Development of Serum and Intestinal Antibody Response to Rotavirus After Naturally Acquired Rotavirus Infection In Man

Transcription

1 Journal of Medical Virology 8: (1981) Development of Serum and Intestinal Antibody Response to Rotavirus After Naturally Acquired Rotavirus Infection In Man Marie Riepenhoff Talty, Sara Bogger Goren, Philip Li, Patricia J. Carmody, Helen J. Barrett, and Pearay L. Ogra Department of Pediatrics (M.R.-T., S.B.-G., P. L., P.J. C., H.J. B" P. L.0.) and Department of Microbiology (M.R.- T., P.L.O.), State University of New York at Buffalo, School of Medicine and Division of Clinical Infectious Diseases and Virology, Children's Hospital, Buffalo The temporal characteristics of the response of rotavirus specific IgM, IgG, 19A in serum and secretory antibody in feces to rotavirus were studied in 77 hospitalized patients with rotavirus induced gastroenteritis. The response in serum was characterized by the sequential appearance of rotavirus specific IgM, IgG, and 19A antibody. The IgM antibody appeared to be higher in the acute phase of the disease and was subsequently replaced by the IgG and 19A antibodies. However, the titers of IgG rotavirus antibody in convalescent specimens of serum were found to be statistically significantly lower in patients with severe or prolonged rotavirus infection than in specimens from subjects with mild or moderate disease. Most fecal specimens collected during both the acute and convalescent phase of illness contained virus specific secretory 19A. Higher concentrations of antibody were measured in convalescent samples from patients with prolonged diarrhea and virus shedding. These observations suggest a possible relationship between the severity of rotavirus infection and the nature of systemic and secretory antibody response. Key words: rotavirus, serum and intestinal antibody response, rotavirus gastroenteritis Accepted for publication August 24, Address reprint requests to Marie Riepenhoff-Talty, PhD, Children's Hospital, Division of Infectious Diseases and Virology, 219 Bryant Street, Buffalo, NY /81/ $ Alan R. Liss, Inc.

2 216 Riepenhoff-Talty et al INTRODUCTION Human rotavirus (HRV) is an important cause of diarrheal disease in infants and young children. Over of children 3 years of age have rotavirus specific serum antibody, which persists into the adult life [Jesudoss, 1978; Kapikian, 1975; Yolken, 1978]. Whereas rotavirus infection in adults is generally asymptomatic or mild, the presence of specific antibody in the serum alone has not been shown to prevent the mucosal reinfection [Wenman, 1979], and human and animal studies have shown that local antibody in the gut may be more directly responsible for protection. The facts that substantial quantities of rotavirus specific IgA have been found in human breast milk, and breast-fed babies appear to be infected less frequently than bottle fed babies, raise the possibility that rotavirus specific secretory IgA may be important in determining the outcome of rotavirus infection [Cameron 1978, Totterdell, 1976]. Although earlier studies have demonstrated the appearance of rotavirus specific antibody in serum, as well as, intestinal secretions [Blacklow, 1976; Wenman, 1979], the precise role of the immune response in the mechanism of the pathogenesis of or protection against rotavirus infection remains to be determined. The present studies were initiated to examine the temporal sequence of serum and secretory antibody response to rotavirus after naturally acquired infection in hospitalized infants and young children. In addition, an attempt was made to explore the nature of the immune response relative to severity of clinical disease in this patient population. MATERIALS AND METHODS Study Population Seventy-seven children hospitalized with rotavirus gastroenteritis were included in the study. The patient population was selected at random and consisted of 45 males and 32 females. Twenty-seven patients were under 6 months of age, 23 were 6 to 12 months; 19 were 13 to 24 months and 8 were over 2 years of age. The mean age was 12.6 months. Diarrheal illness was classified as mild-moderate, severe, or prolonged. The parameter for placing a child in a category included frequency and severity of various symptoms, physical signs, and biochemical abnormalities, indicated in Table I. There were 42 patients with mild-moderate disease, 23 with severe disease, and 12 subjects with diarrhea persisting longer than 2 weeks. Specimen Collection The diagnosis of rotavirus enteritis was made by electron microscopic examination of stool specimens collected soon after admission to the hospital (see below). Specimens of serum and feces were collected subsequently and tested for the presence of rotavirus and the appearance of specific antibody. Multiple specimens were collected during the acute and convalescent phase of the infection in 53 patients. However, in some patients adequate volume of sample was not available to make determinations for antibody activity in all immunoglobulin isotypes. As a result, all antibody data are presented in terms of different specimens relative to the time of collection rather than individual patients.

3 TABLE I. Scoring for Severity of Disease* Antibody Response to Rotavirus Infection 217 Symptom Symptom/score Symptom/score Symptom/score Diarrhea No. of stools per day Duration of diarrhea in days <4 = I 1-4 = 1 5 = = 2 >8 = = 3 Vomiting No. per day Duration in days 1-3 = I 2 = = = 2 >6 = 3 >5 = 3 Dehydration In degree <5070 = 1 2:5070 = 2 2: = 3 Fever Degree in centigrade <38.5 = 1 2:38.5 = 2 2:39 = 3 Acidosis Electrolyte imbalance "Scores were graded as follows: Severity Total score Mild-moderate 0-12 Severe Prolonged = Diarrhea for 2: 14 days = 2 <7.30 = 3 POS = 3 Electron Microscopic Examination of Fecal Samples A small amount of feces, mixed with 2070 ammonium acetate, was placed on a carbon-formvar coated copper mesh grid (Ernest Fullam Co., Schenectady, NY). After removal of excess sample with filter paper, a drop of 1% phosphotungstate (PTA), ph 7.2, was added to the grid. After approximately 1 minute the excess PTA was removed and the grid was placed 2 inches from an ultraviolet light for 3 to 5 min to inactivate viral infectivity. The grids were examined at X 30,000 using a JEOL loos electron microscope (leol, Tokyo, Japan). Preparation of Fecal Samples for Antibody Testing Fecal specimens were diluted 1:4 in phosphate buffered saline (PBS), mixed, and clarified by low speed centrifugation. The resultant supernatant fluid was filtered through a 0.2 p.m filter (Millipore Corp, Bedford, ME) and heat inactivated at 56 C for 30 min and stored at - 20 C until used in assays. All samples were tested for immunoglobulins and processed further for determination of rotavirus specific antibody by immunofluorescence. Indirect Immunofluorescence for Rotavirus Specific IgM, IgG, IgA, and Secretory IgA Monolayers of Buffalo Green Monkey (BGM) kidney cell culture were infected with Lincoln strain of Nebraska Calf diarrhea virus (NCDV). Smears of infected and uninfected (control) cell cultures, prepared on glass slides, were acetone fixed and incubated with serial twofold dilutions of patient serum or fecal extracts

4 218 Riepenhoff-Talty et al for 30 min at 37 C. After extensive washing, the smears were incubated with either fluorescein isothiocyanate (FITC) conjugated goat anti-human IgG, IgM (Meloy Chem, Springfield, VA) or secretory component (SC) or IgA (Accurate Chemical Co, Hicksville, NY) for 30 min at 37 C. After repeated additional washings with PBS the slides were examined for virus specific immunofluorescence employing a Leitz ortholux microscope (Wetzlar, Germany) using a B36-12 excitation filter. In order to rule out nonspecific reactions, appropriate controls and blocking experiments were carried out. Positive and negative control sera were included in each assay. Colostrum, containing secretory IgA to rotavirus, (titer 2': 64) was included as a positive control in the assays testing fecal extracts for rotavirus secretory IgA. Hyperimmune goat anti-rotavirus sera (kindly supplied by R. Yolken) was used to block specifically the fluorescence of the rotavirus infected BGM cells. The antibody activity in fecal samples was determined after standardization of each sample to 10 mg/dl of immunoglobulin isotype. Immunoglobulin Quantitation Immunoglobulin levels in feces were determined by the radial immunodiffusion technique (LC-Partigen Ig Kit, Behring Diagnostics, Somerville, NJ). The immunoglobulins in the serum were quantitated by the nephelometric method [Killingworth 1972; Ritchie 1967). RESULTS Relationship of Virus Shedding and Diarrhea The mean duration of virus excretion in feces after onset of symptoms was approximately 11 days (Fig. 1). However, some patients shed the virus for only 2 to 3 days after the onset of symptoms or as long as 30 days. An association between resolution of diarrhea and cessation of viral shedding was observed, and as a rule the diarrhea stopped within 2 to 3 days after the cessation of viral shedding. Concentrations of Rotavirus Specific Antibody in Serum and Feces The antibody responses in serum and feces expressed as GMTs of all the major classes of rotavirus specific antibody found in serum or feces are summarized in Table II. Convalescent titers include all specimens obtained from 2 to 16 weeks after the onset of symptoms. Predictably, the serum and fecal IgM titers were highest during the acute phase and declined during convalescence. Low levels of serum and fecal IgO and IgA rotavirus antibody activity were observed during acute phase of the illness and with increasing titer of antibody during convalescence (Table II). Relationship of Antibody Response to the Severity of Illness The local and systemic rotavirus antibody responses relative to the severity of disease are in Figure 2. In all patients regardless of severity of illness, the serum IgM antibody responses were similar. However, fecal IgM antibody levels in acute phase samples appeared to be three- to fourfold higher (P < 0.01) in patients with mild disease. No significant differences were observed in fecal IgM levels in convalescent samples or in levels of serum IgG in acute phase samples from patients with varying degrees of illness. Of 49 acute samples, 22 contained specific IgO.

5 30 Antibody Response to Rotavirus Infection ~.. 5.r Day ofcessation of Diarrhea Fig. I. Length of diarrhea compared to virus shedding. Each point represents one subject and depicts the day on which the diarrhea was resolved and the fecal sample was negative for rotavirus. Significant differences in IgO levels of convalescent phase serum samples were observed primarily between patients with mild to moderate disease and patients with severe disease. The mean serum IgO antibody titer was 22 in patients with mild-moderate disease compared to 3 in patients with severe disease (P <0.001) (Fig. 2). Only (22 of 82) fecal samples tested contained detectable levels of specific IgO. Appreciable levels of specific IgA antibody were observed in convalescent serum and fecal samples of all patients regardless of severity or chronicity of disease. However, highest fecal IgA concentrations were in the convalescent samples from patients who had diarrhea for 2 or more weeks. In order to determine if lower levels of specific rotavirus serum IgO were simply an indication of lower total levels of serum IgO in patients with protein loss due to severe intestinal wall damage, all samples were assayed for total levels of immunoglobulins. Table III compares the age related differences in serum levels between patients with mild-moderate, severe, or prolonged rotavirus induced gastroenteritis. Although the serum IgO immunoglobulin content appeared to be somewhat lower in all patients, no significant differences were observed in the IgO content between patients with different severity of clinical disease. DISCUSSION Mild gastroenteritis due to rotavirus can occur in adults and neonates in the presence of virus specific serum antibody [Wenman, 1979; Totterdell, 1976]. However, the reported incidence of rotavirus reinfection is relatively rare, and it is reasonable to assume that primary infection results in some form of specific protection. The role of intestinal or local antibody in the protection from or limita-

6 220 Riepenhoff-Talty et at J IgM 64r j '~l±h~lil~l 0 64~N'86 11 JI -11 J _ e_ -e-. <' L _ IgG li~t 16 N'88 qmo 11.~~~~~* ~ ~ I- <2 ~ ~ -er u, IgA l~kljlj:l~j 101 l~~t jl: J= l N't38 ~ 8 ~ !!! ~_ < I11III Mild- Severe Prolonged Mild- Severe Prolonged Moderote Acute *EJ,pressed per 10mgof Immunoglobulin per dl Moderote Convalescent Fig. 2. Geometric mean titer, as indicated by the bar, of classes of antibody to rotavirus detected in serum and feces. Subjects were classed as mild-moderate, severe, or prolonged (see Table I). tion of rotavirus infection is not well understood. We have attempted to characterize the serum and intestinal antibody responsesin terms of antibody classes relative to the severity of disease. Antibody response in the serum was consistent with those typically observed in other natural viral infections with early development of IgM and later development of IgG and IgA. Previous reports have described the development of antibody to rotavirus in feces [Watanabe, 1978], with the demonstration of both specific IgG and IgA from children shedding the virus. IgA was most often found

7 Antibody Response to Rotavims Infection 221 TABLE II. Levels of Rotavirus Specific Antibody IgM, IgG, and IgA in Serum and Fecal Specimens Obtained After Rotavirus Infection IgM IgG IgA No. Antibody titer" No. Antibody titer" No. Antibody titer" Specimen specimens Acute Convalescent Specimens Acute Convalescent specimens Acute Convalescent Serum 80 5 ± ± ± ± ± ± 0.3 Fecesv 86 7 ± 3 2 ± <2 2 ± ± ± 0.4 "Geometric mean titer ± SE. bantibody content in fecal specimens was expressed as 10 mg of Ig/d\. TABLE III. Age-Related Differences in Levels of Serum in IgG Patients With Rotavirus Enteritis Age range 1-3 mos. 4-6 mos mos mos. -24 mos. Concentration of serum IgG" I mg I dl No. Mild I No. No. samples moderate samples Severe samples Prolonged Age matched normal levels a ± SD for all subjects in that age group. bsingle sample. "Indicates significant difference. 301 ± b ± b ± 299" ± ± 216" ± 21" ± 61" ± 226" 323b 642 ± ± ± 40" 430 ± ± ± ± ± 183 coating the rotavirus present in the feces. In the present study, the highest concentrations of virus specific secretory IgA were observed in specimens from patients with prolonged disease. It appeared that the presence of secretory IgA alone did not limit the duration of virus shedding or of diarrhea. Possibly the overwhelming amount of rotavirus antigen produced in this infection is greater than the neutralizing capacity of the secretory IgA in the gut. However, no information is available regarding the nature of activity and response of this antibody relative to the severity of disease. An interesting finding in the present study was the reduced serum IgG antibody response to rotavirus in association with severe clinical disease. The mechanism for virus infection related decreases in IgG in patients with severe or prolonged rotavirus enteritis and a brisk secretory IgA response in the feces in patients with prolonged diarrhea cannot be explained on the basis of the data presented. It is conceivable that during the acute stages of mucosal inflammation there is considerable loss of immunoglobulin through the intestinal lumen. Although the immunoglobulin G levels were generally lower in the entire patient population, the individual and mean serum IgG concentration in patients with severe or prolonged disease were similar to those with relatively mild disease. In addition, no significant differences were observed in levels of other serum immunoglobulins; therefore, increased intestinal loss of IgG on a selective basis would be highly unlikely. It is possible that continued antigenic stimulation may induce suppressive mechanisms for specific IgG synthesis. This possibility is supported by the findings of prolonged virus shedding and increased IgA response in the intestine in

8 222 Riepenhoff-Talty et al patients with reduced IgG antibody response in the serum. Although viral infections have been known to induce varying degrees of alterations in the regulating mechanisms of immune response [Dolin, 1973; Kantzler, 1974; Mangi, 1974; Munyer, 1975], the relative potential for the generation of virus specific suppression or help during active rotavirus infection has not been clearly defined. ACKNOWLEDGMENTS These studies were supported in part by grantsfor the National Institute of Allergy and Infectious Diseases (AI-15939), National Heart, Lung, and Blood Institute, (HL-21829), and Biomedical Research Grant Support (BRS No. 2838) for United States Public Health Service. REFERENCES Barron AI, Olshevski C, Colin MM (1970): Characteristics of BGM line of cells from African green monkey kidney. Archives Gesamte Virusforsch 32:389. Blacklow NR, Echeverria P, Smith DH (1976): Serological studies with reovirus-like enteritis agent. Infection and Immunity 13: Cameron DJS, Bishop RRF, Veenstra AA, Barnes GL (1978): Noncultivable viruses and neonatal diarrhea: Fifteen month survey in a newborn special care nursery. Journal of Clinical Microbiology 8: Dolin R, Reichman RC, Fauci AS (1973): Lymphocyte population in acute viral gastroenteritis. Infection and Immunity 14: Jesudoss ES, John JT, Mathan M, Spence L (1978): Prevalence of rotavirus antibody in infants and children. Indian Journal of Medical Research 68: Kantzler GB, Lauteria SF, Cusumano CL, Lee JD, Granguly R, Walman RH (1974): Immunosuppression during influenza virus infection. Infection and Immunity 10: Kapikian AZ, Cline WL, Mebus CA, Wyatt RG, Kalika AR, James HD, Van Kirk D, Chanock RM, Kim HW (1975): New complement-fixation test for the human reovirus-like agent of infantile gastroenteritis: Nebraska Calf diarrhea virus used as antigen. Lancet I: Killingworth LM, Savory J (1972): Manual of nephelometric methods for immunochemical determination of immunoglobulins IgG, 19A, and IgM in human serum. Clinical Chemistry 18: Mangi RJ, Dwyer JM, Niederman JC, Kantor FS (1974): Decreased circulating T cells during viral pharyngitis. Annals of Internal Medicine 81: Munyer TP, Mangi RJ, Dolin T, Kantor FS (1975): Depressed lymphocyte function after measles and mumps and rubella vaccination. Journal of Infectious Diseases 132: Ritchie RF (1967): A simple, direct and sensitive technique for measurement of specific protein in a dilute solution. Journal of Laboratory and Clinical Medicine 70: Totterdell BM, Chrystie IL, Banatvala JE (1976): Rotavirus infections in a maternity unit. Archives of Diseases of Childhood. 51: Watanabe H, Gust ID, Holmes I (1978): Human rotavirus and its antibody: Their coexistence in feces of infants. Journal of Clinical Microbiology 7: Wenrnan WM, Feltham Hinde D, Gurwith M (1979): Rotavirus infection in adults. New England Journal of Medicine 301: Yolken RH, Wyatt RG, Zissis G, Brand CD, Rodiguez WN, Kim HW, Parrott RH, Urrutia JJ, Mata L, Greenberg HB, Kapikran AZ, Chanock (1978): Epidemiology of human rotavirus types I and 2 as studied by enzyme-linked immunosorbent assay. New England Journal of Medicine 299: