Chlorella vulgaris microalgae

Lyophilized C. vulgaris biomass was obtained from the algal biotechnology unit, National Research Centre (Egypt). Inoculum was prepared as described by El-Sayed et al. (Reference El-Sayed, Abdalla and Abdel-Maguid2001) using BG-II growth medium (Stainer et al. Reference Stainer, Kunisawa, Mandel and Cohen-Bazire1971). Continuous light illumination was provided from daylight lamps (10 × 40 w). Aeration was achieved using an oil-free air compressor (HIBLOW AIR PUMP, type SPP-100GJ-H, Japan) through a 3-mm polyethylene tube. Room temperature was adjusted to 27 ± 2 °C during the whole incubation period. Incubation was carried out using fully transparent polyethylene bags (75 × 5 cm² and 100 µm thickness) containing 2·5 litres of algal broth. Mass production of C. vulgaris was performed within a 1000-litre Zigzag photobioreactor (El-Sayed et al. Reference El-Sayed, Battah and Wehedy2015). For harvesting and cleaning of the obtained biomass, a series of precipitation and washing was performed using tap water and a cooling centrifuge (RUNNE, HIDEBERG, RSV-20, Germany).

Chlorella vulgaris was analysed for fatty acids using fatty acid methyl esters prepared by base-catalysed methanolysis of the glycerides (potassium hydroxide [KOH] in methanol). A Perkin-Elmer chromatograph (model 8420, Beaconsfield, Perkin Elmer, Beaconsfield, UK) equipped with a flame ionization detector was used. Helium was the carrier gas at 30 cm/sec, column temperature was initially 150 °C for 1 min, and then increased by 4 °C/min to 200 °C, and was then held at 200 °C for 10 min. Peaks in chromatographs were identified and quantified using pure methyl ester standards (Sigma-Aldrich Inc., Bellefonte, PA, USA). The fatty acid profile was (/kg DM): 18·5 g C14:0, 525·6 g C16:0, 158·8 g C17:1, 88·3 g C18:0, 54·2 g C18:1, 65·6 g C18:2, 88·7 g C18:3, 632·4 g total saturated fatty acids (SFA) and 367·3 g total UFA.

The amino acid profile of C. vulgaris was determined as described by Spackman et al. (Reference Spackman, Stein and Moor1958). Analysis was performed using an LC 3000 amino acid analyser (Eppendorf-Biotronik, Maintal, Germany) at the Central Service Unit, National Research Centre, Egypt. The analysis was based on the separation of amino acids using strong cation exchange chromatography followed by the ninhydrin colour reaction and photometric detection at 570 nm. The amino acids content was (/kg DM): 105 g aspartic acid, 52·4 g threonine, 50·8 g serine, 107 g glutamic acid, 51·0 g glycine, 84·4 g alanine, 64·4 g valine, 15·0 g methionine, 50·1 g isoleucine, 68·4 g leucine, 52·0 g tyrosine, 42·0 g phenylalanine, 50·2 g histidine, 56·0 g lysine, 82·0 g arginine and 64·0 g proline.

Goats, feeding and experimental design

One week after parturition, 15 multiparous lactating Damascus goats (44 ± 0·8 kg body weight) were assigned randomly to three experimental groups (five goats per treatment) for 12 weeks. The experimental design was a completely randomized design. Goats were housed individually in soil-surfaced pens (1·5 m²/goat), with unrestricted access to water and experimental diets to meet their energy and nutrient requirements according to NRC (2007) recommendations. Adjustments were made to the diets to ensure collection of orts.

The basal diet fed to the goats contained (/kg DM) 500 g of Egyptian berseem clover (Trifolium alexandrinum) and 500 g of concentrates feed mixture (containing per kg DM: 250 g un-decorticated cotton seed meal, 350 g wheat bran, 300 g maize, 30 g rice bran, 30 g molasses, 20 g limestone, 10 g urea and 10 g salt). Chemical composition (g/kg DM) of ingredients fed to the Damascus goats is shown in Table 1. Chemical composition (g/kg DM) and nutritive value of experimental diets fed to the Damascus goats are shown in Table 2.

Table 1.

Chemical composition of ingredients fed to lactating Damascus goats (g/kg DM)

* Consisted of (/kg DM): 250 g cotton seed meal, 350 g wheat bran, 300 g maize, 30 g rice bran, 30 g molasses, 20 g limestone, 10 g urea and 10 g salt.

Table 2.

Chemical composition and nutritive value of experimental diets fed to the lactating Damascus goats (g/kg DM)

* The basal diet based on 500 g/kg DM of concentrates feed mixture and 500 g/kg DM berseem clover (Trifolium alexandrinum) without addition of Chlorella vulgaris microalgae (Control) or with addition of Chlorella vulgaris microalgae at 5 (Alg05) and 10 (Alg10) g/goat/day.

† Calculated according to (NRC 2001) using the obtained nutrients digestibility coefficients.

Goats were supplemented with C. vulgaris at 0 (Control), 5 (Alg05) and 10 g/goat/day (Alg10). Diets were offered to each goat individually at 08.00 and 16.00 h in two equal portions. Daily allocation of C. vulgaris for each goat was mixed with the diet individually and fed at 08.00 h. Samples of the berseem clover and concentrate mixtures were taken daily, composited weekly and dried at 60 °C in a forced-air oven for 48 h. Samples were ground to pass a 1-mm screen using a Wiley mill and then stored for chemical analysis.

Digestibility and chemical analysis

Three digestibility trials were conducted at the 4th, 8th and 12th week (7 days for each collection period) of the experiment. During these periods, DM intake was recorded daily by weighing feeds offered and orts. Faecal grab samples were collected twice daily at 07.00 and 15.00 h, dried at 60 °C in a forced-air oven for 72 h (AOAC 1997, method 930·15) and pooled by goat within period. An acid-insoluble ash was used as the internal indigestibility marker (Sales & Janssens Reference Sales and Janssens2003). Acid-insoluble ash contents of feeds and faeces were determined gravimetrically after drying, ashing, boiling of ash in hydrochloric acid (HCl), filtering and washing of the hot hydrolysate, and re-ashing. Coefficients of digestion were calculated according to Ferret et al. (Reference Ferret, Plaixats, Caja, Gasa and Prio1999). Digestible nutrients and energy value of diets were calculated according to NRC (2001).

Feed, orts and faecal samples were ground through a 1 mm screen using a Wiley mill (Arthur H. Thomas, Philadelphia, PA, USA) and analysed for DM (method 930·15), ash (method 942·05), nitrogen (N; method 954·01) and ether extract (EE; method 920·39), according to AOAC (1997) official methods. Neutral detergent fibre (NDF) was determined by the procedure of Van Soest et al. (Reference Van Soest, Robertson and Lewis1991) without the use of alpha amylase but with sodium sulphite. Acid detergent fibre (ADF) was analysed according to AOAC (1997; method 973·18). Acid detergent lignin (ADL) was analysed by solubilization of cellulose with sulphuric acid in the ADF residue according to Van Soest et al. (Reference Van Soest, Robertson and Lewis1991). Non-structural carbohydrates (NSC; [NSC (g/kg DM) = 1000 − ((NDF (g/kg DM) + crude protein (CP; g/kg DM) + EE (g/kg DM) + ash (g/kg DM))], cellulose [cellulose (g/kg DM) = NDF (g/kg DM) − ADL (g/kg DM)], hemicelluloses [hemicellulose (g/kg DM) = NDF (g/kg DM) − ADF (g/kg DM)] and organic matter (OM; [OM (g/kg DM) = 1000 − ash (g/kg DM))] were calculated.

Sampling and analysis of rumen fluid

On day 7 of the 8th and 12th week of the experiment, ruminal contents were sampled at 3 h post-morning feeding to determine pH and concentration of fermentation end-products. Rumen contents (~100 ml) were collected once from the ventral sac by using a stomach tube and hand pump, and then composite samples taken from each goat were strained through four layers of cheesecloth. The pH of ruminal fluid was measured immediately using a pH meter (HI98127 pHep^®4 pH/Temperature Tester, Hanna^® instrument, Italy).

A sub-sample of 5 ml was preserved in 5 ml of 0·2 M HCl for ammonia-N analysis and 0·8 ml of rumen liquor was mixed with 0·2 ml of a solution containing 250 g of metaphosphoric acid/l for total volatile fatty acid (VFA) analysis. Samples were stored at −20 °C until laboratory analysis. The concentration of ruminal ammonia-N was determined according to AOAC (1997; method 954·01). Total VFA concentration in samples was determined by titration, after steam distillation of a 4 ml sample according to Annison (Reference Annison1954). The molar proportions of the individual VFA were measured by gas-liquid chromatography (Varian 3700; Varian Specialties Ltd, Brockville, Ontario, Canada). Separation was carried out with a capillary column (30 m × 0·25 mm internal diameter, 1 m film thickness, Supelco Nukol; Sigma–Aldrich, Mississauga, ON, Canada) and flame ionization detection. The column temperature was adjusted to 100 °C for 1 min, then increased by 20 °C/min to 140 °C, then by 8 °C/min to 200 °C and held at this temperature for 5 min. The injector temperature was 200 °C while the detector temperature was 250 °C. Helium was used as the carrier gas.

Sampling and analysis of blood serum

On day 7 of the 8th and 12th week, blood samples (10 ml) were taken 4 h after feeding from the jugular vein of each goat into a clean dry tube, without anti-coagulants. Blood samples were centrifuged at 4000  g at 4 °C for 20 min. Serum was separated into 2-ml clean dried Eppendorf tubes and frozen at −20 °C until analysis.

Blood serum samples were analysed for concentrations of total protein, albumin, urea-N, glutamate-oxaloacetate transaminase (GOT), glutamate-pyruvate transaminase (GPT), glucose and cholesterol using specific kits (Stanbio Laboratory, Boerne, Texas, USA) following manufacturer's instructions. Globulin concentration was calculated by subtracting albumin values from their corresponding total protein values.

Milk sampling, milk composition and fatty acid analysis

Goats were milked by hand twice daily at 09.00 and 21.00 h in the 4th, 8th and 12th week (7 days per week), and samples (100 g/kg of recorded milk yield) were collected at each milking. A mixed sample of milk (proportional to amounts produced in the morning and evening) was taken daily. Milk samples were analysed for total solids, fat, protein and lactose using infrared spectrophotometry (Foss 120 Milko-Scan, Foss Electric, Hillerød, Denmark). The ash content of milk was determined by heating samples in a muffle furnace at 550 °C for 8 h. Fatty acids in milk were determined as described previously in Kholif et al. (Reference Kholif, Khattab, El-Shewy, Salem, Kholif, El-Sayed, Gado and Mariezcurrena2014) using methyl esters prepared by base-catalysed methanolysis of the glycerides (KOH in methanol) according to International Standards (ISO-IDF 2002) on a Perkin-Elmer chromatograph (model 8420, Beaconsfield, Perkin Elmer, Beaconsfield, UK) equipped with a flame ionization detector. Athrogenicity index (AI) was calculated according to Ulbricht & Southgate (Reference Ulbricht and Southgate1991) as:

$$AI\, = \,(C12:0\, + \,4\, \times \,C14:0\, + \,C16:0)/\sum {\rm of \;UFA}.$$

Average yield (g/d) of each milk component was calculated by multiplying milk yield by the component content (g/kg). Gross energy content in milk was calculated according to Tyrell & Reid (Reference Tyrell and Reid1965) as: Milk energy content (MJ/kg) = 4·184 × 2·204 × [(41·63 × fat (g/100 g) + 24·13 × protein (g/100 g) + 21·60 × lactose (g/100 g) − 11·72)/1000]. Milk energy output (MJ/d) was calculated as milk energy (MJ/kg) × milk yield (kg/d).

Energy corrected milk (ECM) was calculated according to Sjaunja et al. (Reference Sjaunja, Baevre, Junkkarinen, Pedersen, Setala, Crettenand, Moll, Mosconi and Wegmann1991) as: ECM (kg/d) = milk (kg/d) × [38·3 × fat (g/kg) + 24·2 × protein (g/kg) + 16·54 × lactose (g/kg) + 20·7]/3140.

Statistical analysis

Data on intake, apparent diet digestibility, ruminal fermentation, blood chemistry parameters, milk yield and composition (including fatty acid profile) were analysed using a completely randomized design with repeated measurements in time, where each goat was an experimental unit using PROC MIXED of SAS (2006) with the following model:

$$Y_{ijk{\rm l}} = m + T_i + A_j(T_i) + W_k + (T\times W)_{ik} + {\rm} E_{ijk{\rm l}}$$

where Y _ijkl expressed each observation of the jth goat in the kth sampling time given ith treatment, T _i expressed the treatments effect, A(T) _ji expressed the goat within treatments, W _k expressed the sampling week effect, (T × W) _ik expressed the interaction between the treatments and sampling week, and E _ijkl expressed the experimental error. Treatments × sampling week interactions were insignificant for most of the measured parameters and therefore are not reported; only the main effects of diets are reported. The Tukey test was used for multiple comparisons of means, and polynomial (linear and quadratic) contrasts were used to examine dose responses to increasing levels of C. vulgaris in the diet. Significance was declared at a level of P < 0·05.