Eur J Pediatr (1999) 158: 469 473 Springer-Verlag 1999

IMMUNOLOGY/ ALLERGOLOGY

M. A. Lopez-Gonzalez P. D az F. Delgado M. Lucas

Lack of lymphoid cell apoptosis in the pathogenesis of

tonsillar hypertrophy as compared to recurrent tonsillitis

Received: 8 July 1997 / Accepted in revised form: 2 April 1998

Abstract he pathogenic mechanism of tonsillar hypertrophy is unknown and lacks a

proper infectious or immunological explanation. Epidemiological studies point to polluted
environments as the main cause of tonsillar hypertrophy in the adaptation of the juvenile
organism. Tonsils and adenoids of 67 children aged 216 years (mean 5.9 years) were
divided into three groups: recurrent tonsillitis (n 21), recurrent tonsillitis with tonsillar
hypertrophy (n 21) and tonsillar hypertrophy without history of tonsillitis (n 25). The
following biological markers were studied: anti-streptolysin O antibody and anti-deoxy
ribonuclease B antibody serology, microbiology and cell count of granulocytes in tonsils
and adenoids as well as lymphocyte subsets and ``ex vivo'' endonuclease activity in tonsils.
Anti-streptolysin O antibody and anti-deoxyribonuclease B antibody titres were signi-
cantly raised in recurrent tonsillitis. Positive bacterial cultures for Streptococus pyogenes
were rare in cases of tonsillar hypertrophy. T-lymphocytes counts were lower and the
proportion of basophils was higher in hypertrophic tonsils than in recurrent tonsillitis.
Two parameters of apoptosis were studied; the activation of endonuclease, inducing
breakdown of DNA resulting in cell death, and the sensitivity to thapsigargin, known to
trigger the cleavage of DNA by apoptotic endonuclease. In children with tonsillar hy-
pertrophy both parameters were decreased contrasting with those with recurrent tonsillitis
where apoptosis is increased. It may be speculated that the increase of basophils in children
with tonsillar hypertrophy results in increased release of interleukin-4, which could prevent
lymphoid apoptosis and lead to cell proliferation in tonsillar tissue.
Conclusion Whereas recurrent tonsillitis is characterised by apoptotic death of lymphoid
tissue, tonsillar hypertrophy is caused by environmental pollution agents that trigger the
chronic inammatory process without apoptotic cell death.

Key words Tonsils Adenoids T-cells Endonuclease Basophils

Abbreviations ADB anti-deoxyribonuclease B antibody ASLO anti-streptolysin O

antibody RT recurrent acute tonsillitis RT-TH recurrent acute tonsillitis with tonsillar
hypertrophy TH tonsillar hypertrophy

of throat infections [1, 4] which may lead to obstructive

Introduction sleep apnoea [19, 20]. The pathogenesis of TH is poorly
understood although obstructive sleep apnoea is an in-
Idiopathic tonsillar hypertrophy (TH) is characterized dication in children for tonsillectomy and adenoidec-
by massively enlarged tonsils in children without history tomy [2, 11]. Since recurrent tonsillitis (RT) is caused by

M.A. Lopez-Gonzalez F. Delgado M. Lucas (&) P. D az

Department of Pediatric Otorhinolaryngology, Molecular Biology Service,
Virgen del Roc o University Infantile Hospital, Virgen Macarena University Hospital,
c/Manuel Siurot, s/n, c/Sanchez Pizjuan, 4, E-41 009- Sevilla, Spain,
E-41 013- Sevilla, Spain Fax: +34-95-455 74 81
470
Streptococcus pyogenes [4, 5], RT and TH could be
considered as two quite dierent pathogenic entities.
TH is associated with an increase in lymphoid ele-
ments without alteration of the connective stroma.
Nonetheless, the disruption of the normal relationship
of the lymphocyte subsets has not been related to TH,
neither has a proper infectious nor immunological ex-
planation of TH been postulated [21]. The life-span of
the lymphocyte includes a nal step in the secondary
lymphoid tissue. Many cells survive and the rest are
committed to apoptotic cell death [9, 23]. Activation of
an endogenous endonuclease has been described asso-
ciated with apoptosis in many cell types. The endonu-
clease is present constitutively in some cells (rodent
cortical thymocytes, T-lymphocytes) in which apoptosis
is triggered by dierent agents, but is inducible in others
[7, 22]. We questioned whether pathological conditions
in diseased tissues could modify the pattern of endonu-
clease activity in secondary lymphoid tissues.
Basophils are known to participate in the immediate
and late phase of the allergic response [8, 17]. Migration
of basophils to the aected lymphoid tissue could ex-
plain the pathogenesis of TH by mechanisms dependent
on the allergic reactions. Experiments were designed to
study the activity of endonuclease (a marker of apop-
totic cell death) in tonsillar cells and its sensitivity to be
triggered ``in vitro'' by the apoptotic agent thapsigargin
(a trigger of DNA breakdown [10]).
Materials and methods
Patients
The group of patients consisted of 67 children (23 females and 44
males) aged 216 years (mean 5.9 years). The main symptoms were
recurrent bouts of acute tonsillitis (more than four times a year for
more than 1 year) and/or TH leading to obstructive sleep apnoea
and dysphagia. This study excluded children receiving antibiotics
within at least 2 weeks before surgery. The patients were divided
into three groups based on tonsil size and history of infections. RT
included children (n 21) with recurrent infection and small tonsils
that occupied less than 20% of the oropharyngeal airway. The
group recurrent tonsillitis with tonsillar hypertrophy (RT-TH)
contained children (n 21) with recurrent infections and massively
enlarged tonsils that occupied 5075% of the oropharyngeal air-
way. The group TH consisted of children (n 25) with enlarged
tonsils, occupying up to 100% of the oropharyngeal airway, but no
history of infection.
Bacteriology
A culture swab of the surface and the core of the tonsils and ade-
noids was sent to the microbiology laboratory. Positive cultures
contained more than 100,000 colony-forming units per plate.
Measurements of anti-streptolysin O
and anti-deoxyribonuclease B antibodies
Anti-streptolysin O (ASLO) and anti-deoxyribonuclease B (ADB)
antibodies were measured in serum of the patients within the
week before adenotonsillectomy. ASLO was quantied by laser
immunonephelometry (Behring, Germany), ADB by enzyme inhi-
bition assay (Wampole Laboratories, USA) according to the pro-
tocols of the manufacturers. The upper limit of normal values was
333 and 660 U/ml for ASLO and ADB respectively.
Dierential count of white cells and the subsets of lymphocytes
in tonsils and adenoids
Pieces of tonsils and adenoids were minced with scissors in
phosphate-buered saline and cells released into the medium by
gentle shaking. The cellular suspension was recovered from the
supernatant by sedimentation. Dierential count of leucocytes
was done by optic microscopy after Giemsa staining. Lymphocyte
subpopulations were analysed by ow cytometry after labelling of
cells with uorescein isothiocyanate or phycoerytrin-conjugated
monoclonal antibodies. The antibodies to cluster of dierentia-
tion were purchased from Becton Dickinson (CD3, CD4,
CD5, CD8, CD11b, CD14, CD16, CD19, CD25, CD38, CD45,
CD62L, T-cell receptor), from Inmunotech (CD29, CD45RA,
CD45RO), from Knickerboker (CD4) and from Immunoquality
(plasmocytes). Cells were analysed in a Becton Dickinson cyto-
uorometer equipped with a 15 mW argon laser and CellQuest
software.
Analysis of DNA breakdown
Tonsillar lymphocytes were isolated and incubated at 37C under
5% CO2. DNA was extracted and the molecular size determined by
1.8% agarose gel electrophoresis [10]
Chemicals
Agarose was from BRL; 8-DNA/HindIII digested molecular
weights markers, proteinase K and ribonuclease A were from
Boehringer-Mannheim; N-laurylsarcosine and ethidium bromide
were from Sigma; RPMI medium, fetal bovine serum, strepto-
mycin and penicillin were from Flow. Thapsigargin was from
Calbiochem.
Results
Microbiology and serology
Microbiological studies showed that the percentage of
positive cultures for Streptococcus pyogeras of tonsillar
surface and core and of adenoids was signicantly
higher in case of RT and RT-TH (close to 30%) than in
TH (4%). Raised ASLO and ADB titres were found in
100% of RT and RT-TH cases whereas only 10% of
these with TH had raised values (Fig. 1).
Basophil count in tonsils and adenoids
Polymorphonuclear leucocytes in the diseased lymphoid
tissues showed no dierences in the percentage of neu-
trophils and eosinophils whereas the basophil content
was raised in both tonsils and adenoids in cases of TH.
The basophil count was quite similar in tonsils and ad-
enoids within each patient. The degree of increase in
basophils in the hypertrophied adenoids and tonsils

Fig. 1 Titres of anti-streptolysin O antibody (ASLO) and anti-
deoxyribonuclease B antibody (ADB) in patient serum. Results,
given in dilution units, are the mean ``SEM of the grouped values:
RT (21), RT-TH (21), TH (25). Values of ASLO and ADB were
higher in RT than in RT-TH and TH (P < 0.005) and higher in
RT-TH than in TH (P < 0.05)
ranged from 65% to 140% compared with cases of RT
and RT-TH respectively (Fig. 2).
T-cell subsets
The distribution of lymphocyte subpopulations varied
signicantly among the infectious and obstructive ton-
sillar pathology only in the percentage of T-lympho-
cytes(CD3+ and TCR alpha+/beta+) that increased in
cases of RT, RT-TH compared to TH. We found
no other dierences in the cellular subpopulations
studied in cases of infectious and obstructive tonsillar
pathology.
471
Fig. 2 Basophil content of tonsils and adenoids in recurrent acute
tonsillitis and tonsillar hypertrophy. Filled circles refer to tonsils
and empty circles to adenoids. Single and mean values are given as
a percentage of total leucocytes. The number (in parentheses) of
children studied are distributed in the following groups: (a) Tonsils,
RT (10), RT-TH (16), TH (21); (b) adenoids, RT (4), RT-TH (9),
TH (11). The basophil count was lower in RT than in RT-TH and
TH in tonsils (P < 0.01) and in adenoids (P < 0.05). The basophil
count was higher in TH than in RT-TH in tonsils (P < 0.01) and
in adenoids (P < 0.05)
DNA cleavage
Figure 3 shows the pattern of endonuclease activation in
TH, RT-TH and RT. RT was characterized by a high
spontaneous sensitivity to DNA breakdown under
control conditions contrasting with a low sensitivity to
thapsigargin. RT-TH showed a quite dierent pattern
b
Fig. 3 Analysis of DNA breakdown in tonsillar lymphocytes from
RT, RT-TH and TH children. Cells were incubated for 24 h in the
presence and in the absence of 0.5 lM thapsigargin to determine:
(a) the spontaneous DNA cleavage in cells cultured without
thapsigargin (lled bars); (b) thapsigargin-dependent cleavage of
DNA (low grey bars) and; (c) the lack of sensitivity of the cultured
lymphocytes to endonuclease (dark grey bars). The last group refers
to cells with high molecular weight DNA both under control
conditions and in the presence of thapsigargin. Results were
obtained in lymphocytes prepared from RT (n 6), RT-TH (n 7)
and TH (n 7) tonsils. The DNA was extracted after 18 h
incubation and the size analysed as described in Methods and
Fig. 4
472
Fig. 4 The pattern of internucleosomal breakdown. Lymphocytes
isolated from tonsils were incubated for 0 h (lanes 1 and 3), 8 h
(lane 4) and 18 h (lanes 2 and 5) in the absence (lanes 1 and 2) and
in the presence (lanes 3, 4 and 5) of 0.5 lM thapsigargin. DNA was
isolated and aliquot run in 1.8% agarose gel stained with ethidium
bromide. The right-hand lane was loaded with N174/haeIII
digested DNA; molecular sizes are given in bp
since the spontaneous breakdown of DNA was excep-
tional whereas the cells were highly sensitive to
thapsigargin which triggered DNA cleavage by the
endonuclease. In TH, both a low spontaneous cleavage
of DNA and a low sensitivity to thapsigargin were
concordant. This was particularly signicant when
compared to the RT-TH group.
Figure 4 illustrates the typical ladder of oligonu-
cleosomal fragments following the breakdown of DNA
by the endonuclease.
Discussion
Very signicant ndings were the raised percentage of
basophils in TH and the quite dierent pattern of
endonuclease activity among infectious and hypertro-
phic tonsillar pathologies. Given these results and the
dierences in bacteriology and lymphocyte subsets, RT
and TH should be regarded as dierent pathogenic en-
tities. The mechanism underlying TH could be ascribed
to the general pathogenesis of inammation. Basophils
are known to participate in the immediate and late phase
of the allergic response [8, 17] and therefore may con-
tribute to the sustained inammation of the aected
lymphoid tissues.
Dierent forms of the endonuclease have been de-
scribed and thapsigargin is known to trigger the apop-
totic endonuclease [10]. Activation of the endonuclease
contributes to DNA breakage and genomic instability
and has even been involved in some type of tumours
[15]. The nal mechanism of TH is dicult to assess but
the release of interleukin-4 by basophils [14, 16] could
regulate the activity of the apoptotic endonuclease and
lead to cell proliferation in the tonsillar and adenoid
tissue. In fact, interleukin-4 prevents the induction of
apoptosis of T-, B- and natural killer lymphocyte subsets
even in leukaemic cell lines [3, 6, 12, 13, 18]. However, a
role for other cytokines, i.e. tumour necrosis factor al-
pha cannot be ruled out.
We propose that RT is caused by the immune re-
sponse to pathogenic bacteria that, in addition to the
inammatory process, induce the breakdown of DNA
by activating the apoptotic endonuclease. In contrast,
TH could be caused by environmental agents, non-
bacterial or allergic, that result in the recruitment of
basophils which in turn trigger the chronic inammatory
process and the inhibition of apoptotic cell death.
Nevertheless, the possible involvement of viral infections
remains to be studied.
Acknowledgements Supported by grant 96/0278 from the Fondo
de Investigaciones Sanitarias. We thank Manoli Lopez and Salud
Ponce for clerical and technical assistance.
References
1. Beck EG, Schmidt P (1993) Umweltmedizinische gruppen-
diagnostische Kinder-untersuchungen 19821990. Ubersicht
Zentralbl Hyg Umweltmed 193:395418
2. Berkowitz RG, Halzal GH (1990) Tonsillectomy in children
under 3 years of age. Arch Otolaryngol Head Neck Surg
116:685686
3. Dancescu M, Rubio-Trujillo M, Biron G, Bron D, Delespesse
G, Sarfati M (1992) Interleukin 4 protects chronic lymphocytic
leukemic B cells from death by apoptosis and upregulates Bcl-2
expression. J Exp Med 176:13191326
4. DeDio RM, Tom LWC, McGowan KL, Wetmore RF, Han-
dler SD, Potsic WP (1988) Microbiology of the tonsils and
adenoids in a pediatric population. Arch Otolaryngol Head
Neck Surg 114:763765
5. Farocki MA (1967) Bacteriology and histology of tonsillar
parenchyma in tonsillectomized specimens. Eye Ear Nose
Throat Mon 46:310312
6. Illera VA, Perandones CE, Stunz LL, Mower DA Jr, Ashman
RF (1993) Apoptosis in splenic B lymphocytes. Regulation by
protein kinase C and IL-4. J Immunol 151:29652967
7. Khodarev NN, Ashwell JD (1996) An inducible lymphocyte
nuclear Ca2+/Mg2+-dependent endonuclease associated with
apoptosis. J Immunol 156:922931
8. Lichtenstein LM (1994) The role of basophil in allergic reac-
tions. In: (Van Moerbeke D, Mangelinckex C, Todt A, eds)
Conversations In allergy. 20th Symposium of the Collegium
Internationale Allergologicum. Nantucket, Massachusetts
9. Lucas M, Solano F, Sanz A (1991) Induction of programmed
cell death (apoptosis) in mature lymphocytes. FEBS Lett
1:1920
10. Lucas M, Sanchez-Margalet V, Sanz A, Solano F (1994)
Protein kinase C activation promotes cell survival in mature
lymphocytes prone to apoptosis. Biochem Pharmacol 47:667
672
11. Meredith GM (1990) Adenotonsillectomy. Arch Otolaryngol
Head Neck Surg 116:741
12. Migliorati G, Pagliacci C, Moraca R, Crocicchio F, Nicoletti I,
Riccardi C (1991) Interleukins modulate glucocorticoid-in-
duced thymocyte apoptosis. Int J Clin Lab Res 21:300303
13. Migliorati G, Nicoletti I, D'Adamio F, Spreca A, Pagliacci C,
Riccardi C (1994) Dexamethasone induces apoptosis in mouse
natural killer cells and cytotoxic T lymphocytes. Immunology
81:2126
14. Paul WE (1991) Interleukin-4 production by Fc epsilon R+
cells. Skin Pharmacol Suppl: 814
15. Russo CA, Weber TK, Volpe CM, Stoler DL, Petrelli NJ,
Rodr guez-Bigas M, Burhans WC, Anderson GR (1995) An
anoxia inducible endonuclease and enhanced DNA breakage

as contributors to genomic instability in cancer. Cancer Res
55:11221128
16. Schroeder JT, MacGlashan DW Jr, Kagey-Sobotka A, White
JM, Lichtenstein LM (1991) IgE-dependent IL-4 secretion by
human basophils. The relationship between cytokine produc-
tion and histamine release in mixed leukocyte cultures. J Im-
munol 153:18081817
17. Schroeder JT, Kagey-Sobotka A, MacGlashan DW, Lichten-
stein LM (1995) The interaction of cytokines with human ba-
sophils and mast cells. Int Arch Allergy Immunol 107:7981
18. Spinozzi F, Nicoletti I, Agea E, Belia S, Moraca R, Migliorati
G, Riccardi C, Grignani F, Bertotto A (1995) IL-4 is able to
reverse CD2-mediated negative apoptotic signal to CD4-CD8-
alpha beta and/or gamma delta T lymphocytes. Immunology
86:379384
473
19. Steinschneider A (1975) Nasopharyngitis and prolonged sleep
apnoea. Pediatrics 56:967971
20. Stradling JR, Thomas G, Warley ARH, Williams P, Freeland
A (1990) Eect of adenotonsillectomy on nocturnal hypo-
xaemia, sleep disturbances and symptoms in snoring children.
Lancet 335:240253
21. Sugiyama M, Sasaki T, Nakai Y, Otani S, Kinoshita Y (1982)
Studies on the cell mediated immune response of tonsillar
lymphocytes with regard to the clinical course of patient's age.
Int J Pediatr Otorhinolaryngol 4:193208
22. Wyllie AH, Kerr JFR, Currie AR (1980) Cell death: the sig-
nicance of apoptosis. Int Rev Cytol 68:251256
23. Wyllie AH, Arends MJ, Morris RG, Walker SW, Evan G
(1992) The apoptosis endonuclease and its regulation. Semin
Immunol 4:389397
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.