Depletion of nuclear histone H2A variants is associated with chronic DNA damage signaling upon drug-evoked senescence of human somatic cells

Modeling of acute DNA damage and genotoxic stress (drug-evoked) senescence in human somatic HCA2 cells

Senescence is triggered by various cellular stresses that result in genomic lesions and activation of chronic DNA damage signaling. However, the role of chromatin and its properties upon induction of senescence remains highly controversial [37-41, 83]. Several chemo-therapeutic drugs, including etoposide [41, 42], bleomycin [36, 43], and doxorubicin [43] are known to induce premature senescence in a variety of human somatic and stem cells. In our study, we have used well-characterized model of human neonatal fibroblasts (HCA2) [44-46] to assess the nuclear H2A histone levels under three different conditions: 1) normally proliferating cells, 2) acute DNA damage, and 3) drug-evoked senescence.

Cycling early passage diploid HCA2 fibroblasts were cultivated as outlined in Materials and Methods Section, and treated with the chemomimicking agent, bleomycin (50 μg/ml), for 3 h to generate conditions of acute DNA damage. Drug-evoked senescence was induced when the cells were cultured in normal growth media after the removal of bleomycin for 3-5 days [36, 44]. A functional biomarker that is used most often to detect activated DDR is the histone variant, H2A.X, which when phosphorylated by ATM/ATR on serine 139 is commonly known as γH2A.X [47, 48] (Figure 1A). Under the condition of acute DNA damage, about 90% of the cells were positive for defused global nuclear staining in immunostaining experiments with γH2A.X-specific antibodies (Figure 1B). Despite the initial wave of DNA repair, leading to progressive reduction in time of the bleomycin-induced γH2A.X foci, about 80% of the cells underwent drug-evoked cellular senescence after 3-5 days post-bleomycin treatment. This was indicated by the formation of the predominantly focal pattern of gH2A.X and activation of lysosomal senescence-associated b-actin (Figure 1C), which is in accordance with the previously reported observations in the same cell system [49]. These events are known to result in constitutive DNA damage signaling, which is a phenomenon well documented in all major types of senescence [23, 50, 83].

Western analysis of HCA2 cells upon proliferation, acute DNA damage, and drug-evoked senescence revealed a significant increase in γH2A.X representation in the whole cell extracts during acute DNA damage (Figure 2A), and only a marginal increase in γH2A.X upon drug-evoked senescence, when compared with the proliferating cells. However, no significant change in the total protein level of histone H2A was observed by western blot analysis in the tested samples (Figure 2B). Most antibody-based tests cannot provide the requisite specificity for the accurate quantification of proteins in question, because multiple isoform or variants of the same protein could be recognized by the same antibody. The data also demonstrate that different exposures in western blot analysis can significantly impair interpretations (Figure 2A, compare upper and middle panels). The antibodies used are listed in Materials and Methods Section.

Development of multiplexed SRM assay for quantitative measurement of the histone H2A family, H2A.X and γH2A.X

Over the past few years, mass spectrometry has emerged as a technology to complement and potentially replace standard immunoassays in routine research and clinical tests. Mass spectrometry based SRM is rapidly becoming a preferred technology for the development of quantitative protein or peptide assays for clinical and academic research because it delivers high sensitivity, selectivity, throughput, and accurate quantification [35]. In an SRM experiment, the two mass analyzers of a triple-quadrupole mass spectrometer are used to isolate the peptide ion of interest and the derived fragment ion(s). This enables the precise quantification of peptides in a complex biological background, and thus indirectly the corresponding proteins present in the samples. The ability to multiplex the analytes allows effective quantification of protein panels across multiple samples by monitoring specific transitions (precursor-fragment ion pairs) during a predefined elution window. Targeted proteomics approaches based on SRM coupled to high-resolution LS-MS/MS discovery workflows have proven effective for the analysis of biomarkers in plasma and serum, successfully permitting a systematic, quantitative analysis of putative marker candidates [51, 52].

In this study, we developed a multiplexed SRM assay for quantitative measurement of: 1) nuclear representa-tion of histone H2A family proteins (histone H2A variant unbiased approach), 2) abundance of histone H2A.X (variant-specific approach), and 3) histone H2A.X phosphorylation (γH2A.X) (PTM-specific approach). Peptides selection was based on three different criteria: literature references, previously obtained LC-MS/MS discovery data (not shown), and algorithmic prediction. An initial exhaustive list of peptides and transitions was iteratively optimized as previously described [52-54]. The peptides sequences and transitions for monitoring peptides and heavy standards are listed in Table 1 and Supporting Information Table S1. Calibration curves with limit of detection (LOD) and limit of quantification (LOQ) for the peptides are shown in Supporting Information Figure S1.

In the unbiased approach, the internal monitoring peptides AGLQFPVGR and HLQLAIR are non-unique and common to all of the H2A histone family members (Figure 3A) with the exception of macro-H2A histone. These peptides allowed quantification of the abundance of all H2A variants in the tested samples. In the variant-specific approach, the other internal monitoring peptide, TSATVGPK, allowed for the specific quantification of H2A.X histone, regardless of its phosphorylation status. The C-terminal peptides ATQAS₁₃₉QEY and KATQAS₁₃₉QEY (missed cleavage), both phosphorylated and unmodified, allowed for a quantitative measurement of the serine 139 phosphorylated and unmodified tryptic products unique to histone H2A.X in our PTM-specific approach (Figures 3A and 5A). Simultaneous quantitative measurement of all of these tryptic peptides allowed us to monitor overall nuclear representation of histone H2A family members, H2A.X, and, importantly, allowed us to estimate the rate of phosphorylation of H2A.X histone by monitoring γH2A.X peptide in the tested samples.

Expression of H2A histone family members is sensitive to acute and chronic DNA damage signaling

Having established that all, except macro-H2A.1, histone H2A family members are highly conservative in their globular domain at the level of their protein sequence, we subsequently investigated the genomic sequence similarities among the histone H2A family members (shown in Supporting Information Figure S2). Histone variant specific primers were designed wherever possible using NCBI Primer-BLAST, and verified on the UCSC genome browser (Supporting Information Figure S3). Not all variants yielded unique primer pairs. The qPCR analysis was performed as described in Materials and Methods Section to investigate the dynamics of the variants-specific expression in normally proliferating HCA2 cells as well as in the same cells under condition of acute DNA damage and drug-evoked senescence. The results are shown in Figures 3B and 5B. Using gene-specific single-stranded DNA preparation coupled with qPCR, we were able to selectively assess the transcriptional activities of seven genes of H2A histone variants family: H2A type 3, H2A type 1, H2A.Z, H2A.V, H2A.J, H2A.X, and marco-H2A.1. We noted that the overall level of transcriptional activity of these genes varied, which can be arbitrary sub-divided into three distinct categories, where only H2A type 1 genesexhibited high-level expression. This cluster of histone variants probably accounts for the majority of histone H2A protein within the chromatin of HCA2 cells (Figure 3B). Both histones H2A.J (Figure 3B) and H2A.X (Figure 5B) exhibited medium levels of gene expression with only marginal differences in the amounts of mRNA generated from these genes, when compared with β-actin gene expression. H2A type 3, H2A.Z, H2A.V, and marco-H2A.1 expression was barely detectable in HCA2 fibroblasts. Thus, we rationalized that histone H2A type 1, H2A.J, and H2A.X are the major contributors to the chromatin packaging in HCA2 cells.

Having established the major transcribed members of the histone H2A family, we subsequently examined the changes in transcriptional activities of these genes upon acute and chronic DNA damage signaling. It has been previously demonstrated that biosynthesis of histones H3 and H4 drastically declined in the tested somatic models of ex-vivo replicative and drug-evoked senescence, IMR90 and WI38 [36]. In accordance with these observations, our data demonstrated that acute and chronic DNA damage signaling also impacts the expression of histones H2A type 1 and H2A.X, resulting in statistically significant (p-values=3.85×10⁻⁷ and 1.22×10⁻⁵, respectively) downregulation of these histones at the transcriptional level (Figures 3B and 5B) under both acute DNA damage and drug-evoked senescence conditions. Therefore, we concluded that expression of these histones is sensitive to DNA damage signaling and activation of chronic DDR, in particular. However, no changes in the transcriptional activity of H2A type 3, H2A.Z, H2A.V, H2A.J, or marco-H2A.1 genes were observed under the same conditions.

The above-mentioned data further expand the previous observation of changes in the histone H3 and H4 transcription during ex-vivo replicative aging and drug-evoked senescence of fibroblasts [36], and demonstrate that similar trends can be recorded for another indispensible nucleosomal packaging histone family, H2A. Comparison of the expression levels of the different members of the histone H2A family revealed that not all of the histone family members respond equally to the chronic DNA damage signaling. Particularly, histone H2A.J transcription showed slight, but significant transcriptional upregulation.

Drug-evoked senescence is associated with significant depletion of the histone H2A variants in the chromatin of human primary fibroblasts

To biochemically quantify the histone H2A variants' protein levels in the nucleus of cells subjected to either acute or chronic DNA damage signaling, we deployed SRM, which allowed us to monitor the abundance of H2A type 1, H2A.X, and H2A.J gene products at the protein level. The optimized SRM assay was used to interrogate nuclear fraction samples isolated from proliferating (−bleomycin) HCA2 cells, as well as from the cells under condition of acute DNA damage (+bleomycin, 3 h) and drug-induced senescence (+bleomycin, 5 days). The nuclei fractions were prepared as described in the Materials and Methods Section. Eight independent experimental samples were obtained for each condition. All the samples were normalized by DNA content. Quantitative data were obtained for all samples and then analyzed in “Pinpoint” software to determine the differential abundance patterns of all peptides described in the previous section (Table 1).

Our data revealed a prominent and statistically significant loss of the histone H2A family proteins in the HCA2 human fibroblasts under conditions of chronic DDR upon drug-evoked senescence (p-value= 1.22×10−15 for AGLQFPVGR monitoring peptide and p-value= 2.12×10−14 for HLQLAIR monitoring peptide) (Figure 4). This observed downregulation of H2A protein abundance in the nuclei of senescent cells correlates to the decline in the accumulation of mRNA of the HCA2-expressed histone H2A variants shown inFigure 3B, suggesting that either biosynthesis or their delivery into the nuclei was abrogated upon chronic DDR. The raw quantitative data for all sample measurements are given in Supporting Information Table S2.

Interestingly, our data also suggest a biphasic mode for nuclear dynamics of these histone H2A variants in response to drug-induced genotoxic stress. Prior to decline in the nuclear H2A protein levels in senescent cells, there was a statistically significant increase (p-value= 5×10⁻¹⁰) in the representation of these histones within the nuclei of the HCA2 cells during acute DDR (3 h after the addition of bleomycin to the media). We speculate that this increase in the nuclear H2A histones might be due to the formation of a histone supply “buffer zone,” which assists the active chromatin remodeling and histone exchange events coupled at this stage with efficient DNA damage repair, similar to the one described in [55]. While we observed an initial increase (about 36%) in the nuclear abundance of histone H2A family members, our data clearly depict a second trend: Histone H2A family nuclear representation dropped by 70%, when compared with the stage of acute DNA damage and 60%, when compared with the initial amount of histones H2A in normally proliferating HCA2 cells. One reason for such dynamics could be that the DNA repair machinery encountered genomic regions within which DNA damage could not be resolved efficiently, perhaps due to the failure of either site-specific or global action of the histone chaperones [34, 55, 56]. In general, histone chaperones are responsible for both the delivery of histones into the nucleus and integration of histones into chromatin, to faithfully restore the chromatin landscape following DNA repair. Irrespective of the nature of these events, it is important to emphasize that this biphasic behavior would not be detected by traditional western blot analysis of histone H2A, because these changes were not detectable in the western blot analysis experiments using total cellular HCA2 protein extracts (Figure 2B).

Together, these data support the previous observations in yeast and mammalian somatic cells, where replicative aging has been noted to be associated with prominent loss of major histones from the chromatin [33, 36]. In addition, our data also indicate that the SRM approach has a higher discriminating potential than traditional antibody-based assays.

Measurement of histone H2A.X abundance and H2A.X phosphorylation upon acute and chronic DNA damage signaling

H2A.X is constitutively expressed throughout the cell cycle. This histone variant differs from any other core histone H2A family member by the presence of a short C-terminal tail and a highly conserved SQE motif (serine, glutamine, glutamic acid) (Figures 3A and 5A). It has been reported that the H2A.X expression levels vary between mammalian tissues, ranging from 2 to 10% of total histone H2A expression [57]. This histone variant undergoes rapid phosphorylation of serine at position 139 (site of γ-phosphorylation) by members of the PI-3 kinase-like family in response to the DSBs [55]. The phosphorylation event serves as an important cue for the stable retention of DNA damage repair and DDR mediator proteins in close proximity to DSBs, resulting in the formation of microscopically discernible foci that function as a platform for interactions between multi-component signal transduction complexes, facilitating the activation of the cell cycle checkpoint [58-60]. The formation of γH2A.X foci can occur in response to genotoxic insults, such as drug treatment, UV exposure, metabolic stress, reactive oxygen species (ROS) [55], as well as under conditions of inappropriate over-activated growth and signaling pathways [82].

While both the timing and order of DDR events preceding and following H2A.X phosphorylation are well documented in the context of acute DNA damage and senescence in human cells [50], there is a lack of direct quantitative measurements of nuclear γH2A.X. Our multiplexed SRM approach allows for variant-specific quantification of H2A.X protein and its phosphorylation (PTM-specific quantification) in the tested samples. The H2A.X-specific monitoring peptides are listed in Table 1 and shown in red in Figure 5A.

Consistent with the overall depletion in histone H2A variants' protein level in the nuclei of HCA2 cells (Figure 4) following chronic DNA damage signaling, we observed a robust decrease in protein representation (p-value= 3.05×10⁻¹⁰) of H2A.X histone variant (Figure 5C). This was also accompanied by statistically significant decrease in H2A.X at transcriptional level (p-value= 1.22×10⁻⁵), as demonstrated by qPCR analysis (Figure 5B).

Interestingly, our measurements indicate that histone H2A.X comprises only about 0.4% of the total H2A protein in the nuclei of normally proliferating HCA2 cells (124 pmole of total H2A family member proteins shown in Figure 4 vs. 450 fmole of total H2A.X protein presented in Figure 5). This nuclear abundance is 10-fold lower than that previously suggested in mammalian tissues based on immunostaining analysis and 2D gel electrophoresis [57]. H2A.X nuclear protein dynamics were noted to follow a similar biphasic mode, as shown in Figure 4 and discussed earlier. We observed a relatively small, but statistically significant increase (~36%, p-value= ~2.20×10⁻¹¹) in the amount of nuclear H2A.X in the condition of acute DNA damage.

The reproducibility of SRM quantification is dependent on the efficiency of trypsin digestion. Experimental measurements may be skewed if protein modifications block the expected proteolytic cleavage site due perturbations in the protein folding. Therefore, in addition to fully tryptic peptides, we also monitored the missed cleavage peptides KATQASQEY and KATQASphQEY, because these were detected in our discovery spectra. The relative amounts of the fully tryptic and missed cleavage peptides in our experiments indicated that although about 10% of the unphosphorylated protein in all of our samples had one missed cleavage, the relative amounts of semi-tryptic KATQASQEY across all three samples were similar to the fully tryptic peptide ATQASQEY (Figure 5D and 5E, Table 1). Similar results were obtained with the phosphorylated peptides ATQASphQEY and KATQASphQEY (Figure 5D and 5E, Table 1).

As expected, SRM-targeted analysis of the H2A.X-specific tryptic peptides encompassing the serine 139 site for the damage-induced phosphorylation, ATQASQEY, and KATQASQEY revealed a decrease in the unphosphorylated form of H2A.X during acute DNA damage and in senescent cells (Figure 5D and 5E). This result coincides with the increased DNA damage-propelled serine 139 phosphorylation (γH2A.X) (Figures 1A, 1B and 2A) confirmed by quantification of ATQASphQEY and KATQASphQEY peptides (Figure 5A, 5D, and 5E). Our data indicate that the fraction of the phosphorylated H2A.X (γH2A.X) is about 0.5% relative to the total amount of H2A.X in the nucleus of the normally proliferating HCA2. Acute DNA damage has been found to result in about 3% of total H2A.X phosphorylation (comparable with that reported in [57]), where only about 2% of the total nuclear H2A.X has been noted to exist in the phosphorylated form upon drug-evoked senescence.

Our results, therefore, suggest that HCA2 fibroblasts had partially resolved bleomycin-inflicted DNA damage triggered by chronic DNA damage signaling and senescence.

Together, these data, although consistent with immunostaining and western blot analysis shown in Figures 1A, 1B, and 2A, represent the first truly quantitative measurement of the degree of H2A.X phosphorylation in human primary somatic cells under conditions of acute DNA damage and drug-evoked senescence. The observed degree of H2A.X phosphorylation in the senescent human fibroblasts is also in agreement with our previous, genome-wide nucleosomal analysis of γH2A.X distribution, which estimated that only 1.2% of the total chromatin contains γH2A.X-bearing nucleosomes upon replicative senescence of human adult stem cells (ADSCs) [24].

In accordance with the nuclear loss of generic H2A (Figure 4), our measurements of total nuclear quantitates of H2A.X (both phosphorylated and unphosphorylated) indicate a nuclear depletion of this histone variant upon chronic DDR in drug-evoked model of senescence (400 fmol in proliferating cells vs. 260 fmol in senescent cells). This decline in nuclear protein abundance of histone H2A.X variant, together with robust decrease in the transcription activity of this gene (Figure 5B), suggest that chronic DNA damage signaling might be associated with downregulation of histone H2A.X biosynthesis or its trafficking from the cytoplasm to the nucleus. This observed change also corroborates with the previous reports of a decrease in the synthesis of H2A.X upon ex-vivo replicative aging of human primary lung fibroblasts.

Samples

HCA2 cells were cultured in the presence and absence of 50 μg/ml of bleomycin as previously published [49, 79]. The nuclei were harvested by centrifugation through a 30% sucrose cushion as described in [80]. This resulted in three sample groups: 1. No DNA damage, 2. Early DNA damage, 3 h after bleomycin treatment, and 3. Drug-evoked DNA damage: after initial 3 h in 50 μg/ml of bleomycin, the cell media was changed to normal growth media and the cells were further cultured for 5 days and then subjected to nuclei preparation and SRM analysis. The nuclei pellets from approximately 10M cells were lysed in 400 ml 8M Gu-HCL + DTT. The samples were subsequently enzymatically digested and processed for SRM analysis as previously published [35, 52].

Immunohistochemistry

As described in the Samples Section, 1×10⁴ HCA2 cells/well in 4-well slides were treated with bleomycin, and then fixed with 4% paraformaldehyde and permeabilized with PBS and 0.5% Triton X-100. The blocking and antibody incubations (γH2A.X; Millipore #05-636) were performed in 4% normal donkey serum (NDS; Jackson Immunochemicals) in PBS. The nuclei were counterstained with 100 ng/ml of 4', 6-diamidino-2-phenylindole (DAPI; Sigma), and the slides were mounted in ProLong Gold antifade aqueous mounting medium (Invitrogen). Epifluorescence images were acquired on an Olympus BX60 fluorescence microscope with Spotfire 3.2.4 software (Diagnostics Instruments).

Senescence-associated β-galactosidase assay

As described in the Samples Section, 1×10⁴ HCA2 cells/well in 4-well slides were treated with bleomycin. β-galactosidase detection was carried out using a commercial senescence detection kit (BioVision #K320-250), according to the manufacturer's recommended protocol. The slides were incubated with staining solution for 14 h prior to visualization.

Western blot analysis

A total of 1×10⁵ HCA2 cells were cultured on 10-cm dishes and treated with bleomycin, as described in the Samples Section. Triplicates were performed for each condition. The cells were collected by cell scraper and lysed with lysis buffer (50 mM Tris-HCl; pH 7.4; 100 mM NaCl; 1% NP-40; 0.1% SDS; 0.5% sodium deoxycholate; Halt protease inhibitor cocktail, Thermo Scientific #87786) at 4°C. The lysates were rotated at 4°C for 45 min, and then sonicated vigorously to shear DNA. Protein concentrations were determined by Bradford assay, and normalized lysate volumes were loaded onto a NuPage 4-12% Bis-Tris gel (Novex #NP0336) in NuPage LDS sample buffer (Invitrogen #NP0007), and run at 200 V for 50 min. The proteins were transferred to nitrocellulose membrane using the XCell II blot module (Invitrogen #EI9051). The membranes were blocked with 5% milk solution for 30 min at room temperature, before being incubated with primary antibodies overnight in TBS-T at 4°C with gentle rocking. The antibodies used were as follows: γH2A.X (Millipore #05-636), Histone H2A (AbCam #AB13923), and β-actin (AbCam #AB6276). HRP-conjugated secondary antibodies were used before subsequent ECL exposure. The γH2A.X membranes were exposed for short (1 min) and long (10 min) periods, and the representative results are shown in Figure 2A. The band intensities were calculated using Fiji (Image J), and normalized to that of β-actin.

RT-PCR

A total of 1×10⁵ HCA2 cells were cultured on 10-cm dishes and treated with bleomycin as described in the Samples Section. The procedure was performed in triplicates for each condition. The cells were collected by cell scraper and RNA was isolated using Trizol reagent (Life Technologies #15596). DNA contamination was removed with TURBO DNase (Applied Biosystems #AM2238) at 37°C for 30 min, followed by 10 min at 70°C in the presence of 15mM EDTA to inactivate the enzyme. Reverse transcription was carried out using gene-specific primers (see the following paragraph) and SuperScript III reverse transcriptase (Invitrogen #18080-051) at 55°C for 1 h. qPCR was carried out using Fast SYBR Green Master Mix (Applied Biosystems #4385612). The reactions were run in a Roche Lightcycler 480 system under the following conditions: 1 × 95°C for 20 s; 40 × 95°C for 3 s, and 60°C for 30 s. Crossing point (Cp) cycle values for each reaction were calculated by Absolute Quantification/Second Derivative Maximum calculations. The Cp values for each primer/sample were normalized to that of β-actin expression.

Primers

Also refer to Supplemental Figure S1. The primers were designed using NCBI Primer-Blast software using default parameters. This software was used to design primers that were specific for individual H2A variants (wherever possible). H2A.X: F - ACGAGGAGCTCAACAAGCTG, R - GTGGCGCTG GTCTTCTTG; H2A.J: F - GCTTTTGAATGTGCTGG ATG, R - ACCTCCCCGCTAGATGTCAC; H2A type 3: F - CGGTTGCCGTTGTCTTTTT, R - GAGCGCG ACTTAGCCTTG; H2A type 1C: F - TTAGAGTACC TGACCGCCGA, R - GCGAGTCTTCTTGTTGTCGC; Macro-H2A.1: F - AGACGTCCAGGTCTGCCA, R - CCAGAATCTCCGCTGTCAGG; H2A.V: F - CTACA GTTTCCTGTGGGCCG, R - TCCAGAATCGCAGCA CTGTA; H2A.Z: F - CAGGGAAGGCGGGAGACAG, R - CCTTACCGCCAGCCATTTCG.

Trypsin digestion

The samples were thawed on ice and processed as previously described [35]. Experiments using a concatenated, heavy-labeled synthetic protein standard demonstrated consistent digestion [51]. The samples were normalized to total protein concentration.

SRM assay development

SRM assays were developed on a Thermo Scientific TSQ Vantage triple quadrupole mass spectrometer, Accela MS pump, CTC autosampler, and IonMax source equipped with a low-flow metal needle, as previously described [53, 54]. The samples were analyzed in triplicate. Reversed-phase separations were carried out on a 150 × 1-mm Thermo Scientific Hypersil Gold 3.0-μm C18-particle column. Solvent A was LC-MS grade water with 0.2% (v/v) formic acid. Solvent B was LC-MS grade acetonitrile with 0.2% (v/v) formic acid. Thermo Scientific Pinpoint software was used to predict candidate peptides, select fragment ions, and build the MS instrument method.

Light- and Heavy-labeled peptides

Synthetic isotopically labeled peptides were used as internal standards for quantification. Light and heavy versions (98% purity) of each target peptide were synthesized (Thermo Fisher Scientific). Heavy peptides had sequences identical to those of the light peptides, but the C-terminal Lysine or Arginine was fully-labeled (>98.5%) with 13C or 15N (see Supporting Information Table S1 for peptide sequences and transitions). The peptides were identified by coeluting light and heavy transitions.

Calibration curve generation

Seven-point calibration curves (0.5, 1, 5, 10, 50, 100, and 500 fmol, Figure 2 a-i) were developed with a pool (30 μg) of all samples as a background matrix. Each point on the calibration curve (and every sample analyzed) included 100 fmol of heavy-labeled peptides. We used reverse curves to generate calibration curves, because we did not have the matrix free of endogenous analytes. Complete details of this type of curve have been presentedpreviously [81]. In addition, all the samples were dissolved in a solution of 200 μg/ml of glucagon in 97% water, 3% ACN, and 0.2% FA to minimize binding to plastic surfaces [54]. Finally, the calibration curve replicate points were run adjacently. As each calibration point contained the same amount of light peptide (endogenous) variance in the heavy peptide (or endogenous peptide), these quantities were used to calculate the run-to-run variance. Also, heavy peptides were spiked into each clinical sample and the low run-to-run variance added confidence to the analysis.