4.2.3. Protein Analysis

The abundance of AGAT, GAMT, BBCK and CKMT1A in placental protein extracts (*n* = 19 control and *n* = 19 PE) was measured by western immunoblotting, as previously described [17]. Briefly, 40 μg protein/sample was separated on 4%–15% Criterion ™ TGX Stain-Free ™ precast gels in 10 × Tris/Glycine/SDS bu ffer solution (Bio-Rad, Gladesville, Australia) (for sample arrangemen<sup>t</sup> see Supplementary Information Tables S1 and S2). Two micrograms of human liver protein extract (Santa Cruz Biotechnologies sc-363766, Dallas, TX, USA) were included on gels for AGAT and GAMT assessment, as a positive control. Gels for creatine kinase B type and creatine kinase u-type mitochondrial assessment included 2 μg of human brain protein extract (Santa Cruz Biotechnologies sc-364375, Dallas, TX, USA) as the positive control. Proteins samples were transferred for 30 min onto Immobilin-FL PVDF membranes (Millipore, Billerica, MA, USA) and membranes scanned to quantify the total protein transferred using a Bio-Rad Gel Doc ™ XR + (Bio-Rad Laboratories, Hercules, CA, USA). Membranes were then blocked for 1 h with 5% skim milk powder/10% Tris-bu ffered saline with 0.1% Tween 20 (TBST) at room temperature. After blocking, membranes were incubated overnight at 4 ◦C with primary antibodies for AGAT (1:100; Atlas Antibodies HPA026077, Bromma, Sweden), GAMT (1:1000; Monash Antibody Technologies Facility, Melbourne, Australia); Creatine kinase B type (1:5000; Abcam BBCK ab92452, Cambridge, UK) or creatine kinase u-type mitochondrial (1:1000; Atlas CKMT1A HPA043491) in 5% skim milk powder/TBST.

Membranes were incubated for 1 h with fluorescent secondary antibodies (Anti-Mouse IgG (H+L) DaylightTM 680 Conjugate or Anti-Rabbit IgG (H+L) DaylightTM 800 Conjugate; Cell Signalling Technologies ®, Danvers, MA, USA). Membranes were exposed on an Odyssey ® Infrared Imaging System (LI-COR Biosciences, Lincoln, NE, USA) and individual protein band optical densities were determined using Image Studio Lite software (V5.2.5; LI-COR Biosciences, Lincoln, Nebraska, USA). Optical densities for each sample were quantified and then normalized to the total protein transferred onto the blot for that sample [49]. To account for any inter-blot variability, results were then normalized further using the optical density generated by an internal control (healthy term human placenta sample) run on each blot. Data is expressed relative to the control cohort.
