Vitamin E concentration in alpaca meat and its impact on oxidative traits during retail display

Accepted Manuscript Vitamin E concentration in alpaca meat and its impact on oxidative traits during retail display

Melanie A. Smith, Courtney L. Nelson, Tamara E. Biffin, Russell D. Bush, Evelyn J.S. Hall, David L. Hopkins PII: DOI: Reference:

S0309-1740(18)30873-8 https://doi.org/10.1016/j.meatsci.2019.01.004 MESC 7753

To appear in:

Meat Science

Received date: Revised date: Accepted date:

13 September 2018 3 December 2018 15 January 2019

Please cite this article as: Melanie A. Smith, Courtney L. Nelson, Tamara E. Biffin, Russell D. Bush, Evelyn J.S. Hall, David L. Hopkins , Vitamin E concentration in alpaca meat and its impact on oxidative traits during retail display. Mesc (2019), https://doi.org/10.1016/ j.meatsci.2019.01.004

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Vitamin E concentration in alpaca meat and its impact on oxidative traits during retail display Melanie A. SmithA,* , Courtney L. NelsonA, Tamara E. BiffinA, Russell D. BushA, Evelyn J. S. HallA, and David L. HopkinsB The University of Sydney, Sydney School of Veterinary Science, Faculty of Science,425

PT

A

Centre for Red Meat and Sheep Development, NSW Department of Primary Industries,

SC

B

RI

Werombi Road, Camden, NSW, 2570, Australia

NU

Cowra, NSW 2794, Australia

AC C

EP T

ED

MA

*Corresponding author: [email protected]

ACCEPTED MANUSCRIPT Abstract. The longissimus thoracis et lumborum (LL), and adductor femoris (AF) muscles from 39 castrated, 23 (± 1) month old huacaya alpacas were used to determine vitamin E content and the impact on lipid oxidation levels. At 24 h post death the LL and AF muscles were removed and sampled for meat quality analysis and subjected to simulated retail display. Vitamin E content of either muscle had no significant impact on colour stability or oxidation

PT

traits during retail display. This is thought to be due to the high levels of vitamin E (> 5.4 mg/kg) in both muscles. Lipid oxidation levels were 0.2 mg MDA/kg higher in both muscles

RI

post retail display. However, overall differences in TBARS levels detected pre and post display were very low (< 1.19 mg MDA/kg) and well below sheep threshold values of > 3 mg

SC

MDA/kg. The mechanism behind why alpaca meat has such high vitamin E levels compared

NU

to other species requires further investigation.

AC C

EP T

ED

MA

Key words: alpaca, meat colour, meat quality, oxidation, vitamin E

ACCEPTED MANUSCRIPT 1.

Introduction Currently, the factors contributing to the stability of alpaca (Vicugna pacos) meat

quality parameters during product display (Smith, Bush, van de Ven, Hall, et al., 2017) and the influence of naturally occurring antioxidants such as vitamin E are unknown. During retail display meat products are vulnerable to product deterioration, resulting in unfavourable odours, aromas and increased meat surface browning (Mancini & Hunt, 2005). In common red meat species, this change in surface colour during retail display has been linked to

PT

reduced consumer appeal and discounted retail price (Hopkins, Lamb, Kerr, van de Ven, & Ponnampalam, 2013; Khliji, van de Ven, Lamb, Lanza, & Hopkins, 2010). The majority of

RI

deterioration is caused by oxidisation where lipids and proteins come into contact with

SC

oxygen (Mancini & Hunt, 2005). The rate of oxygen consumption of meat varies between muscles and species due to the proportion of aerobic or anaerobic muscle fibres (Lawrie,

NU

1998). However, the mechanisms driving the degree of oxidation in meat are primarily determined by the availability of antioxidants, the amount of oxygen available, the intramuscular fat (IMF) content and proportion of poly-unsaturated fatty acids within the IMF

MA

(Falowo, Fayemi, & Muchenje, 2014). Lipid soluble antioxidants, including vitamin E (αtocopherol), play an important role in slowing down the rate of oxidation by reducing the

ED

amount of free radicals which would otherwise initiate oxidation (Descalzo & Sancho, 2008). Alpaca meat has low levels of IMF (< 1 %) with correspondingly low levels of lipid oxidation (< 2.44 mg MDA/kg meat) and minimal surface browning across 72 h of simulated

EP T

retail display (Smith, Bush, van de Ven, Hall, et al., 2017). These low levels of lipid and colour oxidation are favourable, when compared to other red meat species, as the product appears to remain in a saleable state without having to be discounted due to lipid oxidation

AC C

and/or unacceptable levels of product brownness (Hopkins et al., 2013). Currently, it is not known if naturally occurring antioxidants, such as vitamin E, are contributing to the low levels of oxidation observed in alpaca meat (Smith, Bush, van de Ven, Hall, et al., 2017). Vitamin E content of meat has been shown to help reduce oxidation (Bekhit, Hopkins, Fahri, & Ponnampalam, 2013) and that it can be naturally obtained by animals through the consumption of green pasture (Daley, Abbott, Doyle, Nader, & Larson, 2010). There has been limited research into α-tocopherol (vitamin E) levels in alpaca meat with only one Peruvian study reporting very low levels (0.31 (± 0.21) µg g-1 ) with high variability in longissimus thoracis et lumborum (LL) (Salvá, Zumalacárregui, Figueira, Osorio, & Mateo, 2009). However, any potential correlations between vitamin E content and lipid and protein oxidation are absent. Given that alpacas are primarily produced in a pasture

ACCEPTED MANUSCRIPT based system that is likely to naturally provide adequate levels of vitamin E, further research is required to explain if vitamin E content is contributing to the low oxidation levels observed in alpaca meat. The aims of this study were to firstly investigate the levels of vitamin E in two alpaca muscles; and secondly determine if vitamin E impacted on lipid oxidation and meat colour stability traits across a retail display period of 72 h.

Materials and Methods

PT

2.

RI

2.1. Experimental Design

Thirty nine castrated 23 (± 1) month old huacaya alpacas were randomly selected from

SC

a commercial property on the Southern Tablelands of New South Wales, Australia. The animals were randomly assigned to one of two processing groups (n = 24 animals assigned to

NU

group 1 and n = 15 assigned to group 2). The two groups were processed a fortnight apart. The unbalanced animal allocation was due to the muscle samples being obtained from control

MA

animals included in a larger processing experiment which prevented balanced numbers across both processing dates. To overcome this, processing day was included within the statistical models to account for any potential variation. This was an observational study, in that there

ED

were no true nutritional treatments applied to influence the vitamin E levels. The animals included within the study were representative of the Australian alpaca population and were

EP T

grazed under the same conditions prior to the study commencing to ensure all animals had the same dietary back ground.

Animal Measurements

AC C

2.2

Individual animal measurements were recorded on all animals on farm at 0, 5 and 8 weeks before the processing of each group as per standard husbandry procedures. Each individual was identified (animal I.D.) by unique numbers on ear tags and radio frequency identification (RFID) tags. Animal liveweight (LW) measurements were recorded for all animals using GallagherT M Weigh Scales and Livestock Manager TSi 2 Data Recorder (Gallagher

Group

Limited,

Epping

VIC,

Australia).

Body

condition

score

(BCS)

measurements were taken using a conventional 0.25 increment 1 – 5 scale, measured on the lumbar region of each animal, similar to the sheep method described by Keinprecht et al. (2016). 2.3

Pasture

ACCEPTED MANUSCRIPT 2.3.1 Pasture Collection Pasture samples were collected at 0, 5 and 8 weeks prior to the animals being processed. A randomised quadrate sampling approach was used to determine qualitative and quantitative pasture traits. A total of 6 samples were taken from representative areas of the paddock per collection and were cut from ground level in a 30 cm x 30 cm (900 cm2 ) quadrate sampling area. A fresh pasture weight was recorded at the point of collection prior to samples being frozen (- 20 ºC) and transported back to the laboratory for dry matter (DM)

PT

determination.

A secondary vitamin E sample was collected at the same time as the quadrate sample.

RI

At each collection site an additional grab sample of pasture was taken and pooled to generate

SC

one representative vitamin E sample per collection. Air was removed from the sample, the bag sealed, and frozen at - 20 ºC to minimise sample oxidation prior to analysis.

NU

2.3.2 Pasture Analysis

Samples from each collection date were pooled, dried at 100ºC for 24 h and ground to a

MA

1 mm particle size. Samples were then analysed in duplicate for: analytical dry matter, neutral detergent fibre (NDF), acid detergent fibre (ADF) and crude protein (CP) following the official methods of analysis protocols (AOAC, 2005). The NDF and ADF analysis was

ED

conducted based on guidelines developed by Van Soest (1963) using F57 filter bags (ANKOM, New York, USA) and an automated ANKOM 200 fibre analyser (A2000

EP T

ANKOM Technologies, New York, USA). Crude protein was calculated by determining the nitrogen (N) content of pasture by combustion using a Leco-428 Analyser (Michigan, USA). The CP content of the pasture was calculated on a standard conversion of CP = N x 6.25

AC C

(McDonald, Edwards, Greenhalgh, & Morgan, 2002). The metabolisable energy (ME) content of pasture was determined from the fibre and nitrogen content of pasture using methods previously described by NSW Agriculture (1983) and Oddy, Robards, and Low (1983). Where digestible dry matter (DDM) was firstly calculated according to the formula: DDM % = 83.58 – 0.824 (ADF %) + 2.62 (N %) and then ME was calculated using the formula: ME (MJ/ kgDM) = 0.17 (DDM %) – 2. Pasture vitamin E content was analysed through high performance liquid chromatography (HPLC) measuring α-tocopherol content using methods described by McMurray & Blanchflower (1979). 2.4

Animal Processing Procedure Each group of animals were transported (approximately 3 h driving) to a commercial

ACCEPTED MANUSCRIPT camelid certified abattoir on the South Coast of New South Wales for processing. Prior to processing, all animals were subjected to a 24 h period of lairage with unrestricted access to water. Animals were processed in their respective groups using methods previously described by Smith, Bush, Thomson, & Hopkins (2015). 2.5

Meat Quality Analysis

2.5.1 Muscle Sampling

PT

The m. adductor femoris (AF) and m. longissimus thoracis et lumborum (LL) (n = 39 samples/muscle) were removed from one side of each carcass 24 h after death (average

RI

chiller temperature 3.1 ºC and humidity 92 %). The AF and the cranial (rack) portion of the

SC

LL were sub sectioned into samples for; vitamin E (~ 20 g), 10 day aged ultimate pH (pHu) (~ 5 g), and the remaining block was vacuum packed and aged for 5 days (average

NU

temperature 3.2 ºC and humidity 79 %). An additional intra muscular fat (IMF) sample (~ 40 g) was taken from the LL muscle prior to aging. Only the LL muscle was tested for IMF due to the smaller size of the AF limiting the number of samples that could be taken from this

MA

muscle. The vitamin E and IMF samples were immediately diced, placed into respective tubes and stored at - 20 °C to prevent degradation.

ED

After 5 days of aging, each aging block was divided into a 4 cm retail display colour block, an ultimate pH (~ 5 g) sample, and a pre display thiobarbituric acid reactive substances

EP T

(TBARS) (~ 5 g), and peroxidase activity (PA) (~ 5 g) sample. After retail display, a post display TBARS and PA sample (~ 5 g) was obtained from the colour block.

AC C

2.5.2 Ultimate pH, Vitamin E and Intramuscular Fat (IMF) analysis Approximately 1 g of frozen muscle was used from each sample to determine the pHu using methods described by De Brito et al. (2016). Each sample was homogenised (Ystral homogeniser: series X10/25, Ystral, Germany) in an iodoacetate buffer at 19,000 rpm for 15 seconds prior to being placed into a water bath (22 °C). A pH meter (smartCHEM-CP, TPS Pty Ltd., Brisbane, Australia) was calibrated using pH 4.01 and 6.86 standard solutions (TPS Pty Ltd., Brisbane, Australia) prior to the pH of each sample being measured. Vitamin E was analysed on a wet muscle basis using methods previously described above by McMurray & Blanchflower (1979) using high performance liquid chromatography (HPLC) to identify αtocopherol content and reported on a mg/kg basis. The LL IMF samples were freeze dried and ground using a FOSS KnifetechTM 1095 sample mill and stored in airtight containers at

ACCEPTED MANUSCRIPT - 20 °C until analysis. The IMF content was measured according to methods described by Hopkins et al. (2014) where 3 g of freeze dried sample was extracted in a Soxtec machine with 85 ml of hexane for 80 minutes. The samples were then removed, dried and the residues weighed. 2.5.3 Retail Colour Analysis Retail colour analysis was conducted according to methods described by Smith, Bush,

PT

van de Ven, Hall, et al. (2017). In summary, a 3 cm steak was cut from each LL and AF 5 d aged sample. Each sample was placed on an individual styrofoam tray and covered with 15

RI

µm polyvinyl chloride film and displayed under fluorescent lights (set at ~ 1000 lx) for 72 h.

SC

Light reflectance of each sample was measured using a Hunter Lab Mini ScanTM XE Plus (Cat No. 6352, model no. 45/0-L, reading head diameter of 37 mm) at 4 time periods (0, 24,

NU

48 and 72 h) throughout the simulated retail display. The initial measurement (0 h) was taken 40 mins after the fresh surface was cut to allow the sample to bloom. Prior to each measurement the Hunter Lab was calibrated with a white enamel tile and black glass as per

MA

operating instructions. Each reading recorded L* (lightness), a* (redness), b* (yellowness) and ratio (ratio of wavelengths 630 nm/580 nm; from now on this will be referred to as

2.5.4 Oxidation Assays

ED

oxy/met ratio) values.

EP T

2.5.4.1 Thiobarbituric Acid Reactive Substances (TBARS) Assay The protocol utilized for the TBARS quantification assay was based upon methods described by Holman, Coombs, Morris, Bailes, & Hopkins (2018) and expressed as mg

AC C

malondialdehyde (MDA) per kg muscle. In summary, 100 mg of muscle were homogenised using micro-tube pestles in 500 µl of radioimmunoprecipitation assay (RIPA) buffer (no. 10010263, RIPA buffer concentrate, Cayman Chemicals

TM

Ltd., Michigan, USA). After

each sample was centrifuged the supernatant was analysed as per the OXI-tek TBARS Assay Kit (no. ALX-850-287-K101, Enzo® Life Sciences Inc., New York, USA) technical instructions. Absorbance was measured at 532 nm using a benchtop spectrophotometer (AMR- 100 microplate reader, Allsheng, Hangzhou, China). 2.5.4.2 Peroxidase Activity The peroxidase activity of meat was measured through analysis of the catalaseperoxidase reaction and was expressed as U/g where U = the amount of peroxidase (nmole)

ACCEPTED MANUSCRIPT that reduces 1.0 µmol H2 O2 per minute per gram of muscle at 37 °C and followed methods outlined by Holman et al. (2018). In brief, 20 mg muscle samples were homogenised using micro-tube pestles in 200 µL of RIPA buffer (no. 10010263, RIPA buffer concentrate, Cayman ChemicalsT M Ltd., Michigan, USA). All samples were then centrifuged for 10 minutes at 5600 rpm after which the supernatant was analysed against peroxide standards following the technical notes provided in the Peroxidase Activity Assay Kit (no. MAK092, Sigma- Aldrich Ltd., Missouri, USA). Absorbance was measured at 570 nm on a micro plate

PT

reader (FLUOstar OPTIMA, BMG LabtechnologiesT M Ltd., Victoria, AUS) every 2-3

Statistical Analysis

SC

2.6

RI

minutes for a total of six measurements.

The statistical analysis was conducted using the software package Genstat (16th

NU

edition) (VSN International, 2011). Summary statistics were generated on animal BCS data and qualitative pasture measurements (DM, CP, NDF, ADF, DMD, ME and vitamin E). Animal liveweight data was analysed in a linear mixed model (LMM) with the fixed terms:

MA

constant + collection date and random terms: animal I.D. + random error. Multiple LMM were generated for the colour, oxidation and meat quality parameters

ED

and included the same random traits for each model unless otherwise specified. The random terms for each model included: processing day + animal I.D. / carcass side + random error.

EP T

The fixed effects within each LMM were determined at the 0.05 level according to the Wald test. Non-significant (P > 0.05) terms were eliminated from the model using a stepwise backward elimination approach.

AC C

Identical LMM were run individually for muscle vitamin E and pHu data where the fixed terms included: constant + muscle (AF and LL). The pHu model included the additional random trait of pHu batch to account for test batch variation; this was not required for vitamin E as all samples were run in the same batch. The IMF LMM included fixed terms: constant + processing day, with processing day being removed from the random terms and replaced with IMF batch. Two separate colour LMMs were generated to analyse the retail colour stability of alpaca meat over the length of retail display (colour model 1) and determine if co-variates (pHu and vitamin E) sampled at time 0 h had a significant impact on initial colour trait values (colour model 2). Colour models 1 and 2 were identically run for each colour trait L*

ACCEPTED MANUSCRIPT (lightness), a* (redness), b* (yellowness) and oxy/met ratio. The fixed terms of model 1 were; muscle (AF and LL) x retail display (0, 24, 48 and 72 h) interaction + constant. Colour model 2 included fixed traits: muscle (LL and AF) x initial colour value (0 h) + pHu + vitamin E + constant and included the additional random trait of pHu batch to account for test batch variation. This was not required for vitamin E as all samples were run in the same batch.

PT

The TBARS and PA oxidation data were analysed individually using identical LMMs. Fixed terms included: muscle (AF and LL) x retail display (pre and post) interaction + co-

RI

variates vitamin E (AF and LL) + LL IMF. The random terms included in addition to the above traits test batch effect to account for test batch variation of the respective test (TBARS

3.1

Results

NU

3.

SC

or PA).

On farm animal and pasture measurements

MA

Animal live weight increased significantly (Table 1) with an average weight gain of 2.2 kg per week. The animals maintained an average BCS of 3 (ranging from 2.5 – 4) throughout the experiment. Pasture quality improved throughout the collection period with CP and ME

ED

values increasing from 8 % and 7.1 MJ/kg DM respectively in collection one up to 21 % CP and 9.2 MJ/kg DM ME in collection 5 (Table 2). Vitamin E content of pasture varied from

EP T

19.1 mg/kg in collection 4 through to 178.5 mg/kg in collection 7 (Table 2). The largest variation in vitamin E was observed after the peak in pasture ME levels, with vitamin E levels increasing from 22.2 mg/kg in collection 5 through to 129.5 and 178.5 mg/kg for

3.2

AC C

collection 6 and 7 respectively (Table 2). Meat Quality

3.2.1 Muscle traits

Ultimate pH levels were significantly lower in the LL muscle than the AF muscle (Table 3). Muscle vitamin E levels were higher in the AF than the LL muscle (Table 3). The average IMF content of the LL muscle was 3.1 (± 0.24) %, with no differences observed between slaughter (processing) days (P > 0.05). 3.2.2 Retail colour Vitamin E content had no significant effect on L*, a*, b* or oxy/met ratio values at the

ACCEPTED MANUSCRIPT start of retail display (Table 4). Ultimate pH levels at the start of retail display had a significant effect on both L* and b* values, such that for every unit increase in ultimate pH, L* values decreased by 8.1 (± 3.99) units and b* values decreased by 4.5 (± 2.17) units. There were significant differences in L*, a*, and oxy/met ratio levels between the AF and LL muscles (Table 4). At 0 h display the LL had higher L* (lighter) and a* (redder) readings and lower ratio (browner) appearance than the AF muscle. There was no statistical difference

PT

between muscles for b*. The oxy/met ratio was the only colour trait to have a significant Muscle X Retail

RI

Display interaction (Table 5). Vitamin E levels had a positive correlation to oxy/met ratio levels, such that for every unit increase in vitamin E there was 0.01 unit increase in ratio

SC

values (P > 0.05). Initial (0 h) oxy/met ratio values were higher in the AF than the LL and by 24 h and thereafter the difference between muscles was minimal (Table 6; P < 0.001). The

NU

highest oxy/met ratio values were observed at 24 h and as length of retail display continued a consistent trend in declining oxy/met ratio values occurred (Table 6).

MA

Lightness (L*) and redness (a*) levels varied between LL and AF muscles (Table 5), such that the AF muscle was darker in colour (lower L*) and redder (higher a*) than the LL

ED

muscle (Table 7). There was no variation between muscle yellowness (b*) levels (Table 5 and 7). Length of retail display had a significant impact on lightness (L*), redness (a*) and yellowness (b*) values (Table 5). Lightness (L*) and redness (a*) values followed the same

EP T

trend with the lowest values recorded at 0 h and the highest values observed at the 24 h measurement, followed by a decline as length of display continued (Table 8). Yellowness (b*) levels followed a similar trend with a significant increase observed between the first (0

AC C

h) and second (24 h) measurement and then a decline after a peak at 48 h (Table 8). 3.2.3 Oxidation Assays

There was a significant TBARS retail display x muscle interaction (Table 9). Both muscles (AF and LL) followed the same trend of higher post retail display TBARS levels (Table 9). The LL had lower oxidation levels than the AF, such that pre display AF TBARS values were similar to the LL TBARS post display values (Table 9). Pre retail display vitamin E content of the LL and AF had no effect on TBARS (P = 0.064 and P = 0.973 respectively). However there was a trend between TBARS levels and vitamin E, such that for every unit increase in vitamin E, pre display TBARS values were predicted to decrease by 0.36 (± 0.067) and 0.17 (± 0.059) mg MDA/kg for the LL and AF respectively and by 0.142 (±

ACCEPTED MANUSCRIPT 0.142) mg MDA/kg in the LL post display. No impact was observed in the AF post display. The LL had higher PA activity (0.68 ± 0.11 U/g) than the AF muscle (0.56 ± 0.09 U/g). Peroxidase activity was higher (P = 0.596) pre retail display (0.65 ± 0.12) than post (0.59 ± 0.11 U/g). There was no effect of IMF (LL) content on TBARS (P = 0.194) or PA (P = 0.261) levels.

PT

4. Discussion All animals were on a rising plane of nutrition in the lead up to slaughter, as indicated

RI

by an increase in animal live weight gain of 2.2 kg/week and maintained BCS (score 3).

SC

This was irrespective of the pasture biomass and quality fluctuations over the sampling period, demonstrating that the combined nutritional value and quantity of feed was greater

NU

than animal requirements over this period.

There were minimal trends observed between pasture quality (in particular ME and CP levels) traits and pasture vitamin E levels. The pasture vitamin E levels in this study

MA

varied between sampling periods, ranging between 19 – 179 mg/kg pasture. This was a greater range than reported by De Brito et al. (2017) for five different forage types, but

ED

these authors also reported that levels between sampling times varied by up to 82%. Overall, the pasture vitamin E levels in this study appeared higher than reported supplementation baseline α- tocopheryl intake values of 10 - 20 mg/kg feed (Guidera,

EP T

Kerry, Buckley, Lynch, & Morrissey, 1997; Ripoll, Joy, & Muñoz, 2011) and below the majority of α- tocopheryl treatment rates of 176 - 1000 mg/kg feed (Bellés et al., 2018; Guidera et al., 1997; Jose, Jacob, Pethick, & Gardner, 2016; Ripoll et al., 2011). It is clear

AC C

that forage levels of vitamin E can vary widely and that this will not be reflected in muscle levels as found in the current study and also by De Brito et al. (2017). Alpaca muscle vitamin E levels were well above the minimum 2 mg/kg meat oxidation threshold values established in lamb (Guidera et al., 1997; Ponnampalam et al., 2017) and beef (Arnold et al., 1992). These threshold values represent the minimum of vitamin E in muscle required to have an impact on reducing protein and lipid oxidation rates in both lamb and beef, irrespective of production system (pasture or grain). In this study vitamin E content had no significant impact on any colour stability or lipid oxidation traits and the overall protein and lipid oxidation levels were low. This is potentially due to the high concentrations of vitamin E reported in both alpaca muscles (> 4.6 mg/kg meat) and may be partially explained by research conducted in lamb, which found that the maximum effectiveness of

ACCEPTED MANUSCRIPT vitamin E on oxy/met ratio colour levels is reached at 4 mg/kg, after which it plateaus (Jose et al., 2016). Previous studies have shown that vitamin E levels of 3 mg/kg in sheep (Guidera et al., 1997) and 3.5 mg/kg in beef (Arnold et al., 1992) improve oxy/met ratio colour stability traits and when dietary vitamin E intake levels are at 3.45 mg/kg muscle, lipid oxidation (TBARS) in lamb can be reduced down to acceptable levels of below 2.4 mg MDA/kg meat (Ponnampalam et al., 2014). Alpaca meat undergoes protein oxidation to a lesser extent than other red meat species,

PT

as indicated by the smaller rate of decline observed in oxy/met ratio levels during retail colour display. When meat is exposed to oxygen the pigment primarily associated with

RI

colour, myoglobin, undergoes oxidation and as a result the meat surface colour changes from

SC

a red oxymyoglobin state to a brown metmyoglobin state, as indicated by lower oxy/met ratio values (Mancini & Hunt, 2005). In common red meat species this change in surface colour

NU

during retail display has been linked to reduced consumer appeal and discounted retail product (Hopkins et al., 2013). The favourable oxy/met ratio results in this study supports previous trends observed by Smith, Bush, van de Ven, Hall, et al. (2017) and show that ratio

MA

levels, maintained at or higher than 4.2 throughout the entire display period for both the LL and AF muscles, indicate low levels of brownness. The values reported in this study are well

ED

above lamb consumer acceptability thresholds of 3.3 (Khliji et al., 2010). This suggests alpaca meat might not follow trends reported in lamb, where unacceptability threshold levels can be reached within 48 h of retail display (Jose et al., 2016). Although there is some

EP T

variation in oxy/met ratio levels during retail display, indicating protein oxidation and metmyoglobin formation, it may not be at a level that would compromise retail sales by being discounted at the retail level due to lack of visual appearance. Further research should be

AC C

conducted to identify key consumer oxy/met ratio threshold values for alpaca meat and explore the reasons behind these low oxidation levels, with a view to determine if they can be informative to other species, such as lamb and beef, to reduce the financial cost and wastage incurred from high

oxidation levels during retail display. In addition future work should be

conducted on longer shelf life periods (> 72 h) to determine the point where oxidation levels start to reach unacceptability thresholds. This paper supports the well-established negative effects that increasing muscle pH can have on meat colour and shows that as pH levels increase, muscle lightness (L*) and yellowness (b*) values decline, becoming darker, and less yellow (Abril et al., 2001; Lawrie, 1998). Similarly it was observed that different muscles have different colour profiles at the start and during retail display, which is primarily due to the variation in muscle fibre

ACCEPTED MANUSCRIPT composition (Lawrie, 1998). Overall, the colour stability trends observed support previous studies on alpaca meat by Smith, Bush, van de Ven, Hall, et al. (2017), with peaks over the 24 and 36 h of retail display. The TBARS (< 1.2 mg MDA/kg) reported in this study are lower than previously reported for alpaca, ranging from 1.5 – 2.2 mg MDA/kg post 72 h retail display (Smith, Bush, van de Ven, Hall, et al., 2017). The LL IMF content had no impact on post TBARS readings irrespective of the higher IMF content reported in this study (3.1 %) than previous studies

PT

which have reported levels < 1% LL IMF (Smith, Bush, van de Ven, Hall, et al., 2017; Smith, Bush, van de Ven, & Hopkins, 2017). This may be explained by the combination of relatively

RI

low IMF content and the high levels of vitamin E found in the muscle, reducing the

SC

propagation of free radicals in membrane and plasma lipoproteins during retail display (Traber & Stevens, 2011).

NU

Muscle TBARS levels were higher in both muscles post retail display, indicating oxidation occurring during display. However, overall TBARS levels remained low (< 1.2 mg MDA/kg) and within acceptability threshold values (< 2.0 mg MDA/kg) reported in beef

MA

(Campo et al., 2006). In other species consumers have been able to determine lamb lipid oxidation (TBARS) levels when TBARS levels were greater than 2.4 mg MDA/kg muscle

ED

(Ponnampalam et al., 2014). However, Hughes, McPhail, Kearney, Clarke, & Warner (2015) found TBARS levels between 2.60 and 3.11 mg MDA/kg in long term aged beef striploins were still acceptable to consumers. The variation between studies could be partially explained

EP T

by varying concentrations of volatile compounds, such as hexanal, which were not reported but have been shown to be responsible for rancid aromas in cooked meat (Lawrie, 1998). Exploring these aroma and flavour compounds in the future might help to determine the

AC C

correlation between TBARS, flavour compound levels and consumer acceptability of alpaca meat, with an aim of developing acceptability thresholds for both raw and cooked alpaca meat.

The PA results indicated higher levels of oxidative enzyme activity prior to retail display than after retail display. This finding is unexpected as the samples had been vacuum packed and not exposed to oxygen prior to sampling. Therefore, low peroxidase levels would have been expected at the start of retail display. However, it should be noted that the PA results in this test indicated a high level of variability within the assay due to the large s.e. observed and the low level means. This suggests that more work should be conducted to validate the PA assay for alpaca meat and to determine if PA is a suitable test to indicate oxidation levels in alpaca meat.

ACCEPTED MANUSCRIPT

5. Conclusion Alpaca meat contains high levels of vitamin E, in both the LL and AF muscle, compared to other red meat species. Vitamin E content had no significant impact on colour stability or lipid oxidation traits during the 72 h retail display. This is thought to be due to the levels of vitamin E being above previously reported red meat threshold values, resulting in low levels

PT

of increased surface browning and lipid oxidation during retail display. This gives alpaca meat an advantage if a retail market develops, given consumers discount oxidised meat.

RI

Further research is warranted on a large sample size to explore these favourable relationships between antioxidant concentrations and alpaca meat quality traits with a view

SC

to better understanding changes in other red meat species.

NU

6. Acknowledgments

This project was funded by AgriFutures Australia and Illawarra Prime Alpaca. The authors

MA

would like to thank the NSW DPI technical staff and the staff at Milton District Meats, for their assistance.

ED

Table 1

Predicted liveweight (kg) of alpacas grazing in a mixed pasture based system, in the Southern Tablelands of New South Wales Australia prior to meat processing.

Period

Liveweight (kg) (predicted mean ± s.e.)

30/12/2015

*

11/01/2016

44.6 ± 1.0a

18/01/2016

51.5 ± 1.0b

01/02/2016

52.8 ± 1.0c

08/02/2016

54.7 ± 1.0d

6

22/02/2016

55.8 ± 1.0d

7

7/03/2016

57.8 ± 1.0e **

2 3 4 5

AC C

1

Date

EP T

Collection

Means with different letters differ significantly as calculated by the difference between LSD values. *Data from the first collection not included due to the scales malfunctioning. **Collection 7 data only contained animals from the second processing group.

ACCEPTED MANUSCRIPT Table 2 Nutritional parameters (calculated from raw data) of pasture grazed prior to animal processing. Traits measured include; dry matter (DM), crude protein (CP), neutral detergent fibre (NDF), acid detergent fibre (ADF), digestible dry matter (DDM), metabolisable energy (ME), and vitamin E. Collection time points were taken on the same day as animal liveweight. CP

NDF

ADF

%

%

%

%

83

8

68

41

2

11/01/2016

78

9

67

40

3

18/01/2016

87

10

68

4

01/02/2016

33

17

60

5

08/02/2016

44

21

54

6

22/02/2016

64

18

7

7/03/2016

63

14

ED EP T AC C

%

ME

Vitamin E

(MJ/kg

(mg/kg)

DM)

54

7.1

52.8

54

7.2

57.2

42

54

7.1

96.3

37

61

8.3

19.1

32

66

9.2

22.2

52

39

60

8.1

129.5

61

33

62

8.5

178.5

SC

30/12/2015

MA

1

DDM

PT

Period

DM

RI

Date

NU

Collection

ACCEPTED MANUSCRIPT Table 3 Ultimate pH (pHu) and vitamin E predicted means (± standard error) for two alpaca muscles (m. longissimus et lumborum (LL) and m. adductor femoris (AF)). Trait

Muscle

Predicted Mean ± s.e.

pHu

AF

5.54 ± 0.016a

LL

5.47 ± 0.016b

PT

6.1 ± 0.62A

Vitamin E AF (mg/kg) LL

5.4 ± 0.62B

RI

Predicted means for each trait generated on an individual trait basis. Different letters per trait

AC C

EP T

ED

MA

NU

SC

indicate significance between means (P < 0.05).

ACCEPTED MANUSCRIPT Table 4 Model based P values illustrating the effect of co-variates, vitamin E and ultimate pH (pHu) on muscle lightness (L*), redness (a*), yellowness (b*) and oxy/met ratio colour traits of two alpaca muscles (m. longissimus et lumborum (LL) and m. adductor femoris (AF)) at 0 h retail display.

a*

< 0.001

0.905

b*

0.696

0.791

Oxy/met Ratio

< 0.001

0.386

0.396 0.043

RI

Muscle 0.002

PT

Model based co-variate P values Vitamin E pHu 0.536 0.046

AC C

EP T

ED

MA

NU

SC

Y variates Colour Traits (0 hr) L*

0.753

ACCEPTED MANUSCRIPT Table 5 Model based P values of colour traits (L*, a*, b*, and oxy/met ratio) measured on two alpaca muscles (m. longissimus et lumborum (LL) and m. adductor femoris (AF)) across 0, 24, 48, 36 and 72 h (time) retail display. Y variates

Model based P values

Colour Traits

Muscle * Time

Muscle

Time

0.616

< 0.001

< 0.001

a*

0.051

< 0.001

< 0.001

b*

0.580

0.721

< 0.001

< 0.001

-

-

AC C

EP T

ED

MA

NU

SC

Oxy/met Ratio

RI

L*

PT

interaction

ACCEPTED MANUSCRIPT Table 6 The Oxy/Met (630/580 nm wavelength) colour ratio predicted means (± s.e.) from two alpaca muscles (longissimus et lumborum (LL) and adductor femoris (AF)) during a simulated 72 h retail colour display. Muscle

Time of retail display (predicted mean ± s.e. ) 0 hrs

24 hrs

48 hrs

72 hrs

5.27 ± 0.27ab

5.29 ± 0.28ab

4.63 ± 0.24c

4.17 ± 0.22e

LL

4.60 ± 0.24c

5.00 ± 0.26bd

4.56 ± 0.24cd

4.25 ± 0.22e

PT

AF

RI

Predicted means with different letters differ significantly (P < 0.05) as determined by the

AC C

EP T

ED

MA

NU

SC

LSD.

ACCEPTED MANUSCRIPT Table 7 Retail colour lightness (L*), redness (a*) and yellowness (b*) wavelengths (predicted means ± standard error) of two different alpaca muscles m. longissimus et lumborum (LL) and m. adductor femoris (AF) adjusted for length of retail display. Muscle

Colour Trait

PT

AF LL a L* 35.3 ± 1.03 38.4 ± 1.04b a* 18.4 ± 0.56a 17.7 ± 0.56b a b* 16.4 ± 0.46 16.4 ± 0.46a Predicted means with different letters per colour trait (across row) differ significantly (P <

AC C

EP T

ED

MA

NU

SC

RI

0.05).

ACCEPTED MANUSCRIPT Table 8 Changes in alpaca meat retail colour lightness (L*), redness (a*) and yellowness (b*) wavelengths (predicted mean ± standard error) over a 72 h simulated retail display period and adjusted for muscle. Colour Trait

Time of retail display (predicted mean ± s.e.) 0 hrs

24 hrs

48 hrs

72 hrs

RI

PT

L* 33.4 ± 1.07a 36.4 ± 1.07b 35.6 ± 1.07ab 35.4 ± 1.07ab a b b a* 15.5 ± 0.57 19.7 ± 0.57 19.0 ± 0.57 18.0 ± 0.57c b* 14.0 ± 0.47a 17.3 ± 0.47b 17.5 ± 0.47b 16.8 ± 0.47c Predicted means with different letters per colour trait (across row) differ significantly (P <

AC C

EP T

ED

MA

NU

SC

0.05) as determined by the LSD.

ACCEPTED MANUSCRIPT Table 9 Thiobarbituric acid reactive substances (TBARS) oxidation levels (predicted means ± standard error) of two alpaca muscles (m. longissimus et lumborum (LL) and m. adductor femoris (AF)) pre (0 h) and post (72 h) retail display. Oxidation test

Muscle

Retail Display (predicted mean ± s.e.) Pre

Post

PT

TBARS AF 1.01 ± 0.08a 1.19 ± 0.08b (MDA/kg) LL 0.79 ± 0.08c 1.01 ± 0.08a Predicted means with different letters differ significantly (P < 0.05) as determined by the

SC

RI

LSD.

NU

7. References

AC C

EP T

ED

MA

Abril, M., Campo, M. M., Önenç, A., Sañudo, C., Albertı,́ P., & Negueruela, A. I. (2001). Beef colour evolution as a function of ultimate pH. Meat Science, 58(1), 69-78. doi: https://doi.org/10.1016/S0309-1740(00)00133-9 AOAC. (2005). Offical Methods of Analysis (18 ed.). Gaithersburg, Maryland, USA: Association of Official Analytical Chemists (AOAC). Arnold, R. N., Scheller, K. K., Arp, S. C., Williams, S. N., Buege, D. R., & Schaefer, D. M. (1992). Effect of long- or short-term feeding of alpha-tocopheryl acetate to Holstein and crossbred beef steers on performance, carcass characteristics, and beef color stability. Journal of Animal Science, 70(10), 3055-3065. doi: https://doi.org/1992.70103055x Bekhit, A. E.-D. A., Hopkins, D. L., Fahri, F. T., & Ponnampalam, E. N. (2013). Oxidative Processes in Muscle Systems and Fresh Meat: Sources, Markers, and Remedies. Comprehensive Reviews in Food Science and Food Safety, 12(5), 565-597. doi: https://doi.org/10.1111/15414337.12027 Bellés, M., Leal, L. N., Díaz, V., Alonso, V., Roncalés, P., & Beltrán, J. A. (2018). Effect of dietary vitamin E on physicochemical and fatty acid stability of fresh and thawed lamb. Food Chemistry, 239, 1-8. doi: https://doi.org/10.1016/j.foodchem.2017.06.076 Campo, M. M., Nute, G. R., Hughes, S. I., Enser, M., Wood, J. D., & Richardson, R. I. (2006). Flavour perception of oxidation in beef. Meat Science, 72(2), 303-311. doi: https://doi.org/10.1016/j.meatsci.2005.07.015 Daley, C. A., Abbott, A., Doyle, P. S., Nader, G. A., & Larson, S. (2010). A review of fatty acid profiles and antioxidant content in grass-fed and grain-fed beef. Nutrition Journal, 9(1), 10. doi: https://doi.org/10.1186/1475-2891-9-10 De Brito, G. F., McGrath, S. R., Holman, B. W. B., Friend, M. A., Fowler, S. M., van de Ven, R. J., & Hopkins, D. L. (2016). The effect of forage type on lamb carcass traits, meat quality and sensory traits. Meat Science, 119, 95-101. doi: https://doi.org/10.1016/j.meatsci.2016.04.030 De Brito, G.F., McGrath, S.R., Holman, B.W.B., Friend, M.A., De Alencar, M.M., van de Ven, R.J., & Hopkins, D.L. (2017). The effect of forage diets on the fatty acid profile, lipid and protein oxidation, and retail colour stability of muscles from White Dorper lamb. Meat Science, 130, 81-90. doi: https://doi.org/10.1016/j.meatsci.2017.04.001 Descalzo, A. M., & Sancho, A. M. (2008). A review of natural antioxidants and their effects on oxidative status, odor and quality of fresh beef produced in Argentina. Meat Science, 79(3), 423-436. doi: https://doi.org/10.1016/j.meatsci.2007.12.006 Falowo, A. B., Fayemi, P. O., & Muchenje, V. (2014). Natural antioxidants against lipid–protein

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

PT

oxidative deterioration in meat and meat products: A review. Food Research International, 64, 171-181. doi: https://doi.org/10.1016/j.foodres.2014.06.022 Guidera, J., Kerry, J. P., Buckley, D. J., Lynch, P. B., & Morrissey, P. A. (1997). The effect of dietary vitamin E supplementation on the quality of fresh and frozen lamb meat. Meat Science, 45(1), 33-43. doi: https://doi.org/10.1016/S0309-1740(96)00086-1 Holman, B. W. B., Coombs, C. E. O., Morris, S., Bailes, K., & Hopkins, D. L. (2018). Effect of long term chilled (up to 5weeks) then frozen (up to 12months) storage at two different sub-zero holding temperatures on beef: 2. Lipid oxidation and fatty acid profiles. Meat Science, 136, 915. doi: https://doi.org/10.1016/j.meatsci.2017.10.003 Hopkins, D. L., Clayton, E. H., Lamb, T. A., van de Ven, R. J., Refshauge, G., Kerr, M. J., Bailes, K., Lewandowski, P., & Ponnampalam, E. N. (2014). The impact of supplementing lambs with algae on growth, meat traits and oxidative status. Meat Science, 98(2), 135-141. doi: http://dx.doi.org/10.1016/j.meatsci.2014.05.016 Hopkins, D. L., Lamb, T. A., Kerr, M. J., van de Ven, R. J., & Ponnampalam, E. N. (2013). Examination of the effect of ageing and temperature at rigor on colour stability of lamb meat. Meat Science, 95(2), 311-316. doi: http://dx.doi.org/10.1016/j.meatsci.2013.04.041 Hughes, J. M., McPhail, N. G., Kearney, G., Clarke, F., & Warner, R. D. (2015). Beef longissimus eating quality increases up to 20 weeks of storage and is unrelated to meat colour at carcass grading. Animal Production Science, 55(2), 174-179. doi: https://doi.org/10.1071/AN14304 Jose, C. G., Jacob, R. H., Pethick, D. W., & Gardner, G. E. (2016). Short term supplementation rates to optimise vitamin E concentration for retail colour stability of Australian lamb meat. Meat Science, 111, 101-109. doi: https://doi.org/10.1016/j.meatsci.2015.08.006 Keinprecht, H., Pichler, M., Pothmann, H., Huber, J., Iwersen, M., & Drillich, M. (2016). Short term repeatability of body fat thickness measurement and body condition scoring in sheep as assessed by a relatively small number of assessors. Small Ruminant Research, 139, 30-38. doi: https://doi.org/10.1016/j.smallrumres.2016.05.001 Khliji, S., van de Ven, R., Lamb, T. A., Lanza, M., & Hopkins, D. L. (2010). Relationship between consumer ranking of lamb colour and objective measures of colour. Meat Science, 85(2), 224229. doi: http://dx.doi.org/10.1016/j.meatsci.2010.01.002 Lawrie, R. A. (1998). Lawrie's Meat Science (6th ed.). Cambridge, England: Woodhead Publishing Limited. Mancini, R. A., & Hunt, M. C. (2005). Current research in meat color. Meat Science, 71(1), 100-121. doi: http://dx.doi.org/10.1016/j.meatsci.2005.03.003 McDonald, P., Edwards, R. A., Greenhalgh, J. F. D., & Morgan, C. A. (2002). Animal Nutrition (6th Edition ed.). Essex, UK: Peasons Education Limited. McMurray, C. H., & Blanchflower, W. J. (1979). Application of a high-performance liquid chromatographic fluorescence method for the rapid determination of α-tocopherol in the plasma of cattle and pigs and its comparison with direct fluorescence and high-performance liquid chromatography-ultraviolet detection methods. Journal of Chromatography A, 178(2), 525-531. doi: https://doi.org/10.1016/S0021-9673(00)92511-1 NSW Agriculture. (1983). Summary of Feed Composition. Glenfield: Nutrition and Feeds Evaluation Unit. Oddy, V. H., Robards, G. E., & Low, S. G. (1983). Prediction of in vivo digestibility of ruminant feeds from fibre and nitrogen content. Paper presented at the Proc 2nd International Conference of the International Network of Feeds Information Centres. Ponnampalam, E. N., Norng, S., Burnett, V. F., Dunshea, F. R., Jacobs, J. L., & Hopkins, D. L. (2014). The Synergism of Biochemical Components Controlling Lipid Oxidation in Lamb Muscle. Lipids, 49(8), 757-766. doi: https://doi.org/10.1007/s11745-014-3916-5 Ponnampalam, E. N., Plozza, T., Kerr, M. G., Linden, N., Mitchell, M., Bekhit, A. E.-D. A., . . . Hopkins, D. L. (2017). Interaction of diet and long ageing period on lipid oxidation and colour stability of lamb meat. Meat Science, 129, 43-49. doi: https://doi.org/10.1016/j.meatsci.2017.02.008 Ripoll, G., Joy, M., & Muñoz, F. (2011). Use of dietary vitamin E and selenium (Se) to increase the shelf life of modified atmosphere packaged light lamb meat. Meat Science, 87(1), 88-93. doi:

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

PT

https://doi.org/10.1016/j.meatsci.2010.09.008 Salvá, B. K., Zumalacárregui, J. M., Figueira, A. C., Osorio, M. T., & Mateo, J. (2009). Nutrient composition and technological quality of meat from alpacas reared in Peru. Meat Science, 82(4), 450-455. doi: http://dx.doi.org/10.1016/j.meatsci.2009.02.015 Smith, M. A., Bush, R. D., Thomson, P. C., & Hopkins, D. L. (2015). Carcass traits and saleable meat yield of alpacas (Vicugna pacos) in Australia. Meat Science, 107(0), 1-11. doi: http://dx.doi.org/10.1016/j.meatsci.2015.04.003 Smith, M. A., Bush, R. D., van de Ven, R. J., Hall, E. J. S., Greenwood, P. L., & Hopkins, D. L. (2017). The impact of gender and age on the nutritional parameters of alpaca (Vicugna pacos) meat, colour stability and fat traits. Meat Science, 123, 21-28. doi: http://dx.doi.org/10.1016/j.meatsci.2016.08.012 Smith, M. A., Bush, R. D., van de Ven, R. J., & Hopkins, D. L. (2017). The effect of grain supplementation on alpaca (Vicugna pacos) production and meat quality. Small Ruminant Research, 147, 25-31. doi: https://doi.org/10.1016/j.smallrumres.2016.11.024 Traber, M. G., & Stevens, J. F. (2011). Vitamins C and E: Beneficial effects from a mechanistic perspective. Free radical biology & medicine, 51(5), 1000-1013. doi: https://doi.org/10.1016/j.freeradbiomed.2011.05.017 Van Soest, P. J. (1963). Use of detergents in the analysis of fibrous feeds. II. A rapid method for the determination of fibre and lignin. Journal of the Association of Official Agricultural Chemists, 46(5), 829 - 835. VSN International. (2011). GenStat for Windows 14th Edition: VSN International, Hemel Hempstead, UK.