Engineered plants provide a photosynthetic platform for the production of diverse human milk oligosaccharides

Abstract

Human milk oligosaccharides (HMOs) are a diverse class of carbohydrates which support the health and development of infants. The vast health benefits of HMOs have made them a commercial target for microbial production; however, producing the approximately 200 structurally diverse HMOs at scale has proved difficult. Here we produce a diversity of HMOs by leveraging the robust carbohydrate anabolism of plants. This diversity includes high-value and complex HMOs, such as lacto-N-fucopentaose I. HMOs produced in transgenic plants provided strong bifidogenic properties, indicating their ability to serve as a prebiotic supplement with potential applications in adult and infant health. Technoeconomic analyses demonstrate that producing HMOs in plants provides a path to the large-scale production of specific HMOs at lower prices than microbial production platforms. Our work demonstrates the promise in leveraging plants for the low-cost and sustainable production of HMOs.

Fig. 1. Production of all three HMO classes in planta.

a, HMOs are composed of Glc, Gal, Fuc, GlcNAc and/or Neu5Ac connected via Gal-ß-1,3/4, GlcNAc-ß-1,3/6, Fuc-α-1,2/3/4 or Neu5Ac-α-2,3/6 glycosidic linkages. b, HMOs can be divided into branched, unbranched, type I and/or type II HMOs. c, HMO biosynthetic pathways used in this study for HMO production in planta. d, Extracted ion chromatograms (EIC) showing the identification of 2′FL, 3′SL, 6′SL, LNFPI, LSTa, LSTc and LNT/LNnT in extracts of individual plant leaves using LC–MS/MS (Q Exactive, Thermo Fisher Scientific). Red, blue and purple colouring denote fucosylated, neutral and acidic HMOs, respectively. Some further peaks are present due to in-source fragmentation of larger oligosaccharides with no available standards.

Fig. 2. Manipulation of nucleotide sugar biosynthetic pathways modulates HMO profiles in planta.

a, HMO biosynthetic pathway expressed for the production of LNFPI. b, Nucleotide sugar biosynthetic pathways expressed with LNFPI pathway. c, Quantification of LNFPI production through expression of LNFPI biosynthetic pathway alongside combinatorially expressed nucleotide sugar biosynthetic pathways using an internal calibration curve obtained with an Agilent 6530 Accurate-Mass Q-ToF MS. The middle bar represents the median. Upper and lower whiskers correspond to the largest and smallest values within 1.5 × the interquartile range, respectively. Upper and lower hinges represent the third and first quartiles, respectively. Statistical analysis was conducted using a heteroscedastic two-tailed Student’s t-test with the LNFPI pathway expressed alone used as the reference group. *P < 0.05. P values are: LNFPI + fucose, 0.030; LNFPI + GlcNAc, 0.012; LNFPI + fucose + GlcNAc, 0.01; LNFPI + GlcNAc + Gal, 0.01; LNFPI + fucose + GlcNAc + Gal, 0.043. A sample size of three leaves was used for each experiment. d, Effect of nucleotide sugar biosynthetic pathway overexpression on HMO profile produced using the LNFPI pathway. Values reflect normalized peak area. Hexose, HexNAc, deoxyhexose (Deoxyhex) determined using m/z and MS/MS fragmentation. We performed mass spectral analysis on an Agilent Q-ToF MS.

Fig. 3. Production of HMOs in stably transformed plants.

a, Constructs used in creation of stable lines containing biosynthetic enzymes for the production of LNFPI. b, Photos of 4-week-old transgenic N. benthamiana. c, Concentration of LNFPI produced in leaves of each stable line. d, Concentration of 2′FL produced in leaves of each stable line. For quantification, three leaves from each plant were analysed separately. Quantification of LNFPI and 2′FL obtained with a Thermo Fisher Scientific Q Exactive Mass spectrometer. LB, left border; 2A , 2A peptide; RB, right border; P_x, promoter; T_x, terminator; WT, wild type. The middle bar represents the median. Upper and lower whiskers correspond to the largest and smallest values within 1.5 × the interquartile range, respectively. Upper and lower hinges represent the third and first quartiles, respectively. A sample size of three leaves was used for each experiment.

Fig. 4. Optimized purification protocol developed for functional analysis of plant-produced HMOs.

a, Workflow of extraction, purification and characterization of HMOs from N. benthamiana leaves. b, Growth curves of HMO-consuming (BLI 15697) and control (BAL 27536) strains in media supplemented with HMO isolated from breast milk (HMO) or HMOs isolated from plants (pHMO). Error bars represent standard deviation.

Fig. 5. Plant-based platform improves the economics of producing the HMO, LNFPI.

Estimated MSP of LNFPI produced using biomass sorghum as a model production platform in two bioethanol price scenarios. Error bands represent final values calculated with ±20% of input parameters.