A human embryonic kidney 293T cell line mutated at the Golgi alpha-mannosidase II locus

Crispin M, Chang VT, Harvey DJ, Dwek RA, Evans EJ, Stuart DI, Jones EY, Lord JM, Spooner RA, Davis SJ. (2009), J Biol Chem. 284, 21684-95

Abstract

Disruption of Golgi alpha-mannosidase II activity can result in type II congenital dyserythropoietic anemia and induce lupus-like autoimmunity in mice. Here, we isolated a mutant human embryonic kidney (HEK) 293T cell line called Lec36, which displays sensitivity to ricin that lies between the parental HEK 293T cells, in which the secreted and membrane-expressed proteins are dominated by complex-type glycosylation, and 293S Lec1 cells, which produce only oligomannose-type N-linked glycans. Stem cell marker 19A was transiently expressed in the HEK 293T Lec36 cells and in parental HEK 293T cells with and without the potent Golgi alpha-mannosidase II inhibitor, swainsonine. Negative ion nano-electrospray ionization mass spectra of the 19A N-linked glycans from HEK 293T Lec36 and swainsonine-treated HEK 293T cells were qualitatively indistinguishable and, as shown by collision-induced dissociation spectra, were dominated by hybrid-type glycosylation. Nucleotide sequencing revealed mutations in each allele of MAN2A1, the gene encoding Golgi alpha-mannosidase II: a point mutation that mapped to the active site was found in one allele, and an in-frame deletion of 12 nucleotides was found in the other allele. Expression of the wild type but not the mutant MAN2A1 alleles in Lec36 cells restored processing of the 19A reporter glycoprotein to complex-type glycosylation. The Lec36 cell line will be useful for expressing therapeutic glycoproteins with hybrid-type glycans and as a sensitive host for detecting mutations in human MAN2A1 causing type II congenital dyserythropoietic anemia.

Key figure: Screening of ricin-resistant clones generated by EMS mutagenesis

A, ricin resistance of ricin-selected cell lines. Individual cell lines were treated with graded doses of ricin for 4 h, and their subsequent ability to synthesize proteins was determined. Blue, parental HEK 293T (n = 24);open circles and dotted lines, HEK 293S Lec1 (n = 4); gray (n= 1–4), orange (n = 1, 2, 3, or 6), violet (n = 1–3), and green (n = 6 or 7) indicate cell lines belonging to various selection screen-defined classes. Error bars have been omitted for clarity. B, nonselected sensitivity of all of these cell lines to PEx. Cell lines were treated with graded doses of PEx for 4 h, and their subsequent ability to synthesize proteins was determined. Blue (n = 10), open circles (n = 2), gray (n = 1–4), violet (n = 1 or 3), and green(n = 3) are same as in A; orange (n = 1–5) indicates cells with ∼5-fold or greater resistance to PEx. Error bars have been omitted for clarity. C, scatter plot of PEx resistance versus ricin resistance for all of the cell lines generated. Sensitivity of the cell lines to toxin was determined as described in A by measuring the concentration (IC₅₀, ng/ml) of toxin required to reduce protein synthesis to 50% of that of untreated cells, and resistance (-fold) was calculated from the ratio of the IC₅₀ for the cell line versus the IC₅₀ for HEK 293T cells. PEx-resistant cells (orange) cluster as a group with 2–9-fold ricin resistance; green, a cluster of cell lines with very high ricin resistance. D, scatter plot of relative number of ricin B chain-binding sites versus ricin resistance for 62 of the cell lines generated (7 “gray” and 1 “green” cell line were not tested). The relative number of ricin-binding sites on the cell lines was determined by flow cytometry using Alexa 647-labeled ricin B chain. Apart from the highly ricin-resistant lines (green), the cell lines cluster into two broad groups, with either medium to high (group I) or low to medium (group II) numbers of ricin-binding sites. Violet-colored symbols, candidate HEK 293T Lec36 cells.