Mass spectrometry enumeration of filamentous M13 bacteriophage

Abstract

In the last decade, filamentous M13 bacteriophage has emerged into numerous biotechnological applications as a promising nontoxic and self-assembling biomaterial with specific binding properties. This raises a question about its upscale production that consequently requires an accurate phage enumeration during the various protocol developments. However, traditional methods of measuring phage concentration are mainly biological in nature and therefore time and labor intensive. These traditional methods also demonstrate poor reproducibility and are semi-quantitative at best. In the present work, we capitalized on mass spectrometry based absolute protein quantitation. We have optimized the quantitation conditions for a major coat protein, pVIII. Enumeration of M13 bacteriophage can be further performed using the determined molar concentration of pVIII, Avogadro’s number, and known copy number of pVIII per phage. Since many different phages have well-defined copy number of capsid proteins, the proposed approach can be simply applied to any phage with known copy number of a specific capsid protein.

The filamentous M13 bacteriophage (M13 phage) contains a circular single-stranded DNA genome that encodes only 10 proteins. The wild type genome is enclosed in a cylindrical protein coat consisting of 2700 copies of major coat protein, pVIII (UnitProtKB - P69541). At either end of the virion, there are a number of minor coat proteins. However, pVIII constitutes 98% of total protein. Traditionally, M13 phage is the most commonly used vector to create random peptide and protein display libraries [1–4]. More recently, due to its nontoxic, self-assembling and specific binding properties, M13 phage applications have emerged into novel research areas, such as cell imaging and targeting [5, 6], vaccine development [7, 8], bio-chemical and bio-optical sensing [9, 10], tissue and biomaterials engineering [11, 12], and broad field of nanobiotechnology [13, 14]. Furthermore, the single-stranded genome of M13 phage is commonly used as a scaffold for DNA origami [15]. All of these applications rely on upscale production of M13 phage and optimizing various steps of device fabrication and use. To achieve these goals, the field urgently needs a quantitative method for the accurate measurement of M13 phage concentration.

There are three general groups of reported methods to quantify phages. These include techniques of measuring phage infectivity, measuring phage nucleic acid or proteins, and counting physical phage particles [16]. Not all reported techniques received further attention, especially those with high labor and expensive cost. For the M13 phage, the most commonly used techniques are (i) plaque counts on agar plates seeded with the bacteria in which the M13 phage can propagate and (ii) quantitative polymerase chain reaction (qPCR) [17, 18]. Unfortunately, these traditional approaches are prone to serious limitations. Plaque assay is time consuming, suffers from poor reproducibility, has ambiguity in morphology characterization and plaques counting, and has a large error range. Although qPCR continues to be a promising method to determine the absolute quantity of phage, qPCR tends to report values higher than the actual ones because the number of phage genomes in a sample is always much higher than the number of intact phage particles [19].

Since each of the current methods has several limitations, an improved phage quantitation by mass spectrometry was explored here. We report the development of a multiple reaction monitoring (MRM) assay [20] to quantitate the pVIII coat protein of M13 phage and demonstrate how the measured molar concentration of pVIII can be used for phage enumeration.

Materials and method

Source and purification of M13 phage

Ph.D.−12 filamentous bacteriophage display peptide library kit was purchased from New England BioLabs (Ipswich, MA, USA). E. coli K12 ER2738 from the kit was streaked onto an LB/tetracycline plate and cells were allowed to grow overnight at 37 °C. The next day, LB medium (3 mL) was inoculated with a single colony and, after overnight growth at 37 °C, was used for further propagation depending on the amount of cells required for experiments. To amplify the M13 phage, freshly grown E. coli K12 ER2738 were transfected with various amounts of phage and incubated at 37 °C with 250 rpm shaking for various periods of time. The cell media were then subjected to 12000 g centrifugation for 10 min to remove E. coli cells. All centrifugations were performed at 4 °C to ensure complete E. coli removal, and 12000 g centrifugation was performed twice. The M13 phage was then precipitated with 4 % PEG-8000/0.5 mol/L NaCl overnight at 4 °C, collected the next day by centrifugation at 12000 g for 10 min, and re-suspended in phosphate buffer saline. Various preparations of M13 phage were kept at 4 °C until further use.

An AKTA FPLC instrument (Amersham Biosciences, Piscataway, NJ, USA) was used to purify M13 phage on a size exclusion Superdex 200 Increase 10/300 GL column in phosphate buffer saline. The flow rate was 0.4 mL/min. The M13 phage was recovered from the void volume fraction and was subjected to analysis on a 15 % SDS-PAGE gel to assess the purity of preparation. Proteins were stained with Pierce silver stain for mass spectrometry kit (Thermo Scientific, Rockford, IL, USA). The pVIII concentration in this sample was determined by amino acid analysis (New England Peptide Inc., Gardner, MA) and was used in all further experiments.

Sample processing for mass spectrometry

The total protein concentration in M13 phage samples was measured using the DC protein assay kit (Bio-Rad Laboratories, Hercules, CA) and bovine serum albumin as a standard. Selected amounts of phage were processed in 50 mmol/L NH₄HCO₃/0.1 % Rapi-Gest. M13 phage samples were first treated with 10 mmol/L dithiothreitol. After incubation for 60 min at room temperature, samples were treated with 30 mmol/L iodoacetamide for another 60 min in the dark and precipitated with chloroform/methanol [21]. The pellets obtained from precipitation were sonicated in 100 μL of 50 mmol/L NH₄HCO₃/0.1 % RapiGest and supplemented with an isotope labeled internal standard (IS). Two different peptides were used as ISs; Ala-Glu-Gly-Asp-Asp-Pro-Ala-Lys (AEGDDPAK or IS-short) and Ala-Glu-Gly-Asp-Asp-Pro-Ala-Lys-Ala-Ala-Phe-Asn-Ser-Leu (AEGDDPAKAAFNSL or IS-long). Both versions of IS carry isotope-labeled Pro (¹³C₅, ¹⁵N) and were synthesized by Biomatik Corp. (Cambridge, Ontario, Canada). The 1 mg/mL concentration of peptides was provided by Biomatik Corp. and reconfirmed by amino acid analysis (New England Peptide Inc., Gardner, MA). Finally, the samples were treated with trypsin (1:10, w/w) for 15 h at 37 °C. After trypsinolysis, the samples were acidified with 0.5 % trifluoroacetic acid (TFA) for 30 min at 37 °C and centrifuged at 20000 g for 30 min to collect the supernatant. Samples were then dried in a vacuum centrifuge (Eppendorf AG, Hamburg, Germany), re-dissolved in 100 μL of water, and dried again.

LC-MS/MS analysis

For LC-MS/MS analysis, dried peptides were reconstituted in 3 % acetonitrile/ 0.1 % formic acid or in 3 % acetonitrile/0.1 % TFA when an Agilent Zorbax Eclipse Plus C18 reversed-phase RRHD column (2.1 mm × 50 mm, 1.8 μm particle) was used for LC separation. Peptides were eluted at a flow rate of 200 μL/min using the following gradient of solvent B in solvent A: 3 % B for 5 min, 3 % to 10 % B for 8 min, 10 % to 50 % B for 2 min, and 50 % to 3 % B for 4 min. Solvent A was water containing either 0.1 % formic acid or 0.1 % TFA and solvent B was acetonitrile containing 0.1 % formic acid or 0.1 % TFA. When Imtakt Unison UK-C18 reversed-phase MF column (2 mm x 100 mm, 3 um particles) was used for LC separation, dried peptides were reconstituted in 0.1 % formic acid in water. Peptides were eluted at a flow rate of 200 μL/min using the following gradient of solvent B in solvent A: 0 % B for 2 min, 0 % to 15 % B for 6 min, and 15 % to 0 % B for 2 min. Solvent A was water containing 0.1 % formic acid and solvent B was acetonitrile containing 0.1 % formic acid.

MRM assay was performed on an Agilent 6490 iFunnel Triple Quadrupole LC/MS system (Santa Clara, CA). Protein concentrations were calculated from the ratio of the light and heavy MRM peak areas multiplied by the known amount of stable isotope-labeled IS spiked into the sample prior to the digestion with trypsin. Three transitions per peptide detected at the same retention time were monitored. The overall details of MRM assay are illustrated in the Supplementary Figure. The acquisition method used the following parameters in positive ion mode: fragmentor 380 V, collision energy 15 V, dwell time 100 ms, cell accelerator 4 V, electron multiplier 500 V, and capillary voltage 3500 V. MRM transitions for 2+ charged precursor ions and 1+ charged product ions were predicted using PinPoint software (Thermo Fisher Scientific, Waltham, MA).

Data analysis

MRM peak area integration was performed using Skyline (4.1.0.11714) (University of Washington). Excel was used to calculate peak area ratios. Peak integration was manually inspected and adjusted, if necessary. The peak ratios from transitions were averaged to yield the peptide ratios. All experiments were performed in duplicate with three replicate injections. Data are represented as the mean ± SD.

Results and discussion

Optimizing conditions of MRM assay

We first assessed the purity of M13 phage preparation after size-exclusion chromatography. Fig. 1A shows that heavily loaded sample of void volume fraction (lane 2) is represented by a single silver-stained pVIII band on the SDS-PAGE.

Fig. 1.

15 % SDS-PAGE (A) and amino acid sequence (B) of pVIII. (A) Samples of M13 phage before (1) and after (2) size exclusion chromatography. (B) Target tryptic peptide of pVIII for MRM assay (IS-short) is shown in red. C-terminal flanking region extension for IS-long is shown in blue.

The pVIII is a 50 amino acid residue protein. In silico digestion of pVIII with trypsin produces only one peptide suitable for MRM assay, which is AEGDDPAK, the N-terminal sequence of pVIII (Fig. 2B, shown in red). We and others had reported that addition of natural flanking sequences on both sides of tryptic peptides used as IS can improve quantitative performance of this standard in MRM assays [22–26]. Therefore, we evaluated two peptides as IS for pVIII quantitation and refer to them as IS-short and IS-long. IS-short is AEGDDPAK while IS-long has a 6 amino acid residue extension at the C-terminus, AEGDDPAKAAFNSL, which is expected to better mimic the full-length pVIII protein. Both versions of IS carry isotope-labeled Pro (¹³C₅, ¹⁵N). Initial MRM runs revealed the three most intense transitions for AEGDDPAK: y3 (401.7/315.2 and 404.7/321.2), y6 (401.7/602.3 and 404.7/608.3), and y4 (401.7/430.2 and 404.7/436.2). Y-transition is defined as a pair of parental ion and product y-ion. The m/z values for doubly charged precursor/singly charged fragment ions are shown in parentheses for light and heavy transitions, respectively.

Fig. 2.

Optimizing LC for MRM assay. Three transitions (y3 in green, y6 in blue, and y4 in red) are shown. (A), (B), and (C) show light transitions for AEGDDPAK after digestion of M13 phage with trypsin. (D) shows heavy transitions for IS-short. LC was performed on an Agilent Zorbax Eclipse Plus C18 RRHD in the presence of formic acid (A) or TFA (B). LC was performed on a Unison UK-C18 MF column in the presence of formic acid (C and D).

Fig. 2A shows y3, y6, and y4 transitions from trypsin-digested pVIII in green, blue, and red, respectively. However, it becomes clear that the short and hydrophilic AEGDDPAK is not well retained on the typical C18 column (Agilent Zorbax Eclipse Plus C18 RRHD (2.1 mm x 50 mm, 1.8 μm particles)) in the presence of formic acid. It eluted at 1.44 min, which is essentially the void volume. We replaced formic acid in the mobile phase with TFA, resulting in improved retention as shown in Fig. 2B. However, this replacement also caused a 10-fold loss of sensitivity. The decrease in signal intensity can be seen by comparing the y-axis in Fig. 2A and 2B. These shortcomings are still acceptable for quantitation of abundant amounts of pVIII, but may severely affect analysis of very low pVIII concentrations. Because no alternative peptides were available for MRM quantitation of pVIII,, we attempted to identify chromatographic conditions that would yield optimal retention of AEGDDPAK in the presence of formic acid. For this purpose, we investigated a special class of C18 column that is compatible with 100% aqueous conditions, such as Unison UK-C18 MF column (2 mm x 100 mm, 3 um particles) from Imtakt USA, Portland, OR. The gradient for this stationary phase can start with 0% acetonitrile and acceptable retention at 6.2 min was observed for AEGDDPAK using this column, as shown in Fig. 2C and 2D. Fig. 2C and 2D also show that relative ratios of the three transitions derived from real M13 phage sample and from IS are identical. This points to the absence of matrix interference from biological samples for quantitation of pVIII based on MRM analysis of AEGDDPAK.

When optimizing sensitivity, it is always beneficial to have multiple strong transitions. Therefore, the next step in optimizing the MRM assay for AEGDDPAK was to compare the ratio of transitions at different collision energy (CE) values. All initial experiments were performed at CE 20 V and yield only one strong transition (y3, shown in green). Fig. 3 shows that different CEs affect the ratio of transitions by suppressing or enhancing individual transitions and that it is possible to find the best combination by optimizing CE. For our instrumental setup, the best CE was 15 V with two equally strong transitions (y3 and y6 shown in green and blue, respectively).

Fig. 3.

Optimizing CE value for MRM assay. Three transitions (color coordinated with Figure 2) for AEGDDPAK after digestion of M13 phage with trypsin are shown at different CE values.

The sensitivity and linearity of the assay were further evaluated. A sample of pVIII was supplemented with either IS-short or IS-long at 2/1 molar ratio and digested with trypsin. Various dilutions of this sample were then used to evaluate assay sensitivity. Fig. 4A shows the data for 10 femtomole of pVIII injected per LC-MS/MS run. For both ISs used, the proportions of light and heavy transitions demonstrate good consistency even at this low amount of sample on column.

Fig. 4.

Sensitivity (A) and linearity (B) of MRM assay. (A) Light and heavy transitions for 10 femtomol of pVIII on the column are shown. M13 phage sample was supplemented with either IS-short or IS-long and digested with trypsin. (B) Response curves for samples with various concentration of pVIII supplemented with the same amount of either IS-short or IS-long. The experiments were performed in duplicate with three replicate injections. Data are represented as the mean ± SD.

Overall, for several dilutions of this sample, the expected light to heavy 2/1 ratio was accurately determined with SD values ± 5 % and ± 8 % for IS-short and IS-long, respectively.

For linearity experiments (Fig. 4B), 8 dilutions of pVIII were supplemented with the same amount of IS-short or IS-long, digested with trypsin, and MRM data were plotted as log of pVIII concentration versus log of light/heavy peak area ratios. The response curves show linearity and low scatter over the 2 orders of magnitude of pVIII concentration range tested. Both ISs performed well in these experiments although IS-long remains reproducibly associated with slightly higher scatter than found for IS-short. In total, we concluded that both ISs, IS-short and IS-long, can be equally used for quantitation of pVIII.

Quantitation of pVIII in biological samples

We have optimized an MRM assay to quantitate the molar concentration of pVIII and applied it to M13 phage samples.

Protocols of phage amplification are usually performed in a large volumes of E. coli culture and have a common step of amplified phage precipitation to concentrate the obtained sample. Overnight treatment on ice with 4 % PEG-8000/0.5 mol/L NaCl is typically used to precipitate phage. For smaller volumes of phage, high-speed ultracentrifugation can be used to pull down and concentrate phage. We performed a side-by-side comparison of the efficiency of both methods. First, we measured the pVIII concentration using the MRM assay and took the amount of pVIII in the sample as a 100 % (Fig. 5A). We then used chloroform/methanol precipitation to pull down pVIII and found 98.6 % of pVIII in the pellet. Precipitation with PEG-8000/NaCl resulted in 81.8 % recovery of pVIII while ultracentrifugation at 208,000 g for 60 min yielded only 68.7 % recovery (Fig. 5A). Depending on the requirements for the final preparation of phage, all of these approaches can be useful. Chloroform/methanol precipitation of the small aliquot of phage followed by MRM analysis could be the most accurate way to quantitate phage concentration for analytical purposes. PEG-8000/NaCl precipitation remains a good approach to precipitate phage from large volumes. Despite the relatively low recovery, high-speed centrifugation still could be valuable when the yield itself is less important than avoiding any extra phage treatments.

Fig. 5.

M13 phage precipitation (A) and amplification (B). (A) Treatments include chloroform/methanol precipitation (1), PEG-8000/NaCl precipitation (2), and 208,000 g centrifugation (3). The experiments were performed in duplicate with three replicate injections. Data are represented as the mean ± SD. (B) Time dependence of M13 phage amplification. Data for three replicate injection of a representative experiment are presented as the mean ± SD.

Many M13 phage applications will require substantial upscaling of production. The classical protocol for M13 phage amplification in E. coli cells recommends collection of phage approximately 4.5 hours after transfection. We have quantitated pVIII in samples collected up to 8 hours after transfection and plotted the data in Fig. 5B. A plateau occurs from 4 hours to 5 hours after transfection, followed by exponential growth of phage release. Basically, postponing phage collection up to 8 hours after transfection resulted in at least 6-fold increase in total M13 phage collected.

Enumeration of M13 phage

Wild type M13 phage has a cylindrical protein coat consisting of 2700 copies of the major coat protein, pVIII. One of the distinct advantages of the MRM approach is that the molar concentration of pVIII can be converted into the number of M13 phages using Avogadro’s number (N_A) and the copy number of pVIII per phage (the equation is shown below).

$$\left[\text{M13}\text{phage/L}\right]=\left[\text{pVIII},\text{mol/L}\right]x{N}_{\text{A}}/2700$$

As a demonstration of this concept, we have measured the pVIII concentration in the commercial Ph.D.−12 filamentous bacteriophage display peptide library from New England BioLabs (Ipswich, MA, USA) and calculated that the amount of M13 phage there is 1.77 ± 0.15 × 10¹⁰ phage/μL for three biological replicates from the same library. This value concurs well with manufacturer provided number of around 10¹⁰ pfu/μL and serves to validate the MRM approach proposed here.

Many wild type phages have a well-defined copy number of capsid proteins in their structure. For example, the total number of 48.7-kDa gp23 monomers in the bacteriophage T4 capsid is 930 [27] and mature Head II of bacteriophage HK97 contains 420 copies of cleaved gp5 [28]. Both of these phages can be enumerated based on MRM quantitation of 48.7-kDa gp23 or cleaved gp5, respectively. Therefore, the MRM assay for enumeration of M13 phage proposed here has a broad application to many other phages. However, we also need to mention that the size of the phage capsid depends on the size of nucleic acid encapsulated. Consequently, the copy number of a specific capsid protein determined for a wild type phage may be different in mutant phages with changes in the size of nucleic acid. Nevertheless, mutant phages can also be enumerated after verifying the copy number of capsid proteins in these phages.

Abbreviations

M13 phage

M13 bacteriophage

qPCR

quantitative polymerase chain reaction

MRM

multiple reaction monitoring

TFA

trifluoroacetic acid

CE

collision energy

IS

internal standard