Rotavirus Infection of Infant and Young Adult Nonobese Diabetic Mice Involves Extraintestinal Spread and Delays Diabetes Onset▿

  1. Barbara S. Coulson*
  1. Department of Microbiology and Immunology, The University of Melbourne, Melbourne, Victoria 3010, Australia
  1. FIG. 1.

    Detection of rotavirus in the intestine, liver, and pancreas (A) and in blood cells, serum, and spleen (B) of male and female infant NOD mice inoculated orally with RRV. RRV was detected by culture followed by EIA and by direct EIA. In panel B, infectious RRV was detected by culture then EIA in blood cells and spleen. Blood cells that contained infectious RRV also were positive for RRV antigen by direct EIA, and no other blood cell samples contained RRV antigen detectable by EIA. Rotavirus antigen was detected in serum by direct EIA, but infectious RRV was not detected in serum by culture then EIA. The number of mice examined at each day postinfection for each sample type (when these differed between sample types) is indicated above each panel.

  2. FIG. 2.

    Detection of rotavirus in stools and organs of young adult NOD mice inoculated with RRV. Approximately equal numbers of female and male mice were infected orally (A) or i.p. (B). RRV was detected by culture followed by EIA and by direct EIA. The number of mice examined at each day postinfection for each sample type (when these differed between sample types) is indicated above the panel.

  3. FIG. 3.

    RRV antigen was detected in cells outside islets in the pancreas of infant NOD mice and colocalized with resident and nonresident macrophages. Pancreases were dissected at 5 days postinfection from mice infected orally at 5 days of age. In panel A, OCT-fixed, frozen pancreatic sections were stained with rotavirus antibody-negative control rabbit serum (I and V), rabbit antiserum to RRV (II and VI), or anti-insulin antibody (III and VII). The merged images of panels II and III and VI and VII, provided in panels IV and VIII, respectively, show no colocalization of rotavirus antigen and insulin. In panel B, pancreatic sections were stained with rabbit antiserum to RRV (IX and XII) or rat antibody to the mouse macrophage marker F4/80 (X and XIII). The merged images of panels IX and X and XII and XIII are shown in panels XI and XIV, respectively. The large arrows show examples of macrophages containing rotavirus antigen, and the small arrow indicates a cell containing rotavirus antigen that did not stain as a macrophage. Panels XV, XVI, and XVII are magnifications at ×2.2 of the boxed areas in panels XII, XIII and XIV, respectively. Original magnification, ×400.

  4. FIG. 4.

    Modulation of diabetes development by RRV infection in infant and young adult NOD mice. Female mice were infected with RRV or mock infected as infants (A and C) or young adults (B and D) orally (A and B) or i.p. (C and D) and monitored for diabetes development until 30 weeks of age. The number of mice in each experimental group (n) is indicated in the legend to each panel. As assessed by seroconversion to RRV, all mock-infected mice were free of rotavirus infection during the course of the experiment. All RRV-inoculated mice seroconverted to RRV.

  5. FIG. 5.

    Effect of oral RRV infection on pancreatic insulitis development in female infant NOD mice. The histopathology in representative sections of pancreas dissected from mice 12 days or 12 weeks after oral inoculation at 5 days of age is illustrated. Pancreases from RRV-infected and control mice that had not developed hyperglycemia at these times were fixed in Bouin's solution, sectioned, and stained with hematoxylin and eosin. UI, uninfiltrated islet; Peri, peri-islet infiltrate; Intra, intraislet infiltrate. Definitions of these states are given in Materials and Methods. Pancreatitis or pancreatic damage was not seen in any section examined. Original magnification, ×200.

| Table of Contents