Persistent human bocavirus 1 infection and tonsillar immune responses

Abstract Background Persistent human bocavirus 1 (HBoV1) infection is a common finding in patients suffering from chronic tonsillar disease. However, the associations between HBoV1 infection and specific immune reactions are not completely known. We aimed to compare in vivo expression of T‐cell cytokines, transcription factors, and type I/III interferons in human tonsils between HBoV1‐positive and ‐negative tonsillectomy patients. Methods Tonsil tissue samples, nasopharyngeal aspirate (NPA), and serum samples were obtained from 143 immunocompetent adult and child tonsillectomy patients. HBoV1 and 14 other respiratory viruses were detected in NPAs and tonsil tissues by polymerase chain reaction (PCR). Serology and semi‐quantitative PCR were used for diagnosing HBoV1 infections. Expression of 14 cytokines and transcription factors (IFN‐α, IFN‐β, IFN‐γ, IL‐10, IL‐13, IL‐17, IL‐28, IL‐29, IL‐37, TGF‐β, FOXP3, GATA3, RORC2, Tbet) was analyzed by quantitative reverse‐transcription (RT)‐PCR in tonsil tissues. Results HBoV1 was detected by PCR in NPA and tonsils from 25 (17%) study patients. Serology results indicated prior nonacute infections in 81% of cases. Tonsillar cytokine responses were affected by HBoV1 infection. The suppression of two transcription factors, RORC2 and FOXP3, was associated with HBoV1 infection (p < 0.05). Furthermore, intratonsillar HBoV1‐DNA loads correlated negatively with IFN‐λ family cytokines and IL‐13. Conclusions Our study shows distinctively decreased T‐helper17 and T‐regulatory type immune responses in local lymphoid tissue in HBoV1‐positive tonsillectomy patients. HBoV1 may act as a suppressive immune modulator.

Results: HBoV1 was detected by PCR in NPA and tonsils from 25 (17%) study patients. Serology results indicated prior nonacute infections in 81% of cases.

Conclusions:
Our study shows distinctively decreased T-helper 17 and T-regulatory type immune responses in local lymphoid tissue in HBoV1-positive tonsillectomy patients. HBoV1 may act as a suppressive immune modulator.

K E Y W O R D S
cytokine, forkheadbox protein 3, FOXP3, human bocavirus, RAR-related orphan receptor C 2, RORC2, T-helper 17 , T-regulatory, tonsil, transcription factor 1 | BACKGROUND Human bocavirus 1 (HBoV1) is a non-enveloped single-stranded DNA virus belonging to the Parvoviridae family. 1 It was originally discovered in 2005 in respiratory-tract specimens. 2 HBoV1 is typically found in such specimens from young children suffering with lower respiratory-tract infections such as bronchiolitis, wheezing, asthma, or pneumonia. 1,3 The diagnosis of acute bocavirus infection is based on quantitative polymerase chain reaction (PCR), expression of HBoV1 messenger RNA (mRNA), or serology, as the virus DNA may persist for weeks, even months. 1,[3][4][5][6] Interestingly, a few studies have found HBoV1 DNA in tonsil squamous cell carcinoma tumors, inducing speculations of a possible causal association. 7,8 Tonsils in the oropharynx are in contact with inhaled and ingested allergens and pathogens. Viral infection detected by PCR is a common finding in tonsil tissue of routine tonsillectomy patients. 9 The prevalence of HBoV1 DNA in fairly asymptomatic tonsillectomy patients suffering chronic tonsillar disease has been 4%-27% in the tonsil tissue and 11%-18% in the nasopharyngeal aspirate (NPA) samples. 6,[9][10][11][12][13] In our earlier study, the tonsillar cytokine expressions were associated with viral infections in general and clinical characteristics. 14 Moreover, HBoV1 has been shown to modulate rhinovirusinduced cytokine responses in wheezing children. 15 Persistent viral infections of the tonsil tissue seem to alter the immune response, but the specific role of HBoV1 in chronic tonsillar disease remains unclear.
Human tonsils serve as an in vivo model for investigating cytokine responses at the local lymphoid tissue level. However, limited data are available on the bocavirus-associated T-cell and innate immune responses 16 and immunologic effects overall. The investigation of these events is important as earlier studies suggest that HBoV1 may have immunomodulatory effects. 15

| Sample collection
The internal part of the tonsil tissue removed from the patient was instantly cut into 3-4 mm cubes and stored in RNAlater (Qiagen, Hilden, Germany), first at +2 to 8°C until the next working day, and then at −80°C. 14 NPA samples were collected through a nostril using a standardized procedure. 18 If the aspirate yield was small, the collection was repeated after administering 2 ml of physiologic saline into a nostril.
Peripheral blood samples were drawn for serology and allergy tests.
The first of the paired serum samples was collected during the tonsillectomy anaesthesia and the follow-up sample was taken after a median of 58 days (range 36-104). For viral analyses, a piece of the removed tonsils and the NPA were stored at −80°C. 14 Study patients completed a standard questionnaire to collect information on health, medication, and respiratory symptoms within 30 days before the operation. 14

| Virus diagnostics
Respiratory viruses in NPA and tonsil tissue samples were detected by PCRs (including the reverse transcription step when applicable) on nucleic acid extracts. 18,19 In-house real-time PCRs were used as described for HBoV1, enterovirus, rhinovirus, and respiratory syncytial virus. [20][21][22] Primer sequences of HBoV1 have been previously described. 9 Seeplex RV12 ACE Detection (Seegene, Seoul, Korea) multiplex PCR assay was applied for adenovirus, coronaviruses (229 E/NL63 and OC43/HKU1), influenza A and B viruses, metapneumovirus, parainfluenza virus types 1-3, respiratory syncytial virus groups A and B, and rhinovirus, according to manufacturer's instructions. Quantitative PCR (qPCR) was used to measure tonsil HBoV1 DNA load (copies/g). 23 Serological tests for HBoV1-specific IgM and IgG were performed for 123 patients. 5,23 To verify that the IgG results were HBoV1 specific, the serum samples were blocked with HBoV2 and HBoV3 antigens. PCR tests were done at the Department of Virology, University of Turku, Finland. Serology was analyzed at the Department of Virology, University of Helsinki, Finland.

| Cytokine analysis
Tonsil tissue was stabilized and then homogenized in grinding tubes containing CK28 ceramic beads by a Precellys 24 homogenizer (Bertin Technologies) two times at 6000 rpm for 50 s. 14

| Patient characteristics
HBoV1 DNA in NPA and tonsil tissue was detected by PCR in 25/143 (17%) patients with a median age of 6 years; see Tables 1 and 3. The median age of HBoV1-negative patients (118/143%, 83%) was 18 years (p < 0.001). The sex distribution between the study groups was equal; 56% were male in the HBoV1-positive group and 52% in the -negative group. There was no difference in self-reported allergy, sensitization, allergic diseases, active/passive smoking, or symptoms on the operation day between the groups. Tonsillar hypertrophy was a more common indication for the operation in the HBoV1-positive group (16%, 64%) compared to the HBoV1-negative group (32%, 27%) (p < 0.001). In contrast, recurrent tonsillitis was a more common indication in the HBoV1-negative group compared to thepositive group (40%, 34% vs. 3%, 12%) (p = 0.03). See more patient characteristics in Table 1.   Table 3.

| Cytokine and transcription factor findings
The cytokine and transcription factor findings were different in  Chronic white patches in tonsils (n = 2), recurrent fever (n = 1), throat abscess (n = 1), teeth braces (n = 1), and food remnants in tonsils (n = 1). b One or more of the following: mild rhinitis, cough, symptoms of otitis, throat pain, upper airway obstruction symptoms. c HBoV1 DNA-positive: two tonsil tissues, one NPA, one tonsil and NPA. Note: One case was barely above cut-off for HBoV1 IgM and intratonsillar HBoV1-PCR positive.

T A B L E 2 Human bocavirus serology findings
(0.019), and IL-13 (p = 0.013) and tended to correlate negatively with IFN-β (p = 0.058) and IFN-γ (p = 0.088). Age had no effect on the DNA load correlation (Figure 2). There was no correlation between RORC2 and FOXP3 with Type 3 interferons in HBoV1 positive patients (data not shown).  pathogens and autoimmunity. Overall, IL-17A and IL-17F are associated with, for example, rheumatoid arthritis, inflammatory bowel disease, psoriasis, allergic asthma, and atopic dermatitis. 25,26 In our study, the expression of RORC2 in the tonsil tissue was lower within the HBoV1-positive group. This finding indicates that HBoV1 is associated with Th17-type cytokine reactions by affecting the transcription factor RORC2. However, there was no difference in the expression levels of IL-17 between HBoV1-positive and -negative groups. This might indicate that part of IL-17 production occurs in NK cells, neutrophils, CD8 + T cells, and γð T cells. 25 It has been shown earlier that co-infection of HBoV1 and rhinovirus results in a modified non-Th2-type response, which is different than the cytokine response of either virus alone. 15  cytokines. TGF-ß is strongly associated with autoimmune diseases, Alzheimer disease, cardiovascular pathologies, cancer, allergic rhinitis, and fibrosis. 25 FOXP3+ T-reg cell function is highly studied due to the potential in clinical practice. 28 It has been demonstrated that the frequency of FOXP3+ T-reg cells in tonsil tissue is three times higher than in peripheral blood. 29 Interestingly, in our study, the expression of T-reg transcription factor FOXP3 was lower in the tonsil tissues of patients in the HBoV1 DNA-positive group.

| DISCUSSION
Furthermore, our study shows a negative correlation between intratonsillar HBoV1-DNA loads and IL-28, IL-29 (IFN-λ family). The secretion of IL-28 and IL-29 and other IFNs is an important defence mechanism against respiratory viral infections. 25 It has been shown that IFN-λ provides antiviral defence especially at the epithelium surface of the respiratory tract. 30 Contoli et al. showed a clear inverse correlation between rhinovirus load and IFN-λ production in bronchial epithelial cells. 31 HBoV1 may also reduce IFN responses as shown in vitro. 17 Our study is the first one that compares IFN-λ and HBoV1-DNA loads.
Moreover, the correlation between intratonsillar HBoV1-DNA loads and IL-13 was also negative. IL-13 is part of T-helper 2 type immune response and is especially associated with allergic diseases such as asthma and rhinitis. 25 Interestingly, a recent study shows an increase of type 2 immune effectors including IL-13 in severe COVID-19 compared to a moderate one. 32 IL-13 seems to relate to virus-induced immune reactions. Thus, low IFN and IL-13 expressions could promote persistence of viral infections. In earlier study, HBoV has been associated with upregulation of type 2 immune response in bronchoalveolar lavage fluid. Especially cytokines linked with lung fibrosis and tumor development were expressed, prompting speculations of possible causal correlation. 33 Furthermore, this finding supports the role of HBoV as an immunomodulator.
The strengths of this study are the direct analyses of tonsil tissue samples and the thorough characterization of the study subjects.
Additionally, persistent HBoV1 infection was carefully diagnosed by PCR and serology. The limitations of the study include a relatively small number of study subjects (all n = 143, HBoV1 positive n = 25) and that the data set was not complete. HBoV1-DNA loads were measured from 9 of the 12 HBoV1 positive tonsil tissue samples.
Serum samples were obtained from 123 cases. Due to ethical reasons, we did not study a healthy control group.

| CONCLUSION
In conclusion, our study shows that persistent HBoV1 infection is associated with suppression of tonsillar cytokine responses, especially transcription factors RORC2 and FOXP3. This finding supports the earlier study 15 that HBoV1 infection may have immunosuppressive capacity and warrants further study.