вход по аккаунту


Determination of immunoglobulin content of CSF based on light chain characteristics.

код для вставкиСкачать
Determination of Immunoglobulin
Content of CSF Based on Light
Chain Characteristics
K. Eickhoff, MD, and R. Heipertz, MD
The immunoglobulin light chain (typeskappa and lambda) content of normal cerebrospinal fluid is similar to that of
normal serum. In inflammatory diseases of the central nervous system a shift in the k/l ratio in comparison t o serum
values, usually a relative increase of Igk, can be observed. This increase of the W1 ratio, though not specific for any
disease, is most commonly found in multiple sclerosis. There is no correlation between the I g G content and the W I
ratio. The methods described &re measure bound and free light chains simultaneously. The calculations show that
free light chains are present in both normal and idammatory CSF and that they appear t o be of polyclonal origin.
The detection of light chain abnormalities in CSF can be taken as an indicator of endogenous immunoglobulin
production in the CNS and is of significance for the diagnosis of inflammatory CNS processes, especially when other
signs of endogenous immunoglobulin production are absent.
Eickhoff K,Heipem R Determination of immunoglobulin content of CSF based
on light chain characteristics. Ann Neurol3509-512, 1978
The quantitative determination of immunoglobulin
concentrations in cerebrospinal fluid is well established in the diagnosis of inflammatory diseases of the
central nervous system. Commonly these are measured on the basis of their heavy chain specificity as
IgG, IgA, and lgh4. Since all immunoglobulins have
light chains of either the kappa or lambda type in
common, their concentrations can be quantitatively
determined by their antigenic light chain characteristics, i.e., as Igk and Igl [Z-4, 10, 13, 141. Link e t d[8,
11, 161 were the first to investigate CSF for light
chains, and they found an increase of the k/l ratio in
50% of their patients with multiple sclerosis (MS). In
the present study the light chain concentrations (Igk
and Igl) were measured in both normal and inflammatory CSF, and the k/l ratio was calculated and related to IgG and albumin concentrations.
“normal,” in which the relative IgG content in CSF was
normal; and Group B (N= 54), from patients with inflammatory disease of the CNS (meningitis, encephalitis, MS,
lues cerebrospinalis),in which 41 specimens (76%) showed
a pathological increase of relative IgG concentration in CSF
that was not due to passive filtration from blood serum
caused by defective blood-brain barrier function, but was
produced by synthesis within the CNS (Tables 1, 2).
The analytical method of radial imrnunodiffusion for the
determination of light chain concentrations has been described before [2]. The LC-Partigen plates were specially
prepared by Behringwerke, Marburg, and calibrated with
isolated IgG containing 43 units per liter of Igk and 35
units per liter of Igl. The absolute light chain concentrations of this I& preparation were determined by the
same technique against standard human serum (Behringwerke, No. 974). The results of the CSF analyses
were converted to milligrams per liter according to a previously published formula [2]. All statistical calculations were
performed on a programmed desk calculator (Wang 600).
Material and Methods
Lumbar CSF from patients with a variety of neurological
disorders was investigated for cell content, total protein
(modified Biuret reaction), and concentrations of albumin,
I&, IgA, and IgM (LC-Partigen plates, Behringwerke,
Marburg, Federal Republic of Germany). The CSF samples
were divided into two groups: Group A (N = 39). from
patients having disorders other than an idammatory disease of the CNS (seizures, cerebrovascular disease, atrophy, lumbar disc herniation, psychosis), and termed
The results of the CSF analyses were compared with
the results of serum tests previously published [2].
Like serum, CSF also shows a significant correlation
From the Neurochemical Laboratory, University of Gijmngen,
Gottingen, and the Department of Neurology, University of
Hamburg, Hamburg, Federal Republic of Germany.
Address reprint requests to Dr Heipem. Neurologische Univ.
Klinik, Martinistr. 52, 2 Hamburg 20, Federal Republic of Germany.
between light chain concentration and total immunoIgVIgG IgA IgM;
globulin concentration (I&
r = 0 . 8 6 ; ~= 22.3
0 . h ) . Figure 1 shows histograms of W1 quotients in CSF specimens from the two
Accepted for publication Jan 17, 1978.
0364-5134/7810003-0608$01.25 @ 1978 by K. Eickhoff
Table I . Charactcrirtics of the Patientr in Group A
(Controls, No aui&nn of CNS Inflammation)
No. of
(N= 39)
Patient Data
Peripheral neuropathy
Cerebrovascular disease
Parkinson syndrome
Amyotrophic lateral sclerosis
Lumbar disc herniation
CSF characteristics
Cell count less than 1213 ( cells/mn:
Total protein less than 600 mghter
IgG immunoglobulins less than 60 mglliter
Table 2 . Characteriktics of the Patients in Group B
(Infiammatoty CNS Disease)
No. of
(N = 54)
Patient Data
Cerebrospinal syphilis
Tuberculous meningitis
Bacterial meningitis
Viral meningitis
Facial palsy
Transverse myelitis
Optic neuritis
Multiple sclerosis, definite or probable
Fig 1 . Histogram of RIl quotients in CSF from the two diagnostic groups: (A)patients with noninflrmmatoq direascr; (8)
patients with inflammatory diseases. Above, the range of kll
quotientsfrom normalsera.
CSF characteristics
Cell count increased to more than 12/3 in
32 (60%)
Total protein increased to more than 600
mg per liter in 33 (62%)
I g G immunoglobulins increased to more
than 60 mg per liter in 45 (84%)
Relative IgG increase (overproportional
increase of I g G in relation to albumin)
in 41 (76%)
analyzed here, 11% ( 5 MS, 1viral encephalitis) were
above and 4% ( 2 MS)were below the 95% confidence limits. There was no statistically significant correlation between the IgG concentration and the value
of the W1 quotient using Spearman’s rank correlation
It is also possible to express the light chain concentrations as a relative percentage (I& + Igl = 100%).
I n Group A, values similar to those in normal sera
were found (mean Igk in CSF,61 f 7.8%; mean I&
in serum, 63 ~ t _7.1%). Group B shows considerable
deviation, with I& values in CSF between 18 and
100%. The distribution histograms are shown in Figure 2. Figure 3 demonstrates the results obtained
when the total light chain concentration (Igk Igl) is
expressed as a percentage of total immunoglobulin
concentration (IgG I d ’ + IgM = 100%) and correlated with the corresponding albumin concentration in CSF. Light chain concentrations amounting to
more than 100% of total immunoglobulin concentration are found only in CSF specimens containing less
than 700 mg per liter of albumin, i.e., those in which
the blood-brain barrier is functioning. I n all cases in
which total light chain concentration amounted to
diagnostic groups. The quotients in Group A, the
“normal” group, show normal distribution with an
arithmetic mean of 1.65 and standard deviation of
0.55. The 95% confidence limits (mean f 2SD),
which range from 0.55 to 3.3, are slightly wider than
the corresponding serum values (0.68 to 2.88). In
contrast, some inflammatory CSF specimens (Group
B) show a definite shift of W1 quotients. O f those
510 Annals of Neurology Vol 3
No 6 June 1978
more than 80% of total imihunoglobulin cbncentration, the relative distribution of I& and Igl was similar to that in normal SelllflCL (mean I&, 63 f 12%;
N = 18).Only Group 3 cotitained an instahce of total
light chain contentration amounting to 143% of total
immunoglobulin concentration and consisting almost
exclusively (93%) of k chains.
Frick add Scheid-Seydel[5] were the first to demonstrate that immunoglobulins in CSF are filiered
through the blood-brain barrier from blood plasma.
A similar relationship between serum and CSF for
the relative light chain concentrations and the W1
quotients with corresponding values could be demonstrated here for Group A, the control group.
Changes in the k/l ratio of CSF independent of serum
have been described in inflammatory conditions of
the CNS 11, 6, 7, 9, 11, 12, 161 but also in noninflammatory CNS processes [71. This W1 shift
appears to be associated most commonly with MS
[ll]. The relatively small number of cases with altered k/l ratio in our study could be due to the fact
that our Group B contained specimens from other
inflammatory conditions besides MS, such as encephalitis, meliingitis, and cerebrospinal syphilis.
Like other authors [7, 121, we found both increased
and diminished k/l quotients; there was no correlation with the presence of an overproportional IgG
elevation in CSF.
Alteration of total light chain concentration and W1
quotients in CSF independently of serum can be
taken as evidence for the endogenous production of
immunoglobulins within the CNS [6, 161. Although
these changes are by no means specific for i d a h matory CNS disease, our results show that they are
most commonly associated with an inflammatory
process, so that they may be termed typical for CNS
Fig 2 . Histograms of the percentage fractions of lgk (IgG IgA
+ IgM = 1 a0%) in CSF from the two diagnostic gmnps; (A)
patients with noninjammatory diseases; (8)patilnts with
inflammatory diseases. Above, the nsrltsjbrn normalsera.
F i g 3. Relationship between albrrnin concentration in CSF (as
an indicator of blood-brain barrierfnnction) and nkative light
chain concentration (Igk and I g l as a percentage of total immnnoglobulins).Dottcdline marks the meatt 2 SD range from
nomral sera. Straight line mpnsents the mpssion line of CSF
values: y = 79.9 - 0.027x;r= -0.206.
Eickhoff and Heipertz: CSF Immunoglobulins 51 1
inflammation. Altogether, our study showed light
chain abnormalities of CSF exclusively in 14
patients-5 with altered W1 ratio, 6 with significantly
increased or decreased total light chain concentration, and 3 with both. Similar changes were not observed in CSF from Group A. Only 6 of our 14 patients with CSF light chain abnormalities also had an
overproportional increase of IgG in CSF. In the
other 8 cases (5 definite and 3 probable MS), the
alterations in CSF light chain concentration were the
only evidence of an immunological phenomenon involving the CNS.
Free light chains of polyclonal origin can be detected in serum [ 151. Because of their small molecular size, they readily diffuse through the blood-brain
barrier into CSF [ l , 61. Light chains of intact immunoglobulin molecules contain blocked antigenic sites
known as “hidden determinants,” which normally do
not react with antisera but may cause distortion of
results with an increase of free light chains (e.g., multiple myeloma or uremia) [2, 103. Since the method is
based on determination with antisera against Igk and
Igl, which react both with light chains bound to intact
immunoglobulins and with hidden determinants, it is
impossible to differentiate between light chaincharacterized immunoglobulin and free light chains.
Thus a certain percentage of total immunoglobulins,
in a few cases over 10096, appears to consist of light
chains. This phenomenon could be interpreted to
demonstrate the existence of free light chains in CSF
because the light chain concentration derived from
intact immunoglobulins would be expected to
amount to no more than about 40 to 60% of total
immunoglobulin concentration. Double ring formation as an indicator of free light chains [1, 61 could
not be observed in this series. O u r calculations show
that free light chains are present in inflammatory CSF
and in CSF from other diseases; apart from 1 case in
Group B, they must be of polyclonal origin because
kappa and lambda occur in the same ratio as in blood
serum or as in CSF not containing any significant
amounts of free light chains. We therefore suggest
that the determination of free light chains by absorbed antisera against all determinants should be
complemented by biochemical isolation of free light
We conclude that the demonstration of light chain
abnormalities in CSF independent from serum can be
of assistance in the diagnosis of an inflammatory CNS
condition. In a small proportion of cases such abnormalities may be the only evidence for localized immunoglobulin production within the CNS.
512 Annals of Neurology
Vol 3
No 6 June 1978
We wish to thank the Behringwerke, Marburg, Federal Republic
of Germany, for kindly supplying the special LC-Partigen plates,
and Miss S. Zachmann for her expert technical assistance.
1. Bollengier F, Lowenthal A, Henrotin W: Bound and free light
chains in subacute sclerosing panencephalitis and multiple
sclerosis serum and cerebrospinal fluid. J Clin Chem Clin
Biochem 13:305-310, 1975
2. Eickhoff K, Heipern R The determination of serum immunoglobulin concentrations on the basis of their light-chain antigenic properties. Clin Chim Acta 78343-349, 1977
3. Fahey J L Two types of 6.6 s y-globulins, P,A-globulins and
18 s y,-macroglobulins in normal serum and y-microglobulins
in normal urine. J Immunol 91:438-447, 1963
4. Fahey JL, McKelvey EM: Quantitative determination of
serum immunoglobulins in antibody-agar plates. J Imrnunol
9484-90, 1965
5. Frick E, Scheid-Seydel L: Untersuchungen mit
y-Globulin zur Frage der Abstammung der LiquoreiweiB
korper. Klin Wochenschr 36857-863, 1958
6. Iwashita H, Grunwald F, Bauer H : Double ring formation in
single radial immunodiffusion for kappa chains in multiple
sclerosis cerebrospinal fluid. J Neurol 207:45-52, 1974
7. Kolar 0, Anthony E: Cerebrospinal fluid and serum light
polypeptide chains in 160 patients with various nervous system disorders. J Neurol 200:6-17, 1971
8. Link H: Immunoglobulin G and low molecular weight proteins in human cerebrospinal fluid. Chemical and irnrnunological characterization with special reference to multiple sclerosis. Acta Neurol Scand 43:Suppl 28:l-136, 1967
9. Link H, Panelius M, Salmi AA: Immunoglobulins and measles
antibodies in subacute sclerosing panencephalitis. Arch
Neurol28:23-30, 1973
10. McKelvey EM, Fahey J L Immunoglobulin changes in disease:
quantitation on the basis of heavy polypeptide chains, IgG
(yG), IgA (yA), and I g M (yM), and of Light polypeptide
chains, type K (I) and type L (11). J Clin Invest 44:1778-1787,
11. Olsson JE, Link H: Immunoglobulin abnormalities in multiple
sclerosis. Arch Neurol 28:392-399, 1973
12. Roberts-Thompson PJ, Esiri MM,Young AC, et al: Cerebrospinal fluid immunoglobulin quotients, kappdlambda ratios,
and viral antibody titres in neurological disease. J Clin Pathol
29:1105-1115, 1976
13. Skvaril F, Barandun S, Kuffer F, et al: Veranderungen des
kappdlambda-Verhaltnisses der menschlichen Serumimmunoglobdine im Verlaufe der Enwicklung. Blur 33:281284, 1976
14. Skvaril F, Barandun S, Morell A, et al: Imbalances of kappa/
lambda immunoglobulin light chain ratios in normal individuals and in immunodeficient patients, in Protides of the Biological Fluids, 23rd Collegium, Brugge. New York, Pergamon
Press, 1975, pp 415-420
15. Solling K Free light chains of immunoglobulins in normal
serum and urine determined by radioimmunoassay. Scand J
Clin Lab Invest 35:407-412, 1975
16. Zettervall 0, Link H: Electrophoretic distribution of kappa
and lambda immunoglobulin light chain determinants in
serum and cerebrospinal fluid in multiple sclerosis. Clin Exp
Immunol 7:365-372, 1970
Без категории
Размер файла
323 Кб
contents, base, characteristics, chains, determination, immunoglobulin, light, csf
Пожаловаться на содержимое документа