In a previous blog outlining the advantages of high-resolution accurate mass measurements for protein quantitation using the TripleTOF 6600, it was noted that although the triple-stage quadrupole demonstrated high sensitivity when operated in multiple reaction monitoring mode (MRM), the relatively low-resolution measurement of m/z failed to discriminate Rituximab response from nominally isobaric interferences given the complexity of the proteolytically digested samples (June 28/2016). While the accurate mass filtering capabilities of the TripleTOF 6600 represents one mechanism for achieving increased selectivity over MRM, the triple quadrupole/linear ion trap (LIT) hybrid platform represented by the QTRAP^® 4500, 5500, 6500 and 6500+ systems provides an alternative technique by leveraging a third stage of MS, often referred to as MRM³. In this blog, we outline the MRM³ scan function and survey several large molecule applications which utilize the additional stage of fragmentation in the LIT to yield significant improvements in achievable detection limits when compared to MRM.

Since the introduction of the first QTRAP in 2002, a number of LIT innovations have collectively enhanced overall performance, including:

• Improved fragment-ion trapping efficiency through increased rf drive frequency
• Increased sensitivity via implementation of Q3 auxiliary axial electrodes
• Introduction of pulsed gas to increase ion-trap pressure, thereby increasing collisional frequency and reducing excitation time whilst favorably biasing the population of ions which are fragmented as opposed to ejected during resonance excitation

The actual implementation of MRM³ is typically prompted by high baseline or persistent interference observed during MRM analysis, where the chosen transition is simply not discriminating enough to isolate a primary fragment from other ions with very similar m/z ratios. To generate additional specificity in these instances, the LIT isolates a primary fragment ion (incoming from Q2) and then dissociates it into a new array of secondary fragments, which are then rapidly scanned out to the detector (Figure 1). By selecting a primary fragment and a subsequent secondary fragment that exclusively characterizes the surrogate peptide, a unique ion trail is created enabling abstraction from the milieu of other compounds in the proteolytic digest, with each fragmentation step providing an additional dimension of specificity. Since all secondary fragment ions are scanned out of the LIT, an extracted ion chromatogram can be derived by summing several m/z species, potentially increasing overall sensitivity.

While the vast majority of large molecule quantitation endeavors using a bottom-up strategy utilize MRM, there are several reported instances where the additional specificity and selectivity offered by MRM³ proved advantageous in achieving lower detection limits whilst simplifying sample preparation strategies. For example, the applicability of MRM³ has been demonstrated by researchers at the Universite de Lyon for targeting protein biomarkers in the low ng/mL range in non-depleted human serum, most notably PSA, with results correlating well with established ELISA tests.1 Prior to MRM³, an LC-MRM method was developed for PSA using tryptic digestion monitoring the optimal surrogate peptide transition m/z 636.8 > 943.5 for LSEPAELTDAVK. However, the method used HAS depletion, which proved the most difficult step to automate, and detection was limited to 50 ng/mL due to background interference. In contrast, a clinically relevant 4 ng/mL detection limit was achieved using MRM³ when monitoring the transition m/z 636.8 > 943.5 > 627.3 + 698.3 + 799.4 + 797.4 (Figure 2), and the immunodepletion step could be eliminated from the sample preparation strategy, thus improving sample throughput. In general, researchers noted that in moving from MRM to MRM³, limits of quantitation increased on average three-to-six fold for trypsin hydrolyzed bacterial protein models TP171, TP574, TP435, and core NS4 (Table 1).

Other examples of MRM³ improving the LOQ of proteotypic peptides in complex plasma digest include two biomarkers for inflammation (C-reactive protein, CRP) and metalloproteinase inhibitor 1 protein (TIMP-1), while two urinary putative protein biomarkers of kidney injury quantified using MRM³ include aquaporin 2 and podacin.2 Depending on the targeted protein, LOQs were improved two-to-four fold when compared to MRM (Table 1).

In the case of CRP, despite careful selection and optimization for the proteotypic peptide ESDTSYVSLK, the presence of intense contaminating signals within all the MRM transitions was problematic. Rather than introduce complicated antibody strategies for depletion of interferences or enrichment of the targeted protein or peptide (SISCAPA), MRM³ was implemented monitoring the transition m/z 564 > 696 > 347. The XIC of the second-generation progeny ion m/z 347 lacked any interfering signals and served to illustrate the striking difference between MRM³ and MRM scan functions towards the detection specificity of minor peptides dominated by major compounds when analyzing whole plasma digest.

Most recently, MRM³ on the QTRAP 6500 platform was implemented for the quantitation of crustacean allergens derived from shrimp and lobster with applicability demonstrated at trace concentrations in complex food matrix.³ Protein extraction followed by tryptic digestion and non-targeted LC-HR MS/MS using a TripleTOF 5600 led to the identification of three peptide biomarkers from lobster and four from shrimp, with all identified peptides corroborated as shellfish allergens. When analyzing complex industrially prepared foods with the potential for contamination with crustacean, the most selective and specific biomarkers monitored by MRM were only detectable at concentrations of 1000 μg of crustacean meat/g of food (Table 2). Below this detection limit, researchers noted that most relevant MRM transitions were indistinguishable from background response. In contrast, the additional specificity of MRM³ allowed lobster and shrimp biomarker detection at concentrations down to 25 μg of crustacean meat/g of food due to the suppression of non-specific signals from untargeted peptides. This limit of detection was sufficient to report crustacean contamination at levels relevant for sensitive allergic individuals.

MRM³ Quantitation Considerations: Optimizing Duty Cycle and Sensitivity
The population of trapped ions in an MRM³ experiment can be optimized with a Dynamic Fill Time (DFT) algorithm that inversely varies the LIT injection period against the flux of ions sampled from the source. In this manner, trapping time is reduced for high ion flux to prevent overpopulating the LIT, and increased for low ion flux to optimize response at low concentration. However, prolonged fill time at low concentration can result in poor chromatographic sampling rates resulting in an insufficient number of data points representing a peak profile, ultimately limiting the achievable LOQ due to poor precision and accuracy.

To circumvent the sampling rate variation associated with DFT, shorter fixed injection periods in combination with ion trapping in Q0, the quadrupole that serves as an ion guide at the instrument’s orifice, can be implemented. In combination with fixed fill time, sensitivity is maintained by accumulating sufficient ions in Q0 while the LIT is occupied with isolating, dissociating and scanning a preceding batch of ions. The ions accumulated in Q0 are released to the recently emptied LIT, resulting in larger ion populations filling the trap for dramatically improved sensitivity (≥10-fold). One drawback to this process is that fixed injection periods optimized for the LOQ can result in detector saturation at analyte concentrations below the desired upper limit of quantitation, resulting in a diminished dynamic range. However, the IonDrive™ High Energy Detector in recent QTRAP platforms allows ultra-fast pulse counting (10⁸ cps) with a higher saturation point without loss in low-end sensitivity. This feature provides the versatility required to accommodate a larger ion flux when using fixed injection time, thereby delivering dynamic ranges of four to five orders of magnitude.

Another approach for enhancing MRM³ sensitivity involves decreasing the resolution of the first resolving quadrupole (Q1), thereby improving the transmission efficiency of parent ions into the collision cell. In this manner, greater populations of primary fragment ions are generated for further dissociation in the LIT. Any losses in Q1 selectivity due to transmitting a wider window of ions is often overridden by the specificity gains of the resonance excitation event in the LIT, coupled with the mass selectivity of the excitation waveform. In fact, as demonstrated in Figure 3, the selectivity of the single-frequency waveform used for secondary ion dissociation of a monoisotopic peak is discerning enough to exclude ¹³C-isotopes⁴.

Conclusion
While MRM-based assays often provide the highly sensitive and selective quantitation requirements for analyzing surrogate peptides from complex biological background, there are instances when co-eluting interferences cannot be removed by these assays, and/or background response remains high, compromising achievable detection limits. The MRM³ functionality on SCIEX QTRAP Systems offers a promising approach for mining low-abundance peptides from high background noise while simplifying sample preparation strategies, thereby conferring a new dimension of selectivity and specificity to large molecule LC-MS assays.

1. Fortin T, Salvador A, Charrier J.P, Lenz C, Bettsworth F, Lacoux X, Choquet-Kastylevsky G, Lemoine J (2009) Multiple reaction monitoring cubed for protein quantification at the low nanogram/milliliter level of nondepleted human serum. Anal Chem 81: 9343-9352
2. Jeudy J, Salvador A, Simon R, Jaffuel A, Fonbonne C, Leonard J-F, Gautier J-C, Pasquier O, Lemoine J (2014) Overcoming biofluid protein complexity during targeted mass spectrometry detection and quantification of protein biomarkers by MRM cubed. Anal Bioanal Chem 406: 1193-1200
3. Korte R, Monneuse J-M, Gemrot E, Metton I, Humpf H-U, Brockmeyer J (2016) New high performance liquid chromatography coupled mass spectrometry method for the detection of lobster and shrimp allergens in food samples via multiple reaction monitoring and multiple reaction monitoring cubed. J Agric Food Chem 64: 6219-6227
4. Plomley JB, Makhloufi M (2013) Improved assay selectivity for the determination of hydroxymidazolam in capillary microsampling extracts using LC-MS³ on a hybrid linear ion trap. 61st ASMS Conference on Mass Spectrometry and Allied Topics, Minneapolis, MN