Variation in nutritional quality in UK retail eggs

Fatty acid (FA), carotenoid and vitamin contents of UK eggs were assessed for four production systems (caged (CA), free-range (FR), organic (OR) and extensive organic (EO)) as well as season. The impact of enforced housing, due to avian influenza, was also investigated. Production system did not alter vitamin D 3 , B 2 or B 9 content, but significantly influenced nutritionally desirable FA, carotenoid and vitamins A and E - concentrations decreased as production intensity increased, although for most, CA and FR did not differ significantly. Vitamin E and FA profiles for OR and EO were also similar, although carotenoids were higher in EO eggs. In contrast, FA, carotenoids, vitamins E and B 9 were consistent throughout the year, unlike vitamins A, D 3 and B 2 , which fluctuated with season; D and B 2 were higher in July than January and lower vitamin A was the only detected implication from enforced housing of FR and OR birds


INTRODUCTION
Eggs are widely recognised to be a relatively inexpensive dietary source for a range of nutrients including high quality protein with an ideal amino acid profile, essential polyunsaturated fatty acids (PUFA), carotenoids and vitamins (British Egg Industry Council, 2020;Réhault-Godbert, Guyot, & Nys, 2019).Egg consumption in the UK has increased steadily in recent years (British Egg Industry Council, 2020), possibly driven by their low cost and dismissal of their historic link to cardiovascular disease (Guo, Kliem, Lovegrove, & Givens, 2017).Consumer studies show, whilst price is an important driver of egg purchases, perceptions of bird welfare and nutritional quality are also relevant (Pettersson, Weeks, Wilson, & Nicol, 2016).For example, recent UK buying habits have seen a transition from a market dominated by eggs from caged birds to greater purchase of free-range eggs, increasing from only 32% of sales in 2004 to 74.5% in 2021 (British Egg Industry Council, 2020).
With respect to nutrients essential to consumer health, some studies (Guo et al., 2017;Mierlită, 2020;Mugnai et al., 2014) report little difference in composition between eggs from fully housed birds compared with free range or organic eggs, and Marelli et al. (2021) and Mierlită (2020) report similarities between free range and organic eggs.In contrast, other results (Lordelo et al., 2017;Marelli et al., 2021) show significant differences comparing eggs from caged birds with those from floor-based barn systems, whereas others found clear differences for some fatty acids (FA) comparing free range and organic eggs (Lordelo et al., 2017;Mierlită, 2020;Mugnai et al., 2014).These discrepancies might be explained by differences in layer diet and management across the systems and countries involved, as well as vegetation cover and/or access by birds to outdoor ranges.Herbage intake is known to have a strong influence on egg quality (Mugnai et al., 2014) yet theoretical access to an outdoor range does not always lead to foraging behaviour and hence consumption.Pasture activity by hens is driven by flock size, internal and external stocking densities as well as predators, range quality and weather conditions among other factors (Gilani, Knowles, & Nicol, 2014); systems or seasons encouraging foraging behaviour and grazing intake might enhance egg nutritional quality.
Due to the influence of foraged flora and fauna, egg nutritional quality may be altered if access to the range is restricted.For example, recent winters have seen enhanced biosecurity of outdoor poultry across Europe, leading to enforced housing of free-range birds, to control the seasonal spread of Avian Influenza from migratory birds, which appears to be getting worse year on year (Meurens et al., 2021).While housed, most European countries operate derogations, permitting eggs to be marketed as 'free range' or 'organic', up to 12-16 weeks after restrictions come into play.However, there is little or no information on potential changes to egg quality during these periods of forced housing.
This study compared (i) the nutritional quality of UK retail eggs from four main management systems; caged (CA), free range (non-organic) (FR), supermarket sourced organic (OR) and small scale, direct sales extensive organic (EO), (ii) seasonal variations and how this might relate to the production systems and (iii) the impact of enforced housing on egg quality from "free-range" systems during periods of avian influenza restrictions.Since the nutrient intake by hens can influence egg composition, the hypothesis tested here is the content of desirable constituents (beneficial fatty acids, carotenoid antioxidants, and vitamins A, B 2 , B 9 ,D and E) will be higher in free range and organic eggs, especially in the summer.

Sample Collection
Medium or mixed sized eggs were collected from four UK production systems: CA, FR, OR and EO.Details of eggs sourcing, and collection dates are shown in table 1.For FR and OR eggs, three half dozen boxes of own-brand eggs were purchased every three months, five times between April 2017 and April 2018 (specified as A-E), from eight supermarkets in Newcastle upon Tyne, UK.Caged eggs were purchased on the same dates but only available from 4 of the shops.
The EO eggs, along with production records, were collected directly from five farms in Southern England and Wales, just prior to supermarket purchases (except for the last date when only 4 farms had eggs available).For three farms, each with three flocks of varying ages, twelve eggs were collected from each flock (classed as 20-36, 37-55 and over 56 weeks old), sampling 36 eggs per farm, on all dates.For the other 2 farms, with single-aged flocks, 21 eggs were collected each time.At first collection (A, April 2017), all poultry were confined inside as a precaution against Avian Influenza.However, since EO farms released birds prior to egg collection, this date was excluded from the main statistical analysis.Immediately after collection, eggs were transferred in polypropylene opaque boxes to Newcastle University and stored at ambient temperature until further processing.

Pooled egg sample preparation
Three random supermarket eggs, one per box, were pooled and mixed.For EO flocks, one egg per age group (from multiple flock farms) or three random eggs were pooled.These whole eggs were mixed, freeze-dried and used for vitamin B 2 and B 9 analysis.For fatty acid analysis, fat soluble vitamins (retinol, E and D) and carotenoids analysis, yolks were separated from albumen, from another 3 eggs per batch, pooled and immediately stored at -80°C until analysis.All samples for analysis were frozen or/and freeze-dried before their "best before date".

Fatty acid analysis of yolk
Fatty acid (FA) analysis and quantification was as reported by Chatzidimitriou et al. (2022).This is explained briefly in this text here with more details given in supplementary material.Following lipid extraction and gravimetric quantification fatty acid methyl esters (FAME) were prepared, separated by gas chromatography (Shimadzu, GC-2014, with a flame ionization detector and Agilent CP-Sil 88 column -100m x 0.25mmID x 0.20 µm film thickness).Quantification, using 52-FAME standards (GLC46) from Nu-Chek Prep Inc, US, was based on peak areas of individual FA, expressed as a percentage of the total area under peaks for known quantified FA.
2.4 Carotenoids, vitamin A and E analyses of yolk

Sample preparation
All procedures were undertaken under dim lighting with echinenone as an internal standard.As with fatty acid analysis, a brief description of the method is given here with details described in supplementary materials.
Yolk preparation followed a modification of methods described by Kurilich and Juvik (1999) and Hammershøj, Kidmose, and Steenfeldt (2010).In a glass tube 0.2g±10mg of yolk was weighed, 1 ml of ultrapure water (18.2Ω) was added and vortexed.Two ml of ethanol of 5% pyrogallol, 1% BHT and the internal standard (around 0.01%) were added and vortexed.Saponification took place with the addition of 0.5 ml of KOH and after filling the tubes with nitrogen, incubating for 30 min at 45•C.Tubes were cooled for 1 min, 1 ml of ultrapure water was added and vortexed.
Two ml of hexane was added before being vortexed for 1 min.After centrifuging at 2000 g for 5 min the upper hexane layer was transferred to a clean tube using a glass Pasteur pipette.This hexane extraction was repeated before the pooled extracts were dried under a steady nitrogen stream, then reconstituted in 1 ml of HPLC mobile phase and filtered using a 0.45 µm syringe PTFE into a centrifuge tube.Aliquots of 300 μl were transferred to amber glass vials and immediately run in the HPLC, kept at 4 •C during analysis.
Analytical standards of echinenone, lutein, zeaxanthin, canthaxanthin, β-cryptoxanthin, βcarotene and α-carotene were purchased from CaroteNature GmbH (Ostermundigen, Switzerland).Stock solutions of standards retinol and tocopherols were dissolved in ethanol (containing 0.1% BHT), while carotenoids were dissolved in chloroform (containing 0.1% BHT).The concentration of standard stock solutions was measured after appropriate dilution by measuring the absorbance in either ethanol or hexane with a UV spectrophotometer as described by Liu et al. (2011).Calibration curves for each compound were then made with a range bracketing the expected concentrations, using a Shimadzu HPLC system with a Phenomenex column (250mm x 4.6mm at 30•C), coupled with photodiode array and fluorescence detectors (FD).Briefly, using a linear gradient for 20 minutes with a mobile phase of 65% acetonitrile, 35% methanol and 0.065% trimethylamine and a flow rate of 1.5 ml/min, injecting 20 μl samples.Quantification was at 450 nm (for lutein, zeaxanthin, beta-cryptoxanthin and canthaxanthin), 325 nm for retinol and, for tocopherols (alfa-, beta-, gamma-and delta-) FD, at excitation and emission wavelengths of 295nm and 325 nm, respectively, validating results against stock dilutions.
2.6 Vitamin B 2 and B 9 analyses of eggs Unlike procedures described above, which were carried out in-house, this analysis was conducted at an accredited commercial laboratory.Freeze-dried composite eggs from July 2017 and January 2018 were transferred on dry ice to SGS Institute Fresenius GmbH (Germany).Vitamin B 2 as the mass fraction of total riboflavin including its phosphorylated derivatives, which was extracted by acid and enzymatic dephosphorylation treatments, and quantified by HPLC-fluorimetry using external standard, as described in protocol BS EN 14152:2003, and vitamin B 9 as the total folate content by turbedimetric detection of the growth of the microorganism Lactobacillus casei (ATCC 7469).Results for both were expressed as μg/100g egg.and E).For FR and OR eggs, comparative analysis for fatty acids, carotenoids, vitamins A and E was performed between April 2017 (when all hens were housed) and April 2018.

RESULTS
All differences mentioned and discussed are statistically significant (p<0.05),unless specified otherwise.

Lipid content and fatty acid profiles
Concentrations of total lipid and proportion of nutritionally relevant FA in eggs (comprising > 90% of total) show considerable differences between the 4 production systems (Table 2).In contrast, egg FA profiles were consistent over the 5 sampling dates throughout the year (results not shown), with the sole significant difference being the ratio of omega-6 PUFA (n-6) with omega-3 PUFA (n-3) (n-6:n-3).This was lowest in July at 7.3:1 and highest in January at 8.7:1 (p<0.05),although neither differed from other dates.
The lowest lipid content was for FR eggs (33.4%); 7% lower than EO eggs (35.9%) (p<0.05), and neither differed from CA and OR eggs.Although many individual FA concentrations differed between production systems, on the whole results for OR and EO eggs were very similar, with no significant differences in concentrations and only minor variation in values.There were also strong similarities in FA profiles for CA and FR eggs, since the only differences reaching significance were for polyunsaturated fatty acids (PUFA), oleic acid (c9C18:1, OA) and c11C18:1 concentration (12% higher, 4% and 7% lower respectively in FR compared with CA eggs).

Carotenoids, vitamin A and E
Production system also influenced the carotenoids and vitamins A and E content of eggs (Table 3).For all carotenoids, γ-tocopherol and total vitamin E, there were substantial incremental increases in concentrations going from CA, FR, OR to EO eggs.For all carotenoids, concentrations in EO eggs were 1.5-2.1 times those in OR eggs, which were 1.5-2.5 times those of FR eggs and, although concentrations in FR eggs ranged from 1.5-2.8times the levels in CA eggs, these differences were not significant.A similar stepped increase in concentrations was seen for γtocopherol, with EO being almost 3 times higher than in CA eggs,.On the whole, vitamin contents were more consistent across systems although differences for vitamin A did reach significance when comparing CA with both FR (+10%) and OR (+9%) eggs.Similarly, OR eggs contained 24% more Vitamin E than CA eggs.

Vitamin D in yolk and B 2 and B 9 in egg
Unlike the fatty acids, carotenoids, and vitamin E, sampling dates significantly influenced concentrations of vitamin B 2 , and D (Table 5).The production system made little difference to concentrations (results not shown), except for a trend (p<0.1) that EO eggs were lower in 25-OH D 3 than other types (1.5 vs 1.7-1.9µg/100g yolk).Whilst vitamin B 9 was consistent for July and January (at 30.9 μg/100g egg) both D 3 and B 2 activity differed.In summer, July eggs had 54% more D 3 and almost 20% more 25-OH D 3 than in January.July eggs were also 10% higher in vitamin B 2 compared with those purchased in January.

DISCUSSION
The study hypothesis, that the content of beneficial fatty acids, carotenoid antioxidants (lutein, zeaxanthin, beta-cryptoxanthin and tocopherols), and vitamins (A, D, E, B 2 and B 9 ) will be higher in free range and organic eggs, especially in the summer, was both supported and disproven.
Whereas most nutrients assessed did differ between the production systems, others were did not and the same applied for variation throughout the year -many did not change but others differed between sampling dates.
However, in this study significant differences were identified for eggs available at supermarkets in NE England, as well as from small scale organic producers selling at farm-gates.This category was included in the survey to represent relatively small flocks encouraging birds' foraging behaviour, since herbage intake has been shown to enhance nutritional quality of eggs (Mugnai et al., 2014).
Our sampling throughout the year (April, July and October in 2017, January and April 2018) covered potential seasonal differences for ranging behaviour, pasture availability, and possible differences in feed consumption from the ranges.Sampling continued to April 2018, to compare eggs from the previous April, when free range and organic birds were fully housed under avian influenza protection and unable to access the ranges.Although limited to a single sample date, this comparison gives an indication of possible implications to egg composition from more stringent housing orders across Europe, to reduce the risk of avian influenza for farmed birds.
The strong relationship between layers' diet and egg composition is well recognised (Banaszewska, Biesiada-Drzazga, Marciniuk, Hrnčár, Arpášová, & Kaim-Mirowski, 2020;Kowalska et al., 2021), despite the ability of hens to substantially moderate nutrient transfer into the egg (Chatzidimitriou et al., 2022).Hence, since feeding practices for hens in different production systems do vary between countries, it is perhaps not surprising that studies comparing eggs from hens under different management in different countries report inconsistent results (Banaszewska et al., 2020;Baykalir et al., 2020;Dalle Zotte et al., 2021;Guo et al., 2017;Marelli et al., 2021;Mierlită, 2020).This study found composition differences in FA, lutein, zeaxanthin, betacryptoxanthin, tocopherols and retinol between eggs from different systems, and for vitamin D and B 2 throughout the year.However, whilst, some results confirm previous findings, others are contradictory.The differences identified here in egg composition from contrasting management systems, at different times of the year, are likely due to variation in feed ingredients as well as access to (and consumption of) supplementary vegetation and fauna from range pastures.
However, bird genotype and age profile also differs between production systems (Leenstra et al., 2014); both of which have been shown to influence egg composition (Kowalska et al., 2021), again contributing to potential differences in nutrient supply from eggs produced under differing management or times of the year.Although relevant, it is unfortunate this information is rarely available in a 'basket-type' survey such as this.
4.1 Egg fatty acid profiles

Comparison with published results
Published studies considering egg composition from differing production systems show inconsistent results, possibly due to management differences within the systems compared, and an inevitable influence of variation in feed ingredients across the countries.Dietary FA intake has a strong influence over egg FA profiles (Mugnai et al., 2014) and whereas there might be some common ingredients (such as soy), there will also be differences between diets in various countries, especially for organic, or other systems, attempting to maximising 'home produced' or 'local' sourcing.As a consequence, the profile of egg FA varies widely between countries and production systems, making a useful comparison with these results challenging.Although not a comprehensive review of published work on egg FA profiles, Figure 1 compares the mean values (for SFA, MUFA and PUFA as well as the ratio between n-6 with n-3) for 18 different 'systems' covered in 5 recent comparative studies (Baykalir et al., 2020;Dalle Zotte et al., 2021;Islam et al., 2021;Lordelo et al., 2017;Marelli et al., 2021;Mierlită, 2020) with findings in this UK comparison.
All mean values for SFA and PUFA concentrations in this study fall within the ranges in published system means, although MUFA concentrations were higher and the ratio between n-6:n-3 was generally lower (considered to be better for consumer health).However, since the other studies were from countries with contrasting pedo-climatic and agronomic background conditions (Italy, Romania, Pakistan, Portugal and Turkey) the greater variability compared with this UK survey is not surprising.This is particularly noticeable for SFA (similar for all UK systems) and the ratio of n-6:n-3, which, although lower for EO and OR compared with FR eggs, mean values for all UK systems in this study (7.5-9.1:1) were lower than most published system means, with the exception of FR eggs from Pakistan (7.7:1) and FR and OR eggs from Romania (5.2:1 and 4.2:1 respectively).

Comparison between systems in this study
Although concentrations of many beneficial FA showed incremental increases going from CA, FR, OR to EO, the most striking aspect of results here is the similarity in composition of OR and EO eggs on one hand and CA and FR eggs on the other -with no significant differences for the FA considered in the organic comparison.For many nutritionally relevant, unsaturated FA, concentrations in both OR and EO were significantly higher than both CA and FR eggs, which were similar in composition to each other.Although this confirms results from Italian flocks (Marelli et al., 2021;Mugnai et al., 2014), it contrasts with the Romanian study, reporting similarities in FA profiles between FR and OR eggs, both of which differed from 'conventional' eggs (Mierlită, 2020), as well as findings from Portugal, reporting similar PUFA and n-6 levels in FR and OR eggs -again both differing from CA and barn eggs (Lordelo et al., 2017).
Management under free range and organic production may appear similar since birds can access outdoor ranges under both systems, so perhaps composition differences identified between these two egg types is surprising.Organic eggs from the supermarkets were 40-50% higher in individual and groups of n-3, with 21% more PUFA, compared with FR eggs (which were also 7% ( 1 Results from: Baykalir et al., 2020;Dalle Zotte et al., 2021;Islam et al., 2021;Lordelo et al., 2017;Marelli et al., 2021;Mierlită, 2020) Key: SFA= saturated fatty acids, MUFA= monounsaturated fatty acids, PUFA= polyunsaturated fatty acids, Ratio n-6:n-3 = ratio of the sum of all omega-6 fatty acids to the sum of all omega-3 fatty acids Figure 1.Results from this and 5 other studies 1 comparing egg fatty acid profiles from contrasting production systems higher in the detrimental C14:0).These differences have 3 likely explanation: (a) access to a range does not necessarily translate into movement and feeding behaviour, (b) in the absence of mineral fertiliser and other agrochemicals, the vegetation on organic land, relying on forage legumes like clovers for nitrogen supply, will be more diverse than conventional farms and (c) all feed offered in organic farming has to be certified and traceable to permitted production systems, so is to differ from that offered in non-organic free range systems.On the whole, ranging activity by organic flocks is likely to be higher than non-organic systems, encouraged by smaller flocks (Campbell, Dyall, Downing, Cohen-Barnhouse, & Lee, 2020) and possibly inadequate essential amino acids supply in the offered diet (Leenstra et al., 2014), which has been shown to encourage foraging (Skřivan & Englmaierová, 2014).Although to date, there are no reports of the impact on poultry meat or egg composition from diverse vegetation consumption, this has been shown to raise n-3 and PUFA levels in organic milk (Loza et al., 2023) and beef (Butler, Ali, Oladokun, Wang, & Davis, 2021).
In contrast to the strong impact of management on egg FA profiles, seasonality had a very minor effect, even when birds were denied range access in April 2017.This, and the similarity in composition between CA and FR eggs, possibly suggests that flora and fauna consumption from the range may have a weaker influence over egg lipid composition than the diets hens are offered in the housing.

Comparison with published composition
Published data on the carotenoid content of eggs also appears quite variable across studies, although inconsistency in reporting units makes comparisons challenging.There is general agreement that carotenoid content is dominated by lutein and zeaxanthin, which increase with herbage consumption (Mugnai et al., 2014;Skřivan et al., 2014).Values for lutein and zeaxanthin identified here (combining to give 78-82% of the total carotenoids) cover a similar range to some studies (Mugnai et al., 2014;Schlatterer & Breithaupt, 2006) but they are considerably less than the 25.5-56.2and 21.8-42.4mg/kg yolk dry matter (respectively) reported by Skřivan et al. (2014) with the lower values of these ranges found in eggs produced by hens with zero vegetation on their range.The lutein content of OR and EO egg yolks exceed 'typical' values of 550 and 575 µg/100 g yolk quoted in UK food composition tables (Public Health England) which are comparable with values for FR eggs in this study, but higher than CA eggs.

Comparison between systems in this study relative to other studies
As with FA results, the carotenoid and fat-soluble vitamin concentrations in eggs from contrasting systems were very different, although largely unaffected by seasonality -except for vitamin A. In April 2017, when under Avian Influenza restrictions, eggs were lower in vitamin A than in April 2018.Again, CA and FR eggs had similar carotenoid and fat-soluble vitamin contents, with the only difference reaching significance was for Vitamin A -although significant, the 9% superiority of FR eggs is dwarfed by much higher concentrations in both types of organic eggs.For the 4 carotenoids and fat-soluble vitamin assessed, OR eggs were higher than CA eggs, almost by a degree of magnitude-they had 7 times as much zeaxanthin, 5 times the β-cryptoxanthin and over twice as much lutein, 'unknown carotenoids' and canthaxanthin, as well as δ-tocopherol and γtocopherol, resulting almost twice the vitamin E activity as CA eggs.Despite the apparent similarity of range access by bird in FR and OR systems, results for carotenoids and fat soluble vitamins reinforce the theory of contrasting use of the range and possibly offered feed composition, since OR had more than twice as much zeaxanthin and canthaxathin and 50% more lutein than FR eggs.
Both FR and OR eggs were higher in vitamin A than CA eggs and OR eggs also had higher vitamin E activity, compared with both CA and FR eggs.However, vitamin A and E in EO eggs were statistically similar to all other systems, suggesting greater variability between these farms than from larger flocks supplying supermarkets.In the secondary April comparison, birds with access to the range in 2018 produced eggs with more vitamin A than a year earlier, when confined to housing.
Like other results, there is agreement and conflict between these findings and other publications.Karsten, Patterson, Stout, and Crews (2010) reported higher levels of both vitamin A and E in eggs from 'pastured' compared with caged hens.In contrast, Skrivan et al. (2014) found vitamin A content did not vary but vitamin E in eggs was higher, if hens had access to vegetation compared with non-grazing hens.As discussed, these discrepancies are likely to be driven by differences in feed and range composition, although possibly also by genetics.A comparison of barn and FR eggs using 2 genotypes identified an interaction between genetics and management; FR eggs from only one genotype were higher vitamin A content than eggs from barn hens (Krawczyk, Sokołowicz, & Szymczyk, 2011).These researchers also report eggs (from both genotypes) had similar vitamin E content under both production systems which contrasts with Mugnai et al. (2014), who reported EO eggs were higher in vitamin E than either OR and CA eggs for most of the year, except winter, when concentrations were similar for all systems.Interestingly, they noted egg vitamin E levels reflected the pattern of vitamin E in available herbage, which changed over the seasons.
The lack of seasonal variation in carotenoid content, especially when birds were confined to housing, was surprising considering its apparent link to herbage consumption.This is in contrast with the study by Mugnai et al. (2014) who found an interaction between management and season, with greater differences in zeaxanthin levels between egg types in winter, which diminished in summer.The lack of season differences here suggests the extent of foraging by birds is either consistent throughout the year (despite fluctuations in daylight, herbage growth and weather) and/or there are compensatory changes in the carotenoid content and availability from the vegetation -which have been shown for Italian pastures (Mugnai et al., 2014).
However, in the absence of any information on the feeds offered, range composition or genetic make-up of birds involved in any of the management systems makes further discussion speculatory.

Vitamin D in yolk and comparison with published results
In contrast to results already discussed, vitamin D 3 and 25-OH D 3 did not differ between the management systems but summer yolks (July) were significantly higher in both, compared with those purchased in January.There is limited information on vitamin D content of eggs and its specific forms of D 3 and 25-OH D 3 , although there has been another study investigating variation across management systems in the UK market and 2 papers looking at eggs in Australia.Results from these studies are very variable and confounded by differences in reporting units.The current results for both D 3 and 25-OH D 3 were similar to those of Guo et al. (2017) but they align poorly with UK food composition tables (Public Health England) which suggest D 3 content to be three fold higher and 25-OH D 3 to be 10-fold lower than these results.The difference may be due The lack of management effect on vitamin D 3 in this study accords with German commercial eggs assessed by Kühn, Schutkowski, Kluge, Hirche, and Stangl (2014), who found no difference between FR and barn eggs and Dunlop et al. (2017) who found no difference in vitamin D 3 concentrations between Australian free range and cage eggs (possibly because of considerable variation in the former).However, these results for retail eggs contrast with a controlled trial by Kühn et al. (2014) reporting 3-4 fold higher D 3 and OH D 3 content in eggs by excluding birds from housing for 9 hours each day for 4 weeks during summer (compared to eggs from fully housed birds).Guo et al. (2017) also found FR and OR eggs were higher in D 3 compared with indoor eggs and OR eggs were higher in 25-OH D 3 compared with FR and CA eggs.
Results showing higher D 3 and 25-OH D 3 contents in July eggs, indicate the production of vitamin D due to sun UV exposure (as shown by Kühn et al. (2014).Examining the interaction for vitamin D 3 between management and date, considering numerical differences and not only statistical significance, it is clear that seasonal variation for D 3 exists only for systems which have access outside housing.This agrees with Guo et al. (2017) who showed greater seasonal differences for both D 3 and 25-OH D 3 egg contents in FR and OR eggs, than for indoor systems.Dunlop et al. (2017) looked at FR eggs from a low-stocking density farm (130 hens/ha) and found these had higher vitamin D 3 (but not 25-OH D 3 ) content than eggs from farms with higher stocking densities, suggesting lower stocking densities might encourage range access and greater sun exposure (assuming that the feed vitamin D content was similar).However, such a relationship between OR and EO eggs was not found in this study since the only statistical difference between seasons was for EO eggs.This suggests the extensive system, and possibly all 'free-range' flock, might benefit from winter supplementation to maintain egg quality across seasons.Keeping egg vitamin D content high during winter months is potentially more relevant for consumer health than for summer eggs; limited sun exposure and vitamin D synthesis increases the potential benefit from boosting dietary intake.

Vitamins B 2 and B 9
To our knowledge, this is the first report on the impact of management and season on vitamins B 2 and B 9 content in eggs, and there are few published reports relating to other foods.
Concentrations of B 2 (riboflavin) in this study (mean of 0.41 mg/100 g egg) were slightly less than the 0.50 mg/100 g egg reported in the UK food composition tables (Public Health England), even for summer eggs, which were significantly higher than those bought in winter.In contrast, to the consistent egg vitamin B 2 content across production systems, organic milk has been reported to be higher in B 2 than non-organic milk (Poulsen, Rybicka, Poulsen, Larsen, Andersen, & Larsen, 2015).The higher levels of B 2 in summer eggs is also in contrast to findings for organic milk; Poulsen et al. (2015) found concentration higher in winter than summer.However, riboflavin metabolism in cows is very different to poultry since a healthy rumen microbiome synthesis sufficient B vitamins for most ruminants (McDonald, 2011).On the other hand, egg concentrations reflect supplemental and natural dietary supply to hen (White Iii Iii, Armstrong, & Whitehead, 1986), with the latter potentially higher in the summer due to vegetation and fauna consumption by ranging FR and OR birds.
Concentrations of egg B 9 (folate) (mean 31 μg/100g egg) did not vary with season or system and were considerably lower (-50%, -34% respectively) than reported by Hoey et al. (2009) in UK eggs or UK feed composition tables (Public Health England) and also lower (~ -50%) than reported for Swedish eggs (Strandler, Jastrebova, & Mattisson, 2011) (based on 15.5g yolk weight as measured).As with B 2 , supplementary folate can significantly increase yolk levels (Bunchasak & Kachana, 2009) and 95% of egg folate is found in yolk (Public Health England).Folate is also found in green leafy vegetables and legumes (Food Standards Agency 2002) and therefore potentially variable from seasonal changes in vegetation folate concentrations.Indeed, the folate content of EO eggs was lower in winter than summer.As mentioned, EO hens possibly consume more legume forages with more extensively grazing the ranges than other systems.This theory is possibly confirmed by the stable folate content of CA eggs, summer and winter.

Conclusions
Bird management had a stronger influence over egg content of FA, carotenoid and vitamin E than seasonality.Eggs from OR and EO systems showed a FA profile more beneficial to human health, having less SFA, more total and long chain n-3 and a lower n-6:n-3 ratio than FR eggs.Compared with CA eggs, whilst PUFA and n-3 were higher in OR and EO eggs, the difference between their n-6-:n-3 ratios was not significant.Carotenoids were also higher in eggs from hens with increasing opportunity to forage, with EO eggs having the highest levels.Conventional CA and FR systems produced eggs with similar carotenoid and vitamin E levels, while vitamin A was higher in eggs from ranging hens (both OR and FR) compared with eggs from CA systems.Eggs from EO systems were more variable for vitamin A and E content although did not differ from other systems.The lack of interactions for FA and carotenoids between systems and dates highlights the weight of the impact that management practices have rather than seasonal influences.
Although egg vitamin D 3 did not differ between management systems, date had a clear effect confirming the positive impact of free-range systems in summer due to sun exposure to produce vitamin D 3 in hens and hence eggs.However, again a greater variation in EO eggs was noted, suggesting winter supplementation within these systems might be beneficial to maintain standard egg quality throughout the year.Vitamins B 2 and B 9 did not vary between management systems, while a seasonal effect with higher levels in summer was shown for B 2 , whereas B 9 was numerically higher during summer within FR systems.
Preliminary results suggest enforced housing for 3 months to protect hens from Avian Influenza infection could have limited impact on egg FA and carotenoid profiles although their Vitamin A content might be reduced.The impact of more prolonged housing however needs to be investigated.
variance (ANOVA), using linear mixed-effects models, were performed in the R statistical environment(R Development Core Team, 2009).This technique assesses the impact of fixed factors (suspected to influence the outcome) but also accommodates the potential effect of random factors (such as egg sourcing, in this case) within a hierarchical dataset.Both main factors and their interactions were assessed, with management systems (CA, FR, OR and EO) and sampling date (B, C, D and E) as fixed factors and sample ID (supermarket/farm) as a random factor.Pairwise comparisons of means were performed using post-hoc Tukey's honestly significant difference test.Vitamin B 2 , B 9 and vitamin D 3 , were examined for two dates only (B to changes in feeding source from vitamin D 3 to 25-hydroxyvitamin D 3 (COMMISSION REGULATION (EC) No 887/2009).Comparison of vitamin D activity reported in former studies are omitted as EFSA (2023) has recently reported an activity factor for 25-hydroxyvitamin D 3 compared to vitamin D 3 of 2.5, which is different from the former used factor between 1 and 5.

6
Funding: This work was supported by funding from Newcastle University and the Sheepdrove Trust.7 Author contributions: Conceptualization -CL, CS, GB; Data curation -EC, MB, HD; Formal analysis -EC, JJ, CS; Funding acquisition -CL, GB; Investigation -EC, HD; Methodology-CL, CS, GB,EC; Project administration -GB, CL; Supervision -CL, CS GB; Roles/Writing -original draft -EC, GB; and Writing -review & editing -all.8 Conflict of interest: none of the authors identifies any conflict of interest with this work.

Table 1 .
Egg collection plan with productions systems, sampling dates and replicate numbers

Table 3 .
Means concentrations ± SE and ANOVA significance levels by production system for carotenoids, vitamin A and E (µg/100 g yolk) in eggs collected from supermarkets (caged, free range and organic systems) and extensive organic farmsIn the limited comparison of April eggs (Table4), differences between systems were confirmed but again, minimal influence of the dates over egg carotenoids or vitamin E content was identified.The exception was for vitamin A which was consistent between FR and OR eggs but 16% higher (p<0.001) in 2018 (compared with 2017), when birds could range outdoors.