Contaminants and Biomarker Assays in Beluga Whales from the Eastern Chukchi Sea

CONTAMINANTS AND BIOMARKER ASSAYS IN BELUGA WHALES FROM THE EASTERN CHUKCHI SEA

NORTH SLOPE BOROUGH DEPARTMENT OF WILDLIFE MANAGEMENT

PROGRESS REPORT, CONTRACT 2013-017

Submitted to: Dr. Robert Suydam North Slope Borough Department of Wildlife Management Barrow, Alaska 99723 (907) 852-0350

Submitted by: Mote Marine Laboratory Directorate of Marine Biology & Conservation 1600 Ken Thompson Parkway Sarasota, FL 34236 (941) 388-4441 (941) 388-4223 FAX

Dr. Dana L. Wetzel and Dr. John E. Reynolds

Mote Marine Laboratory, 1600 Ken Thompson Pkwy, Sarasota, FL 34236

2 March 2015

Mote Marine Laboratory Technical Report Number 1880

PROGRESS REPORT SUBMITTED TO THE NORTH SLOPE BOROUGH,

DEPARTMENT OF WILDLIFE MANAGEMENT

PROJECT TITLE: Contaminants and Biomarker Assays in Beluga Whales from the Eastern Chukchi Sea

CONTRACT NUMBER: 2013-017

DATE: 2 March 2015

SUBMITTED TO: Dr. Robert Suydam

SUBMITTED BY: Dr. Dana Wetzel and Dr. John Reynolds Mote Marine Laboratory, Sarasota, Florida

PROJECT END DATE: The anticipated end date is December 2015. At that time all funds will be spent and all analyses will be completed and interpreted. A final report will be submitted by 1 January 2016.

Introduction:

The project has three primary components, namely 1) age determination, 2) analysis of polycyclic aromatic hydrocarbons and other classes of organic contaminants, and (3) assessment of selected biomarkers of effects of stressors. The results achieved to date will be described separately in this report. Samples (eye lens, serum, blubber, and liver) were acquired by a team of scientists led by Dr. Robert Suydam (North Slope Borough, Department of Wildlife Management; NSB) during the 2012 subsistence harvest of beluga whales (Delphinapterus leucas) by the community of Point Lay, Alaska. The list of samples collected for age determination and biomarker analyses, together with morphometric and other data for each whale appears in Table 1. In addition, samples of liver and blubber were collected and sent to Mote Marine Laboratory for the contaminant analyses.

The overarching goals of the project are to (1) assess the results of a range of analyses (to assess beluga whale age, contaminant levels, and sublethal biomarkers of effects of stressors) in association with other data (e.g., morphology, body condition) to better understand the status of the Chukchi Sea beluga whale stock, and (2) create “baseline” values against which to assess changes in beluga whale status in the future.

Age determination:

Background and Goals:

The goals of this component of the project were as follows:

---to use amino acid racemization (aar) techniques to assess the age of individual beluga whales harvested for subsistence in Point Lay, Alaska; and

---to compare ages determined using aar with ages determined using counts of growth layer groups in the teeth of the same individual whales.

Suydam sent frozen eyes from 15 whales (Table 1) harvested prior to the involvement of Wetzel/Reynolds in the Point Lay hunt, or from whales harvested elsewhere, to Mote Marine Laboratory for analysis. This sample included one fetus (LDL-1F-09).

Under the current contract, the eyes from an additional 8 whales were taken for subsistence were collected and frozen for later shipment to and analysis at Mote Marine Laboratory (Table 1).

Eyes of individual belugas were processed and analyses conducted as described by Wetzel et al. (2014; Appendix 1) for bowhead whales (Balaena mysticetus). The analyses include artificial aging studies of samples from the lens of one of the beluga whales sampled. Wetzel et al. (2014) is provided (Appendix 1) in its entirety.

The chemical analyses of the hydrolyzed and derivative eye lens nucleus of each whale yield values for the D form of aspartic acid and the L form of aspartic acid (D/L). Kasp represents the rate constant for the conversion of the L form to the D form in the living animal; that rate is temperature dependent, so the calculated age for each whale is also temperature related.

The analyses provide the necessary data to apply the following equation to calculate age of each whale:

log (1 (DL /LD ) ) / (1 ( / ) )log(1(D/L))/ (1 (/D L) )  ii  00 Agei  2Kasp

Kasp is calculated using the results of the heating experiments (Wetzel et al. 2014) using the Arrhenius equation:

KEasp Aexp a / (RT )

Here, R is the universal gas constant, and T is the temperature of the living whale in degrees Kelvin. The parameters A and Ea are estimable from the heating experiment data.

Results:

As noted, lens temperature in the living whale has a significant influence on the calculated age of the whale. To provide a better understanding of lens temperature in living belugas, Hans Thewissen and Andrew VonDuyke (members of the Point Lay scientific group) used a probe and thermistor to determine the lens temperature within two hours after death of six whales harvested in summer, 2013. Those temperatures (Thewissen, pers. comm. 2013; Table 2) averaged 18.57°C. In 2012, Thewissen (pers. comm. 2012) acquired lens temperatures from three harvested whales that had been dead between 50 and 80 minutes; for those whales the lens temperatures were 23.7°C, 18.8°C, and 20.1°C (mean = 20.866°C, a figure that corresponds well with the average temperature [=20.85°C] of the two whales sampled most quickly after death in 2013; Table 2). Note that these measurements provide usual insight and confirm that beluga lenses are colder than is core body temperature, but given circumstance of the measurements (i.e., the relatively high temperature of the ocular fat, and the time interval warming between death and the temperature measurements), it seems quite possible that lens temperature in the living whales is colder.

Table 3 provides replicate data for the measured D form and L form of aspartic acid for each beluga whales sampled in 2012 for this study (i.e., individuals with a field number of LDL##12), as well as additional individuals for which eyes were archived. The measurements are extremely precise, as demonstrated by the very small standard deviations. In Table 3, several values of Kasp are provided at plausible lens temperatures to provide the calculated age of each whale. Note that one column of possible ages used the derived value of Kasp based on the empirical measurements taken in 2012 of lens temperature described above (i.e., mean = 20.866°C).

Discussion and Implications:

Although beluga eyes appear not to be as cold as those of bowhead whales (estimated to be around 11.3°C or lower based on some direct measurements following harvest; see Wetzel et al. 2014; Sformo et al. 2011), it appears that the beluga eyes are likely to be considerably cooler (perhaps around 15°C or less) than the “normal” mammalian body temperature and beluga whale core temperature.

The whales from the 2012 harvest for which age was derived (Table 3) were generally young animals, with only two exceeding 10 years of age based on calculations using a lens temperature of 15°C. Sexual maturity in beluga whales occurs at age 9-12 years in females and sometime later in males (O’Corry-Crowe 2009). The animals turn white in color a couple years earlier than that in both sexes, and the ages of the larger, white whales sampled in 2012 approach or equal the age at which white pigmentation predominates.

In K-selected species, mature adult females constitute the most valuable demographic group to sustain population size, so if the whales we examined are representative of those taken for subsistence along the northwestern coast of Alaska, it appears that the hunt may largely be targeting immature animals that travel together. This suggests that the harvest may avoid

targeting the most valuable demographic group of belugas, an important management consideration.

There are several factors that enter into accurate calculations of age using D:L ratios. One is the D:L ratio in the lens nucleus of a whale at age 0. In this study we had access to an eye from only one fetal beluga; additional fetal eyes would be valuable in calculating D:L for whales at age 0.

A compelling reason for doing this study is to compare ages derived from amino acid racemization methods (described here) with ages derived from counting growth layer groups (GLGs) in the teeth of belugas. The latter approach requires validation for beluga whales for which there are uncertainties regarding deposition rates of GLGs and loss of GLGs with age (e.g., Lockyer et al. 2007). The NSB-DWM hoped that this study would help to resolve the questions about the utility of GLGs. The comparative GLG data for the whales we examined will be provided by Suydam, but are not yet available.

Next Steps:

The obvious next steps for the age determination component of the project include (a) acquiring age estimates based on GLG analysis, (b) comparing the GLG-based ages with the aar-based ages, and (c) developing a manuscript for a peer-reviewed journal. However, two specific steps that would be useful for calculating ages based on the aar approach include: (a) acquiring frozen eyes of at least two additional fetuses so better document the D:L ratio at age 0, and (b) acquiring, if possible a frozen eye from Naku, a known age adult female whale that recently died at Mystic, along with temperature of the water in her tank.

Contaminant Analyses:

Background and Goals:

The effects of contaminants are a major concern in the Arctic from the standpoints of both human health and wildlife health. Given the current and past extent of oil and gas exploration and development, it is extremely important to assess levels of PAHs (polycyclic aromatic hydrocarbons), which include the most toxic components of oil. In addition, it is very important to assess levels of other classes of organic contaminants including the PBDEs (polybrominated diphenyl ethers), a group of flame retardants which represent a relatively new contaminant threat that is persistent and quite toxic.

The goals of this component of the project were to:

---assess levels of PAHs, PBDEs, PCBs (polychlorinated biphenyls), and OCs (organochlorine pesticides) in blubber and liver samples taken from whales harvested for subsistence in Point Lay; and

---use the data both to create “baselines” against which to assess possible changes in contaminant levels in the future and to assess whether current levels of these contaminants may pose a threat to the people who depend on beluga whales for subsistence.

Results:

PAH analyses: All samples have been analyzed, but the data need to be interpreted. That process is underway.

PBDE, PCB, and OC analyses: Sample analyses and interpretations are scheduled to start shortly.

Next steps:

The next steps for this component of the project are to complete all chemical analyses, conduct interpretations of the data, and produce a final report for NSB.

Biomarkers:

Background and Goals:

Biomarkers can provide extremely useful insights into the sublethal effects of stressors on an exposed organism, in this case beluga whales. The biomarkers we have selected to use generally address significant biological functions, such as reproduction or immune system function.

The goal of this component of the project is to improve assessments of beluga whale health by using biomarkers of immune function, reproduction/fertility, and acute or chronic stress.

Results:

The acute phase protein markers (C-reactive protein [CRP], serum amyloid A [SAA], and haptoglobin) were assayed by the University of Miami’s Acute Phase Protein Laboratory (APPL); the data are summarized in Table 4 and full reports are appended for each whale (Appendix 2).

The results of fertility hormone analyses appear in Table 5. The results are for inhibin B and anti-Müllerian hormone (AMH). Results for inhibin A are not yet available. The fertility hormone results are provided for three captive belugas (Juno, Kela, and Naku) from Mystic Aquarium, as well as for several of the whales harvested in 2012 at Point Lay. For some of the Point Lay whales, there was insufficient serum to conduct these analyses.

Results for a number of cytokines are presented in Table 6. For several cytokines (e.g., GM-CSF, IL10), most whales had undetectable levels in their serum, but some cytokines (e.g., IL-1RA, IL-18) were present at elevated levels in the majority of samples.

Discussion:

Acute phase proteins:

In normal animals, CRP levels are negligible (APPL; Appendix 2). SAA levels are also negligible in normal animals, and SAA is especially useful as a marker of acute inflammation. Haptoglobin levels do not change as dramatically as do levels of CRP and SAA; haptoglobin is a better marker of chronic inflammation than are CRP and SAA.

The APPL indicates that normal reference intervals of SAA for most species is <10-20 mg/L; by that standard, three of the belugas sampled in 2012 were suffering from some level of acute inflammation. This preliminary suggestion should be treated with caution, however, since the normal reference intervals can differ among species (e.g., for Florida manatees, Trichechus manatus latirostris, a range of <10-50 mg/L is acceptable [Hart et al. 2006] and the reference interval for belugas may not be the same as that provided by APPL. We are investigating whether scientists or veterinarians have established normal reference values for SAA in wild or captive belugas exist.

The levels of CRP in sampled belugas were always below 2.0 mg/L, and the range of levels for haptoglobin was 0.04-0.45 mg/ml. Interpretations of those values will be done, as possible by consulting with beluga whale biologists and veterinarians.

Fertility assays:

The results for the captive whales (Juno, Kela, and Naku) were generally different from those of the whales taken during the 2012 harvest. At the time of sampling, Juno (a male) was 9 years old, and Kela and Naku (females) were both 29-30 years old. The oldest whale sampled in the 2012 hunt in Point Lay was a female (LDL0612; Mote ID number PLAY-12-

0285) estimated to be around 15.6 years old; all other Point Lay animals were much younger than the captive animals. In addition, it is interesting to note that the mature female captive whales did not produce calves in their lifetimes. Thus, with an admittedly small sample size, the age difference between the captive and wild whales and the lack of reproductive success will affect interpretations.

The inhibin B levels in the Point Lay whales were dramatically higher than those of the captive animals at Mystic. For the Point Lay animals (including the oldest female, LDL0612), all animals had inhibin B values above 100 pg/mL, whereas the values for the Mystic animals were all less than 9 pg/mL.

The AMH levels for two large males, one captive (Juno) and one wild (LDL0112) were comparably high (360 ng/mL and 753 ng/mL), although Juno was older at the time of sampling. For all the other whales for which AMH values were acquired, captive whale values were 1-2 orders of magnitude higher than were the wild whale values.

Cytokines:

When cytokine gene expression is normal, circulating levels of cytokines are generally at very low or even undetectable levels. Thus, consistently high serum levels indicate some sort of immune function stressor. Two cytokines that were consistently elevated among the sampled whales were interleukin 1 receptor antagonist and interleukin 18 (IL-1RA and IL-18, respectively). IL-18 is a pro-inflammatory cytokine production of which leads to release by T cells and natural killer cells of interferon gamma, which activates macrophages and certain other immune cells. IL-1RA is an inhibitor of the pro-inflammatory response associated with acute phase cytokines.

Other cytokines (e.g., IL-12, IL4) were also elevated in a number of individuals. We will conduct one additional analysis of all cytokines to comprehensively validate the data in Table 6; at that time, we will provide a full interpretation of what the cytokine data mean with regard to immune function of the harvested belugas.

Next steps:

Laboratory analyses for inhibin A of sampled beluga whales will be conducted, and the results interpreted.

As noted (see discussion, above) the cytokine assays will be re-run to comprehensively validate the results to date. Results of all cytokine analyses and fertility analyses will be comprehensively interpreted.

The results of the analyses will be interpreted to identify animals for which health or reproduction may be impaired.

Biomarker results will be interpreted with regard to contaminant levels and age of the beluga whales.

References:

Hart, K., J. Harvey, R. Bonde, D. Murphy, M. Lowe, M. Menchaca, E. Haubold, and R. Francis-Floyd. 2006. Comparison of methods used to diagnose generalized inflammatory disease in manatees (Trichechus manatus latirostris). Journal of Zoo and Wildlife Medicine 37(2):151-159.

Lockyer, C., A.A. Hohn, D.W. Doidge, M. P. Heide-Jorgensen, and R. Suydam. 2007. Age determination in belugas (Delphinapterus leucas): A quest for validation of dentinal layering. Aquatic Mammals 33(3):293-304.

O’Corry-Crowe, G.M. 2009. Beluga whale Delphinapterus leucas. Pp 108-112. IN: W.F. Perrin, B. Würsig, and J.G.M. Thewissen (eds.), Encyclopedia of Marine Mammals, Second Edition. Elsevier, San Diego, CA.

Sformo, T., C. Hanns, and J.C. George. 2011. Preliminary findings on lens temperatures in the bowhead whale (Balaena mysticetus). Unpublished report, North Slope Borough Department of Wildlife Management, Barrow, Alaska.

Wetzel, D.L., J.E. Reynolds, P. Mercurio, G.H. Givens, E.L. Pulster and J.C. George. 2014. Age estimation for bowhead whales, Balaena mysticetus, using aspartic acid racemization with enhanced hydrolysis and derivatization procedures. Paper SC/65b/BRG05 presented to the Scientific Committee of the International Whaling Commission, May 2014.

Tables:

Table 1: Identification information, morphometrics (reported in cm), and other attributes of beluga whales sampled during the July 2012 subsistence harvest in Point Lay. Under “Color”, W stands for white and GW stands for gray-white. Matrices included serum (for biomarker analyses) and an eye lens (for aar-based age determinations). In addition, liver and blubber samples were taken for contaminant analyses.

Data Collected Whale Tail MML ID sample Date Collected Sex Color Length Matrix Width number

PLAY-12-0204 LDL0112 9‐Jul‐12 M W 383 88 Serum

PLAY-12-0205 LDL0212 9‐Jul‐12 F W 355 76 Serum

PLAY-12-0206 LDL0312 9‐Jul‐12 F W 362 78 Serum

PLAY-12-0207 LDL0412 9‐Jul‐12 F GW 279 64 Serum

PLAY-12-0208 LDL0512 9‐Jul‐12 M W 335 79 Serum

PLAY-12-0209 LDL0612 9‐Jul‐12 F W 355 80 Serum

PLAY-12-0210 LDL0712 9‐Jul‐12 F GW 280 59 Serum

PLAY-12-0211 LDL0812 9‐Jul‐12 M W 385 93 Serum

PLAY-12-0212 LDL0912 9‐Jul‐12 F GW 317 75 Serum

PLAY-12-0213 LDL1012 9‐Jul‐12 F GW 298 77 Serum Eye PLAY-12-0280 LDL0112 9‐Jul‐12 M W 383 88 lens Eye PLAY-12-0281 LDL0212 9‐Jul‐12 F W 355 76 lens Eye PLAY-12-0282 LDL0312 9‐Jul‐12 F W 362 78 lens Eye PLAY-12-0283 LDL0412 9‐Jul‐12 F GW 279 64 lens Eye PLAY-12-0284 LDL0512 9‐Jul‐12 M W 335 79 lens Eye PLAY-12-0285 LDL0612 9‐Jul‐12 F W 355 80 lens Eye PLAY-12-0286 LDL0712 9‐Jul‐12 F GW 280 59 lens Eye PLAY-12-0287 LDL0812 9‐Jul‐12 M W 385 93 lens Eye PLAY-12-0288 LDL0912 9‐Jul‐12 F GW 317 75 lens Eye PLAY-12-0289 LDL1012 9‐Jul‐12 F GW 298 77 lens 10

Table 2: Eye temperatures from beluga whales landed in Point Lay, Alaska, 4 July 2013. Temperatures (reported in °C) were taken by H. Thewissen and A. VonDuyke. Harvests occurred between 0345 and 0415.

Beluga Corneal Lens Ocular Measure number surface surface fat time (am)

LDL‐1‐13 15.1 21.4 25.1 4:48 LDL‐2‐13 15.6 20.3 23.7 4:53 LDL‐3‐13 15.1 17.7 26.5 4:57 LDL‐4‐13 14.2 16.1 22.4 5:03 LDL‐7‐13 14.1 16.9 23.5 5:57 LDL‐9‐13 14.1 19 28 5:54

Water Temp (at 04:48) 12.9 Water Temp (at 05:04) 12.8 Water Temp (at 05:48) 12.7 Air Temp (at 05:05) 10.7

Table 3: Age estimates for beluga whales, based on different eye lens temperatures. The lens temperature of 20.866°C corresponds to the average lens surface temperature measured by Thewissen (2012). The possible lens temperature of 15°C corresponds with corneal temperature at death measured in 2013 by Thewissen and VonDuyke (pers. comm.; Table 2). Morphometric and other data associated with the 2012 Point Lay samples appear in Table 1. In this Table, D and L refer to the two forms of aspartic acid that were measured in triplicate for each sample. Precision of the measurements was very good, based in values for standard deviation.

Table 4: Results of analyses by University of Miami of levels of the acute phase proteins C-reactive protein, serum amyloid A, and haptoglobin.

C‐reactive protein Serum amyloid A Haptoglobin Whale ID (mg/L) (mg/L) (mg/ml) PLAY0205 < 2.0 13.9 0.4 PLAY0206 < 2.0 22.31 0.31 PLAY0207 < 2.0 9.88 0.15 PLAY0208 < 2.0 5.17 0.04 PLAY0209 < 2.0 38.31 0.45 PLAY0210 < 2.0 20.24 0.07 PLAY0211 < 2.0 < 0.1 0.26 PLAY0212 < 2.0 7.49 0.17

Table 5: Fertility hormone (inhibin B and anti-Müllerian hormone [AMH]) levels in captive and wild belugas. The captive animals (Juno, Kela, and Naku) are housed at Mystic Aquarium.

Inhibin B Inhibin B Inhibin B AMH AMH AMH Inhibin A Inhibin A Inhibin A Collection MML# Field ID Gender Name pg/mL pg/mL pg/mL ng/ml ng/ml ng/ml pg/mL pg/mL pg/mL %CV date Rep 1 Rep 2 average Rep 1 Rep 2 average Rep 1 Rep 2 average

FPP‐11‐0153 Y0L07 Male Juno 30‐Mar‐11 2.413 2.926 2.6695 398.32 321.84 360.08

FPP‐11‐0143 6177 17‐Jun‐10 4.927 6.835 5.881 2.111 1.477 1.794

FPP‐11‐0144 6413 28‐Jul‐10 7.47 5.061 6.2655 0.631 0.632 0.6315 Female Kela

FPP‐11‐0145 6564 26‐Jan‐11 9.577 6.812 8.1945 0.757 0.731 0.744

FPP‐11‐0146 X0L05 23‐Mar‐11 7.51 8.855 8.1825 0.659 0.645 0.652

FPP‐11‐0147 6224 3‐Mar‐10 2.514 2.489 2.5015 1.026 1.157 1.0915

FPP‐11‐0148 6319 18‐May‐10 3.234 2.393 2.8135 0.613 0.672 0.6425

FPP‐11‐0149 6407 21‐Jul‐10 2.292 2.319 2.3055 0.376 0.369 0.3725 Female Naku

FPP‐11‐0150 6443 2‐Sep‐10 2.56 2.771 2.6655 0.563 0.425 0.494

FPP‐11‐0151 6513 12‐Nov‐10 2.777 2.788 2.7825 0.319 0.364 0.3415

FPP‐11‐0152 X0L04 23‐Mar‐11 3.651 2.495 3.073 0.662 0.509 0.5855

PLAY‐12‐0204 LDL0112 Male 9‐Jul‐12 112.54 99.16 105.85 742.32 763.04 752.68 17.48 18.377 17.9285 3.178

PLAY‐12‐0205 LDL0212 Female 9‐Jul‐12 169.14 153.58 161.36 0.057 0.06 0.0585

PLAY‐12‐0206 LDL0312 Female 9‐Jul‐12 135.64 106.18 120.91 0.055 0.061 0.058

PLAY‐12‐0207 LDL0412 Female 9‐Jul‐12 125.57 100.99 113.28 0.046 0.041 0.0435

PLAY‐12‐0208 LDL0512 Male 9‐Jul‐12 131.24 119.46 125.35 OVER OVER OVER

PLAY‐12‐0209 LDL0612 Female 9‐Jul‐12 222.72 216.84 219.78 0.066 0.063 0.0645 8.728 8.171 8.4495 4.19

PLAY‐12‐0210 LDL0712 Female 9‐Jul‐12

PLAY‐12‐0211 LDL0812 Male 9‐Jul‐12

PLAY‐12‐0212 LDL0912 Female 9‐Jul‐12

PLAY‐12‐0213 LDL1012 Female 9‐Jul‐12

Table 6: Levels of selected cytokines measured in serum of harvested Point Lay beluga whales. KEY: GM-CSF: granulocyte macrophage colony-stimulating factor; IL: interleukin; TNFa: tumor necrosis factor alpha. Blanks represent undetectable levels.

CYTOKINES GM‐CSF IL‐10 IL‐12 IL‐18 Samples Rep 3AverageRep 1Rep 2Rep 3 Stdev Average Std Error %CV Rep 1Rep 2Rep 3 Stdev Average Std Error %CV Rep 1Rep 2Rep 3 Stdev Average Std Error %CV PLAY‐12‐0204 15.43 11.08 6.61 3.076 13.255 2.175 23.21 138.24 151.73 116.14 17.968 135.37 10.374 13.273 PLAY‐12‐0205 38.14 25.32 35.88 6.843 33.113 3.9509 20.67 313.63 289.86 339.44 24.797 314.31 14.317 7.8893 PLAY‐12‐0206 7.62 6.61 7.62 0.583 7.2833 0.3367 8.006 167.64 170.26 140.96 16.213 159.62 9.3606 10.157 PLAY‐12‐0207 90.29 90.29 35.88 28.1 30.4 3.997 31.46 2.3076 12.7 211.38 180.69 221.43 21.224 204.5 12.253 10.378 PLAY‐12‐0208 81.38 62.95 99.04 18.046 81.12333 10.419 22.246 PLAY‐12‐0209 57.85 37.82 20.32 12.374 29.07 8.75 42.567 23.93 16.87 21.59 3.596 20.797 2.0763 17.29 442.1 424.14 419.63 11.887 428.6233 6.863 2.7733 PLAY‐12‐0210 29.94 26.02 23.46 3.264 26.473 1.8843 12.33 69.19 69.19 PLAY‐12‐0211 149.05 165.01 101.93 32.798 138.6633 18.936 23.653 PLAY‐12‐0212 26.71 13.03 14.47 1.018 13.75 0.72 7.405 172.88 165.01 124.51 25.955 154.1333 14.985 16.839 PLAY‐12‐0213 270.62 260.91 289.86 14.734 273.7967 8.5067 5.3814 CYTOKINES IL‐1alpha IL‐1b IL‐1ra IL‐2 Samples Rep 1Rep 2Rep 3 Stdev Average Std Error %CV Rep 1AverageRep 1Rep 2Rep 3 Stdev Average Std Error %CV Rep 1Rep 2Rep 3 Stdev Average Std Error %CV PLAY‐12‐0204 2.09 2.09 PLAY‐12‐0205 44.61 44.61 56.58 42.79 51.13 6.945 50.167 4.0099 13.84 PLAY‐12‐0206 1.84 1.84 33.03 36.52 36.52 2.015 35.357 1.1633 5.699 PLAY‐12‐0207 1.71 3.87 1.527 2.79 1.08 54.744 138.82 50.58 85.57 37.65 112.2 26.625 33.56 PLAY‐12‐0208 PLAY‐12‐0209 2.34 2.34 70.99 57.66 77.8 10.24 68.817 5.9146 14.89 PLAY‐12‐0210 48.92 56.58 47.26 4.971 50.92 2.8703 9.763 167.45 167.5 PLAY‐12‐0211 3.87 3.68 3.42 0.226 3.657 0.1304 6.1779 37.67 37.67 PLAY‐12‐0212 2.59 3.36 0.544 2.975 0.385 18.302 50.58 56.58 90.7 4.243 53.58 3 7.918 242.66 289.84 633.11 33.361 266.3 23.59 12.53 PLAY‐12‐0213 1.46 1.33 0.092 1.395 0.065 6.5895 586.11 553.61 647.07 47.45 595.6 27.393 7.966 CYTOKINES IL‐4IL‐6IL‐8TNFa Samples Rep 1Rep 2Rep 3 Stdev Average Std Error %CV Rep 1Rep 2Rep 3 Stdev AverageStd Error %CV Rep 1Rep 2Rep 3StdevAverageStd Error %CV Rep 1Rep 2Rep 3StdevAverageStd Error %CV PLAY‐12‐0204 18.92 18.92 0 18.92 0 0 PLAY‐12‐0205 47.6 37.43 27.83 9.886 37.62 5.7079 26.28 PLAY‐12‐0206 14.17 14.17 PLAY‐12‐0207 47.6 75.66 19.84 61.63 14.03 32.19 PLAY‐12‐0208 PLAY‐12‐0209 153.53 127.06 129.42 14.65 136.67 8.4575 10.72 35.13 31.08 27.96 3.595 31.39 2.076 11.45 PLAY‐12‐0210 112.06 118.42 114.45 3.213 114.98 1.855 2.794 PLAY‐12‐0211 124.71 122.37 88.5 20.26 111.86 11.7 18.12 19.47 24.27 10.84 3.394 21.87 2.4 15.519 PLAY‐12‐0212 53.37 18.92 24.36 36.145 17.225 67.39 PLAY‐12‐0213 84.17 96.18 84.17 6.934 88.173 4.0033 7.864 8.49 10.55 11.13 1.387 10.057 0.801 13.796 35.1 34.26 35.93 0.835 35.097 0.4821 2.379

Appendix 1: The following reference provides details of age determination methodology used with the beluga whale eyes in the current contract:

Age estimation for bowhead whales, Balaena mysticetus, using aspartic acid racemization with enhanced hydrolysis and derivatization procedures

D.L. WETZEL1, J.E. REYNOLDS, III1, P. MERCURIO1,2, G.H. GIVENS3, 1 4 E.L. PULSTER AND J.C. GEORGE

Contact e-mail: [email protected]

ABSTRACT: Important information for determining effective conservation strategies for bowhead whales (Balaena mysticetus) includes accurate and precise determination of the ages of individual whales. Attempts to develop reliable methods to determine age for this species have included baleen carbon cycling analysis, aspartic acid racemization measurements, and assessments of corpora accumulation. Each of these methods has limitations affecting its utility. The objectives of this work were to identify the deficiencies and limitations of the existing methods for aspartic acid racemization measurement and to improve the effectiveness of such procedures to obtain consistent, reproducible results; this has been problematic to date. Using D/L ratios of aspartic acid in bowhead whale lens nuclei obtained from optimized analytical methods, we estimate whale ages employing a previously published aspartic acid racemization rate. The improved methods significantly increase the precision of D/L measurements. The resulting age estimates are also more precise than estimates generated from corpora count data. The enhanced techniques for measuring D/L may serve equally well for other hard-to-age species including other mysticetes, birds, and even homeothermic ectotherms. KEY WORDS: BOWHEAD WHALE, AGE DETERMINATION, ASPARTIC ACID RACEMIZATION

INTRODUCTION

Conservation of great whales and other marine mammals can be hindered by a lack of knowledge regarding the health of individuals, population demography and the extent to which threats are under control (Reynolds et al. 2005). Long-term survival of bowhead whales (Balaena mysticetus) and other ice-adapted species will be particularly influenced by direct effects of climate change and resultant changes in human activities. Ice-free seas will foster expanded commercial fishing and shipping, and oil and gas production, among other

1Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, Florida 34236. 2 Current Address: National Research Centre for Environmental Toxicology (Entox), The University of Queensland, Queensland Health Precinct, 39 Kessels Road, Coopers Plains, Qld 4108, Australia. 3 Department of Statistics, Colorado State University, Fort Collins, CO 80523. 4 Department of Wildlife Management, North Slope Borough, Barrow, AK 99734.

17 changes (Arctic Climate Impact Assessment 2004), which could affect life history parameters, carrying capacity and even the sustainability of these species (Reeves et al. 2012). Models to evaluate impacts of threats on marine mammals, including the great whales, must include data on factors such as age-specific reproductive performance and survival rate, population size and trend, and animal health (e.g., Runge et al. 2004). An important component of evaluating the effects of environmental change and consequent human activities on bowhead whales involves precise and accurate estimation of ages of individual whales. For most marine mammals, age determination is done by counting growth layer groups (GLGs) in teeth (Hohn 2009). Certain marine mammals, however, either lack teeth altogether (e.g., the members of the Suborder Mysticeti [baleen whales; Order Cetacea]) or replace teeth at an uncertain rate (manatees [Family Trichechidae; Order Sirenia]) making GLG counts inappropriate (Hohn 2009). Bowhead whales are mysticetes, so an alternative to counting GLGs must be used to estimate age. Attempts to develop reliable age estimation methods for this species have used baleen carbon cycling (Lubetkin et al., 2008; 2012), amino acid racemization (AAR) (George et al. 1999; Rosa et al. 2013), and corpora counting (Olsen and Sunde, 2002; George et al., 2011). However, there are shortcomings associated with each of the techniques used to date. The accuracy and precision of the corpora counting method are low for young animals, and are not applicable for males. Moreover, resulting age estimates can have high standard errors (e.g., Olsen and Sunde, 2002; George et al. 1999; 2011). To appropriately apply corpora aging techniques, it is necessary to know life history and other parameters including age of sexual maturity, age of onset of senescence (or even whether there is senescence), ovulation rate (and potential changes thereof) and whether corpora albicantia persist through the life of the animal (Olsen and Sunde, 2002). One might also consider using baleen length to estimate age. However, baleen is continuously worn down as bowheads grow older and wear rates need to be estimated to apply the technique, hence accurate age estimates can only be made for young whales. In contrast, there is significant information available in baleen carbon cycling analysis (Lubetkin et al, 2008; 2011). However, this technique has significant age range limitations and can only reliably be used for young whales. Measuring body length is also not an effective method for estimating age. The correlation between body length and whale age is poor, as we will see below, and the relationship is sex- specific. Aspartic acid racemization of the lens nucleus is an alternative method used for calculating ages of bowhead whales. Whereas the baleen cycling and corpora count estimation methods are fundamentally based on biological processes of an individual whale, AAR is based on physical chemistry of the rate of a chemical reaction. During gestation, aspartic acid is laid down in the nucleus of the lens where no metabolic activity occurs that would convert the L form of aspartic acid to the D form (Bada et al., 1980). Thus, in absence of metabolism, the conversion of L to D aspartic acid takes place only due to racemization over time. The kinetics of the rate of reaction constants of aspartic acid enantiomers can be ascertained using the Arrhenius equation which accounts for the effect of a change of temperature on the rate constant and, therefore, on the rate of the reaction. Whereas the

mechanics of the mathematical treatment for the Arrhenius equation are clear and simple to apply, the data generation needed to calculate the reaction rate, Kasp, can be compromised by analytical error. Although the use of AAR has shown considerable promise and led to several published analyses specifically for bowhead whales, a number of limitations and problems have been encountered when using this approach for ageing whales. The most serious of these problems seems to be with reproducibility of the D/L ratio data as a consequence of both analytical chemistry protocols and sample/standard instability. In light of such problems, a primary focus of our efforts has been to examine heretofore untested assumptions with regard to the efficacy of AAR approaches that have been used in previous studies of bowhead ages, starting with Bada et al. (1980). Wetzel et al. (2007) performed such testing and modified methods used by George et al. (1999) and Rosa et al. (2004, 2013) to provide bowhead age estimates that are significantly more precise than those of the earlier studies. This paper describes further refinement of methods to promote improved precision for age determination of bowhead whales, which has implications for enhanced ability to determine age of individuals for other taxa including other mysticete species. When this approach is applied to a large number of bowhead whales of various body sizes, sampled during different harvests, managers will have data that will improve their ability to assess both fundamental life history parameters and the relationships of threats to the ages of affected whales, as well as detect changes in such relationships over time. Having an accurate aging technique for both sexes will allow scientists and managers to understand whether particular demographic groups are most vulnerable to ship strikes and entanglements; monitor changes in life history and other parameters; assess changes over time in the extent to which fishing, shipping and other threats (e.g., contaminants) affect particular age groups; and inform effective mitigation actions before consequences of threats become critical.

MATERIALS AND LAB METHODS

Sample acquisition For approximately three decades, scientists with the North Slope Borough, Department of Wildlife Management (NSB-DWM), have worked closely with Alaska Native hunters to examine bowhead whales taken during subsistence hunts. During this period, a large number of eyes from freshly-dead whales have been preserved (frozen, intact) along with other tissue samples and basic biological data for each whale. For this study we selected lenses from 67 bowhead whales, ranging from fetuses to very large adults. We analyzed the D/L ratios for each of the 67 lenses and then used lens material from two whales to begin a pilot artificial aging study to evaluate the efficacy of using artificial aging to estimate Kasp and, ultimately, to use such methods to estimate age.

Figure 1: D/L ratio versus time for hydrolysis at two hydrolysis temperatures, 80ºC (solid) and 100ºC (dashed).

Time (min) % A (methanol) % B (acetonitrile) % C (N-acetyl-L-cysteine) 5 5.0 0.0 95.0 25 23.0 0.4 76.6 30 0.0 60.0 40.0 40 5.0 0.0 95.0

were run ‘blind’. For each eye, the lens nucleus was removed and trimmed. One half of the nucleus was retained frozen in a clean glass vial, and the other half was hydrolysed.

Eye lens removal and analysis of extracts The methods employed for acquisition of lens nuclei were generally similar to those described by George et al. (1999), Olson and Sunde (2002), and Rosa et al. (2004, 2013). Prior to the analyses, no information about whale size and sex was known– i.e., the samples were run ‘blind’. For each eye, the lens nucleus was removed and trimmed. One half of the nucleus was retained frozen in a clean glass vial, and the other half was hydrolyzed. Sample extracts were hydrolyzed and derivatized using methods which were modified from previous studies (George et al. 1999; Olson and Sunde 2002; Rosa et al. 2013).

Analyses for D and L isomers of aspartic acids were done in triplicate on a HyperClone reverse phase C18 column (120A, 250 x 4mm, 5micron; Phenomenex, Torrance, CA, USA) using a high performance liquid chromatograph (HPLC; Agilent Technologies, Santa Clara, CA, USA), equipped with an autosampler and scanning fluorescence detector (ƛex=230nm; ƛem=445nm). The HPLC flow rate was 1.5 ml/min, the column temperature was set to 30ºC and methanol (A), acetonitrile (B) and N-acetyl-L-cysteine (C) were used as eluants with the gradient program in Table 1.

Hydrolysis George et al. (1999) and others have followed the methodological lead of J.L. Bada, a pioneer in the application of amino acid racemization to age estimation for large whales (e.g., Bada et al. 1980). These scientists typically conducted hydrolysis of bowhead lens samples in 6M HCl at 100ºC for 6 hours. In contrast, our study tested a range of temperatures and durations for hydrolysis to find the point at which hydrolysis is optimized for the species in question (in this case, the bowhead whale). That optimum occurred using 6M HCl at 80ºC for 8 hours. Specifically, our study found that hydrolysis produces stable results after 8 hours, but more variable outcomes over shorter time periods. These results are illustrated in Figure 1. Neither 100ºC nor 80ºC is a temperature at which aspartic acid structure is affected in a manner that affects D/L ratios (Goodfriend 1997; Goodfriend and Meyer 1991).

Standards. Under most chemical analysis scenarios, the normal quality assurance and quality control information regarding chemical standard calibration is rather rote and unexceptional. However, in many previous studies (e.g., Olsen and Sunde 2002; Rosa et al. 2013) researchers have found that responses in calibration curves of the D and L isomers of aspartic acid were notably atypical. In almost all previous whale AAR research, day-to-day instrument responses were reported as inconsistent, totally absent, impaired or just “touchy”, to the point where, in one study, multiple models for standard responses needed to be developed using robust regression and then model selection techniques were applied (e.g. George et al. 1999). Because of this, we paid particular attention to designing a robust calibration curve comparison, bracketing each set of three lens samples with a standard and running all analyses in triplicate. All standards in this study were corrected for cross contamination as there were residual D isomers in the L isomer aspartic acid standard and vice versa. This cross contamination is expected and is due to the fact that the chemical reaction (racemization) is constant. The standards used for the calibration curve consisted of seven different ratio mixtures of D to L isomers of aspartic acid which were analyzed in triplicate for each set of 10 lens sample analyses. For each standard calibration curve analysis, we required the regression coefficient of determination (R2) to be at least 0.99 or a new standard calibration mixture was made and analyzed until our acceptable R2 value was achieved.

Derivatization Previous studies (e.g., George et al., 1999; Rosa et al., 2004) conducted derivatizations in which the amino acid extract being assessed was diluted 1:1000 with distilled water, and a subsample of the dilution was placed in a centrifuge with 10ul of OPA-NAC (ortho- phthaldialdehyde and N-acetyl-L-cysteine). This mixture was shaken for 20 seconds, and

centrifuged for 15 minutes, at which time 475 ul of 0.05 M sodium acetate buffer was added. Finally, 200 ul of this was analyzed by HPLC using methanol and sodium acetate for mobile phase at 1ml/min. In contrast to the multi-step process described above, our study followed a different approach (e.g., Kaufman and Manley 1998). We used OPA-IBLC (ortho-phthaldialdehyde and N-isobutyryl-l-cysteine) instead of OPA-NAC; our amino acid extract was rehydrated with 0.01M HCL and sodium azide (antibacterial) at 0.04ml/mg lens; and our extract was placed on the HPLC where derivatization was performed in a single step within the autosampler syringe. Conducting the derivatization in a single step within the syringe eliminates the possibility of inconsistencies or errors being introduced at each step of more traditional processes and the chemical stability problems that have been observed with previous methods.

Artificial aging studies The assessment of possible age-related changes in the racemization rate in lenses of a particular species can be done through artificial aging studies (Goodfriend 1997; Goodfriend and Meyer 1991). Fundamentally, heating of the lens extracts mimics the aging process, but at a faster reaction rate. Following heating of the hydrolysed extracts over a range of lengths of time and at three incubation temperatures, D/L ratios of the extracts are measured. The standard analysis (Goodfriend 1997; Goodfriend and Meyer 1991) applies the Arrhenius equation to these data to assess how the aspartic acid racemization rate (i.e., Kasp) changes with age. To conduct an artificial aging experiment for bowheads, lens samples were heated to three temperatures: 80ºC, 100ºC, and 120ºC, for a total of 50 samples analyzed in triplicate. After incubation of the extracts for different time periods at each temperature, the D/L ratio of aspartic acid was measured, as described above.

STATISTICAL METHODS

Age estimation Estimates of (D/L)i for the ith whale and (D/L)0 are used to estimate age according to the equation log (1 (DL /LD ) ) / (1 ( / ) )log(1(D/L))/ (1 (/D L) )  ii  00 Agei  2Kasp

(George et al., 1999). The (D/L)0 value was estimated using an inverse variance weighted average of five values. The first value is 0.0250 (s.e. 0.0013) from Rosa et al. (2013). This is estimated from a regression model using D/L data mostly for young whales of known ages (using corpora counts, baleen growth increments and fetus data). The remaining values are means of triplicated D/L measurements for four fetuses included in the present dataset. These values ranged from 0.0256 to 0.0292 with standard errors ranging from 0.0001 to 0.0005. For the ith whale, the observed data value of (D/L)i is taken to be the average of our three replicated measurements.

Variance estimation used a hybrid parametric and non-parametric bootstrap approach (Davison and Hinkley, 1997). The variance and 95% confidence interval were estimated separately for each whale. Also separately, for each whale, we re-sampled the three independent D/L measurements uniformly with replacement. Within each bootstrap iteration we also employed parametric re-sampling of (D/L)0 and Kasp. The approximate correlation between the estimates of (D/L)0 and Kasp is 4e-9, so this was ignored during re-sampling. Together, these bootstrap sampled quantities were used to generate one bootstrap pseudo- estimate of Agei. We used 10,000 bootstrap replications for each whale. Confidence intervals were generated using the percentile method. Using these age estimates and other estimates from 161 whales based on aspartic acid racemization, corpora counts, and baleen aging methods (Lubetkin et al., 2008; George et al., 2011; International Whaling Commission, unpublished), we fit the two-stage von Bertalanffy II (1938) model to estimate sex-specific growth curves. This is the same approach used by Lubetkin et al. (2012) except that we did not include growth spurts.

Artificial aging studies The standard statistical analysis for our artificial aging study would follow the procedures developed by Goodfriend (1997), estimating Kasp for the three temperatures and then using

the Arrhenius equation KEasp Aexp a / (RT ) to estimate the relationship between Kasp and temperature. Here, Kasp is the rate constant for the reaction from the L form to the D form of aspartic acid, R is the universal gas constant, and T is the temperature of the living whale in degrees Kelvin. The parameters A and Ea are estimable from the heating experiment data via linear regression based on the log transform of the above model. The results may be extrapolated to estimate Kasp for lens temperatures appropriate for bowheads. However, our results indicated that this standard analysis would not be appropriate for the bowhead data obtained from our heating experiment for reasons discussed further below.

RESULTS AND DISCUSSION

Age estimation Table 2 provides estimated ages, bootstrap standard errors and bootstrap 95% confidence intervals for the whales analysed in this study. Negative age estimates are sensible because the D/L values include measurement uncertainty; such estimates should be interpreted as ‘very young’. The most striking aspect of these results is that there is evidence that some bowhead lifetimes may extend nearly 200 years or beyond. This is consistent with previous findings from other researchers. Excluding negative age estimates, 80% of the coefficients of variation for these estimates are within 0.01 of 0.16, indicating a very consistent degree of uncertainty on the log scale. The ages of eleven of the whales we analysed here have previously been estimated by other researchers, using a variety of techniques including corpora counting and baleen isotope cycle analysis. Table 3 compares our estimates to these previous results. Our aspartic acid racemization estimates are consistent with the baleen cycle estimates from Lubetkin et al. (2008). However, the corpora counting age estimates of George et al. (2011) appear to be generally higher than the estimates from other methods.

Figure 2 shows a plot of the age estimates for the whales we analysed and 172 additional whales previously aged by other researchers. Each whale is represented by a dot (the point estimate) and a horizontal bar (spanning the 95% confidence interval). Red bars correspond to females, and males are represented by blue. The whales aged in our study are shown with heavier lines than for the whale ages from other researchers. The black lines in Figure 2 show the fitted sex-specific von Bertalanffy II (1938) growth curves, with female whales being larger than males of the same age.

Artificial aging experiment Applying the Goodfriend (1997) heating experiment method to the bowhead data yields an estimated relationship between Kasp and temperature. Lens temperature for living bowhead whales is unknown, however Sformo et al. (unpublished report, 2011) attempted to measure the temperature of lenses from a few very freshly dead bowhead whales. Although those measurements occurred some time after death, and the data should be considered quite preliminary, the mean temperature was 11.3ºC (SE + 1.9ºC), which is considerably lower than the core body temperature of 33.8º (George, 2009). Sformo et al. suggested that (a) lens temperature may be higher if peripheral cooling after death occurs, or (b) lower lens temperatures are possible if extrapolations are made from cornea temperatures. Using the estimated relationship between Kasp and temperature, if we set Kasp=0.000977, which is the estimate of Rosa et al. (2013), our analysis yields age estimates that resemble the estimates of those authors, but correspond to a lens temperature that is considerably lower than the conventional wisdom or the range suggested by the scant empirical data of Sformo et al. (unpublished report, 2011). Conversely, if we set lens temperature to be similar to the estimate of Sformo et al. and solve for Kasp, the resulting age estimates are surprisingly low. We have not presented the estimated parameters of the model or ages here. There are two key reasons why.

First, due in part to the increased precision of our D/L measurements, we have discovered that the heating experiment yields data indicating a statistically significant nonlinear relationship between duration of heating exposure and D/L. This nonlinearity is most pronounced for the 80º data, and diminishes as temperature increases. Such a relationship was unexpected and prevents application of the Goodfriend (1997) method unless the nonlinearity is ignored. Second, it is important to recognize that application of the estimated Arrhenius equation to bowheads requires an enormous extrapolation from temperatures around 100º to perhaps 10º or less. The modest estimation uncertainty in the vicinity of tested temperatures is greatly magnified by this extrapolation, even assuming that the linear regression model is still appropriate at such extreme untested temperatures. Thus, it is apparent that the artificial heating data contain a strong statistical signal, but that further work must be done to develop an appropriate analysis approach akin to that of Goodfriend (1997) for simpler data. Such research is ongoing.

CONCLUSIONS

Considerable effort has gone into development and application of methods to age bowhead whales. Several approaches have emerged, including analysis of AAR of the nucleus of the eye lens. Whereas earlier studies using AAR have provided valuable insights into bowhead longevity and life history, the methods described here present an opportunity to further improve such analyses. The novel methodological changes described herein provide D/L data that are remarkably consistent. The specific modifications developed by this study include (a) species specific hydrolysis time and temperature that optimizes the preparation of the lens aspartic acid for further analysis; (b) stable and consistent calibration curves to eliminate the need for modelling standards; (c) use of single step, within-syringe derivatization; and (d) investigation of artificial aging experiments that may shed light on questions such as constancy of Kasp with age, lens temperature, and ultimately age estimates using data based on physical chemistry rather than more plastic features. In concert with the Kasp estimate of Rosa et al. (2013), the data presented here provide age estimates for 64 whales not previously aged, and the results are strongly consistent with previous studies that support the hypothesis of extreme longevity for some bowhead whales. Further research with AAR, other aging techniques and, particularly, artificial aging experiments offers promising avenues for continuing to improve our understanding of bowhead life history.

ACKNOWLEDGEMENTS

This study could not have been done without funding provided by the North Slope Borough, Department of Wildlife Management, and the National Marine Fisheries Service. In addition to funding, the Borough provided incredible logistic support in the field for the PIs, as well as generous use of office space and computers. We are especially grateful to Ms. Cyd Hanns, Dr. Robert Suydam and Gay Sheffield for their support and specimen collection. We also thank the whaling captains and crews for allowing us to work with them during spring and fall hunts to acquire the specimens that were used in this study. We thank Susan Lubetkin for her suggestions regarding analysis of the heating experiment data with the Arrhenius equation approach. Finally, we thank Dr. Judith Zeh (University of Washington) for sharing her vast expertise on bowheads.

REFERENCES Arctic Climate Impact Assessment (2004) Impacts of a Warming Arctic. Cambridge University Press, England Bada, J.L., Brown, S. and Masters, P.M. (1980) Age determination of marine mammals based on aspartic acid racemization in the teeth and lens nucleus. p. 113-118. In: W.F. Perrin and A.C. Myrick (eds) Age Determination of Toothed Whales and Sirenians. Rep. int. Whal. Commn, Special Issue 3, Cambridge, England. Davison, A.C. and Hinkley, D.V. (1997) Bootstrap Methods and Their Applications. Cambridge University Press, Cambridge, United Kingdom.

George, J.C. 2009. Growth, morphology and energetics of bowhead whales (Balaena mysticetus). Dissertation, University of Alaska Fairbanks George, J.C., Bada, J., Zeh, J., Scott, L., Brown, S.E., O’Hara, T. and Suydam, R. (1999) Age and growth estimates of bowhead whales (Balaena mysticetus) using aspartic acid racemization. Canadian J Zool. 77: 571-580. George, J.C, Follmann, E., Zeh, J., Sousa, M., Tarpley, R.J. and R. Suydam. 2011. A new way to estimate whale age using ovarian corpora counts. Canadian Journal of Zoology. Can. J. Zool. 89: 840–852. Goodfriend, G.A. (1997) Aspartic acid racemization and amino acid composition of the organic endoskeleton of the deep-water colonial anemone Gerardia: determination of longevity from kinetic experiments. Geochimica et Cosmochimica Acta 61:1931-1939. Goodfriend, G.A. and Meyer, V.R. (1991) A comparative study of the kinetics of amino acid racemization/epimerization in fossil and modern mollusk shells. Geochimica et Cosmochimica Acta 55:3355-3367 Hohn, A.A. (2009) Age estimation. p. 11-17. In: W.F. Perrin, B. Würsig and J.G.M. Thewissen (eds). Encyclopedia of Marine Mammals. Elsevier, Inc., San Diego, CA. Kaufman, D.S. and Manley, W.F. (1998) A new procedure for determining DL amino acid ratios in fossils using reverse phase liquid chromatography. Quat Geochronology, 17:987-1000. Lubetkin, S.C., Zeh, J.E., Rosa, C., and George, J.C. (2008) Age estimation for young bowhead whales (Balaena mysticetus) using annual baleen growth increments. Canadian J Zool., 86:525-538. Lubetkin, S.C., Zeh, J., and George, J. (2012) Statistical modeling of baleen and body length at age in bowhead whales. Canadian Journal of Zool., 90(8):915-931. Olson E. and Sunde, J. (2002) Age determination of minke whales (Balaenoptera acutorostrata) using the aspartic acid racemization technique. Sarsia, 87:1-8. Reeves, R., Rosa, C., George, J.C., Sheffield, G. and Moore, M. (2012) Implications of Arctic industrial growth and strategies to mitigate future vessel and fishing gear impacts on bowhead whales. Marine Policy 36:454–462. Reynolds, J.E., Perrin, W.F., Reeves, R.R., Ragen, T.J. and Montgomery S. (eds) (2005) Marine Mammal Research: Conservation Beyond Crisis. Johns Hopkins University Press, Baltimore, MD. Rosa, C., George, J.C., Zeh, J., Botta, O., Zauscher, M., Bada, J., O’Hara, T.M. (2004) Update on age estimation of bowhead whales (Balaena mysticetus) using aspartic acid racemization. Paper SC/56/BRG6 presented to the Scientific Committee of the International Commission, June, 2004. Rosa, C., Zeh, J., George, J.C., Botta, O., Zauscher, M., Bada, J. and T.M. O'Hara. (2013) Age estimates based on aspartic acid racemization for bowhead whales (Balaena mysticetus) harvested in 1998-2000 and the relationship between racemization rate and body temperature. Marine Mammal Science, Runge, M.C., Langtimm, C.A. and Kendall, W.L. (2004) A stage-based model of manatee population dynamics. Marine Mammal Sci. 20:361-385. Sformo, T., Hanns, C. and George, J.C. (2011) Preliminary findings on lens temperatures in the bowhead whale (Balaena mysticetus). Unpublished report, North Slope Borough Department of Wildlife Management, Barrow, AK.

Von Bertalanffy, L. (1938) A quantitative theory of organic growth (Inquiries on growth laws, II). Human Biology, 10(2):181-213. Wetzel, D.L., Mercurio, P., Reynolds, J.E. and George, J.C. (2007) The pursuit of precise and accurate methods to determine ages of bowhead whales (Balaena mysticetus). Proc. 17th Biennial Conference on the Biology of Marine Mammals, 29 November-3 December, Cape Town, South Africa.

Whale Age 2.5% 97.5% SE Whale Age 2.5% 97.5% SE 02B17 7.5 5.7 10.2 1.2 07B9F -2.1 -3.0 -1.6 0.4 02B2 52.0 39.4 71.9 8.3 07G3 39.9 30.2 55.1 6.5 02B21 12.6 9.4 17.4 2.0 07G4 29.3 22.2 40.6 4.8 02B22 1.7 1.2 2.4 0.3 07S1 9.5 7.1 13.3 1.6 02B3 106.3 80.6 146.0 16.9 07S2 7.5 5.6 10.4 1.2 02B5 4.8 3.6 6.6 0.8 07S3 17.7 13.3 24.2 2.8 03B6 19.4 14.6 26.7 3.2 07S4 34.8 26.3 47.7 5.5 03B9 68.3 51.6 93.5 11.0 08B14 27.4 20.6 37.8 4.4 04B4 22.2 16.8 30.4 3.5 08S3 187.7 141.9 258.0 29.8 04B5 80.1 60.4 111.0 13.0 09KK1 -3.1 -4.3 -2.4 0.5 04B8 24.0 18.3 32.9 3.9 10B15 20.1 15.2 27.6 3.2 04B9 18.6 14.1 25.7 3.0 11B3 56.1 42.3 77.8 9.1 04G2 4.4 3.3 6.0 0.7 11B4 2.5 1.3 4.1 0.7 04KK1 123.3 93.2 169.3 19.9 11B5 37.2 28.2 51.3 6.0 05B11 17.5 13.3 24.2 2.9 11B6 71.3 53.8 98.1 11.3 05B12 28.2 21.3 38.8 4.6 11B7 157.3 118.8 214.8 25.0 05B21 7.2 5.3 10.1 1.2 12B15 -0.6 -1.2 -0.1 0.3 05B25 14.8 11.2 20.3 2.4 12S2 23.3 17.7 32.2 3.7 05B8 1.3 0.8 2.0 0.3 12S2F -1.6 -2.3 -1.0 0.3 05B8 0.3 0.1 0.6 0.1 12S3 1.2 0.6 2.0 0.4 05S5 47.4 36.1 65.7 7.6 81WW2 73.2 55.6 100.1 11.5 05S7 81.5 61.6 112.7 13.0 96B5 121.4 91.8 166.6 19.5 06B10c -5.1 -6.9 -3.9 0.8 97B10 58.1 44.1 79.6 9.2 06B18 54.5 41.3 75.1 8.8 97B12 67.4 50.9 93.1 10.9 06B6 28.0 21.2 38.6 4.5 97B5 5.0 3.8 6.9 0.8 07B10 37.9 28.7 52.2 6.1 97B7 13.1 10.0 18.1 2.1 07B11 78.1 58.7 108.1 12.6 97B8 18.6 14.0 25.5 3.0 07B12 32.3 24.4 44.6 5.2 98B20 16.2 12.2 22.5 2.6 07B13 88.6 66.9 122.2 14.3 98B21 48.4 36.4 66.5 7.8 07B16 28 4 21 5 39 0 45 98B4 22 1 16 8 30 5 36

Figure 2: Fitted von Bertalanffy growth curves. Each whale is represented by a dot (the point estimate) and a horizontal bar (spanning the 95% confidence interval). Red bars correspond to females, and males are represented by blue.

Whale Age s.e. Method Citation 52.0 8.3 AAR Here 02B2 79 18 Corpora G11 65.9 12.0 Corpora DAA 12.6 2.0 AAR Here 02B21 11.7 2.3 Baleen L08 106.3 16.9 AAR Here 02B3 139 38 Corpora G11 114.1 23.5 Corpora DAA 68.3 11.0 AAR Here 03B9 102 26 Corpora G11 85.0 15.9 Corpora DAA 24.0 3.9 AAR Here 04B8 31 6 Corpora G11 18.6 3.0 AAR Here 04B9 43 8 Corpora G11 28.2 4.6 AAR Here 05B12 38 7 Corpora G11 121.4 19.5 AAR Here 96B5 125 38 Corpora G11 114.1 23.5 Corpora DAA 58.1 9.2 AAR Here 97B10 65 14 Corpora G11 55.2 9.3 Corpora DAA 18.6 3.0 AAR Here 97B8 31 6 Corpora G11 27.5 5.1 Corpora DAA

Table 3: Comparison of age estimates from various studies. L08 refers to Lubetkin et al. (2008). G11 refers to George et al. (2011). DAA refers to data available from the International Whaling Commission under its Data Availability

Appendix 2: Levels of three acute phase proteins (C-reactive protein—CRP, serum amyloid A—SAA, and haptoglogin) in serum samples from beluga whales taken in the Point Lay subsistence hunt in summer 2012. Samples were analyzed by University of Miami