composition prepared by the process of any one of claims 1, 2 or 3 and comprising phytosterols and co-occurring compounds selected from Striterpenes, long chain alcohols and other alcohol soluble organic compounds. SDated this 15th day of November 1999 THE UNIVERSITY OF BRITISH COLUMBIA By their Patent Attorneys COLLISON CO.