A Study of Serum Immunoglobulin

Levels in Elderly Persons That Provides

New Insights into B Cell
Immunosenescence
FLORINDA LISTÌ,a GIUSEPPINA CANDORE,a
MARIA ASSUNTA MODICA,a MARIANGELA RUSSO,a
GABRIELE DI LORENZO,b MARIA ESPOSITO-PELLITTERI,b
GIUSEPPINA COLONNA-ROMANO,a ALESSANDRA AQUINO,a
MATTEO BULATI,a DOMENICO LIO,a CLAUDIO FRANCESCHI,c
AND CALOGERO CARUSOa
a Gruppo di Studio sull’Immunosenescenza, Dipartimento di Biopatologia e
Metodologie Biomediche, Università di Palermo, Palermo, Italy
b Dipartimentodi Medicina Clinica e delle Patologie Emergenti, Università di
Palermo, Palermo, Italy
c Dipartimentodi Patologia Sperimentale and Centro Interdipartimentale
“L. Galvani,” Università di Bologna, Bologna, Italy, and Istituto Nazionale di
Riposo e Cura per Anziani, Ancona, Italy

ABSTRACT: The literature on immunosenescence has focused mainly on

T cell impairment. With the aim of gaining insight into B cell immunose-
nescence, we investigated the serum immunoglobulin levels in a cohort of
166 subjects (20–106 years). Serum IgG (and IgG subclasses) were quan-
tified by the nephelometric technique, IgE by CAP system fluorescence
enzyme immunoassay, and IgD by radial immunodiffusion (RID). There
was an age-related increase of IgG and IgA; the IgG age-related increase
was significant only in men, but IgG1 levels showed an age-related in-
crease both in men and women, whereas IgG3 showed an age-related
increase only in men. IgE levels remain unchanged, whereas IgD and
IgM serum levels decreased with age; the IgM age-related decrease was
significant only in women, likely due to the relatively small sample of
aged men. Thus, in the elderly the B cell repertoire available to respond
to new antigenic challenge is decreased. A lot of memory IgD− B cells are
filling immunological space and the amount of naı̈ve IgD+ B cells is dra-
matically decreased. This shift away from a population of predominantly
naı̈ve B cells obviously reflects the influences of cumulative exposure to
foreign pathogens over time. These age-dependent B cell changes indicate

Address for correspondence: Giuseppina Candore, Ph.D., Gruppo di Studio sull’Immunosenescenza,

Dipartimento di Biopatologia e Metodologie Biomediche, Corso Tukory 211, 90134 Palermo, Italy.
Voice: +39-09-1655-5932; fax: +39-09-1655-5933.
e-mail: gcandore@unipa.it

Ann. N.Y. Acad. Sci. 1089: 487–495 (2006). 

C 2006 New York Academy of Sciences.

doi: 10.1196/annals.1386.013

487
488 ANNALS NEW YORK ACADEMY OF SCIENCES

that advanced age is a condition characterized by lack of clonotypic im-

mune response to new extracellular pathogens. In any event, the increase
of memory B cells and the loss of naı̈ve B cells, as measured by serum
IgD levels, could represent hallmarks of immunosenescence and could
provide useful biomarkers possibly related to the life span of humans.

KEYWORDS: B cell; centenarians; elderly; immunosenescence; immu-

noglobulins

INTRODUCTION

The modifications of the immune system in the elderly are generally seen
as deterioration of the immune system, that is, immunosenescence. On the
other hand, immunosenescence is a complex process involving multiple re-
organizational and developmentally regulated changes, rather than a simple
unidirectional decline of whole functions. However, immunosenescence is
largely responsible for the diminished ability of older individuals to overcome
infection.1–4
Immunosenescence is claimed to be principally a result of the declining
effectiveness of T cells. Lifelong and chronic antigenic load is the major driv-
ing force of immunosenescence, which has an impact on human life span by
reducing the number of virgin antigen-nonexperienced T cells, and, simultane-
ously, filling the immunological space with expanded clones of memory and
effector, antigen-experienced T cells. Gradually, the T cell population shifts
to a lower ratio of naı̈ve to memory cells, the thymus pumps out fewer naı̈ve
T cells with age and those T cells remaining, especially the CD8+ subset, also
show increased oligoclonality with age. So, the repertoire of cells available to
respond to antigenic challenge from previously unencountered pathogens is
shrinking.4–8
In contrast to T cells, no evidence for a loss of B cell function has been
seemingly found as neither the total number of B cells or immunoglobulin
(Ig) secreting cells have been shown to be profoundly decreased with age.9
However, the B cell repertoire is influenced by aging during an actual im-
mune response, where the spectrum of expressed Ig genes, as well as the
frequency of somatic mutations, affects the quality, though not necessarily the
quantity, of the antibody response. What appears as an intrinsic defect in so-
matic mutations seems to be caused by suppressive influences exerted in vivo
by aged CD4+ T cells,3,9–11 possibly reflecting both the age-associated shift
from type 1 to type 2 cytokine patterns4,12 and the age-related impairment
of the CD40–CD40L system.11,13 Finally, one of the most dramatic examples
of age-associated repertoire change is the appearance of oligoclonal expan-
sions of CD5 B cells producing antibodies against self-antigens, albeit with no
known pathophysiological consequences.3,9,11
LISTÌ et al.: B CELL IMMUNOSENESCENCE 489

In a recent paper, we have focused on B cells in the aged by studying the

expression of some surface markers. In particular, in the elderly and in cente-
narians there was an increase of CD27+ B cells with a decrease of CD27− B
lymphocytes.14 These data seem of some importance. In fact, CD27 is con-
sidered a marker of primed memory cells and its engagement promotes the
differentiation of memory B cells into plasma cells.15 The decrement of virgin
CD27− B lymphocytes cells and the concurrent increase of memory CD27+
B lymphocytes can be one of the events that might have an impact on the
antibody repertoire of the elderly.14 In fact, chronic antigenic load might fill
immunological space with expanding clones of memory, antigen-experienced
B cells, which affect the clonotypic immune response to new extracellular
pathogens.
In this study, with the aim of gaining insight into B cell immunosenescence,
we investigated the serum Ig levels in young and old persons and in centenar-
ians.

MATERIALS AND METHODS

Sera were collected from Sicilian people: 46 healthy young adults (age <66
years, 30 women, and 16 men), 85 healthy aged persons (66–96 years, 63
women and 22 men), and 35 centenarians (≥ 99 years, 24 women and 11
men). None of the selected subjects was affected by neoplastic, infectious,
or autoimmune disease, nor had they received any drug influencing immune
functions at the time of the study. The study was approved by the University
Hospital Ethics Committee and all the subjects gave written informed consent.
Blood venous samples were withdrawn from the subjects under basal condition
at 9:00 AM and allowed to clot; after centrifugation, serum was stored in aliquots
at –70◦ C until analysis.
The concentrations of IgM, IgA, IgG, and IgG subclasses were determined
by a nephelometric technique (Array System 360, Beckman Coulter, Cassina
De’ Pecchi, Provincia di Milano, Italy). The nephelometric quantification is
based upon the reaction of a monospecific anti-immunoglobulin-specific anti-
serum with the human Ig to be determined. The generated immune complexes
are quantified by measuring the side-scattered light. The nephelometer is cal-
ibrated by measurement of a series of turbidity standards. The monospecific
antisera were purchased from Beckman Coulter. The concentrations in the test
samples are calculated relative to the calibration curves, obtained with the
nephelometric standard sera. The amount of IgD in each serum sample was
determined by the use of radial immunodiffusion (Medic, Pavone Canavese,
Provincia di Torino, Italy). Serum samples (20 L) were deposited in the
gel with monoclonal antibodies anti-IgD and incubated at room temperature
(∼
= 20◦ C) in a damp chamber for 48 h. The IgD concentration was determined
490 ANNALS NEW YORK ACADEMY OF SCIENCES

TABLE 1. Age- related serum concentrations of immunoglobulins (mean and 95% CI)
21–43 years (n = 46) 66–96 years (n = 85) 99–108 years (n = 35)
IgG mg/L 9,780 (8,711–10,861) 10,688 (9,911–10,865) 13,045 (12,047–14,042)
IgA mg/L 1,898 (1,650–2,146) 3,211 (2,834–3,588) 4,739 (4,049–4,218)
IgM mg/L 1,407 (1,203–1,611) 1,079 (927–1,232) 1,011 (893–1,128)
IgD IU/mL 23.7 (14.2–33.2) 7.3 (2.4–12.3) 3.2 (2.5–3.8)
IgE kU/L 119.2 (66.5–171.9) 189.5 (57.5–321.5) 172.1 (71.1–273.2)
Spearman’s rank significant correlation between age and serum immunoglobulins concentrations:
IgG = 0.0004; IgA < 0.0001; IgM = 0.002; IgD < 0.0001.

by evaluating the diameters of immunoprecipitates and converting them with

a table conversion. Total serum IgE levels were quantified by CAP-system
fluorescence enzyme immunoassay (FEIA) (Pharmacia, Uppsala, Sweden),
using UNICAP 100, according to the manufacturer’s instruction. Serum con-
centrations of IgD are presented as IU/mL. Total serum IgE values are ex-
pressed in kU/L. All the other Ig serum concentrations are presented as mg/L.
The results were expressed as mean (95% CI) and differences between gen-
ders were analyzed by the Mann–Whitney U-test. The rank correlations be-
tween concentration of serum Igs and age were calculated by Spearman’s rank
coefficient.

RESULTS

TABLE 1 shows the summary of age-related Ig serum concentrations (mean

and 95% CI). There was a significant positive correlation with age for IgG
and IgA values (respectively = 0.0004 and < 0.0001 by Spearman’s rank
coefficient), whereas there was a significant negative correlation with age for
IgM and IgD values (respectively = 0.002 and < 0.0001). No significant age-
related changes in IgE values were observed. By analyzing data according
to the gender (TABLE 2), significant age-related correlation was maintained in
women for IgA (< 0.0001), IgM ( = 0.001), IgD (< 0.0001) and in men for IgG
( = 0.0001), IgA (< 0.0001), and IgD ( = 0.01). Serum IgM concentrations
were reduced in young men when compared with young women (P < 0.05).
No other differences were found between women and men.
TABLE 3 shows the summary of age-related serum concentrations of IgG
subclasses (mean and 95% CI). There was a significant positive correlation
with age for IgG 1 and IgG 3 values (respectively = 0.0009 and = 0.009). No
significant age-related changes in IgG 2 and IgG 4 values were observed. By
analyzing data according to gender (TABLE 4), significant age-related corre-
lation was maintained in women for IgG 1 ( = 0.01) and in men for IgG 1
( = 0.01) and IgG 3 ( = 0.02). No differences were found between women and
men.
LISTÌ et al.: B CELL IMMUNOSENESCENCE 491

TABLE 2. Age- and gender-related serum concentration of immunoglobulins (mean and

95% CI)
Women (n = 117)
21–43 years (n = 30) 66–96 years (n = 63) 99–108 years (n = 24)
IgG mg/L 10,284 (9,056–11,512) 10,350 (9,546–1,158) 12,820 (11,515–14,125)
IgA mg/L 1,857 (1,600–2,113) 3,060 (2,652–3,468) 5,292 (3,849–6,734)
IgM mg/L 1,584 (1,312–1,857) 1,166 (973–1,356) 1,055 (885–1,224)
IgD IU/mL 28.5 (15.6–41.5) 4.6 (2.4– 6.8) 3.2 (2.3–3.9)
IgE kU/L 68.2 (26.8–109.6) 103.2 (49.5–156.9) 131.9 (22.7–241.0)

Men (n = 49)
21–43 years (n = 16) 66–96 years (n = 22) 99–108 years (n = 11)
IgG mg/L 8,858 (6,653 –11,063) 11,652 (9,648–1,365.) 13,535 (11,848–15,222)
IgA mg/L 1,974 (1,398–255) 3,643 (2,736–4,550) 5,292 (3,849–6,734)
IgM mg/L 1,075 (835–1,316) 831 (646–1,016) 915 (838–992)
IgD IU/mL 14.5 (1.5–27.6) 15.1 (3.7–34.0) 3.3 (2.0–4.6)
IgE kU/L 214.9 (89.2–340.6) 436.5 (64.4–937.6) 259.9 (19.3–500.6)
Serum IgM concentrations were reduced in young men compared with young women (P < 0.05).
No other differences were found between women and men in the same groups.
Spearman’s rank significant correlation between age and serum immunoglobulins concentrations
in women: IgA < 0.0001; IgM = 0.001; IgD < 0.0001.
Spearman’s rank significant correlation between age and serum immunoglobulins concentrations
in men: IgG = 0.0001; IgA < 0.0001; IgD = 0.01.

Concerning the difference that we observed between women and men, our
data are in agreement with previous results showing that IgA, IgM, and IgE
serum levels are influenced by sex, although we cannot exclude the existence
of regulator genes of immunoglobulin synthesis on heterochromosome since
B cell diseases and immunoglobulin deficiencies are linked genetically to the
X chromosome.16

DISCUSSION

It is well known that immune function declines with aging. It is assumed that
this decline concerns cell-mediated immunity more than humoral responses.

TABLE 3. Age-related serum concentrations of IgG subclasses of (mean and 95% CI)
21–43 years (n = 46) 66–96 years (n = 85) 99–108 years (n = 35)
IgG 1 mg/L 7,239 (6,603–7,874) 8,802 (7,478– 10,126) 9,542 (8,631–9,176)
IgG 2 mg/L 5,849 (404–4,643) 4,657 (404–5,270) 4,758 (4,308–5,208)
IgG 3 mg/L 1,127 (633–1,621) 920 (814–1,026) 1,090 (876–914)
IgG 4 mg/L 676 (450–902) 752 (606–898) 709 (572–847)
Spearman’s rank significant correlation between age and serum subclasses of IgG concentrations.
IgG 1 = 0.0009; IgG 3 = 0.009.
492 ANNALS NEW YORK ACADEMY OF SCIENCES

TABLE 4. Age- and gender-related serum concentration of IgG subclasses (mean and
95%CI)
Women (n = 117)
21–43 years (n = 30) 66–96 years (n = 63) 99–108 years (n = 24)
IgG 1 mg/L 6,890 (6,159–7,621) 8,605 (7,132–10,077) 9,412 (8,102–10,722)
IgG 2 mg/L 5,158 (4,360–5,956) 4,633 (3,897–5,370) 4,925 (4,352–5,498)
IgG 3 mg/L 1,358 (613–2,123) 905 (782–1,027) 1,057 (834–1,280)
IgG 4 mg/L 592 (453–731) 722 (556–888) 707 (515–899)

Men (n = 49)
21–43 years (n = 16) 66–96 years (n = 22) 99–108 years (n = 11)
IgG 1 mg/L 7,891 (6,626–7,576) 9,367 (6,255–12,478) 9,825 (8,970–10,668)
IgG 2 mg/L 7,120 (1,707–12,533) 4,725 (3,543–5,907) 4,394 (3,604–5,184)
IgG 3 mg/L 675 (538–812) 965 (740–1,190) 1,164 (613–1,714)
IgG 4 mg/L 834 (198–551) 839 (512–1,165) 715 (538–891)
No differences were found between females and males in the same groups.
Spearman’s rank significant correlation between age and serum subclasses of IgG concentrations
in women: IgG 1 = 0.01.
Spearman’s rank correlation between age and serum immunoglobulins concentrations in men:
IgG 1 = 0.01; IgG 3 = 0.02.

However, the B cell system is also affected, as suggested by the increased inci-
dence of autoantibodies, monoclonal gammopathies, and chronic lymphocytic
leukemia in the elderly.3–11 In particular, immunoglobulin levels may show
age-related changes. Several reports in the past have regarded the study of
serum immunoglobulin levels in the elderly, with the aim of both determin-
ing the normal values for clinical laboratory and of exploring the age-related
pathophysiology of the humoral responses.17 The number of subjects under
study and their age was different in the various studies; nevertheless, three
decades ago, hyperimmunoglobulinemia was claimed to be a characteristic of
the elderly.18 On the whole, the results demonstrated that IgG and IgA serum
levels increase with age, whereas IgM levels remain unchanged and IgE and
IgD decrease with age. Besides, among IgG subclasses, IgG1, 2, and 3 showed
a significant increase, whereas IgG4 did not.17,19,20
To the best of our knowledge no papers before ours have been published on
the relationship between age and all immunoglobulin class serum levels in the
same sample under study; besides, our paper includes the largest number of
centenarians studied so far. So, taking into account our results and the above
discussed findings, we can conclude that there is an age-related increase of IgG
and IgA; the IgG age-related increase was significant only in men, but IgG1
levels showed an age-related increase both in men and women, whereas IgG3
showed an age-related increase only in men. On the other hand, IgE levels
remain unchanged, as already shown by us,21 whereas IgD and IgM serum
levels decreased with age; the IgM age-related decrease was significant only
LISTÌ et al.: B CELL IMMUNOSENESCENCE 493

in women, likely due to the relatively small sample of aged men. Present and
previous data on unchanged levels of IgE make questionable the statement
that the incidence of allergic diseases decreases with age.21 Most important
are the data on the other immunoglobulins. In fact, the age-related increase of
IgG and IgA and, conversely, the age-related decrease of IgM and particularly
of IgD seem to be of some importance in order to gain insight into B cell
immunosenescence.
Secreted IgD is present in small amounts in the human serum. Regula-
tion of the synthesis still remains uncertain, but it has been argued that an
independent mechanism, possibly genetically determined, is important in the
basal production, although its levels seemingly depend on the amount of IgD+
B cells.22–24 Within the germinal centers activated naı̈ve IgD+ B cells undergo
vigorous proliferation, somatic hypermutation of Ig V region genes, isotype
switching, interaction with antigens, antigen-driven selection, and differentia-
tion into memory B cells, which proliferate rapidly in response to antigens and
produce large amounts of antibodies.15,25 The classical criterion of memory B
cells is the lack of expression of membrane IgD and CD38 or the expression of
switched IgG, IgA, and IgE; however, CD27 antigen represents a key marker
for memory B cells and the definitive marker of all memory B cells is the
presence of somatically mutated high-affinity antigen receptors.25,26 So, adult-
circulating B cells may be separated into three subpopulations: IgD+CD27−
naı̈ve B cells, nonswitched memory B cells IgD+CD27+, and IgDCD27+
switched memory B cells.27,28 In fact, CD27+ B cells carry somatic mutated
V-region genes, indicating that they are memory cells, representing unclass-
switched memory B cells.26 Thus, IgD+CD27+ B cells are an independent
subpopulation of memory B cells and may play a crucial role in secondary
immune response by their prompt synthesis of high-affinity IgM. The percent-
ages of IgD+CD27+ memory B cells increase during childhood and adult-
hood, peak at about 40 years of age, and then decline in the aged, determining
a poor secondary humoral immunity by IgM.28
The present results, showing that the levels of IgM and IgD are negatively
age-related whereas IgG and IgA are positively related, demonstrate, then, that
in the elderly a B cell repertoire available to respond to new antigenic challenge
is decreased. We can presume that a lot of memory IgD− B cells are filling
immunological space and the amount of naı̈ve IgD+ B cells is dramatically
decreased. This shift away from a population of predominantly naı̈ve B cells
obviously reflects the influences of cumulative exposure to foreign pathogens
over time. These age-dependent B cell changes documented by present and
previous studies14,28 indicate that advanced age, per se, is a condition charac-
terized by lack of clonotypic immune response to new extracellular pathogens.
In any event, our results are suggesting that the increase of memory B cells
and the loss of naı̈ve B cells, as measured by serum IgD levels, could represent
hallmarks of immunosenescence and could provide useful biomarkers possi-
bly related to the life span of humans. Since information on the senescence of
494 ANNALS NEW YORK ACADEMY OF SCIENCES

B cells is of obvious interest, further studies are necessary to confirm these

results by extending the number of markers used to characterize the cells.

ACKNOWLEDGMENTS

This work was supported by grants from the Italian Ministry of Education,
University and Research (ex 605) to G.C., G.D.L., G.C.R., D.L., and C.C. Funds
from the Italian Ministry of Health (Determinanti Immunogenetici di Salute
nell’anziano: un confronto interregionale) to C.C. are also acknowledged. The
“Immunosenesence Research Group” coordinated by Prof. C. Caruso in as-
sociation with INRCA was enlarged through a joint contract (Longevity and
Elderly Disability Biological Markers). F.L. and M.E.P. are Ph.D. students at
Pathobiology Ph.D. course (directed by C.C.) of Palermo University and this
work is submitted in partial fulfillment of the requirement for the Ph.D.

REFERENCES

1. FRANCESCHI, C. et al. 1995. The immunology of exceptional individuals: the lesson

of centenarians. Immunol. Today 16: 12–16.
2. COSSARIZZA, A. et al. 1997. Cytometric analysis of immunosenescence. Cytometry
27: 297–313.
3. GLOBERSON, A. & R.B. EFFROS. 2000. Ageing of lymphocytes and lymphocytes in
the aged. Immunol. Today 21: 515–521.
4. PAWELEC, G. et al. 2002. T cells and aging. Front. Biosci. 7: d1056–1183.
5. FRANCESCHI, C. et al. 2000. Human immunosenescence: the prevailing of innate
immunity, the failing of clonotypic immunity, and the filling of immunological
space. Vaccine 18: 1717–1720.
6. FAGNONI, F.F. et al. 2000. Shortage of circulating naive CD8(+) T cells provides
new insights on immunodeficiency in aging. Blood 95: 2860–2868.
7. PAWELEC, G. et al. 2004. Is immunosenescence infectious? Contribution of per-
sistent herpes viruses to immunosenescence and influence on human longevity.
Trends Immunol. 25: 406–410.
8. PAWELEC, G. et al. 2005. Human immunosenescence: is it infectious? Immunol.
Rev. 205: 257–268.
9. WEKSLER, M.E. 2000. Changes in the B-cell repertoire with age. Vaccine 18:
1624–1628.
10. MILLER, R.A. 1996. The aging immune system: primer and prospectus. Science
273: 70–74.
11. WEKSLER, M.E. & P. SZABO. 2000. The effect of age on the B-cell repertoire. J.
Clin. Immunol. 20: 240–249.
12. CARUSO, C. et al. 1996. Cytokine production pathway in the elderly. Immunol.
Res. 15: 84–90.
13. LIO, D. et al. 1998. Interleukin-12 release by mitogen-stimulated mononuclear
cells in the elderly. Mech. Ageing Dev. 102: 211–219.
14. COLONNA-ROMANO, G. et al. 2003. B cells in the aged: CD27, CD5, and CD40
expression. Mech Ageing Dev. 124: 389–393.
LISTÌ et al.: B CELL IMMUNOSENESCENCE 495

15. AGEMATSU, K. et al. 2000. CD27: a memory B-cell marker. Immunol. Today 21:
204–206.
16. MODICA, M.A. et al. 1989. Blood IgA, IgM and IgE levels are influenced by sex
and HLA phenotype. Exp. Clin. Immunogenet. 6: 251–257.
17. PAGANELLI, R. et al. 1994. Humoral immunity in aging. Aging Clin. Exp. Res. 6:
143–150.
18. RADL, J. et al. 1975. Immunoglobulin patterns in humans over 95 years of age.
Clin. Exp. Immunol. 22: 84–90.
19. TIETZ, N.W. et al. 1992. Laboratory values in fit aging individuals—sexagenarians
through centenarians. Clin. Chem. 38: 1167–1185.
20. HARALDSSON, A. et al. 2000. Serum immunoglobulin D in infants and children.
Scand. J. Immunol. 51: 415–418.
21. DI LORENZO, G. et al. 2003. A study of age-related IgE pathophysiological changes.
Mech. Ageing Dev. 124: 445–448.
22. LIO, D. et al. 1992. IgD serum levels are influenced by HLA-DR phenotype. Dis.
Markers 10: 105–108.
23. LITZMAN, J. et al. 1997. Serum IgD levels in children under investigation for and
with defined immunodeficiency. Int. Arch. Allergy Immunol. 114: 54–58.
24. PREUD’HOMME, J.L. et al. 2000. Structural and functional properties of membrane
and secreted IgD. Mol. Immunol. 37: 871–887.
25. KLEIN, U. et al. 1998. Somatic hypermutation in normal and transformed human
B cells. Immunol. Rev. 162: 261–280.
26. BANCHEREAU, J. & F. ROUSSET. 1992. Human B lymphocytes: phenotype, prolif-
eration, and differentiation. Adv. Immunol. 52: 125–262.
27. GROTH, C. et al. 2002. Impaired up-regulation of CD70 and CD86 in naive (CD27-)
B cells from patients with common variable immunodeficiency (CVID). Clin.
Exp. Immunol. 129: 133–139.
28. SHI, Y. et al. 2003. Functional analysis of human memory B-cell subpopulations:
IgD+CD27+ B cells are crucial in secondary immune response by producing
high affinity IgM. Clin. Immunol. 108: 128–137.