Determination of Estriol 16-glucuronide in human urine with surface plasmon resonance and lateral flow immunoassays Xiuqian Jiang, ab Mark Waterland, ab Len Blackwell b and Ashton Partridge * ab Received 6th November 2009, Accepted 25th January 2010 First published as an Advance Article on the web 8th February 2010 DOI: 10.1039/c001532f A rapid quantitative immunoassay for estriol-16-glucuronide by Surface Plasmon Resonance (SPR) sensing has been developed and applied to urine samples from non-pregnant and pregnant subjects. The assay was based on a partially-purified polyclonal antibody (pAb) raised in sheep, which showed negligible cross-reactivity with estrone-3-glucuronide and estriol-17-glucuronide. Colloidal gold coated by the pAb was used as the signal generator in the SPR-based inhibition immunoassay. An estriol-16-glucuronide-ovalbumin conjugate with an oligoethylene glycol (OEG) as linker was used to immobilize the steroid on the biosensor chip surface. The SPR assay had a limit of detection of 0.016 ng/mL, and could be performed rapidly giving results in two minutes. The assay can be carried out directly on any urine samples without complicated sample pretreatment. A one-step lateral flow strip test was also developed using the same pAb nanogold conjugates and bovine serum albumin estriol-16-glucuronide conjugates as the capture agent spotted onto a nitrocellulose membrane as the test line. A sensitive and repeatable lateral flow assay was achieved with a limit of detection of 0.49 ng/mL in time-diluted urine using a low coating concentration of the polyclonal antibody. Despite the strip sensor displaying adequate sensitivity in a standard curve generated by exposure to estriol-16- glucuronide in a spiked urine blank, the application of the strip sensors to real urine samples was not so successful due to matrix effects. Introduction Surface plasmon resonance (SPR) sensing is a highly sensitive label-free analytical technique, in which the sensor response is based on changes in refractive index that occur at the sensor surface upon binding of the target analyte. The high sensitivity, speed of response and the requirement for minimal sample highlights the applicability of SPR for immunoassays. Lateral flow immunoassays (LFIA) are cost-effective, relatively easy to perform and provide a more user-friendly analysis than most laboratory based assays. They can be stored for long periods without refrigeration, making them suitable alternatives for use in off-laboratory or resource-poor settings. Moreover, compared to SPR biosensor-based assays they can be rapidly brought to market with a relatively small investment. Estriol-16-glucuronide (E3-16G) is a steroid derived from ovarian estradiol by hepatic hydroxylation and conjugation and has been proposed as an alternative urinary metabolite to estrone glucuronide for the monitoring of ovarian function. 1 Metabolism of estradiol produces three estriol glucuronides, one conjugated at position 3, one at position 17 and one at position 16. Of the three, E3-16G is the preferred metabolite for monitoring fertility, since it is excreted rapidly whereas the estriol-3-glucuronide undergoes a complex entero-hepatic re-circulation before excre- tion and hence is delayed with respect to E3-16G. 2 E3-16G is also produced by the feto-placental unit during pregnancy and is the predominant estrogen in the urine of pregnant women increasing in parallel with the healthy growth of the foetus. 3 It has also been associated with breast cancer. 4 Much work has been done on development of assays for the measurement of estriols in the urine of pregnant women, 5 but there are few direct assays with the sensitivity to measure E3-16G levels in the urine of non-pregnant women. Detection methods for E3-16G have traditionally been based on radioimmunoassay (RIA), 6,7 high pressure liquid chromatography (HPLC), 8 and liquid chroma- tography (LC) coupled with fluorescence measurement, 9 mass spectrometry (MS) or UV spectrophotometry. 10,11 However, these analytical methods require many clean-up steps, are time- consuming and may be expensive. An accurate nanoparticle-enhanced SPR biosensor-based assay suitable for measurement of E3-16G in liquid samples has been described and validated in a previous study with a sensi- tivity close to 0.1 nmol/24 h (14 pg/mL). 12 Baker (1979) showed that the mean value of the mid cycle peak of E3-16G excretion using a direct radioimmunoassay method was approximately 75 nmol/24 h. 13 Therefore, the sensitivity of the SPR assay is more than sufficient to measure the low levels of E3-16G encountered in urine samples during the human menstrual cycle. A direct assay such as the SPR assay and LFIA is simpler and cheaper to perform than methods which need extraction or derivation but it introduces the possibility of matrix effects. All immunoassays are subject to matrix effects which can interfere with antibody binding reactions and hence give rise to false values. Measurement of a physiological marker such as E3-16G, which is present in low levels in a bodily fluid such as urine, a MacDiamid Institute for Advanced Materials and Nanotechnology, Private Bag, 11222 Palmerston North, New Zealand. E-mail: A. Partridge@massey.ac.nz; Fax: +64-6-350-5682; Tel: +64-6-350-5918 b Institute of Fundamental Science, Massey University, Private Bag, 11222 Palmerston North, New Zealand 368 | Anal. Methods, 2010, 2, 368–374 This journal is ª The Royal Society of Chemistry 2010 PAPER www.rsc.org/methods | Analytical Methods