Comparison of protein capture from a human cancer cell line by genomic G-quadruplex DNA sequences toward aptamer discovery.

A genome-inspired route to aptamer discovery that expands the sequence space beyond that available in traditional, combinatorial selection approaches is investigated for discovery of DNA-protein interactions in cancer. These interactions could then serve as the basis for new DNA aptamers to cancer-related proteins. The genome-inspired approach uses specific DNA sequences from the human genome to capture proteins from biological protein pools. The use of naturally occurring DNA sequences takes advantage of biological evolution of DNA sequences that bind to specific proteins to perform biological functions. Linking aptamer discovery to nature increa`ses the chances of uncovering protein-DNA affinity binding interactions that have biological significance as well as analytical utility. Here, the focus is on genomic, G-rich sequences that can form G-quadruplex (G4) structures. These structures are underrepresented in combinatorial libraries used for conventional aptamer selection. Additionally, G4-forming sequences are prone to inefficient PCR amplification, further biasing aptamer selection away from these structures. Nature provides a large diversity of G4-forming sequences throughout the human genome. They are prevalent in gene promoter regions, especially in oncogene promoters, and are therefore promising candidates for aptamers to regulatory proteins in cancer. The present work investigates protein capture from nuclear and cytoplasmic extracts of the breast cancer cell line MDA-MB-468 by G4-forming sequences from the CMYC, RB, and VEGF gene promoters. The studies included the effects of modifications of the VEGF sequence on the selectivity of protein capture, from which we identified promising aptamer candidates, subject to further refinement, to the proteins nucleolin and RPL19, both of which play important regulatory functions related to cancer.

Mechanism of sequence-based separation of single-stranded DNA in capillary zone electrophoresis.

Separation of DNA by length using CGE is a mature field. Separation of DNA by sequence, in contrast, is a more difficult problem. Existing techniques generally rely upon changes in intrinsic or induced differences in conformation. Previous work in our group showed that sets of ssDNA of the same length differing in sequence by as little as a single base could be separated by CZE using simple buffers at high ionic strength. Here, we explore the basis of the separation using circular dichroism spectroscopy, fluorescence anisotropy, and small angle X-ray scattering. The results reveal sequence-dependent differences among the same length strands, but the trends in the differences are not correlated to the migration order of the strands in the CZE separation. They also indicate that the separation is based on intrinsic differences among the strands that do not change with increasing ionic strength; rather, increasing ionic strength has a greater effect on electroosmotic mobility in the normal direction than on electrophoretic mobility of the strands in the reverse direction. This increases the migration time of the strands in the normal direction, allowing more time for the same-length strands to be teased apart based on very small differences in the intrinsic properties of the strands of different sequence. Regression analysis was used to model the intrinsic differences among DNA strands in order to gain insight into the relationship between mobility and sequence that underlies the separation.

Nucleotide Selectivity in Abiotic RNA Polymerization Reactions.

In order to establish an RNA world on early Earth, the nucleotides must form polymers through chemical rather than biochemical reactions. The polymerization products must be long enough to perform catalytic functions, including self-replication, and to preserve genetic information. These functions depend not only on the length of the polymers, but also on their sequences. To date, studies of abiotic RNA polymerization generally have focused on routes to polymerization of a single nucleotide and lengths of the homopolymer products. Less work has been done the selectivity of the reaction toward incorporation of some nucleotides over others in nucleotide mixtures. Such information is an essential step toward understanding the chemical evolution of RNA. To address this question, in the present work RNA polymerization reactions were performed in the presence of montmorillonite clay catalyst. The nucleotides included the monophosphates of adenosine, cytosine, guanosine, uridine and inosine. Experiments included reactions of mixtures of an imidazole-activated nucleotide (ImpX) with one or more unactivated nucleotides (XMP), of two or more ImpX, and of XMP that were activated in situ in the polymerization reaction itself. The reaction products were analyzed using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) to identify the lengths and nucleotide compositions of the polymerization products. The results show that the extent of polymerization, the degree of heteropolymerization vs. homopolymerization, and the composition of the polymeric products all vary among the different nucleotides and depend upon which nucleotides and how many different nucleotides are present in the mixture.

Sequence-based separation of single-stranded DNA at high salt concentrations in capillary zone electrophoresis.

DNA separation by fragment length can be readily achieved using sieving gels in electrophoresis. Separation by sequence has not been as simple, generally requiring adequate differences in native or induced conformation between single or hybridized strands or differences in thermal or chemical stability of hybridized strands. Previously, it was shown that four single-stranded DNA (ssDNA) 76-mers that differ by only a few A-G substitutions could be separated based solely on sequence by adding guanosine-5'-monophosphate to the running buffer in capillary zone electrophoresis (CZE). The separation was attributed to interactions of the ssDNA with self-assembled guanine-tetrad structures; however, subsequent studies of an expanded set of ten 76-mers showed that the separation was a more general phenomenon that occurred at high salt concentrations. With the long-term goal of using experimental and computational methods to provide insight into the basis of the separation, a set of ssDNA 15-mers was designed including a poly(dT) 15-mer and nine variants. Separations were performed using fluorescent-labeled ssDNA in CZE with laser-induced fluorescence detection. Results show that separation improves with increasing buffer concentration and decreasing temperature, due at least in part to longer separation times. Migration times increase with increasing purine content, with A having a much larger effect that G. Circular dichroism spectra of the mixtures of the strands suggest that the separation is not due to changes in conformation of the ssDNA at high salt concentrations.

In situ imidazole activation of ribonucleotides for abiotic RNA oligomerization reactions.

The hypothesis that RNA played a significant role in the origin of life requires effective and efficient abiotic pathways to produce RNA oligomers. The most successful abiotic oligomerization reactions to date have utilized high-energy, modified, or pre-activated ribonucleotides to generate strands of RNA up to 50-mers in length. In spite of their success, these modifications and pre-activation reactions significantly alter the ribonucleotides in ways that are highly unlikely to have occurred on a prebiotic Earth. This research seeks to address this problem by exploring an aqueous based method for activating the canonical ribonucleotides in situ using 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and imidazole. The reactions were run with and without a montmorillonite clay catalyst and compared to reactions that used ribonucleotides that were pre-activated with imidazole. The effects of pH and ribonucleotide concentration were also investigated. The results demonstrate the ability of in situ activation of ribonucleotides to generate linear RNA oligomers in solution, providing an alternative route to produce RNA for use in prebiotic Earth scenarios.

Sequence-based separation of single-stranded DNA using nucleotides in capillary electrophoresis: focus on phosphate.

DNA analysis has widespread applicability in biology, medicine, biotechnology, and forensics. DNA separation by length is readily achieved using sieving gels in electrophoresis. Separation by sequence is less simple, generally requiring adequate differences in native or induced conformation or differences in thermal or chemical stability of the strands that are hybridized prior to measurement. We previously demonstrated separation of four single-stranded DNA 76-mers that differ by only a few A-G substitutions based solely on sequence using guanosine-5'-monophosphate (GMP) in the running buffer. We attributed separation to the unique self-assembly of GMP to form higher order structures. Here, we examine an expanded set of 76-mers designed to probe the mechanism of the separation and effects of experimental conditions. We were surprised to find that other ribonucleotides achieved the similar separation to GMP, and that some separation was achieved using sodium phosphate instead of GMP. Potassium phosphate achieved almost as good separations as the ribonucleotides. This suggests that the separation medium provides a physicochemical environment for the DNA that effects strand migration in a sequence-selective manner. Further investigation is needed to determine whether the mechanism involves specific interactions between the phosphates and the DNA strands or is a result of other properties of the separation medium. Phosphate generally has been avoided in DNA separations by capillary gel electrophoresis because its high ionic strength exacerbates Joule heating. Our results suggest that phosphate compounds should be examined for separation of DNA based on sequence.

Potential pitfalls in MALDI-TOF MS analysis of abiotically synthesized RNA oligonucleotides.

Demonstration of the abiotic polymerization of ribonucleotides under conditions consistent with conditions that may have existed on the prebiotic Earth is an important goal in "RNA world" research. Recent reports of abiotic RNA polymerization with and without catalysis rely on techniques such as HPLC, gel electrophoresis, and MALDI-TOF MS to analyze the reaction products. It is essential to understand the limitations of these techniques in order to accurately interpret the results of these analyses. In particular, techniques that rely on mass for peak identification may not be able to distinguish between a single, linear RNA oligomer and stable aggregates of smaller linear and/or cyclic RNA molecules. In the case of MALDI-TOF MS, additional complications may arise from formation of salt adducts and MALDI matrix complexes. This is especially true for abiotic RNA polymerization reactions because the concentration of longer RNA chains can be quite low and RNA, as a polyelectrolyte, is highly susceptible to adduct formation and aggregation. Here we focus on MALDI-TOF MS analysis of abiotic polymerization products of imidazole-activated AMP in the presence and absence of montmorillonite clay as a catalyst. A low molecular weight oligonucleotide standard designed for use in MALDI-TOF MS and a 3'-5' polyadenosine monophosphate reference standard were also run for comparison and calibration. Clay-catalyzed reaction products of activated GMP and UMP were also examined. The results illustrate the ambiguities associated with assignment of m/z values in MALDI mass spectra and the need for accurate calibration of mass spectra and careful sample preparation to minimize the formation of adducts and other complications arising from the MALDI process.

G-quadruplex guanosine gels and single walled carbon nanotubes.

Solubilization of single walled carbon nanotubes (SWNTs) in aqueous gel phases formed by reversible, G-quadruplex self-assembly of guanosine monophosphate (GMP) alone or with guanosine (Guo) is described. Unlike other media and methods for aqueous solubilization of SWNTs, the guanosine gels ("G-gels") are found to readily disperse high (>mg/mL) concentrations of individual rather than bundled SWNTs. SWNT dispersions in GMP alone precipitate in several hours and re-form upon shaking; however, dispersions in the binary GMP/Guo gels are indefinitely stable. Increasing GMP or KCl concentration in the binary gels increased the relative abundance of large diameter and semi-conducting SWNTs. Different gel compositions also displayed different selectivities toward SWNTs of different chiralities. These results indicate a strong connection between the self-assembled G-gels and the dimensions and structures of the SWNTs that they solubilize.

Capture and identification of proteins that bind to a GGA-rich sequence from the ERBB2 gene promoter region.

The ERBB2 gene (HER2/neu) is overexpressed in many human breast cancers. It is an important therapeutic target and its product protein is a key biomarker for breast cancer. A 28-bp GGA repeat sequence (Pu28-mer) in the nuclease hypersensitive site of the ERBB2 promoter region may play an important role in the regulation of ERBB2 transcription, possibly involving the formation of a G-quadruplex. In order to investigate this possibility, an affinity MALDI-MS approach was used for in vitro protein capture from nuclear extracts from cultured MCF-7 and BT-474 cancer cells at Pu28-mer and control oligonucleotide-modified surfaces. Captured proteins from MCF-7 cells were analyzed by LC-MS/MS. Based on these results, Western blot was then used to interrogate captured proteins from both MCF-7 and the Her-2/neu-positive BT-474 cells. Results support the formation of a G-quadruplex by Pu28-mer, indicated by circular dichroism spectroscopy, that selectively captures transcription factors including Ku70, Ku80, PURA, nucleolin, and hnRNP K. Chromatin immunoprecipitation confirmed binding of Ku70, Ku80, PURA, and nucleolin to ERBB2 promoter in the live BT-474 cells. These findings may lead to a better understanding of the role of non-duplex DNA structures in gene regulation and provide a more complete picture of the regulation of ErbB2 expression in breast cancer. The results also provide a blueprint for development of "genome-inspired" aptamers based on the Pu28-mer sequence for in vitro and in vivo detection of proteins related to regulation of ERBB2 gene expression and breast cancer.

Chiral selectivity of guanosine media in capillary electrophoresis.

Gels formed by self-association of monomeric guanosine compounds join numerous other agents such as cyclodextrins, crown ethers, chiral surfactants, antibiotics, proteins, and polysaccharides for chiral separations. Guanosine gels (G-gels) are self-assembled networks of hydrogen-bonded tetrads formed by guanosine nucleotides and their derivatives. The tetrads stack upon themselves to form columnar, helical aggregates that are stabilized by π-π interactions and centrally located cations. Previous work showed the effectiveness of G-gels formed by guanosine-5'-monophophate for separation of the enantiomers of the cationic drug propranolol using capillary electrophoresis. Subsequently, it was found that not all chiral compounds could be resolved into their enantiomers, leading us to investigate in this work the structural features that appear to be correlated to enantiomerically selective interactions of chiral compounds with G-gels. For those compounds (anionic 1,1'-binaphthyl-2,2'-diyl hydrogen phosphate and zwitterionic tryptophan) for which enantiomeric resolution was achieved, the effects of experimental conditions and G-gel composition were examined. For other compounds with no net charge (hydrobenzoin and zwitterionic amino acids and derivatives), the migration times were used as an indicator of the extent of interaction with the G-gel run buffer. It was found that the extent of interaction alone does not determine the chiral selectivity of the G-gel, indicating that the mechanism of chiral separation involves particular structural characteristics of the chiral compounds.

Incorporation of guanosine gels into sieving matrices for length- and sequence-based separation of DNA in capillary electrophoresis.

Sieving gels are used in capillary gel electrophoresis to resolve DNA strands of different lengths. For complex samples, however, such as those encountered in metagenomic analysis of microbial communities or biofilms, length-based separation may mask the true genetic diversity of the community since different organisms may contribute same-length DNA with different sequences. There is a need, therefore, for DNA separations based on both the length and sequence. Previous work has demonstrated the ability of guanosine gels (G-gels) to separate four single-stranded DNA 76-mers that differ by only a few A/G base substitutions. The goal of the present work is to determine whether G-gels could be combined with commercial sieving gels in order to simultaneously separate DNA based on both length and sequence. The results are given for the four 76-mers and for a standard dsDNA ladder. Commercial sieving gels were used alone and in combination with G-gels. For the 76-mers, the combined medium was less efficient than the G-gel alone but was able to achieve partial resolution. The combined medium was at least as effective as the sieving gel alone at resolving the denatured DNA ladder and showed indications of sequence-based resolution as well, as supported by MALDI-MS. The results show that the combined sieving gel/G-gel medium retains the selectivity of the individual media, providing a promising approach to simultaneous length- and sequence-based DNA separation for metagenomic analysis of complex systems.

Association of insulin-like growth factor 2 with the insulin-linked polymorphic region in cultured fetal thymus cells.

The insulin-linked polymorphic region (ILPR) is a regulatory sequence in the promoter region upstream of the human insulin gene and is widely recognized as a locus of type 1 diabetes susceptibility. Polymorphism of the ILPR sequence can affect expression of both insulin and the adjacent insulin-like growth factor 2 (IGF-2) gene. Several ILPR variants form G-quadruplex DNA structures in vitro that exhibit affinity binding to insulin and IGF-2. It has been suggested that the ILPR may form G-quadruplexes in vivo as well, raising the possibility that insulin and IGF-2 may bind to these structures in the ILPR in chromatin of live cells. This work establishes the presence of IGF-2 in the nucleus of cells cultured from human fetal thymus and its association with the ILPR in the chromatin of these cells. In vitro experiments support the involvement of G-quadruplex DNA in the binding interaction.

A genome-inspired DNA ligand for the affinity capture of insulin and insulin-like growth factor-2.

The insulin-linked polymorphic region (ILPR) of the human insulin gene contains tandem repeats of similar G-rich sequences, some of which form intramolecular G-quadruplex structures in vitro. Previous work showed affinity binding of insulin to an intramolecular G-quadruplex formed by ILPR variant a. Here, we report on interactions of insulin and the highly homologous insulin-like growth factor-2 (IGF-2) with ILPR variants a, h, and i. Circular dichroism indicated intramolecular G-quadruplex formation for variants a and h. Affinity MALDI MS and surface plasmon resonance were used to compare protein capture and binding strengths. Insulin and IGF-2 exhibited high binding affinity for variants a and h but not i, indicating the involvement of intramolecular G-quadruplexes. Interaction between insulin and variant a was unique in the appearance of two binding interactions with K(D) approximately 10(-13) M and K(D) approximately 10(-7) M, which was not observed for insulin with variant h (K(D) approximately 10(-8) M) or IGF-2 with either variant (K(D)s approximately 10(-9) M). The results provide a basis for the design of DNA binding ligands for insulin and IGF-2 and support a new approach to discovery of DNA affinity binding ligands based on genome-inspired sequences rather than the traditional combinatorial selection route to aptamer discovery.

Tunable thermoassociation of binary guanosine gels.

It is well-known that aqueous solutions of individual guanosine compounds can form gels through reversible self-assembly. Typically, gelation is favored at low temperature and acidic pH. We have discovered that binary mixtures of 5'-guanosine monophosphate (GMP) and guanosine (Guo) can form stable gels at neutral pH over a temperature range that can be tuned by varying the relative proportions of the hydrophobic Guo and the hydrophilic GMP in the mixture. Gelation was studied over the temperature range of 5-40 degrees C or 60 degrees C at pH 7.2 using visual detection, circular dichroism (CD) spectroscopy, and CD thermal melt experiments. Solutions with high GMP/Guo ratios behaved similar to solutions of GMP alone while solutions with low GMP/Guo formed firm gels across the entire temperature range. Most interesting were solutions between these two extremes, which were found to exhibit thermoassociative behavior; these solutions are liquid at refrigerator temperature and undergo sharp transitions to a gel only at higher temperatures. Increasing the GMP/Guo ratio and increasing the total concentration of guanosine compounds shifted the onset of gelation to higher temperatures (ranging from 20 to 40 degrees C), narrowed the temperature range of the gel phase, and sharpened the reversible phase transitions. The combination of self-assembly, reversibility, and tunability over biologically relevant temperature ranges and pH offers exciting possibilities for these simple and inexpensive materials in medical, biological, analytical, and nanotechnological applications.

A genome-inspired, reverse selection approach to aptamer discovery.

Limitations of Systematic Evolution of Ligands by Exponential Enrichment (SELEX) and related methods that depend upon combinatorial oligonucleotide libraries have hindered progress in this area. Our laboratory has introduced a new approach to aptamer discovery that uses oligonucleotides with sequences drawn from the human genome to capture proteins from biological samples. Specifically, we have focused on capture of proteins in nuclear extracts from human cell lines using G-quadruplex (G4) forming genomic sequences. Previous studies identified capture of several proteins both in vitro and in live cells by the Pu28-mer sequence from the ERBB2 promoter region. Here we provide a more comprehensive study of protein capture from BT474 and MCF7 human breast cancer cells using G4-forming sequences from the CMYC, RB, VEGF and ERBB2 human oncogene promoter regions. Mass spectrometric analysis and Western blot analysis of protein capture at oligonucleotide-modified surfaces revealed capture of nucleolin by all three of the oligonucleotides in BT474 and MCF7 cells, and also of ribosomal protein L19 (RPL19) in BT474 cells. Chromatin immunoprecipitation (ChIP) analysis confirmed the interaction of nucleolin with all three promoter sequences in MCF7 cells and with RB in BT474 cells. ChIP also revealed interactions of RPL19 with CMYC in BT474 cells and of both RPL19 and ribosomal protein L14 (RPL14) with ERBB2 in BT474 cells. These results offer the basis for development of new aptamers based on the G4 sequences from the CMYC, RB, VEGF, and ERBB2 promoters toward proteins including nucleolin, RPL19 and RPL14. These interactions also may have biological and therapeutic significance.

Aptamer stationary phase for protein capture in affinity capillary chromatography.

The thrombin-binding DNA aptamer was used with thrombin as a model system to investigate protein capture using aptamer stationary phases in affinity capillary chromatography. The aptamer was covalently attached to the inner surface of a bare fused-silica glass capillary to serve as the stationary phase. Proteins were loaded onto the capillary via an applied pressure. The capillary was then washed to remove unbound and non-specifically associated proteins. Finally, the bound protein was released and eluted using 20 mM Tris buffer containing 8 M urea, pH 7.3, at 50 degrees C. Eluate was collected after each step (load, wash and elute) and relative amounts of protein each were compared using fluorescence spectroscopy. The identity of the protein in the collections was confirmed using matrix assisted laser desorption ionization-time-of-flight (MALDI-TOF) mass spectrometry. The experiment was repeated for thrombin on a bare (unmodified) capillary and a capillary coated with a scrambled-sequence, non-G-quartet forming oligonucleotide that does not bind with thrombin. The results show that the aptamer stationary phase captures approximately three times as much thrombin as the control columns. The experiment was also repeated using human serum albumin (HSA) alone and in an equimolar mixture with thrombin. HSA was not retained on the aptamer capillary, nor did it affect the capture of thrombin from the mixture.

Multiplex single strand conformation polymorphism analysis by capillary electrophoresis with on-the-fly fluorescence lifetime detection.

This paper describes the use of on-the-fly fluorescence lifetime detection (OFLD) for multiplex single strand conformation polymorphism (SSCP) analysis by capillary electrophoresis (CE). The dye labels studied for multiplex SSCP-OFLD-CE analyses included RG, NBD, and BODIPY-FL. The dyes were first investigated for a model system of "Wild Type" and "Mutant" 43-base fragments designed to vary by a single A/T substitution. Two dye pairs, BODIPY-FL/ RG and BODIPY-FL/NBD, were then used to detect the G20210A mutation in the human prothrombin gene. Mobility correction was required for the BODIPY-FL/RG system. Three "blind" analyses were performed of three mixtures that combined a control fragment (wild type-BODIPY-FL) with two "unknown" fragments selected among four possibilities (wild type or mutant labeled with NBD or RG). In each multiplex analysis, the "origin" of the unknown fragments was correctly identified on the basis of fluorescence lifetime of the dye label and the presence or absence of the mutation was correctly determined on the basis of conformation-induced differences in migration time.

RNA Oligomerization in Laboratory Analogues of Alkaline Hydrothermal Vent Systems.

Discovering pathways leading to long-chain RNA formation under feasible prebiotic conditions is an essential step toward demonstrating the viability of the RNA World hypothesis. Intensive research efforts have provided evidence of RNA oligomerization by using circular ribonucleotides, imidazole-activated ribonucleotides with montmorillonite catalyst, and ribonucleotides in the presence of lipids. Additionally, mineral surfaces such as borates, apatite, and calcite have been shown to catalyze the formation of small organic compounds from inorganic precursors (Cleaves, 2008 ), pointing to possible geological sites for the origins of life. Indeed, the catalytic properties of these particular minerals provide compelling evidence for alkaline hydrothermal vents as a potential site for the origins of life since, at these vents, large metal-rich chimney structures can form that have been shown to be energetically favorable to diverse forms of life. Here, we test the ability of iron- and sulfur-rich chimneys to support RNA oligomerization reactions using imidazole-activated and non-activated ribonucleotides. The chimneys were synthesized in the laboratory in aqueous "ocean" solutions under conditions consistent with current understanding of early Earth. Effects of elemental composition, pH, inclusion of catalytic montmorillonite clay, doping of chimneys with small organic compounds, and in situ ribonucleotide activation on RNA polymerization were investigated. These experiments, under certain conditions, showed successful dimerization by using unmodified ribonucleotides, with the generation of RNA oligomers up to 4 units in length when imidazole-activated ribonucleotides were used instead. Elemental analysis of the chimney precipitates and the reaction solutions showed that most of the metal cations that were determined were preferentially partitioned into the chimneys.

Capillary electrochromatographic separation of bovine milk proteins using a G-quartet DNA stationary phase.

DNA oligonucleotides that form G-quartet structures were used as stationary phase reagents for separation of bovine milk proteins, including alpha-casein, beta-casein, kappa-casein, alpha-lactalbumin and beta-lactoglobulin. Both artificial protein mixtures and a skim milk sample were analyzed. The separations were performed using open-tubular capillary electrochromatography, in which the oligonucleotides were covalently attached to the inner surface of a fused-silica capillary. Better resolution was achieved using the G-quartet-coated capillaries than was achieved using either a bare capillary or a capillary coated with an oligonucleotide that does not form a G-quartet structure. A 4-plane G-quartet-forming stationary phase was able to resolve three peaks for alpha-casein and to detect thermal denaturation of the proteins in the milk sample. The results suggest that G-quartet stationary phases could be used to separate very similar protein structures, such as those arising from genetic variations or post-translational modifications.

On-the-fly fluorescence lifetime detection of humic substances in capillary electrophoresis.

On-the-fly fluorescence lifetime detection was investigated as a tool for studying humic substances in capillary zone electrophoresis (CZE). Humic substances are complex, heterogeneous mixtures of natural products that tend to migrate in a single, broad CZE peak. The intrinsic fluorescence lifetime of five humic substances from the International Humic Substances Society (IHSS) was monitored using excitation at 488 or 364 nm to produce intensity-lifetime electropherograms for each of the substances. Each frequency-domain lifetime measurement, collected at subsecond intervals during the CZE run, contains the equivalent of a complete decay profile. Lifetime analysis of each decay profile was used to construct a lifetime-resolved electropherogram for each lifetime component, from which the variation in relative intensity contributions of each lifetime across the broad CZE peak could be determined. Absorption spectra, fluorescence excitation-emission spectra, and lifetime profiles of batch solutions of the samples were determined as well. It was found that, whereas absorption and fluorescence spectral characteristics tended to discriminate between humic acids and fulvic acids, the batch solution lifetime profiles discriminated instead between samples from different sources, regardless of fraction. On-the-fly lifetime detection provided a more detailed view of the fluorescence decay of the samples, including greater resolution of lifetimes for two of the fulvic acids and greater discrimination among samples based on lifetime profiles across the CZE peaks.