PHOSPHOLIPID FATTY ACID PROFILES OF PLASMA AND ERYTHROCYTE MEMBRANES IN DOGS FED WITH COMMERCIAL GRANULATED FOOD

Intake of long-chain n-3 polyunsaturated fatty acids (PUFA) bene ﬁ ts human and animal health. Our study aimed to analyze the long-chain n-3 PUFA content of two types of food and their effect on plasma and erythrocyte phospholipids of Belgian Shepherd dogs. A total of 10 dogs were fed commercial granulated food (Food 1), and another 10 were provided commercial Premium granulated food of high quality (Food 2). All the analyses were performed using gas-liquid chromatography. Our results showed that Food 1 contained more n-3 PUFA than Food 2, which was re ﬂ ected in higher n-3 PUFA in plasma and erythrocyte phospholipids. Because long-chain n-3 PUFA in phospholipids are precursors for antioxidative molecules, further studies should investigate the effects of the analyzed commercial granulated food rich in n-3 on oxidative stress parameters in dogs.


INTRODUCTION
In recent years, many pet owners have abandoned conventional, veterinaryrecommended commercial diets in search of more "natural" and "homemade" choices [1].Still, it is more common practice nowadays for owners to use commercialized labeled food for pets such as canines.Among other ingredients, polyunsaturated fatty acids (PUFA) are important for human and animal health.The conversion of short chain to long chain PUFA is rate-limiting and varies between species [2].The balance of n-6/n-3 ratio in phospholipids, as well as the balance between reactive oxygen species and reactive nitric species on one side and antioxidative defense, on the other, is important for the normal physiological function of organisms.In dogs, fewer infl ammatory mediators were produced when fed diets ratios (n-6/n-3) of 5:1 and 10:1 in comparison with being fed an n-6-rich diet with a fatty acid ratio of 100:1 [3].
Mammals are not able to synthesize fatty acids (FA) with double bonds at C-9, but are able to some extent to elongate and further saturate the aliphatic chain [4].It is well known t hat diet could affect fatty acid profi les in plasma and erythrocyte membranes (EM) and that changing its composition causes changes in many parameters, which is confi rmed in animal and human studies [3].Since the FA composition of the EM correlates with that of other cell membranes, the effect of dietary FA supplementation may be analyzed by studying EM.It is well known that plasma phospholipids profi les refl ect short-term changes in the diet while changes in erythrocyte membrane phospholipids refl ect long-term dietary habits (up to 3 months prior analysis).Comparing those two profi les could lead us to a conclusion about the diet habits of examined animals in general [5].The FAs profi le of the serum phospholipids is related to the average dietary FAs intake during the last 3 to 6 weeks, while the composition of erythrocyte phospholipids depends on the dietary fat intake during the preceding months [5].The FAs profi le in the tissues partly refl ects not only the dietary fat intake but also the effi ciency of FAs metabolism in the body [6].
The FAs profile of tissues and triglycerides (TG) is known to be influenced by many factors, including dietary intake, age, gender, and endogenous metabolism [7].Considerable interest exists in the possible health benefits of increasing dietary intake of n-3 PUFA [8].Three families of long-chain PUFAs with different biological roles exist, n-3, n-6 and n-9 and they are derived from the shortest non-synthesizable precursors: linoleic acid (LA) (18:2, n-6) and alpha-linolenic acid (ALA) (18:3, n-3).Humans can desaturate and elongate ALA, as a precursor of n-3 series, to eicosapentanoic acid (EPA) and docosahexanoic acid (DHA).This process is dependent on aging, presence and type of disease, inflammation processes, and other factors [4].In rats, which are often used as animal models, the rate of conversion of ALA to DHA is high in the liver, although Δ5 and Δ6 desaturases are expressed in many other rodent tissues besides the liver [9].
There are few commercial pet dog foods with EPA and DHA concentrations adequate for the treatment of disease or some vulnerable conditions.Target ranges for EPA and DHA vary quite widely for different conditions but typically fall between 50 and 220 mg/kg body weight.Commercial diets with n-3 fatty acids typically provide less EPA and DHA than desirable and may be advertised as containing fl axseed or canola oil (rich in ALA) [10].
In fact, there are adverse effects associated with the use of n-3, and an increase in the concentration of EPA and DHA in commercial pet food (dogs) makes the topic important to revisit.Those effects include altered platelet function, gastrointestinal adverse effects, potential effects for nutrient excess, weight gain, altered immune function, and effects on glycemic control and insulin sensitivity.As far as dogs and specifi c abnormalities are concerned decreased epithelization of wounds after 5 days (n-6/n-3=0.3:1),increased plasma and urine thiobarbituric reactive substances (n-6/n-3=5.4:1),decreased plasma vitamin E (n-6/n-3=1.4:1), decreased skin and neutrophil leukotriene B4/increased leukotriene B5, lower delayed-type hypersensitivity response (n-6/n-3=1.4:1),decreased CD4+T lymphocyte count (n-6/n-3=1.4:1),decreased lymphocyte proliferation (EPA/DHA=0.8)[11].Metabolically, however, fatty acid patterns of plasma phospholipid fractions again revealed a sparing effect of ALA on LA.It should be mentioned that a direct effect of ALA on improvements of skin and coat could not completely be ruled out in these studies while long-chain n-3 PUFAs from fi sh oil or other marine sources appear to be especially capable of modifying infl ammatory and immune responses [12].
The aim of the study was to analyze the fatty acid content of two different diets for dogs and to examine the effects of their daily consumption on the erythrocyte membrane and plasma phospholipid profi les.

Animals
This study which lasted for 12 weeks, was approved by the Ethics Commission of the Faculty of Veterinary Medicine are included dogs of the Belgian Shepherd (Malinoa) breed from two kennels (1 and 2).In kennels, we selected 10 dogs, (5 females and 5 males), age categories of 3 to 7 years, with body weight 30.2±2.2 kg.We measured weight gain monthly and it changed up to 1 kg/dog.Male dogs weight gain was slightly higher than females but without statistical difference.By the basic examination of the dogs in both kennels (blood pressure, pulse, temperature, breathing, skin condition, and skin cover), there was a constant absence of disease, otherwise, all dogs had neatly managed health cards.According to the constitution they fell into 3 categories, which means they have an ideal weight corresponding to this breed of dog.The dogs had their activities in the morning and in the evening for 60 minutes (walking, running), otherwise, these dogs are considered as working dogs.In kennel number 1, the dogs were fed commercial granular foods that normally satisfi es the standard nutritional needs of dogs (I).In kennel number 2, the dogs were fed Premium granulated food of high quality, this being the most sold dog food in Serbia (II).The amount of food (400 g/day/dog) was divided into two meals one in the morning and the other in the evening at the same time each day.

Sample collection and analysis
Granulated foods samples (I and II) (from four representative large markets) were analyzed.The primary sample was generated by mixing an equal portion of four samples taken from different markets.Five replicate samples of the composite sample were analyzed by standard laboratory methods to measure the concentration of proteins, carbohydrates, lipids, minerals, and water.Fatty acid composition from the lipids was done by standard laboratory procedure, as described below.At the end of the study blood samples were taken from the (vena cephalica antebrachii), with the aid of EDTA vacuum blood collection tubes.Erythrocytes and plasma were separated and stored at a temperature of -80 C 0 .After blood sampling by the routine method, analysis of erythrocytes and plasma was done by gas-liquid chromatography (GC).
Nutritional analysis was carried out by an accredited chemical laboratory at the Institute of Public Health in Požarevac.Ash content was determined by the direct gravimetric method which includes ashing of the samples in an oven at 550 °C until a constant weight was attained.Moisture was determined gravimetrically [13].Crude protein content was estimated based on the total nitrogen content of the sample determined by the Kjeldahl method (AOAC 955.04D) [13].Crude fat content was determined gravimetrically (Soxhlet extraction, AOAC method [13].Total carbohydrate content, crude "by difference", was calculated by the following formula: total carbohydrate (%) = 100% -% (protein + ash + fat + moisture).The energy content of food was calculated based on determining content by the following formula: Energy value (estimated, kJ/100 g) = [4 x protein (%)] + [4 x carbohydrate (%)] + [9 x fat (%)].

Isolation of lipids
The method consists of homogenization of plasma with a 2:1 chloroform/methanol mixture.Washing of the mixture with a 5 times smaller volume of water or saline (0.9 g NaCl in 100 ml of water).The resulting mixture was separated into two phases.The lower phase was the total pure lipid extract.
First the lipids present in the volume of 0.5 ml of erythrocytes were extracted with a mixture of chloroform isopropanol (7:11) following the prescribed procedure [14].After that we isolated the phospholipids from the lipid subclasses by thin-layer chromatography on silica-gel plates using petroleum ether, diethyl ether and glacial acetic acid (87:12:1, by volume).In a modifi ed procedure by Christopherson and Glass [15], we made a direct transesterifi cation of fatty acids.Lipid extracts derived from hexane were evaporated under a nitrogen stream to full vapor and dry bottom of Eppendorf 's tubes.After that, the residue was dissolved in 10µl hexane.We took 1 µl and injected it into a chromatograph.Methyl esters of fatty acids were analyzed by gas-liquid chromatography in a Shimadzu chromatograph GC 2014 (Kyoto, Japan) equipped with a fl ame ionization detector on an Rtx 2330 column (60mm x 0,25 mmID, fi lm thickness 0,2 µm, Restek, Bellefonte, PA, USA).Adequate separation was obtained over a 50 min period with an initial temperature of 140 0 C maintained for 5 min.The temperature was then increased to 220 0 C at a rate of 3 0 C/min and kept at the fi nal temperature for 20 min.The identifi cation of fatty acid methyl esters (FAME) was made by comparing peak retention times with standard mixtures (PUFA-2 and/ or 37 FAMEs mix, Supelco, Bellefonte, PA, USA).Finally, the content of fatty acids, from 16:0 through 22:6n-3, was expressed as a percentage of the total content of identifi ed fatty acids.

Statistical analysis
Statistical analysis was done by one-way ANOVA test and Student t-test (signifi cance p<0.05) in SPSS.All results were expressed in percentages of a total of 100%.

Biochemical parameters
Analyzed biochemical parameters such as triglycerides and plasma cholesterol did not show sta s cal signifi cance in the plasma of canines treated with two diff erent diets.Although there were some changes in LDL cholesterol, previously reported by Ravic et al [16].

Analyzed food
There were differences in FA percentage between the examined foods.There was an equal percentage of palmitic (16:0) acid, but stearic acid was more abundant in Food 1 (p<0.01).Monosaturated FA, specially palmitoleic acid, was present at a higher percentage in Food 2 (p<0.01),while oleic acid (18:1 n-9) was detected almost in the same percentage.Linoleic acid (18:2 n-6) was also present in more than 10% in both Food 1 and Food 2. Long-chain PUFAs were present in a very low percentage with statistically signifi cant differences between Foods.ALA (18:3n3) was statistically signifi cant in a higher percentage in Food 1 (p<0.001),as well as EPA (p<0.001) and DHA (p<0.001) compared to Food 2 (Table 1).

Dog plasma and erythrocyte membrane phospholipid fatty acids
It is well known that plasma phospholipids fatty acids profi les represent the shortterm (up to several days) food intake, while erythrocytes present data composition of fatty acids for long term (last 3 months) food intake.In our study, we compare dogs' FAs profi les in plasma and erythrocytes membrane phospholipids in two kennels (Table 2-4).
In the erythrocyte membrane, phospholipids´ percentage of fatty acids has almost the same distribution and percentages as in plasma phospholipids between Kennels.Saturated fatty acids (PA and SA) (p<0.01) were in higher percentages in Kennel 1-fed dogs as well as monounsaturated FA (p<0.1).Percentage of EPA (p<0.001),DPA (p<0.01), and DHA (p<0.001) were increased in Kennel 1 in plasma phospholipids, cont.Table 1.
while ALA concentration did not signifi cantly change.These results go along with the plasma phospholipids FA distribution already mentioned.
These results suggest that the differences are negligible between plasma and erythrocytes phospholipids fatty acids percentages.

DISCUSSION
Generation of SFA, MUFA and PUFA fatty acids is connected with mediators such as prostaglandins, leukotrienes and others, which are able to infl uence metabolic changes in dogs in health and disease.The n-6/n-3 balance is important in everyday balanced diet and is recommended to be as low as it could be.Following up on intakes of linoleic acid, as n-6 family fatty acid, and its metabolic pathway to DGLA and arachidonic acid (AA) and cell membrane incorporation is of great importance as information of biochemical or clinical parameters for the evaluation of dog´s health [5].Omega-6 linoleic and its transformation to ARA affect not only from the cell structural point but also affect the membrane response to stimuli, thus membrane fatty acid composition can be useful as information of pro and anti-infl ammatory predisposition of the canine organism [17].
By correlating plasma and erythrocyte phospholipids FA composition after the treatment, we investigated if FA from both plasma and erythrocytes could represent markers of dietary n-3 intake.
Our discussion is more based on n-3 PUFA signifi cance and intake in canines.Certain results indicate a benefi cial effect of n-3 long-chain polyunsaturated fatty acids (LC PUFA), especially eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) on canine health.Several studies have examined the effects of fi sh oil, linseed oil, or combinations on plasma phospholipids fatty acids profi le in canine species [18], lymphocyte proliferation, or neutrophil composition in laboratory animals [19,20].
The nutrient requirements of canine athletes who have a greater capacity for fat oxidation are unique concerning that canine metabolism is unique [10].Relative to body mass size dogs metabolize free fatty acids at twice the rate observed in men.Consequently, the muscles are more adapted to use fat than human muscles [21].
Usual daily food requirements are water, energy, major nutrients, fi bers, minerals, and vitamins.High fat diet, fatty acids chain length and saturation affect a variety of issues from infl ammation to oxidative stress in animals [22,23].However, there is very little information regarding optimal dietary fats intake for canine athletes [23].
Medium-chain triglycerides yield 8 carbon to 12 carbon-free fatty acids (coconut, palm oil) when digested and are directly transported through absorption into the blood, bound to albumin, to the liver via the portal circulation, leading to sparing of glycogen [24].The fatty acid composition also infl uences detection in scent-trained dogs, hunting dogs, and service dogs.It is well known that polyunsaturated fatty acids may modestly improve performance in dogs that require activity as a part of their work [25].The use of carbohydrates as a major dietary substrate is essential.Soluble fi bers may alter the large intestinal microfl ora which produces short-chain fatty acids.A small amount of soy fi ber and fructooligosaccharides are components of gastrointestinal veterinary diets.Defi ciencies in major minerals have been observed in dogs fed with non-traditional diets.It is recommended that bones should be ground into the diet to improve calcium and phosphorus balance.Vitamins (fat-soluble or water-soluble) are involved in cellular metabolism or coenzymes in the citric acid cycle.
A study by Poel at al (2017) [26] was conduc ted to determine the stability of two preparations of GAA (granulated and crystallized) and CMH in a moist and a dry dog food formulation during production and storage.Most commercial dog foods contain vitamins above the minimum requirement.Diets high in PUFA (fi sh-based sled dog diets) should contain more vitamin E to prevent lipid peroxidation.In the study by Vessecchi et al., the study aimed to evaluate the macronutrient composition, fatty acids and amino acids profi le, and essential minerals contents of vegan pet foods available in the Brazilian market, and to assess their compliance with recommended allowances for dogs and cats [27].
Omega -3 fatty acids incorporat ed in cell membranes, especially EPA and DHA, decrease clinical signs of osteoarthritis in dogs.EPA serves as a substrate for the COX and LOX enzymes.Omega-6 FA are involved in physiological processes while infl ammation contributes to the formation of "proinfl ammatory" prostaglandins and leukotrienes.Omega-3 produces less infl ammatory prostanoids and 5-leukotrienes.Dogs have limited ability to convert ALA to DHA.That is the reason for diet EPA and DHA consumption directly.Diets that contain EPA and DHA have been recommended for degenerative joint diseases, aging in general, as anti-infl ammatory supplements, and in growing puppies [25].Currently, omega-3 FAs are used in managing many diseases including neoplasia, dermatologic diseases, hyperlipidemia, cardiovascular, gastrointestinal, and orthopedic diseases [11].There are few commercial pet foods with EPA and DHA adequate for the treatment of disease.Joint diets, renal diets dermatologic conditions contain more omega-3 fatty acids than maintenance diets.
Commercial diets with omega-3 provide less EPA and DHA than desirable and contain fl axseed instead.Because diets with ALA have different effects when compared with diets enriched in EPA and DHA composition information may very much contribute to results.Purushotman et al. study on beagles fed with fl axseed added to a basal diet (57% alfa-linolenic acid, ALA at the rate 100ml/kg for 3 weeks) showed that plasma ALA, EPA, and LA increased steadily and signifi cantly from 0-22 days while arachidonic acid showed no difference.Plasma DHA showed no signifi cant changes over time which agrees with previous studies in dogs [2].Dunbar et al., concluded that hepatic conversion of DHA is slow in the canine and after its production from ALA it is likely to relocate to neurological tissue [29].Extensive studies are required for clarifying this issue and to confi rm how different breeds metabolize PUFA.Waldron et al. confi rm earlier reports in their study on dogs fed a fi sh oil diet that ALA was converted to EPA and further elongated to DPA but was not converted to DHA in plasma phospholipids [18].The benefi cial effect of n-3 FA due to their competition with n-6 for cellular membrane incorporation leads to the suggestion that n-6: n-3 ratio could be used as a dietary index to modulate cell composition and cellular function [30].
Still, the type and amount of n-3 PUFA, but not the n-6:n-3 ratio in diets, contribute to membrane n-3's highly unsaturated fatty acid composition and further changes in cell function [18].Stoeckel et al. (2011) concluded that in dogs an increase of dietary n-3 FA content leads to a rapid inclusion of n-3 into the erythrocyte membrane, regardless of whether the n-3 FA is offered as an enriched diet or as a normal diet supplemented with an n-3 FA additive [31].
The cardiovascular benefi ts of n-3 PUFA could originate from their ability to improve lipid metabolism and reduce the synthesis of proinfl ammatory eicosanoids derived from n-6 PUFA.At an adequate level of incorporation, EPA and DHA infl uence the membrane fl uidity as well as membrane protein-mediated reactions, generation of lipid-mediators, cell signaling, and gene expression in different cells [32].Data on cats, together with the results of lipoprotein analysis, indicate possible disturbances in the hepatic transformation of LDL and VLDL, and a high risk of atherogenic events [33].Dogs have the capacity to metabolize n-3 fatty acids and the effects of omega-3 fatty acids on the skin and coat, infl ammatory responses, and neurologic development in puppies are quite visible [34].
Concerning specifi c benefi cial effects that fi sh oil or EPA and DHA have on canines, it would be useful to add it to the examined food or as an everyday supplement to those dogs.Diagnostic tests performed in dogs provide valuable information [35] and our data can contribute to it.
Our further examination and studies will address it, and we planned a study on the treatment of PUFA in canines (EPA +DHA).That study could be useful and give us a more complete conclusion about the signifi cance of EPA and DHA in canine feeding.

Table 3 .
Estimated desaturase 9,6,5 activities in plasma and erythrocytes in dogs in Kennel 1 and 2.

Table 4 .
Percentage of overall n-3, n-6 and n-6/n-3 ratio in Kennel 1 and Kennel 2 in plasma and erythrocytes of examined dogs.