Development of a New Progesterone Metabolite EIA/ELISA for Non

Development of a New Progesterone Metabolite EIA/ELISA for Non-Invasive Reproductive Monitoring by Maheeka Embogama, Bobbi O’Hara and Russ Hart, Arbor Assays Inc., Ann Arbor, MI

Introduction K068-H1/H5 Assay Validation

Free and conjugated steroid hormones (estrogens, androgens, and progestogens) can be detected in urine and feces K068-H1/H5 Typical Standard Curve

of female animals across reproductive stages. As steroids are passed through the gut they can be transformed to more 100 1.2 soluble metabolites, via reduction, hydroxylation and/or converted into water-soluble glucuronides. EIA/ELISA testing of 90 these fecal steroids, specially progesterone metabolites is a convenient method to analyze large numbers of samples 1.0 both in the laboratory and in field settings to assess reproductive status. However fecal metabolites vary considerably 80 %/0 from species to species and it has been suggested that there should be development of specific antibodies and assays 70 Net OD 0.8 for each species. It is obviously a long, expensive and complex task to develop an assay with the correct reactivities to 60 every species fecal metabolite. 50 0.6 In this study we have developed an EIA/ELISA specifically for progesterone metabolites. We conjugated two 40 3-hydroxyprogesterone derivatives, 5α- and 5β-pregnan-3β (3α)-ol-20-one hemisuccinates, to keyhole limpet hemocyanin 0.4 (KLH) (for immunization), bovine serum albumin (BSA) for evaluating titers and horseradish peroxidase (HRP) for conjugate 30

use. We discuss the development, evaluation and testing of these antibodies and the ensuing progesterone metabolite 20 0.2 EIA/ELISA kit. 10

0 0.0 100 1,000 10,000 Methods Cross Reactivity, AAL15-CL425 Comparison Antibody Generation K068-H1/H5 K025-H1/H5 5α-pregnan-3β-ol-20-one hemisuccinate and 5β-pregnan-3α-pregnan-3β-ol-20-one hemisuccinate in anhydrous DMF AAL15 CL425 Reactant Progesterone Metabolite Monoclonal Progesterone were activated using N-hydroxysuccinimde and dicyclocarbodiimide in DMF overnight at 4°C desiccated. The activated Progesterone 100 100 steroids were added in excess to keyhole limpet hemocyanin (KLH) for immunization, BSA for evaluating titers, and HRP for conjugate use, all dissolved in borate buffer, pH 8.5 containing DMF. After 1 hour stirring at room temperature the KLH and 3β-hydroprogesterone - 172

BSA conjugates were dialyzed overnight at 4°C against PBS. 3α-hydroprogesterone - 188

The HRP conjugate was purified on an Amersham PD10 column using PBS as 5β-dihydroprogesterone 61.9 - 5ß HRP Conjugate 3 the elution buffer. Ten 1mL fractions were collected. Analysis using the 404/275nm

275 nm 5α-dihydroprogesterone 56.7 7 optical density ratio was used to identify conjugate fractions. Antibodies were 2.5 404 nm 404/275nm Ratio Pregnanolone (5β-Pregnan-3α-ol-20-one) 41.2 0.12* 2 raised to the KLH and BSA conjugates at a facility in Pennsylvania. Three rabbits each were immunized with the KLH 5α- and 5β- conjugates. Antibody titers were 1.5 Epiallopregnanolone (5α-pregnan-3β-ol-20-one) 38.3 - OD assessed by both using BSA-conjugated 5α- and 5β- steroid conjugates and in 1 Allopregnanolone 27.3 - titer assays using the 5α- and 5β-HRP conjugates. Any positive rabbit bleeds 0.5 were tested for steroid displacement using Arbor Assays (AAI) Goat anti-rabbit Pregnenolone 17.6 5.9 0 1 2 3 4 5 6 7 8 9 10 IgG plates, X016-1EA, with AAI Assay Buffer (AB10), X065, Wash Buffer, X007, Fraction Number Epipregnanolone 10.2 - TMB Substrate, X019, Stop Solution (1M HCl), X020, and the HRP conjugate diluted into AAI Conjugate Diluent, X076. 17α-hydroxyprogesterone 5.7 0.38

We choose antibody AAL15, an antibody raised to 5β-pregnan-3α-pregnan-3β-ol-20-one, because of sensitivity along with 11α-hydroxyprogesterone 4.9 147 reactivity to structurally related steroids. AAL15 also showed zero or low reactivity to stress steroids such as cortisol, 11β-hydroxyprogesterone - 2.7 androgens such as testosterone, estrogens such as estradiol and E1G (data not shown). Rabbit AAL15 is still alive and 20α-hydroxyprogesterone 0.34 - giving us production bleeds of serum. She was initially immunized in the summer of 2017. She gets boosted and bleed on a regular schedule. AAL15 serum also has quite high titer, approximately 1:40,000 dilution, which ensures long lasting supply. Allopregnandiol 0.29 -

Progesterone Metabolite Assay Testosterone 0.18 - Cortisol <0.1 <0.04 The assay follows our standard template. 50 µL of samples or standards diluted in AB10 are added to the secondary antibody coated plate, a goat anti-rabbit IgG clear plate, X016. 25 µL of conjugate diluted in conjugate diluent, X076 is added and 17β-Estradiol <0.1 - the assay initiated by addition of the rabbit polyclonal antibody. The plate is shaken at RT for 2 hours, followed by washing Corticosterone <0.1 <0.1 and addition of 100 µL of TMB substrate, X019. Color generation is carried out at room temperature for 30 minutes when the reaction is terminated by addition of 50 µL stop solution. Generated signal is read at 450 nm. Data was analyzed using Androstenedione <0.1 <0.1 a Molecular Devices SoftMax 4PLC template. * run by Coralie Munro

Factors tested during research and development and validation were: Alcohol Interference, Cross reactivity; Linearity; Precision Precision, and Sample performance. Intra Assay Precision: Three human samples were diluted with Assay Buffer and run in replicates of 20 in an assay. The mean and precision of the calculated Progesterone concentrations were:

Sample Progesterone Conc. (pg/mL) %CV 1 2,377 4.9

Sample Data 2 1,517 7.2 from a Maned Wolf, (Chrysocyon brachyurus) (kind gift of Dr. Rachel Santymire, Lincoln Park Zoo, Chicago, IL) Urine samples 3 710.3 8.0 were diluted in the kits assay buffer, AB10, and run in our monoclonal CL425 based assay kit, K025-H1/H5 and the progesterone metabolite kit, K068-H1/H5. Creatinine levels were determined on the urine samples using Arbor Assays Urinary Creatinine kit, Inter Assay Precision: Three human samples were diluted with Assay Buffer and run in duplicates in nineteen assays run K002-H1/H5. The time course of pregnancy and the correlation data are shown below. over multiple days by six operators. The mean and precision of the calculated Progesterone concentrations were:

Maned Wolf Urine, pre, during & post pregnancy Sample Progesterone Conc. (pg/mL) %CV K025-H1/H5 CL425 Ab (Blue) and K068-H1/H5 AAL15 Ab (Red) 200 100 K025-H1/H5 Ab = CL425 1 2,442 6.8 180 90 100 K068-H1/H5 Ab = AAL15

90 y = 0.3848x - 1.4161

5 160 80

2 2 1,665 6.6 4 80 2 L 140 70 R = 0.8353 C 70 a i 120 60 3 844.6 8.8 v

) 60 t

r 100 50

C P4 AAL1550 g 80 40 m

/ 40 Linearity: Linearity was determined by taking two urine samples diluted with Assay Buffer, one with a low diluted progesterone

g 60 30 n P4 AAL15

( 30

e 40 20 level of 179.2 pg/mL and one with a higher diluted level of 1,729 pg/mL, and mixing them in the ratios given below. The n

o 20 r

e 20 10 t measured concentrations were compared to the expected values based on the ratios used.

s 10 e 0 0 g

o 0

r 1 3 4 2 19 3 12 P 106 117 122 130 123 218 313 317 10 1122 1214 1229 0 20 209 0 0 0 04 1009 7 0 0 7 7 7 0407 0 50 100 150 200 05110051109051118 0511 2 06 06 06 7 7 7 050913050916 0512005122506 0 060 060 060 06 06 0 0 0 0 0 0 0 0 Low Urine High Urine Expected Conc. (pg/mL) Observed Conc. (pg/mL) % Recovery P4 CL425 80% 20% 489.1 519.9 106.3

Note the high level of correlation between the measurements with the sample from the Maned Wolf. The measured 60% 40% 799.1 797.5 99.8

concentrations are lower with the progesterone metabolite kit, K068-H1/H5, generating about 38.5% of the concentration as 40% 60% 1,109 1174 105.9 our CL425-based assay. 20% 80% 1,419 1451 102.2 Fecal samples from a pregnant Iberian Lynx, (Lynx pardinus) (kind gift of Prof. Martin Dehnhard, Leibniz Institute for Zoo Mean Recovery 103.6% and Wildlife Research, Berlin, Germany) were extracted, concentrated and diluted in assay buffer and run in our monoclonal CL425 based assay kit, K025-H1/H5 and the progesterone metabolite kit, K068-H1/H5. The time course of pregnancy and 1600 the correlation data for the Iberian Lynx are shown below.

1400 Time Course from Iberian Lynx Fecal Extracts in K025-H1/H5 and K068-H1/H5

Data from Extracts of Fecal Samples from an Iberian Lynx 1200 Measured in CL425 and AAL15 EIA Assays

1,400 5,000

CL425 ng/mL 1000 AAL15 ng/mg 4,500 y 1.0223x 10.241 1,200 R 0.99604

4,000 800 Observed Conc. (pg/mL) 1,000 3,500 600 3,000 800

2,500 400 400 600 800 1000 1200 1400 1600 600 2,000 Expected Conc. (pg/mL) Progesterone ng/mL AAL15 1,500 Progesterone ng/mL CL425 Progesterone 400

1,000 Alcohol Interference: The assay will tolerate up to 5% ethanol in sample volume before any significant reduction in binding 200 500 is observed.

0 0 40 45 50 55 60 65 70 75 Days Post Mating Data from Extracts of Fecal Samples from an Iberian Lynx Measured in CL425 and AAL15 EIA Assays

1,400 5,000 Correlation of Progesterone Measured in CL425 and AAL15 assays on dilutions of fecal extracts from an Iberian Lynx Summary CL425 ng/mL AAL15 ng/mg 4,500 5,000 1,200 We have developed an EIA/ELISA based upon antibodies raised to 5β-Pregnan-3α-ol-20-one. The assay utilizes a polyclonal 4,500 y = 2.6633x + 676.59 4,000 R = 0.32339 antibody from a rabbit that we are maintaining and obtaining production quantities of sera from on a scheduled basis. We

4,000 1,000 currently have on hand in Ann Arbor enough polyclonal antibody for over 500,000 plates, so any supply issues will not be a 3,500 3,500 problem for many years.

3,000 3,000 800 The assay shows good sensitivity for measuring progesterone and its metabolites in at least 2 widely used sample matrices, 2,500 2,500 urine and fecal extracts. Interestingly for urine samples from a pregnant wolf the commonly used monoclonal CL425 based

2,000 600 assay and this new antibody show very similar profiles. Fecal extracts from a felid showed a biphasic pattern about 2 weeks 2,000 Prgesterone ng/mL AAL15 Prgesterone 1,500Lorem ipsum prior to birth. The two assay systems gave good correlation with time from mating when the first 62 days were evaluated with Progesterone ng/mL AAL15 1,500 Progesterone ng/mL CL425 Progesterone 400 1,000 a slope of 4.28 higher concentration with the new antibody than CL425 and a correlation coefficient of 0.894. If the last 13

500 days of data is included the correlation coefficient drops to 0.323 as the CL425 measured signal drops and the new antibody 1,000

200 0 shows its highest concentration just prior to parturition. 500 0 200 400 600 800 1,000 1,200 1,400 Progesterone ng/mL CL425

0 0 40 45 50 55 60 65 70 75 Days Post Mating www.ArborAssays.com