**5.4 Cooking chicken breast**

Chicken breast samples of 150 g were used. Before cooking, each sample was doused either in the pure olive oil (control) or in oil with a carotenoid. The oven was pre-heated to 275°C. A probe thermometer with a digital reading was placed in the thickest point of the sample. The sample was then placed in the oven and a reading taken and recorded every 300 sec for the 900-sec duration of the cooking.

## **5.5 Cooking salmon**

Wild pacific Keta salmon fillets, 115 g each, were oven-cooked in individual small aluminium foil containers. The following preparation conditions were applied before cooking: either addition of 25 mL of the olive oil containing Lycopene 0.23 mg/mL) or control—addition of 25 mL of pure olive oil. Salmon fillets were turned several times making sure that they are completely covered with the added oil. After that, 500 mg of salt and 5 mL of fresh lemon juice were applied on the surface of each portion of fish. During the cooking process, the measurement of the internal (doneness) temperature of the fish was performed using a digital thermometer at the following time points: 8 min, 12 min, 16 min, and 20 min. All the containers had to be taken out of the oven for temperature measurements; hence, only the time when the fish was in the oven was counted as the cooking time. Once the internal temperature of the fish

reached 62°C (doneness temperature for salmon), a small (about 1 g) fragment of fish was immediately taken from the relevant piece (using a scalpel blade and thumb forceps) and placed in a 15-mL laboratory tube containing 1 mL of distilled water. An additional sample of fish was also taken before cooking.

#### **5.6 Salmon sample preparation**

Samples were transferred to the laboratory and weighed using analytical scales (Discovery DV114C, OHAUS Corp.). Distilled water was added to each sample apart from the 'juice' and 'sauce' samples to provide the ratio of 9 mL of water per 1 g of sample (these samples were regarded as 1/10 dilutions). 1 mL of distilled water was added to the 'juice' and 'sauce' samples to produce 1/2 dilutions. Following this step, all the samples were homogenised using IKA T10 basic Ultra-Turrax homogeniser system at maximum speed (30,000 RPM). After the homogenisation of each sample, the homogeniser was disassembled, and both its rotor and stator were carefully cleaned in order to prevent sample cross-contamination. Following homogenisation, second dilutions with four volumes of distilled water (1-mL homogenate + 4 mL of water) were prepared from the homogenates of the raw fish, fish cooked with water and fish cooked with either pure olive oil, or containing lycopene, or astaxanthin (resulting in 50× dilutions). Similar dilutions were made from salmon 'juice' and 'sauce' samples described above (resulting in 10× dilutions). Finally, additional fivefold dilutions were made from the 50× dilutions of homogenates prepared from all salmon samples (resulting in 250× dilutions).

#### **5.7 Vitamin B12 concentration determination**

All samples were analysed within 48 hours following cooking experiments. Vitamin B12 BioAssayTM ELISA Kit (US Biological) was used for Vitamin B12 concentration determination. Vitamin B12 concentrations in the samples from the cooking experiment were determined in both undiluted supernatants (1 g of liver + 19 mL of buffer) and dilutions 1/2, 1/4 and 1/8 (the latter only for the samples cooked without lemon juice). The dilutions were prepared using sample dilution buffer (phosphatebuffered saline—PBS) supplied with the kit. B12 concentration evaluation was performed in 50 μL of solution according to the protocol provided with the kit.

Vitamin B12 concentration determination was performed using Multiscan FC microplate photometer (Thermo Fisher Scientific) by measuring optical light absorbance at 450 nm (reference wavelength 620 nm) as recommended by the kit manufacturer. All the calibration standards were measured in duplicates. Measurement results were analysed using SkanIt software for Multiscan FC system (a fourparameter logistic algorithm was applied). Vitamin B12 concentrations in the original samples were obtained by re-calculation taking into account sample dilutions during material processing. Once all the measurements were completed, the results of the two cooking experiments were combined by taking average concentration value for each set of conditions.

#### **5.8 Formation of cholesterol crystals**

A solution of 198 mL of 1 g of cholesterol (Sigma) in 99% ethanol was divided in two equal parts. In one part, 1 mL of ethanol containing 100 μg of the dissolved lycopene was added. This provided a ratio of lycopene to cholesterol as 1:106 .

Into the other part, the control, 1 mL of the ethanol itself was added.

Then, both the samples were left in a dark room for evaporation under 20–22°C. Recording of the status of both the samples was made at least daily or at even shorter intervals.

#### **5.9 Disassembly of cholesterol crystals**

20 mL of ethanol solutions with different concentrations of lycopene were added to 100 mg of crystallised cholesterol. As a control, 20 mL of the same ethanol was used but without any lycopene. After gentle stirring for a couple of minutes, at the room temperature of about 20–22°C, recording of the results was made.

#### **5.10 Ex vivo experiment with cholesterol crystals in the arterial wall**

It was important to try to check whether carotenoids, and lycopene in particular, could affect the folding of cholesterol crystals, which are not just synthesised by a manufacturer, but produced naturally, and particularly those, which are developed during pathological process in human. For this purpose, we used pieces of atherosclerotic abdominal aorta, which were obtained, during a combined vascular graft and bypass surgery.

Comparable types of atherosclerotic lesions were collected. This was in terms of their stages of development, with prominently featured cholesterol crystals of similar size, embedded into the atheromatous tissues of the aorta wall. As a control material, we collected the pieces of atherosclerotic abdominal aorta containing calcium phosphate crystals.

First, an ethanol solution 1 μg/mL of lycopene was prepared. It was then diluted by PBS 10-fold. As a control, solution was made with the same ratio ethanol to PBS but without lycopene. Then, the pieces of aorta were incubated, in light-protected containers at a room temperature of about 20–22°C for 13 days.

#### **5.11 Clinical parameters**

The body mass index (BMI) of the participants was measured in the morning and calculated in kg/m2 . Pulse rate and systolic and diastolic blood pressure were measured three times in the left arm of the seated patient after 15 min of rest. The time between measurements was no less than 2 min. The mean number for each parameter was calculated.

All the body and vascular parameters were measured in the morning between 8 and 10 am.

#### **5.12 Blood biochemistry**

Biochemistry and inflammatory markers, glucose, total cholesterol (TC), triglycerides (TG), high-density cholesterol (HDL), low-density cholesterol (LDL), C-reactive protein (CRP) and oxidised LDL (LDL-Px) were measured using commercially available analytical kits according to manufacturers' instructions (BioSystems, Medac, R&D Systems). Inflammatory oxidative damage (IOD—malondialdehyde) was measured applying a colorimetric method [7].

#### **5.13 Statistics**

For the assessment of normally distributed parameters, the Shapiro-Wilkinson method was used. Student's *t*-test was then applied for both the paired and unpaired samples.

In cases where parameters were not normally distributed, the Mann-Whitney U test and the Kruskal-Wallis test were used.

The analysis of variance (ANOVA) and the analysis of covariance (ANCOVA) were used with post hoc analysis (Statistica9 suit, StatSoft; Inc.). Statistical significance between two-tailed parameters was considered to be p<0.05.
