Journal o/ the neuroloqical Sciences, 1975, 26:277-281 ~, Elsevier Scientific Publishing Company, Amsterdam
277 Printed in The Netherlands
On the t-Protein in Cerebrospinal Fluid P. VERHEECKE* D~Tartment oj Neuroloyy ( Pro£ H. rander Eecken), University of Ghent, Ghent (Belyium) (Received 14 March, 1975)
INTRODUCTION
Ever since the t-protein ofcerebrospinal fluid (CSF) with its electrophoretic mobility between fll and ? was discovered, its origin has been a subject of controversy. Pette and Stupp (1960) described the change of the electrophoretic mobility of transferrin from fla towards t after incubation with neuraminidase. Clausen and Munkner (1961) described the iron-binding capacity of t-protein. Parker and Bearn (1962) studied different variants of transferrin and their iron-binding neuraminidase-induced products and showed all transferrins to shift from their original position to the t-position over 3 intermediate positions, keeping always their iron-binding capacity, apparently after losing 1 to all 4 of their neuraminic acid moieties. They concluded that 2 transferrins existed in CSF, one being more prone to neuraminidase-induced changes than the other. This was confirmed by Frick and Scheid-Seydel (1963) who found different specific activities of t31 I-labelled transferrin in serum and in CSF, after injection of the labelled substance and allowing 4 days for equilibration. They introduced the name of "cerebrogenic" transferrin. However, Laterre (1965) challenged this interpretation, stating that the difference between fl x-transferrin and t-transferrin was that the former was freshly derived from serum, whereas the latter had been for a longer time in the CSF and thus derived from/31 by neuraminidase activity. Electrophoresis of unconcentrated CSF throws a new and unexpected light upon this problem. METHODS AND MATERIALS
128 CSF samples were obtained by lumbar puncture. About 5 ~tl of each sample, with 150-600 ktg protein/ml, was subjected to agar gel electrophoresis and stained as described elsewhere (K6r6nyi and Gallyas 1972) with minor modifications as proposed by myself (Verheecke 1974). The remaining 44 of these samples were then * On leave from the Department of Physio!3gical Chemistry (Prof. L. Vandendriessche), University of Ghent.
278
P. VERHEECKE
frozen, k e p t at - 3 2 ~ C a n d later t h a w e d to be c o n c e n t r a t e d a n d e l e c t r o p h o r e s e d at the B o r n - B u n g e F o u n d a t i o n in B e r c h e m - A n t w e r p (Dr. A. L6wenthal). T h e r e the classical techniques were used, i.e. with a b o u t 5 ttl c o n c e n t r a t e with 50- 60 m g p r o t e i n ml for each p h e r o g r a m ( L 6 w e n t h a l 1964). Relative m o b i l i t i e s (m,.) o f r - f r a c t i o n s were m e a s u r e d o n the d e n s i t o g r a m , t a k i n g the / ] l - a l b u m i n d i s t a n c e as ttnltv It soon b e c a m e clear t h a t with u n c o n c e n t r a t e d C S F it was u n n e c e s s a r y to c o m p a r e with a s t a n d a r d s o l u t i o n o f a l b u m i n - t r a n s f e r r i n ( h u m a n a l b u m i n a n d t r a n s f e r r i n were purc h a s e d f r o m Sigma). T h e m o b i l i t y o f the a l b u m i n is the s a m e in C S F as in a s t a n d a r d s o l u t i o n o f purified a l b u m i n . N o C S F c o u l d be f o u n d with an a l b u m i n lransferrin d i s t a n c e which was m e a s u r a b l y different f r o m that in the s t a n d a r d solution. T h e classical increase o f the m o b i l i t y o f C S F a l b u m i n c o m p a r e d to serum a l b u m i n is a c o n c e n t r a t i o n artifact (Fig. 1 ). Nevertheless, all C S F s a m p l e s were elect r o p h o r e s e d t o g e t h e r with o t h e r s a m p l e s or with s e r u m s a m p l e s in a Pleuger I m m u n o p h o r (5 at a t i m e ) ; thus the o c c a s i o n a l [31 which h a d an a b n o r m a l m o b i l i t y was readily discovered when c o m p a r e d to the o t h e r d e n s i t o g r a m s o f the set.
B
O
Fig. l. Agar electropherograms of unconcentrated (A) and concentrated (B) CSF. The same sample of the same patient was used. Both A and B are electrophoresed along with a standard mixture of purified human albumin and transferrin in water. In A the mixture was loaded beside the main application cleft in the lower margin, in B the mixture is in the upper margin. The cathode is to lgft, the anode to the right. A : with the distance albumin-transferrin of the standard solution as 1.000 the albumin of the unconcentrated sample is at 1.000, there is a/~2-peak at -0.163 and a small r-peak at -0.230 B: the remainder of the sample was concentrated, electrophoresed and stained with Amido-Schwarz. The albumin peak is at 1.067 and the z-peak at -0.253. There is no tq2-peak.
279
ON THE T-PROTEIN IN CEREBROSPINAL FLUID
REsULTs
Histograms were made with the m r of the f12- and T-peaks plotted against the number of peaks (Fig. 2). Several facts are already obvious after this short series of experiments. The 44 CSF concentrates have their z-fraction with a mean m r o f about - 0 . 2 5 5 and a normal distribution. When the same 44 samples are electrophoresed unconcentrated, even the slowest m r o f t is nearer to fll than - 0.255 and the distribution is not normal at all. Furthermore, the 44 unconcentrated samples have 54 peaks: 10 proteinograms have 2 peaks in the fl2-z-region. The proteinograms o f the 44 concentrates have no fl2peaks, only one Z-peak each. Finally, even when the fl2-peaks are not taken into account, the distribution o f t h e mrs of the z-peaks is still not normal. This distribution can be better analysed when a larger series of pherograms is examined (Fig. 3). Here the f12- and z-peaks of 128 pherograms of unconcentrated CSF samples have their r n r plotted against their number. It seems that the histogram has its slowest peak at m r -0.230, with a shoulder at roughly -0.200. Another peak seems to be present at - 0.155 and nearer to fll the picture is confused. The quantitative importance of the ill-peak has a role when one is to discover a peak with an m r less than - 0 . 1 0 0 different. A small peak in that region near a huge fl~ will remain undiscovered. Therefore, one cannot say a lot about the region with an m r between 0 and - 0.125. Even so there are 4 clearly visible peaks : fll (mr = 0) and the peaks at - 0. 155, -0.200, - 0 . 2 3 0 and perhaps a fifth at -0.105. No
1 o -]
A
o 0
-0025 - 0 0 5 0
~ -0075 -0100
-0125
-0150 -0175
0
.0.025 -0.050 -0.075 - 0 . I 0 0
-0125
-0.150 - 0 7 7 5 - 0 2 0 0 - 0 2 2 5
-0200
0.225 -0.250 -0.275
mr
No
-0.250 -0.275
mr
Fig. 2. Histograms of the number of f12- and z-peaks plotted against the mr. A: 44 unconcentrated samples; B: concentrates of the same 44 samples. No 20 15 10 5 0 - 0 0 2 5 -(2050 -0.075
-0.100
-0.125 -0.150 -0.175 -0.200 -0.225 -0.250 -0 275
mr
Fig. 3. Histogram of the number of f12o and r-peaks from 128 unconcentrated samples, plotted against their mr.
280
p. VERHEECKE DISCUSSION
These observations are not in agreement with any explanation of the presence of r-proteins in CSF published up to now. Laterre's (1965) explanation does not fit because the presence o f a z-protein with a constant rn r is an in vitro artifact and not the consequence o f in vivo neuraminidase activity. Parker and Bearn's (1962) explanation does not fit either: although there are 3 intermediate peaks between 31 and r, they do not have the regular intervals between them which Parker and Bearn so elegantly showed in their experiments. There are probably even more peaks in the obscure region between 0 and - 0.125, not to mention the classical peak at - 0 . 2 5 5 which appears when CSF concentrates are electrophoresed. So there are far too many peaks to fit into Parker and Bearn's theory. Thus the question remains: why does part of the transferrin remain at the original [~lposition, whereas another part moves to slower regions? All possible explanations are to be divided into 2 groups. First, negatively-charged groups can be split off the molecule; second, positively-charged groups can be adsorbed to the molecule. Up to now the first group has had most of the attention. But not only neuraminidase must be looked at; other enzymes can split: off negatively-charged groups or even negatively-charged parts of the polypeptide molecule : it could just be trivial proteolytic activity. One argument in favour of this idea is that a further slowdown of the [~2 migrating band to r is apparently quicker than transition from/31 to 32- This is in favour of the idea that the 3z migrating bands are damaged molecules, more prone to digestion than whole transferrin molecules. Why then does this not happen in serum? Perhaps some inhibitors which are absent in CSF are present in serum or. on the contrary, and more probably, are the breakdown products broken down more quickly and thoroughly in serum'? The second group of possibilities, although less probable, must be kept in mind. However, no known positively-charged group is at this moment a likely candidate to be bound to transferrin. On the other hand, negatively-charged groups bound to albumin, which is known to bind small molecules easily, could account for the change of mr of that molecule.
CONCLUSION
This report makes two conclusions possible. The albumin in CSF has the same electrophoretic mobility as the serum albumin. The z-protein in vivo has no constant m r with a normal distribution when measured on many pherograms. Apparently it is a breakdown product of transferrin, some breakdown products (the classical zprotein) being more stable than others (the intermediate peaks). The mechanism is possibly proteolytic. ACKNOWLEDGEMENTS
I wish to thank Mrs. A. Vanneste and Mrs. De Smet for electrophoresing and staining the unconcentrated samples and Mrs. G. Van Soom for concentrating the CSF and
ON THE C-PROTEIN IN CEREBROSPINAL FLUID
281
performing the electrophoresing on the concentrates. I am grateful to Prof. A. LOwenthal for many useful discussions and to Prof. H. van der Eecken for constant support. SUMMARY
128 cerebrospinal fluid samples were obtained. About 5/d of each of the samples was subjected unconcentrated to agar gel electrophoresis. The remaining 44 of these samples were concentrated and the concentrates were electrophoresed. The proteins were stained. Analysis of the relative mobilities of the peaks makes two conclusions possible: the albumin in cerebrospinal fluid has the same electrophoretic mobility as purified serum albumin. The r-protein possibly arises from transferrin by proteolytic breakdown, some products (such as the classical z-protein) being less unstable than others (the intermediate peaks). REFERENCES CLAUSEN, J. AND T. MUNKNER (1961) Transferrin in normal cerebrospinal fluid, Nature (Lond.), 189: 6(~61. FRlCK. E. AND L. SCHEn)-SEYDEL (1963) Untersuchungen mit j~31 markiertem Transferrin zur Frage der Abstammung der Liquoreiweissk6rper, Klin. Wschr., 41 : 589-593. K~R~NVt, L. AND F. GALLYAS (1972) A highly sensitive method for demonstrating proteins in electrophoretic, immuno-electrophoretic and immunodiffusion preparations, Clin. chim. Acta, 38: 465-467. LATERRE, E. C. (1965) Les Prot~ines du Liquide COphalorachidien d I'Etat Normal et Pathologique, Arscia, Brussels, pp. 172-175. LOWENTHAL, A. (1964) Agar Gel Eleetrophoresis in Neurology, Elsevier, Amsterdam. PARKER, W. C. AND A. G. BEARN (1962) Studies on the transferrin of adult serum, cord serum and cerebrospinal fluid The effect of neuraminidase, J. exp. Med., 115:83 105. PETTE, D. AND I. STUPP (1960) Die r-Fraktion im Liquor cerebrospinalis, Kiln. Wschr., 38:109 110. VERHEECKE, P. (1974) Agar gel electrophoresis of unconcentrated cerebrospinal fluid--- The degenerative type, Acta neurol, bel¢l., 74: 376-382.
277 Printed in The Netherlands
On the t-Protein in Cerebrospinal Fluid P. VERHEECKE* D~Tartment oj Neuroloyy ( Pro£ H. rander Eecken), University of Ghent, Ghent (Belyium) (Received 14 March, 1975)
INTRODUCTION
Ever since the t-protein ofcerebrospinal fluid (CSF) with its electrophoretic mobility between fll and ? was discovered, its origin has been a subject of controversy. Pette and Stupp (1960) described the change of the electrophoretic mobility of transferrin from fla towards t after incubation with neuraminidase. Clausen and Munkner (1961) described the iron-binding capacity of t-protein. Parker and Bearn (1962) studied different variants of transferrin and their iron-binding neuraminidase-induced products and showed all transferrins to shift from their original position to the t-position over 3 intermediate positions, keeping always their iron-binding capacity, apparently after losing 1 to all 4 of their neuraminic acid moieties. They concluded that 2 transferrins existed in CSF, one being more prone to neuraminidase-induced changes than the other. This was confirmed by Frick and Scheid-Seydel (1963) who found different specific activities of t31 I-labelled transferrin in serum and in CSF, after injection of the labelled substance and allowing 4 days for equilibration. They introduced the name of "cerebrogenic" transferrin. However, Laterre (1965) challenged this interpretation, stating that the difference between fl x-transferrin and t-transferrin was that the former was freshly derived from serum, whereas the latter had been for a longer time in the CSF and thus derived from/31 by neuraminidase activity. Electrophoresis of unconcentrated CSF throws a new and unexpected light upon this problem. METHODS AND MATERIALS
128 CSF samples were obtained by lumbar puncture. About 5 ~tl of each sample, with 150-600 ktg protein/ml, was subjected to agar gel electrophoresis and stained as described elsewhere (K6r6nyi and Gallyas 1972) with minor modifications as proposed by myself (Verheecke 1974). The remaining 44 of these samples were then * On leave from the Department of Physio!3gical Chemistry (Prof. L. Vandendriessche), University of Ghent.
278
P. VERHEECKE
frozen, k e p t at - 3 2 ~ C a n d later t h a w e d to be c o n c e n t r a t e d a n d e l e c t r o p h o r e s e d at the B o r n - B u n g e F o u n d a t i o n in B e r c h e m - A n t w e r p (Dr. A. L6wenthal). T h e r e the classical techniques were used, i.e. with a b o u t 5 ttl c o n c e n t r a t e with 50- 60 m g p r o t e i n ml for each p h e r o g r a m ( L 6 w e n t h a l 1964). Relative m o b i l i t i e s (m,.) o f r - f r a c t i o n s were m e a s u r e d o n the d e n s i t o g r a m , t a k i n g the / ] l - a l b u m i n d i s t a n c e as ttnltv It soon b e c a m e clear t h a t with u n c o n c e n t r a t e d C S F it was u n n e c e s s a r y to c o m p a r e with a s t a n d a r d s o l u t i o n o f a l b u m i n - t r a n s f e r r i n ( h u m a n a l b u m i n a n d t r a n s f e r r i n were purc h a s e d f r o m Sigma). T h e m o b i l i t y o f the a l b u m i n is the s a m e in C S F as in a s t a n d a r d s o l u t i o n o f purified a l b u m i n . N o C S F c o u l d be f o u n d with an a l b u m i n lransferrin d i s t a n c e which was m e a s u r a b l y different f r o m that in the s t a n d a r d solution. T h e classical increase o f the m o b i l i t y o f C S F a l b u m i n c o m p a r e d to serum a l b u m i n is a c o n c e n t r a t i o n artifact (Fig. 1 ). Nevertheless, all C S F s a m p l e s were elect r o p h o r e s e d t o g e t h e r with o t h e r s a m p l e s or with s e r u m s a m p l e s in a Pleuger I m m u n o p h o r (5 at a t i m e ) ; thus the o c c a s i o n a l [31 which h a d an a b n o r m a l m o b i l i t y was readily discovered when c o m p a r e d to the o t h e r d e n s i t o g r a m s o f the set.
B
O
Fig. l. Agar electropherograms of unconcentrated (A) and concentrated (B) CSF. The same sample of the same patient was used. Both A and B are electrophoresed along with a standard mixture of purified human albumin and transferrin in water. In A the mixture was loaded beside the main application cleft in the lower margin, in B the mixture is in the upper margin. The cathode is to lgft, the anode to the right. A : with the distance albumin-transferrin of the standard solution as 1.000 the albumin of the unconcentrated sample is at 1.000, there is a/~2-peak at -0.163 and a small r-peak at -0.230 B: the remainder of the sample was concentrated, electrophoresed and stained with Amido-Schwarz. The albumin peak is at 1.067 and the z-peak at -0.253. There is no tq2-peak.
279
ON THE T-PROTEIN IN CEREBROSPINAL FLUID
REsULTs
Histograms were made with the m r of the f12- and T-peaks plotted against the number of peaks (Fig. 2). Several facts are already obvious after this short series of experiments. The 44 CSF concentrates have their z-fraction with a mean m r o f about - 0 . 2 5 5 and a normal distribution. When the same 44 samples are electrophoresed unconcentrated, even the slowest m r o f t is nearer to fll than - 0.255 and the distribution is not normal at all. Furthermore, the 44 unconcentrated samples have 54 peaks: 10 proteinograms have 2 peaks in the fl2-z-region. The proteinograms o f the 44 concentrates have no fl2peaks, only one Z-peak each. Finally, even when the fl2-peaks are not taken into account, the distribution o f t h e mrs of the z-peaks is still not normal. This distribution can be better analysed when a larger series of pherograms is examined (Fig. 3). Here the f12- and z-peaks of 128 pherograms of unconcentrated CSF samples have their r n r plotted against their number. It seems that the histogram has its slowest peak at m r -0.230, with a shoulder at roughly -0.200. Another peak seems to be present at - 0.155 and nearer to fll the picture is confused. The quantitative importance of the ill-peak has a role when one is to discover a peak with an m r less than - 0 . 1 0 0 different. A small peak in that region near a huge fl~ will remain undiscovered. Therefore, one cannot say a lot about the region with an m r between 0 and - 0.125. Even so there are 4 clearly visible peaks : fll (mr = 0) and the peaks at - 0. 155, -0.200, - 0 . 2 3 0 and perhaps a fifth at -0.105. No
1 o -]
A
o 0
-0025 - 0 0 5 0
~ -0075 -0100
-0125
-0150 -0175
0
.0.025 -0.050 -0.075 - 0 . I 0 0
-0125
-0.150 - 0 7 7 5 - 0 2 0 0 - 0 2 2 5
-0200
0.225 -0.250 -0.275
mr
No
-0.250 -0.275
mr
Fig. 2. Histograms of the number of f12- and z-peaks plotted against the mr. A: 44 unconcentrated samples; B: concentrates of the same 44 samples. No 20 15 10 5 0 - 0 0 2 5 -(2050 -0.075
-0.100
-0.125 -0.150 -0.175 -0.200 -0.225 -0.250 -0 275
mr
Fig. 3. Histogram of the number of f12o and r-peaks from 128 unconcentrated samples, plotted against their mr.
280
p. VERHEECKE DISCUSSION
These observations are not in agreement with any explanation of the presence of r-proteins in CSF published up to now. Laterre's (1965) explanation does not fit because the presence o f a z-protein with a constant rn r is an in vitro artifact and not the consequence o f in vivo neuraminidase activity. Parker and Bearn's (1962) explanation does not fit either: although there are 3 intermediate peaks between 31 and r, they do not have the regular intervals between them which Parker and Bearn so elegantly showed in their experiments. There are probably even more peaks in the obscure region between 0 and - 0.125, not to mention the classical peak at - 0 . 2 5 5 which appears when CSF concentrates are electrophoresed. So there are far too many peaks to fit into Parker and Bearn's theory. Thus the question remains: why does part of the transferrin remain at the original [~lposition, whereas another part moves to slower regions? All possible explanations are to be divided into 2 groups. First, negatively-charged groups can be split off the molecule; second, positively-charged groups can be adsorbed to the molecule. Up to now the first group has had most of the attention. But not only neuraminidase must be looked at; other enzymes can split: off negatively-charged groups or even negatively-charged parts of the polypeptide molecule : it could just be trivial proteolytic activity. One argument in favour of this idea is that a further slowdown of the [~2 migrating band to r is apparently quicker than transition from/31 to 32- This is in favour of the idea that the 3z migrating bands are damaged molecules, more prone to digestion than whole transferrin molecules. Why then does this not happen in serum? Perhaps some inhibitors which are absent in CSF are present in serum or. on the contrary, and more probably, are the breakdown products broken down more quickly and thoroughly in serum'? The second group of possibilities, although less probable, must be kept in mind. However, no known positively-charged group is at this moment a likely candidate to be bound to transferrin. On the other hand, negatively-charged groups bound to albumin, which is known to bind small molecules easily, could account for the change of mr of that molecule.
CONCLUSION
This report makes two conclusions possible. The albumin in CSF has the same electrophoretic mobility as the serum albumin. The z-protein in vivo has no constant m r with a normal distribution when measured on many pherograms. Apparently it is a breakdown product of transferrin, some breakdown products (the classical zprotein) being more stable than others (the intermediate peaks). The mechanism is possibly proteolytic. ACKNOWLEDGEMENTS
I wish to thank Mrs. A. Vanneste and Mrs. De Smet for electrophoresing and staining the unconcentrated samples and Mrs. G. Van Soom for concentrating the CSF and
ON THE C-PROTEIN IN CEREBROSPINAL FLUID
281
performing the electrophoresing on the concentrates. I am grateful to Prof. A. LOwenthal for many useful discussions and to Prof. H. van der Eecken for constant support. SUMMARY
128 cerebrospinal fluid samples were obtained. About 5/d of each of the samples was subjected unconcentrated to agar gel electrophoresis. The remaining 44 of these samples were concentrated and the concentrates were electrophoresed. The proteins were stained. Analysis of the relative mobilities of the peaks makes two conclusions possible: the albumin in cerebrospinal fluid has the same electrophoretic mobility as purified serum albumin. The r-protein possibly arises from transferrin by proteolytic breakdown, some products (such as the classical z-protein) being less unstable than others (the intermediate peaks). REFERENCES CLAUSEN, J. AND T. MUNKNER (1961) Transferrin in normal cerebrospinal fluid, Nature (Lond.), 189: 6(~61. FRlCK. E. AND L. SCHEn)-SEYDEL (1963) Untersuchungen mit j~31 markiertem Transferrin zur Frage der Abstammung der Liquoreiweissk6rper, Klin. Wschr., 41 : 589-593. K~R~NVt, L. AND F. GALLYAS (1972) A highly sensitive method for demonstrating proteins in electrophoretic, immuno-electrophoretic and immunodiffusion preparations, Clin. chim. Acta, 38: 465-467. LATERRE, E. C. (1965) Les Prot~ines du Liquide COphalorachidien d I'Etat Normal et Pathologique, Arscia, Brussels, pp. 172-175. LOWENTHAL, A. (1964) Agar Gel Eleetrophoresis in Neurology, Elsevier, Amsterdam. PARKER, W. C. AND A. G. BEARN (1962) Studies on the transferrin of adult serum, cord serum and cerebrospinal fluid The effect of neuraminidase, J. exp. Med., 115:83 105. PETTE, D. AND I. STUPP (1960) Die r-Fraktion im Liquor cerebrospinalis, Kiln. Wschr., 38:109 110. VERHEECKE, P. (1974) Agar gel electrophoresis of unconcentrated cerebrospinal fluid--- The degenerative type, Acta neurol, bel¢l., 74: 376-382.
