Cynthia Williams
Cynthia Williams; Anne Fenton; Lauren Nagy; Qiuting Hong; L.Renee Ruhaak; Satya Dadenkar; Carlito Lebrilla
UC Davis, Davis, CA
NOVEL ASPECT: Plasma and IgG glycan profiles, but not IgG protein abundances are altered in HIV infection.
Human immunodeficiency virus (HIV) is a retrovirus that infects and weakens the immune system and can progress to AIDS, which can be fatal. Extensive HIV studies have shown that immunoglobulin expressions are abnormal and display altered glycosylation patterns. However, compositional glycomic profiling of human plasma and quantitation of specific protein glycosylation in HIV samples has not been widely explored. In this study, we profile the plasma glycans to determine glycan biomarkers for HIV infection. In addition, we have developed a method for quantification of glycosylated proteins, specifically immunoglobulin G (IgG) to examine whether IgG-specific glycosylation varies in HIV infection.
Plasma samples from 26 male patients as well as 11 controls were collected; HIV infected patients were divided into 3 groups: no therapy (n=11), therapy (n=11), and long term no therapy (n=4). For the glycomic analysis, the glycoproteins were denatured and glycans were enzymatically released by PNGase F. The N-glycans were purified using graphitized carbon SPE and analyzed on an Agilent 6520 HPLC-Chip/QTOF MS. Accurate mass, retention time, and a retrosynthetic library was used for glycan identification. For IgG quantification, plasma samples were treated with DTT and IAA followed by an overnight tryptic digestion at 37˚C. Quantification was performed using an Agilent 6490 QqQ MS coupled with an Agilent 1290 UPLC system.
Total plasma N-glycan profiles were generated for all 37 samples using our high-throughput nano-LC separation method. Namely, N-glycans were chromatographically separated on a PGC chip. Over 300 glycan features were resolved and identified which corresponded to over 100 glycan compositions. Statistical differences were found in neutral, fucosylated and sialylated, as well as high mannose glycans. Specifically, Hex9HexNAc2 or Man 9 was found to be decreased in HIV patients but increased in those with little to no therapy, with therapy treatment, and in control cases. With Man 9, t-tests revealed a p= 7.46*10-4 between healthy and no therapy samples. Conversely, Hex3HexNAc5Fuc1 was found to decrease with therapy treatment versus healthy controls (p=8.9*10-3). Previous studies in our lab have shown that Man 9 and Hex3HexNAc5Fuc1 are both associated with immunoglobulins, specifically IgM, and IgM, IgG, and IgA, respectively. Quantification of IgG glycosylation was examined to determine whether the observed differences originated from glycan or protein expression. A method for IgG quantitation using QqQ-MS was developed for quantitating peptides and applied to the 37 HIV sample set.  Quantitation of the peptide that is common to IgG 1, 2, 3, and 4, using MRM, revealed no statistical differences in the IgG levels between the different HIV sample groups. However, after normalizing the protein glycosylation to the total protein content, many glycopeptides revealed significant statistical differences. Galactose-deficient glycopeptides were observed at much higher intensities in HIV patients compared to healthy cases, which is a trend that has already been reported in literature. These results suggest that glycosylation undergoes specific changes during infection but not necessarily protein expression.