Alterations in protein regulators of neurodevelopment in the cerebrospinal fluid of infants with posthemorrhagic hydrocephalus of prematurity

Abstract

Neurological outcomes of preterm infants with posthemorrhagic hydrocephalus are among the worst in newborn medicine. There remains no consensus regarding the diagnosis or treatment of posthemorrhagic hydrocephalus, and the pathological pathways leading to the adverse neurological sequelae are poorly understood. In the current study, we developed an innovative approach to simultaneously identify potential diagnostic markers of posthemorrhagic hydrocephalus and investigate novel pathways of posthemorrhagic hydrocephalus-related neurological disability. Tandem multi-affinity fractionation for specific removal of plasma proteins from the hemorrhagic cerebrospinal fluid samples was combined with high resolution label-free quantitative proteomics. Analysis of cerebrospinal fluid obtained from infants with posthemorrhagic hydrocephalus demonstrated marked differences in the levels of 438 proteins when compared with cerebrospinal fluid from age-matched control infants. Amyloid precursor protein, neural cell adhesion molecule-L1, neural cell adhesion molecule-1, brevican and other proteins with important roles in neurodevelopment showed profound elevations in posthemorrhagic hydrocephalus cerebrospinal fluid compared with control. Initiation of neurosurgical treatment of posthemorrhagic hydrocephalus resulted in resolution of these elevations. The results from this foundational study demonstrate the significant promise of tandem multi-affinity fractionation-proteomics in the identification and quantitation of protein mediators of neurodevelopment and neurological injury. More specifically, our results suggest that cerebrospinal fluid levels of proteins such as amyloid precursor protein or neural cell adhesion molecule-L1 should be investigated as potential diagnostic markers of posthemorrhagic hydrocephalus. Notably, dysregulation of the levels these and other proteins may directly affect ongoing neurodevelopmental processes in these preterm infants, providing an entirely new hypothesis for the developmental disability associated with posthemorrhagic hydrocephalus.

Fig. 1.

Pathway analysis of neurodevelopmental proteins identified in CSF of posthemorrhagic hydrocephalus. Network 1 resulting from the Ingenuity Pathways Analysis comprising the proteins identified in PHH CSF, including APP, L1CAM, BCAN, and NCAM-1. Additional networks that were mapped to gene objects in the IPA database are shown in supplemental Figs. S2–S11. As shown in the legend, nodes represent genes, with specific shapes denoting protein functional classes. Direct and indirect relationships between nodes are shown as unbroken or dashed lines indicating, respectively.

Fig. 2.

Tandem multi-affinity fractionation yields consistent samples for LC-MS analysis with limited technical variability. A, Experimental design for label-free protein quantification of two CSF procurements S1 and S2 obtained 7 days apart from a single infant after initiation of treatment for PHH. Four multi-affinity fractionations were performed for each sample and analyzed in a randomized fashion as follows: S1.2, S1.1, S1.4, S2.3, S2.2, S2.1, S2.4, and S1.3. B, SDS-PAGE analysis of the four MAF replicates from S1 and S2. The distributions of the coefficient of variation for the peptide ion intensities from LC-MS analysis is plotted for the observed peptide signals as isotope groups. C, Peptide ion intensities for bovine serum albumin from the indicated internal standard peptide from S1 (blue bars) and S2 (red bars, and the individual samples as designated in A. D, Peptide ion intensities of the indicated L1CAM peptide from S1 (blue bars) and S2 (red bars, and the individual samples as designated in A). E, Hierarchical cluster analysis of significant peptide intensities (p > 0.001) after ANOVA statistical analysis as described under “Experimental Procedures.” The LC-MS analysis order was as follows: S1.3, S2.3, S2.2, S1.4, S1.1, S1.2, S2.4, and S2.1.

Fig. 3.

Tandem MAF-quantitative proteomics demonstrates consistent changes at the peptide and protein level in the CSF of preterm infants during treatment for PHH. Replicate fractions from two CSF procurements (S1 and S2) obtained 7 days apart from a single infant after initiation of treatment for PHH. Shown are the standardized peptide intensities of the 6 L1CAM peptides (A), the 3 APP peptides (B), and the eight chromagranin A peptides (C). Bar coloring is used to differentiate standardized peptide intensities by replicate fraction for both S1 and S2.Standardized intensities were calculated as described under “Experimental Procedures.” The peptides and their normalized intensity values are given in supplemental Table S3. D, Hierarchical cluster analysis of the mean intensity z score for all proteins that were identified with two or more peptides, both with a Mascot score of > 40. Several proteins of interest are highlighted (KRT: Keratin; HBB: Hemoglobin beta; HBA2: Hemoglobin alpha 2; HBG2: Hemoglobin gamma G; CHGA: Chromogranin A; BCAN: Brevican; NCAM1: Neural cell adhesion molecule 1; APP: Amyloid precursor protein; L1CAM: L1 cell adhesion molecule).

Fig. 4.

Confirmation of immunodepletion in tandem multi-affinity fractionation. A, The concentration of ApoA4 shown in naive (before tandem MAF) and immunodepleted (ID, after tandem MAF) S1 and S2 CSF samples. B, The relative levels of the full length isoform of beta-actin (ActB) in naive and ID S1 and S2 CSF. Note that after tandem MAF, the levels of ApoA4 and beta-actin were dramatically and significantly reduced.

Fig. 5.

Changes in the CSF levels of protein mediators of neurodevelopment after initiation PHH treatment are protein-specific. A, Volcano plot showing changes in the relative abundances of CSF protein mediators of neurodevelopment identified with tandem MAF proteomics after initiation of PHH treatment. The p values from the Benjamini-Hochberg-corrected ANOVA results are plotted against the log₂ fold change. The proteins shown are the quantifiable proteins from the neurodevelopmental pathway shown in Fig. 1. After initiation of PHH treatment, proteins displayed a decrease (left of 0 on the axis), increase (right of 0), or no significant change at all (below the hashed red line). Note the decrease in APP and L1CAM over the 7 day interval from S1 to S2, whereas ActB and ApoA4 increase. Fold-change measurements for the proteins demonstrating a significant difference between samples are shown in Table II. B, Conventional laboratory techniques including Western blot (L1CAM, APP, and ActB) and ELISA (ApoA4) confirm the trends in protein levels observed with tandem MAF process. The immunoblot for beta actin was shown in the previous figure. Asterisks denote p < 0.05, n = 3 per group.

Fig. 6.

The CSF levels of several key protein mediators of neurodevelopment are elevated in PHH but return to control levels after initiation of PHH treatment. A–D, (insets). Using tandem MAF-proteomics, CSF from the initial PHH sample (PHH_IS) demonstrated 4.96, 10.4, 8.17, and 2.55 fold increase in L1CAM, APP, BCAN, and NCAM-1 over control CSF, respectively. At TEA, there was no significant difference detected when comparing PHH (PHH_TEA) with control for three of the aforementioned proteins; NCAM-1 was decreasing, but at a slower rate and therefore was still different between PHH and control groups (p = 0.02). n = 3 for each group in the tandem MAF proteomics data. A–D, Graphs plotting protein levels versus EGA in weeks are shown. ELISAs for L1CAM (PHH: n = 12; control: n = 31), APP (PHH: n = 12; control: n = 25), and NCAM-1 (PHH: n = 12; control: n = 19), and a Western blot for BCAN (PHH: n = 12; control: n = 12), verified the increases in CSF levels of these proteins observed in PHH infants compared with age-matched controls. There was variability in the time course of return to control levels for these proteins, such that the increase in L1CAM appeared to resolve earlier than NCAM-1. The color red denotes the PHH group and the color blue was used for the control group. The thick solid lines are the best fit regression lines with 95% confidence interval boundaries (thinner lines).