JOURNALOF THE PENNSYLVANIA ACADEMY OF SCIENCE

Founded on April 18, 1924

December 2016Volume 90(2)

ISSN: 1044-6753

PAS Home Page: http://pennsci.org

Carl R. Pratt, Ph.D.

Editor

Department of Biology

Immaculata University

Immaculata, PA 19345

Contents

EDITORIAL POLICY AND FORMAT i

PROBING THE UTILITY OF ISOXAZOLINES AS QUORUM SENSING INHIBITORS OF GRAM-NEGATIVE BACTERIA

31

JESSICA R. KUEHNE, CONNOR M. LINK, IVONNE N. SCHNEIDER, AMY M. DANOWITZ

THE INFLUENCE OF SHIFTED SPECTRAL SENSITIVITIES ON PREDATOR DETECTION ABILITY OF LAKE MALAWI AFRICAN CICHLIDS

37

KELLY M. HANSON AND BRIDGETTE E. HAGERTY

DETECTION OF BORRELIA BURGDORFERI AND BORRELIA MIYAMOTOI IN WHITE-FOOTED MICE, PEROMYSCUS LEUCOPUS, AND BLACKLEGGED TICKS, IXODES SCAPULARIS, IN MONROE COUNTY, PENNSYLVANIA, A LYME ENDEMIC REGION

44

MEAGHAN BIRD AND JANE E. HUFFMAN

CONJUGATIVE COMPETENCY OF MERCURY RESISTANT BACTERIA ISOLATED FROM ONONDAGA LAKE, NY

48

COLLEEN WALSH AND LAURIE F. CASLAKE

TICKS OF BLACK BEARS (URSUS AMERICANUS) FROM NORTHEASTERN PENNSYLVANIA AND TICK-BORNE PATHOGENS

56

ELIZABETH MCGOVERN, J. MISCHLER, M. BIRD, AND J. HUFFMAN

i

The Journal of the Pennsylvania Academy of Science publishes original papers, research notes, commentary, editorials, view points, and review ar-ticles in the natural, physical, engineering, and social sciences. All papers must discuss the relevance of the data presented and a clear interpretation of its meaning in view of current knowledge of the discipline concerned. Helpful references for the author are: (1) Day, R.A. 1983. How to write a scientific paper. 2nd ed. ISI Press, Philadelphia, xv + 181 pp.; (2) O’Connor, M. and F.P. Woodford. 1976. Writing scientific papers in English, Elsevier, Amsterdam, vii + 108 pp.; (3) MacGregor, A.J. 1979. Graphics simplified; and (4) How to plan and prepare effective charts, graphs, illustrations, and other visual aids, University of Toronto Press, Toronto, 1-64 pp.

Authors are requested to examine recent issues of the Journal in order to conform to the general style of the journal. Papers are accepted for consid-eration at any time. Submitted manuscripts are accepted for review with the understanding that the same work has not been published, copyrighted or submitted for publication elsewhere and that all persons cited as a personal communication have consented to be cited. Additionally, submission of the manuscript is a representation that all the authors for the said manuscript and the institution where the research was carried out have approved its publica-tion. Signed authorization will be required as appropriate. Authors are billed for page charges to partially defray the costs of publishing.

Submit names, email addresses, as well as the professional area of exper-tise of 4 possible reviewers who have agreed to review your manuscript. The reviewers must be outside the author’s institution, possess knowledge of current research in the area of study, and generally be professionally quali-fied to referee the paper. The peer reviewing process is the Editor’s responsi-bility, and the reviewers are selected at the discretion of the Editor.

All authors are requested to conform to the following:

1. General Format. All manuscripts should be typed, and double spaced, with 3 cm margins all around. Do not use single spacing anywhere (includ-ing Literature Cited). Images should be submitted as jpegs or tif and in the English language. Manuscripts should be organized as follows: (1) an un-numbered cover sheet with Title, Authors, their institutions and addresses, and name, address, and telephone number of the author to receive proof, (2) an unnumbered sheet with an Abstract, (3) Introduction, (4) Materials and Methods, (5) Results, (6) Discussion, (7) Acknowledgements, and (8) Literature Cited. All pages of the text, Introduction through Literature Cited, are to be numbered, and the names of authors should appear in the upper right-hand corner of each page. The text should begin in the middle of the first numbered page. Manuscripts need to be submitted in English.

2. Headings. All headings are in CAPITAL letters and centered.3. Title. Brief and to the point. It should inform the reader of the subject of the paper.

JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE

PENNSYLVANIA ACADEMY OF SCIENCEJOURNAL INFORMATION FOR AUTHORS

EDITORIAL POLICY AND FORMAT

4. Byline. Include author’s name, name of institution, department, address and zip code.

5. Abstract. A clear and concise paragraph which summarizes the research.

6. Introduction. The introduction should be concise and offer only that infor-mation necessary to orient the reader to the purpose and scope of the paper. It should state the reasons for the work and cite only published literature relevant to the subject.

7. Materials and Methods. Describe materials, methods, and equipment. Avoid repeating previously published details, unless modifications are ex-tensive. The necessity of conciseness should not lead to omission of im-portant experimental details necessary for others to repeat the work. When applicable, describe the experimental design and justify its use.

8. Results and Discussion. The Results section is a clear and concise account of the findings. Data should be presented in the most efficient manner, either in text, tables, or illustrations. All tables and illustrations must be referenced in the text. The Discussion section should extend or contradict current pub-lished information on the subject. Limit the discussion to the relevant subject and avoid speculation.

9. Acknowledgements. The source of any financial support received for the work being published must be indicated in the Acknowledgments section. The usual format is as follows: “This work was supported by Public Health Service grant CA-01234 from the National Cancer Institute.”

Recognition of personal assistance should be given as a separate para-graph, as should any statements disclaiming endorsement or approval of the views reflected in the paper or of a product mentioned therein.

10. Appendixes. Appendixes that contain additional material to aid the reader are permitted. Titles, authors, and reference sections that are distinct from those of the primary article are not allowed. If it is not feasible to list the author(s) of the appendix in the byline or the Acknowledgments section of the primary article, rewrite the appendix so that it can be considered for publication as an independent article, either full-length paper or Note style. Equations, tables, and figures should be labeled with the letter “A” preceding the numeral to distinguish them from those cited in the main body of the text.

11. Literature Cited and Footnotes. Except for the title and author reference at the beginning of the paper, and superscript notation in tables, do not use footnotes. Create separate Appendices or an Endnotes section if additional supplementary text material is required. Place Endnotes section just before the Literature Cited section. Number each endnote within the Endnote sec-tion using Arabic numbers in the order in which they are referred to in the

iiJOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(1), 2016

other sections of the manuscript. In other sections of the manuscript, place endnotes reference numbers in parentheses, and use the text style of type and not superscript. Place appendices after the Literature Cited section. Include a Literature Cited section: list references in alphabetical order by first author. Include only published references cited in the manuscript; unpublished work normally will be cited as personal communication (pers. comm.) in other sections of the manuscript, e.g., J.R. Halma (pers. comm.) or (J.R. Halma, pers. comm.). List all authors and full citation in the Literature Cited sec-tion. Use the most recent issue of the recognized abstracting authority to determine the correct abbreviations of periodical names (e.g., for biology use BIOSIS, Bioscience Information Service, Philadelphia, PA). If in doubt, do not abbreviate serial names. Use the following format and style for the Literature Cited section:

Journal- Monmonier, M. 1987. Title. J. Pa Acad. Sci. 62:73-77.

Book (Select pages)-Snedecor, G. W. and W. G. Cochran. 1976. Statistical Methods. The Iowa State Univeristy Press. Ames, IA, 237-238.

Book (Complete work)-Snedecor, G. W. and W. G. Cochran. 1976. Statistical Methods. The Iowa State Univeristy Press. Ames, IA, xix + 593 pp.

For Internet citation, choose either MLA, APA, Chicago or another appropri-ate style, but stay consistent in the manuscript.

In all but the Literature Cited section, cite all works by author and year. For works with one or two authors, include names in each citation, e.g., (Smith and Reif 1984), or, if authors’ names are used in the text- Smith and Reif (1984); for works by three or more authors, use et al. after the first author, e.g., (Gur et al. 1983), or, if the authors’ names are used in the text- Gur et al. (1983). Research Notes with fewer than five references should be cited within the other sections of the manuscript thereby eliminating the need for a Literature Cited section. When references are cited within the text of other sections, include authors by last name only, and do not use et al. in the citation, e.g., for a journal article- (Genys, Harman and Fuller 1984, Proc. Pa. Acad. Sci. 58:67-69), or if authors are used in the text- Genys, Harman and Fuller (1984, Proc. Pa. Acad. Sci. 58:67-69); for a book-(Snedecor and Cochran, 1976, Statistical Methods, The Iowa State University Press, Ames, IA, 237-238), or, if authors are used in the text-Snedecor and Cochran (1976, The Iowa State University Press, Ames, IA, 237-238).

12. Research Notes. Papers submitted as short communications with an abstract are classified as Research Notes. Research Notes must contain the same basic quality of content and order of presentation as more substantial papers having content separated by section. Citations must follow the same format as articles.

13. Case Reports. While a full-length article or a Note may contain a case report section when the report is incidental to the rest of the paper, a specif-ic Case Report format must be used when the report constitutes the entire article.

A Case Report must include an abstract of no more than 50 words. The text starts with presentation of the case under the section heading “Case Report”; there is no introductory text before the Case Report heading. After the case is presented, the rest of the text follows in a separate sec-tion after a ruled line to separate the sections. No separate head is used

for this short discussion section, but paragraph lead-ins are permitted. The total number of tables and figures (combined) must not exceed 3.

14. Editorials. These are commissioned only articles. Original papers should not be submitted under this article type.

15. Commentaries. These are commissioned-only articles. Original pa-pers should not be submitted under this article type. They are generally not peer-reviewed.

16. View point. These articles are academic papers which address an issue of current public or professional debate and/or approach an issue from a more personal perspective than standard academic writing.

17. Tables and Illustrations. Tables must have a title, be numbered, and typed on a separate page. Tables must be created using the tables tool in Microsoft Word or as text with tabs separating columns and paragraph re-turns separating rows. Computer generated images or scans should be saved as Tiff, EPS, or JPEG files. Resolution (at the dimensions to be published) should be at an absolute minimum of 300 dpi. Provide the legends for all illustrations in consecutive order on a separate page.

18. If animals are used in the research, the author(s) must state in the ma-terial and method section that the study was conducted in accordance with the guidelines laid down by the U.S. Office of Laboratory Animal Welfare (or individual country of origin) and that the research was approved by the Institutional Animal Care and Use Committee. All human subject research, and all other activities, which in part involve human subject research, must be reviewed and approved by the institutions IRB committee.

19. There are no page charges for publication in the journal, however one of the authors must be a member of the Pennsylvania Academy of Science. Galley proofs will be sent to the authors for checking; they must be returned to the Editor within a week after receipt.

20. Editorial Policy. Every paper is reviewed by the Editor and selected professional referees. Manuscripts will be returned or rejected if considered unsuitable for publication.

21. Manuscripts and Correspondence. Address all inquiries relating to pub-lication in the Journal to the Editor: Dr. Carl Pratt, Immaculata University, Department of Biology, 1145 King Road, 210B Loyola Hall, Immaculata, PA 19345, CPratt@immaculta.edu

Journal of the Pennsylvania Academy of Science 90(2): 31-36, 2016

PROBING THE UTILITY OF ISOXAZOLINES AS QUORUM SENSING INHIBITORS OF GRAM-NEGATIVE BACTERIA1

JESSICA R. KUEHNE, CONNOR M. LINK, IVONNE N. SCHNEIDER, AMY M. DANOWITZ2

Mercyhurst University, Erie, Pennsylvania 16504

ABSTRACT

One of the most serious threats to human health is the increase in the number of bacteria that are resistant to most or all current antibiotics. To combat this issue, new antibiotics that have novel modes of actions against their bacterial targets are needed. One bacterial system that may provide a fruitful target for new antibiotics is the quorum sensing system. Quorum sensing (QS) is the process by which bacteria use low molecular weight ligands to sense their population densities. Once a sufficient density has been achieved, certain population dependent phenotypes such as activation of virulence factors and production of factors that allow for biofilm formation are activated. Gram-negative bacteria species use small molecule acyl homoserine lactones as their natural QS ligands. In addition to these natural ligands, non-natural small molecules have been synthesized and used to effectively modulate quorum sensing. The work reported here centers on using isoxazolines as a scaffold for designing QS modulators. A focused library of functionalized isoxazolines was synthesized and their ability to modulate QS was tested in Vibrio fischeri (a Gram-negative bacteria species with a well-understood QS mechanism). Of the twelve isoxazolines synthesized, seven were seen to act as QS inhibitors in V. fischeri.[ J PA Acad Sci 90(2): 31-36, 2016 ]

INTRODUCTION

According to the Centers for Disease Control (2013), “each year in the United States, at least 2 million people become infected with bacteria that are resistant to antibiotics and at least 23,000 people die each year as a direct result of these infections.” With multi-drug resistant bacteria on the rise, new and more effective ways of treating these diseases are in

high demand. As many previously targeted bacterial systems are becoming increasingly difficult to hit, it is important to find not just new antibiotics, but also new bacterial targets that have not yet been exploited. One system that may prove to be a fruitful target for new classes of antibiotics is the quorum sensing (QS) system (Clatworthy, A. et. al. 2007). Quorum sensing is the process by which bacteria use low molecular weight ligands to monitor their population densities. Once a certain density has been reached, bacteria activate various group behaviors including bioluminescence, release of virulence factors, and biofilm formation. It is thought that if QS can be inhibited, the virulence of the bacteria that rely on this system could be attenuated (for reviews on this subject, see: Bassler, B.L. 2002; Bassler, B.L. and Lossick, R.L 2006).

As such, there have been many studies aimed at determining how to modulate bacterial QS with Gram-negative bacteria being the most well studied. The native QS ligand for Gram-negative bacteria is the acyl homoserine lactone (AHL) which was first discovered in the marine bacterial species Vibrio fischeri (Elberhard, A. et. al. 1981) (Fig. 1 A). Since this discovery, many groups have focused on synthesizing non-natural QS ligands. Through these efforts, several classes of molecules were discovered that have the ability to modulate quorum sensing (Stevens, A. et. al. 2011). The most common classes of molecules include non-natural AHLs (Geske, G., et. al. 2005; Geske, G. et.al. 2007; Geske, G. et.al. 2007; McInnis, C., et. al. 2011; O’Loughlin, C., et. al. 2013; Stacy, D., et. al. 2012) (Fig. 1 B), halogenated furanones (Koch, B., et. al. 2006; Manefield, M. et.al. 2002) (Fig. 1 C), tetrazoles (Müh et al. 2006) (Fig. 1 D), and terphenyls (O’Reilly, M.C. and Blackwell, H.E. 2016) (Fig. 1 E). It is currently unknown if these classes of molecules represent privileged scaffolds for QS modulation or if other scaffolds could also prove to be useful in this respect.

As many of the known QS inhibitors feature 5-membered heterocycles, it was decided to try to expand the scope of QS inhibitors using an additional heterocyclic scaffold; the five-membered isoxazoline ring. Isoxazolines were chosen because they provide the dual benefits of synthetic ease and high possibility for functionalization. A focused library of twelve isoxazolines with varying functionality at the C3 position was synthesized and characterized, ten of which were tested for their ability to modulate QS in the Gram-

1Accepted for publication August 2016.2Corresponding Author: Mercyhurst University 501 East 38th St. Erie, Pa 16504 Ph: 814-824-2029 email: adanowitz@mercyhurst.edu

32JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

negative bacteria species Vibrio fischeri (Table 1). V. fischeri were used as they provide three major benefits for this type of testing. First, their QS system has been extensively studied. In brief, V. fischeri employ a LuxI/R type QS system. AHLs are produced in the cytoplasm of cells by the LuxI enzyme. The AHLs then exit the cell through passive diffusion where they can be taken up by other V. fischeri. At high cellular densities, sufficient quantities of AHL are produced such that they can bind the LuxR protein receptor. Once bound to an AHL, LuxR is able is activate the transcription of the luxICDABEG operon which encodes for a number of genes including those responsible for bioluminescence and LuxI. The production of additional LuxI leads to the production of more AHL and enhances a positive feedback loop for QS (Lupp, C. et. al. 2003). Second, they are a naturally bioluminescent bacterial species and their bioluminescence is controlled by QS. This provided an ideal way to monitor QS modulation as changes in luminescence could be easily measured. Finally, V. fischeri is a non-virulent species, meaning that they can be handled on the bench-top without the extensive precautions used with more virulent bacterial species. It should be noted that while V. fischeri are not dangerous to human health, LuxI/R homologs have been discovered in dozens of bacterial species (including species that act as human pathogens) (Case, R.J. et. al. 2008), which indicates that results discovered in V. fischeri may be relevant to these species.

MATERIALS AND METHODS

Isoxazoline synthesis

General: All chemicals used in these syntheses were purchased from Sigma Aldrich, Fisher Scientific, Post Apple Scientific, or SPC Science and were used without additional purification. Purification of isoxazolines was carried out using flash silica gel chromatography (Dynamic Adsorbents 32-63 microns) according to the procedure of Still, Kahn, and Mitra (1978). All reactions were run under ambient conditions. 1H NMR was recorded in CDCl3 on a 60 MHz Anasazi instrument. IR spectra were taken using a Perkin-Elmer Spectrum One with universal ATR attachment.

General procedure for oxime synthesis: To a stirring solution of toluene (50 mL) was added hydroxylamine hydrochloride (62 mmol, 6.2 eq) and sodium carbonate (62 mmol, 6.2 eq). Aldehyde (10 mmol, 1 eq) was also added. The reaction was allowed to stir at room temperature overnight. Once starting material was no longer present via TLC analysis (3:1 hexanes: ethyl acetate), the reaction was quenched with DI water (25 mL), and the organic and aqueous layers were separated. The aqueous layer was washed three times with ethyl acetate (25 mL). All organic layers were combined, dried over Na2SO4, filtered and concentrated using a rotary evaporator. The crude mixture was then carried over.

Figure 1. Molecules that have been used to modulate quorum sensing. A. The natural AHL from V. fischeri. B. An example of a non-natural AHL that is capable of inhibiting quorum sensing. C. An example halogenated lactone. D. A tetrazole containing quorum sensing inhibitor. E. A terphenyl quorum sensing inhibitor.

33

General procedure for isoxazoline synthesis: Toluene (50 mL) was added to a foil covered three-neck flask and was cooled to 0 °C in an ice/water bath for 15 minutes. Once cooled, oxime (carried over as crude from the previous step, 1 eq) and allyl alcohol (100 mmol, 10 eq) were added to the flask. Bleach (Great Value obtained from a local Wal Mart) (100 mL to give a final hypochlorite concentration of 0.1 M) was added dropwise via a separatory funnel over the course of 30-45 minutes. The reaction was allowed to run for 12-36 hours until starting material was no longer present by TLC (3:1 hexanes to ethyl acetate). The reaction was quenched with DI water (25 mL), and the organic and aqueous layers were separated. The aqueous layer was then washed 3 times with ethyl acetate (25 mL). The organic layers were combined, dried over Na2SO4, filtered and concentrated using a rotary evaporator. The resulting product was then purified using flash chromatography and characterized by 1H NMR and IR.

Characterization data for isoxazolines:

(3-phenyl-4,5-dihydro-5-isoxazolyl)methanol (Table 1, entry 1): Prepared via the general procedure but scaled up 2x in 98% yield (1.58 g) as a white solid. 1H NMR: δ 2.18 (s, 1H), 3.41 (dd, J = 1.1 Hz; 9.2 Hz, 2H), 3.78 (m, 2H), 4.87 (m, 1H), 7.58 (m, 5H). IR: 3386, 3104, 2947, 2845, 1596, 1446, 1354 cm-1.

[3-(4 -chlorophenyl)-4,5 -dihydro-1,2-oxazol-5 -yl]methanol (Table 1, entry 2): Prepared via the gener-al procedure in 8% yield (0.050 g) as an off-white solid. 1H NMR: δ 1.4 (s, 3H), 1.5 (s, 3H), 3.2 (dd, J = 0.96 Hz; 9.2 Hz), 3.70 (m, 2H), 7.5 (m, 5H). IR: 3380, 2923, 2853, 1598, 1496 cm-1.

[3-(4-methoxylphenyl)-4,5-dihydro-1,2-oxazol-5-yl]methanol (Table 1, entry 3): Prepared via the general procedure in 66% yield (0.360 g) as a solid. 1H NMR: δ 2.7 (bs, 2H), 3.4 (dd, J= 0.9 Hz; 9.4 Hz, 2 H), 3.9 (m, 5H), 6.9 (d, J =9.1 Hz, 2H), 7.5 (d, J = 9 Hz, 2H). IR: 3361, 3004, 2923, 2846, 1608, 1518 cm-1.

[3-(4-methylphenyl)-4,5 -dihydro-1,2-oxazol-5-yl]methanol (Table 1, entry 4): Prepared via the general procedure in 86% yield (0.305 g) as an off white solid. 1H NMR: δ 1.97 (s, 3H), 2.00 (dd, J = 1.4 Hz; 9.3 Hz, 2H), 2.5 (m, 2H), 5.9 (d, J = 7.6, 2H), 6.3 (d, J = 8.04, 2H). IR: 3494, 2936, 2919, 2865, 1606, 1515 cm-1.

[3-(4-isopropylphenyl)-4,5-dihydro-1,2-oxazol-5-yl]methanol (Table 1, entry 5): Prepared via the gener-al procedure in 23% yield (0.056 g) as a yellow oil. 1H NMR: δ 1.3 (d, J = 6.9 Hz, 8H), 3.24 (s, 1H), 3.2 (dd, J = 1.0 Hz; 9.1 Hz), 3.4 (m, 2H), 7.3 (d, J = 13 Hz, 2H), 7.4 (d, J = 13 Hz, 2H).

[3 - (4 -f lurophenyl) - 4 ,5 -dihydro-1,2-oxazol-5 -yl]methanol (Table 1, entry 6): Prepared via the general procedure in 91% yield () as a white solid. 1H NMR: δ 1.7 (m, 1H), 2.0 (m, 1H), 3.5 (d, J = 17 Hz, 2H), 4.0 (m, 2H), 5.0 (m, 1H), 7.5 (m, 2H), 7.8 (m, 2H). IR: 3382, 2929, 2850, 1602, 1513 cm-1.

[3 - (2-f lurophenyl) - 4 ,5 -dihydro-1,2-oxazol-5 -yl]methanol (Table 1, entry 7): Prepared via the gener-al procedure in 66% yield (0.260 g) as a yellow oil. 1H NMR: δ 0.87 (s, 1H), 1.1 (s, 1H), 2.2 (m, 2H), 2.4 (m, 3H), 3.6 (m, 1H), 6.1 (m, 5H), 6.6 (td, J = 2.0 Hz, 6.9 Hz, 1H). IR: 3406, 2925, 1613, 1596, 1500 cm-1.

[3-(2-chlorophenyl)-4,5-dihydro-1,2-oxazol-5-yl]metha-nol (Table 1, entry 8): Prepared via the general procedure in 62% yield (0.230 g) as a yellow oil. 1H NMR: δ 2.0 (m, 2H), 2.2 (m, 2H), 3.5 (dd, J = 1.3 Hz; 7.7 Hz), 3.8 (m, 2H), 7.4 (m, 3H). IR: 3398, 2933, 2871, 1709, 1596, 1560 cm-1.

[3-(2,5-dichlorophenyl)-4,5-dihydro-1,2-oxazol-5-yl]methanol (Table 1, entry 9): Prepared via the general procedure in 69% yield (0.24 g) as a yellow oil. 1H NMR: δ 1.8 (m, 3H), 2.1 (m, 1H), 3.6 (m, 1H), 3.9 (m, 2H), 5.0 (m, 1H), 7.2 (m, 2H). IR: 3400, 2935, 2872, 1709, 1582, 1559 cm-1.

3-butyl-4,5-dihydro-5-isoxazolemethanol (Table 1, entry 10): Prepared via the general procedure on a 2.5 mmol scale in 23% yield (0.11 g) as a yellow oil. 1H NMR: δ IR: 3238, 2958, 2929, 2862, 1495, 1465 cm-1.

3-pentyl-4,5-dihydro-5-isoxazolemethanol (Table 1, en-try 11): Prepared via the general procedure on a 4 mmol scale in 63% yield (0.43 g) as a yellow oil. 1H NMR: δ (note, impurities associated with grease were found in this NMR making the alkyl region difficult to interpret. Further purification was not carried out as the molecule was not viable for the biological testing). 1H NMR: δ 2.1 (m, 3H), 3.76 (d, J = 8.4, 2H), 3.5 (t, J = 3.4, 3H), 4.5 (m, 1H). IR: 3364, 3029, 2920, 1604, 1495 cm-1.

3-hexyl-4,5-dihydro-5-isoxazolemethanol (Table 1, en-try 12): Prepared via the general procedure on a 4 mmol scale in 63% yield (0.45 g) as a yellow oil. 1H NMR: δ 1.7 (m, 13H), 2.2 (m, 1H), 3.0 (m, 1H), 3.8 (m, 2H). IR: 3387, 2900, 2857, 1457, 1434 cm-1.

Biological testing

General: Vibrio fischeri were obtained from Carolina Biosciences and were cultured according to manufacturer instructions. All bacteria work was carried out on the bench top using standard safety protocols for handling biohazard level I organisms. Photobacterium broth and agar plates were made and sterilized using standard protocols. All assays

34JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

were carried out in 96-well format and readings were taken using a BioTek plate reader. All data analysis was performed in Microsoft Excel and GraphPad.

Assaying for quorum sensing inhibition: V. fischeri were cultured in the dark overnight with shaking (25 °C, 200 rpm) in 10 mL culture tubes containing photobacterium broth. The following morning, an initial luminescence reading was taken to ensure growth. If luminescence exceeded the detection maximum of the plate reader, bacteria were further diluted with LBS (up to a 1:10 ratio) to ensure that luminescence would be within the dynamic range of the instrument. Bacteria in LBS (248 µL) were then added to each well. Premade stock solutions of isoxazoline in DMSO were then added to the wells (2.5 µL per well) to reach then desired final concentration (1-0.01 mM). DMSO alone was tested as vehicle control. All experiments were run in quadruplicate.

To ensure that the isoxazolines were soluble in the LBS media, an additional 96-well plate was made containing just LBS (248 µL) and compound in DMSO (2.5 µL per well). These readings served as a background correction factor for the absorbance readings.

Both plates were incubated with shaking (25 °C, 200 rpm) for 3 hours. After this time, the luminescence and absorbance at 600 nm of the plates were read.

The luminescence/background corrected absorbance was determined for each individual well, and then these readings were averaged between replicate wells. The absorbance was factored in to ensure that any decrease in luminescence was due to QS inhibition and not cytotoxicity. The average lum/abs for each concentration was then divided by the lum/abs for the DMSO control to give a final fold value. The final fold values for each concentration were then used to determine the EC50 for each compound.

RESULTS AND DISCUSSION

Ten of the twelve isoxazolines that were synthesized were tested in the biological assay. Of these compounds, seven showed a dose-dependent decrease in luminescence with most EC50 values in the 102-103 µM range (Fig. 2). While the exact mode of action is not yet known, these results indicate that isoxazolines may be a useful scaffold for use in designing QS inhibitors for Gram-negative bacteria species.

In addition to demonstrating that isoxazolines can be used as QS inhibitors, this study also sheds some light on the isoxazoline substitution pattern that is most effective at modulating bacterial QS. Isoxazolines with para- substituted aryl rings tended to give the best inhibition. As can be seen in Table 1 entries 1-5, as the steric bulk at the para-position of the aryl ring increases, the potency of the molecule also generally increases. While the exact cause of this trend is still under investigation, this data indicates that large substituents may be able to better fit into the AHL binding site on LuxR.

A crystal structure of an AHL bound to the LuxR homolog LasR shows that the tail region of the AHL extends into a hydrophobic binding pocket in LasR (Bottomley, M.J. et. al. 2007). It is reasonable to suspect that a similar mechanism may be at play here, where the bulkier substituents are able to fit better into the LuxR binding site.

The exception to this trend is the substituent with para-fluoro substituted aryl ring which showed significant cytotoxicity (Table 1 entry 6). While the exact reasons for this are still being examined, the lack of activity of the ortho-fluorinated compound coupled with the cytotoxicity of the para-fluorinated compound indicate that fluorinated isoxazolines should be avoided in designing QS inhibitors. The cytotoxicity of these specific compounds is still under investigation. With the exception of the compounds shown in Table 1 entries 6 and 8, cytotoxic effects were not observed at the concentrations tested (generally 1-0.01 mM). Concentrations above 1 mM were not used in the QS assay due to the compounds becoming largely insoluble in the

Table 1. Structures, % yields, and EC50 for isoxazolines tested as part of this study. a. EC50 could not be determined for these compounds due to significant cytotoxic effects. b. EC50 could not be determined for these compounds due to a lack of solubility in the aqueous bacterial growth media. The EC50 curves can be found in Figure 2.

35

photobacterium media at these concentrations. It is possible that the remaining compounds are cytotoxic at higher concentrations but this possibility was difficult to assess due to the solubility concerns.

Isoxazolines with C3 aryl rings that had substituents at the ortho and di-ortho positions were also tested (Table 1 entries 7-9). In most cases, modest to no activity was observed with these compounds. More work must be done to determine if this effect is generalizable to all ortho- and di-ortho- substitutions or limited to halogen substituents.

Isoxazolines containing aryl substituents at the C3 position were generally more potent inhibitors than those that contained an alkyl chain at that position. This was largely attributed to a decreased solubility in the aqueous growth media of the alkyl-substituted molecules as increased absorbance at 600 nm was measured for the control wells containing these compounds (Table 1 entries 10-12).

From this initial work, it can be concluded that isoxazolines with aryl rings in the C3 position can serve as effective quorum sensing inhibitors for V. fischeri. Specifically, it seems that isoxazolines with para-substituted aryl rings are the most efficient inhibitors. Further study is needed to determine why certain isoxazolines are more effective inhibitors than others and to determine the exact mode of action being employed by these inhibitors.

ACKNOWLEDGEMENTS

This work was supported by Mercyhurst University.

LITERATURE CITED

Bassler, B.L. 2002. Small talk: cell-to-cell communication in bacteria. Cell. 109:421-424.

Bassler, B.L. and Losick, R.L. 2006. Bacterially speaking. Cell. 125:237-246.

Bottomley, M.J.; Muraglia, E.; Bazzo, R.; Carfi, A. 2007. Molecular insights into quorum sensing in the human pathogen Pseudomonas aeruginosa from the structure of virulence regulator LasR bound to its autoinducer. J. Biol. Chem. 282:13592-13600.

Case, R.J.; Labbate, M.; Kjelleberg, S. 2008. AHL-driven quorum-sensing circuits: their frequency and function among the Proteobacteria. ISME. 2:345-349.

Centers for Disease Control and Prevention. Antibiotic re-sistance threats in the United States, 2013. Web. 10 Apr. 2016.

Clatworthy, A.E.; Pierson, E.; Hung, D.T. 2007. Targeting virulence: a new paradigm for antimicrobial therapy. Nat. Chem. Biol. 3:541-548.

Eberhard, A.; Burlingame, A.L.; Eberhard, C.; Kenyon, G.L.; Nealson, K.H.; Oppenheimer, N.J. 1981. Structural identification of autoinducer of Photobacterium fischeri luciferase. Biochemistry. 20:2444-2449.

Geske, G.D.; Wezeman, R.J.; Siegel, A.P.; Blackwell, H.E. 2005. Small molecule inhibitors of bacterial quo-rum sensing and biofilm formation. J. Am. Chem. Soc. 127:12762-12763.

Geske, G.D.; O’Neil, J.; Blackwell, H.E. 2007. N-Phenylac-etanoyl-L-homoserine lactones can strongly antagonize or superagonize quorum sensing in Vibrio fischeri. ACS. Chem. Biol. 2: 315-320.

Geske, G.D.; O’Neil, J.C.; Miller, D.M.; Mattmann, M.E.; Blackwell, H.E. 2007. Modulation of bacterial quorum sensing with synthetic ligands: systematic evaluation of N-acylated homoserine lactones in multiple species and new insights into their mechanisms of action. J. Am. Chem. Soc. 129:13613-13625.

Koch, B.; Liljefors, T.; Persson, T.; Nielsen, J.; Kjelleberg, S.; Givskov, M. 2006. The LuxR receptor: the sites of in-teraction with quorum-sensing signals and inhibitors. Mi-crobiol. 141:3580-3602.

Figure 2. EC50 curves generated for isoxazolines. Then entry number refers to the entry in Table 1.

36JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

Lupp, C.; Urbenowski, M.; Greenberg, E.P.; Ruby, E.G. 2003. The Vibrio fischeri quorum-sensing systems ain and lux sequentially induce luminescence gene expres-sion and are important for persistence in the squid host. Mol. Microbiol. 50:319-331.

Manefield, M.; Rasmussen, T.B.; Henzter, M.; Andersen, J.B.; Steinberg, P.; Kjelleberg, S.; Givskov, M. 2002. Ha-logenated furanones inhibit quorum sensing through ac-celerated LuxR turnover. Micrbiol. 148:1119-1127.

McInnis, C.E. and Blackwell, H.E. 2011. Thiolactone mod-ulators of quorum sensing revealed through library de-sign and screening. Bioorg. Med. Chem. 19:4820-2828.

Müh, U., Schuster, M., Heim, R., Singh, A., Olson, E. R., Greenberg, E. P. 2006. Novel Pseudomonas aeru-ginosa quorum-sensing inhibitors identified in an ul-tra-high-throughput screen. Antimicrobial Agents and Chemotherapy, 50:3674-3679.

O’Loughlin, C.T.; Miller, L.C.; Siryapom, A.; Drescher, K.; Semmelhack, M.F.; Bassler, B.L. 2013. A quorum-sens-ing inhibitor blocks Pseudomonas aeruginosa virulence and biofilm formation. Proc. Natl. Acad. Sci. 110:17981-17986.

O’Reilly M.C. and Blackwell, H.E. 2016. Structure-based design and biological evaluation of triphenyl scaf-fold-based hybrid compounds as hydrolytically stable modulators of a LuxR-type quorum sensing receptor. ACS Infect. Dis. 1:32-38.

Stacy, D.M.; Welsh, M.A.; Rather, P.N.; Blackwell, H.E. 2012. Attenuation of quorum sensing in the pathogen Acinetobacter baumannii using non-native N-acyl homo-serine lactones. ACS Chem. Biol. 7:1719-1728.

Stevens, A.M.; Queneau, Y.; Soulѐre, L.; von Bodman, S.; Doutheau, A. 2011. Mechanisms and synthetic modu-lators of AHL-dependent gene regulation. Chem. Rev. 111:4-27.

Still, W.C.; Kahn, M.; Mitra, A. 1978. Rapid chromato-graphic technique for preparative separations with mod-erate resolution. J. Org. Chem. 43: 2923-2925.

Journal of the Pennsylvania Academy of Science 90(2): 37-43, 2016

ABSTRACT

Colorblindness is often perceived as a disadvantageous trait because affected individuals distinguish fewer colors in their surroundings. However, the frequency of colorblindness in mammalian populations, especially humans, raises the question as to whether the trait is beneficial in certain environments or situations. Here, we explored the idea that colorblindness aids in the detection of color-camouflaged objects with respect to the visual systems of Lake Malawi African cichlids. Short-wavelength-shifted species Cynotilapia afra and long-wavelength-shifted species Cyrtocara moorii were shown an artificial predatory stimulus hidden against a red camouflage background. We quantified cichlid activity level by counting the number of line crossings in a five-minute period. The short-wavelength-shifted C. afra, which is unable to detect the color red, was better at detecting the camouflaged predatory stimulus than the long-wavelength-shifted C. moorii. When shown the camouflaged predatory stimulus, C. afra exhibited significantly reduced activity as compared to control C. afra trials with no predatory stimulus present, perhaps in an effort to avoid detection. Conversely, C. moorii activity levels remained as high as in trials with no predatory stimulus present, indicating that this species was unable to detect the camouflaged predator. Therefore, this particular aspect of visual systems is likely at work in cichlids, which corresponds with previous primate-based studies exploring the idea that colorblindness is advantageous. Variation in spectral sensitivities may affect Lake Malawi cichlids’ evolutionary fitness and therefore be a driving factor in their rapid evolution and speciation.[ J PA Acad Sci 90(2): 37-43, 2016 ]

THE INFLUENCE OF SHIFTED SPECTRAL SENSITIVITIES ON PREDATOR DETECTION ABILITY OF LAKE MALAWI AFRICAN CICHLIDS1

KELLY M. HANSON2 AND BRIDGETTE E. HAGERTY

York College of Pennsylvania, Department of Biological Sciences, York, PA

INTRODUCTION

The extraordinarily diverse African cichlids of Lake Malawi are the epitome of rapid speciation with roughly 500 species, all of which diverged from a common ancestor within the last one million years (Kocher 2004). Although these cichlids are closely related due to their recent diversification and speciation, the variation observed between the species is striking (Ribbink et al. 1983, Kornfield and Smith 2000). They vary greatly with regards to the lake depth and habitat occupied, prey and diet, territoriality and aggression, coloration and markings, and mating rituals and reproductive cycles (Genner et al. 1999, Albertson 2008).

Beyond the more commonly described differences in behavior and external appearance, Malawi cichlid species also differ from each other with regards to color vision (Carleton and Kocher 2001, Hofmann et al. 2009). Color vision enables an animal to distinguish among light wavelengths as well as different light intensities. The appearance of color vision occurred about 450 million years ago, and now vertebrate visual systems consist of retinas containing both rods and cones, with few exceptions (Bowmaker and Hunt 2006). Rods are photoreceptor cells expressing only one type of visual pigment protein, or opsin, and therefore these rods only detect light intensity. Conversely, cones are photoreceptor cells expressing combinations of several opsins that are activated by different wavelengths of light, and so create the sensation of color (Cronin et al. 2014).

Though one might perceive the ability to distinguish fewer colors as a disadvantageous trait, the frequency of colorblindness in human populations raises the question as to whether the trait may be beneficial in some capacity (Pokorny et al. 1979). The prevalence of congenital color vision deficiencies in European males, approximately 8% in European males on average, demonstrates that these phenotypes are not uncommon (Delpero et al. 2005, Birch 2012). The most common form of color vision deficiency in humans results from one or more non-synonymous point mutations in one of the opsins that are responsible for detecting red and green (Neitz and Neitz 2011). These individuals are known as dichromats, as opposed to the color-normal trichromats. In humans and other primates, dichromats distinguish camouflaged objects from their

1Accepted for publication December 2016.2Current address: University of Rochester Medical Center, Department of Environmental Medicine, Box EHSC, 601 Elmwood Avenue, Rochester, New York 14642.

38JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

surroundings more easily than trichromats. In fact, during World War II, the U.S. Army employed colorblind soldiers to distinguish color-camouflaged enemies more effectively, which many have engendered exploration of this topic (Reit 1978). For example, a study in Costa Rica found that dichromatic white-faced capuchin monkeys capture exposed invertebrates more efficiently and spend less time visually foraging than trichromats of the same species (Melin et al. 2010). Additionally, a team of researchers trained brown capuchins and long-tailed macaques to choose a certain shape out of four choices for a food reward. When the stimuli were camouflaged, those without color-normal vision chose the correct shape at rates significantly above chance, unlike color-normal individuals (Saito et al. 2005). Similar studies showed the same trend in human subjects (Morgan et al. 1992, Saito et al. 2006). There are two hypotheses that have been proposed to explain why colorblind individuals may have an advantage in these situations. First, a dichromat lacking red-green vision is hypothesized to have an advantage over trichromats when an object has the same amounts of red and green as its surroundings, but it is poorly camouflaged with respect to yellows, blues, and greys (Judd 1943). Second, dichromats may forgo neural stimulation from color, instead segregating an object from its surroundings based upon other visual cues, such as incongruous shadows, movements, and patterns in relation to the background (Morgan et al. 1992).

Here, we propose that one factor that drives the evolution of Malawi cichlids is spectral sensitivity diversity. In these populations, differential expression of three of the seven genes that code for color-detecting opsin proteins results in interspecific variations in color vision phenotype (Parry et al. 2005, O’Quin et al. 2012, Cronin et al. 2014). Previous studies in primates suggest that if some cichlid species have lost expression of certain opsins over the course of evolution, thereby losing the ability to discern a certain color, then they may be better able detect color-camouflaged stimuli, such as predators. We chose two Malawi cichlid species for our study, one with spectral sensitivity shifted towards long wavelengths, and one towards short wavelengths. First, our long-wavelength-shifted species was the blue dolphin cichlid, Cyrtocara moorii, which is a haplochromine cichlid occupying open, sandy habitats near the surface of Lake Malawi (Alderton 2003). Second, the cichlid species with short-wavelength-shifted spectral sensitivity in this study was Cynotilapia afra, a shallow rock-dwelling mbuna species in Lake Malawi (Ribbink et al. 1983). Mbuna cichlids of Lake Malawi are known for their bright color patterns, aggression, rocky habitats, and vegetarian diet, whereas haplochromine species like C. moorii usually have less patterning, express less aggression, live in open waters, and are carnivorous (Alderton 2003). The spectral sensitivity of C. moorii is skewed toward red; its long wavelength sensitive (LWS) opsin expression, which translates into seeing red color, accounts for 52.8% of all of its expressed opsins. In contrast, C. afra only exhibits 0.1% LWS opsin

expression due to their visual abilities being shifted toward shorter wavelengths. (Hofmann et al. 2009). These two species were chosen because they exhibited an extreme example of differential opsin expression – in this case, the LWS opsin. A larger difference between species’ expression of a particular opsin would allow better statistical power to detect differences in behavioral parameters.

Since previous studies mentioned above show that dichromatic primates detect camouflaged objects better than trichromatic counterparts, we explored the possibility that this phenomenon extends to non-mammalian vertebrates, such as cichlids. Based on these primate studies and the differences in opsin gene expression between C. afra and C. moorii, we predicted that upon exposure to a red-camouflaged predatory stimulus, C. afra would be more likely to exhibit cautionary behavior than C. moorii, which would be manifested as a change in activity levels. This difference in activity levels would indicate that the red-blind C. afra possesses a heightened ability to break camouflage, possibly due to possessing short-wavelength-shifted spectral sensitivity and therefore lacking LWS opsin expression.

MATERIALS AND METHODS

Experimental setup

Five C. moorii and five Ntekete Bay C. afra were purchased from a cichlid breeder (“Cichlids Are Special” located in York, PA). All individuals were unsexed pre-adolescents aged 3-4 months with sizes ranging from 5.0-6.5 cm. The ages of the fish were determined based on birth records from the breeder. The species were roughly evenly split between two 40 gallon tanks with a rocky habitat at 26.0-27.5 ºC. Alkalinity and pH were kept at 2.8-3.5 dkh and 7.8-8.0, respectively. Before testing commenced, subjects were allowed to mature until 8-9 months of age, the approximate age at which sexual dimorphism appears for both species. All research protocols in this study abide by the U.S. Office of Laboratory Animal Welfare guidelines as well as the American Society of Ichthyologists and Herpetologists standards outlined in The Guidelines for the Use of Fishes in Research.

A 10 gallon trial tank was marked with a grid of squares, each square measuring 10 cm x 10 cm, on the bottom of the tank to record the position of the test subject. The tank was filled to 10 cm depth, equal to the width and length of each of the grid squares to limit the movement of the fish vertically (Fig. 1). Each trial was video recorded with Logitech webcam software. Like the home tank environment, the water was aerated, heated to 27 ºC, and had an approximate pH of 7.8 and an approximate alkalinity of 2.6 dkh. These water specifications and other important water quality aspects were achieved by filling the trial tank with 2/3 home tank

39

water and 1/3 dechlorinated tap water. Small rock substrate like that of the home tank covered the bottom of the tank in such a way that the grid lines were still visible. A 32 watt white fluorescent light bulb (Phillips 800 Series #F32T8/TL835) illuminated the trial tank from above and was the only light source in the room during trials.

Three of the four faces of the tank were obscured with black cardboard. The fourth face was variable. It either presented a custom camouflage pattern composed entirely of shades of red (color camouflage: CC) or blank white glossy printer paper (no color camouflage: NCC). During the trials, these two backgrounds would serve as surroundings for the predatory stimulus (PS), or a control with no predatory stimulus (NPS). The red camouflage pattern was generated in MS Paint v6.1 using the following hexadecimal shades of red (and relative ratios): #AD0304 (18.8%); #960202 (18.2%); #C70707 (15.2%); #7E0303 (14.4%); #F11E1E (11.2%); #610101 (6.4%); #ED4949 (6.3%); #D82828 (5.2%);

and #BA2A29 (4.3%). Relative ratios were determined using Image Color Summarizer v0.7. This pattern was printed onto poster paper using a HP DesignJet Z6200 Photo Production large format printer.

Following guidelines from Rowland (1979) for making realistic fish dummies, the predatory stimulus was constructed from terra cotta clay to create a generalized fish silhouette: body, mouth, dorsal fins, ventral fins, and tail. This model was then covered on both faces with the same red color camouflage poster paper as the CC trial tank background (described above). Small areas on the model that could not be covered with this pattern were covered with red acrylic paint. The model, or predatory stimulus, measured 15 cm in length and was pinned to the CC or NCC backgrounds using dissection pins covered in red lab tape.

Measuring effects of color camouflage

There were four rounds of trials in this study, resulting from combinations of color camouflage (CC) or no color camouflage (NCC) and predatory stimulus (PS) or no predatory stimulus (NPS). The true control trials lacked both the predator and the camouflage (NCC-NPS). Two other rounds of partial control trials lacked one or the other (NCC-PS and CC-NPS). The experimental round of trials showed the subjects the predatory stimulus disguised against the color camouflage background (CC-PS) (Fig. 2). Because subjects were not tagged due to species-related constraints, the order of trials that each individual experienced could not be randomized.

During a single trial, the trial tank was first prepared as described above. At first, a piece of black cardboard obscured the variable stimuli from the subject. A subject was randomly captured from the home tank and placed in a 1 L translucent plastic beaker with home tank water. The beaker was then placed upright in the tank until the beaker water temperature precisely matched the tank water temperature. Then the subject was released into the tank by setting the beaker gently on its side. The beaker was kept in the tank during the trial for better acclimation, as it provided shelter.

After release into the trial tank, the subject was allowed to acclimate to the new surroundings for 45 min with the white light on. Noise in the lab was kept at a minimum. This acclimation time length was determined in preliminary trials based upon how long it took an undisturbed individual to accept food offered (data not shown). Preliminary trials had also shown that lights should be kept on during the acclimation period to promote exploration of the new environment. After the 45 min acclimation period, the lights were turned off, the stimulus was revealed by removing the visual barrier, and video recording commenced. The subject was kept in darkness for 5 min to minimize capturing reactions from the sounds of removing the visual barrier. After this small acclimation period, the lights were turned

Figure 1. Aerial (A) and lateral (B) views of the aquarium setup for each trial. A webcam positioned above the tank recorded the subjects’ location with regards to the gridlines seen in A.

40JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

on again and the activity of the subject was recorded for 5 min. After returning the subject to the home tank, the trial tank was emptied and scrubbed with warm water to prevent secreted stress hormones from interfering with the response of the next subject. All trials were conducted during the same approximate time of day (late afternoon) to account for any potential differences due to light cycle interference.

A commonly accepted method of assessing predator detection in fish is to quantify subjects’ activity levels (Ferrari et al. 2007, Milano et al. 2010, O’Connor et al. 2015). For each video recording of the 40 total trials (5 individuals per species, 2 species, 4 types of trials), we observed the number of times a subject crossed from one grid square into another by using a mechanical counter. A line crossing was recorded if at least 75% of the subject’s body (up to the posterior edge of its dorsal fin) moved into another square. The posterior end of the subject’s body alone moving into another square was not counted as a line crossing, because most instances of this were due to water current rather than deliberate movement.

Statistical analysis

Mean activity level was compared among the treatments using GraphPad Prism v6.0 software (α = 0.05). For each group (n = 5), no outliers were identified and variances were not significantly different. Since the two-way analysis of variance (ANOVA) is robust to the assumption of normality given that no outliers are present, a two-way ANOVA for

fixed factors (Model 1) was conducted. A Tukey’s multiple comparisons test followed to compare mean activity levels between species and treatment types.

RESULTS

During the control trials (NCC-NPS), mean activity levels were similar between C. afra (95% CI = 42.7-145.3) and C. moorii (95% CI = 66.9-154.7) (Fig. 3). When the subjects were shown the predator without camouflage (NCC-PS), there was no difference detected between C. afra (95% CI = 28.2-89.4) and C. moorii (95% CI = 35.5-87.3) with the Tukey’s multiple comparisons test (Table 1). When the subjects were shown only the red camouflage with no predatory stimulus (CC-NPS), C. afra (95% CI = 78.3-100.5) and C. moorii (95% CI = 67.4-128.2) maintained activity levels that were similar to the control trial activity levels, which were not significantly different based on a Tukey’s multiple comparisons test.

When the predatory stimulus was hidden within the red camouflage (CC-PS), the species showed different reactions, supporting a significant interaction between species and stimulus (F(3,31) = 3.887; P = 0.0181). The red-blind C. afra exhibited a much lower activity level that was similar to the C. afra NCC-PS trials, even though the predatory stimulus was camouflaged (95% CI = 33.2-74.8). Conversely, when shown the camouflaged predatory stimulus (CC-PS),

Figure 2. Variable stimuli shown to all subjects during the four types of trials with the respective abbreviations used throughout this study. The white or red camouflage backgrounds were on sheets of cardboard, and the predatory stimulus was constructed of terra cotta clay, covered with identical red camouflage pattern and held onto the cardboard background with dissection pins covered in red lab tape.

Figure 3. Activity level of subjects during each type of trial. Activity level was defined as the number of gridline crossings of subjects during a 5 min trial. Median is indicated as the centerline and whiskers indicate the minimum and maximum values. The subjects were either shown a white background (NCC-NPS), a predatory stimulus (NCC-PS), a red camouflage pattern (CC-NPS), or the predatory stimulus hidden within the red camouflage pattern (CC-PS). Illustrations above the data for each type of trial show which stimulus was displayed to the subjects within that treatment group. ** indicates a significant difference between species within a treatment based on Tukey’s multiple comparisons test (p < 0.01), while ns indicates no difference.

41

C. moorii had significantly higher activity levels than C. afra (95% CI = 78.0-183.6), which were comparable to C. moorii control trials (NCC-NPS).

DISCUSSION

Our objective was to examine whether shifted spectral sensitivities in two species of Lake Malawi cichlids might affect their ability to detect a camouflaged predatory stimulus. We hypothesized that low LWS opsin expression of C. afra (0.1%) would cause this species to see the camouflaged predatory stimulus better than the C. moorii subjects, which have very high (52.8%) LWS opsin expression (Hofmann et al. 2009). We also expected that the type of stimulus displayed would affect the activity levels of the species. Finally, we predicted that similar activity levels would be observed between species for the white background (NCC-NPS), predator only (NCC-PS), and camouflage only (CC-NPS) trials, while activity levels would differ between species during the camouflaged predator (CC-PS) trials.

We observed activity levels during our experiment that supported our hypotheses. The difference in mean activity levels between C. afra and C. moorii exposed to the camouflaged predator was the most revealing. The low activity levels of the red-blind C. afra in this trial suggested that they were better able to see the camouflaged predator than C. moorii. Because the mean activity was comparable to the trials with the non-camouflaged predator, C. afra appeared able to see the predator as if it were not camouflaged. Similarly, C. moorii retained higher activity levels, suggesting that they were unable to detect the camouflaged predator.

Both species had lower activity when shown the non-camouflaged predator as compared to the control and camouflage-only trials. Although neither species’ activity levels between the control and predator-only trials were significantly different, we may not have had the statistical power, with a small sample size per species, to detect that difference. Additionally, we were unable to track each fish as an individual during each trial due to species-related constraints. Though the two-way ANOVA is conservative

and fairly robust in correcting for the lack of repeated measures, being able to identify individuals across trials would have increased the precision of this study (Schmider et al. 2010).

Previous studies on this topic have used primate subjects, with the camouflaged stimulus being either an inanimate pattern that the subjects were trained to identify or a scurrying insect for which the subjects were scavenging (Morgan et al. 1992, Saito et al. 2005, Saito et al. 2006, Melin et al. 2010). Similar to our study, these experiments supported the hypothesis that colorblindness aids in the detection of camouflaged objects. However, our study is the first to show that this particular aspect of visual systems is present in cichlids as well as the previously studied humans and various non-human primates. Since this phenomenon has now been identified in multiple taxa, the colorblind advantage may not be limited to primates and instead may be present in many more diverse species’ color vision systems.

By using an immobile stimulus that was predatory in nature, we evoked a natural stress response, in contrast to the subjects that were trained to identify certain stimuli over others (humans, Morgan et al. 1992 and Saito et al. 2006; nonhuman primates, Saito et al. 2005). Evoking an instinctive response further supports the possibility that these differences between species may reflect true differences in real world populations. Similarly, Melin and colleagues (2010) observed insect foraging behavior of white-faced capuchin monkeys without creating an unnatural environment. However, motion can aid in object detection, so this factor may have influenced the ability of Melin’s capuchin monkeys to see scurrying insects (Schaerer and Neumeyer 1996, Land and Nilsson 2012). We deliberately avoided using motion of the predatory stimulus, but found that the non-camouflaged stimulus still caused a stress response in both species, in the form of diminished activity levels.

The interspecific differences in color vision ability within cichlid populations exist due to factors such as habitat depth, diet, sexual selection, and water turbidity (Crescitelli et al. 1985, Van der Meer et al. 1995, Gerl and Morris 2008). For example, although generally cichlid species in Lake Malawi have short-wavelength-shifted spectral sensitivities like

Comparison(C. afra vs. C. moorii) Mean difference 95% CI of difference P-value

NCC-NPS -16.8 -76.55 to 42.95 >0.05NCC-PS -2.6 -58.94 to 53.74 >0.05CC-NPS -8.4 -64.74 to 47.94 >0.05CC-PS -76.8 -133.1 to -20.46 <0.01

Table 1. Abbreviated output from a Tukey’s multiple comparisons test comparing mean activity levels between cichlid species for all stimulus types (α = 0.05).

42JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

C. afra, C. moorii may be able to detect longer wavelengths like red because it inhabits sandy-bottom waters. Turbidity tends to affect sandy habitats more so than rocky habitats, altering the photic environment and hence affecting the color vision of species evolving there (Cronin et al. 2014). Another selective pressure that may have influenced the shift of C. moorii spectral sensitivities away from short wavelengths is that of camouflaged predator detection. The relatively low expression of blue-detecting S2A opsin (6.9% of total opsin expression) by C. moorii may enable this species to better detect blue-camouflaged predators, such as Champsochromis caeruleus, a considerably large predatory cichlid whose habitat overlaps with that of C. moorii (Kazembe and Makocho 2006, Hofmann et al. 2009).

Interestingly, different color vision sensitivities may not be simply a product of evolution; but also a potential cause of evolution. Color plays a very important role in sexual selection, and differences in visual sensitivities could greatly influence cichlid speciation (Parry et al. 2005, Gerl and Morris 2008). However, visual differences likely affect more than mate choice. We propose that different spectral sensitivities also influence a species’ evolutionary fitness in situations where detecting camouflaged objects is advantageous, such as predator evasion, foraging, or hunting. The advantage of certain visual sensitivities with regards to predator detection, as evidenced by this study, is one of the many factors contributing to the complex system of cichlid evolution, both past and present.

ACKNOWLEDGMENTS

This research was supported by the Department of Biological Sciences at York College of Pennsylvania as part of the undergraduate student research program.

The authors would like to thank B. Rehnberg, C.R. Tracy, R.P. Phipps, and two anonymous reviewers for reviewing previous drafts of this manuscript.

LITERATURE CITED

Albertson, R. 2008. Morphological divergence predicts habitat partitioning in a Lake Malawi cichlid species complex. Copeia. 3:689-698.

Alderton, D. 2003. Cichlids: Understanding Angelfish, Os-cars, Discus, and Others. BowTie Press. Irvine, CA, 7-46.

Birch, J. 2012. Worldwide prevalence of red-green color de-ficiency. J. Opt. Soc. Am. A. 29(3):313-320.

Bowmaker, J.K. and D.M. Hunt. 2006. Evolution of verte-brate visual pigments. Curr. Biol. 16(13): R484-R489.

Carleton, K. and T. Kocher. 2001. Cone opsin genes of Afri-can cichlid fishes: Tuning spectral sensitivity by differen-tial gene expression. Mol. Biol. Evo. 18(8):1540-1550.

Crescitelli, F., M. McFall-Ngal and J. Horwitch. 1985. The visual pigment sensitivity hypothesis: Further evidence from fishes of varying habitats. J. Comp. Physiol. B. Bio-chem. Syst. Environ. Physiol. 157(1): 323-333.

Cronin, T.W., S. Johnsen, N.J. Marshall and E.J. Warrant. 2014. Visual Ecology. Princeton University Press. Princ-eton, NJ, 27-43.

Delpero, W.T., H. O’Neill, E. Casson and J. Hovis. 2005. Aviation-relevant epidemiology of color vision deficiency. Aviat. Space Envir. Md. 76(2):127-133.

Ferrari, M.C.O., A. Gonzalo, F. Messier and D.P. Chivers. 2007. Generalization of learned predator recognition: An experimental test and framework for future studies. P. Roy. Soc. Lond. B Bio. 274(1620):1853-1859.

Genner, M.J., G.F. Turner and S.J. Hawkins. 1999. Foraging of rocky habitat cichlid fishes in Lake Malawi: Coexis-tence through niche partitioning? Oecologia. 121:283-292.

Gerl, E.J. and M.R. Morris. 2008. The causes and conse-quences of color vision. Evo. Edu. Outreach 1:476-486.

Hofmann, C.M., K.E. O’Quin, N.J. Marshall, T.W. Cronin, O. Seehausen and K.L. Carleton. 2009. The eyes have it: Regulatory and structural changes both underlie cichlid visual pigment diversity. PLoS Biol. 7(12): e1000266.

Judd, D.B. 1943. Colorblindness and the detection of cam-ouflage. Science 97(2529): 544-546.

Kazembe, J., and P. Makocho. 2006. Champsochromis caeruleus. The IUCN Red List of Threatened Species 2006. eT60928A12424792.

Kocher, T. 2004. Adaptive evolution and explosive specia-tion: The cichlid fish model. Nat. Rev. Genet. 5(4):288-298.

Kornfield, I. and P. Smith. 2000. African cichlid fishes: Model systems for evolutionary biology. Annu. Rev. Ecol. Evol. Syst. 31(1):163-196.

Land, M.F., and D. Nilsson. 2012. Animal Eyes. 2nd ed. Oxford University Press. Oxford, NY, 231-232.

Melin, A.D., L.M. Fedigan, H.C. Young and S. Kawamura. 2010. Can color vision variation explain sex differenc-es in invertebrate foraging by capuchin monkeys? Curr. Zool. 56(3):300-312.

Milano, D., M. Lozada and H. Zagarese. 2010. Predator-in-duced reaction patterns of landlocked Galaxias macula-tus to visual and chemical cues. Aquat. Ecol. 44(4):741-748.

Morgan, M.J., A. Adam and J.D. Mollon. 1992. Dichromats detect color-camouflaged objects that are not detected by trichromats. Proc. R. Soc. B. 248(1323):291-295.

43

Neitz, J. and M. Neitz. 2011. The genetics of normal and de-fective color vision. Vision Res. 51(7):633-651.

O’Connor, C.M., A.R. Reddon, A. Odetunde, S. Jindal and S. Balshine. 2015. Social cichlid fish change behavior in response to a visual predator stimulus, but not the odour of damaged conspecifics. Behav. Process. 121:21-29.

O’Quin, K.E., J.E. Schulte, Z. Patel, N. Kahn, Z. Naseer, H. Wang, M.A. Conte and K.L. Carleton. 2012. Evolution of cichlid vision via trans-regulatory divergence. Evo. Biol. 12(1):251-261.

Parry, J.W.L., K.L. Carleton, T. Spady, A. Carboo, D.M. Bowmaker and J.K. Hunt. 2005. Mix and match color vision: Tuning spectral sensitivity by differential op-sin gene expression in Lake Malawi cichlids. Curr. Biol. 15(1):1734-1739.

Pokorny, J., V.C. Smith, G. Verriest and A.J.L.G. Pinckers. 1979. Congenital and Acquired Color Vision Defects: Current Ophthalmology Monographs. Grune & Stratton, Inc. New York, NY, 1 + 432 pp.

Reit, S. 1978. Masquerade: The Amazing Camouflage De-ceptions of World War II. Hawthorn Books. Portland, OR, 1 + 255 pp.

Ribbink, A.J., B. Marsh, A. Marsh, A.C. Ribbink and B.J. Sharp. 1983. A preliminary survey of the cichlid fishes of rocky habitats in Lake Malawi. S. Afr. J. Zool. 18(3):149-310.

Rowland, W. 1979. Some methods of making realistic fish dummies for ethological research. Behav. Res. Meth. In-str. 11(6):564-566.

Saito, A., A. Mikami, S. Kawamura, Y. Ueno, C. Hiramat-su, K.A. Widayati, B. Suryobroto, M. Teramoto, Y. Mori, K. Nagano, K. Fujita, H. Kuroshiman and T. Hasegawa. 2005. Advantage of dichromats over trichromats in dis-crimination of color-camouflaged stimuli in nonhuman primates. Am. J. Primatol. 67(1):425-436.

Saito, A., A. Mikami, T. Hosokawa and T. Hasegawa. 2006. Advantage of dichromats over trichromats in discrimi-nation of color-camouflaged stimuli in humans. Percept. Mot. Skills 102(1):3-12.

Schaerer, S. and C. Neumeyer. 1996. Motion detection in goldfish investigated with the optomotor response is “col-or blind.” Vision Res. 36(24):4025-4034.

Schmider, E., M. Ziegler, E. Danay, L. Beyer and M. Büh-ner. 2010. Is it really robust? Reinvestigating the robust-ness of ANOVA against violations of the normal distri-bution assumption. Eur. J. Res. Meth. Behav. Soc. Sci. 6(4):147-151.

Van der Meer, H.J., G.C. Anker and C.D.N. Barel. 1995. Ecomorphology of retinal structures in zooplantivorous haplochromine cichlids (Pisces) from Lake Victoria. En-viron. Biol. Fishes 44(1):115-132.

Journal of the Pennsylvania Academy of Science 90(2): 44-47, 2016

DETECTION OF BORRELIA BURGDORFERI AND BORRELIA MIYAMOTOI IN WHITE-FOOTED MICE, PEROMYSCUS LEUCOPUS, AND BLACKLEGGED TICKS, IXODES SCAPULARIS, IN

MONROE COUNTY, PENNSYLVANIA, A LYME ENDEMIC REGION1

MEAGHAN BIRD AND JANE E. HUFFMAN2

Northeast Wildlife DNA Laboratory, Department of Biological Sciences, East Stroudsburg University, East Stroudsburg, PA

ABSTRACT

The causative agents of Lyme disease (Borrelia burgdorferi) and tick-borne relapsing fever (Borrelia miyamotoi) are both vectored by the blacklegged tick (Ixodes scapularis) and maintained in the rodent reservoir, the white-footed mouse (Peromyscus leucopus). The prevalence of B. burgdorferi and B. miyamotoi in P. leucopus and larval I. scapularis was investigated in Monroe County, Pennsylvania. Borrelia burgdorferi and B. miyamotoi DNA was detected by polymerase chain reaction (PCR) and compared seasonally. Of 60 mice captured throughout 2013, 17 (28.3%) were found to be infected with B. burgdorferi only, 2 (3.3%) with B. miyamotoi only, and 2 (3.3%) were co-infected. Two of twelve (16.7%) larval I. scapularis pools were found to be infected with B. miyamotoi. This study is the first report of B. miyamotoi infection in white-footed mice and larval blacklegged ticks in Pennsylvania.[ J PA Acad Sci 90(2): 44-47, 2016 ]

INTRODUCTION

Vector-borne pathogens are a major source of emerging diseases. Lyme disease, caused by the bacteria Borrelia burgdorferi, was discovered in 1977 in Lyme, Connecticut and is the most common vector-borne disease in the United States (Barbour and Fish, 1993;Steere et al. 2004; de la Fuente et al., 2008; Hamer et al., 2010). Up to 30,000 cases of Lyme disease are reported annually to Centers for Disease Control, (CDC, 2015) with many more that go unreported. Lyme disease is endemic in the Northeast and upper Midwest of the United States and in 2013 the number of confirmed cases of Lyme disease in Pennsylvania was 4981 (CDC, 2015).

This is also an area of emerging tick-borne pathogens, such as B. miyamotoi, along with increasing abundance of Ixodes scapularis (Scoles et al. 2001; Diuk-Wasser et al. 2014).

Borrelia miyamotoi is a spirochete, genetically related to B. burgdorferi (Scoles et al. 2001). The first confirmed human case of B. miyamotoi infection in North America occurred in Hunterdon County, NJ in 2013 (Gugliotta et al., 2013). Since its discovery (Platonov et al. 2011) as the causative agent of tick-borne relapsing fever, Krause et al. (2013) reported a significant prevalence of exposure among people in Lyme disease endemic areas of the United States. Krause et al. (2015) reported on B. miyamotoi in nature and humans. Continued research has indicated that the white-footed mouse, Peromyscus leucopus, acts as reservoir host and the blacklegged tick Ixodes scapularis, as vector for both B. burgdorferi and B. miyamotoi (Barbour et al., 2009). Notably, unlike B. burgdorferi, B. miyamotoi can be transmitted transovarially from female ticks to progeny, leading to populations of infected larval ticks (Scoles et al., 2001; Barbour et al., 2009).

Borrelia burgdorferi poses a serious risk to public health and the number of cases in the US has been increasing in recent years ( Platonov et al., 2011; CDC, 2015). The emerging pathogen B. miyamotoi is also becoming a public health threat, although there is much yet to be learned about its natural history in regions of high tick density and tick-borne disease endemicity. The present study reports on the occurrence of B. burgdorferi and B. miyamotoi in Monroe County, Pennsylvania in the white-footed mouse and black-legged tick.

MATERIALS AND METHODS

White-footed mice were trapped in various locations of Monroe County, Pennsylvania using Victor (Lititz, PA) brand snap traps from January to December 2013. At necropsy, specimens were weighed and sexed, and the spleens were removed and stored in 1.5 ml microcentrifuge tubes at -20 °C. Trapping and necropsy protocols were approved by the East Stroudsburg University IACUC committee. Larval ticks were collected by the dragging method with a 2X3 foot

1Accepted for publication August 2016.2Corresponding Author: jhuffman@esu.edu

45

white drag cloth, from a wood lot in Monroe County, PA. The geographic locations of the tick collection was in the area where mice were trapped.

DNA was extracted from splenic tissue using the MoBio Tissue and Cells DNA Extraction kit and protocol (MoBio Laboratories, Carlsbad, CA). Samples were incubated for 2 hours at 65 °C. Larval ticks were pooled into groups of 20 and, extracted using Qiagen DNeasy Blood and Tissue kit with an overnight incubation and 20 µl Proteinase K. Both larval tick and mouse tissue DNA extractions were quantified using an Eppendorf Biophotometer and frozen at -20 °C.

Extracted genomic DNA was amplified using a 15 µl nested PCR reaction. The primer sets used were described by Bunikis et al. (2004) and are Borrelia genus specific, targeting the intergenic spacer region of the Borrelia genome, which is a non-coding region used in many phylogenetic studies. Outer nested primers (IGS1), at a 10 µM concentration consist of forward- 5′-GTATGTTTAGTGAGGGGGGTG-3′ and reverse- 5′-GGATCATAGCTCAGGTGGTTAG-3′. First nested reaction conditions are comprised of 35 cycles of 94 °C denaturation for 30 seconds, 56 °C annealing for 30 seconds, and a 72 °C extension for 1 minute. The inner nested primers (IGS2), at a 10 µM concentration, consist of forward- 5′-AGGGGGGTGAAGTCGTAACAAG-3′ and reverse- 5′-GTCTGATAAACCTGAGGTCGGA-3′. Second nested reaction conditions are 40 cycles of 94 °C denaturation for 30 seconds, 60 °C annealing for 30 seconds, and 72 °C extension for one minute. Both reactions were preceded by a 5-minute burn-in at 94 °C and a final extension of 5 minutes at 72 °C before being held at 4 °C. Primer sets were purchased from Applied Biosystems Custom Oligo Synthesis Service. The IGS primers are able to differentiate between these two species of Borrelia by the size of the DNA fragment. The band produced by the PCR product of B. burgdorferi DNA is approximately 988 base pairs and B. miyamotoi is about 547 (Scott et al., 2010). Borrelia andersoni, another species found in the Northeastern United States, can also be amplified by these IGS primers and will produce a band of about 1000 base pairs (Scott et al. 2010).

Following PCR, DNA was visualized using gel electrophoresis. Amplified DNA was dyed with Qiagen CoralLoad dye and separated in a 1% TAE agarose gel stained with ethidium bromide (0.5 µl/ml). Gels were visualized under UV light on a Bio-Rad Mini Transilluminator. All Borrelia spp. positive samples were sequenced using an ABI3130 Genetic Analyzer (Applied Biosystems, Valencia, CA) and sent to Cornell University Institute of Biotechnology for sequence confirmation. PCR products were purified using the Exo SAP-It enzymatic PCR cleanup reagent and protocol (Affymetrix, Santa Clara, CA). Further PCR product purification took place through Big-dye Terminator reaction and dye clean-up, using LifeTechnologies’ reagents and protocol. Cycle sequencing with the ABI3130 Genetic analyzer also utilizes the reagents and protocol from LifeTechnologies. Larval ticks were sequenced for species

confirmation following amplification with arthropod specific primers LCO-HCO (Vrijenhoek, 1994). All sequences were analyzed through GenBank’s Basic Local Alignment Search Tool (BLAST). A Kruskal-Wallis test was used to analyze seasonal Borrelia infection rates in white-footed mice.

RESULTS

A total of 60 white-footed mice were collected from Pennsylvania, 22 in winter, 5 in spring, 19 in summer, and 14 in fall. Borrelia sp. was isolated from 21 (35%) of all the mice tested. Seventeen (28.3%) were infected with only B. burgdorferi and 2 (3.3%) with only B. miyamotoi. Two (3.3%) specimens were co-infected with both species of Borrelia. Five (22.7%) B. burgdorferi positive samples were found during winter trapping, 1 (20%) in spring, 6 (31.6%) summer, and 5 (35.7%) in the fall. No specimens from winter, spring, or fall collection were positive for B. miyamotoi only and 2 (10.5%) were positive from the summer. One (5.3%) co-infected specimen was collected during the summer and 1 (7.1%) during the fall. Infection totals for each season can be found in Figure 1. There was no significant difference between seasons for infection with B. burgdorferi, B. miyamotoi, or co-infected samples using the Kruskal-Wallis test.

Twelve (57.1%) of the 21 positive samples were successfully sequenced. Both specimens positive for only B. miyamotoi, 9 positive for only B. burgdorferi and one co-infected specimen for B. burgdorferi and B. miyamotoi.

Fifty-seven larval ticks were collected and divided into 12 pools, depending on date of collection. Larval pools were identified through genetic sequencing as I. scapularis. Two (16.7%) pools tested positive for B. miyamotoi. All pools were negative for B. burgdorferi.

Figure 1. Total number of white-footed mice infected with Borrelia burgdorferi, B. miyamotoi, both, or neither, collected seasonally in 2013.

46JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

DISCUSSION

The results finding B. burgdorferi infection in P. leucopus in this study are consistent with other studies conducted in similar areas of Pennsylvania (Devlin, 2009; James, 2009). In the current study 14.3% of larval tick pools were positive for B. miyamotoi, but none were positive for B. burgdorferi supporting previous research which reports transovarial transmission of this pathogen but not of B. burgdorferi (Scoles et al., 2001;Barbour et al., 2009). A novel aspect of this study is the detection of B. miyamotoi in both white-footed mice and larval ticks in Monroe County, PA. Low prevalence of B. miyamotoi infection in ticks was reported in nymphal I. scapularis from the nearby Lehigh Valley region of Pennsylvania (Edwards et al., 2015). However, the present study documents the first time B. miyamotoi infection has been detected in both larval ticks and white-footed mice within the state and is the first study of its kind specifically in Monroe County.

Co-infection with B. burgdorferi and B. miyamotoi has been documented in white-footed mice (Barbour et al., 2009). The current study reported co-infection but confirmation was difficult. The two white-footed mouse samples determined to be co-infected in this study tested positive for both pathogens during successive PCR reactions and there was no instance of concurrent amplification in the same reaction. Serial dilution testing of B. burgdorferi and B. miyamotoi positive controls revealed no significant competitive inhibition between species within the same reaction. It is possible that this finding is misleading due to the purity and concentration of laboratory produced positive controls and the likelihood that Borrelia infection in nature would occur at different concentrations. Barbour et al. (2009) found that at extreme ratios of co-infection, the Borrelia of a lower concentration was difficult or impossible to detect. In this and the study by Scott et al. (2010) examining B. miyamotoi infection, in wild turkeys, reported that co-infections may be more prevalent than what is shown by PCR analysis due to a highly concentrated species inhibiting a lesser concentrated species. Therefore, the two co-infected white footed mice documented in this study may be a conservative result.

There was no significant correlation between Borrelia infection in P. leucopus and season; however this could have been affected by the differences in sample size between seasons. The prevalence of B. miyamotoi in P. leucopus showed a trend in significance as, 3 of the 4 samples that tested positive for B. miyamotoi were collected in the summer indicating a potential correlation. A possible explanation for this trend involves the occurrence of transovarial transmission of B. miyamotoi in I. scapularis. Larval I. scapularis often feed on rodents, primarily the white-footed mouse, and are most abundant during summer months (Ostfeld et al., 2006). An increased density of larval ticks, potentially infected with B. miyamotoi through transovarial transmission may elicit a higher instance of

infection in the reservoir host. Further study is necessary to confirm this, including acquiring a larger sample size, examining incubation period of the bacteria, and the ability of the mouse immune response to eliminate the infection.

All tick and mouse positive samples were subjected to sequencing, yet not all samples were positively sequenced. The final prevalence for mice was 23.7%, which was based on PCR and gel electrophoresis results, while only slightly more than half of these (57%) were successfully confirmed by sequence. A possible reason for this discrepancy is the use of splenic tissue to determine infection with Borrelia spp. Studies have shown that splenic tissue is ineffective in performing disease diagnostics because it contains large quantities of the pathogen DNA (James, 2009). DNA in the extracted product is often too concentrated, which can hinder DNA amplification and sequencing. In future studies, DNA extraction from additional organs should be attempted and may produce a product that is more easily analyzed.

This study reports on the prevalence of 2 important tick borne pathogens in an area of high tick density and Lyme disease endemicity. It expands on previous knowledge of B. burgdorferi prevalence and reports on the previously unknown distribution of B. miyamotoi in Pennsylvania white-footed mice.

ACKNOWLEDGMENTS

We thank Thomas Rounsville, Krista Smith, and Katie Donahue for their assistance in the field and laboratory. This work was funded by the Northeast Wildlife DNA Lab of East Stroudsburg University and grants from the Pennsylvania Academy of Sciences and Commonwealth of Pennsylvania University Biologists. Positive controls and technical advice were graciously provided by Dr. Sarah Hamer (Texas A&M), Dr. Samuel Telford (Tufts University), and Dr. Marten Edwards (Muhlenberg College).

LITERATURE CITED

Barbour, A.G., Bunikis, J., Travinsky, B., Hoen, A.G., Di-uk-Wasser, M.A., Fish, D., Tsao, J.I., 2009. Niche parti-tioning of Borrelia burgdorferi and Borrelia miyamotoi in the same tick vector and mammalian reservoir species. Am. J. Trop. Med. Hyg. 81, 1120–1131. doi:10.4269/ajt-mh.2009.09-0208

Barbour, A.G., Fish, D., 1993. The biological and social phenomenon of Lyme disease. Science 260, 1610–1616.

BLAST 2016. blast.ncbi.nlm.nih.gov

47

Bunikis, J., Garpmo, U., Tsao, J., Berglund, J., Fish, D., Barbour, A.G., 2004. Sequence typing reveals exten-sive strain diversity of the Lyme borreliosis agents Bor-relia burgdorferi in North America and Borrelia afzelii in Europe. Microbiology 150, 1741–1755. doi:10.1099/mic.0.26944-0

CDC. 2015. Reported cases of Lyme disease by state or lo-cality, 2004-2013†. CDC [WWW Document], 2015. URL http://www.cdc.gov/lyme/stats/chartstables/reportedcas-es_statelocality.html (accessed 3.10.15).

de la Fuente, J., Estrada-Pena, A., Venzal, J.M., Kocan, K.M., Sonenshine, D.E., 2008. Overview: Ticks as vec-tors of pathogens that cause disease in humans and ani-mals. Front. Biosci. J. Virtual Libr. 13, 6938–6946.

Devlin, L.J., 2009. Prevalence of Borrelia burgdorferi sen-su lato and Anaplasma phagocytophium in Peromyscus leucopus in a region of endemicity in Monroe County, PA. Master’s Thesis East Stroudsburg University.

Diuk-Wasser, M.A., Liu, Y., Steeves, T.K., Folsom-O’Keefe, C., Dardick, K.R., Lepore, T., Bent, S.J., Usmani-Brown, S., Telford, S.R., Fish, D., Krause, P.J., 2014. Monitoring human babesiosis emergence through vector surveillance New England, USA. Emerg. Infect. Dis. 20, 225–231. doi:10.3201/eid2003.130644

Edwards, M.J., Barbalato, L.A., Makkapati, A., Pham, K.D., Bugbee, L.M., 2015. Relatively low prevalence of Babesia microti and Anaplasma phagocytophilum in Ix-odes scapularis ticks collected in the Lehigh Valley re-gion of eastern Pennsylvania. Ticks Tick-Borne Dis. 6, 812–819. doi:10.1016/j.ttbdis.2015.07.009

Gugliotta, J.L., Goethert, H.K., Berardi, V.P., Telford, S.R. 2013. Meningoencephalitis from Borrelia miyamotoi in an immunocompromised patient. N. Engl. J. Med. 368, 240–245. doi:10.1056/NEJMoa1209039

Hamer, S.A., Tsao, J.I., Walker, E.D., Hickling, G.J., 2010. Invasion of the Lyme Disease vector Ixodes scapularis: implications for Borrelia burgdorferi endemicity. Eco-Health 7, 47–63. doi:10.1007/s10393-010-0287-0

James, T.N., 2009. The prevalence of Anaplasma phago-cytophilum and Borrelia burgdorferi from three biopsy locations in Peromyscus leucopus. Master's Thesis East Stroudsburg University.

Krause, P.J., Narasimhan, S., Wormser, G.P., Rollend, L., Fikrig, E., Lepore, T., Barbour, A., Fish, D., 2013. Human Borrelia miyamotoi Infection in the United States. N. Engl. J. Med. 368, 291–293. doi:10.1056/NEJMc1215469

Krause, P.J., Fish, D, Narasimbhan, S, Barbour, A. 2015. Borrelia miyamotoi infection in nature and in humans. Clin. Microbio. and Infection 21,631-639.

Ostfeld, R.S., Canham, C.D., Oggenfuss, K., Winch-combe, R.J., Keesing, F., 2006. Climate, deer, rodents, and acorns as determinants of variation in Lyme-Disease risk. PLoS Biol 4, e145. doi:10.1371/journal.pbio.0040145

Platonov, A.E., Karan, L.S., Kolyasnikova, N.M., Makh-neva, N.A., Toporkova, M.G., Maleev, V.V., Fish, D., Krause, P.J., 2011. Humans infected with relapsing fever spirochete Borrelia miyamotoi, Russia. Emerg. Infect. Dis. 17, 1816–1823. doi:10.3201/eid1710.101474

Scoles, G.A., Papero, M., Beati, L., Fish, D., 2001. A re-lapsing fever group spirochete transmitted by Ixodes scapularis ticks. Vector-Borne Zoonotic Dis. 1, 21–34. doi:10.1089/153036601750137624

Scott, M.C., Rosen, M.E., Hamer, S.A., Baker, E., Ed-wards, H., Crowder, C., Tsao, J.I., Hickling, G.J., 2010. High-prevalence Borrelia miyamotoi infection among wild urkeys (Meleagris gallopavo) in Tennessee. J. Med. Entomol. 47, 1238–1242. doi:10.1603/ME10075

Steere, A.C., Coburn, J., Glickstein, L., 2004. The emer-gence of Lyme disease. J. Clin. Invest. 113, 1093–1101. doi:10.1172/JCI200421681

Vrijenhoek, R., 1994. DNA primers for amplification of mi-tochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Mar Biol Biotechnol 3, 294–299.

Journal of the Pennsylvania Academy of Science 90(2): 48-55, 2016

1Accepted for publication November 2016.2Department of Biology, 309 Kunkel Hall, 705 Sullivan Road, Easton PA 18042. Phone: 610-330-5462. E-mail: CaslakeL@Lafayette.edu

CONJUGATIVE COMPETENCY OF MERCURY RESISTANT BACTERIA ISOLATED FROM ONONDAGA LAKE, NY1

COLLEEN WALSH AND LAURIE F. CASLAKE2

Department of Biology, Lafayette College, Easton PA 18042-1778

ABSTRACT

Onondaga Lake has a 100-year history of industrial and domestic pollution resulting in its addition to the EPA National Priorities List in 1994. Early research efforts in Onondaga Lake focused on the numbers of heterotrophic bacteria and the proportion displaying phenotypic mercury resistance. Bacterial resistances to diverse metals, including mercury, are often physically linked to genes encoding antibiotic resistance. Given the occurrence of multiple resistance genes on the same plasmid or transposon, it is possible that heavy metal selection could cause the spread of antibiotic resistance through the microbial population by conjugation. We tested our existing culture collection of both mercury-resistant and mercury-susceptible bacteria from Onondaga Lake for the ability to transfer antibiotic resistance via conjugation to the Escherichia coli C600 recipient strain. The frequency of transfer of antibiotic resistance via conjugation was statistically higher for mercury-resistant isolates than mercury-susceptible isolates. This has long-term implications for the increasing spread of antibiotic resistance through the microbial community in Onondaga Lake.[ J PA Acad Sci 90(2): 48-55, 2016 ]

INTRODUCTION

Mercury is among the class of metals with no known biological function. The majority of mercury and mercury-containing compounds are released into water and air through fossil fuel combustion and the electrodes used in the chlor-alkali industrial process. This mercury is usually Hg2+, the ionic form that results from the interaction of volatile Hg0

with ozone in the presence of water (Barkay et al. 2003). In the environment, microbes can convert mercury compounds

from one form into another, including by methylation and demethylation, within the global mercury cycle. While exposure to ionic mercury can induce cellular toxicity, many bacteria exhibit mercury resistance via reduction of Hg2+ to the volatile form, Hg0, mediated by an inducible NADPH-dependent disulfide oxidoreductase. The genes that code for the proteins responsible for this biochemical transformation are organized into the mer operon and occur broadly, being found in both Gram-negative and Gram-positive organisms (Barkay et al. 2003).

Onondaga Lake in Syracuse, NY (Figure 1) received mercury-laden waste from two chlor-alkali plants located south of Ninemile Creek on the lake's western shore (reviewed in Effler 1996). These chlor-alkali plants discharged an estimated 76,000 kg (170,000 lb) of mercury to the lake between 1946 and 1970 (Effler and Matthews 2003). Although Allied Signal Inc. was ordered to reduce emissions in 1970 and the plants closed in 1977 and 1988, total mercury levels in the water, sediment, and fish tissue continue to exceed EPA standards (Effler 1996). In addition to the original mercury contamination from the chlor-alkali

Figure 1. Line drawing of Onondaga Lake showing the sampling sites (triangles) at Ley Creek (S1) and North Barge Canal (S2). The Syracuse Metropolitan Wastewater Treatment Plant (METRO) and the Bristol-Myers plant (BMS) are shown (circles). Line drawing produced by Mr. John Clark, Digital Scholarship Services at Lafayette College, from data at the Cornell University Geospatial Information Repository.

49

plants, an estimated 14.1 kg of mercury entered Onondaga Lake in 1992 with Ninemile Creek being the highest source (7.1 kg Hg yr -1; Henry et al. 1995). Two additional sources, the Syracuse Metropolitan Wastewater Treatment Plant, METRO, (3.8 kg Hg yr -1) and Onondaga Creek (1.8 kg Hg yr-1), also contributed mercury to the lake (Klein and Jacobs 1995). Historically, surface sediment mercury concentrations in Onondaga Lake ranged from 0.15 mg Hg kg -1 to 68.9 mg Hg kg -1 (NYS Dept of Health 1995); the highest measurement far exceeds the 0.2 mg Hg kg -1 level considered to be of “low risk to aquatic life” set by the New York State Department of Environmental Conservation (NYS DEC 2014). Water column mercury concentrations ranging from 7 to 35 ng Hg L-1 exceed the 1.3 ng Hg L-1

typical of relatively pristine systems (Bloom and Effler 1990; Effler 1996).

Recently, the EPA collected soil and swale samples along the Ley Creek tributary (U.S. EPA, 2014). The highest level of mercury in soils, 4.1 mg Hg kg -1, was found in the surface layer (0-12” deep) on the northern bank of Ley Creek. Elevated levels of mercury (3.5 mg Hg kg -1) were also found more deeply deposited (30-36” deep) on the southern bank of the creek (U.S. EPA 2014). The removal of contaminated sediment and the reduction of mercury levels to below 0.18 mg Hg kg -1 (in the top 2 ft) and below 2.8 mg Hg kg -1 (in the subsurface) is the objective of the current remedial measures initiated in Ley Creek in 2011 (US EPA 2014).

Unlike organic pollutants, toxic metals are not subject to degradation and may remain as agents of selective pressure for a long period of time (Stepanauskas et al. 2006). It is well known that microorganisms are capable of growth in the presence of mercury and mercury-resistant bacteria are distributed worldwide in both pristine and contaminated environments (Osborn et al. 1995). Early work by Barkay & Olson (1986) revealed that the proportion of the heterotrophic microbial population of Onondaga Lake that was resistant to 0.05 mg ml-1 mercury ranged between 0.007% and 0.55%, depending on the location, and a strong positive correlation existed between sediment mercury concentration and the presence of the mer operon (r = 0.96). From these data the authors concluded that horizontal gene transfer of the mer operon among microbes contributed to the ability of the bacteria to adapt to the presence of mercury (Barkay and Olson 1986).

Six years after its declaration as a Superfund site, and twelve years after the last plant closed, we quantified the proportion and distribution of mercury resistant organisms in Onondaga Lake and collected a population of organisms for future analyses. This collection of mercury-susceptible and mercury-resistant Onondaga Lake isolates from both Ley Creek, an inlet, and the lake outlet at the North Barge Canal were tested for phenotypic antibiotic resistance and the ability to transfer this antibiotic resistance to Escherichia coli C600 through conjugation. While both mercury-resistant and mercury-susceptible microbes were phenotypically

antibiotic resistant, the ability to transfer this resistance to E. coli C600 was statistically greater in the mercury-resistant population.

MATERIALS AND METHODS

Onondaga Lake

Onondaga Lake is located at latitude 43° 06' 54", longitude 76° 14' 34" north of the City of Syracuse, NY (Figure 1). The lake has a surface area of 12 km2, a volume of 1.31 x 108 m3, a mean depth of 10.9 m and a maximum depth of 19.5 m (Effler & Matthews 2003). Onondaga Lake receives surface runoff from a drainage basin of approximately 250 square miles. Two major tributaries, Ninemile Creek and Onondaga Creek, provide approximately 70% of the water input to Onondaga Lake; other tributaries, including Ley Creek, contribute an additional 10% (Effler & Harnett 1996). METRO, the Metropolitan Syracuse Wastewater Treatment Plant, which can treat up to 126 million gallons per day through the full secondary and tertiary treatment processes, contributes 20% of the annual flow. The lake flushes approximately four times per year on a completely mixed basis (Effler & Matthews 2003). Onondaga Lake discharges through the North Barge Canal, a 1200 m channel, into the Seneca River that combines with the Oneida River becoming the Oswego River that enters Lake Ontario (approximately 56 km north) at the city of Oswego (Figure 1 inset).

Sample collection and enumeration of mercury-resistant bacteria

Ley Creek (site 1) and the North Barge Canal (site 2; see Figure 1) were chosen for ease of access and were sampled in July 2000. Seven water samples and three sediment samples were collected from each site into sterile specimen containers (120 mL, Fisher Scientific, Pittsburgh, PA). Sample containers were first rinsed with lake water from the sampling site and then filled completely with either water from a depth of approximately 20 cm or sediment from the top 2 cm to 3 cm within 1 m of the lake perimeter. Samples were stored for less than 6 hours on ice until processed back in the laboratory. Water temperature was obtained during sampling; water pH was determined in the laboratory (Beckman Instruments, Fullerton, CA).

Excess water was aspirated from each sediment sample; one gram (wet weight) was diluted with 5 ml of sterile saline (0.85% NaCl). Samples were agitated on a rotary shaker (180 rpm, 10 min) at room temperature and the aqueous phase was decanted. Water samples and the aqueous phase from sediment samples were serially diluted in sterile saline.

50JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

Enumeration of total culturable aerobic heterotrophic bacteria was performed in triplicate with plate count agar (PCA). Enumeration of aerobic mercury-resistant bacteria was performed in triplicate with PCA amended with 100 µM HgCl2 (PCA/Hg; Mindlin et al. 2001; Rochelle et al. 1989) using filter-sterilized mercury stock solution (0.1 M) added to autoclaved, tempered agar prior to dispensing. All plates were incubated at 25 °C for 36 to 48 hrs prior to colony counting. Average bacterial counts and standard errors were calculated by location and source. Outliers were eliminated through 1.5 IQR analysis (Grubbs 1969). The percent of bacterial mercury resistance was calculated as the percent of aerobic mercury-resistant bacteria among total aerobic heterotrophs. A random sample of mercury-susceptible and mercury-resistant isolates from the water and sediment from each site were frozen at -80 °C in cell freezing media (2X stock: 1 gm yeast extract, 10 ml DMSO, 10 ml glycerin, 80 ml 0.2 M NaPO4 pH 7.0) for future analyses.

Antibiotic sensitivity testing using Kirby Bauer method.

Mercury-susceptible and mercury-resistant isolates were streaked from frozen stock onto PCA or PCA/Hg plates to test for viability. Isolates were grown at 22 °C in 1 ml nutrient broth or nutrient broth plus 100 µM HgCl2, respectively, on a rotary shaker at 180 rpm for 48 hrs. Liquid cultures, 400 µl, were spread onto 150 x 15 mm PCA plates. Sensitivity discs for eight antibiotics (Table 1) with diverse modes of action were placed onto the plates with sterile forceps: ampicillin (10 µg), ciprofloxacin (5 µg), naladixic acid (30 µg), sulfisoxazole (25 µg), erythromycin (15 µg), kanamycin (30 µg), streptomycin (10 µg), and tetracycline (30 µg). Plates were incubated at 25 °C for 48 hrs. Isolates

were characterized as antibiotic sensitive or resistant (CLSI 2005). To determine the level of antibiotic resistance of individual isolates, the multiple antibiotic resistance (MAR) index was calculated by dividing the number of antibiotics to which an isolate was resistant by the total number of antibiotics to which the isolate was exposed (Krumperman 1983). A MAR value ≥0.25 indicated a multiply antibiotic-resistant bacterium.

Conjugation

Onondaga Lake isolates from the water fractions from both sampling sites (n = 88) were conjugated with Escherichia coli K-12, C600 (Rifr, Nalr, Lac-, F-), a standard restriction modification-defective laboratory strain (a gift from Dr. Anne Summers, University of Georgia), which served as the recipient for conjugation experiments. Escherichia coli C600 is mercury susceptible (HgS) and resistant to 100 µg ml-1 rifampin (Rif) and 50 µg ml-1 naladixic acid (Nal) and sensitive to other antibiotics. Donors were randomly chosen mercury-susceptible and mercury-resistant isolates carrying resistance to several antibiotics. Donor strains and E. coli C600 were grown separately in 1 ml nutrient broth (Becton, Dickinson & Co., Franklin Lakes, NJ), at 30 °C with shaking at 180 rpm for 24 hrs; 5 µl of each culture was added to 1 ml nutrient broth, vortexed on low for 5 sec and incubated at 30 °C for 24 hrs without shaking. Post-mating selection was performed by spreading 100 µl aliquots of each mating mixture onto appropriate double-selective medium and incubating at 30 °C for 24 hrs (Wireman et al. 1997). The double antibiotic selection for the transconjugants ensured that none of the donor strains would grow. From each mating, four individual E. coli C600 transconjugants were randomly

Percentage of isolates exhibiting antimicrobial agent resistance, by type of sample

AntibioticSource: Ley Creek (Lake input) Source: North Barge Canal (Lake outflow)

Water Sediment Water SedimentHgS (n=37) HgR (n=35) HgS (n=41) HgR (n=32) HgS (n=37) HgR (n=44) HgS (n=45) HgR (n=34)

Ampicillin 64.9% 68.6% 43.9% 71.9% 94.6% 72.7% 66.7% 94.1%Ciprofloxacin 5.4% 0.0% 0.0% 0.0% 2.7% 4.5% 4.4% 0.0%Naladixic Acid 8.1% 5.7% 2.4% 9.4% 13.5% 25.0% 15.5% 5.9%Sulfisoxazole 62.2% 88.6% 58.5% 71.9% 100.0% 95.4% 82.2% 97.1%Erythromycin 43.2% 60.0% 24.4% 34.4% 94.6% 86.4% 42.2% 97.1%Kanamycin 48.6% 42.9% 7.3% 40.6% 18.9% 2.3% 22.2% 8.8%Streptomycin 24.3% 25.7% 7.3% 9.4% 56.8% 6.8% 15.6% 23.5%Tetracycline 0.0% 5.7% 2.4% 3.1% 16.2% 0.0% 8.9% 5.9%

Table 1. The percentage of mercury-susceptible (HgS) and mercury-resistant (HgR) isolates from Ley Creek and the North Barge Canal expressing resistance to each antibiotic.

51

selected, grown in 1 ml nutrient broth at 30 °C with shaking at 180 rpm for 24 hrs, spread on NA/Rif/Nal plates (to select for E. coli C600), and tested for the presence of additional non-selected markers by the Kirby-Bauer agar-diffusion method. Isolates were judged according to CLSI criteria (2005) and were determined to be positive for conjugation if E. coli C600 became resistant to any antibiotics, other than rifampin and naladixic acid, following conjugation.

Statistical Analyses

Chi-Square Analyses. Chi-square analysis tested the null hypothesis that mercury resistance is independent of multiple antibiotic resistance among bacterial isolates from both sample sites in Onondaga Lake; a p-value less than or equal to 0.05 was considered statistically significant.

Antibiotic Resistance. Antibiotic resistance load between populations of organisms from the water and sediment at each site were compared using a one-tailed Students’ t test. To account for multiple pairwise tests, the alpha level was adjusted based on four comparisons for more rigor. Therefore a p-value less than or equal to 0.0125 (0.05 ÷ 4 tests) was deemed statistically significant.

Conjugation Frequencies. Conjugation frequencies were compared statistically with Student's t test adjusted for frequency data and calculated using pooled variance for unequal cell means. Alpha level was adjusted for non-independent tests. Means were significantly different when sample t exceeded critical t at probability of 0.0125.

RESULTS

The presence and distribution of mercury-resistant bacteria in Onondaga Lake

Heterotrophic organisms in the water and sediment from the two sampling sites in Onondaga Lake were analyzed for phenotypic mercury resistance, determined by growth on PCA/Hg. Water and sediment samples from Ley Creek contained approximately one order of magnitude more microbes than the water and sediment samples from the North Barge Canal (Figure 2A, 2B). The proportion of mercury-resistant organisms in the water fraction at Ley Creek (Figure 2C) was higher than at the North Barge Canal, the outlet of the lake. The proportion of mercury-resistant organisms in the sediment was similar between the two sites (Figure 2C).

The occurrence of antibiotic resistance among Onondaga Lake microorganisms from Ley Creek and the North Barge Canal

Mercury-susceptible (n=160) and mercury-resistant (n=145) isolates were tested for sensitivity to eight antibiotics using the Kirby-Bauer disc diffusion method (Table 1). Only 18 isolates (5.9%) were susceptible to all eight antibiotics; 13 of these were mercury-susceptible while five were mercury-resistant (Figure 3A). There was a statistically significant association between mercury-resistance and antibiotic resistance in the sediment at the North Barge Canal

Figure 2. Heterotrophic microbial counts in samples collected from the (A) water and (B) sediment of Ley Creek and the North Barge Canal; (C) the distribution of phenotypic mercury-resistance among these organisms. The standard error of counts from all Ley Creek samples was < 9%; the standard error of counts from the North Barge Canal samples ranged from 11% (water) to 20% (sediment).

52JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

(p=0.002). In the isolates from the water fraction at the North Barge Canal, there was a statistically significant difference in the antibiotic resistance load between mercury-resistant and mercury-susceptible organisms (p=0.003); however, resistance to three or more antibiotics was more common in mercury-susceptible organisms. There was no statistically significant association between mercury-resistance and antibiotic-resistance in the population of organisms from Ley Creek water (p=0.118) or sediment (p=0.422).

None of the organisms tested were resistant to all eight antibiotics; however, one mercury-susceptible organism isolated from the North Barge Canal sediment was resistant to seven antibiotics (Figure 3A). Multiply-resistant isolates (i.e., resistant to two or more antibiotics) predominated among all isolates from the two sampling sites (Figure 3A). In general, resistance to three or more antibiotics (MAR > 0.38) occurred with greater frequency among isolates that were also mercury-resistant with the lone exception of isolates from the water fraction at the North Barge Canal where 84% of mercury-resistant isolates were resistant to three or more antibiotics while 97% of mercury-susceptible isolates carried that level of resistance (p=0.003; Figure 3B). Irrespective of the mercury-resistance status, a higher proportion of isolates from the water and sediment samples from the North Barge Canal were resistant to three or more antibiotics compared to organisms isolated from water and sediment from Ley Creek (Figure 3B).

Conjugation

Both mercury-susceptible and mercury-resistant microbial isolates from water samples at each site (n=88) that exhibited resistance to one or more antibiotics served as a genetic donor to E. coli C600. Conjugative competency was analyzed by testing for antibiotic resistance transfer. Mercury-resistant isolates from both Ley Creek and the North Barge Canal exhibited a significantly greater frequency of antibiotic resistance transfer to the E. coli C600 recipient strain through conjugation than did mercury-susceptible isolates from the same site (Figure 4). Irrespective of the mercury-resistance status, isolates from the North Barge Canal exhibited a significantly greater frequency of antibiotic resistance gene transfer than isolates from Ley Creek.

DISCUSSION

Twenty years ago, Onondaga Lake was added to the Federal Superfund National Priorities List. The long-term goal is to reduce the level of industrial chemicals, including mercury, that contaminate the sediment and to return the lake and its tributaries to their original splendor (U.S. EPA 2015). Over the past 20 years, major efforts to restore the native habitat of the lake have included the removal of eight

tons of mercury from the Linden Chemicals and Plastics (LCP) property and the dredging of contaminated sediment from the lake bottom. Dredging began in 2012 with oversight by the NYS Department of Environmental Conservation, the EPA, and NYS Department of Health and was completed in 2014 (U.S. EPA 2015).

The proportion of mercury-resistant bacteria at various sites in Onondaga Lake was analyzed when samples were collected in 2000. Isolates from Ley Creek sediment demonstrated an increase of two-orders of magnitude over

Figure 3A. Percentage of isolates displaying resistance to increasing numbers of antibiotics (A) and to three or more antibiotics (B). Percentage was calculated for eight bacterial groups. Mercury-resistant (n=35) and mercury-susceptible (n=37) isolates from water samples and mercury-resistant (n=32) and mercury-susceptible (n=41) isolates from sediment samples from Ley Creek. Mercury-resistant (n=44) and mercury-susceptible (n=37) isolates from water samples and mercury-resistant (n=34) and mercury-susceptible (n=45) isolates from sediment samples from North Barge Canal. In Panel A, R is the numbers of antibiotics that isolates resist.

53

data documented nearly 20 years earlier (2.2% in 2000 vs. 0.023% in 1984; Barkay & Olson, 1986). Although, the percentage of isolates from Ley Creek demonstrating mercury resistance had increased dramatically over 1984 levels, the numbers of viable sediment heterotrophic organisms in 2000 (~2.2 x 105 CFU gm-1) did not differ significantly from 1984 cell counts of 2.97 x 105 CFU gm-1. While there may be differences attributable to the exact collection location and recent weather events, the increase in mercury resistance may also reflect the impact of mercury loading from tributaries and METRO and the ability of organisms to participate in horizontal gene transfer. The continuous input of mercury between 1984 and 2000 impacts the sediment ecology by favoring mercury-resistant microorganisms; these organisms may lower the proximal ionic mercury levels and thus provide a benefit to other aquatic organisms. In an earlier study, macrophytes from Onondaga Lake were compared to those from a non-mercury impacted lake, Lake Nockamixon, PA. Both populations of macrophytes were grown in sediment from Onondaga Lake (adjusted to 0.68 mg Hg kg-1). After six weeks, macrophytes from Lake Nockamixon exhibited chlorosis and tissue sloughing typical of mercury toxicity (Mhatre and Chapehekar 1985) while macrophytes from Onondaga Lake, replete with an intimate association of mercury-resistant bacteria, remained healthy green in appearance (Caslake et al. 2006). These results suggest that mercury-resistant bacteria may play a vital role in detoxifying mercury compounds to decrease the effects of mercury bioaccumulation.

Presently, the mer operon is one of the best-understood mechanisms of resistance to mercury. The mer operon is found in both Gram-positive and Gram-negative bacteria on transposons and plasmids making it possible to be transferred from cell-to-cell. Bacterial gene transfer occurs throughout the biosphere, especially in nutrient rich sites such as aquatic systems, sediments, and in sewage treatment plants. An earlier report of a positive relationship between mercury stress and the gene-mobilizing capacity of soil bacteria (Alonso et al., 2001) suggested to us that the increased conjugative activity might lead to the concomitant transfer of one or more antibiotic resistance genes. We focused our efforts on isolates from the water samples from Ley Creek and the North Barge Canal due to the relatively high proportion of mercury-resistance in isolates from Ley Creek (5.1%) and the relative paucity of mercury-resistance in isolates from the North Barge Canal (0.2%) – the level of antibiotic resistance at least among mercury-resistant organisms, was similar between the two sites. Conjugal transfer of antibiotic resistances to the E. coli C600 recipient strain was observed more frequently from mercury-resistant isolates than from mercury-sensitive isolates within both sites (although the use of saline and standard nutrient media in conjugation experiments may have introduced hypertonic stress to these freshwater organisms, this is a standard method for handling medically important bacteria and was necessary to support E. coli growth during conjugation experiments). As a percentage, mercury-resistant isolates are relatively rare at the North Barge Canal (0.2% of total); however, these organisms carry more antibiotic resistant genes (84% carry resistance to three or more antibiotics) and are more likely to transfer these resistances to E. coli C600 under standard laboratory conditions. While our data indicate that mercury-resistant planktonic isolates transferred antibiotic resistance relatively easily in liquid culture, a more detailed study showed that transfer efficiency can vary by as much as a million-fold, depending on the growth phase of the recipient and the type of plasmid (Król et al. 2013).

In the present study, high levels of antibiotic resistance were observed in both mercury-susceptible and mercury-resistant organisms from both sampling sites. In general, resistance to erythromycin was more common in mercury-resistant organisms than in mercury-susceptible organisms with the lone exception being the water environment at the North Barge Canal where the vast majority of mercury-susceptible and mercury-resistant organisms expressed erythromycin resistance. More than 50% of organisms across all sites were resistant to sulfisoxazole, indicating the presence of an alternate enzyme for dihydrofolic acid synthesis that is insensitive to this antibiotic. At most of our sampling sites, a higher proportion of organisms were resistant to erythromycin than kanamycin. While both kanamycin and erythromycin inhibit protein synthesis, kanamycin binds the 70S ribosome inhibiting translocation and eliciting miscoding while erythromycin binds the 50S subunit

Figure 4. Mean (n = 14 to 32, + 1 SE) Relative Conjugation Frequency. Conjugation frequencies were calculated as: # of resistances transferred/total # of resistances known to be present for transfer. Conjugation frequencies were compared statistically with Student's t test adjusted for frequency data and calculated using pooled variance for unequal cell means. Alpha level was adjusted for non-independent tests. Means were significantly different when sample t exceeded critical t at probability of 0.0125; means with the same letter do not differ statistically.

54JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

and blocks translocation from the A-site to the P-site. In enterococci from human and animal sources, erythromycin resistance is typically due to target modification by the erythromycin resistance methylase genes, primarily ermB (Frye and Jackson 2013). The observed level of resistance to antibiotics was higher than expected; however, the measured levels in Onondaga Lake are comparable to other environments. When untreated groundwater was tested, 100% of noncoliforms and 87% of coliforms were resistant to at least one antibiotic, with an average of 28% of organisms exhibiting resistance to five or more antibiotics (McKeon et al. 1995). West et al. (2011) reported an average of 81% of isolates from upstream of two-wastewater treatment plants in Lake Michigan were phenotypically ampicillin-resistant compared to 89% from downstream of the same wastewater treatment plants. However, in their study a higher proportion of isolates from upstream (19%) than downstream (10.9%) of the wastewater treatment plants were resistant to two or more antibiotics (West et al. 2011).

Waste input treated in the Syracuse Wastewater Treatment Plant may have contributed to the observed high levels of antibiotic resistance. In 1943, Bristol-Myers purchased Cheplin Biological Laboratories Inc. (see Figure 1) and began producing penicillin in Syracuse. Over the decades, the plant manufactured up to 70% of the penicillin produced in the United States (Effler et al. 2001; Effler et al. 2013) with waste being processed through METRO. Therefore, in addition to normal cycling of antibiotics through humans and into the waste stream, manufacturing may have contributed to the antibiotic load through METRO.

Bacteria exchange DNA in their natural environment using a variety of processes including conjugation. In the present study, with one exception, isolates from the water column at both Ley Creek and the North Barge Canal exhibited a higher percentage of multi-drug resistance (MAR > 0.38) than did samples from the sediment. This suggests that transfer of antibiotic resistance may occur more easily in water than in sediment. This hypothesis is supported by data from Titus and Pfister (1984) who report a higher conjugation frequency of the plasmid-encoded cadmium resistance operon between Ottawa River (Ohio) isolates in a water column than between the same isolates in a sediment column.

Improvements to the wastewater treatment plant and other rehabilitation efforts have dramatically improved the water quality and ecological characteristics of Onondaga Lake and surrounding river system. While there is limited input of mercury from external sources, internal sources (e.g., turnover from the hypolimnion) remained an important source of mercury to the food chain. Capping the sediment may reduce the vertical diffusion from the hypolimnion, but a concomitant reduction in mercury-resistant organisms may prove elusive due to various selective pressures acting on these multiply resistant organisms. Future studies characterizing the mercury- and antibiotic-resistant status of organisms in this environmental system are warranted.

ACKNOWLEDGEMENTS

We thank three anonymous reviewers for critical comments on this manuscript along with Drs. John Freeman and Nancy McCreary Waters for comments on an earlier version. We thank Mr. John Clark for production of Figure 1 from data at the Cornell University Geospatial Information Repository and Ms. Stephanie Giordano for assistance with sample collection. The EXCEL Scholars Program is acknowledged for funding to C. W. with additional funding from the Lafayette College Department of Biology.

LITERATURE CITED

Alonso, A., P. Sanchez, and J. Martinez. 2001. Environmen-tal selection of antibiotic resistance genes. Environ. Mi-crobiol. 3:1-9.

Barkay, T., S.M. Miller, and A.O. Summers. 2003. Bacterial mercury resistance from atoms to ecosystems. FEMS Mi-crobiol. Rev. 27:355-384.

Barkay, T., and B.H. Olson. 1986. Phenotypic and genotyp-ic adaptation of aerobic heterotrophic sediment bacterial communities to mercury stress. Appl. Environ. Microb. 52:403-406.

Bloom, N.S., and S.W. Effler. 1990. Seasonal variability in the mercury speciation of Onondaga Lake (New York). Water Air Soil Poll. 53:251-265.

Caslake, L.F., S.S. Harris, C. Williams, and N.M. Waters. 2006. Mercury-resistant bacteria associated with macro-phytes from a polluted lake. Water Air Soil Poll. 174:93-105.

CLSI, Clinical and Laboratory Standards Institute, 2005. Methods for antimicrobial dilution and disk susceptibility testing of infrequently isolated or fastidious bacteria; pro-posed guideline. CLSI Document M45-P [ISBN 1-56238-583-6]. Clinical and Laboratory Standards Institute, 940 West Valley Road, Suite 1400, Wayne, Pennsylvania USA.

Effler, S.W. 1996. Synthesis and perspectives, p. 789-811. In S.W. Effler (ed.), Limnological and engineering analysis of a polluted urban lake. Springer Verlag, New York, Inc.

Effler, S.W., R.K. Gelda, C.M. (Brooks) Matthews. 2001. Implications of industrial loads for ammonia pollution in an urban lake. Water Environ. Res. 73:192-203.

Effler, S.W. and G. Harnett. 1996, Chapter 1. ‘Background’, in S.W. Effler (ed), Limnological and engineering analysis of a polluted urban lake, Springer-Verlag, Inc, New York, NY, USA, pp. 1-31.

55

Effler, S.W., and D.A. Matthews, 2003. Impacts of a soda ash facility on Onondaga Lake and the Seneca River, NY. Lake Reserv. Manage. 19:285-306.

Eff ler, S.W., S.M. O’Donnell, A.R. Prestigiacomo, D.A. Matthews, and M.T. Auer. 2013. Retrospective analyses of inputs of municipal wastewater effluent and coupled impacts on an urban lake. Water Environ. Res. 85:13-26.

Frye, J.G., and C.R. Jackson. 2013. Genetic mechanisms of antimicrobial resistance identified in Salmonella enteri-ca, Escherichia coli, and Enterococcus spp. isolated from U.S. food animals. Frontiers in Microbiology. 4:1-22.

Grubbs, F.E. 1969. Procedures for detecting outlying obser-vations in samples. Technometrics 11:1-21.

Henry, E.A., L.J. Dodge-Murphy, G.N. Bigham, S.W. Klein, and C.C. Gilmour. 1995. Total mercury and meth-ylmercury mass balance in an alkaline, hypereutrophic urban lake (Onondaga Lake, NY). Water Air Soil Poll. 80:509-518.

Klein, S.M., and L.A. Jacobs. 1995. Distribution of mercury in the sediments of Onondaga Lake, NY. Water Air Soil Poll. 80:1035-1038.

Król, J.E., A.J. Wojtowicz, L.M. Rogers, H. Heuer, K. Smalla, S.M. Krone, and E.M. Top. 2013. Invasion of E. coli biofilms by antibiotic resistance plasmids. Plasmid 70:110-119.

Krumperman, P.H. 1983. Multiple antibiotic resistance in-dexing Escherichia coli to identify risk sources of fae-cal contamination of foods. Appl. Environ. Microbiol. 46:165-170.

McKeon, D.M., J.P. Calabrese, and G.K. Bissonnette. 1995. Antibiotic resistant gram-negative bacteria in rural groundwater supplies. Water Res. 29:1902-1908.

Mhatre, G.N., and S.B. Chapehekar. 1985. The effect of mercury on some aquatic plants. Environ. Pollut. A, 39:207–216.

Mindlin, S., G. Kholodii, Z. Gorlenko, S. Minakhina, L. Minakhin, E. Kalyaeva, A. Kopteva, M. Petrova, O. Yurieva and V. Nikiforov. 2001. Mercury resistance transposons of gram-negative environmental bacteria and their classification. Res. Microbiol. 152:811-822.

New York State Department of Health, 1995. Public Health Assessment & Health Consultations, Onondaga Lake, Syracuse, Onondaga County, New York. CERCLIS No. NYD986913580. (http://www.atsdr.cdc.gov/HAC/pha/PHA.asp?docid=245&pg=0)

New York State Department of Environmental Conserva-tion. Division of Fish, Wildlife and Marine Resources, Bureau of Habitat. 2014. Screening and Assessment of Contaminated Sediment. June 24, 2014 (www.dec.ny.gov/docs/fish_marine_pdf/screenasssedfin.pdf)

Osborn, A.M., K.D. Bruce, P. Strike, and D.A. Ritchie. 1995. Sequence conservation between regulatory mercu-ry resistance genes in bacteria from mercury polluted and pristine environments. Syst. Appl. Microbiol. 18:1-6.

Rochelle, P.A., J.C. Fry and M.J. Day. 1989. Factors affect-ing conjugal transfer of plasmids encoding mercury resis-tance from pure cultures and mixed natural suspensions of epilithic bacteria. J. Gen. Microbiol. 135:409-424.

Stepanauskas, R., T.C. Glenn, C.H. Jagoe, R.C. Tuckfield, A.H. Lindell, C.J. King and J.V. McArthur. 2006. Cose-lection for microbial resistance to metals and antibiotics in freshwater microcosms. Environ. Microbiol. 8:1510-1514.

Titus, J.A., and R.M. Pfister. 1984. Bacteria and cadmium interactions in natural and laboratory model aquatic sys-tems. Arch. Environ. Con. Tox. 13:271-277.

U.S. Environmental Protection Agency. 2014. Record of Decision, Lower Ley Creek Subsite of the Onondaga Lake Superfund Site, City of Syracuse/Town of Salina, Onondaga County, New York. U.S. EPA. Accessed on September 1, 2015.

U.S. Environmental Protection Agency. 2015. First Five-Year Review Report Onondaga Lake Bottom Sub-site of the Onondaga Lake Superfund Site, Onondaga County, New York. U.S. EPA. Accessed on October 9, 2015. Available at: http://semspub.epa.gov/src/docu-ment/02/372861

West, B.M., Liggit, P., Clemans, D.L., and S.N. Francoeur. 2011. Antibiotic resistance, gene transfer, and water qual-ity patterns observed in waterways near CAFO farms and wastewater treatment facilities. Water Air Soil Pollut 217:473-489.

Wireman, J., C.A. Liebert, T. Smith and A.O. Summers. 1997. Association of mercury resistance with antibiot-ic resistance in the gram-negative fecal bacteria of pri-mates. Appl. Environ. Microbiol. 63:4494-4503.

Journal of the Pennsylvania Academy of Science 90(2): 56-61, 2016

ABSTRACT

There are three major human-biting tick species in the northeastern United States: Ixodes scapularis (blacklegged tick), Amblyomma americanum (lone star tick), and Dermacentor variabilis (American dog tick). While all three species have been found on American black bears (Ursus americanus), there are no published reports on the prevalence and diversity of ticks on black bears in northeastern Pennsylvania. Ticks were collected from black bears post-mortem during the hunting season in November 2015 in northeastern Pennsylvania. Four-hundred eighteen ticks were collected from 54 black bears at PA Game Commission hunter check stations. Only adult Ixodes scapularis were found on these bears. The mean number of I. scapularis collected per bear was 7.74 ± 1.21 (SE) and 70.6 % of the collected ticks were female. A subset of females (n = 75) were measured to determine the duration that each tick fed on it’s host. DNA was isolated from this subset of ticks and subjected to polymerase chain reaction (PCR) to investigate the prevalence of select pathogens that cause disease in humans. Over half of the measured females (54.7 %) fed for more than 4 days on their black bear hosts. Borrelia burgdorferi was detected in 13 I. scapularis (17.3 %), B. miyamotoi in 1 (1.3 %), and 5 ticks were PCR-positive for Anaplasma spp. (6.7 %).[ J PA Acad Sci 90(2): 56-61, 2016 ]

INTRODUCTION

Since the late 1900’s, American black bear (Ursus americanus) populations have been increasing in size and expanding their range across the eastern United States (Williamson 2002). Currently, the population estimate for black bears in Pennsylvania is 18,000 whereas the estimate was 4,000 in the 1970’s (PA Game Commission 2016). Throughout it's range, the American black bear is

TICKS OF BLACK BEARS (URSUS AMERICANUS) FROM NORTHEASTERN PENNSYLVANIA AND TICK-BORNE PATHOGENS1

ELIZABETH MCGOVERN2, J. MISCHLER3, M. BIRD2, AND J. HUFFMAN2,4

2Department of Biological Sciences Northeast Wildlife DNA Laboratory, East Stroudsburg University, East Stroudsburg, PA 3Department of Biology King’s College, Wilkes – Barre, PA

documented as a host of several human-biting tick species, including Ixodes scapularis (blacklegged tick), Amblyomma americanum (lone star tick), and Dermacentor variabilis (American dog tick) (Rogers 1975, Burguess and Huffman 2005, Yabsley et al. 2009).

Since the late 1900’s, Ixodes scapularis has been increasing in abundance in Pennsylvania and is now present in all 67 counties (Hutchinson et al. 2015). Ixodes scapularis is the primary vector of Borrelia burgdorferi, Anaplasma phagocytophilum, and Babesia microti in the northeastern United States (Keirans et al. 1996, Jin et al. 2012). All 3 of these tick-borne diseases are a threat to human health, and the number of cases are increasing as are the geographical ranges of cases for each disease (CDC 2015b). Lyme disease is the most commonly reported vector-borne disease in the United States (CDC 2015a). In 2013, 36,307 cases of Lyme disease were reported in the United States whereas relatively few anaplasmosis (2,782) and babesiosis (1,796) cases were reported (CDC 2015b). An emerging human disease pathogen, Borrelia miyamotoi, has been detected in I. scapularis indicating that this tick species is also the vector responsible for B. miyamotoi infection in humans (Barbour et al. 2009, Krause et al. 2013, CDC 2015c).

Adult I. scapularis feed on medium and large-sized mammal hosts (Keirans et al. 1996, Stafford 2007), including American black bears (King 1960, Rogers 1975, Yabsley et al. 2009, Leydet and Liang 2013, Zolnik et al. 2015). Ixodes scapularis mating may occur on or off a host (Oliver 1989, Kiszewski et al. 2001, Stafford 2007), and generally occurs in the fall or spring (Yuval and Spielman 1990, Daniels et al. 1996). Adult females must obtain a blood meal to lay eggs in the spring or early summer (Oliver 1989, Yuval and Spielman 1990, Daniels et al. 1996), whereas males do not need to feed and generally die soon after copulation (Stafford 2007, Troughton and Levin 2007). Females likely play a more significant role in I. scapularis range expansion and local prevalence because a single female can lay thousands of eggs (Stafford 2007).

Zolnik et al. (2015) reported that American black bears are capable of supporting I. scapularis populations because black bears, like white-tailed deer (Odocoileus virginianus), can provide a blood meal that is adequate for successful reproduction. More studies investigating the role of this

1Accepted for publication, December 2016.4Corresponding Author: jhuffman@esu.edu

57

large mammal in supporting I. scapularis populations are needed because black bear populations are increasing (PA Game Commission 2016, Williamson 2002), and therefore their role as a host of Ixodes scapularis may also become more significant.

There are few published studies that have tested ticks collected from black bears for pathogens that cause disease in humans. Leydet and Liang (2013) detected B. burgdorferi, among other pathogens, in ticks collected from black bears in Louisiana. Yabsley et al. (2009) detected Ehrlichia chaffeensis and Rickettsia parkeri in ticks collected from black bears in southern Georgia and northern Florida. Bove (2012) found Babesia spp. in I. scapularis collected from black bears in northeastern New Jersey. The objective of our study was to assess tick burdens of the American black bear in northeastern Pennsylvania and determine the prevalence of select tick-borne pathogens in collected ticks.

METHODS

Tick Collection and Identification

Ticks were collected from black bears at a PA Game Commission hunter check station in November 2015. The bears were brought by hunters from several counties in northeastern Pennsylvania: Carbon, Columbia, Lackawanna, Luzerne, Lycoming, Monroe, Montour, Schuylkill, Sullivan, and Wyoming. Each bear was searched for ticks for 10 minutes around the ears, armpits, chest, groin, and neck. Ticks were collected in plastic centrifuge tubes and stored in a cooler at the hunter check station. The ticks were identified

to species, developmental stage, and sex using pictorial keys by Keirans and Litwak (1989) and were stored at -22 °C until DNA extraction.

Determination of Tick Feeding Durations

To confirm that female Ixodes scapularis obtained a blood meal from their black bear host, the duration of feeding was determined for a subset of adult females (n = 75). Tick body length and maximum scutum width were measured using a stereomicroscope with an ocular micrometer and the scutal index or the ratio between body length and maximum scutum width was calculated. Regression equations developed by Yeh et al. (1995) were used to determine feeding durations using the calculated scutal indices.

Tick DNA Extraction and Polymerase Chain Reaction

DNA was extracted from the same subset of adult female I. scapularis (n = 75) that were measured to determine feeding durations using whole ticks and a Qiagen DNeasy Blood & Tissue Kit (Qiagen, Redwood City, CA) according to the manufacturer’s protocols. The extracted tick DNA was analyzed for B. burgdorferi, B. miyamotoi, Anaplasma spp., and Babesia spp. by PCR using specific primers for each pathogen (Table 1). Tests for Borrelia spp. and Anaplasma spp. utilized 15 µL, nested PCR, while the amplification of Babesia spp. was a 20 µL, single reaction. Borrelia spp. can be differentiated through amplicon size when amplified with the described IGS primers. The PCR product of B. burgdorferi DNA is 988 base pairs and B. miyamotoi is 547 base pairs (Scott et al. 2010).

Table 1. List of primer sequences for Anaplasma phagocytophilum, Babesia spp., Borrelia burgdorferi, Borrelia miyamotoi and the target site.

Species Primer Primer Sequence Base Pairs Target rRNA ReferenceAnaplasma spp. GE3a

GE10rF-CACATGCAAGTCGAACGGATTATTC R-TTCCGTTAAGAAGGATCTAATCTCC

932 16S Massung et al. 1998

A. phagocytophilum GE9f GE2

F-AACGGATTATTCTTTATAGCTTGCT R-GGCAGTATTAAAGCAGCTCCAGG

546 16S Massung et al. 1998

Babesia spp. BAB1 BAB4

F-CTTAGTATAAGCTTTTATACAGC R-ATAGGTCAGAAACTTGAATGATACA

238 18S Persing et al. 1992

Borrelia spp. IGS1 F-GTATGTTTAGTGAGGGGGGTG R-GGATCATAGCTCAGGTGGTTAG

1029 16S-23S Bunikis et al. 2004

B. burgdorferi IGS1 IGS2

F-AGGGGGGTGAAGTCGTAACAAG R-GTCTGATAAACCTGAGGTCGGA

988 16S-23S Bunikis et al. 2004

B. miyamotoi IGS2 F-AGGGGGGTGAAGTCGTAACAAG R-GTCTGATAAACCTGAGGTCGGA

547 16S-23S Bunikis et al. 2004

58JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

Positive controls for B. burgdorferi and B. miyamotoi were provided by M. Edwards (Muhlenberg College) and S. Telford (Tufts University) and A. phagocytophilum by J.E. Huffman (East Stroudsburg University). All negative controls used nuclease free water instead of a DNA template.

RESULTS

Four-hundred eighteen adult Ixodes scapularis were collected from 54 American black bears. No other tick species were detected, and there were no immature life stages of I. scapularis collected. Two-hundred ninety-five (70.6 %) were female and 123 (29.4 %) were male. The number of I. scapularis detected per bear was between 1 and 41 with a mean of 7.74 ± 1.21 (SE) per bear (Figure 1). Over half of the 75 measured females fed on their black bear host for over 4 days (54.7 %) (Figure 2).

None of the seventy-five females tested were positive for Babesia spp. Thirteen were positive for B. burgdorferi (17.3 %), 1 for B. miyamotoi (1.3 %), and 5 for Anaplasma spp. (6.7 %). There were no cases of coinfection.

DISCUSSION

In our study, we determined I. scapularis burdens for black bears in northeastern Pennsylvania and tick feeding durations and the prevalence of select pathogens for a subset of ticks collected from these black bears. We found, only I. scapularis and only adults of this species on the 54 black

bears searched. This lack of species and life stage diversity is consistent with other black bear ecto-parasite studies conducted during the fall in the northeastern United States (i.e. Bove 2012, Zolink et al. 2015), and also with the seasonal variations in tick species and life stage abundance (Schulze et al. 1986, Jin et al. 2012).

Zolnik et al. (2015) found that black bears, like white-tailed deer, provide blood meals sufficient for the successful production of offspring and may therefore be capable of supporting and maintaining I. scapularis populations. The tick burdens observed in our study provide further evidence that the black bear may play a role in maintaining I. scapularis populations. We found a mean tick burden of 7.74 ± 1.21 (SE) adult I. scapularis per black bear during November 2015. The mean number of ticks per bear reported represents the minimum tick burden per bear because searches were limited to 10 minutes and only specific areas of the body were examined. Zolnik et al. (2015) reported a mean of 3.07 I. scapularis per bear between late September and late October 2012 in northern New Jersey, and searched each bear for 10 minutes. Adult I. scapularis are most abundant between mid-October and early December (Schulze et al. 1986), and this seasonality in abundance, in addition to differences in study area, among other factors, likely contributed to the difference in tick burdens observed between studies.

In our study, both female and male Ixodes scapularis were collected from black bears. Although feeding durations were not computed for males, all the males appeared unfed. Troughton and Levin (2007) observed the life cycle of I. scapularis in the laboratory and found that the adult males rarely fed. Most of the females collected in our study, however, were visibly engorged. It is probable that I. scapularis mate on black bears and that the engorged females were feeding in order to lay eggs.

Figure 1. Number of Ixodes scapularis per American black bear collected post-mortem in November 2015. A mean of 7.74 ± 1.21 (SE) I. scapularis were collected per bear (418 in total from 54 bears).

Figure 2. Feeding durations of female Ixodes scapularis collected from American black bears post-mortem in November 2015 (n = 75).

59

Seventy-five I. scapularis were tested for select pathogens, and because of this low sample size, our results likely represent the minimum prevalence for each pathogen in northeastern Pennsylvania. The pathogen with the greatest prevalence was B. burgdorferi (17.3 %) followed by Anaplasma spp. (6.7 %), and B. miyamotoi (1.3 %). Babesia spp. infection was not observed in our study. During a state-wide tick pathogen study in Pennsylvania, Hutchinson et al. (2015), found no difference between male and female I. scapularis pathogen prevalence. Therefore, although only females were tested in our study, our results represent the minimum prevalence for pathogens of adult I. scapularis from northeastern PA.

Our results are consistent with other I. scapularis pathogen prevalence studies from Pennsylvania which detect B. burgdorferi most often relative to other tick-borne pathogens. Hutchinson et al. (2015) found state-wide infection rates of 47.4 % for B. burgdorferi, 3.5 % for B. microti, and 3.3 % for A. phagocytophilum. In south-central Pennsylvania, Han et al. (2014) reported a 21 % infection rate for B. burgdorferi and detected no B. microti or A. phagocytophilum. The pathogen prevalence data reported in our study is also consistent with tick-borne disease occurrence in Pennsylvania in that anaplasmosis and babesiosis are infrequent relative to Lyme disease cases (Courtney et al. 2003, Hutchinson et al. 2015). Borrelia miyamotoi, an emerging pathogen in the United States (Krause et al. 2013), was found in one I. scapularis. PCR analysis does not distinguish between A. phagocytophilum, the pathogen that causes anaplasmosis in humans and AP-Variant 1, which is not known to cause disease in humans (Courtney et al. 2003), therefore, from our analyses, we could not determine if the 5 ticks that tested positive for Anaplasma spp. carried the pathogen that causes anaplasmosis.

The geographic distributions of I. scapularis and the pathogens it transmits have increased in the eastern United States (CDC 2015b, Kugeler et al. 2015, Eisen et al. 2016). Likewise, black bear populations are increasing and their range is expanding in the eastern United States, notably into suburban areas (Williamson 2002). Animals with high tick burdens and large home ranges are most effective at dispersing ticks (Madhav et al. 2004). Black bears have larger home ranges than white-tailed deer (Whitaker and Hamilton 1998), and can likely support a greater tick load than deer.

Ixodes scapularis abundance and pathogen prevalence in ticks differ by region (Hamer et al. 2010, Hutchinson et al. 2015). In Pennsylvania, I. scapularis populations are established in every county (Hutchinson et al. 2015, Eisen et al. 2016). Several states within the black bear range, including West Virginia, Virginia, and Kentucky, however, lack I. scapularis occurrence records (Garshelis et al. 2008, Eisen et al. 2016). Black bear movement could play a role in I. scapularis expansion into these areas.

Zolnik et al. (2015) detected the DNA of A. phagocytophilum and B. microti in the blood of black bears from New Jersey, Shaw et al. (2015) reported Babesia spp. in New Jersey black bears, and Westmoreland et al. (2016) detected A. phagocytophilum in the blood of black bears from North Carolina. Although studies have detected tick-borne pathogen infections in black bears, information on the reservoir competence of the black bear is lacking and was not investigated in this study. The translocation of female I. scapularis by black bears, however, could result in the establishment or enhancement of I. scapularis populations which could increase the incidence of tick-borne diseases in humans (Hamer et al. 2010). Unlike B. burgdorferi and A. phagocytophilum, transovarial transmission of B. miyamotoi can occur (Ogden et al. 1998, Homer et al. 2000, Rollend et al. 2013). Thus, the geographic range of B. miyamotoi infections in humans could increase if an infected female I. scapularis is translocated and lays eggs where no or relatively little B. miyamotoi is present.

Characterizing the ticks present on black bears and the prevalence of pathogens in these ticks may serve as an alternative surveillance technique to track the distribution and relative abundance of I. scapularis and the human disease pathogens it transmits. Yearly, state run black bear hunter check stations and trapping efforts provide opportunities to collect ticks, from an otherwise inaccessible species. Surveillance of I. scapularis populations and the populations of other tick species, their hosts, and the pathogens they transmit, is critical in assessing human health risks and taking actions towards disease prevention.

ACKNOWLEDGEMENTS

The authors would like to thank personnel at the PA Game Commission for access to samples. This work was supported by the Northeast Wildlife DNA Laboratory at East Stroudsburg University in East Stroudsburg, PA and the Department of Biology at King’s College in Wilkes-Barre, PA.

LITERATURE CITED

Barbour A.G., J. Bunikis, B. Travinsky, A.G. Hoen, M.A. Diuk-Wasser, D. Fish, and J.I. Tsao. 2009. Niche parti-tioning of Borrelia burgdorferi and Borrelia miyamotoi in the same tick vector and mammalian reservoir species. Am. J. Trop. Med. Hyg. 81, 1120 – 1131.

Bove, D. 2012. The occurrence of tick-borne pathogens in black bears (Ursus americanus) in New Jersey. Master’s Thesis. East Stroudsburg University, East Stroudsburg, PA.

60JOURNAL OF THE PENNSYLVANIA ACADEMY OF SCIENCE Vol. 90(2), 2016

Bunikis J., U. Garpmo, J. Tsao, J. Berglund, D. Fish, and A.G. Barbour. 2004. Sequence typing reveals extensive strain diversity of the Lyme borreliosis agents Borrelia burgdorferi in North America and Borrelia afzelii in Eu-rope. Microbiol. 150: 1741–1755.

Burguess, K.I. and J. E. Huffman. 2005. Diseases of bears, Pp. 298-322, In S.K. Majumdar, J.E. Huffman, F.J. Bren-ner, and A. I. Panah, (Eds.). Wildlife diseases: Landscape epidemiology, spatial distribution, and utilization of re-mote sensing technology. Pennsylvania Academy of Sci-ence, Easton, PA, 506 pp.

CDC. 2015a. Lyme disease: Data and Statistics. Available online at http://www.cdc.gov/lyme/stats/index.html. Ac-cessed 23 March 2016.

CDC. 2015b. Morbidity and mortality weekly report (MMWR): Summary of notifiable infectious diseases. Available online at http://www.cdc.gov/mmwr/mmwr_nd/. Accessed 04 April 2016.

CDC. 2015c. Ticks: B. miyamotoi. Available online at http://www.cdc.gov/ticks/miyamotoi.html. Accessed 24 March 2016.

Courtney, J.W., R.L. Dryden, J. Montgomery, B.S. Schnei-der, G. Smith, and R.F. Massung. 2003. Molecular char-acterization of Anaplasma phagocytophilum and Borrelia burgdorferi in Ixodes scapularis ticks from Pennsylva-nia. J. Clin. Microbiol. 41: 1569 – 1573.

Daniels, T.J., R.C. Falco, K.L. Curran, and D. Fish. 1996. Timing of Ixodes scapularis (Acari: Ixodidae) oviposition and larval activity in southern New York. J. Med. Ento-mol. 33: 140 – 147.

Eisen, R.J., L. Eisen, and C.B. Beard. 2016. County-scale distribution of Ixodes scapularis and Ixodes pacificus (Acari:Ixodidae) in the continental United States. J. Med. Entomol. 53: 349 – 386.

Garshelis, D.L., D. Crider and F. van Manen, (IUCN SSC Bear Specialist Group). 2008.Ursus americanus: The IUCN red list of threatened species. Available online at http://www.iucnredlist.org/details/41687/0. Assessed on 05 July 2016.

Hamer, S.A., J.I. Tsao, E.D. Walker, and G.J. Hickling. 2010. Invasion of the Lyme disease vector Ixodes scapu-laris: Implications for Borrelia burgdoferi endemicity. EcoHealth 7: 47 – 63.

Han, G.S., E.Y. Stromdahl, D. Wong, and A.C. Weltman. 2014. Exposure to Borrelia burgdorferi and other tick-borne pathogens in Gettysburg National Military Park, south-central Pennsylvania, 2009. Vector Borne Zoonotic Dis. 14: 227 – 233.

Homer, M.J., I. Aguilar-Delfin, S.R Telford III, P.J. Krause, and D.H. Persing. 2000. Babesiosis. Clin. Microbiol. Rev. 13: 451 – 469.

Hutchinson, M.L., M.D. Strohecker, T.W. Simmons, A.D. Kyle, and M.W. Helwig. 2015. Prevalence rates of Bor-relia burgdorferi (Spirochaetales: Spirochaetaceae), Anaplasma phagocytophilum (Rickettsiales: Anaplasma-taceae), and Babesia microti (Piroplasmida: Babesiidae) in host-seeking Ixodes scapularis (Acari: Ixodidae) from Pennsylvania. J. Med. Entomol. 52: 693 – 698.

Jin, H., F. Wei, Q. Liu, and J. Qian. 2012. Epidemiology and control of human granulocytic anaplasmosis: A systemat-ic review. Vector Borne Zoonotic Dis. 12: 269 – 274.

Keirans, J. E. and T. R. Litwak. 1989. Pictorial key to the adults of hard ticks, family Ixodidae (Ixodida: Ixo-doidea), east of the Mississippi River. J. Med. Entomol. 26: 435–448.

Keirans, J.E., H.J. Hutcheson, L.A. Durden, and J.S. Klompen. 1996. Ixodes (Ixodes) scapularis (Acari: Ixo-didae): Redescription of all active stages, distribution, hosts, geographical variation, and medical and veterinary importance. J. Med. Entomol. 33: 297-318.

King J.M. 1960. Pathology, parasitology, and hematology of the black bear in New York. New York Fish and Game J. 7:99–101.

Kiszewski, A.E., F.R. Matuschka, and A. Spielman. 2001. Mating strategies and spermiogenesis in ixodid ticks. Annu. Rev. Entomol. 46: 167 – 182.

Krause, P.J., S. Narasimhan, G.P. Wormser, L. Rollend, E. Fikrig, T. Lepore, A. Barbour, and D. Fish. 2013. Human Borrelia miyamotoi infection in the United States. N. Engl. J. Med. 368: 291-293.

Kugeler, K.J., G.M. Farley, J.D. Forrester, and P.S. Mead. 2015. Geographic distribution and expansion of human Lyme disease, United States. Emerg. Infect. Dis. 21: 1455 – 1457.

Leydet, B.F., Jr. and F.T. Liang. 2013. Detection of hu-man bacterial pathogens in ticks collected from Louisi-ana black bears (Ursus americanus luteolus). Ticks Tick Borne Dis. 4: 191 – 196.

Madhav, N.K., J.S. Brownstein, J.I. Tsao and D. Fish. 2004. A dispersal model for the range of blacklegged tick (Acari: Ixodidae). J. Med. Entomol. 41:842-852.

Massung, R. F., K. Slater, J. H. Owens, W. L. Nicholson, T. N. Mather, V. B. Solberg, and J. G. Olson. 1998. Nest-ed PCR assay for detection of granulocytic ehrlichiae. J. Clin. Microbiol. 36: 1090-1095.

Ogden N.H., K. Bown, B.K. Horrocks, Z. Woldehiwet, and M. Bennett. 1998. Granulocytic ehrlichia infection in ix-odid ticks and mammals in woodlands and uplands of the U.K. Med. Vet. Entomol. 12: 423–429.

Oliver, J.H., Jr. 1989. Biology and systematics of ticks (Acari: Ixodida). Ann. Rev. Ecol. Syst. 20: 397 – 430.

61

PA Game Commission. 2016. Population growth of black bears. Available online at www.pgc.pa.gov/Wildlife/WildlifeSpecies/BlackBear/Pages/BlackBearPopula-tionGrowth.aspx. Accessed 5 December 2016.

Persing, D. H., D. Mathiesen, W. F. Marshall, S. R. Telford, A. Spielman, J. W. Thomford, and P. A. Conrad. 1992. Detection of Babesia microti by polymerase chain reac-tion. J. Clin. Microbiol. 30: 2097-2103.

Rogers, L.L. 1975. Parasites of black bears of the Lake Su-perior region. J. Wildl. Dis. 11: 189-192.

Rollend, L., D. Fish, and J.E. Childs. 2013. Transovarial transmission of Borrelia spirochetes by Ixodes scapu-laris: A summary of the literature and recent observa-tions. Ticks Tick Borne Dis. 4: 46–51

Schulze, T.L., G.S. Bowen, M.F. Lakat, W.E. Parkin, and J.K. Shisler. 1986. Seasonal abundance and hosts of Ix-odes dammini (Acari: Ixodidae) and other Ixodid ticks from an endemic Lyme disease focus in New Jersey, USA. J. Med. Entomol. 23: 105-109.

Scott, M.C., M.E. Rosen, S.A. Hamer, E. Baker, H. Ed-wards, C. Crowder, J.I. Tsao, and G.J. Hickling. 2010. High-prevalence Borrelia miyamotoi infection among wild turkeys (Meleagris gallopavo) in Tennessee. J. Med. Entomol. 47:1238–1242.

Shaw, M., N. Kolba, and J.E. Huffman. 2015. Babesia spp. in Ursus americanus (black bear) in New Jersey. North-east Nat. 22: 451-458.

Stafford, K.C., III. 2007. Tick management handbook: An integrated guide for homeowners, pest control operators, and public health officials for the prevention of tick-asso-ciated disease. Connecticut Agricultural Experiment Sta-tion. Bulletin 1010: 78 pp.

Troughton, D.R. and M.L. Levin. 2007. Life cycles of seven ixodid tick species (Acari: Ixodidae) under standardized laboratory conditions. J. Med. Entomol. 44: 732 – 740.

Westmoreland, L.S., M.K. Stoskopf, and R.G. Maggi. 2016. Prevalence of Anaplasma phagocytophilum in North Car-olina eastern black bears (Ursus americanus). J. Wildl. Dis. PMID:27479929 DOI:10.7589/2016-02-036.

Whitaker Jr., J.O. and W.J. Hamilton Jr. 1998. Mammals of the eastern United States, 3rd Edition. Cornell University Press, Ithaca, NY. 583 pp.

Williamson, D.F. 2002. In the black: Status, management, and trade of the American black bear (Ursus americanus) in North America. Available online at http://www.njfis-handwildlife.com/pdf/bear/policy_lit/williamson_traffic.pdf. Accessed 13 March 2016.

Yabsley, M. J., T. N. Nims, M. Y. Savage, and L.A. Durden. 2009. Ticks and tick-borne pathogens and putative sym-bionts of black bears (Ursus americanus floridanus) from Georgia and Florida. J. Parasitol. 95: 1125-1128.

Yeh, M., J.M. Bak, R. Hu, M.C. Nicholson, C. Kelly, and T.N. Mather. 1995. Determining the duration of Ixodes scapularis (Acari: Ixodidae) attachment to tick-bite vic-tims. J. Med. Entomol. 32: 853 – 858.

Yuval, B. and A. Spielman. 1990. Duration and regulation of the developmental cycle of Ixodes dammini (Acari: Ix-odidae). J. Med. Entomol. 27: 196 – 201.

Zolnik, C. P., A. M. Makkay, R. C. Falco, and T. J. Dan-iels. 2015. American black bears as hosts of blacklegged ticks (Acari: Ixodidae) in the northeastern United States. J. Med. Entomol. 52: 1103-1110.