An In-Depth Comparison of the Male Pediatric and Adult Urinary Proteomes

Abstract

In this study, we performed an in-depth characterization of the male pediatric infant urinary proteome by parallel proteomic analysis of normal healthy adult (n =6) and infant (n =6) males and comparison to available published data. A total of 1584 protein groups were identified. Of these, 708 proteins were identified in samples from both cohorts. Although present in both cohorts, 136 of these common proteins were significantly enriched in urine from adults and 94 proteins were significantly enriched in urine from infants. Using Gene Ontology, we found that the infant-enriched or specific subproteome (743 proteins) had an overrepresentation of proteins that are involved in translation and transcription, cellular growth and metabolic processes. In contrast, the adult enriched or specific subproteome (364 proteins) showed an overexpression of proteins involved in immune response and cell adhesion. This study demonstrates that the non-diseased male urinary proteome is quantitatively affected by age, has age-specific subproteomes, and identifies a common subproteome with no age-dependent abundance variations. These findings highlight the importance of age-matching in urinary proteomics.

Introduction

Urine is a valuable body fluid for clinical proteomics [1, 2]. It has the following advantages: i) readily available in large amounts; ii) can be obtained noninvasively; iii) can be repeatedly prospectively sampled from the same individual for longitudinal studies; iv) it is relatively stable[3] and v) it is less complex in comparison to other systemic body fluids, such as blood-derived body fluids[3]. The urinary proteome is comprised of proteins from ultrafiltration of plasma by the kidneys, in addition to the local secretion and/or shedding of proteins from the genitourinary tract[4], suggesting it may reflect local and/or distal proteomic changes. Furthermore, the content of the urinary proteome can be significantly affected by normal physiologic and environmental variations [1, 2]. Therefore, a detailed examination of the non-diseased, i.e. “normal” population will advance the understanding of the normal variations of the urinary proteome.

Research to define the non-diseased adult urinary proteome has been ongoing for the past decade and the number of the identified proteins has constantly increased. The vast majority of previous efforts have focused on the adult urinary proteome. Adachi et al. defined one of the most comprehensive catalogues of the adult urinary proteome with more than 1500 proteins identified in the urine from 10 healthy adult donors[5]. Marimuthu and colleagues recently found over 600 new proteins in urine by using substantial sample fractionation [6].

While there have been significant efforts to characterize the adult urinary proteome, there have been comparatively few efforts to characterize the pediatric urinary proteome outside of biomarker discovery studies [7, 8]. To date, there have been no studies that determine if the pediatric urinary proteome is distinct from the adult urinary proteome. In rodents, we previously demonstrated that the urinary proteome changes drastically with normal postnatal maturation [9]. In this study we perform an in-depth analysis of the normal male infant urinary proteome and then compare it to normal healthy adult males and the current reported literature. This analysis identifies proteins that are enriched in or specific to infant urine, and proteins that are present independent of age. We identified three distinct subproteomes: 1) infant subproteome, 2) adult subproteome, and 3) common subproteome. Using spectral counting, we identified a subset of proteins in the common subproteome to be enriched or diminished based on age. To our knowledge, this is the first in-depth interrogation of the normal male infant urinary proteome and demonstrates that the non-diseased urinary proteome changes between infants and adults. This information is critical to future proteomic-based clinical studies on infant urine.

Experimental Section

Urine sample acquisition and storage

Urine samples and demographic data were obtained from a urine specimen registry at Boston Children’s Hospital using an IRB approved protocol. For the adults, clean catch voided samples were obtained from six healthy adult males. All adult patients completed self-reported health questionnaires to determine eligibility. No patients had reported urologic or other major systemic abnormality. All infant urine samples were obtained via a sterile, 5 French (1.67 mm) pediatric feeding tube utilizing standardized protocols at Boston Children’s Hospital. The feeding tube is approximately 1/4 the diameter of a normal pediatric male urethra. This method is the current clinical gold standard for collecting a one-time urine sample in non-toilet trained pediatric patients, because it is atraumatic and sterile. Catheterized urine samples have the lowest risk of introduction of contamination from skin or skin flora bacteria, as opposed to other methods such as a bagged urine sample. Samples were obtained from six healthy infant males undergoing minor penile surgery. Detailed medical histories were obtained from the patient record to determine eligibility. All patients were full term babies, with no prenatal or postnatal abnormalities except for a minor distal penile cosmetic abnormality. Macroscopic urinalysis and culture were negative for all samples. Samples were centrifuged at 4000 × g for 60 minutes at 4°C to remove debris. Protein concentration was measured by spectrophotometry (BioRad, CA) and samples were divided into 100 μg aliquots and stored at −80°C for one time use. Table 1 details the demographic data of each individual per cohort.

Table 1.

The demographic information and number of proteins and peptides identified in each of the (A) adult and (B) infant samples. Cohort-wide numbers of unique proteins and peptides are also reported along with their % coefficient of variations (cv) and the mean ages of each cohort.

	Adult 1	Adult 2	Adult 3	Adult 4	Adult 5	Adult 6	Adult Total*	cv
Age (yrs)	23	31	33	36	32	35	31.67 (mean)	--
Proteins	547	435	609	521	599	629	936	12.9
Peptides	3912	2934	4582	3786	4775	5628	10170	21.8
Spectra	13840	9889	16333	15888	21725	25159	102834

	Infant 1	Infant 2	Infant 3	Infant 4	Infant 5	Infant 6	Infant Total*	cv
Age (yrs)	0.8	0.6	1 5	2.1	0.5	0.7	1.03 (mean)	--
Proteins	640	572	878	483	644	693	1356	20.3
Peptides	5625	3956	7745	4164	5340	6315	17569	25.5
Spectra	19063	12760	33066	19079	14298	15355	113621

Ethanol Precipitation

9 volumes of cold ethanol were added to each urine aliquot, centrifuged at 2500 × g for 25 minutes at 4°C. All of the supernatant except for ~200 μl was removed. The pellet was resuspended in the remaining supernatant, transferred to a 1.5 ml Eppendorf tube, and then dried in the vacuum centrifuge (Speedvac, Thermo Savant) at room temperature.

One dimensional gel electrophoresis (1-DE)

100 μg of protein were resuspended in 21 μl of 100 mM ammonium bicarbonate (NH₄HCO₃) and 8 μl 4X LDS sample buffer (Invitrogen). 1 μl of 1 M DTT was added prior to incubation for 50 minutes at 50°C to reduce the sample. Samples were alkylated with 2 μl of 40% acrylamide at room temperature for 10 minutes. Samples were heated at 70°C for 10 minutes and then all samples from one age group were loaded onto a single 4–12% Bis-Tris precast gel (Invitrogen). Gels were run at 100 V and stained overnight in colloidal blue as per manufacturer protocol (Invitrogen).

Each gel lane was cut into twenty-four similar sized pieces. Gel fractions were washed for 30 minutes in 200 mM NH₄HCO₃, pH 7.8 at 37°C. These fractions were then dehydrated by incubation for 30 minutes in 200 mM NH₄HCO₃, pH 7.8/ACN (4:6 v/v) at 37°C, followed by rehydration for 30 minutes in 50 mM NH4HCO3, pH 7.8 at 37°C. Following incubation in 100% acetonitrile for 2 minutes, 0.1 μg trypsin in 50 mM NH₄HCO₃, pH 7.8 was added to each sample. The proteolysis was performed overnight at 37°C. The supernatant was subsequently recovered into microcentrifuge tubes and a second peptide extraction from these gel pieces was carried out with 0.1% trifluoroacetic acid (TFA) in 60% ACN for 5 minutes. Peptide-containing liquid fractions were pooled, dried under vacuum and resuspended in 25 μl 0.5% formic acid/5% Acetonitrile.

LC-MS/MS

All LC-MS/MS analyses were performed on an LTQ-FT Ultra mass spectrometer (Thermo). The LC columns (15 cm × 100 μm ID) were packed in-house (Magic C18, 5 μm, 100 Å, Michrom BioResources, into PicoTips, New Objective, Woburn, MA). Samples were analyzed with a 60 minute linear gradient (0–35% acetonitrile with 0.2% formic acid) and data were acquired in a data dependent manner, in which MS/MS fragmentation is performed on the 6 most intense peaks of every full MS scan.

Database Searching and Validation

The 200 most intense fragment ions of each raw product ion spectrum were used for searches against the homo sapiens database (2012_05) using Protein Pilot. Default search settings were utilized. A fragment ion search tolerance of 0.8 Dalton (Da) was permitted. The peptide level score cutoff for each of the runs was adjusted to ensure a 1% false discovery rate throughout the experiments[10], and proteins were regrouped to ensure parsimony. All identified proteins required at least 2 or more unique peptides within in at least one of the samples studied.

Spectral Count Quantification

Spectral counts were used as a surrogate measure of protein abundances[11]. We implemented a data analysis method that accounts for variability between samples to normalize relative protein abundance. Briefly, the total number of tandem MS spectra matching peptides from a protein was divided by the total sum of the spectral count values in that sample. The normalized spectral counts for each protein were then compared across the infant and adult cohorts. Normalization is particularly important in urine due to a potentially high level of variation between individuals[12]. Statistically significant proteins were determined by QSPEC software, which uses a Bayesian statistical test to determine the significance of quantitative differences[13]. Results were filtered at a 1% FDR.

Gene Ontology Analysis

To investigate the cellular component and biological process of our urinary proteins, three lists were created: 1) infant subproteome and enriched proteins (742 total entries); 2) adult subproteome and enriched proteins (367 total entries); and 3) common subproteome without quantitative variations (478) total entries. These numbers are slightly different from the set of proteins which were uniquely identified below, as they also include cohort-enriched proteins. BiNGO software was used to find statistically over- and under-represented GO categories in our three urinary subproteome datasets[14]. The full UniprotKB database was utilized as a reference set. All GO terms that were significant within a protein list with p < 0.0001 (after Benjamini-Hochberg multiple testing correction) were selected as over or underrepresented and are reported in Supporting Information. Selected terms are presented in supplementary materials.

Results and Discussion

The urinary proteome can change with disease or because of normal physiologic variations. Normal variations are most likely due to diet, hydration, circadian rhythm, metabolic and catabolic processes, exercise, sex, disease and other environmental factors. Previous work on normal rat urine demonstrated that the urinary proteome undergoes significant changes from birth (0 days) to adulthood (> 30 days) [9]. During days 0 to 14 of life in the rat, proteins associated with cell proliferation and differentiation, cell adhesion, and tissue remodeling were prominent in the urine. These upregulated proteins may reflect normal systemic or urogenital track maturation, such as nephron development, which in the rodent (in contrast to humans) continues to occur after birth.

However, to date, no studies have specifically looked at the effect of age and maturation on the non-diseased human urinary proteome. In this study we performed an in-depth analysis of the normal infant urinary proteome to determine if age affects the normal proteome. We compared the urinary proteome of 6 healthy adult males (mean age: 31.7 years) to 6 healthy full-term infant males (mean age: 1.0 year). Better characterization of these two cohorts and the effect of age will improve our understanding of normal variation of the urinary proteome and future urine proteomic studies utilizing normal cohorts.

Identification of age-specific proteins

The SDS-PAGE patterns of urinary proteins from the infant and adult samples are shown in Figure 1 and suggest putative distinctive features in each proteome. Comparison of their proteomes by GeLC-MS demonstrated significant differences in the proteins identified in each age group. The number of identified proteins per sample is presented in Table 1. In total, 10,170 unique peptides were identified in the adult cohort that mapped to 936 non-redundant proteins. In the 6 infant samples, 17,569 unique peptides were identified, corresponding to 1,356 proteins. As shown in Figure 2A, we identified three datasets (adult, infant, and common) between the two age groups. A total of 708 proteins were commonly identified in both cohorts; 228 proteins were detected only in the adult samples; and 648 proteins were identified exclusively in infants in our study. The complete list of identified proteins and peptides are provided in Supporting Information.

Figure 1.

Urinary proteins of 6 adult (A1–A6) and 6 infant (I1–I6) samples separated by SDS-PAGE. The amount of protein loaded on each gel has been normalized to 30 μg.

Figure 2.

A. Venn diagram of the three urinary subproteomes in our study: adult subproteome, infant subproteome and common subproteome. The subset of commonly identified proteins which were quantitatively enriched in one cohort is also shown.

B. Fractions of adult, infant, and common subproteomes that were not identified in the healthy adult urinary proteomic literature. Sixty-six percent of the infant subproteome was not previously identified in previously published adult urinary proteomics studies.

To further curate the infant-specific dataset, we cross-validated these protein identifications with two other prominent contemporary proteomic studies [5, 6]. These studies interrogated the non-diseased adult urine proteome. When cross-validated with our data, 66% of the proteins identified in our infant-specific subproteome were not identified in these studies, whereas 47% of adult and only 10% of the common proteins were not identified in these studies (Figure 2B). Respectively, this corresponded to 430 infant, 71 common, and 108 Adult entries.

Although it is difficult to prove the absence of a protein in a proteomics experiment, several factors support the infant-specific status assigned in this study. First, a high degree of interpersonal variability supports the identification of unique proteins in a particular age cohort, as even a single identification in the opposite cohort would nullify a proteins status as “unique”. Second, although other studies have achieved comparable or greater depth of the urinary proteome, almost two thirds of the infant-specific proteins identified in this study were not identified in the other studies explored. Third, 64% of the cohort- specific proteins are identified in multiple individuals within a single cohort, further supporting their status as cohort-specific. Supplementary materials present data for the individual proteins separately for facile future reference. Overall, the median relative standard deviation was determined to be.

Considering these factors, the identification of infant-specific proteins posits compelling biological hypotheses for future studies. For example, additional studies on appropriately aged cohorts could identify proteins associated with puberty, or identify markers of normal maturation. Certainly, these data suggest that future studies of urinary biomarkers must be performed with appropriately age-matched controls.

Quantitation of the Common Urinary Proteome and Creation of Age-Specific Subproteomes

A large number of proteins were found in both adults and infants (708) and spectral counting analysis of these proteins indicated that certain common proteins exhibit higher levels of expression between the cohorts studied. The majority of these (478) demonstrated no significant difference in our spectral counting analysis (FDR = 1%). However, 94 proteins were enriched in infant urines and 136 proteins were enriched in adult urines after statistical testing and multiple hypothesis correction. All statistically significant proteins are presented in Table 2, along with their fold-changes.

Table 2.

Proteins identified with statistically significant quantitative values between the two cohorts. Protein accession, protein name, bayes factor, enriched cohort and fold change are listed.

Accession	Name	Bayes Factor	Enriched Cohort	Fold Change
Q7Z3B1	Neuronal growth regulator 1	8.3E+10	Ad	10.2
P11279	Lysosome-associated membrane glycoprotein 1	3.2E+07	Ad	9.4
P19961	Alpha-amylase 2B	7.1E+09	Ad	8.9
Q12794	Hyaluronidase-1	9.7E+03	Ad	7.2
O75594	Peptidoglycan recognition protein 1	9.9E+05	Ad	7.1
P05451	Lithostathine-1-alpha	8.7E+04	Ad	6.7
Q14315	Filamin-C	2.0E+03	Ad	6.4
P01859	Ig gamma-2 chain C region	1.5E+12	Ad	6.1
Q15904	V-type proton ATPase subunit S1	2.9E+03	Ad	6.1
Q5IJ48	Crumbs homolog 2	1.7E+02	Ad	6.0
B9A064	Immunoglobulin lambda-like polypeptide 5	2.3E+03	Ad	5.9
P0C0L5	Complement C4-B	3.6E+04	Ad	5.9
P01621	Ig kappa chain V-III region NG9 (Fragment)	1.5E+02	Ad	5.8
Q9BYE9	Cadherin-related family member 2	5.0E+09	Ad	5.7
Q5D862	Filaggrin-2	1.2E+05	Ad	5.6
Q6P531	Gamma-glutamyltransferase 6	8.5E+01	Ad	5.5
P04746	Pancreatic alpha-amylase	6.9E+09	Ad	5.0
P20138	Myeloid cell surface antigen CD33	5.8E+03	Ad	4.9
P19440	Gamma-glutamyltranspeptidase 1	1.1E+06	Ad	4.9
P01871	Ig mu chain C region	1.7E+03	Ad	4.8
P22792	Carboxypeptidase N subunit 2	2.2E+05	Ad	4.8
Q9BXP8	Pappalysin-2	5.7E+04	Ad	4.7
Q9H6B4	CXADR-like membrane protein	6.6E+03	Ad	4.6
Q92520	Protein FAM3C	1.8E+04	Ad	4.6
Q9Y2E5	Epididymis-specific alpha-mannosidase	1.1E+04	Ad	4.6
Q9HBB8	Cadherin-related family member 5	3.0E+07	Ad	4.5
P01619	Ig kappa chain V-III region B6	4.3E+02	Ad	4.5
P02788	Lactotransferrin	3.3E+03	Ad	4.4
Q86UD1	Out at first protein homolog	6.6E+02	Ad	4.3
P23142-4	Isoform C of Fibulin-1	1.2E+03	Ad	4.3
P00915	Carbonic anhydrase 1	3.6E+02	Ad	4.2
P31997	Carcinoembryonic antigen-related cell adhesion molecule 8	5.8E+06	Ad	4.1
Q8WZ42	Titin	9.3E+03	Ad	4.0
P39060	Collagen alpha-1(XVIII) chain	3.7E+03	Ad	4.0
P32004	Neural cell adhesion molecule L1	2.5E+01	Ad	4.0
P09603	Macrophage colony-stimulating factor 1	1.2E+10	Ad	3.8
P43121	Cell surface glycoprotein MUC18	4.4E+02	Ad	3.7
P19801	Amiloride-sensitive amine oxidase [copper-containing]	2.2E+05	Ad	3.7
O94919	Endonuclease domain-containing 1 protein	1.5E+06	Ad	3.7
Q15828	Cystatin-M	2.9E+03	Ad	3.6
Q7Z794	Keratin, type II cytoskeletal 1b	1.9E+02	Ad	3.4
P07858	Cathepsin B	1.2E+02	Ad	3.4
P30626	Sorcin	8.4E+01	Ad	3.4
Q9NPG4	Protocadherin-12	1.2E+02	Ad	3.4
P26992	Ciliary neurotrophic factor receptor subunit alpha	1.6E+04	Ad	3.4
Q00796	Sorbitol dehydrogenase	1.6E+03	Ad	3.2
P19652	Alpha-1-acid glycoprotein 2	2.8E+02	Ad	3.2
P03951	Coagulation factor XI	6.2E+01	Ad	3.2
P08185	Corticosteroid-binding globulin	1.4E+02	Ad	3.1
P12273	Prolactin-inducible protein	2.4E+02	Ad	3.1
Q9NR99	Matrix-remodeling-associated protein 5	1.2E+02	Ad	3.1
P00790	Pepsin A	1.9E+02	Ad	3.0
Q12860	Contactin-1	1.0E+11	Ad	3.0
P11597	Cholesteryl ester transfer protein	3.3E+02	Ad	2.9
Q9UNN8	Endothelial protein C receptor	2.1E+08	Ad	2.9
P30039	Phenazine biosynthesis-like domain-containing protein	1.8E+01	Ad	2.9
Q12913	Receptor-type tyrosine-protein phosphatase eta	1.8E+04	Ad	2.8
P41222	Prostaglandin-H2 D-isomerase	4.3E+17	Ad	2.7
P01764	Ig heavy chain V-III region VH26	9.8E+01	Ad	2.7
Q92673	Sortilin-related receptor	4.6E+08	Ad	2.6
Q96S96	Phosphatidylethanolamine-binding protein 4	8.0E+01	Ad	2.6
P05090	Apolipoprotein D	5.0E+16	Ad	2.6
O75144	ICOS ligand	6.9E+04	Ad	2.5
P02649	Apolipoprotein E	1.5E+04	Ad	2.5
P02750	Leucine-rich alpha-2-glycoprotein	2.7E+02	Ad	2.5
O14773	Tripeptidyl-peptidase 1	2.0E+05	Ad	2.4
P05362	Intercellular adhesion molecule 1	2.4E+01	Ad	2.4
P62158	Calmodulin	2.2E+01	Ad	2.4
Q9HAT2	Sialate O-acetylesterase	4.1E+01	Ad	2.4
Q86UN3	Reticulon-4 receptor-like 2	6.3E+01	Ad	2.4
P04180	Phosphatidylcholine-sterol acyltransferase	4.0E+01	Ad	2.4
P12821	Angiotensin-converting enzyme	9.7E+02	Ad	2.4
P01766	Ig heavy chain V-III region BRO	5.8E+02	Ad	2.4
Q8NBJ4	Golgi membrane protein 1	1.6E+06	Ad	2.3
Q7Z4W1	L-xylulose reductase	2.1E+01	Ad	2.3
P02763	Alpha-1-acid glycoprotein 1	6.1E+04	Ad	2.3
Q16651	Prostasin	2.2E+08	Ad	2.3
P08195	4F2 cell-surface antigen heavy chain	1.3E+07	Ad	2.3
P02533	Keratin, type I cytoskeletal 14	7.6E+01	Ad	2.2
P08571	Monocyte differentiation antigen CD14	2.5E+07	Ad	2.2
P15309	Prostatic acid phosphatase	6.7E+09	Ad	2.2
Q9UGT4	Sushi domain-containing protein 2	7.8E+04	Ad	2.2
P01834	Ig kappa chain C region	2.1E+18	Ad	2.1
Q07075	Glutamyl aminopeptidase	7.6E+01	Ad	2.1
P17900	Ganglioside GM2 activator	2.4E+01	Ad	2.0
P13647	Keratin, type II cytoskeletal 5	8.8E+04	Ad	2.0
P07996	Thrombospondin-1	1.3E+02	Ad	2.0
Q9NZP8	Complement C1r subcomponent-like protein	3.4E+02	Ad	2.0
O00241	Signal-regulatory protein beta-1	3.3E+02	Ad	2.0
P05543	Thyroxine-binding globulin	3.6E+05	Ad	2.0
P10153	Non-secretory ribonuclease	4.3E+08	Ad	1.9
Q16706	Alpha-mannosidase 2	1.3E+03	Ad	1.9
O75339	Cartilage intermediate layer protein 1	1.9E+01	Ad	1.9
P15586	N-acetylglucosamine-6-sulfatase	1.1E+04	Ad	1.9
P07998	Ribonuclease pancreatic	2.1E+01	Ad	1.8
P25311	Zinc-alpha-2-glycoprotein	1.0E+07	Ad	1.8
P43251	Biotinidase	1.4E+04	Ad	1.8
P08294	Extracellular superoxide dismutase [Cu-Zn]	1.3E+05	Ad	1.8
Q6GTX8	Leukocyte-associated immunoglobulin-like receptor 1	1.4E+03	Ad	1.8
Q9HB40	Retinoid-inducible serine carboxypeptidase	5.4E+01	Ad	1.7
Q8IV08	Phospholipase D3	1.3E+03	Ad	1.7
O95865	N(G), N(G)-dimethylarginine dimethylaminohydrolase 2	5.2E+02	Ad	1.7
O60494	Cubilin	7.6E+06	Ad	1.7
P16070	CD44 antigen	9.7E+05	Ad	1.6
Q9Y646	Plasma glutamate carboxypeptidase	1.3E+03	Ad	1.6
Q6UX06	Olfactomedin-4	2.6E+01	Ad	1.6
O00187	Mannan-binding lectin serine protease 2	6.1E+03	Ad	1.6
P02766	Transthyretin	6.4E+01	Ad	1.6
P04264	Keratin, type II cytoskeletal 1	6.3E+10	Ad	1.6
P04745	Alpha-amylase 1	7.3E+01	Ad	1.6
Q92820	Gamma-glutamyl hydrolase	3.5E+02	Ad	1.6
Q7Z7M0	Multiple epidermal growth factor-like domains protein 8	1.5E+03	Ad	1.6
O00391	Sulfhydryl oxidase 1	1.2E+03	Ad	1.6
P19835	Bile salt-activated lipase	5.8E+02	Ad	1.6
P55290	Cadherin-13	2.4E+05	Ad	1.5
Q03154	Aminoacylase-1	2.1E+02	Ad	1.5
P06280	Alpha-galactosidase A	4.7E+01	Ad	1.5
Q96DA0	Zymogen granule protein 16 homolog B	6.2E+02	Ad	1.5
P00450	Ceruloplasmin	3.9E+02	Ad	1.5
O95336	6-phosphogluconolactonase	6.9E+01	Ad	1.5
P05062	Fructose-bisphosphate aldolase B	3.2E+01	Ad	1.5
Q8WZ75	Roundabout homolog 4	5.5E+02	Ad	1.5
Q14393	Growth arrest-specific protein 6	3.0E+01	Ad	1.5
O00468	Agrin	6.3E+02	Ad	1.4
P01011	Alpha-1-antichymotrypsin	1.5E+04	Ad	1.4
P10619	Lysosomal protective protein	1.0E+02	Ad	1.4
P27487	Dipeptidyl peptidase 4	5.3E+01	Ad	1.4
P98164	Low-density lipoprotein receptor-related protein 2	7.4E+04	Ad	1.4
P35527	Keratin, type I cytoskeletal 9	3.1E+03	Ad	1.4
P16870	Carboxypeptidase E	9.6E+01	Ad	1.4
P08473	Neprilysin	4.3E+01	Ad	1.4
P16278	Beta-galactosidase	2.6E+01	Ad	1.4
P30530	Tyrosine-protein kinase receptor UFO	2.1E+02	Ad	1.4
P10253	Lysosomal alpha-glucosidase	7.0E+03	Ad	1.3
P05154	Plasma serine protease inhibitor	4.3E+01	Ad	1.2
P58107	Epiplakin	2.2E+01	Ad	1.0
Q99715	Collagen alpha-1(XII) chain	1.5E+24	Inf	11.2
P01133	Pro-epidermal growth factor	1.6E+08	Inf	9.8
P08238	Heat shock protein HSP 90-beta	3.0E+05	Inf	9.1
P14625	Endoplasmin	2.9E+06	Inf	8.5
P07900-2	Isoform 2 of Heat shock protein HSP 90-alpha	3.7E+05	Inf	7.6
P12429	Annexin A3	4.5E+04	Inf	6.9
P15428	15-hydroxyprostaglandin dehydrogenase [NAD+]	9.4E+02	Inf	6.5
P46940	Ras GTPase-activating-like protein IQGAP1	1.6E+05	Inf	5.6
P35237	Serpin B6	4.1E+02	Inf	5.6
P62805	Histone H4	2.2E+04	Inf	5.3
P15313	V-type proton ATPase subunit B, kidney isoform	1.1E+03	Inf	5.0
P38606	V-type proton ATPase catalytic subunit A	3.9E+03	Inf	5.0
P07384	Calpain-1 catalytic subunit	6.8E+01	Inf	4.9
P01034	Cystatin-C	2.5E+06	Inf	4.9
P21802	Fibroblast growth factor receptor 2	1.2E+03	Inf	4.8
Q9Y6W3	Calpain-7	5.3E+02	Inf	4.6
P68371	Tubulin beta-2C chain	1.4E+03	Inf	4.5
P01042	Kininogen-1	1.9E+03	Inf	4.5
P29323	Ephrin type-B receptor 2	8.5E+01	Inf	4.5
Q00610	Clathrin heavy chain 1	1.2E+03	Inf	4.4
P08138	Tumor necrosis factor receptor superfamily member 16	2.1E+03	Inf	4.4
O43852	Calumenin	8.4E+02	Inf	4.4
P13591	Neural cell adhesion molecule 1	1.0E+03	Inf	4.4
P48637	Glutathione synthetase	4.2E+02	Inf	4.4
O43175	D-3-phosphoglycerate dehydrogenase	5.1E+01	Inf	4.0
P25705	ATP synthase subunit alpha, mitochondrial	7.3E+02	Inf	4.0
014818	Proteasome subunit alpha type-7	9.7E+01	Inf	3.9
P62937	Peptidyl-prolyl cis-trans isomerase A	3.8E+08	Inf	3.9
Q05707	Collagen alpha-1(XIV) chain	5.5E+06	Inf	3.8
Q14204	Cytoplasmic dynein 1 heavy chain 1	9.1E+02	Inf	3.8
P05787	Keratin, type II cytoskeletal 8	1.5E+02	Inf	3.7
P08107	Heat shock 70 kDa protein 1A/1B	7.4E+02	Inf	3.5
P04196	Histidine-rich glycoprotein	3.9E+01	Inf	3.3
P12955	Xaa-Pro dipeptidase	6.6E+01	Inf	3.3
P50502	Hsc70-interacting protein	4.0E+01	Inf	3.2
P15121	Aldose reductase	5.7E+01	Inf	3.2
Q14624	Inter-alpha-trypsin inhibitor heavy chain H4	8.6E+07	Inf	3.2
P00338	L-lactate dehydrogenase A chain	5.5E+06	Inf	3.2
Q9Y5Z4	Heme-binding protein 2	3.4E+01	Inf	3.2
P31947	14-3-3 protein sigma	3.6E+01	Inf	3.0
Q8TF66-2	Isoform 2 of Leucine-rich repeat-containing protein 15	1.0E+06	Inf	2.9
Q8IUL8	Cartilage intermediate layer protein 2	1.0E+09	Inf	2.8
P11142	Heat shock cognate 71 kDa protein	4.1E+01	Inf	2.7
P13798	Acylamino-acid-releasing enzyme	4.4E+01	Inf	2.6
P10599	Thioredoxin	1.7E+01	Inf	2.6
P05783	Keratin, type I cytoskeletal 18	2.4E+01	Inf	2.6
P00966	Argininosuccinate synthase	6.0E+06	Inf	2.5
P13645	Keratin, type I cytoskeletal 10	3.8E+01	Inf	2.5
P13489	Ribonuclease inhibitor	2.7E+01	Inf	2.5
O75874	Isocitrate dehydrogenase [NADP] cytoplasmic	1.2E+06	Inf	2.4
P31431	Syndecan-4	3.3E+02	Inf	2.4
P27797	Calreticulin	1.9E+01	Inf	2.4
Q07507	Dermatopontin	5.2E+05	Inf	2.4
P00751	Complement factor B	2.9E+01	Inf	2.4
P04406	Glyceraldehyde-3-phosphate dehydrogenase	3.6E+04	Inf	2.3
P14618	Pyruvate kinase isozymes M1/M2	1.2E+07	Inf	2.3
O00299	Chloride intracellular channel protein 1	8.3E+01	Inf	2.3
Q15113	Procollagen C-endopeptidase enhancer 1	1.8E+03	Inf	2.2
Q9NZH0	G-protein coupled receptor family C group 5 member B	2.3E+05	Inf	2.1
P04075	Fructose-bisphosphate aldolase A	3.9E+01	Inf	2.1
P40121	Macrophage-capping protein	3.6E+01	Inf	2.1
Q96KP4	Cytosolic non-specific dipeptidase	2.7E+04	Inf	2.0
P02765	Alpha-2-HS-glycoprotein	5.9E+03	Inf	2.0
P54793	Arylsulfatase F	2.2E+01	Inf	2.0
P35579	Myosin-9	1.8E+02	Inf	2.0
Q13228	Selenium-binding protein 1	7.7E+04	Inf	2.0
P04792	Heat shock protein beta-1	1.8E+02	Inf	2.0
P09211	Glutathione S-transferase P	2.2E+05	Inf	1.9
P22105	Tenascin-X	2.3E+01	Inf	1.9
P02774	Vitamin D-binding protein	5.3E+04	Inf	1.9
P23528	Cofilin-1	1.4E+03	Inf	1.9
P60174	Triosephosphate isomerase	6.1E+03	Inf	1.9
P62258	14-3-3 protein epsilon	1.3E+03	Inf	1.8
Q9BRK5	45 kDa calcium-binding protein	2.5E+03	Inf	1.8
P00558	Phosphoglycerate kinase 1	3.4E+04	Inf	1.7
P06733	Alpha-enolase	9.2E+02	Inf	1.7
P63104	14-3-3 protein zeta/delta	3.5E+01	Inf	1.7
Q9BRK3	Matrix-remodeling-associated protein 8	1.5E+06	Inf	1.7
P07911	Uromodulin	2.0E+17	Inf	1.7
Q6UVK1	Chondroitin sulfate proteoglycan 4	8.5E+07	Inf	1.6
P12111	Collagen alpha-3(VI) chain	1.4E+03	Inf	1.6
Q5VW32	BRO1 domain-containing protein BROX	1.0E+04	Inf	1.6
Q99816	Tumor susceptibility gene 101 protein	5.2E+02	Inf	1.6
P53990	IST1 homolog	7.0E+03	Inf	1.6
P15311	Ezrin	2.1E+01	Inf	1.5
P09467	Fructose-1,6-bisphosphatase 1	4.0E+01	Inf	1.5
P04004	Vitronectin	7.8E+02	Inf	1.5
P07195	L-lactate dehydrogenase B chain	8.1E+02	Inf	1.5
P02768	Serum albumin	1.7E+17	Inf	1.5
P30086	Phosphatidylethanolamine-binding protein 1	1.8E+01	Inf	1.4
Q8NFZ8	Cell adhesion molecule 4	1.7E+04	Inf	1.4
Q99497	Protein DJ-1	1.8E+01	Inf	1.4
P06870	Kallikrein-1	1.3E+02	Inf	1.4
P12109	Collagen alpha-1(VI) chain	1.1E+05	Inf	1.3

The adult urine was enriched from 3 to 6-fold in several immune system proteins, including Ig gamma-2 chain C, Ig lambda-like polypeptide 5, macrophage colony-stimulating factor 1 and Ig mu chain C. Other adult urine-enriched proteins include proteins originating from the prostate (e.g. prostatic acid phosphatase and prostaglandin-H2 D-isomerase), selected adhesion proteins (Neuronal growth regulator 1, contactin-1, cadherin-2, cadherin-13 and attractin), and some selected circulating proteins from blood (thyroxin binding globulin, alpha-1 antichymotrypsin and ceruloplasmin). In particular, it is not surprising that the male infant has lower concentrations of proteins derived from the prostate considering the lack of maturation of the prostate at one year of age. However, it is too simplistic to say that all proteins which may be synthesized in the prostate are upregulated in adults. For example, epidermal growth factor (EGF) is expressed in the prostate, (among other tissues) however this protein is enriched in the infant cohort approximately 10 fold. This change we are observing in urine is potentially due to EGF being differentially expressed in other maturing tissues, such as the kidney.

The increase of certain adhesion proteins in adults is complicated by the fact that other adhesion proteins are increased in the infant, including dermatopontin (2.4 fold) tenascin-X (1.9 fold), vitronectin (1.5 fold) and cell adhesion molecule 4 (1.4 fold). Previous studies have shown that tenascin-X is present in higher levels in infant serum than adult serum [15]. Proteolysis of vitronectin yields somatomedin-B, a potent growth factor, which interacts with glycosaminoglycans and proteoglycans[16]. Therefore an infant-specific increase of vitronectin and EGF is possibly explained by an increase in cellular growth among this cohort.

This claim is also supported by the increase in several metabolic proteins in infant urine. These include pyruvate kinase isozymes M1/M2, Alpha-enolase, Fructose-biphosphate aldolase A phosphoglycerate kinase 1, fructose-1,6-biphosphate 1, Glyceraldehyde-3-phosphate dehydrogenase, and triosephosphate isomoerase. We postulate that these proteins are enriched in the urine due to increased metabolic activity in the infant cohort.

Cellular and functional characterization of the urinary proteomes based on age

GO characterization of the subproteomes was also performed. Three lists were generated from aggregate qualitative and quantitative proteomic analysis: Infant-specific and enriched proteins, Adult-specific and enriched proteins, and common quantitatively unchanged proteins.

The common and adult subproteomes were enriched in proteins from the extracellular region and plasma membrane, which was in agreement with earlier studies [5, 6]. In contrast, the infant urine-specific subproteome was associated with an increase in intracellular proteins.

GO annotations of the subproteomes also identified functionally enriched categories. Infant urine was enriched in cellular metabolic processes, suggesting an increase in cellular turnover and metabolism. A total of 31 monosaccharide metabolic process proteins were identified in the infant specific or enriched subproteome, while only 8 were expected based on the overall number in the proteome. Taken in concert with the quantitative enrichment of glucose catabolic proteins, a consistent pattern emerges, supporting a proposed increase in metabolism and cell cycle turnover in infants that is reflected in the urinary proteome.

In contrast to the infant, adult urines were enriched in several distinct functional categories. These included peptidase activity (31 entries, 10 expected) polysaccharide binding (14 entries, 2.5 expected), and iron ion homeostasis (6 entries, 0.6 expected). Both the common and the adult subproteomes were functionally enriched in innate and humoral immune response proteins.

To our knowledge, anti-bacterial proteins identified in our data include uromodulin (1.6-fold increase in the infant), lactoferrin (increase in the adult), and neutrophil-assosciated lipocalin (no quantitative change). Also, to our knowledge, the anti-lithiatic proteins include osteopontin (no change), bikunin (a chain of AMBP, no change), prothrombin (no change), Inter-alpha-trypsin inhibitor (no change for H1, H2, H5, H4 up in infants) and calgranulin (S100 proteins, also no change). This suggests that there are negligible differences in these proteins between these cohorts overall, although focused studies may potentially be useful to more fully explore this finding.

Protein Expression Origins

Ultimately, proteins arrive in the urine as a result of complex and varied processes. For example, variations in the urinary proteome could potentially result from an immaturity of infant glomerular and proximal tubular function. Although the pathway of a specific protein into the urine is not always definitive (e.g. blood filtration, renal specific excretion, or lower urinary tract excretion), specific proteins of known provenance are present at similar levels among each cohort. These include abundant circulating plasma proteins, namely Alpha-1-antitrypsin, Complement factor I, hemoglobin, and several others. As these proteins are abundant in plasma and present in each cohort the postulated mechanism of their arrival (e.g. incomplete plasma reabsorption) is likely occurring similarly in each cohort. Furthermore, an examination of the molecular weight distributions of identified proteins (a metric for protein reabsorption) suggests this is the case as there is no molecular weight bias observed between the two cohorts (Supporting Information).

Intracellular proteins are known to be transported into urine, in part, via the formation of urinary exosomes[17]. If there were cohort-specific differences in exosome formation, one might postulate that this is a potential route for intracellular proteins to be differentially expressed into infant urine. To explore possible differences in this regard, our data was compared to an online database, which was formed by the results of two proteomic exosome studies[17, 18]. Of our full set of 1584 proteins, 606 were also identified in the exosome database. Adult proteins had a 46% and infant proteins had a 41% overlap with the exosome data, suggesting there is a negligible qualitative difference for exosome expression globally. Quantitative results also showed negligible differences in exosomal proteins. A total of 354 proteins were identified in both cohorts and also in the exosomal data. Of these, 56 were enriched in the adult and 62 were enriched in the infant, while the majority of proteins (236) were quantitatively unchanged, suggesting very little global quantitative variation of exosomal proteins. Considering our sample preparation methods were not optimized for a study of exosomes[19, 20], an exosome-centric pediatric urinary proteomic study could potentially lend insight into the specific role exosomes may have with regard to the inter-cohort differences observed in the pediatric urinary proteome. Furthermore, as urinary exosome data continues to expand, a clearer trend may emerge between these two cohorts. However, apart from functional enrichment differences in the subproteomes, no obvious systemic expression patterns emerge to explain the varied protein expression differences between these cohorts. Rather, the data presented may best be utilized as a reference for future clinical proteomic biomarker studies. Clearly, these data underscore the importance of age-matched case and control cohorts in urinary proteomic analyses and shed light on important differences in select biological processes between these cohorts.

Spectral counting is only one possible quantitative proteomics methodology. Other MS approaches using internal standards or isotopic labeling could have also been used. These approaches may achieve somewhat lower variances. However, the isotopic labeling reagents add significant financial costs and can potentially introduce systematic bias due to differences in labeling efficiency. In addition, these approaches will not work when the proteomes being compared are qualitatively different, i.e. identify proteins exclusive to one cohort.

Conclusion

This is the first in-depth study of the normal male infant urinary proteome. By comparing it to the urinary proteome of healthy adults, we demonstrated that the male urinary proteome is highly age dependent. We further qualitatively identified a common urinary subproteome and quantitatively determined relative changes in protein abundances between these cohorts. Analysis by GO annotation demonstrated disparate annotations within each age dependent urinary proteome, and identified changes in the enriched subcellular and functional categories in the cohorts. These proteins are involved in compelling biological processes, including metabolism and growth. We hypothesize that the observed proteomic differences reflect differing growth and metabolic states between the adult and infant cohorts. Two known routes of urinary protein expression (incomplete plasma protein reabsorption and exosome formation) showed no systemic differences. The age-specific differences between these proteomes further highlight the importance of understanding the variation among normal, i.e. nondiseased samples to better understand the urinary proteome. In particular, these findings will provide valuable context for future clinical urinary proteomic studies.

Highlights.

• We analyze 12 adult and infant urine samples to >1500 proteins using MS.

• Using spectral counting, we determine over 200 quantitatively varying proteins.

• We identify functional differences associated with each cohort.

• Differences include metabolism, cellular growth, immune response and adhesion.

• These findings are highly relevant for future clinical work on urinary proteomics.

List of Abbreviations

LTQ-FT

Linear Trap Quadrupole Fourier Transform

GeLC-MS

Gel electrophoresis liquid chromatography mass spectrometry

FDR

False Discovery Rate

GO

Gene Ontology

BiNGO

A Biological Network Gene Ontology tool