Director, Clinical Microbiology Laboratory
Brigham and Women's Hospital

andrew.onderdonk@channing.harvard.edu

Our research encompasses several areas related to human microbial flora and its role in health and disease. These areas include the pathogenesis of anaerobic microorganisms, the in vivo and in vitro efficacy of antimicrobials, normal microflora as it relates to the establishment and maintenance of a healthy ecosystem, and immunomodulators that may be used independently or in conjunction with existing therapeutic modalities in the prevention and treatment of serious bacterial sepsis.

Previous studies in this laboratory have reported the relative efficacy of a variety of antimicrobial regimens in preventing abscesses associated with experimental intraabdominal sepsis (IAS). The experimental animal model used was designed to simulate IAS as seen in patients with severe intraabdominal infections associated with accidental fecal contamination of the peritoneal cavity. The model system employs Wistar rats implanted with gelatin capsules containing an inoculum of pooled fecal contents obtained from meat-fed rats plus 10% w/v barium sulfate. This inoculum simulates the intestinal flora normally present in the human large intestine. Sequential assessment of animals implanted in this manner reveals two distinct stages in the infectious process. During the initial 2- to 5-day period, animals develop an acute peritonitis that is associated with a mortality of 40% to 60%. Surviving animals uniformly develop intraabdominal abscesses as shown by examination at necropsy 7 to 10 days after implantation. During the early (peritonitis) stage, Escherichia coli is one of the most common and numerically dominant isolates from the peritoneal fluid and blood cultures. In contrast, bacteriologic examination of abscesses reveals that obligate anaerobes, including Bacteroides fragilis (BF), are the dominant microflora during the second, chronic, stage of the disease. The data obtained from these earlier studies indicated that facultative species, such as E. coli, are responsible for the initial, often fatal, first phase of the disease process, while obligate anaerobes are a prerequisite for the formation of abscesses. Additional studies have included efforts to define experimental infection by using selective supression of the various bacterial components with specific antimicrobial agents. The occurence of mortality and abscesses in surviving animals are the criteria used for evaluation of the acute and chronic phases of this disease, respectively, for all studies involving this model. In addition to supporting the previous microbiologic observations regarding the role of the various microbial components during the disease process, these studies indicated that effective antimicrobial therapy could be achieved with a variety of antimicrobial agents, including the regimen of clindamycin plus gentamicin, which has been used as a benchmark against which other regimens have been compared. Therefore, by using this animal model of IAS, we can determine the relative efficacy of various antimicrobial agents and compare such information with known in vitro susceptibility data and with clinical trial information. Our group is currently collaborating in studies examining the mechanisms of abscess induction by BF, the most frequently isolated obligate anaerobe from intraabdominal infection. Studies from this laboratory have shown that the capsular polysaccharide, which represents this organism's principle virulence factor and confers immunity in experimental models of infection, is composed of two distinct surface-expressed polysacharrides that exist in ionically complex aggregates. Current studies are underway to determine the relationship of these sugars with regard to virulence and immunity of BF in experimental models of IAS.

As part of an ongoing research project examining the effect of various products on the vaginal microflora as it relates to health and disease, we perform quantitative and qualitative bacteriologic studies on samples obtained after use of various catemenial products and douche solutions. These studies are designed to detect changes in the vaginal microflora that may be important for appropriate product use. From these studies we have gathered a large volume of in vitro data that will subsequently be used as a guide to develop an in vitro experimental model, or models, of the vaginal ecosystem. Ultimately, a mathematical model will be formulated using information from both in vivo and in vitro studies that will allow us to predict the outcome of different stresses on the vaginal ecosystem. Currently we are using continuous culture to establish a mixed culture of commonly isolated vaginal isolates to represent the microflora of the vaginal ecosystem. The concentration of each organism, the ratios of organisms to each other, production of metabolic products, and other characteristics of the in vitro system will be compared with the findings of in vivo studies to establish the accuracy of the in vitro model.

A substantial amount of circumstantial evidence suggest that bacteria may contribute to adverse pregancy outcomes. All healthy women (both pregnant and nonpregnant) have microorganisms within the vagina, however, it is not known whether some of these microorganisms contribute to adverse outcomes such as preterm delivery. My laboratory is currently applying our statisitical modeling techniques and quantitative microbiology methods to a study of preterm births. The purpose of the study is to see whether there is any association between concentrations of microorganisms in the vagina of pregnant women, concertrations of lipase and phopholipase potentially produced by vaginal bacteria, and preterm delivery. This NIH-funded study is designed to determine the role of bacteria in preterm birth by using quantitative microbiologic data and statistical modeling to identify specific risk factors associated with preterm birth. The first two years of the study are devoted to subject recruitment and microbiologic analysis of vaginal swab samples obtained at 20 and 30 weeks of gestation. Subjects are placed in one of two high risk cohorts for preterm birth or into a cohort of subjects not considered at risk for preterm birth. Following collection of data from the three study cohorts, all information will be used to build a statistical model identifying specific microbial risk factors for preterm birth. Subsequent intervention studies will be based on the findings.

Much effort has been directed towards finding an etiology for inflammatory bowel disease. Pouchitis has been described as an inflammatory condition associated with the surgical construction of ileal reservoirs for patients with ulcerative colitis. We have undertaken the examination of tissue biopsy samples from patients with and without ileal pouches to evaluate the bacteria that are closely associated with or adherent to tissue biopsy samples from the ileal pouches of patients with and without pouch inflammation.

Ongoing research has identified the potent biologic activity for a variety of charged polysaccharide molecules derived from bacteria and other sources. These uniquely charged molecules, containing both positive and negatively charged substituent groups, are capable of modulating host immune response. When systemically administered material is given to experimental animals, it has been shown that the lethal effects of peritoneal sepsis can be abrogated. Protection against lethal infection is related to specific upregulation of specific cytokines. Because these charged polysaccharides are capable of both upregulating and downregulating host immune response, their potential role as therapeutic agents for a variety of inflammatory processes is being explored.

Pantosti A, Tzianabos AO, Onderdonk AB, Kasper DL. Immunochemical characterization of two surface polysaccharides of Bacteroides fragilis. Infect Immun. 1991 Jun;59(6):2075-82. [abstract]

Onderdonk AB, Dvorak AM, Cisneros RL, McLeod RS, Antionoli D, Silen W, Blair JE, Monahan-Earley RA, Cullen J, Cohen Z. Microbiologic assessment of tissue biopsy samples from ileal pouch patients. Clin Microbiol. 1992 Feb;30(2):312-7. [abstract]

Onderdonk AB, Cisneros RL, Hinkson P, Ostroff G. Anti-infective effect of poly-beta 1-6-glucotriosyl-beta 1-3-glucopyranose glucan in vivo. Infect Immun. 1992 Apr;60(4):1642-7. [abstract]

Geshnizgani AM,Onderdonk AB. Defined medium simulating genital tract secretions for growth of vaginal microflora. J Clin Microbiol. 1992 May;30(5):1323-6. [abstract]

Tzianabos AO, Onderdonk AB, Rosner B, Cisneros RL, Kasper DL. Structural features of polysaccharides that induce intra-abdominal abscesses. Science. 1993 Oct 15;262(5132):416-9. [abstract]

Ross RA, Lee ML, Delaney ML, Onderdonk AB. Mixed-effect models for predicting microbial interactions in the vaginal ecosystem. J Clin Microbiol. 1994 Apr;32(4):871-5. [abstract]

Tzianabos AO, Onderdonk AB, Smith RS, Kasper DL. Structure-function relationships for polysaccharide-induced intra-abdominal abscesses. Infect Immun. 1994 Aug;62(8):3590-3. [abstract]

Tzianabos AO, Onderdonk AB, Zaleznik DF, Smith RS, Kasper DL. Structural characteristics of polysaccharides that induce protection against intra-abdominal abscess formation. Infect Immun. 1994 Nov;62(11):4881-6. [abstract]

Onderdonk AB. Efficacy of trovafloxacin (CP-99-219), a new fluoroquinolone, in an animal model of intraabdominal sepsis. Infect Clin Pract 1996;5:S1-S3. No abstract available.

Qu Z, Ling PR, Tahan SR, Sierra P, Onderdonk AB, Bistrain BR. Protein and lipid refeeding changes protein metabolism and colonic but not small intestinal morphology in protein-depleted rats. J Nutr. 1996 Apr;126(4):906-12. [abstract]

Gibson FC III, Tzianabos AO, Onderdonk AB. The capsular polysaccharide complex of Bacteroides fragilis induces cytokine production from human and murine phagocytic cells. Infect Immun. 1996 Mar;64(3):1065-9. [abstract]

Orni-Wasserlauf R, Onderdonk AB, Winkelman J. Confirmation by chart review of cost savings from application of exclusion criteria to urine culture requests. Lab Med 1996;27:829-832. No abstract available.

Delaney ML, Onderdonk A. Evaluation of the AnaeroPack system for growth of clinically significant anaerobes. J Clin Microbiol. 1997 Mar;35(3):558-62. [abstract]

Gibson FC III, Onderdonk AB, Kasper DL, Tzianabos AO. Cellular mechanism of intraabdominal abscess formation by Bacteroides fragilis. J Immunol. 1998 May 15;160(10):5000-6. [abstract]

Fridkin SK, Yokoe DS, Whitney CG, Onderdonk A, Hooper DC. Epidemiology of a dominant clonal strain of vancomycin-resistant Enterococcus faecium at separate hospitals in Boston, Massachusetts. J Clin Microbiol. 1998 Apr;36(4):965-70. [abstract]

Onderdonk AB, Richardson JA, Hammer RE, Taurog JD. Correlation of Cecal Microflora of HLA-B27 Transgenic Rats with Inflammatory Bowel Disease. Infect Immun. 1998 Dec;66(12):6022-3. [abstract]

Onderdonk AB. Intestinal microflora and inflammatory bowel disease. In: Kirsner JB, editor. Inflammatory Bowel Disease. Philadelphia. W.B. Saunders Company;1999. p144-151. No abstract available.

Onderdonk AB, Gibson FC III, Tzianabos AO, Kasper DL. Protection against abscess formation in a model for intraabdominal sepsis. Kluwer Academic Press. Sepsis 1999; 4:311-315. No abstract available.

Tzianabos AO, Cisneros RL, Gershkovich J, Johnson J, Miller RJ, Burns JW, Onderdonk AB. Effect of surgical adhesion reduction devices on the propagation of experimental intra-abdominal infection. Arch Surg. 1999 Nov;134(11):1254-9. [abstract]

Bates DW, Kuperman GJ, Rittenberg E, Teich JM, Fiskio J, Ma'luf N, Onderdonk A, Wybenga D, Winkelman J, Brennan TA, Komaroff A, Tanasijevic M. A randomized trial of a computer-based intervention to reduce utilization of redundant laboratory tests. Am J Med. 1999 Feb;106(2):144-50. [abstract]

Ross RA, Onderdonk AB. Production of toxic shock syndrome toxin 1 by Staphylococcus aureus MN8 requires both oxygen and carbon dioxide. Infect Immun. 2000 Sep;68(9):5205-9. [abstract]