Skip to main content Accessibility help


  • Access
  • Cited by 1


      • Send article to Kindle

        To send this article to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

        Note you can select to send to either the or variations. ‘’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

        Find out more about the Kindle Personal Document Service.

        Sensitivity of fatty acid desaturation and elongation to plasma zinc concentration: a randomised controlled trial in Beninese children
        Available formats

        Send article to Dropbox

        To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

        Sensitivity of fatty acid desaturation and elongation to plasma zinc concentration: a randomised controlled trial in Beninese children
        Available formats

        Send article to Google Drive

        To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

        Sensitivity of fatty acid desaturation and elongation to plasma zinc concentration: a randomised controlled trial in Beninese children
        Available formats
Export citation


Zn status may affect fatty acid (FA) metabolism because it acts as a cofactor in FA desaturase and elongase enzymes. Zn supplementation affects the FA desaturases of Zn-deficient rats, but whether this occurs in humans is unclear. We evaluated the associations between baseline plasma Zn (PZn) concentration and plasma total phospholipid FA composition, as well as the effect of daily consumption of Zn-fortified water on FA status in Beninese children. A 20-week, double-blind randomised controlled trial was conducted in 186 school age children. The children were randomly assigned to receive a daily portion of Zn-fortified, filtered water delivering on average 2·8 mg Zn/d or non-fortified filtered water. Plasma total phospholipid FA composition was determined using capillary GLC and PZn concentrations by atomic absorption spectrometry. At baseline, PZn correlated positively with dihomo-γ-linolenic acid (DGLA, r 0·182; P=0·024) and the DGLA:linoleic acid (LA) ratio (r 0·293; P<0·000), and negatively with LA (r −0·211; P=0·009) and the arachidonic acid:DGLA ratio (r −0·170; P=0·036). With the intervention, Zn fortification increased nervonic acid (B: 0·109; 95 % CI 0·001, 0·218) in all children (n 186) and more so in children who were Zn-deficient (n 60) at baseline (B: 0·230; 95 % CI 0·023, 0·488). In conclusion, in this study, Zn-fortified filtered water prevented the reduction of nervonic acid composition in the plasma total phospholipids of children, and this effect was stronger in Zn-deficient children. Thus, Zn status may play an important role in FA desaturation and/or elongation.


Tsitsi Chimhashu and Linda Malan are co-first authors.

Zn and essential fatty acids (FA) are required for optimal growth, immune response, gene expression, visual development, neurotransmission and cognition( 1 3 ). The high prevalence of Zn deficiency is a public health concern: 17 % of the world’s population may be at risk of inadequate Zn intake. According to data derived from food balance sheets and demographic data, 17–25 % of the Benin population is at risk of Zn deficiency. Overall, the Beninese population is classified as moderately Zn deficient( 4 6 ).

Zn acts as a cofactor of the FA desaturase and elongase enzymes( 7 , 8 ). The desaturase and elongase enzymes convert the essential linoleic acid (LA) and α-linolenic acid (ALA, 18 : 3n-3) to their long-chain PUFA (LCPUFA) metabolites, as well as the SFA palmitic acid (16 : 0) to nervonic acid (24 : 1n-9)( 9 ). A relationship between Zn and FA metabolism has been confirmed in several animal studies. Zn deficiency altered the FA composition of phospholipids in rat liver, plasma and erythrocytes( 8 , 10 14 ). Some studies have reported higher levels of LA (18 : 2n-6)( 12 , 15 17 ) whilst some have reported lower levels of arachidonic acid (AA, 20 : 4n-6)( 11 , 15 , 16 , 18 20 ) and others high or normal levels of AA in tissue phospholipid of Zn-deficient animals. This relationship was also shown in a recent human study( 21 ). It has also been shown that Zn facilitates the incorporation of FA into phospholipids( 22 ).

Plasma Zn (PZn) is the biomarker of choice for the assessment of Zn status in populations, however the search for a sensitive and specific biomarker for assessing body Zn concentrations at the individual level is still ongoing( 23 ). Although PZn responds well to supplementation, as assessed in depletion-repletion studies( 24 ), PZn has been found to vary due to factors that are unrelated to Zn intake or status, such as infection/inflammation, time of sample collection, sample handling, among others, and might in fact not reflect individual Zn status( 25 ). Recent studies in chickens (Gallus gallus) and in humans have suggested the ratio of LA:dihomo-γ-linolenic acid (DGLA; 20 : 3n-6) or the similarly used DGLA:LA ratio, to be a potential biomarker of Zn status( 21 , 26 ). This is because the conversion of LA to DGLA requires desaturase and elongase enzymes. Reed et al. ( 26 ) and Knez et al. ( 21 ), respectively, found that Zn deficiency was positively associated with the LA:DGLA ratio in erythrocyte and plasma membranes. Knez et al. ( 21 ) found that a lower dietary Zn intake was associated with a higher LA:DGLA ratio in apparently healthy 28–55-year-old adults, but this relationship has not yet been investigated in children.

Zn supplementation can increase dietary Zn intake and improve Zn status( 27 ). Although the results from fortification studies are inconsistent( 28 ), Zn fortification of staple food grain flours such as wheat, millet and maize is a recommended strategy to improve dietary Zn intake( 29 ). However, the bioavailability of Zn from many fortified grain flours is low as they have high phytate:Zn molar ratios( 30 ). Phytic acid is an inhibitory factor that reduces Zn absorption( 31 ). Fortified water as a fortification vehicle for Zn has the benefit of being free of inhibitors of absorption, making it a potentially good source of bioavailable Zn( 32 ). In a randomized, controlled trial, Galetti et al.( 33 ) established that consumption of water from a filter able to fortify water with Zn at physiological concentrations, maintained PZn concentration of Beninese school children who were in the treatment group at baseline levels, whereas the PZn concentration of children in the control group was decreased over the 20-week study period.

In this secondary analysis of data from the Benin study( 33 ), our study aims were to assess: (1) whether PZn and plasma total phospholipid FA composition are correlated in Beninese children; and (2) whether Zn fortification affects the LA:DGLA or DGLA:LA ratio, as a potential Zn status biomarker. Our hypotheses were: (1) that there are baseline associations between PZn and plasma total phospholipid FA composition; and (2) that children in the intervention group have better desaturase and elongase enzyme activities, estimated by product to precursor FA ratios, compared with children in the control group after a 20-week intervention period.


Study design, site and subjects

This study was part of a 20-week double-blind randomised controlled trial conducted in rural Benin in school age children between the months of February and June 2013. Its main objectives were to determine Zn bioavailability from Zn-fortified drinking water providing a mean of 2·8 mg Zn/d, and assess the efficacy of the Zn-fortified water in improving Zn status over a 20-week period( 33 ). At baseline, midpoint, and endpoint we measured PZn and inflammatory markers that may confound PZn, α1-acid glycoprotein (AGP) and C-reactive protein (CRP). The primary outcome of the main study was PZn concentration. We conducted the study at the primary school of Kotopounga, a rural town in north-western Benin, where the climate is a sub-equatorial/tropical savanna with one rainy season (May–June to September–October) and one dry season (October–November to April–May). Study procedures were translated into the local Waama and Otamari languages. Children were excluded from the study if they had: (1) severe anaemia (defined as Hb<70 g/l), (2) major chronic diseases, (3) long-term medication use or (4) recent Zn supplementation. We randomised children (n 277) at enrolment to one of three treatment arms: one group received Zn-fortified, filtered drinking water (Zn+filter group, 2·8 mg Zn/d), one group received non-fortified filtered drinking water (filter group, virtually no Zn) and the third group received local drinking water from a pumped well. In this study, we included only the Zn+filter (n 89) and filter (n 97) groups. Plasma total phospholipid FA composition was determined in 186 subjects (Fig. 1).

Fig. 1 Flow diagram of the secondary study in context of the main study.

Sample size

To our knowledge, no similar study has been previously reported. We therefore based the sample size calculation for the intervention hypothesis of this secondary study on an independent t test analysis in G * power( 34 ). A sample size of fifty-two per group was needed to give the study 80 % power to detect a standardised effect size (0·6) of the intervention if significance was set at P=0·05.


Water was purified daily either with a filter equipped with a Zn fortification chamber (Zn+filter) or with a placebo chamber (filter), and hygienically stored in tanks until consumption. The Zn concentration in water was monitored daily, both with a rapid assessment method (Aquaquant; Merck) in the field, and after the intervention by flame atomic absorption spectrometry (Varian Inc.) at the Laboratory of Human Nutrition at ETH Zurich.

From Monday to Friday, each child received a morning and an afternoon serving of 300 ml filtered water (during the 10.00 hours and the 15.00 hours breaks), with the exception of Wednesdays, when they consumed only the morning serving. The 300-ml Zn-fortified water serving contained a median Zn dosage of 2·4 (interquartile range (IQR) 2·0, 3·2) mg, which resulted in an average daily Zn intake of 4·3 (IQR 3·5, 5·2) mg and 2·8 (IQR 0·0, 4·5) mg on school days and over the entire study, respectively.

Dietary Zn intake was evaluated in a subsample of thirty-six children; mean Zn intake was 8·1 (sd 62·3) mg/d( 4 ). Since the main contributing sources of Zn were maize (38 %), sorghum (21 %), beans and legumes (16 %), vegetables (5 %) and cassava, yam and rice (2·7 % each), the estimated daily intake of absorbable Zn was 1·5 (sd 0·5) mg( 35 ) according to the WHO algorithm for unrefined cereal-based diet and 1·9 (sd 0·3) mg according to the algorithm by Miller et al. ( 36 ).

Blood sampling, biochemical analysis and anthropometry

Two morning (mean time of day: 10:24 (sd 1) h 18 min) fasting (subjects in non-fasting state were 5 and 3·3 % at baseline and endpoint sampling, respectively) venous whole-blood samples from each study participant were collected directly into one trace element–free lithium heparin tube for PZn analysis and one EDTA tube (Sarstedt) for analysis of inflammation markers. Hb was assessed with a portable HemoCue 201+ photometer (HemoCue AB). The blood samples were refrigerated at 4–8°C immediately after collection, centrifuged within 1 h (3000 g for 10 min at room temperature), divided into aliquots in acid-washed plastic vials (Eppendorf AG), and stored at −20°C (within 5 h) until analysis. The samples were shipped to ETH Zurich on dry ice. PZn concentrations were measured at ETH Zurich as described previously( 33 ). Briefly, PZn was measured by flame atomic absorption spectrometer (AA240FS; Varian Inc.) using a commercial aqueous standard (Titrisol 1.009953.0001; Merck) for external calibration, and Seronorm Trace Elements Serum L-2 (Sero AS) as reference material, which delivered values within acceptable ranges as specified by the manufacturer. Standardised protocols by the International Zn Nutrition Consultative Group (IZiNCG) were followed during sample collection, handling, and analysis in order to avoid exogenous or endogenous Zn contamination( 37 ).

CRP (Code A0073; Dako Denmark) and AGP (Code Q032605-2; Dako Denmark) were analysed by using a sandwich ELISA( 38 ). Liquicheck Trilevel (Bio-Rad Laboratories Inc.) was used as control material with each run of analysis, and measured values were within acceptable ranges as specified by the manufacturer.

PZn values were adjusted with correction factors according to the subjects’ respective phases of inflammation: Subjects in incubation (CRP>5 mg/l and/or AGP>1·0 g/l) with 1·08, those in early convalescence (CRP>5 mg/l and AGP>1·0 g/l) with 1·17 and those in late convalescence (CRP<5 mg/l and AGP>1·0 g/l) with 1·06( 39 ). The prevalence of Zn deficiency was calculated by using sex- and age-specific PZn cut-offs suggested by the IZiNCG( 40 ). Low PZn concentration was defined as PZn <9·9 μmol/l for children aged <10 years (morning sampling), <8·7 μmol/l for children aged <10 years (afternoon sampling), <10·7 μmol/l for females aged ≥10 years (morning fasting sampling), <10·1 μmol/l for females aged ≥10 years (morning non-fasting sampling), <9·0 μmol/l for females aged ≥10 years (afternoon sampling), <11·3 μmol/l for males aged ≥10 years (morning fasting sampling), <10·7 μmol/l for males aged ≥10 years (morning non-fasting sampling), and <9·3 μmol/l for males aged ≥10 years (afternoon sampling)( 41 ). Subclinical inflammation was defined≥5 mg/l for CRP and/or 1 g/l for AGP. Severe anaemia was defined as Hb≤70 g/l( 42 ).

Baseline and endpoint plasma samples for each subject were prepared for plasma total phospholipid FA analysis and analysed with GLC together in the same batch (pair-wise analysis) at the Non-Communicable Diseases Research Unit of the South African Medical Research Council. In short, lipids were extracted from plasma with chloroform: methanol (2:1 vol: vol; containing 0·01 % butylated hydroxytoluene) by using a modification of the Folch et al. ( 43 , 44 ) method. Lipid extracts were concentrated, the neutral lipids separated from the total phospholipid fraction by using TLC (silica gel 60 plates, 10×10 cm; Merck), and eluted with diethyl ether: petroleum ether: acetic acid (30:90:1; vol: vol: vol). The lipid band that contained total phospholipids (application origin) was removed from the TLC plate and trans-methylated with methanol: sulphuric acid (95:5 vol: vol) at 70°C for 2 h to yield fatty acid methyl esters (FAME). The resulting FAME were extracted with water and hexane. The organic layer was evaporated, redissolved in hexane, and analysed (sample injection volume 1 μl) by using GLC on a finnigan Focus Gas Chromatograph equipped with flame ionisation detector (Thermo Electron Corporation) and a 30-m capillary column of 0·32 mm internal diameter; BPX70 0·25 μm (SGE International Pty Ltd). Gas flow rates were: N2 (make up gas), 25 ml/min; air, 250 ml/min; H2 (carrier gas), 25 ml/min, with a split ratio of 20:1. GC-oven temperature programming was linear at 4·5°C/min, initial temperature 140°C (hold-time 1 min), final temperature 220°C (hold-time 5 min), injector temperature 220°C, and detector temperature 250°C. The FAME were identified by comparison of the retention times with those of a standard FAME mixture (27 FAME; NuChek Prep Inc.). Relative percentages of FA were calculated by taking the area count (AUC) of a given FAME as a percentage of the total area count of all FA identified in the sample.

Product:precursor FA ratios were used as a proxy to reflect enzyme activity. The GLA:LA (although not widely used as GLA is rapidly converted to DGLA), AA:DGLA and DGLA:LA ratios were used to estimate Δ-5- and Δ-6-desaturase activities( 45 , 46 ). Anthropometric measures were done as described previously( 33 ). Total n-6 LCPUFA was defined as DGLA plus AA.

Statistical analysis

The software package SPSS, version 22 (IBM), together with Excel, version 14 (Microsoft Office) were used to analyse the data of this secondary study. All data were checked for normality using histograms and the Shapiro–Wilk test and any value that departed more than 3 sd from the standardised mean were examined. When data were not normally distributed, values were either logarithmically or square root transformed before statistical analysis. Values in the text and in the tables are represented as mean values and standard deviations for normally distributed data, medians and IQR for non-normal data, and percentages for categorical variables. Baseline differences in participant characteristics between the groups were tested by using independent t tests for continuous variables and Pearson’s χ 2 test for categorical variables. Partial correlation coefficients, adjusted for sex, BMI and age were calculated to determine associations between FA and PZn. The effects of the Zn fortification on plasma total phospholipid FA composition were analysed with one-way ANCOVA by using respective baseline FA values, age, sex and BMI as individual level covariates in the total group, as well as in the children who were Zn deficient at baseline. Parameter estimates, unstandardised β-coefficients (β values) with 95 % CI, were obtained to indicate the size and direction of the intervention effects. Significance was set at P<0·05.

Ethical considerations

The study was conducted according to the principles of the Declaration of Helsinki of 1975, as revised in 1983, and an independent Data Safety Monitoring Board was set up to monitor the intervention. Before the screening, parents gave informed consent by either a written signature or a fingerprint. The ethical committees of, the Beninese National Ethics Committee for Health Research (no. 029; 19 October 2012), and the ETH Zürich, Switzerland (EK-N-47), and the North-West University, Potchefstroom, South Africa (NWU-00086-15-A1) approved the study protocol. This trial was registered at PRS with the identifier NCT01790321.


Study participant characteristics

Table 1 summarises the baseline anthropometric and biochemical characteristics of the study participants. Hb was significantly higher in the Zn+filter group than in the filter group at baseline (P=0·030). Anthropometric measures between groups at baseline did not differ. Median PZn concentration was 10·8 μmol/l (IQR 9·7, 12·1) for the Zn+filter group and 10·5 μmol/l (IQR 9·6, 11·8) for the filter group at baseline, and a prevalence of Zn deficiency in all the children was 32 % (n/N 60/186). There were no differences at baseline between the plasma total phospholipid FA compositions of the treatment groups except for LA (P=0·040) that was higher in the Zn+filter group and the AA:DGLA ratio (P=0·030) that was lower in the Zn+filter group as seen in Table 3.

Table 1 Baseline anthropometric measurements and biochemical characteristics of Beninese children by treatment group* (Percentages and numbers; medians and interquartile ranges (IQR); mean values and standard deviations)

HAZ, height-for-age z score; WAZ, weight-for-age z score; BAZ, BMI-for-age z score; MUAC, mid upper arm circumference; PZn, plasma Zn; CRP, C-reactive protein; AGP, α1-acid glycoprotein.

* Differences between the Zn+filter group and filter groups were examined by using independent t tests for continuous variables and by using the χ 2 test for categorical variables.

Percentage (%, n/N× 100) (all such values). Low PZn concentration defined as PZn <9·9 μmol/ for children aged <10 years (morning sampling), <8·7 μmol/l for children aged <10 years (afternoon sampling), <10·7 μmol/l for females aged ≥10 years (morning fasting sampling), <10·1 μmol/l for females aged ≥10 years (morning non-fasting sampling), <9·0 μmol/l for females aged ≥10 years (afternoon sampling), <11·3 μmol/l for males aged ≥10 years (morning fasting sampling), <10·7 μmol/l for males aged ≥10 years (morning non-fasting sampling), and <9·3 μmol/l for males aged ≥10 years (afternoon sampling).( 40 )

BMI=weight (kg)/height2 (m2).

§ Hb was significantly higher in the Zn+filter group (P=0·03).

Baseline associations of plasma zinc with plasma total phospholipid fatty acid composition

Table 2 shows the corrected correlation coefficients at baseline for the associations between PZn and plasma total phospholipid FA composition, adjusted for age, sex, and BMI. Positive correlations were found between PZn and the DGLA:LA (r 0·293, P≤0·000), and the GLA:LA (r 0·359, P=0·032) ratios. PZn also correlated positively with DGLA (r 0·182, P=0·024), and negatively with LA (r −0·211, P=0·009), and the AA:DGLA ratio (r −0·170, P≤0·032). There were no significant differences at baseline in plasma total phospholipid FA composition between the Zn-sufficient and Zn-deficient groups (data not shown). The adjustment of PZn values with correction factors for inflammation did not affect the main findings.

Table 2 Baseline correlations of plasma zinc with plasma total phospholipid fatty acid composition* (Mean values and standard deviations; medians and interquartile ranges (IQR))

OA, oleic acid; LA, linoleic acid; GLA, γ-linolenic acid; DGLA, dihomo-γ-linolenic acid, AA, arachidonic acid; ALA, α-linolenic acid; LCPUFA, long-chain PUFA.

* Correlations were determined by using linear regressional analysis, adjusted for sex, BMI, age and partial correlation coefficients are shown. PZn values were corrected using adjusted and AGP correction factors( 39 ). Fatty acids with low sample sizes (<39), due to peak area counts below 0·1 % of the total chromatogram were not included in analysis.

Data were log transformed to perform ANCOVA.

Data were squared to perform ANCOVA.

§ Data were reciprocally transformed to perform ANCOVA.

Effects of zinc fortification on plasma zinc and plasma total phospholipid fatty acid composition

As shown in Table 3, Zn fortification significantly prevented the reduction of PZn concentration in the Zn+filter group (P=0·030) compared with the filter group. The effects of Zn fortification on the plasma total phospholipid FA composition of all the children are given in Table 3. When compared with the filter group, Zn fortification prevented reduction of nervonic acid (P=0·047) in the Zn+filter group.

Table 3 Effects of zinc fortification for 20 weeks on plasma total phospholipid fatty acid composition of 6- to 10-year-old Beninese children (Mean values and standard deviations; medians and interquartile ranges (IQR); β-coefficients and 95 % confidence intervals)

PZn, plasma Zn; OA, oleic acid; LA, linoleic acid; GLA, γ-linolenic acid; DGLA, dihomo-γ-linolenic acid; AA, arachidonic acid; ALA, α-linolenic acid; LCPUFA, long-chain PUFA.

* Values changed significantly from baseline to endpoint (P<0·05).

Intervention effects were estimated by one-way ANCOVA, adjusted for respective baseline value, sex, age and BMI. Fatty acids with low sample sizes (<21 in Zn+filter and 21 in filter group), due to peak area counts below 0·1 % of the total chromatogram were not included in analysis.

Data were log transformed to perform ANCOVA.

§ Data were squared to perform ANCOVA.

The effect of Zn fortification on the plasma total phospholipid FA composition in only the Zn-deficient children is shown in the online Supplementary Table S1. In these children, Zn fortification significantly increased nervonic acid (P=0·030) and decreased ALA (P=0·008).


In rural Beninese school age children, PZn concentration at baseline was positively correlated with plasma total phospholipid FA composition for DGLA, DGLA:LA and GLA:LA ratios. Thus, confirming a previously reported relationship between PZn and the DGLA:LA ratio, which is a proxy marker for Δ-6-desaturase activity and has been proposed as a potential biomarker for Zn status. The most unexpected finding, however, was that Zn fortification prevented reduction of the longer-chain n-9 MUFA, nervonic acid in plasma total phospholipids, particularly in children that were Zn-deficient at baseline.

The positive association of PZn concentration with the DGLA:LA ratio at baseline indicates an enhanced conversion of precursor FA to long-chain FA metabolites with increasing PZn concentration. Accordingly, we showed that PZn correlates with DGLA. An inverse relationship between PZn concentration and LA also suggests that a higher availability of circulating Zn results in a more efficient conversion of LA, an essential FA, to longer-chain metabolites. Zn is a cofactor for desaturase enzymes( 47 ), and Zn deficiency may result in a disruption of the rate-limiting Δ-6-desaturase step in FA synthesis, which is necessary for the conversion of LA into DGLA( 48 ). Additionally, Zn deficiency and low dietary Zn intake affect the electron transport of the mitochondrial electron transport chain( 21 ), as microsomal Zn levels directly determine electron transfer chain activity( 49 ). There was a weak positive association between PZn concentration and the GLA:LA ratio; this was likely due to the increased elongation of GLA from LA( 45 ). Subsequently, an improvement in Zn status, as previously shown in other studies, would be expected to result in a positive association between PZn and the AA:DGLA ratio. The proportion of LCPUFA to their PUFA precursors is expected to be increased as the availability of circulating Zn results in an elevated desaturase activity. This study, however, found an inverse relationship between the two, which may indicate that only Δ-6-desaturase activity, but not Δ-5 activity necessary for the conversion of DGLA to AA, is affected by Zn status. This supports the value of the LA:DGLA or DGLA:LA ratio as an emerging biomarker for Zn status, as proposed by Knez et al. ( 21 ) and Reed et al. ( 26 ), although in the latter study it was not clear whether the ratio shift was due to a decrease in feed consumption or decrease in Zn intake.

Contrary to findings from the study conducted by Reed and colleagues, we found no significant difference in the DGLA:LA ratio in the Zn-deficient group when compared with the Zn-sufficient group (0·15 (sd 0·43) and 0·14 (sd 0·42), respectively, P=0·217)( 26 ). However, it should be noted that the study conducted by Reed et al. was an animal study, and the animals were kept in a highly controlled environment and fed a standardised diet.

To our knowledge, this is one of the first human studies to investigate the effect of Zn intervention on FA status. Most previous studies used animal models and found higher LA and lower n-6 LCPUFA in the blood/tissue of Zn-deficient animals, indicating a disruption in desaturation and/or elongation of FA( 8 , 15 , 18 , 19 , 49 ). A novel finding of this study is that Zn fortification prevented the reduction of nervonic acid, a MUFA. This suggests that greater availability of circulating Zn not only increases desaturase activity, but may increase activity of the elongase enzymes in the n-9 synthesis pathway. Nervonic acid is produced from oleic acid, which occurs naturally in various animal and vegetable fats and oils, after undergoing three consecutive chain elongation steps catalysed by elongases. FA elongases, which function with FA desaturases to generate many of the long-chain MUFA- and PUFA, are likely affected by Zn status( 50 ) as an inhibition in the terminal desaturase reaction has been described in Zn deficiency( 49 ). This finding may be physiologically relevant as nervonic acid is required for the biosynthesis of sphingomyelin in the human brain( 51 ).

Overall, the proposed molecular mechanisms by which Zn affects the activity of desaturase and/or elongase enzymes were supported by results from this study. Even with a low dosage of Zn and a small intervention effect( 33 ), Zn fortification had a detectable effect on the FA composition of Zn-deficient children.

The strengths of this study include the randomized and controlled study design, the direct supervision of each child during water consumption to assess compliance, the provision of highly bioavailable Zn which doubled the amount of absorbable Zn intake from the usual diet( 4 ), the strict procedures during collection and processing to avoid Zn contamination of plasma samples( 4 , 33 ) and the extensive characterisation of FA profiles in a relatively large number of children. However, this study had limitations. This study was a secondary analysis of a larger study that was not designed or powered to assess changes in FA profiles( 33 ), this may have introduced bias into our analysis. Furthermore, although PZn concentrations in the Zn+filter group were significantly higher than in the filter group during the intervention, PZn concentrations did not increase in the Zn+filter group. This was attributed to a seasonal increase in infections occurring towards the end of the intervention which cause PZn depression( 33 ), thus reducing the availability of circulating Zn. This may have reduced our ability to differentiate group differences in FA profile during the intervention. It is possible that a more efficacious Zn fortification intervention (or Zn supplementation with higher doses) might have resulted in a more pronounced group differences in the FA profile. Also, although baseline FA composition was similar in the two groups suggesting no large differences in dietary fat intake, potential group differences in dietary fat intake during the intervention could have biased our results, as dietary fat intake is correlated with plasma FA composition( 52 ). finally, our study was limited by the use of plasma for the measurement of FA status( 53 ). The use of erythrocyte membranes or adipose tissue would have been more sensitive to reflect long-term changes in LCPUFA than plasma, as the latter is used as a short term FA biomarker( 54 56 ).

In conclusion, our findings suggest PZn concentrations are positively associated with the activity of FA desaturases (particularly Δ-6-desaturase), and/or elongases in children. Furthermore, Zn fortification for 20 weeks prevented reduction of nervonic acid in plasma total phospholipids, especially in children who were Zn deficient. This was unexpected as previous studies have reported that Zn deficiency mainly affects n-6 FA composition. We provide evidence for the interaction between PZn and plasma total phospholipid FA composition in humans and our findings support the value of the LA:DGLA or DGLA:LA ratio as a potential biomarker of Zn status. Future research should investigate the impact of changes in Zn status on these ratios, and on nervonic acid concentrations, preferably through controlled Zn supplementation studies in Zn-deficient children and adults.


The authors thank the Beninese children who participated in the main study and their parents. Johanna van Wyk is greatly acknowledged for her technical assistance during FA analysis. The authors thank Natasa Mihajlovic for the PZn analysis and Jürgen Erhardt for the CRP and AGP analysis. The authors thank Vestergaard Switzerland for proving the filters and funding the main study, Nestlé Nutrition Institute for funding T. C.’s fellowship at ETH Zurich, the Nestlé Nutrition Institute Africa for the research grant, as well as the South African National Research Fund for awarding a scholarship to T. C.

Supported by Vestergaard Switzerland and Nestlé Nutrition Institute Africa.

T. C. involved in conceptualisation, performed data analyses and interpretation and wrote the manuscript; L. M. involved in data analysis and interpretation, revision of manuscript; J. B. conceptualised secondary study, involved in revision of manuscript; P. J. v. J. supervised FA analyses, cleaned FA data set, data interpretation and revision of manuscript; V. G. and D. M. conceptualised and conducted parent study research and revision of manuscript; C. M. S. data interpretation and revision of manuscript; M. B. Z. initiated and conceptualised the parent study. All authors read and approved the final version of the paper. None of the study funders had any involvement in the study design, analysis or interpretation of data; the writing of the report; or the decision to submit the article for publication.

None of the authors has any conflicts of interest to declare.

Supplementary Material

For supplementary material/s referred to in this article, please visit


1. Calder, PC (2015) Functional roles of fatty acids and their effects on human health. J Parenter Enteral Nutr 39, 18S32S.
2. De Mel, D & Suphioglu, C (2014) Fishy business: effect of omega-3 fatty acids on zinc transporters and free zinc availability in human neuronal cells. Nutrients 6, 32453258.
3. Vallee, BL & Falchuk, KH (1993) The biochemical basis of zinc physiology. Physiol Rev 73, 79118.
4. Galetti, V, Mitchikpè, CES, Kujinga, P, et al. (2016) Rural beninese children are at risk of zinc deficiency according to stunting prevalence and plasma zinc concentration but not dietary zinc intakes. J Nutr 146, 114123.
5. Wuehler, SE, Peerson, JM & Brown, KH (2005) Use of national food balance data to estimate the adequacy of zinc in national food supplies: methodology and regional estimates. Public Health Nutr 8, 812819.
6. Wessells, KR & Brown, KH (2012) Estimating the global prevalence of zinc deficiency: results based on zinc availability in national food supplies and the prevalence of stunting. PLOS ONE 7, e50568.
7. Prasad, AS (2013) Biochemistry of Zinc, vol. 11. New York: Springer Science & Business Media.
8. Cunnane, SC (1988) Evidence that adverse effects of zinc deficiency on essential fatty acid composition in rats are independent of food intake. Br J Nutr 59, 273278.
9. Nakamura, MT & Nara, TY (2004) Structure, function, and dietary regulation of delta6, delta5, and delta9 desaturases. Annu Rev Nutr 24, 345376.
10. Bettger, WJ, Reeves, PG, Moscatelli, EA, et al. (1979) Interaction of zinc and essential fatty acids in the rat. J Nutr 109, 480488.
11. Clejan, S, Castro-Magana, M, Collipp, PJ, et al. (1982) Effects of zinc deficiency and castration on fatty acid composition and desaturation in rats. Lipids 17, 129135.
12. Kudo, N, Nakagawa, Y & Waku, K (1990) Effects of zinc deficiency on the fatty acid composition and metabolism in rats fed a fat-free diet. Biol Trace Elem Res 24, 4960.
13. Eder, K & Kirchgessner, M (1994) Dietary fat influences the effect of zinc deficiency on live. J Nutr 124, 1917.
14. Kirchgessher, K & Manfred, EA (1994) Dietary fat influences the effect of zinc deficiency on liver lipids and fatty acids in rats force-fed equal quantities of diet. J Nutr 124, 19171926.
15. Ayala, S & Brenner, RR (1982) Essential fatty acid status in zinc deficiency. Effect on lipid and fatty acid composition, desaturation activity and structure of microsomal membranes of rat liver and testes. Acta physiol Lat Am 33, 193204.
16. Cunnane, S, Horrobin, D & Manku, M (1984) Essential fatty acids in tissue phospholipids and triglycerides of the zinc-deficient rat. Exp Biol Med 177, 441446.
17. Faure, P, Durand, P, Blache, D, et al. (1995) Influence of a long-term zinc-deficient diet on rat platelet function and fatty acid composition. Biometals 8, 8085.
18. Kramer, TR, Briske-Anderson, M, Johnson, SB, et al. (1986) Polyunsaturated fatty acid patterns in lymphoid and nonlymphoid tissues of zinc deficient and pair-fed rats. Nutr Res 6, 10631074.
19. Dib, A, Clavel, J-P & Carreau, J-P (1989) Effects of gamma-linolenic acid supplementation on lipid composition of liver microsomal membranes. J Clin Biochem Nutr 6, 95102.
20. Wauben, IP, Xing, H-C & Wainwright, PE (1999) Neonatal dietary zinc deficiency in artificially reared rat pups retards behavioral development and interacts with essential fatty acid deficiency to alter liver and brain fatty acid composition. J Nutr 129, 17731781.
21. Knez, M, Stangoulis, JC, Zec, M, et al. (2016) An initial evaluation of newly proposed biomarker of zinc status in humans-linoleic acid: dihomo-γ-linolenic acid (LA: DGLA) ratio. Clin Nutr ESPEN 15, 8592.
22. Gimenez, MS, Oliveros, LB & Gomez, NN (2011) Nutritional deficiencies and phospholipid metabolism. Int J Mol Sci 12, 24082433.
23. King, JC, Brown, KH, Gibson, RS, et al. (2016) Biomarkers of nutrition for development (BOND) – zinc review. J Nutr 146, 858S885S.
24. Wessells, KR, Jorgensen, JM, Hess, SY, et al. (2010) Plasma zinc concentration responds rapidly to the initiation and discontinuation of short-term zinc supplementation in healthy men. J Nutr 140, 21282133.
25. Lowe, NM, Fekete, K & Decsi, T (2009) Methods of assessment of zinc status in humans: a systematic review. Am J Clin Nutr 6, 27230G.
26. Reed, S, Qin, X, Ran-Ressler, R, et al. (2014) Dietary zinc deficiency affects blood linoleic acid: dihomo-gamma-linolenic acid (LA:DGLA) ratio; a sensitive physiological marker of zinc status in vivo (Gallus gallus). Nutrients 6, 11641180.
27. Brown, K & Hotz, C (2004) International Zinc Nutrition Consultative Group (IZiNCG) technical document #1: Assessment of the risk of deficiency in populations and options for its control. Food Nutr Bull 25, S91S204.
28. Das, JK, Kumar, R, Salam, RA, et al. (2013) Systematic review of zinc fortification trials. Ann Nutr Metab 62, 4456.
29. Tripathi, B & Platel, K (2010) Finger millet (Eleucine coracana) flour as a vehicle for fortification with zinc. J Trace Elem Med Biol 24, 4651.
30. Moretti, D, Biebinger, R, Bruins, MJ, et al. (2014) Bioavailability of iron, zinc, folic acid, and vitamin A from fortified maize. Ann N Y Acad Sci 1312, 5465.
31. Sian, L, Hambidge, KM, Westcott, JL, et al. (1993) Influence of a meal and incremental doses of zinc on changes in zinc absorption. Am J Clin Nutr 58, 533536.
32. Polaki, H & Yarla, NS (2014) Water as a new vehicle for nutrition. J Nutr Food Sci 4, 1.
33. Galetti, V, Kujinga, P, Mitchikpè, CES, et al. (2015) Efficacy of highly bioavailable zinc from fortified water: a randomized controlled trial in rural Beninese children. Am J Clin Nutr 102, 12381248.
34. Faul, F, Erdfelder, E, Buchner, A, et al. (2009) Statistical power analyses using G* Power 3.1: Tests for correlation and regression analyses. Behav Res Methods 41, 11491160.
35. World Health Organization & Food and Agriculture Organization (2004) Vitamin and Mineral Requirements in Human Nutrition. Geneva: WHO/FAO.
36. Miller, LV, Krebs, NF & Hambidge, KM (2007) A mathematical model of zinc absorption in humans as a function of dietary zinc and phytate. J Nutr 137, 135141.
37. International Zinc Nutrition Consultative Group (IZiNCG) (2007) Assessing population zinc status with serum zinc concentration. IZiNCG Technical Brief 2, 14.
38. Erhardt, JG, Estes, JE, Pfeiffer, CM, et al. (2004) Combined measurement of ferritin, soluble transferrin receptor, retinol binding protein, and C-reactive protein by an inexpensive, sensitive, and simple sandwich enzyme-linked immunosorbent assay technique. J Nutr 134, 31273132.
39. Thurnham, DI, Mburu, AS, Mwaniki, DL, et al. (2005) Micronutrients in childhood and the influence of subclinical inflammation. Proc Nutr Soc 64, 502509.
40. Hess, SY, Peerson, JM, King, JC, et al. (2007) Use of serum zinc concentration as an indicator of population zinc status. Food Nutr Bull 28, S403S429.
41. Ghosh, AdBB, Darnton-Hill, I, Davidsson, L, et al. (2007) Report of a WHO/UNICEF/IAEA/IZiNCG interagency meeting on zinc status indicators held in IAEA headquarters, Vienna, 9 December 2005. Food Nutrition Bulletin 28, pp. 399–405.
42. World Health Organization (2001) Iron Deficiency Anaemia: Assessment, Prevention and Control: A Guide for Programme Managers. Geneva: World Health Organization.
43. Folch, J, Lees, M & Sloane-Stanley, G (1957) A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 226, 497509.
44. Hon, G, Hassan, M, van Rensburg, SJ, et al. (2009) Immune cell membrane fatty acids and inflammatory marker, C-reactive protein, in patients with multiple sclerosis. Br J Nutr 102, 13341340.
45. Fan, Y-Y & Chapkin, RS (1998) Importance of dietary γ-linolenic acid in human health and nutrition. J Nutr 128, 14111414.
46. Warensjö, E, Rosell, M, Hellenius, M-L, et al. (2009) Associations between estimated fatty acid desaturase activities in serum lipids and adipose tissue in humans: links to obesity and insulin resistance. Lipid Health Dis 8, 37.
47. Horrobin, DF (1981) Loss of delta-6-desaturase activity as a key factor in aging. Med Hypotheses 7, 12111220.
48. Tang, C, Cho, HP, Nakamura, MT, et al. (2003) Regulation of human Δ-6 desaturase gene transcription identification of a functional direct repeat-1 element. J Lipid Res 44, 686695.
49. Cunnane, SC (1988) Zinc: Clinical and Biochemical Significance. Clifton, NJ: CRC Press Inc.
50. Jump, DB (2009) Mammalian fatty acid elongases. Methods Mol Biol 579, 375389.
51. El-Agamy, EI (2009) Bioactive components in camel milk. In Bioactive Components in Milk and Dairy Products, vol. 107, pp. 159–192 [YW Park, editor]. Ames, IA: Wiley-Blackwell.
52. Hodge, AM, Simpson, JA, Gibson, RA, et al. (2007) Plasma phospholipid fatty acid composition as a biomarker of habitual dietary fat intake in an ethnically diverse cohort. Nutr Metab Cardiovasc Dis 17, 415426.
53. Smit, EN, Muskiet, FA & Boersma, ER (2004) The possible role of essential fatty acids in the pathophysiology of malnutrition: a review. Prostagland Leukot Essent Fatty Acids 71, 241250.
54. Baylin, A & Campos, H (2006) The use of fatty acid biomarkers to reflect dietary intake. Curr Opin Lipidol 17, 2227.
55. Baylin, A, Kim, MK, Donovan-Palmer, A, et al. (2005) Fasting whole blood as a biomarker of essential fatty acid intake in epidemiologic studies: comparison with adipose tissue and plasma. Am J Epidemiol 162, 373381.
56. Katan, MB, Deslypere, J, Van Birgelen, A, et al. (1997) Kinetics of the incorporation of dietary fatty acids into serum cholesteryl esters, erythrocyte membranes, and adipose tissue: an 18-month controlled study. J Lipid Res 38, 20122022.