Article pubs.acs.org/jpr
Total Plasma N‑Glycome Changes during Pregnancy L. Renee Ruhaak,*,†,‡ Hae-Won Uh,§ André M. Deelder,† Radboud E. J. M. Dolhain,∥ and M. Wuhrer†,⊥ †
Center for Proteomics and Metabolomics, Leiden University Medical Center, Leiden 2333 ZA, The Netherlands Department of Chemistry, University of California Davis, Davis, California 95616, United States § Department of Medical Statistics and Bioinformatics, Section of Medical Statistics, Leiden University Medical Center, Leiden 2333 ZA, The Netherlands ∥ Department of Rheumatology, Erasmus MC University Medical Center, Rotterdam 3015 CE, The Netherlands ⊥ Division of BioAnalytical Chemistry, VU University Amsterdam, Amsterdam 1081 HV, The Netherlands ‡
S Supporting Information *
ABSTRACT: During pregnancy, the mother faces a major immunological challenge. Most of the major plasma proteins have important immunological functions, and altered levels of these major proteins have been reported during pregnancy, potentially providing immunosuppression. A large number of the high abundance plasma proteins are post-translationally modified by N-glycans, and while it is now understood that these glycans may also affect the immunological functions, their pattern has not been studied in relation to pregnancy. Here, the N-glycosylation profile of 32 pregnant women was determined over the course of their pregnancy using a multiplexed CGE-LIF method. Moreover, for 6 women, the glycosylation profiles of the proteins IgG, IgA, and alpha1-antitrypsin were monitored. For total plasma, 16 glycan signals showed differential expression during pregnancy. In general the levels of largely sialylated bi-, tri-, and tetra-antennary glycans were increased during pregnancy, while biantennary glycans with no more than one sialic acid were decreased. Similarly altered glycosylation profiles were observed for the individual proteins IgG, with a decrease of digalactosylated biantennary glycans after delivery, and alpha1-antitrypsin, on which increased levels of triantennary glycans were observed during pregnancy. Overall, these results show altered glycosylation profiles both for total plasma glycoproteins and on individual proteins during pregnancy, which may contribute to immunosuppression and have other biological functions. KEYWORDS: N-glycans, pregnancy, CGE-LIF, plasma, immunoglobulin, alpha1-antitrypsin
■
INTRODUCTION
remained constant with normal pregnancies, and increases were observed only in pregnancies complicated by acute inflammation.5 However, Honda et al. later observed decreased serum levels of AGP within all trimesters,6 and this latter finding was recently confirmed by Larsson et al.4 The serum levels of AAT have also been studied extensively in pregnant women. Typically, AAT levels are 4- to 6-fold increased during pregnancy.7 Because AAT deficiencies have been associated with preeclampsia8 and spontaneous abortions,9 it is hypothesized that increased levels of AAT have an anti-inflammatory role. The synthesis of acute phase proteins mainly takes place in the liver. Because the levels of several of the acute phase proteins are altered during pregnancy, it is likely that the protein synthesis by the liver is affected. Indeed, it was reported that several liver function tests show abnormal values during pregnancy.10 Interestingly, the decreased levels of AGP
A major immunological challenge in pregnancy is raised by the genetic and antigenic differences between mother and fetus. In order to tolerate the fetus and suppress allo-immunological responses during pregnancy, the maternal immune system has to undergo changes.1−3 A large portion of the highly abundant blood serum proteins have important immunological functions, and it is, therefore, not surprising that the levels of several of these proteins are altered in pregnant women. In a recent study, levels of the acute phase proteins alpha1acid glycoprotein (AGP), alpha1-antitrypsin (AAT), C-reactive protein (CRP), haptoglobin, albumin, and the immunoglobulins IgG, IgA, and IgM were monitored in healthy women, which were sampled during and after their pregnancy.4 Increased levels of AAT, CRP, and IgM were observed during pregnancy, while levels of albumin, AGP, and IgG showed a decrease. Levels of haptoglobin and IgA were reported to remain fairly constant during pregnancy.4 The effect of pregnancy on the levels of AGP has repeatedly been studied, leading to conflicting results. In early studies, AGP serum levels © 2014 American Chemical Society
Received: November 15, 2013 Published: February 14, 2014 1657
dx.doi.org/10.1021/pr401128j | J. Proteome Res. 2014, 13, 1657−1668
Journal of Proteome Research
Article
States Biochemicals (Cleveland, OH). PNGase F was obtained from Roche Diagnostics (Mannheim, Germany). Biogel P-10 was obtained from Bio-Rad (Veenendaal, The Netherlands). Acetic acid, citric acid, and ethanol were from Merck (Darmstadt, Germany). Acetonitrile was purchased from Biosolve (Valkenswaard, The Netherlands). MQ (Milli-Q deionized water; R > 18.2 MΩ cm−1) was used throughout (Millipore, Amsterdam, The Netherlands).
observed during pregnancy were inversely correlated with the serum triglyceride levels of the women during pregnancy, and both changes are supposedly largely liver-mediated.6 A large proportion of the blood serum proteins is posttranslationally modified with N-glycans, and it is well-known that glycosylation has important functions in protein folding and influences protein half-life and protein−protein interactions such as receptor binding.11−13 However, while protein levels have been studied during pregnancy, studies toward the glycan species decorating these proteins are limited. Only the glycosylation profile of immunoglobulin G has been shown to alter during pregnancy; this has been reported both for healthy individuals and patients with rheumatoid arthritis (RA).14−17 Interestingly, RA is associated with lower galactosylation of IgG,18,19 while the levels of galactosylated glycans on IgG increase during pregnancy. Furthermore, this increase in galactosylation during pregnancy is associated with an improvement of RA.14−16,20 Levels of several hormones change significantly during pregnancy. These changes form the basis of the necessary physiological and immunological changes that are required to initiate and maintain a successful pregnancy. Progesterone is one of the major hormones involved in pregnancy and has been shown to influence IgG glycosylation.21 Recent advancements in the technology available for Nglycan analysis allow the rapid acquisition of glycosylation profiles from body fluids such as serum or plasma.22,23 Especially, the use of automated MALDI-MS procedures,24,25 which allow generation of mass fingerprints within seconds, and the use of multiplexed CGE-LIF technology,26,27 which allows separation and detection of 96 glycan profiles within 1 h, greatly enhance the throughput of N-glycan fingerprinting. These technological advancements now allow us to study the protein glycosylation in larger numbers of samples and open up the analysis of larger longitudinal cohorts with several samples per individual. In this study, the overall blood plasma glycosylation profiles of 32 healthy pregnant women were studied at six time points during and after their pregnancy. Moreover, the protein-specific glycosylation profiles of the two immunoglobulins IgG and IgA and the acute phase protein AAT were studied for the six time points in a subset of 6 pregnant women. The protein-specific profiles will indicate whether the altered plasma glycosylation profile is a representation of the overall altered glycosylation profile in all proteins, or whether protein-specific changes occur, which may vastly affect protein function. To obtain the glycosylation profiles, a high-throughput strategy comprising glycan release, aminopyrene-1,3,6-trisulfonic acid (APTS) labeling, hydrophilic interaction liquid chromatography (HILIC) SPE, and multiplexed capillary gel electrophoresis with laser-induced fluorescence (CGE-LIF) analysis was used. This strategy allowed the determination of altered glycosylation profiles during and after pregnancy and is anticipated to provide further insight into the differential glycosylation of blood proteins during pregnancy.
■
Human Plasma Samples
Blood from healthy pregnant Caucasian women without adverse obstetric history (n = 32) was taken during home visits three times during pregnancy (12, 20, and 30 weeks of gestation) and three times postpartum (6 and 12 weeks and between 6 and 9 months) as described earlier.28 The first five visit time points were all met within a deviation of 1 week. The last time point at 6−9 months postpartum was set as baseline value. At that time, we considered the participants as being completely recovered from their pregnancies. Plasma samples were stored at −80 °C. All women participated in the study after informed consent was given. The study was in accordance with the Helsinki II declaration and was approved by the local ethical committee. Capturing of Specific Proteins
The proteins IgG, IgA, and AAT were immune-affinity captured as described earlier.29−31 Briefly, anti-human AAT llama antibody fragments were purified, dialyzed against PBS, and coupled to NHS-activated Sepharose 4 fast flow beads (GE Healthcare) according to the manufacturer’s instructions, at 3 mg/mL overnight to generate anti-AAT beads. Capturing beads were washed three times with 10 volumes of PBS. Fifteen microliters of Protein G beads (GE Healthcare), 6 μL of antiAAT, or 20 μL of anti-IgA beads (Sigma-Aldrich) were applied per well of a 96-well filter plate (Orochem, Lombard, IL). The volume was brought to 200 μL with PBS, and 10 μL of plasma was applied per well. The plate was incubated for 1 h, and then the nonbound protein fraction was removed using vacuum filtration. The beads were washed three times with PBS followed by 2 washes with water. Then, the enriched glycoproteins were eluted into V-bottom 96-well plates (Nunc through VWR, Amsterdam, NL) using 100 mM formic acid. The eluates were subsequently dried by vacuum centrifugation. Preparation of Oligosaccharides from IgG, IgA, and AAT Samples
N-Glycans were released from human protein fractions as previously described.32 Briefly, 2 μL of 2% SDS was added to the proteins. Upon denaturation by incubation at 60 °C for 10 min, 2 μL of 2% nonidet P-40 (NP-40) in 5x PBS containing 0.2 mU of PNGase F was added to the samples. The samples were incubated overnight at 37 °C for N-glycan release. Labeling of oligosaccharides was performed as published26,33 with slight modifications: 2 μL of a freshly prepared solution of label (APTS; 20 mM in 3.6 M citric acid) and 2 μL of freshly prepared reducing agent solution (0.2 M 2-picoline borane in DMSO, see ref 34) were added, the plate was sealed using adhesive tape, and after 5 min of shaking, the samples were incubated at 37 °C for 16 h. To stop the reaction, 50 μL of acetonitrile/water (80:20 v/v) was added to the reaction mixture.
EXPERIMENTAL SECTION
Materials
Dimethylsulfoxide (DMSO), nonidet P-40 (NP-40), formic acid, triethylamine (TEA), APTS, and 2-picoline borane were obtained from Sigma-Aldrich (Zwijndrecht, The Netherlands). Sodium dodecyl sulfate (SDS) was purchased from United 1658
dx.doi.org/10.1021/pr401128j | J. Proteome Res. 2014, 13, 1657−1668
Journal of Proteome Research
Article
Figure 1. CGE-LIF analysis of APTS-labeled total plasma N-glycans: electropherogram with peak annotations. An overlay of 10 electropherograms from 10 individual mothers 6 months postpartum is depicted. Peak numbers have been assigned according to our previous publication.26 Major peaks have been assigned with glycan structures. Further structural information regarding the peak annotations is reported in Table 1. Symbol key: blue square: N-acetyl glucosamine, green ball: mannose, yellow ball: galactose, red triangle: fucose and purple diamond: sialic acid.
Preparation of Oligosaccharides from Human Plasma Samples
deep well plate. The combined eluates were either analyzed immediately by CGE-LIF or stored at −20 °C until usage.
N-Glycans from human plasma were prepared as described previously.25,26 Briefly, 20 μL of 2% SDS was added to 10 μL of plasma. The proteins were denatured at 60 °C for 10 min, and subsequently 10 μL of 4% NP-40 and 0.5 mU of PNGase F in 10 μL of 5x PBS were added to the samples. The samples were incubated overnight at 37 °C for N-glycan release. To label the released glycans, 2 μL of N-glycan solution was mixed with 2 μL of a freshly prepared solution of label (APTS; 20 mM in 3.6 M citric acid). Then, 2 μL aliquots of freshly prepared reducing agent solution (2 M 2-picoline borane34 in DMSO) were added. The samples were labeled by incubation at 37 °C for 16 h. To stop the reaction, 50 μL of acetonitrile/water (80:20 v/v) was added to the reaction mixture.
CGE-LIF Using ABI-3730 DNA Sequencing Equipment
For analysis, 2 μL of N-glycan eluate was added to 60 μL of DMSO. After mixing, samples were analyzed using an ABI3730 DNA sequencer (Applied Biosystems), as previously described.26 The injection voltage was set to 7.5 kV, while the running voltage was 10 kV. The system was equipped with a 48 channel array with capillaries of 50 cm in length, and the capillaries were filled with POP-7 buffer (Applied Biosystems). The 3730 running buffer was obtained from Applied Biosystems. Data were collected with a frequency of 10 Hz for 50 min. Data Processing of DNA Sequencer Data
Data files were converted to xml files using DataFileConverter, which is supplied by Applied Biosystems. Files were then loaded into Matlab software (version 2007a; The Mathworks, Inc., Natick, MA), and after smoothing, the data were cropped to 17,000 data points to reduce the data volume. Alignment using Correlation optimized warping (COW) was performed as reported previously35,36 using a representative electropherogram as the reference file. Segment length and slack size were optimized according to ref 35. After smoothing and background adjustment, the peak areas were determined (30 for total plasma, 19 for IgG, 15 for IgA, 24 for AAT, and 15 for IgA). Peak areas were normalized on the total glycan signal intensity.
HILIC-Solid Phase Extraction (SPE)
Free label and reducing agent were removed from the samples using HILIC-SPE as previously described.26 Briefly, 10 mg of Biogel P-10 in water/ethanol/acetonitrile (70:20:10, v/v) was applied to each well of a 0.45 μm GHP filter plate (Pall Corporation, Ann Arbor, MI). A vacuum manifold was used to remove solvent throughout the procedure. All wells were prewashed using 5 × 200 μL water, followed by equilibration using 3 × 200 μL acetonitrile/water (80:20, v/v). The N-glycan samples were loaded to the wells, and the plate was shaken for 5 min on a shaker to enhance glycan binding. The wells were subsequently washed using 5 × 200 μL acetonitrile/100 mM triethylamine (TEA) adjusted to pH 8.5 with acetic acid (80:20, v/v), followed by 3 × 200 μL acetonitrile/water (80:20, v/v) to remove TEA. For the elution, 100 μL of water was applied followed by 5 min incubation on the shaker (to allow swelling of the Biogel P-10 particles). Thereafter, an additional 200 μL water was added followed by a 5 min incubation on the shaker. The eluates were collected by vacuum into a 96-well V-bottom polypropylene deep well plate (Westburg, Leusden, The Netherlands). Another 200 μL of water was added, followed by a 5 min incubation on the shaker and elution into the same
Statistics
The 192 samples were divided over 3 individual plates to analyze the plasma glycosylation patterns. All six samples of one individual were kept on the same plate. No batch corrections were performed, and p-values
