{"paper_id":"7c380937-be6c-4649-935c-3859fa37ec50","body_text":"Page 1 of 24 \n \nTitle: The effect of non-genetic determinants of human milk oligosaccharide profiles in milk \nof Ugandan mothers \nRunning Title: Non-genetic maternal and infant factors and HMO profiles in Ugandan \nmothers \nAuthors: Tonny Jimmy Owalla 1, Victor Irungu  Mwangi, Sara Moukarzel 3, 4 , Emmanuel \nOkurut1, Chloe Yonemitsu3, Lars Bode3, 4, Thomas G. Egwang1 \nAuthor affiliations: 1Uganda Human Milk and Lactation Center, Med Biotech Laboratories, \nKampala, Uganda;  2Programa de Pós -Graduaçặo em Medicina  Tropical, Universidade do \nEstado do Amazonas (UAE), Manaus , AM, Brazil ; 3Department of Pediatrics, Division of \nNeonatology, University of California San Diego, La Jolla, California, United States of \nAmerica; 4Larsson-Rosenquist Foundation Mother-Milk-Infant Center of Research Excellence \n(MOMI CORE), University of California San Diego, La Jolla, California, United States of \nAmerica. \nAbstract   \nBackground: Human milk oligosaccharide s (HMOs) protect against infection and promote \ngrowth and cognitive development in breastfeeding children. Non -genetic factors which \ninfluence HMO composition in breastfeeding mothers in rural Africa have not been \ninvestigated. \nObjective: We undertook a cross-sectional study to determine the association between HMO \nprofiles and non-genetic maternal factors and children’s sex in Ugandan mother-children pairs. \nMethod: Human milk was  collected from 127 breastfeeding mothers by manual expression. \nHMO analysis was by high performance liquid chromatography. The proportion of each HMO \nper total HMO concentration was calculated. Spearman’s correlation and Mann-Whitney U test \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \nNOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.\n\nPage 2 of 24 \n \nwere used to assess the relationship between individual HMOs and maternal factors and infant \nsex. \nResult: Nineteen HMOs were assayed. The prevalence of secretor and non-secretor status, \nbased on the proportion of mothers with high milk concentrations of 2’FL and LNFP 1 , was \n80.3 % and 19.7 %, respectively. In secretor mothers, 2’FL, DFLac and LNFP I constituted > \n57 % while in non-secretor mothers LNT and LNFPII constituted 46.9 % of the measured total \nHMOs. The median 3’SL concentration in milk of all mothers of male children was \nsignificantly higher than th at in all mothers of female children.  The me dian DFLac \nconcentration in all mothers was significantly higher in multiparous mothers compared to \nprimiparous mothers. Higher FDSLNH and lower LNH concentrations were observed in \noverweight secretor and non-secretor mothers, respectively.  Median concentrations of LNFP \nI and DSLNT were significantly higher in all mothers < 18 years old compared all mothers > \n18 years old. Concentrations of specific HMOs increased, decreased, or remained unchanged \nwith increasing lactation duration in secretor and non-secretor mothers. \nConclusions: Specific HMOs were associated with infant sex and maternal age, parity and \npost-partum BMI in Ugandan mothers but were different from those reported in other \npopulations.  \nKey words: Human milk oligosaccharides, breastfeeding mothers  and children, non -genetic \nfactors, secretor mothers \nIntroduction \nHuman milk oligosaccharides (HMOs), comprises of over 150 non -digestible unconjugated \nglycans and constitutes the third most abundant solid component in human milk-besides lipids \nand fats (1). HMOs play a critical role in protecting infants against infecti on through \nmodulation of gut microbiota and the immune system (2 -6). These HMO effects are \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 3 of 24 \n \ncomposition- and structure-specific (3, 6). The HMOs basic structure consist of a lactose core \nwhich is modified by glycosyltransferases to form 4 main types of HMOs , namely,  non-\nfucosylated neutral HMO s containing  N-acetylglucosamine at the terminal end,  neutral \nfucosylated HMO containing fucose at the terminal position, sialylated HMOs containing sialic \nacid, and HMOs containing both fucose and sialic acid  (1,7,8).  HMO composition and \nconcentration are influenced by maternal genetics. Breastfeeding mothers with an active gene \nencoding the α (1,2)-fucosyltransferase (FUT2) enzyme have milk with characteristically high \nlevels of 2'-FL and Lacto-N-fucopentaose I (LNFPI) and are designated secretor mothers (6). \nWomen who do not express FUT2 have milk lacking these HMOs and are designated non-\nsecretors. The prevalence of secretor mothers vary by geographic from a high prevalence in \nSouth America to the lowest prevalence in African countries ranging from 63% in South Africa \nto 75 % in Malawi and Kenya (9).  \nThere is evidence that environmental, non-genetic maternal and infants' factors are associated \nwith differential HMO concentration in Asian, European, Canadian and American populations \n(9,10, 1 1). However, there is a  dearth of information about  the non -genetic factors that \ninfluence HMO composition and concentrations in breastfeeding mothers in African rural \nsettings. Our study investigated for the first time  non-genetic maternal and infant factors that \ninfluence HMO composition and concentration in Ugandan breastfeeding mother-child pairs \nin rural northeastern Uganda.  \nMethods \nStudy population and design  \nThe study population was part of a cohort who were participating in a prospective longitudinal \nstudy investigating the effectiveness of a malaria vector control intervention in Abwokodia \nParish, Katakwi district in Northeastern Uganda. Details of the study  site have been reported \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 4 of 24 \n \nelsewhere (12). The breast milk sub -study was a clinic based cross -sectional survey of \nbreastfeeding mother-infant pairs. All breastfeeding infants less than two years old and their \nrespective breastfeeding mothers were considered eligible for the study up on consent by one \nor both parents as applicable. The breastfeeding infants were part of 400 children under 5 years \nold who were being followed up for malaria prevalence and incidence.  All the identified 127 \nmother-infant pairs were included in the study (Supplementary Figure 1). \nParticipants recruitment and anthropometric measurements \nThe study was conducted in March 2018. Mothers of all breastfeeding infants participating in \nthe longitudinal study were mobilized by a social scientist with the help of village health teams \n(VHTs). Mother-infant pairs visited the study clinic at St. Anne Health Center III, Katakwi  \ndistrict for screening and enrolment. The objectives of the study were explained to the mothers \nand informed consent obtained. Human milk samples were collected from consenting mothers. \nThe demographic characteristics (infant sex and maternal age, parity, BMI, lactation duration), \nand other anthropometric data for mother-child pairs were recorded on a standardized \nquestionnaire.  \nBreast milk sampling and treatment \nA single 5-mL human milk sample was collected from each of one hundred and twenty-seven \nlactating mothers using manual expression with the help of a senior midwife who is also an \nexperienced lactation nurse. Milk samples were collected into sterile 50 -mL Falcon tubes and \nimmediately frozen in dry ice (approximately -70°C). Milk samples were transported in dry ice \nto Med Biotech Laboratories headquarter in Kampala before being air -freighted in dry ice to \nthe University of California San Diego for HMO high performance liquid chromatography \n(HPLC) analysis.  \nHMO Extraction, Analysis, and Secretor Status Determination \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 5 of 24 \n \nHMO (the 19 well characterized and most abundant) analysis was performed at the University \nof California, San Diego, as previously described (13), by using HPLC after fluorescent \nderivatization. Raffinose was added to each milk sample as an internal standa rd for absolute \nquantification. The total concentration of HMOs was calculated as the sum of the specific \noligosaccharides detected. The proportion of each HMO per total HMO concentration was \ncalculated. Maternal phenotypic secretor status was determined b y the relative abundance \n(secretor) or near absence (non-secretor) of the (α 1-2) linked Fuc (2′(-fucosyllactose (2′FL) in \nthe respective milk samples.  \nStatistical Analyses \nStatistical analyses were performed using GraphPad Prism version 9.0.2 and STATA version \n15. The Shapiro Wilk test were used to evaluate the variables' distribution including the \noligosaccharide concentrations. Normally distributed data were compared using  Student’s t-\ntest, the t -test for paired samples or one -way analysis of variance for groups, while non -\nparametric comparisons were made using the Mann -Whitney U test for paired samples or \nKruskal-Wallis test for groups. The one-way ANOVA and Kruskal-Wallis tests were followed \nup by the appropriate post hoc multiple comparison test. Chi-square or Fisher exact tests were \nalso performed to compare categorical variables. Exploratory analysis were performed to \ncompare median oligosaccharide concentrations according to infant sex, maternal age (years), \ncurrent BMI (kg/m 2), maternal parity (primiparous/multiparous), and duration of lactation \n(weeks). The results were presented as box -plot graphs. Correlations of the oligosaccharide \nconcentrations with other infant or maternal variables were assessed using the Spearman rank \ntest with 95% confidence interval. Significance was defined as P values of less than .05. \nEthical consideration  \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 6 of 24 \n \nThe Uganda National Council for Science and Technology and the Research Ethics Committee \nof the Vector Division, Ministry of Health, approved the original HD4MC study. All mothers \nwho donated milk samples and brought infants to the malaria clinic signed or  thumb-printed \nan informed consent form. \n Results \nCharacteristics of the Study Population \nThe study population comprised of 127 mothers and their breastfeeding children in Katakwi \nDistrict in northeastern Uganda. The mothers had a mean age of 26.6 years (range 15-46) with \na post -partum mean BMI of 23.3 kg/m2 ( Table 1 ). Most of the mothers had had several \nchildren (> 78 %) and had breastfed for a mean of 41.3 weeks (range 3-103). The children had \na mean age of 46.5 weeks (range 3-110) and 52.6 % were females. The mean birth weight was \n3.1 kg (range 1-4.8). The mean children’s heights/lengths and weights at the time of sampling \nwere 70.5 cm (range 50 -98) and 8.80 kg (5 -16), respectively. The mean mid -upper arm \ncircumference for the children was 14.8 cm (range 12.1 -18.7). The mean hemoglobin level in \nchildren was 10.9 g/dL (range 8-11.5). The children had a mean temperature of 36.5 °C (range \n35.4-39.4) and the majority (97.7 %) had normal temperature at the time of sampling. \nHMO profiles and prevalence of secretor and non-secretor mothers  \nMothers with an active secretor (Se) gene that encodes α (1,2)-fucosyltransferase are classified \nas “secretors” and their milk contains significantly higher concentrations of α(1,2)-fucosylated \nHMOs such as 2′FL, DFLac and LNFP I ( 6). The prevalence of Ugandan secretor and non-\nsecretor mothers, based on the proportion of mothers with relatively high milk concentrations \nof 2’FL and LNFP 1 was 80.3 and 19.7 %, respectively (Table 1). In secretor mothers, 2’FL, \nDFLac and LNFP I constituted > 57 %  of the measured total HMOs ( Supplementary Table \n1). By contrast, 2’FL, DFLac and LNFP I constituted only 1.9 % of the measured HMOs in \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 7 of 24 \n \nnon-secretor mothers in whom LNFP II and LNT alone constituted 46.9 %  of the measured \ntotal HMOs.  \nInfant sex is associated with differential concentrations of 3’SL in all mothers  \nThe median 3’SL concentration in the milk of all mothers combined of male infants was \nsignificantly higher than those in milk of all mothers combined of female infants (701.8 (454.5-\n1059.8) nmol/mL versus 516.6 (309.8 -853.8) nmol/mL; P =0.035) by the Mann -Whitney U \ntest (Supplementary Table 2). \nLactation duration is associated with differential HMO composition  \nBreastfeeding for 24 months or more is common in the study population and it was of interest \nto investigate HMO profiles beyond 12 months post-partum. We used Spearman’s correlation \nanalysis to assess the relationship between HMO concentrations and lactation duration defined \nas the time elapsed from the time of first breastfeeding at birth to the time post-partum of breast \nmilk sampling . Five and 10 out of the 19 HMOs had a significant positive and negative \ncorrelation, respectively,  with lactation duration  depending on mothers ’ secretor status \n(Supplementary Table 3 ). The results of Mann -Whitney U tests comparing median HMO \nconcentrations between mothers at different lactation duration s closely paralleled those of \ncorrelation analyses. The concentrations of 3FL, 3’SL and DFLac were significantly higher in \nall mothers combined at > 6 months lactation compared to  ≤ 6 months lactation  (Table 2). By \ncontrast, the concentrations of LNnT, 6’SL, LSTc, LNH, FLNH, DFLNH, and DSLNH were \nsignificantly lower in all mothers at > 6 months lactation duration. When the mothers’ secretor \nstatus was considered, LNFPI, LNnT, 6’SL, LSTc, FLNH, DFLNH, and DSLNH \nconcentrations were significantly lower and 3 FL, 3’SL and DFLac concentrations were \nsignificantly higher in secretor mothers at > 6 months lactation ( Figure 1 ). Similar HMO \nprofiles were observed in non -secretor mothers but LNH and FDSLNH concentrations were \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 8 of 24 \n \nsignificantly lower while LNFP I, FLNH and LNnT concentrations were unchanged in mothers \nat > 6 months lactation ( Supplementary Figure 2). To see if the dynamics of HMO \ncomposition and concentrations persisted beyond 12 months (52 weeks), we  compared HMO \nconcentrations in groups of all mothers combined at lactation durations of  ≤ 12 (N=21), 12-24 \n(N=14), 25-48 (N=40), 49 -72 (36) and > 73 (N=16) weeks, respectively. The same HMOs  \nincreased, decreas ed, or remained unchanged , respectively,  throughout lactation beyond 52 \nweeks (Figure 2). It was not possible to assess HMO profiles in similar groups of secretor and \nnon-secretor mothers due to the small sample sizes. \nMaternal Age is associated with differential concentrations of 2’FL, DFLac and DSLNT  \n2’FL and DFLac concentrations had a significant positive correlation with maternal age only \nin non -secretor mothers  (Supplementary Table 4 ). LNnT had a significant negative \ncorrelation with maternal age in all mothers and secretor mothers but not non-secretor mothers. \nInterestingly, DSLNT concentrations had a significant negative correlation with maternal age \nin secretor mothers but a positive correlation in non -secretor mothers. Median concentrations \nof LNFP I and DSLNT were significantly higher in all mothers < 18 years old (N=4) than in \nall mothers > 18 years old (N= 123) by Mann -Whitney U test  (Supplementary Table 5 ). It \nwas not possible to compare median HMO concentrations between the two age groups by \nsecretor status and lactation duration because of the small sample size for the younger mothers.  \nMaternal Parity is associated with differential concentrations of DFLac in all mothers \nTo confirm if maternal parity affected HMO concentrations, we defined parity as primiparous \nmothers (N= 27) who have had one child versus multiparous mothers (N= 100) who have had \nmore than one child. The median DFLac concentration was significantly higher in multiparous \nall combined mothers compared to primiparous all combined mothers by the Mann-Whitney U \ntest (Supplementary Table 6). It was not possible to compare median HMO concentrations \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 9 of 24 \n \nbetween both parity groups by secretor status and lactation stage because of the small sample \nsize for primiparous mothers. \nMaternal BMI is associated with differential concentrations of FDSLNH and LNH  \nFDSLNH and LNH concentrations positively and negatively correlated with increasing \nmaternal postpartum BMI at the time of sampling in secretor and non -secretor mothers, \nrespectively (Supplementary Table 7). None of the other HMOs had any correlation with \nmaternal BMI. When secretor mothers were classified into normal weight (BMI= 18.5 -24.9, \nN= 81) and overweight (BMI = ≥25, N= 19) categories, median FDSLNH and LNH \nconcentrations were significantly higher a nd lower, respectively, in overweight mothers by \ncomparison with normal weight mothers by the Mann -Whitney U test in secretor and non -\nsecretor mothers, respectively (Supplementary Tables 8 and 9).  \nDiscussion \nNon-genetic factors including infant sex and lactation duration which influence HMO \ncomposition have been investigated in breastfeeding mothers from Brazil, China, Europe, and \nUSA (9,11,15,16,17,19, 20,21,22).  The majority of studies of the effect of lactation duration \non HMO composition investigated limited lactation durations of 3 -6 months; only one \nlongitudinal study followed up breastfeeding mothers for 24 months (35). Despite the \nburgeoning literature on HMOs in breastfeeding mothers in African countries (9,33,35,36, 37, \n38), there is a knowledge gap about non-genetic factors which influence HMO composition in \nmothers in rural and urban African settings. The aim of this study was to determine HMO \nprofiles in 127 rural Ugandan mothers and assess their relationship with variou s maternal and \ninfant factors. HMO profiles in Ugandan mothers demonstrated a high prevalence of secretors \n(80.3 %) with 2’FL and LNFP I predominating in secretors while LNFP II and LNT \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 10 of 24 \n \npredominated in non -secretors (prevalence 19.7 %). Lower secretor prevalence has been \nreported in mothers from some African populations (9).  \nInfant sex and maternal parity, age, post -partum BMI and lactation duration were associated \nwith specific HMO profiles in Ugandan mothers. First, higher concentrations of 3’SL were \nobserved in mothers of male but not female children in this study. By contrast, in Brazilian and \nChinese mothers significantly higher concentrations of LNT were associated with having male \ninfants (15, 16). The fact that infant/child male sex is significantly associated with high \nconcentrations of LNT and 3’SL in different studie s involving different ethnicities suggests \nimportant differential roles for these HMOs in the health and development of male and female \ninfants in the respective populations. Second, DFLac was significantly higher in milk of \nmultiparous Ugandan mothers by comparison with primiparous mothers. Maternal parity has \nbeen associated with increased or decreased concentrations of LNT, LNnT and 3FL in \nEuropean, Chinese, and Brazilian mothers (11, 15, 17). Third, LNFPI and DSLNT \nconcentrations were significantly higher in Ugandan mothers under 18 years old by comparison \nwith older mothers. In a single study involving a limited number of mothers of different \nethnicities, maternal age was associated with increased or increased concentrations of specific \nHMOs which were different from those identified in this study (9). Finally, postpartum BMI \nwas associated with higher and lower concentrations of FDSLNH and LNH, respectively, in \nUgandan mothers. The relationship between pre -partum BMI and HMO composition is \ncontroversial (9, 15, 18, 19, 20,21,22). In this population of Ugandan mothers, the above \nmaternal and infant factors were associated with specific HMOs which are different from those \nreported in other studies (9, 11,15,16,17,18 -22); this observation underscores the im portance \nof population context -specific HMO data. The mechanisms underlying the associations \nbetween specific HMOs with maternal factors and infant sex are not known (10).  \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 11 of 24 \n \nLactation duration was an important non -genetic maternal determinant of HMO profiles in \nUgandan mothers. The fact that concentrations of the same HMOs decreased, increased, or \nremained unchanged with increasing lactation duration beyond 52  weeks in Ugandan mothers \nprobably highlights their importance in the survival and development of breastfeeding infants \nand young children especially in resource-poor rural and urban Uganda where infant morbidity, \nmortality and malnutrition remains disproportion ately high (23, 24, 25).  In this study \npopulation, two HMOs, 2’FL and LNFPI, are associated with protection against malaria in \nbreastfeeding children (Mwangi et al under review). In prior literature,  2’FL has been \nimplicated in multiple functions including protection against infections, reduction of morbidity \nand use of antipyretics and antibiotics (26, 27); stimulation of brain development (28, 29); \nweight gain (19); and improvement of cognitive functions (30). 2’FL also promotes the \nselective growth of bifidobacteria and therefore probably influences the composition of gut \nmicrobiota (31) and microbiota-induced immune functions (32) in breastfeeding infants. 2’FL \nand 3’FL, which also increased during lactation in Ugandan mothers, have been implicated in \nthe prevention of mortality in uninfected Zambian infants born to HIV -positive mothers (33). \nThe exact mechanisms by which HMOs reduce morbidity, prevent mortality and promote \ngrowth and development are unknown but could be associated with their reported anti-infective \nactivities and effects on immunity (31,32, 34). The mechanisms underlying differences in \nHMO profiles during lactation remain speculative (10). There is need for research on the \nregulation of the enzymes involved in HMO biosynthesis during la ctation in Africa, a region \nfraught with the constant threat of malnutrition and food insecurity  which affect HMO \nconcentrations (39). \nThis study has several limitations. First, cause -effect relationships between non -genetic \nmaternal and infant factors cannot be inferred due to the cross -sectional nature of this study. \nSecond, due to the small sample sizes, we were unable to adjust for multiple testing in statistical \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 12 of 24 \n \nanalyses nor make direct comparisons between secretor and non -secretor mothers. This \nlimitation will be addressed in a planned powered longitudinal study of breastfeeding mother-\nchild pairs. Third, milk samples were collected at a single time point, rather than longitudinally \nover several time points during lactation to monitor postpartum changes in HMO profiles over \ntime.  However, this limitation is mitigated by the fact that our cross -sectional milk sampling \nfrom mothers representing a wide range of lact ation stages produced HMO profiles which \nclosely mirrored those reported by a longitudinal study of US mothers followed over 24 months \npost-partum (35). Finally, the HMO profiles in mothers from north -eastern Uganda, who are \npredominantly of the Nilo-Hamitic Atesot tribe, may not be generalizable to mothers in other \nUgandan regions with different environments, diets and tribes.  \nIn conclusion, our studies of non-genetic determinants of HMO composition in rural Ugandan \nmothers identified specific HMOs which were associated with infant sex and maternal age, \nparity and BMI but were different from those reported in other populations, thereby \nunderscoring the importance of generating population context -specific HMO data. Our study \nrevealed HMO profiles in Ugandan mothers  with prolonged lactation durations beyond 52 \nweeks which were remarkably similar  to published HMO profiles from mothers with shorter \nlactation durations. The conservation of these HMO profiles in different populations, regardless \nof environments, diets and ethnicities,  probably underscores their importance in infant/child \ngrowth and survival throughout lactation. \n Acknowledgments: We are very grateful to the mother -infant pairs for their generosity and \nparticipation.  We would like to thank the clinical and laboratory staff of Med Biotech \nLaboratories at St Anne HC III Usuk, Katakwi, Uganda. \nFunding: Thomas Egwang received funding from the Global Innovation Fund and Grand \nChallenges Canada. Lars Bode is UC San Diego Chair of Collaborative Human Milk Research \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 13 of 24 \n \nendowed by the Family Larsson -Rosenquist Foundation (FLRF), Switzerland. The funders \nplayed no role in the study design, data collection and analysis nor in the preparation of the \nmanuscript and decision to publish this paper.   \nConflict of interest and funding disclosure: The authors have no conflicts of interest to \ndisclose. \nAuthors’ contributions to manuscript: TJO and EO conducted the study including sample \ncollection; TJO drafted the manuscript; SM and CY performed the HMO laboratory analysis; \nVIM performed statistical analysis; TGE and LB conceived and directed the study. All authors \nread and approved the final manuscript. \nCorresponding author: Tonny Jimmy Owalla, P.O. Box 9364 Kampala-Uganda, \n+12068543851, owallatonny@gmail.com  \nTJO current address: Department of Global Health, University of Washington, Seattle, WA, \nUSA. \nData Sharing: Data described in the manuscript, code book, and analytic code will be made \navailable upon request from TGE. \nAbbreviations:   \n2’FL, 2’fucosyllactose \n3 FL, 3-fucosyllatose \n3’SL, 3’sialyllactose \n6’SL, 6’sialyllactose \nBMI, body mass index \nDFLac, difucosyllactose \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 14 of 24 \n \nDFLNH, Difucosyllacto-N-hexaose \nDFLNT, Difucosyllacto-N-tetrose  \nDSLNH, disialyllacto-N-tetraose  \nDSLNT, Disialyllacto-N-tetraose \nFDSLNH, Fucosyldisialyllacto-N-hexaose \nFLNH, Fucosyllacto-N-hexaose \nFuc HMO, HMO-bound fucose \nFUT2, α1,2-fucosyltransferase gene \nHMOs, human milk oligosaccharides \nLNFPI, lacto-N-fucopentaose I \nLNFPII, lacto-N-fucopentaose II \nLNFPIII, lacto-N-fucopentaose III \nLNH, Lacto-N-hexaose \nLNnT, Lacto-N-neotetraose \nLNT, Lacto-N-tetraose \nLSTb, Sialyllacto-N-tetraose b \nLSTc, sialyllacto-N-tetraose c \nSia HMO, HMO-bound sialic acid \n \n \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 15 of 24 \n \nReferences \n1. Bode L. The functional biology of human milk oligosaccharides. Early Hum Dev. \n2015;91(11):619-622. doi: 10.1016/j.earlhumdev.2015.09.001 \n2. Triantis V, Bode L and van Neerven RJJ (2018) Immunological Effects of Human \nMilk Oligosaccharides. Front. Pediatr. 6:190. doi: 10.3389/fped.2018.00190 \n3. Ayechu-Muruzabal V, van Stigt AH, Mank M, et al. Diversity of Human Milk \nOligosaccharides and Effects on Early Life Immune Development. Front Pediatr. \n2018; 6:239. Published 2018 Sep 10. doi:10.3389/fped.2018.00239 \n4. Ray C, Kerketta JA, Rao S, et al. Human Milk Oligosaccharides: The Journey Ahead. \nInt J Pediatr. 2019; 2019:2390240. Published 2019 Aug 4. \ndoi:10.1155/2019/2390240 \n5. Gao, X., Wu, D., Wen, Y., Gao, L., Liu, D., Zhong, R., Yang, C., Zhao, C., Antiviral \neffects of human milk oligosaccharides: A review, International Dairy Journal, https:// \ndoi.org/10.1016/j.idairyj.2020.104784 \n6. Bode L, Jantscher-Krenn E. Structure-function relationships of human milk \noligosaccharides. Adv Nutr. 2012;3(3):383S-91S. Published 2012 May 1. \ndoi:10.3945/an.111.001404 \n7. Urashima T, Asakuma S, Leo F, Fukuda K, Messer M, Oftedal OT. The \npredominance of type I oligosaccharides is a feature specific to human breast milk. \nAdv Nutr. 2012;3(3):473S-82S. Published 2012 May 1. doi:10.3945/an.111.001412 \n8. Hegar B, Wibowo Y, Basrowi RW, et al. The Role of Two Human Milk \nOligosaccharides, 2'-Fucosyllactose and Lacto-N-Neotetraose, in Infant Nutrition. \nPediatr Gastroenterol Hepatol Nutr. 2019;22(4):330-340. \ndoi:10.5223/pghn.2019.22.4.330 \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 16 of 24 \n \n9. McGuire MK, Meehan CL, McGuire MA, et al. What's normal? Oligosaccharide \nconcentrations and profiles in milk produced by healthy women vary geographically. \nAm J Clin Nutr. 2017;105(5):1086-1100. doi:10.3945/ajcn.116.139980 \n10. Han SM, Derraik JGB, Binia A, Sprenger N, Vickers MH, Cutfield WS. Maternal and \nInfant Factors Influencing Human Milk Oligosaccharide Composition: Beyond \nMaternal Genetics. J Nutr. 2021;151(6):1383-1393. doi:10.1093/jn/nxab028 \n11. Azad MB, Robertson B, Atakora F, et al. Human Milk Oligosaccharide \nConcentrations Are Associated with Multiple Fixed and Modifiable Maternal \nCharacteristics, Environmental Factors, and Feeding Practices. J Nutr. \n2018;148(11):1733-1742. doi:10.1093/jn/nxy175 \n12. Owalla TJ, Okurut E, Apungia G, et al. Using the Ultrasensitive Alere Plasmodium \nfalciparum Malaria Ag HRP-2™ Rapid Diagnostic Test in the Field and Clinic in \nNortheastern Uganda. Am J Trop Med Hyg. 2020;103(2):778-784. \ndoi:10.4269/ajtmh.19-0653 \n13. Berger PK, Hampson HE, Schmidt KA, Alderete TL, Furst A, Yonemitsu C, \nDemerath E, Goran MI, Fields DA, Bode L. Stability of Human-Milk Oligosaccharide \nConcentrations Over 1 Week of Lactation and Over 6 Hours Following a Standard \nMeal. J Nutr. 2023 Jan 14;152(12):2727-2733. doi: 10.1093/jn/nxac214. PMID: \n36111739; PMCID: PMC9839992. \n14. Ren X, Yan J, Bi Y, et al. Human Milk Oligosaccharides Are Associated with \nLactation Stage and Lewis Phenotype in a Chinese Population. Nutrients. \n2023;15(6):1408. Published 2023 Mar 15. doi:10.3390/nu15061408 \n15. Tonon KM, Miranda A, Abrão ACFV, de Morais MB, Morais TB. Validation and \napplication of a method for the simultaneous absolute quantification of 16 neutral and \nacidic human milk oligosaccharides by graphitized carbon liquid chromatography - \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 17 of 24 \n \nelectrospray ionization - mass spectrometry. Food Chem. 2019; 274:691-697. doi: \n10.1016/j.foodchem.2018.09.036 \n16. Wang M, Zhao Z, Zhao A, et al. Neutral Human Milk Oligosaccharides Are \nAssociated with Multiple Fixed and Modifiable Maternal and Infant Characteristics. \nNutrients. 2020;12(3):826. Published 2020 Mar 20. doi:10.3390/nu12030826 \n17. Samuel TM, Binia A, de Castro CA, et al. Impact of maternal characteristics on \nhuman milk oligosaccharide composition over the first 4 months of lactation in a \ncohort of healthy European mothers. Sci Rep. 2019;9(1):11767. Published 2019 Aug \n13. doi:10.1038/s41598-019-48337-4 \n18. Lagström, H., Rautava, S., Ollila, H., Kaljonen, A., Turta, O., Mäkelä, J., Yonemitsu, \nC., Gupta, J. and Bode, L., 2020. Associations between human milk oligosaccharides \nand growth in infancy and early childhood. The American journal of clinical nutrition, \n111(4), pp.769-778. \n19. Larsson MW, Lind MV, Laursen RP, et al. Human Milk Oligosaccharide \nComposition Is Associated with Excessive Weight Gain During Exclusive \nBreastfeeding-An Explorative Study [published correction appears in Front Pediatr. \n2020 Jan 22; 7:521]. Front Pediatr. 2019; 7:297. Published 2019 Jul 18. \ndoi:10.3389/fped.2019.00297 \n20. Elwakiel M, Hageman JA, Wang W, et al. Human Milk Oligosaccharides in \nColostrum and Mature Milk of Chinese Mothers: Lewis Positive Secretor Subgroups. \nJ Agric Food Chem. 2018;66(27):7036-7043. doi: 10.1021/acs.jafc.8b02021 \n21. Coppa GV, Gabrielli O, Pierani P, Catassi C, Carlucci A, Giorgi PL. Changes in \ncarbohydrate composition in human milk over 4 months of lactation. Pediatrics. \n1993;91(3):637-641. \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 18 of 24 \n \n22. Neville J, Pawlak R, Chang M, Furst A, Bode L, Perrin MT. A Cross-Sectional \nAssessment of Human Milk Oligosaccharide Composition of Vegan, Vegetarian, and \nNonvegetarian Mothers. Breastfeed Med. 2022;17(3):210-217. \ndoi:10.1089/bfm.2021.0259 \n23. Kananura RM, Tetui M, Mutebi A, et al. The neonatal mortality and its determinants \nin rural communities of Eastern Uganda. Reprod Health. 2016; 13:13. Published 2016 \nFeb 16. doi:10.1186/s12978-016-0119-y \n24. English L, Kumbakumba E, Larson CP, et al. Pediatric out-of-hospital deaths \nfollowing hospital discharge: a mixed-methods study. Afr Health Sci. 2016;16(4):883-\n891. doi:10.4314/ahs.v16i4.2 \n25. Amegbor PM, Yankey O, Sabel CE. Examining the Effect of Geographic Region of \nResidence on Childhood Malnutrition in Uganda. J Trop Pediatr. 2020;66(6):598-\n611. doi:10.1093/tropej/fmaa019 \n26. Puccio G, Alliet P, Cajozzo C, et al. Effects of Infant Formula with Human Milk \nOligosaccharides on Growth and Morbidity: A Randomized Multicenter Trial. J \nPediatr Gastroenterol Nutr. 2017;64(4):624-631. \ndoi:10.1097/MPG.0000000000001520 \n27. Sprenger N, Binia A, Austin S. Human Milk Oligosaccharides: Factors Affecting \nTheir Composition and Their Physiological Significance. Nestle Nutr Inst Workshop \nSer. 2019; 90:43-56. doi:10.1159/000490292 \n28. Oliveros E, Ramirez M, Vazquez E, Barranco A, Gruart A, Delgado-Garcia JM, Buck \nR, Rueda R, Martin MJ. Oral supplementation of 2′-fucosyllactose during lactation \nimproves memory and learning in rats. The Journal of nutritional biochemistry. 2016 \nMay 1; 31:20-7. \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 19 of 24 \n \n29. Wu KJ, Chen YH, Bae EK, Song Y, Min W, Yu SJ. Human milk oligosaccharide 2′-\nfucosyllactose reduces neurodegeneration in stroke brain. Translational Stroke \nResearch. 2020 Oct; 11:1001-11. \n30. Berger PK, Plows JF, Jones RB, et al. Human milk oligosaccharide 2'-fucosyllactose \nlinks feedings at 1 month to cognitive development at 24 months in infants of normal \nand overweight mothers. PLoS One. 2020;15(2):e0228323. Published 2020 Feb 12. \ndoi: 10.1371/journal.pone.0228323 \n31. Bajic D, Wiens F, Wintergerst E, Deyaert S, Baudot A, Van den Abbeele P. HMOs \nExert Marked Bifidogenic Effects on Children's Gut Microbiota Ex Vivo, Due to \nAge-Related Bifidobacterium Species Composition. Nutrients. 2023;15(7):1701. \nPublished 2023 Mar 30. doi:10.3390/nu15071701 \n32. Henrick BM, Rodriguez L, Lakshmikanth T, et al. Bifidobacteria-mediated immune \nsystem imprinting early in life. Cell. 2021;184(15):3884-3898.e11. doi: \n10.1016/j.cell.2021.05.030 \n33. Kuhn L, Kim HY, Hsiao L, et al. Oligosaccharide composition of breast milk \ninfluences survival of uninfected children born to HIV-infected mothers in Lusaka, \nZambia. J Nutr. 2015;145(1):66-72. doi:10.3945/jn.114.199794 \n34. Dinleyici M, Barbieur J, Dinleyici EC, Vandenplas Y. Functional effects of human \nmilk oligosaccharides (HMOs). Gut Microbes. 2023;15(1):2186115. \ndoi:10.1080/19490976.2023.2186115 \n35. Plows JF, Berger PK, Jones RB, et al. Longitudinal Changes in Human Milk \nOligosaccharides (HMOs) Over the Course of 24 Months of Lactation. J Nutr. \n2021;151(4):876-882. doi:10.1093/jn/nxaa427 \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 20 of 24 \n \n36. Davis JC, Lewis ZT, Krishnan S, et al. Growth and Morbidity of Gambian Infants are \nInfluenced by Maternal Milk Oligosaccharides and Infant Gut Microbiota. Sci Rep. \n2017;7:40466. doi:10.1038/srep40466 \n37. Paganini D, Uyoga MA, Kortman GAM, et al. Maternal Human Milk Oligosaccharide \nProfile Modulates the Impact of an Intervention with Iron and Galacto-\nOligosaccharides in Kenyan Infants. Nutrients. 2019;11(11):2596. \ndoi:10.3390/nu11112596 \n38. Jorgensen JM, Young R, Ashorn P, et al. Associations of human milk \noligosaccharides and bioactive proteins with infant growth and development among \nMalawian mother-infant dyads. Am J Clin Nutr. 2021;113(1):209-220. \ndoi:10.1093/ajcn/nqaa272 \n39. Moya-Alvarez V, Eussen SRBM, Mank M, et al. Human milk nutritional composition \nacross lactational stages in Central Africa. Front Nutr. 2022;9:1033005. \ndoi:10.3389/fnut.2022.1033005 \n \n \n \n \n \n \n \n \n \n \n \n \n \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 21 of 24 \n \nTABLE 1: Characteristics of the study population \nCharacteristic Mean Range Median   ( IQR) \nMothers    \nAge (years) 26.57 15 - 46 25 (21 - 32) \n*Postpartum BMI (Kg/m2) 23.33 18.7 - 31.8 23.1 (21.9 – 24.4) \nLactation duration (Weeks) 41.28 1 - 104 41 (23 - 60) \nInfants and young children    \nAge (weeks) 46.47 3 - 110   44 (28-66) \nBirth weight (Kg) 3.14 1 - 4.8 3.1 (2.8 - 3.5) \n*Weight (Kg) 8.80 5 - 16 9 (8-10) \n*Height/length (cm)  70.50 50 - 98 70 (64 - 76) \n*Mid upper arm circumference (cm)  14.82 12.1 - 18.7 14.7 (14-15.5) \n*Temperature (oC) 36.5 35.4 - 39.4 36.4 (36.2 – 36.7) \n*Hemoglobin (g/dL) 10.92 8 - 13.5 11 (10.3 - 11.8) \n* = maternal and infant/child factors measured at sampling; Kg/m2 = kilogram per meter \nsquared; % = percent; cm = centimetre; °C = degree Celsius; g/dL = grams per decilitre ; IQR \n= interquartile range; N = number of independent observations. \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 22 of 24 \n \nTABLE 2: Median (IQR) HMO concentrations and lactation duration in all mothers \nHMOs \nnmol/mL \nTotal (N=127) >6 months (n=89) ≤ 6 months (n=38) Unadjusted \np-value \n2'FL  5929.5 (2715.8-\n9271.8) \n6379.9 (3640.8-\n9364.7) \n5189.9 (523.6-\n9044.3) \n0.21 \n3FL  378.9 (263.8-\n606.8) \n452.3 (289.4-\n687.0) \n308.6 (209.2-\n400.8) \n0.002 \nLNnT 805.2 (578.2-\n1033.5) \n751.7 (506.9-\n977.6) \n937.7 (672.9-\n1140.3) \n0.020 \n3'SL 632.0 (353.4-\n987.8) \n731.6 (411.5-\n1101.7) \n470.0 (281.5-\n737.8) \n0.002 \nDFLac 548.4 (287.0-\n881.5) \n651.2 (378.9-\n1128.3) \n403.7 (192.0-\n594.8) \n<0.001 \n6’SL 157.8 (102.2-\n270.9) \n121.1 (82.3-\n171.6) \n403.4 (218.4-\n569.7) \n<0.001 \nLNT 1028.7 (694.8-\n1618.0) \n964.6 (646.1-\n1618.2) \n1134.7 (752.7-\n1366.6) \n0.42 \nLNFP I 599.3 (172.6-\n1863.9) \n493.7 (236.9-\n1707.5) \n1050.3 (109.3-\n2411.5) \n0.38 \nLNFP II 1411.1 (721.0-\n2551.2) \n1428.2 (782.5-\n2492.5) \n1218.7 (557.0-\n2796.8) \n0.89 \nLNFP III 70.0 (55.0-88.8) 69.0 (57.1-83.9) 79.0 (52.7-96.7) 0.18 \nLSTb 123.5 (90.8-\n182.2) \n124.7 (94.8-\n181.9) \n114.2 (90.8-209.9) 0.91 \nLSTc 46.5 (24.3-93.3) 33.8 (19.3-57.5) 133.1 (73.1-242.4) <0.001 \nDFLNT 865.9 (122.3-\n1576.6) \n904.5 (80.4-\n1481.6) \n825.8 (414.1-\n1677.1) \n0.13 \nLNH 71.2 (37.7-106.4) 61.5 (37.2-90.8) 86.1 (52.5-162.0) 0.004 \nDSLNT 417.2 (274.3-\n512.6) \n426.5 (274.3-\n507.4) \n380.8 (280.5-\n512.6) \n0.98 \nFLNH 24.2 (13.2-57.0) 20.5 (11.9-35.6) 45.9 (19.7-101.2) <0.001 \nDFLNH 24.4 (9.4-66.1) 20.3 (9.0-50.9) 32.6 (14.5-80.4) 0.030 \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 23 of 24 \n \nFDSLNH 146.7 (21.1-\n310.1) \n168.4 (16.9-\n290.3) \n108.2 (30.6-450.2) 0.27 \nDSLNH 31.7 (11.6-64.5) 20.7 (6.9-38.5) 81.8 (47.0-125.6) <0.001 \nTotal HMO 16050.6 \n(13219.4-\n17668.9) \n16162.0  \n(13539.2-\n17668.9) \n15434.5  \n(11463.8-17416.4) \n0.74 \nSia HMO  2395.0 (1995.9-\n3149.7) \n2347.5 (1926.0-\n2839.5) \n2766.7 (2062.5-\n3753.1) 0.030 \nFuc HMO 13962.3  \n(10903.9-\n15966.1) \n14074.6  \n(11403.9-\n16138.5) \n13461.9  \n(8989.9-14838.5) \n0.16 \nDiversity 4.7 (3.1-6.0) 4.5 (3.0-5.9) 5.2 (3.4-6.1) 0.16 \nIQR = interquartile range; HMOs = human milk oligosaccharides; N = total number of mothers \nsampled; n = number of independent observations per category; > = greater than; ≤ less than \nor equal to. HMOs with significant unadjusted p values are in bold. \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nPage 24 of 24 \n \nFigure legends \nFIGURE 1 Concentrations of all measured HMOs before and after 6 months postpartum in \nbreast milk from secretor Ugandan mother-child pairs. HMO name abbreviations are indicated \nin the list of abbreviations. Data are shown as box and whisker plots with the medians and the \nlower and upper quartiles or interquartile range (IQR). P values represent the outcomes of \ncomparisons of two groups by the Mann-Whitney U test. \nFIGURE 2  Concentrations of all measured HMOs in five groups of mothers representing \ndifferent lactation stages ≤ 12 (N = 21), 13 -24 (N= 14), 25 -48 (N= 40), 49 -72 (N= 36), ≥73 \n(N= 16) weeks postpartum in breast milk from Ugandan mother -child pairs. A. HMOs whose \nconcentrations do not change from ≤ 12 to ≥73 weeks. B. HMOs whose concentrations decrease \nfrom ≤ 12 to ≥73 weeks. C. HMOs whose concentrations increase from ≤ 12 to ≥73 weeks. \nHMO name abbreviations are indicated in the list of abbreviations. Data are shown as box and \nwhisker plots with the medians and the lower and upper quartiles or interquartile range (IQR). \nP values represent the outcomes of comparisons of two groups by the Mann-Whitney U test.  \nSupplemental Figure 1 Study participants flow chart. \nSupplemental FIGURE 2 Concentrations of all measured HMOs before and after 6 months \npostpartum in breast milk from non - secretor Ugandan mother -child pairs. HMO name \nabbreviations are indicated in the list of abbreviations. Data are shown as box and whisker plots \nwith the medians and the low er and upper quartiles or interquartile range (IQR). P values \nrepresent the outcomes of comparisons of two groups by the Mann-Whitney U test. \n \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\n≤6 > 6\n0\n5000\n10000\n15000\n200002'FL\nns\n≤6 > 6\n0\n500\n1000\n1500\n20003FL\n✱✱\n≤6 > 6\n0\n500\n1000\n1500\n2000\n2500LNnT\n✱✱\n≤6 > 6\n0\n2000\n4000\n60003'SL\n✱\n≤6 > 6\n0\n500\n1000\n1500\n2000\n2500DFlac\n✱✱✱\n≤6 > 6\n0\n500\n1000\n15006'SL\n✱✱✱✱\n≤6 > 6\n0\n1000\n2000\n3000\n4000\n5000LNT\nns\n≤6 >  6\n0\n1000\n2000\n3000\n4000\n5000LNFP I\n✱\n≤6 > 6\n0\n1000\n2000\n3000\n4000\n5000LNFP II\nns\n≤6 > 6\n0\n100\n200\n300LNFP III\nns\n≤6 > 6\n0\n200\n400\n600\n800LSTb\nns\n≤6 > 6\n0\n200\n400\n600\n800\n1000\n1200LSTc\n✱✱✱✱\n≤6 > 6\n0\n1000\n2000\n3000DFLNT\nns\n≤6 > 6\n0\n100\n200\n300\n400LNH\nns\n≤6 > 6\n0\n200\n400\n600\n800\n1000\n1200DSLNT\nns\n≤6 > 6\n0\n200\n400\n600FLNH\n✱✱✱\n≤6 > 6\n0\n100\n200\n300\n400DFLNH\n✱✱\n≤6 > 6\n0\n200\n400\n600\n800FDSLNH\nns\n≤6 > 6\n0\n100\n200\n300\n400\n500DSLNH\n✱✱✱✱\n≤6 > 6\n0\n5000\n10000\n15000\n20000\n25000Total HMOs\nns\n≤6 > 6\n0\n2000\n4000\n6000\n8000Sia HMO\nns\n≤6 > 6\n0\n5000\n10000\n15000\n20000\n25000Fuc HMOs\nns\n≤6 > 6\n0\n2\n4\n6\n8\n10Diversity\nns\nFigure 1.  HMO concentrations and lactation duration in Secretor Mothers\n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\nSuppl Fig 1: Participant flow chart\n400 children < 5years enrolled in malaria incidence follow-up (HD4MC study)\n273 children > 2 years old 127 breastfeeding children < 2 \nyears \n127 breastfeeding children+ \ntheir mothers were enrolled \nin the current study\n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint \n\n6'SL Sia HMO FLNH LSTb 2'FL \nSuppl. Figure 2. HMO concentrations and lactation duration in Non-secretor Mothers \n \n \n \n \n700 \n600 \n500 \n400 \n300 \n200 \n100 \n0 \n \nns \n \n \n≤6 > 6 \n \n \n800 \n \n \n600 \n \n \n400 \n \n \n200 \n \n \n0 \nns \n \n \n≤6 > 6 \n \n \n4000 \n \n \n3000 \n \n \n2000 \n \n \n1000 \n \n \n0 \n \nns \n \n \n≤6 > 6 \n \n \n3000 \n \n \n2000 \n \n \n1000 \n \n \n0 \n \n✱✱ \n \n \n≤6 > 6 \n \n \n400 \n \n \n300 \n \n \n200 \n \n \n100 \n \n \n0 \n \n✱ \n \n \n≤6 > 6 \n \n \n \n1200 \n \n1000 \n \n800 \n \n600 \n \n400 \n \n200 \n \n0 \n \n \n✱✱ \n \n \n \n \n \n \n \n \n \n \n \n≤6 > 6 \n \n \n6000 \n \n \n4000 \n \n \n2000 \n \n \n0 \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n≤6 > 6 \n \n \n200 \n \n150 \n \n100 \n \n50 \n \n0 \n \nns \n \n \n≤6 > 6 \n \n \n5000 \n \n4000 \n \n3000 \n \n2000 \n \n1000 \n \n0 \n \nns \n \n \n≤6 > 6 \n \n \n200 \n \n150 \n \n100 \n \n50 \n \n0 \nns \n \n \n≤6 > 6 \n \n \n500 \n \n400 \n \n300 \n \n200 \n \n100 \n \n400 \n \n300 \n \n200 \n \n100 \n \n1500 \n \n \n1000 \n \n \n500 \n✱✱ \n800 \n \n600 \n \n400 \n \n200 \n \n2000 ns \n \n1500 \n \n1000 \n \n500 \n \n0 \n≤6 > 6 \n \n0 \n≤6 > 6 \n \n0 \n≤6 > 6 \n \n0 \n≤6 > 6 \n \n0 \n≤6 > 6 \n \n \n100 \n \n80 \n \n60 \n \n40 \n \n20 \n \n0 \nns \n \n \n≤6 > 6 \n \n150 \n \n \n100 \n \n \n50 \n \n \n0 \n \n \n \n \n \n \n \n \n \n \n \n \n \n≤6 > 6 \n \n1200 \n \n1000 \n \n800 \n \n600 \n \n400 \n \n200 \n \n0 \n \n✱✱ \n \n \n≤6 > 6 \n \n500 \n \n400 \n \n300 \n \n200 \n \n100 \n \n0 \n \n✱✱ \n \n \n≤6 > 6 \n \n14000 \n \n12000 \n \n10000 \n \n8000 \n \n6000 \n \n \n \n \n \n \n \n \n \n \n \n \n \n≤6 > 6 \n \n \nns \n6000 \nns \n10000 8 \nns \n \n \n4000 \n \n \n2000 \n \n \n0 \n \n \n \n \n \n \n \n \n \n \n≤6 > 6 \n \n8000 \n \n6000 \n \n4000 \n \n2000 \n \n0 \n \n \n \n \n \n \n \n \n \n \n≤6 > 6 \n \n \n6 \n \n \n4 \n \n \n2 \n \n \n0 \n≤6 > 6 \nns \nns \n ✱ \nns \nns \n ns \nLNT DFLNH LSTc 3FL Fuc HMOs \nLNnT DFLNT FDSLNH LNFP I \nDiversity \nLNFP II 3'SL \nLNH DSLNH \nTotal HMOs DSLNT LNFP III DFlac \n . CC-BY 4.0 International licenseIt is made available under a \nperpetuity. \n is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint \nThe copyright holder for thisthis version posted March 22, 2024. ; https://doi.org/10.1101/2024.03.21.24304661doi: medRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}