Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 May 1.
Published in final edited form as: J Zoo Wildl Med. 2023 Jan;53(4):801–810. doi: 10.1638/2021-0080

RELATIONSHIP BETWEEN REPRODUCTIVE AND BONE BIOMARKERS AND OSTEOARTHRITIS IN ZOO ASIAN AND AFRICAN ELEPHANTS

Daniella E Chusyd 1, Janine L Brown 1, Lilian Golzarri-Arroyo Stephanie L Dickinson 1, Virginia B Kraus 1, Jessica Siegal-Willott 1, Timothy M Griffin 1, Janet L Huebner 1, Katie L Edwards 1, David B Allison 1, Steven N Austad 1
PMCID: PMC10150656  NIHMSID: NIHMS1883650  PMID: 36640083

Abstract

Osteoarthritis (OA) is common in zoo Asian (Elephas maximus) and African (Loxodonta africana) elephants. This study investigated the relationship between confirmed or suspected OA with ovarian cyclicity, gonadotropins, progestagens, luteinizing hormone (LH), follicular stimulating hormone (FSH), and collagen type I (CTX-I) in zoo elephants. In Asian elephants, odds of having confirmed or suspected OA decreased with cycling (OR = 0.22, P = 0.016; OR = 0.29, P = 0.020, respectively), however, not when adjusted for age (odds ratio [OR] = 0.31, P = 0.112; OR = 0.58, P = 0.369, respectively). In African elephants, none of the models between confirmed OA and cycling status were significant (P > 0.060), while odds of having suspected OA decreased with cycling (OR = 0.12, P = 0.001), even after adjusting for age (OR = 0.15, P =0.005). Progestagens (Asians: P > 0.096; Africans: P > 0.415), LH (Asians: P > 0.129; Africans: P > 0.359), and FSH (Asians: P > 0.738; Africans: P > 0.231) did not differ with confirmed or suspected OA status, unadjusted. CTX-I concentrations were not related to OA status (P > 0.655). This study concluded hormonal changes may not have a strong impact on OA, so additional investigation into other serologic biomarkers is warranted.

Introduction

Arthritis, referring to inflammation of a joint,19 and specifically osteoarthritis (OA), denoting thinning of joint cartilage, fluid within the joint, and/or bony growth formations around the joint,15 have long been documented as common health concerns in zoo Asian (Elephas maximus) and African (Loxodonta africana) elephants. A recent survey in North America found that 36% of institutions that hold elephants had at least one elephant with arthritis;18 secondary complications and progressive disease often lead to euthanasia. Risk factor identification and early OA diagnosis could facilitate better treatment.

Reproductive status, particularly menopause, has been consistently associated with arthritis in women.5 Endogenous sex hormones may play a protective role against the development and progression of OA by preserving cartilage and subchondral bone.26 Many zoo elephants exhibit abnormal ovarian cycles with baseline progestagen patterns similar to those of post-menopausal women.2 However, studies on whether or how luteinizing hormone (LH), follicle stimulating hormone (FSH), or sex hormones influence the development of OA in elephants are lacking.

Definitive diagnosis of elephant OA is challenging. Typically, radiography is used, but it is time and labor intensive, limited to more distal extremities, and not all elephants are compliant. By contrast, most elephants are trained for blood collection, so identification of serum bone biomarkers to assess OA status could allow early detection and response to treatments. A major component of OA is cartilage degradation and bone turnover. Collagen type I (CTX-I) primarily makes up bone, tendon, and skin;11 serum CTX-I fragments generated during bone loss represent a potential marker of bone catabolism.9 The goal of this study was to investigate whether differences in progestagens, LH, FSH, and CTX-I concentrations were associated with ovarian cyclicity and OA status in elephants.

Methods

This study was approved by the Institutional Animal Care and Use Committees of participating zoos and author institutions. Female elephants (70 Asian, 58 African) housed at 34 and 13 zoological facilities, respectively, which were part of a larger elephant project,4 were included in this study (Tables 1, 2). Serum blood collected in 2012 was stored frozen at −20ºC or colder, with no more than two freeze and thaw cycles prior to analysis.

Table 1.

Descriptive statistics for Asian and African elephants with COA or SOA used in hypotheses: noncycling elephants will have higher rates of COA or SOA, and cycling elephants without COA or SOA will have higher progestagen and gonadotropin concentrations compared with cycling elephants with COA or SOA.a

Asian elephants
Ovarian cyclicity and COA status in Asian elephants
Age (yr) BCS
Overall (n = 57) Cycling elephants (n = 40) Noncycling elephants (n = 17) Overall (n = 53) Cycling elephants (n = 37) Noncycling elephants (n = 16) Hard surfaces (n = 56)
38.5 ± 10.4(12–60) 36.2 ±11.1 (12–52) 43.7 ± 6.3 (36–60) 4.26 ± 0.98 (1–5) 4.27 ± 0.84 (2–5) 4.25 ± 1.29 (1–5) 11.3 ± 14.8% (0–55.9%)
Sex and gonadotropin hormones and COA status in cycling Asian elephants
Overall age (n = 38) Age for elephants with COA (n = 8) Age for elephants without COA (n = 30) Overall BCS (n = 35) BCS for elephants with COA (n = 8) BCS for elephants without COA (n = 27) Hard surfaces (n = 38)
36.7 ± 10.6 (12–52) 44.9 ± 4.1 (40–52) 34.5 ± 10.8 (12–52) 4.23 ± 0.94 (2–5) 4.50 ± 0.76 (3–5) 4.15 ± 0.99 (2–5) 11.3 ± 14.4% (0–55.9%)
Ovarian cyclicity and SOA status in Asian elephants
Age (yr) BCS
Overall (n = 70) Cycling elephants (n = 48) Noncycling elephants (n = 22) Overall (n = 66) Cycling elephants (n = 45) Noncycling elephants (n = 21) Hard surfaces (n = 69)
39.7 ± 10.6(12–64) 36.9 ± 10.5 (12–52) 45.8 ± 8.0 (36–64) 4.24 ± 0.96 (1–5) 4.31 ± 0.82(2–5) 4.10 ± 1.22 (1–5) 10.3 ± 14.0% (0–55.9%)
Sex and gonadotropin hormones and SOA status in cycling Asian elephants
Overall age (n = 45) Age for elephants with SOA (n = 15) Age for elephants without SOA (n = 30) Overall BCS (n = 42) BCS for elephants with SOA (n = 15) BCS for elephants without SOA (n = 27) Hard surfaces (n = 45)
37.2 ± 10.0 (12–52) 42.8 ± 5.8 (29–52) 34.5 ± 10.8 (12–52) 4.29 ± 0.92 (2–5) 4.53 ± 0.74 (3–5) 4.15 ± 0.99 (2–5) 10.5 ± 14.0% (0–55.9%)
African elephants
Ovarian cyclicity and COA status in African elephants
Age (yr) BCS
Overall age (n = 50) Age for cycling elephants (n = 34) Age for noncycling elephants (n = 16) Overall BCS (n = 47) BCS for cycling elephants (n = 31) BCS for noncycling elephants (n = 16) Hard surfaces (n = 50)
33.1 ± 6.3 (7–52) 32.3 ± 6.6 (7–52) 34.8 ± 5.6 (27–44) 4.21 ± 0.75 (3–5) 4.19 ± 0.83 (3–5) 4.25 ± 0.58 (3–5) 10.3 ± 14.1% (0–66.7%)
Sex and gonadotropin hormones and COA status in cycling African elephants
Overall age (n = 19) Age for elephants with COA (n =5) Age for elephants without COA (n = 14) Overall BCS (n = 17) BCS for elephants with COA (n = 4) BCS for elephants without COA (n = 13) Hard surfaces (n = 19)
34.4 ± 6.2 (27–52) 37.4 ± 9.7 (27–52) 33.4 ± 4.5 (27–43) 4.41 ± 080 (3–5) 5.00 ± 0.00 (5–5) 4.23 ± 0.83 (3–5) 10.7 ± 10.7% (0–32.2%)
Ovarian cyclicity and SOA status in African elephants
Age (yr) BCS
Overall (n = 58) Cycling elephants (n = 34) Noncycling elephants (n = 24) Overall (n = 55) Cycling elephants (n = 31) Noncycling elephants (n = 24) Hard surfaces (n = 58)
33.6 ± 6.3 (7–52) 32.3 ± 6.6 (7–52) 35.5 ± 5.4 (27–44) 4.27 ± 0.73 (3–5) 4.19 ± 0.83 (3–5) 4.38 ± 0.58 (3–5) 10.7 ± 15.5% (0–66.7%)
Sex and gonadotropin hormones and SOA status in cycling African elephants
Overall age (n = 20) Age for elephants with SOA (n = 6) Age for elephants without SOA (n = 14) Overall BCS (n = 18) BCS for elephants with SOA (n = 5) BCS for elephants without SOA (n = 13) Hard surfaces (n = 20)
34.4 ± 6.1 (27–52) 36.7 ± 8.9 (27–52) 33.4 ± 4.5 (27–43) 4.39 ± 0.78 (3–5) 4.80 ± 0.45 (4–5) 4.23 ± 0.83 (3–5) 11.7 ± 11.3% (0–32.2%)
Open in a new tab
a

COA, confirmed osteoarthritis; SOA, suspected osteoarthritis; BCS, body condition scores; hard surface: total time spent on hard surfaces over 24 h. Data are presented as means ± SD, with range (minimum-maximum) in parentheses.

Table 2.

Mean and median CTX-I concentrations by OA status in Asian and African elephants.a

Without COA or SOA (n - 18) With COA (n = 9) With SOA (n — 18)
Asian elephants
 CTX-I (ng/ml)
  Mean (SD) 0.75 (0.19) 0.72 (0.33) 0.77 (0.29)
  Median 0.78 0.56 0.70
  [Minimum, maximum] [0.42, 1.05] [0.39, 1.36] [0.39, 1.36]
 Age (yr)
  Mean (SD) 42.4 (4.6) 43.3 (3.9) 42.3 (4.6)
Without COA or SOA (n = 14) With COA (n = 11) With SOA (n = 14)
African elephant
 CTX-I (ng/ml)
  Mean (SD) 1.15 (0.64) 1.05 (0.42) 1.12 (0.49)
  Median 1.06 1.02 1.03
  [Minimum, maximum] [0.39, 2.55] [0.55, 1.70] [0.55, 1.91]
 Age (yr)
  Mean (SD) 36.0 (5.8) 37.1 (7.4) 36.7 (7.0)
Open in a new tab
a

CTX-I, collagen type I; OA, osteoarthritis; COA, confirmed osteoarthritis; SOA, suspected osteoarthritis.

Sample Analyses

Serum progestagens were measured by a solid-phase I25I progesterone radioimmunoassay (RIA; Siemens Medical Solutions Diagnostics, Los Angeles, CA 90045, USA).3 Serum LH was analyzed by an enzyme immunoassay (EIA) using a mouse monoclonal anti-bovine LH antibody (518-B7, Dr. Jan Roser, University of California, Davis, CA 95616, USA), a biotin-conjugated ovine LH label, horseradish peroxidase-conjugated streptavidin (catalog no. 1089153; Roche Diagnostics, Indianapolis, IN 46256, USA) and bovine (NIH-LH-B10, Dr. A. Parlow, National Hormone and Pituitary Program, Harbor UCLA Medical Center, Torrance, CA, 90502, USA) standards.14 FSH was measured by a heterologous 125I double-antibody RIA.3 The assay used an anti-ovine FSH antiserum (JADLER 178, Dr. James Dias, Albany Medical College, Albany, NY, 12208, USA) and ovine FSH label and standards (NIDDK-FSH-S16, Bethesda, MD, 20892, USA). Fecal glucocorticoid metabolites (fGCM) were analyzed by double-antibody enzyme EIA with a polyclonal rabbit anti-corticosterone antibody (CJM006).1 Assay sensitivities were 0.05 ng/ml for progestagens and FSH, 0.3 ng/ml for LH, and 0.14 ng/ml for corticosterone. All intra- and interassay coefficients were <10% and <15%, respectively.

Estrous cycle characteristics were based on serum progestagen profiles using methods previously described.3 Elephants with stable baseline progestagen concentrations (< 0.1 ng/ml) were classified as noncycling. Phase of the cycle was determined using HormLog, R Versions 3.5.28 and RStudio, Version 1.1.423.

Serum CTX-I was measured in duplicate (50 μL) by EIA (Serum CrossLaps® CTX-I, Immunodiagnostic Systems, United Kingdom). The immunoassay was validated through parallelism and accuracy checks. Serial dilutions of pooled elephant samples yielded displacement curves parallel to the standard curve (Asian elephants: y = 0.993x + 0.014, R2 = 0.986, F1,3 = 206.184, P <0.001; African elephants: y = 0.951x – 0.126, R2 = 0.971, F1,4 = 135.137, P <0.001). For accuracy, there was a slight, but consistent underestimation in concentrations, particularly for the African elephants (Asian elephants: y = 0.796x + 0.003, R2 = 0.998, F1,3 = 1423.978, P <0.001: African elephants: y = 0.614x + 0.018, R2 = 0.995, F1,3 = 610.390, P <0.001). Assay sensitivity was 0.02 ng/ml, and the intra- and interassay coefficients of variation were < 3.5% and < 6%, respectively.

Body Condition Score

Elephants were assigned a body condition score (BCS) from 1 to 5, by whole numbers (1 = very poor condition, 5 = very high condition).21

Confirmed OA and suspected OA assessment

Health records provided by participating institutions were reviewed by KLE (primary), with secondary reviews by DEC and JSW for confirmed OA (COA) or suspected OA (SOA) status categorization. Records were searched for the words “arthritis,” “osteoarthritis,” “degenerative joint disease,” “DJD,” “stiffness,” “lameness,” “trauma,” or “reduced range of motion”. COA was defined as veterinary diagnosis based on clinical history (e.g., lameness, swelling, stiffness, decreased range of motion, heat, lower activity levels, changes in typical behaviors)20 or physical examination and evidence on radiographs or necropsy. Elephants with SOA had clinical signs, history, and veterinary statements of suspected OA but no confirmatory radiographs or necropsy. Control elephants (i.e., no history of COA or SOA) had physical exams with no mention of “arthritis,” “osteoarthritis,” “degenerative joint disease,” “DJD,” “stiffness,” “lameness,” “trauma,” or “reduced range of motion” in the medical records. Elephants without full health records or without a veterinary diagnostic statement regarding confirmed or suspected cause of the clinical signs were excluded from the study.

Time spent on hard surfaces

Hard surface (HS) was defined as flooring made of concrete or stone aggregate, and time on HS was calculated using average monthly sums as previously described.20

Statistical Analyses

Statistical analyses were performed using R (hormLong, version 3.5.28 and RStudio, Version 1.1.423) for each species. Unless otherwise stated, separate analyses compared elephants with COA or SOA to controls.

Ovarian cyclicity relationships with OA status (COA or SOA) were evaluated by logistic regression, regressing OA status (0 or 1) on cyclicity status (i.e., cycling or noncycling), and then adjusted for age, BCS, age and BCS, and time spent on HS. The fit of the model and related assumptions were checked for outliers and linearity between log odds and continuous variables. Secondary analyses were conducted to determine relationships between OA status and age, time spent on HS, cyclicity status, and BCS. The prevalence of OA status was also assessed by species.

Descriptive statistics between the luteal and follicular phases were calculated for cycling elephants with and without COA and SOA, and mean data are expressed as ±SD. Linear regression models were used to regress FSH, LH, and progestagen concentrations during the phase of the ovarian cycle on OA status. The primary model was then adjusted for age, BCS, and age and BCS. Secondary analyses were conducted to assess the relationship between FSH, LH, and progestagen concentrations and age by species. For all linear models, normality and equal variance of residuals were assessed. Normality of residuals was determined if the absolute value of skewness <2, and unequal variance was tested with non-significant Levene test and σ1/σ2 >2.12, 13 All assumptions were met.

Differences in CTX-I between elephants with or without COA or SOA were tested with a multivariable regression model, with CTX-I as the dependent variable and OA status as the independent variable, adjusted for both cycling status and age, and conducted separately for each species. Residuals were assessed for normality, and the assumption was satisfied with skewness statistic less than 2.0.13 Descriptive statistics between COA and SOA groups were assessed. Secondary analyses were conducted to determine relationships between age and CTX-I by species, in addition to comparing CTX-I between species and by cycling status.

A linear mixed model was performed to determine whether there was a significant increase in CTX-I over time in elephants with SOA. This was done only in SOA elephants (n = 5 from four institutions) because these individuals with multiple blood samples did not have radiographs or necropsy to confirm OA. Based on date of onset of SOA, banked serum samples from approximate timeframe (within 2–17 mon; mean, 9 mon) and then approximately 2 to 4 yr prior, were analyzed. Significance level was determined at P < 0.05 (two tailed).

Results

Ovarian cyclicity and OA status

Asian elephants

Of 70 Asian elephants, 30 elephants were classified with SOA, with 17 of those cases confirmed by either radiographs or at necropsy. Cycling status was associated with COA and SOA status in the unadjusted logistic model (odds ratio [OR]=0.22, P = 0.016; OR=0.29, P=0.020, respectively); however, it was not after adjusting for age (OR=31, P = 0.112; OR=0.58, P=0.369, respectively), or age and BCS (OR=32, P = 0.129; OR=0.61, P=0.416, respectively). When adjusted for BCS, COA was associated with cycling status (OR=0.27, P = 0.045) and was marginally significant for elephants with SOA (OR=0.35, P = 0.056). When the logistic model was adjusted for HS time, cyclicity status remained significant for both COA and SOA (OR=0.18, P = 0.016; OR=0.23, P=0.016, respectively). Age was positively associated with COA (OR=1.23, P = 0.003) and SOA (OR=1.17, P < 0.001), with older elephants more likely not to cycle (OR=0.91, P = 0.022; OR=0.89, P=0.004, respectively). HS time was marginally significant with COA (OR=0.94, P = 0.059) and negatively associated with SOA (OR=0.94, P = 0.021). When age and HS time were included as predictors for OA status, only age predicted COA (OR=1.24, P = 0.006, OR=0.94, P = 0.063, respectively), while both predicted SOA (OR=1.17, P = 0.001; OR=0.94, P = 0.024, respectively). There was no relationship between COA or SOA status and fGCM concentrations, unadjusted (OR=1.00, P = 0.978; OR=1.00, 0.921, respectively) or adjusted (OR=0.99, P = 0.312; OR=1.00, P=0.645, respectively) for age (Table 1).

African elephants

Of 58 African elephants, 19 elephants were classified with SOA, with 11 of those cases confirmed by either radiographs or at necropsy. Cyclicity status was not associated with COA when unadjusted (OR=0.29, P = 0.078), adjusted for age (OR=0.35, P = 0.160), BCS (OR=0.24, P = 0.060), age and BCS (OR=0.28, P = 0.109), or HS time (OR=0.38, P = 0.210) in the logistic models. However, cyclicity status was associated with SOA when unadjusted (OR=0.12, P = 0.001), adjusted for age (OR=0.15, P = 0.005), BCS (OR=0.10, P = 0.001), age and BCS (OR=0.12, P = 0.004), and HS time (OR=0.14, P = 0.003). Age was positively associated with COA (OR=1.17, P = 0.025) and SOA (OR=1.19, P = 0.007). Surprisingly, HS time was positively associated with COA and SOA status (OR=1.07, P = 0.033; OR=1.05, P=0.036, respectively). When age and HS time were included as predictors for OA status, neither predicted COA (OR=1.11, P = 0.128; OR=1.06, P = 0.098, respectively), while age predicted SOA (OR=1.15, P = 0.033; OR=1.03 P = 0.177, respectively). There was no relationship between COA or SOA status and fGCMs unadjusted (OR=1.01, P = 0.629; OR=1.00, P=0.725, respectively) or adjusted (OR=1.00, P = 0.893; OR=1.00, P=0.977, respectively) for age (Table 1).

Sex and gonadotropin hormones and OA status in cycling elephants

Asian elephants

Of 45 Asian elephants, 15 elephants were classified with SOA, with 8 of those cases confirmed by either radiographs or at necropsy. Average hormone concentrations with and without COA are presented (Table 1, Fig. 1A). Using linear regression models, mean concentrations of FSH, LH, or progestagens did not differ between elephants with and without COA or SOA during the luteal (P = 0.740, 0.371, 0.870; 0.905, 0.129, and 0.469, respectively) or follicular (P = 0.738, 0.409, 0.254; 0.908, 0.622, and 0.096, respectively) phases, respectively. These relationships remained nonsignificant even after adjusting for age, regardless of cycle phase, except for progestagens during the follicular phase, which were higher in elephants with SOA (FSH concentrations during the luteal phase [FSH_L], P = 0.990, 0.769; FSH concentrations during the follicular phase [FSH_F], P = 0.745, 0.876; LH concentrations during the luteal phase [LH_L], P = 0.204, 0.079; LH concentrations during the follicular phase [LH_F], P = 0.152, 0.249; progestogen hormone concentrations during the luteal phase, P = 0.809, 0.332; and progestogen hormone concentrations during the follicular phase, P = 0.133, 0.026). Age was not associated with concentrations of any of the hormones (P > 0.157).

Figure 1.

Figure 1.

Open in a new tab

Mean concentration differences in FSH, LH, and progestagens concentrations by ovarian cycle phase in (A) Asian and (B) African elephants with and without COA. Error bars represent SE. FSH_F: Follicle stimulating hormone concentrations during the follicular phase; FSH_L: Follicle stimulating hormone concentrations during the luteal phase; LH_F: Luteinizing hormone concentrations during the follicular phase; LH_L: Luteinizing hormone concentrations during the luteal phase; PG_F: Progestagen hormone concentrations during the follicular phase; PG_L: Progestagen hormone concentrations during the luteal phase.

African elephants

Of 20 African elephants, 6 elephants were classified with SOA, with 5 of those cases confirmed by either radiographs or at necropsy. Average hormone concentrations with and without COA are presented (Table 1, Fig. 1B). Using linear regression models, mean concentrations of FSH, LH, or progestagens did not differ between elephants with and without COA or SOA during the luteal (P = 0.178, 0.518, 0.541; 0.231, 0.731, and 0.415, respectively) or follicular (P = 0.301, 0.554, 0.881; 0.346, 0.359, and 0.437, respectively) phases, respectively. These relationships remained non-significant even after adjusting for age regardless of phase (FSH_L: P = 0.167, 0.226; FSH_F: P = 0.209, 0.261; LH_L: P = 0.650, 0.875; LH_F: P = 0.769, 0.499; progestogen hormone concentrations during the luteal phase, P = 0.625, 0.476; and progestogen hormone concentrations during the follicular phase, P = 0.899, 0.432). Age was not associated with concentrations of any of the hormones (P > 0.284).

CTX-I in elephants with COA and SOA

Asian elephants

Of 36 Asian elephants, 18 elephants were classified with SOA, with 9 of those cases confirmed by radiographs or at necropsy (Table 2). When using a multivariable regression model, adjusted for cyclicity status and age, there was no difference in CTX-I concentrations between elephants with and without COA or SOA (P = 0.735; 0.848, respectively). Nor was there an association when unadjusted (P = 0.751and 0.858, respectively), or when adjusted for HS time (P = 0.728 and 0.991, respectively). CTX-I was not associated with age (P = 0.697 and 0.886, respectively). Cycling elephants with COA had higher CTX-I concentrations compared to noncycling elephants (P = 0.021); however, this relationship was not observed in SOA elephants (P = 0.400; COA, 0.83 ± 0.25 ng/ml compared with 0.63 ± 0.16 ng/ml; SOA, 0.79 ± 0.23 ng/ml compared to 0.72 ± 0.25 ng/ml, respectively; Table 2). In a subset of SOA elephants (n = 5 from 4 institutions), when using a linear mixed model, CTX-I concentrations did not increase over time approaching SOA detection (P = 0.273).

African elephants

Of 28 African elephants, 14 elephants were classified with SOA, with 11 of those cases confirmed by radiographs or at necropsy (Table 2). When using a multivariable regression model, adjusted for cyclicity status and age, there was no difference in CTX-I concentrations between elephants with and without COA or SOA (P = 0.622 and 0.961, respectively). Nor was CTX-I concentrations associated with COA or SOA status, when unadjusted (P = 0.907 and 0.917, respectively) or adjusted for HS time (P = 0.757 and 0.944, respectively). Cycling elephants with COA or with SOA had higher CTX-I concentrations compared to non-cycling elephants (COA, 1.43 ± 0.62 ng/ml compared with 0.85 ± 0.32 ng/ml; SOA, 1.47 ± 0.61 ng/ml compared with 0.88 ± 0.37 ng/ml, respectively; P < 0.001). CTX-I was not associated with age (P = 0.168 and 0.098, respectively). Independent of OA status, African elephants had higher CTX-I concentrations compared to Asians (P < 0.001; Table 2).

Discussion

This study examined the relationship between OA and ovarian cyclicity status, and CTX-I as a potential biomarker for elephant OA. In both species, noncycling elephants were more likely to exhibit OA compared to cycling elephants; however, the relationship was largely explained by age, as elephant OA increased with age, similar to that observed in people.22 Older elephants also are more likely to be acyclic compared to younger females;2,6 hence the unsurprising relationship between OA and cyclicity status. No associations were observed between serum progestagens, LH, or FSH with OA status.

Antemortem diagnosis of elephant OA typically employs radiography, which has challenges and limitations. Analysis of serum biomarkers for bone and cartilage degeneration would allow for earlier and logistically easier OA diagnosis. CTX-I is a well-conserved biomarker of bone resorption9 and has been associated with cartilage loss. It is higher in women with knee OA,17 although results have been conflicting.23 There was no association between CTX-I with OA status in either elephant species, and CTX-I did not increase longitudinally in the five Asian elephants with samples prior to SOA diagnosis. Concentrations of CTX-I were higher in African than Asian elephants, likely due to their general larger stature and weight. It is possible that CTX-I concentrations may also indicate general skeletal turnover10 and could be influenced by sunlight exposure or vitamin D3 levels.25 However, elephants in this study were housed throughout North America and exposed to varying degrees of sunlight, with no apparent location effect.

Hard surfaces have been associated with overall poor elephant foot and joint health, prompting many institutions to use flooring of softer substances (e.g., sand, rubber).16 Because OA develops over time, using recent cross-sectional data may not accurately represent past environmental contributions to present pathologies. This may explain the negative association between HS time and OA status in Asian elephants and the positive relationship in Africans.

Although the study took advantage of a large elephant welfare data set, limitations in numbers of elephants with COA, and particularly cycling elephants with COA, reduced statistical power. This may also explain the lack of relationship between CTX-I and OA; alternatively, CTX-I may be unrelated to elephant OA or be further influenced by circadian variations and food intake.7,24 Analysis of cartilage turnover biomarkers (e.g., urinary CTX-II) and bone formation biomarkers are recommended in future studies.

In conclusion, although CTX-I was not useful in identifying elephants with OA in this study, identifying an appropriate bone or cartilage blood biomarker would allow caretakers and veterinary staff to diagnose and treat OA earlier in the disease process, improving elephant health and welfare within managed populations. It would also allow for diagnosis and treatment of elephants in remote or free-ranging populations where access to diagnostics are limited.

Acknowledgements:

The authors thank all the participating zoos and their keepers for the commitment they make to their elephants. Specifically, thank you to the following zoos: Africam Safari, Albuquerque Biological Park, Audubon Institute, BREC’s Baton Rouge Zoo, Busch Gardens, Buttonwood Park Zoo, Calgary Zoo, Cameron Park Zoo, Cheyenne Mountain Zoological Park, Cincinnati Zoo and Botanical Garden, Cleveland Metroparks Zoo, Columbus Zoo, Dallas Zoo, Denver Zoo, Dickerson Park Zoo, Disney’s Animal Kingdom, El Paso Zoo, Fresno Chaffee Zoo, Greenville Zoo, Honolulu Zoo, Houston Zoological Gardens, Indianapolis Zoological Society, Inc., Kansas City Zoo, Knoxville Zoological Gardens, Little Rock Zoological Garden, Los Angeles Zoo and Botanical Gardens, Louisville Zoological Garden, Maryland Zoo, Smithsonian National Zoological Park, Niabi Zoo, Oklahoma City Zoological Park, Oregon Zoo, Phoenix Zoo, Point Defiance Zoo and Aquarium, Riverbanks Zoological Park, Rosamond Gifford Zoo at Burnet Park, San Antonio Zoological Gardens and Aquarium, San Diego Zoo, Santa Barbara Zoological Gardens, Saint Louis Zoo, Topeka Zoological Park, Tulsa Zoological Park, Utah’s Hogle Zoo, Virginia Zoological Park, Wildlife Conservation Society—Bronx Zoo, Woodland Park Zoo, Zoo Miami. Thank you to Dr. Kari Morfeld for body condition scoring all the elephants that were used in the present study.

This work was supported in part by the Nathan Shock Center on Aging P30AG050886 (DEC, SNA), R01AG049058 (TMG), and the Molecular Measures Core of the NIH/NIA P30 AG028716 (VBK, JLH). The opinions expressed herein are those of the authors and not necessarily those of any other organization with which the authors are affiliated.

Literature Cited

  • 1.Brown JL, Bray JD, Carlstead K, Dickey D, Farin C, Ange-van Heugten K. Relationships of environmental, management, and health factors to fecal glucocorticoid metabolite concentrations in zoo African and Asian elephants. PloS one. 2019;14(9): e0217326. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Brown JL, Paris S, Pardo-Oviedo NA, Meehan CL, Hogan JN, Morfeld KA, Carlstead K. Reproductive health assessment of female elephants in North American zoos and association of husbandry practices with reproductive dysfunction in African elephants (Loxodonta africana). PloS one. 2016;11(7):e0145673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Brown JL, Walker SL, Moeller T. Comparative endocrinology of cycling and non-cycling Asian (Elephas maximus) and African (Loxodonta africana) elephants. Gen Comp Endocrinol. 2004;136(3):360–370. [DOI] [PubMed] [Google Scholar]
  • 4.Carlstead K, Mench JA, Meehan C, Brown JL. An epidemiological approach to welfare research in zoos: The elephant welfare project. J Appl Anim Welf Sci. 2013;16(4): 319–337. [DOI] [PubMed] [Google Scholar]
  • 5.Cecil RL, Archer BH. Arthritis of the menopause: a study of fifty cases. JAMA. 1925;84(2):75–79. [Google Scholar]
  • 6.Chusyd DE, Brown JL, Hambly C, Johnson MS, Morfeld K, Patki A, Speakman JR, Allison DB, Nagy TR. Adiposity and Reproductive Cycling Status in Zoo African Elephants. Obesity. 2018;26(1):103–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Clowes JA, Hannon RA, Yap TS, Hoyle NR, Blumsohn A, Eastell R. Effect of feeding on bone turnover markers and its impact on biological variability of measurements. Bone. 2002;30(6):886–890. [DOI] [PubMed] [Google Scholar]
  • 8.Fanson B, Fanson K. hormLong: Longitudinal analysis of hormone data. R package version 2. 2014. [Google Scholar]
  • 9.Garnero P, Delmas PD. Biomarkers in OA. Curr Opin Rheumatol. 2003;15(5):641–646. [DOI] [PubMed] [Google Scholar]
  • 10.Garnero P, Delmas PD. Investigation of bone: bone turnover. In: Hochberg MC, Silman AJ, Smolen JS, Weinblatt ME, Weisman MH (eds.) Rheumatology, 3rd ed, London (UK): Harcourt Health Sciences, Ltd; 2003. p. 2043–2057. [Google Scholar]
  • 11.Gelse K, Pöschl E, Aigner T. Collagens—structure, function, and biosynthesis. Adv Drug Deliv Rev. 2003;55(12):1531–1546. [DOI] [PubMed] [Google Scholar]
  • 12.Glass GV, Hopkins KD. Statistical methods in education and psychology. 3rd ed, Boston (MA): Allyn and Bacon; 1996. [Google Scholar]
  • 13.Glass GV, Peckham PD, Sanders JR. Consequences of failure to meet assumptions underlying the fixed effects analyses of variance and covariance. Review of educational research. 1972; doi. 10.3102/00346543042003237. [DOI] [Google Scholar]
  • 14.Graham LH, Bolling J, Miller G, Pratt-Hawkes N, Joseph S. Enzyme-immunoassay for the measurement of luteinizing hormone in the serum of African elephants (Loxodonta africana). Zoo Biology. 2002;21(4):403–408. [Google Scholar]
  • 15.Harari J [Internet]. Osteoarthritis (Degenerative Joint Disease). 01March 2018. Available from https://www.merckvetmanual.com/dog-owners/bone,-joint,-and-muscle-disorders-of-dogs/osteoarthritis-degenerative-joint-disease?query=osteoarthritis
  • 16.Harris M, Sherwin C, Harris S. The welfare, housing and husbandry of elephants in UK zoos: Final report, University of Bristol. DEFRA Science and Research Project; WC05007. 2008. [Google Scholar]
  • 17.Kraus VB, Collins JE, Hargrove D, Losina E, Nevitt M, Katz JN, Wang SX, Sandell LJ, Hoffmann SC, Hunter DJ. Predictive validity of biochemical biomarkers in knee OA: Data from the FNIH OA Biomarkers Consortium. Ann Rheum Dis. 2017;76(1):186–195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Lewis KD, Shepherdson DJ, Owens TM, Keele M. A survey of elephant husbandry and foot health in North American zoos. Zoo Biology. 2010;29(2):221–236. [DOI] [PubMed] [Google Scholar]
  • 19.McIlwraith CW [Internet]. Arthritis in Large Animals. 01June 2022. Available from https://www.merckvetmanual.com/musculoskeletal-system/arthropathies-in-largeanimals/arthritis-in-large-animals?query=arthritis.
  • 20.Miller MA, Hogan JN, Meehan CL. Housing and demographic risk factors impacting foot and musculoskeletal health in African elephants (Loxodonta africana) and Asian elephants (Elephas maximus) in North American zoos. PLoS One. 2016;11(7):e0155223. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Morfeld KA, Meehan CL, Hogan JN, Brown JL. Assessment of body condition in African (Loxodonta africana) and Asian (Elephas maximus) elephants in North American zoos and management practices associated with high body condition scores. PloS one. 2016;11(7):e0155146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Neogi T, Zhang Y. Epidemiology of OA. Rheum Dis Clin. 2013;39(1):1–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Ok S-M, Lee S-M, Park HR, Heong S-H, Ko C-C, Kim YI. Concentrations of CTX I, CTX II, DPD, and PYD in the urine as a biomarker for the diagnosis of temporomandibular joint OA: A preliminary study. Cranio. 2018;36(6):366–372. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Qvist P, Christgau S, Pedersen BJ, Schlemmer A, Christiansen C. Circadian variation in the serum concentration of C-terminal telopeptide of type I collagen (serum CTx): effects of gender, age, menopausal status, posture, daylight, serum cortisol, and fasting. Bone. 2002;31(1):57–61. [DOI] [PubMed] [Google Scholar]
  • 25.Pastoret V, Carstanjen B, Lejeune JP, Fanir F, Remy B, Reginster JY, Serteyn D, Gabriel A. Evaluation of serum osteocalcin and CTX-I in Ardenner horses with special reference to juvenile interphalangeal joint disease. J Vet Med A. 2007;54(9):458–463. [DOI] [PubMed] [Google Scholar]
  • 26.Sniekers YH, Weinans H, van Osch GJ, van Leeuwen JP. Oestrogen is important for maintenance of cartilage and subchondral bone in a murine model of knee OA. Arthritis Research and Therapy. 2010; doi. 10.1186/ar3148. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES