Automated Middle-level Antibody Analysis using On-column FabRICATOR® Digestion with LC-MS

Middle-level analysis by LC-MS is a widely accepted analytical method for rapid characterization of mAbs and other IgG-based biopharmaceuticals. Here, we present a rapid, automated solution for antibody subunit generation and analysis using a standard HPLC-MS setup with only minor modifications. FabRICATOR (IdeS) enzyme was immobilized in an HPLC column format to allow for on-column digestion of IgG-based biologics. The resulting fragments are trapped on the column head of a reverse phase (RP) column connected in-line with the FabRICATOR-HPLC column. After uncoupling the enzyme column from the flow path using a column switching valve, the fragments can be reduced by injection of a reducing agent before being separated by RP-HPLC and analyzed by mass spectrometry. This allowed for a fully automated, completely hands-off workflow for analysis of mAb quality attributes.

Application Notes Middle-level analysis

Application Note

Download the FabRICATOR-HPLC Application Note



FabRICATOR-HPLC is available for ordering. Buy online or visit 'Place an Order'. 


  • Automated subunit generation using FabRICATOR-HPLC
  • Hands-off middle-level LC-MS analysis of IgG based biopharmaceuticals
  • On-column FabRICATOR digestion, reduction and LC-MS analysis
  • Robust performance over >10 days of continuous operation

Fabricator-HPLC Middle-level analysis


Monoclonal antibodies (mAbs) and other IgG-based biopharmaceuticals are increasingly entering clinics and improving patients’ quality of life. The inherent heterogeneity of such biologics necessitates detailed characterization by liquid chromatography and mass spectrometry (LC-MS). While bottom-up peptide mapping is still the gold standard for analysis of critical quality attributes, such approaches are resource and time intensive in both data acquisition and analysis. Top-down and middle-down approaches are therefore gaining in popularity. Antibody subunit analysis has become a widely accepted analytical strategy for rapid characterization of therapeutic antibodies and related products (1). The FabRICATOR (IdeS) enzyme specifically digests IgG just below the hinge (2), which after reduction of disulfide bonds yields antibody fragments of 23-25 kDa that are amenable to high-resolution mass spectrometry. The FabRICATOR-based middle-level LC-MS workflow therefore enables analysis of multiple antibody quality attributes such as glycosylation, oxidation, glycation and C-terminal lysine clipping as well as sequence confirmation.

The quality attributes of a mAb change over the course of a cell culturing process and are highly dependent on the process parameters. Therefore, optimization is needed until a final, robust manufacturing process is developed that produces mAbs with the desired quality. Process development requires the testing of many different parameters and the characterization of the resulting antibodies, leading to a large number of samples to be analyzed. In an effort to reduce sample handling and provide an automated solution for sample preparation, FabRICATOR was covalently immobilized on a resin and packed in PEEK column hardware suitable for HPLC. This analytical tool allowed for a fully automated workflow for middle-level analysis of mAbs by LC-MS. The method is well suited to any standard HPLC-MS system with only minor modifications needed and could potentially be connected directly to a bioreactor for automated monitoring of an on-going mAb production.

Materials & Methods

Sample Preparation

Trastuzumab and adalimumab were obtained in manufacturer’s formulation buffer. Before analysis, each antibody underwent a buffer exchange to 150 mM NH4OAc, pH 7 using a ZebaSpin desalting column (0.5 ml, 40 kDa MWCO, Thermo Scientific) and was diluted to 1 mg/ml. For in-solution digestion with FabRICATOR, 50 µg buffer exchanged antibody was treated with 50 units FabRICATOR in 15 µl PBS for 30 min at 37°C. The digested antibodies were reduced and denatured by adding 25 µl 8 M GdHCl and 10 µl 500 mM DTT (final concentration 1 mg/ml mAb in 4 M GdHCl, 100 mM DTT) and incubated for 30 min at 37°C before analysis.

FabRICATOR-HPLC Column Activation

The column was activated by injecting 3 x 20 µl 20 mM DTT in MQ water into the FabRICATOR-HPLC column at a flow rate of 50 µl/min in 150 mM NH4OAc, pH 7 followed by a column wash at 0.2 ml/min for 2 hours.

Determination of the Digestion Buffer’s Optimal Ionic Strength

The FabRICATOR-HPLC column was coupled directly to the UV detector of the LC system and run at a flow rate of 50 µl/min. Different buffer ionic strengths were achieved by mixing 10 mM NH4OAc, pH 7 and 500 mM NH4OAc, pH 7 at appropriate ratios using the mixing capabilities of the HPLC system’s quaternary pump. 10 µg trastuzumab was injected and isocratic elution was followed for 20 min by measuring UV absorption at 280 nm wavelength.

Automated mAb Subunit Analysis

HPLC setup
Pump: Agilent 1260 Infinity quaternary pump
Columns: FabRICATOR-HPLC, 2.1 x 50 mm (37°C), Thermo Scientific MAbPac RP, 2.1 x 100 mm (70°C)
Valves: 2 x 2pos/6port, divert valve (DV) in the MS source
Mobile phases: A: 0.1% FA in water, B: 0.1% FA in 95% ACN, C: 150 mM NH4OAc, pH 7
Detectors: VWD detector (Abs 280 nm)

MS setup
Instrument: Bruker Impact II ESI-Q-TOF
Mass range: 300-3000 m/z
ESI source voltage: 4.5 kV
Nebulizer gas pressure: 1.8 bar
Drying gas flow rate: 8 l/min
Drying gas temperature: 220°C


1. On-column mAb Digestion

Middle-level analysis 1

Note: The times given are applicable to samples of a concentration of >0.5 mg/ml. For lower concentrations the method has to be adjusted to account for the longer injection time and broader elution peak. Elution times might also depend on the configuration of the LC system and the molecule to be digested. Some optimization might be necessary.


2. FabRICATOR-HPLC Column Wash

Middle-level analysis 2


3. Mobile Phase Change to 0.1% FA, ACN

Middle-level analysis 3


4. Equilibration of the Reverse Phase Column

Middle-level analysis 4

Note: To avoid damage to the column, the flow rate is decreased to 0.05 ml/min before the valve switch and then increased gradually.


5. Reduction

Middle-level analysis 5

Note: The reduction method is repeated four times. The first three times, TCEP is injected followed by a blank injection to allow for the last TCEP injection to pass the column at a low flow rate before desalting.


6. Desalting

Middle-level analysis 6

Note: To avoid damage to the column, increase the flow rate gradually.


7. Analysis

Middle-level analysis 7


8. Re-equilibration of the Reverse Phase Column

Middle-level analysis 8


9. Mobile Phase Change to 150 mM NH4OAc, pH 7

Middle-level analysis 9

Note: The delay volume of the pump must be carefully taken into account with great care required to avoid low pH mobile phases from the reverse phase chromatography reaching the FabRICATOR-HPLC column. The FabRICATOR enzyme is irreversibly inactivated at a pH <5.


10. FabRICATOR-HPLC Column Equilibration

Middle-level analysis 10

Results & Discussion

Influence of the Ionic Strength of the Digestion Buffer on mAb Retention on FabRICATOR-HPLC

Non-volatile buffer salts can cause problems in LC-MS analyses such as ion suppression and adducts. Therefore, the PBS buffer usually used for FabRICATOR digestion is not that well suited when the FabRICATOR-HPLC column is coupled in-line with an LC-MS system. Ammonium acetate (NH4OAc) can be used as an MS-friendly alternative. Digestion performance is unaffected by the new buffer, however, a minimum of 150 mM NH4OAc is necessary to achieve fast and reliable elution of the substrate protein from the FabRICATOR-HPLC column. At low ionic strength of the mobile phase, the mAbs are retained and elute as broad peaks or not at all. Especially the F(ab’)2 fragments exhibit high affinity for the column at low salt concentrations (Fig. 1). All IgG1 mAbs tested exhibited the same elution properties from the FabRICATOR-HPLC column but other IgG subclasses or Fc-fusion proteins might exhibit different retention characteristics and may require some optimization.

Figure 1: Elution from the FabRICATOR-HPLC column at different salt concentrations. 10 µg trastuzumab was injected onto the FabRICATOR-HPLC column at a flow rate of 50 µl/min and elution was monitored for 20 min. In low ionic strength mobile phases, the mAb elutes as low, broad peaks with the Fc/2 eluting first and the F(ab’)2 being retained longer.


Automated Middle-level Analysis of Trastuzumab

FabRICATOR-HPLC allows for on-column digestion of mAbs and related molecules and therefore automated sample preparation for middle-level analysis by mass spectrometry. The HPLC configuration is shown in Figure 2. The FabRICATOR-HPLC column is equilibrated in 150 mM NH4OAc, pH 7 (orange) and a mAb sample is injected onto the column. There the antibodies are digested into Fc and F(ab´)2 fragments (Fig. 2, Step 1). These fragments flow directly to a reverse phase column and are trapped on the column head. At this point the FabRICATOR column is uncoupled from the flow path by means of a column switching valve (Fig. 2. V1) and the mobile phases are changed to 0.1% FA, ACN (teal) suitable for reverse phase chromatography. The mAb fragments are reduced by injection of 50 mM TCEP solution onto the RP column (Fig. 2, Step 2). The high column temperature, together with low pH and organic solvent (20% ACN) in the mobile phase, help to reduce and denature the antibody. This results in separation of the F(ab´)2 fragment into LC and Fd’. The reduction step can be omitted, if only the Fc/2 fragment is of interest (e.g. for analysis of Fc glycosylation). Finally, the mAb subunits can be separated by RP-HPLC (Fig. 2, Step 3) and analyzed by mass spectrometry (Fig. 2, Step 4). This yielded MS spectra virtually indistinguishable from the ones obtained using the standard in-solution FabRICATOR protocol (4; Fig. 3).


Middle-level analysis fig 12


Figure 2: An automated workflow for middle-down analysis of mAbs


Middle-level analysis fig 13

Figure 3. Comparison between digestion on FabRICATOR-HPLC and FabRICATOR in solution. a) UV chromatogram of the trastuzumab subunits generated by the in-solution FabRICATOR protocol (left, orange) and the automated FabRICATOR-HPLC workflow (right, teal). The asterisk marks LC fragments that are not completely reduced with one intramolecular disulfide bridge intact. b) Deconvoluted MS spectra of the Fc/2 fragment generated by in-solution FabRICATOR digestion (left, orange) or the automated FabRICATOR-HPLC workflow (right, teal).


Continuous Operation

A typical cell culturing process for expression of a monoclonal antibody takes between 10-14 days (5). To assess suitability of FabRICATOR-HPLC for monitoring of mAb CQAs during this entire period, we tested the automated middle-level LC-MS approach for 14 days of continuous operation. A new trastuzumab sample (2 µg) was injected every 4 hours and digestion performance was measured by quantification of the intact HC as well as the Fc/2 and Fd’ fragments from the UV chromatogram.
In excess of 95% of the sample was digested during the entire 14-day period (Fig. 4a). We also analyzed Fc glycosylation by middle-level LC-MS and compared it to results obtained from a standard in-solution FabRICATOR workflow (4). The Fc glycosylation profiles we obtained using the automated FabRICATOR-HPLC workflow were stable during the entire 14-day experiment with standard deviations of <0.5% for all glycoforms. The results also corresponded well to a data set produced using the in-solution FabRICATOR protocol (Fig. 4b) as well as previously published data on trastuzumab glycosylation (3).


Figure 4: FabRICATOR-HPLC performance during a 14-day period of continuous operation.a) Digestion efficiency of FabRICATOR-HPLC was followed for 14 days with 6 trastuzumab samples (2 µg each) being analyzed automatically per day. The digestion percentage was determined from the obtained UV chromatograms.
b) Comparison of the obtained Fc glycosylation profiles from middle-level LC-MS analysis using the standard in-solution FabRICATOR protocol (n=10, orange) or the automated FabRICATOR-HPLC workflow (n=28, 2 samples per day, teal). The bars represent the mean, the error bars the standard deviation.


Minimal Carry-over Between Samples

We evaluated the carry-over between samples by alternating injections of trastuzumab and buffer as well as trastuzumab and adalimumab and then quantified carry-over of trastuzumab subunits to the blank run or the adalimumab samples. Carry-over was virtually undetectable by UV-HPLC (<1%, Fig. 5a) and while still detectable by MS, this carry-over did not have any effects on the results of the analysis of the subsequent mAb (Fig. 5b). Further dissection of the system revealed that the carry-over was mostly caused by the RP column and could be lowered further by a more rigorous column washing procedure in-between samples.

Figure 5: Evaluation of sample carry-over on FabRICATOR-HPLC. a) We quantified carry-over of trastuzumab subunits from previous injections to a blank and an adalimumab sample respectively. The UV chromatograms show only very minor signal corresponding to the antibody fragments of the previous injection. b) Comparison between deconvoluted Fc/2 MS spectra of trastuzumab (top, orange) and adalimumab (bottom, teal). Note that none of the trastuzumab glycoforms are detectable in the adalimumab sample and vice versa. The asterisk marks antibody fragments that have not undergone C-terminal lysine clipping.


Method Applicability

The described workflow is applicable to IgG1-based mAbs, ADCs and Fc fusion proteins. However, the HPLC separation and MS acquisition parameters might need to be adjusted to the molecule of
interest. Sample concentrations between 0.025 and 2 mg/ml had no impact on digestion performance or the obtained Fc glycosylation profiles. However, low sample concentrations require method adjustments to account for the longer injection time and consequently a broader elution peak from the FabRICATOR-HPLC column. To date we have only tested this workflow using clean substrate proteins in neutral pH aqueous buffers without any additives such as detergents or chaotropic salts. Unpurified samples or mAbs in low pH buffers (e.g. after elution from a Protein A column) might negatively affect column performance and/or lifetime and require further optimization of the methodology.


Taken together, the results presented here show that FabRICATOR-HPLC allows for fully automated sample preparation for middle-level LC-MS analysis of mAbs, lowering the possibility for sample handling errors and increasing throughput. The results obtained for Fc glycosylation are comparable to those obtained by a well-established in-solution FabRICATOR digestion protocol (3). We achieved constant and reproducible digestion performance (>95%) and consistent mAb analysis results for a period of up to 14 days of continuous operation. Due to the fact that the digested mAb fragments are trapped and concentrated on the RP column prior to analysis we were able to analyze mAb samples of very low concentrations (0.025 mg/ml). This might be of critical importance when monitoring the early stages of a cell culturing process. Finally, the automated middle-level LC-MS workflow using FabRICATOR-HPLC showed very low carry-over between consecutive samples making it suitable for analysis of several different mAbs, for example samples drawn from several bioreactors running in parallel during process development.


  1. Sjögren J., et al. Analyst, 2016 May 23;141(11):3114-25.
  2. von Pawel-Rammingen U., et al. EMBO J. 2002 vol. 21 No 7 pp. 1607-1615
  3. Sanchez-De Melo I., et al. J Proteomics. 2015 Sep 8;127(Pt B):225-33.
  4. Genovis application note A0FR201806: Antibody Fc Glycan Analysis by FabRICATOR® Digestion and LC-MS
  5. Li F., et al. MAbs. 2010 Sep-Oct; 2(5): 466–477.


Watch the FabRICATOR® Movie

Learn more about how FabRICATOR works.


Download FabRICATOR-HPLC Posters

Download posters on FabRICATOR-HPLC here.


Read more about our IgG Proteases

We have other IgG-specific proteases available. Read more here!