Journal of Neuro,mmunology, 34 ( 1991 ) 61-67 © 1991 Elsevier Soence Pubhshers B V All rights reserved 0165-5728/91/$03 50
61
JNI 02046
Enhancement of antigen-specific T-cell reactivity on the affected side in stroke patients E h s a b e t h T a r k o w s k a 1,2, P e t e r E k e l u n d 2 a n d A n d r e j T a r k o w s l o 1,3 Departments of / Chnwal Immunology, 2 Genamcs and ~ Rheumatolog~, UnwerszO' of Goteborg, Goteborg, Sweden (Received 22 October 1990) (Revised, recewed 18 January 1991 and 29 Aprd 1991) (Accepted 30 April 1991)
Key words Stroke, T lymphocyte, B lymphocyte Delayed-type hypersensmvtty, Motor deficit, Sensory defiot
Summary We have analyzed the unpact of stroke with subsequent hemlparesls and sensory loss on in VlVO mediated unmune functions The delayed-type hypersensitivity ( D T H ) reaction to purified protein derlvate (PPD, tuberculin) was used as a measure of antigen-specific T-cell reactivity, and subcutaneous immunization with influenza vaccine was employed to evaluate T-cell-dependent B-cell function Thirtytwo of the 50 stroke patients tested displayed posttlve D T H reaction to PPD All but two showed equal or stronger D T H reaction on the paretlc arm c o m p a r e d to the contralateral side ( p < 00001) This stroke-reduced enhancement of D T H reactivity was evident in patients with combined motor and sensory deficit as well as in subjects with hemlparesls alone In contrast, immunization of stroke patients with influenza vaccine, a T-cell-dependent B-cell antigen, raised equal antigen-specific serum IgG, lgA and IgM antibody responses irrespective of side (paretlc or not paretic) We conclude that stroke enhances antigen-specific T-cell reactivity on the affected side of the body, and that motor but not sensory deficit seems to be required for this e n h a n c e m e n t Ant'lgenspecific B-cell reactivity was not significantly influenced by the hemiparesls
Introduction Multiple extrinsic and intrinsic factors a p p e a r to influence the physiology of the immune system D u n n g the last decade interest has been focused on interactions between the immune and nervous systems Clinical observations have shown
Address for correspondence Dr Ehsabeth Tarkowslo, Department of Chmcal Immunology, Guldhedsgatan 10, S-413 46 Goteborg, Sweden
that pattents with immunologlcally mediated disease such as rheumatoid arthritis (RA) experience amelioration of the disease activity on the paretlc side of the body during a stroke (Thompson et a l , 1962, Lewne et a l , 1987) This and other observations suggest a role for the central nervous system (CNS) in the regulation of the immune system Stroke caused by hemorrhage or thrombosis m a brain vessel is one of the most prevalent vascular diseases in industrialized countries The outcome of stroke will be a paresis a n d / o r loss of
62 sensitivity for e g pain and vibration m a contralateral s~de of the body However, it IS not known ff the loss of these brain functions will have any consequences on the abdlty of the immune system to respond to foreign s~gnals, such as infectious agents or tumors To study the contribution of the CNS to immune reactivity we have analyzed effects of stroke on in wvo mediated immunological functions m an elderly, but apart from cerebrovascular disease, essentially healthy population Delayed-type hypersensltlWty ( D T H ) reaction to tuberculin was chosen as a s~mple m VlVO test of antigen-specific T-cell function whereas analyses of antibody levels after active immunization with influenza vaccine were apphed to study T-cell-dependent B-cell function
to the manufacturer's recommendations Both arms of each study subject were exposed to 2 transmission umts (TU) of PPD and after 72 h the skin reaction was exammed The test was judged positive if their was an induration of more than 6 mm In diameter In case of absence of the reaction, intradermal administration of 5 T U of PPD was performed into both arms Two perpendicular dmmeters of the induration were measured and the sum of these two diameters was used to express the results (Dernevlk, 1984) To study a non-T-cell-mediated induction of vasod~latatton with weal and flare, 100 tzl of histamine (0 1 m g / m l ) was mjected intradermally into the lateral aspect of both arms in each patient The resulting response was recorded after 20 mm Two perpendicular dmmeters of the induration obtained were measured All the tests were performed and read by the same person
Materials and methods
h~fluenza t'accmatlon and et'aluatton of antibody response Thirteen stroke patients received a subcutaneous rejection with 0 5 ml of influenza vaccine (Varagnp, Rhone-Poulenc, Lyon, France) m the non-parettc arm and another 17 patients were ~mmumzed at the same place on the hemlplegic side Serum antibody levels of lgM. lgG and leA class against influenza were analyzed before and 14 days after the vaccination using enzyme-hnked lmmunosorbent assay (ELISA) Briefly, the ELISA test was performed in 96-well polystyrene mlcrotlter plates (Dynatech, Alexandria, VA, U S A ) Plates were coated overmght at 4 ° C with influenza vaccine diluted 1 250 in phosphate-buffered sahne (PBS), washed with PBS and blocked for 1 h with PBS-bovme serum albumin (BSA) 0 5,% The serum samples, dduted m PBS-BSA 0 5%, were added and after overmght incubation at 4 ° C , affinity purified F(ab') 2 fragments of goat anti-human lgG, IgA or lgM (Tago, Burhngame, CA, U S A ) diluted 1 3000 in PBSTween 20 followed by extravldm-horseradlsh peroxadase (05 ~ g / m l ) (Sigma, St Louis, MO, U S A ) were stepwlse apphed Enzyme-substrate solution included 2 5 m g / m l of 2,2-azmo-bm(3ethylbenzthmzohne sulfomc a o d ) (Sigma) m o trate buffer, pH 42, containing 00075% H 2 0 2 The absorbance was measured in a Titertek Multlscan spectrophotometer (Flow, McLean, VA, U S A ) at 414 nm The optical density (OD)
Patwnts Fifty subjects, 26 men and 24 women, 50-93 years old (mean age 7 7 + 9 years), who were patients at Vasa Hospital m Goteborg, were included in the study All patients had a history of cerebral stroke, which occurred between 4 weeks and 17 years before the start of the study Patients who had a history of multiple strokes with deficits on bot,1 sides of the body were excluded from the study Patients w~th malignant diseases, autolmmune diseases, severe infections or on lmmunosuppresstve drugs were also excluded All the patients were evaluated as to the degree of motor deficit m both arms by tests of the extension in the elbow and dorsofleraon of the hand Also, the strength of opposition between thumb and index and the ability to oppose the thumb to the other fingers of the same hand were examlned SensltWlty for touch, pain and vibration were tested in all except for four patients The study was approved by the ethical committee of the Unwerslty of G o t e b o r g Reagents and procedures Tubercuhn and h~stamm tests Purified protern denvatwe (PPD, tubercuhn) was obtained from the State Serum Institute (Copenhagen, D e n m a r k ) and rejected mtradermally according
b3 values m e a s u r e d m the a n t l g e n - s p e o f i c E L I S A were c o m p a r e d to the O D values o b t a m e d from a pool of reference sera All O D values were converted to antigen-specific E L I S A units using cahbratton curves based on the O D o b t a i n e d from serml dilutions of the reference serum The calibration curves were constructed using a c o m p u t e r program based on weighted Ioglt-log models (Russel et a l , 1986) The fold-increase (FI) values were calculated by dividing p o s t l m m u m z a t m n E L I S A umt values by the prelmmunlzatlon E L I S A unit values (Pedersen and Henrlchsen, 1982, Lue et a l , 1988) Skin biopsy In three stroke patients bdateral skin biopsies were obtained 72 h after the immunization with PPD from the central part of mduration Biopsy punch, 3 mm m diameter, was employed and the spectmens were kept in 4% formalin until paraffin e m b e d d i n g followed by cutting and staining with hematoxyhn-eosm Stausncs Statistical analysis regarding the differences between means was carried out by Student's p a i r e d two-tailed t-test A p-value < 0 05 was considered to be statistically slgmficant
Results
Chmcal findings All the patients included in this study displayed varying degree of hemlplegia of the central type The majority ( 3 5 / 5 0 ) had a complete paralysis Fifteen patients were paretlc but able to perform some voluntary movements Eighteen patients showed an Isolated paresis without loss
of sensory quahtles whereas the other subjects displayed loss of both strength and sensitivity In two cases, the patient had paralysis but could not collaborate in tests of the sensory quahttes and three patients were not tested for wbration A m o n g the 27 patients with d e c r e a s e d sensitivity, 16 showed a decrease of p r o p n o c e p t l o n and vibration indicating lesion in the posterior columns and at the same time, a decrease of pain sensation due to involvement of the lateral splnothalamlc tract Six patients had only loss of vibration and five only loss of pare The control arm of all the patients showed no signs of neurological defiot
T-cell reactu,tty Thirty-two of the 50 patients had a positive PPD reaction upon intradermal injection of 2 TU or 5 T U of tuberculin Thirty patients developed positive PPD reactions in both arms whereas one subject was posmve only on the paretlc side and a n o t h e r one only m the control arm Twenty-six of these 32 PPD-postttve patients showed a difference in D T H reactivity between the paretic and the other arm All but two of these 26 showed increased induration area in the paralyzed arm in comparison with the non-pareUc arm ( p < 0 0001) F u r t h e r statistical analysis revealed that significant increases of D T H reactivity on the p a r e u c side were observed both in the group of patients with combined motor and senspry loss ( p = 0 0 0 1 ) as well as in subjects with motor deficit alone ( p = 0 0 0 1 ) Table 1 illustrates that an e n h a n c e m e n t of antlgen-spec,flc T-cell reactivity occurred trrespecuve of the de-
TABLE 1 ANTIGEN-SPECIFIC T-CELL REACTIVITY TO PPD MEASURED BY DTH IN RELATION TO THE DEGREE OF PARESIS IN STROKE PATIENTS The numbers represent mean ± standard error of the mean (SEM) of individual differences between the affected and non-affected arm with regard to PPD response DTH react~wty is expressed in mm as the sum of two perpendicular diameters of induration No of patients
No of patients
No of patients
Differences m D T H
Level of
stud~ed
with posture DTH test
w~th enhanced DTH response on the affected
reactmty (affectednon-affected side)
significance
(ram, mean + SEM)
side Paresis Paralysis
15 35
8 24
8 19
8 1± 15 100 ± 2 1
p = 0001 p < 00001
64 TABLE 2 A N T I G E N - S P E C I F I C T-CELL REACTIVITY TO PPD, M E A S U R E D BY DTH, WITH RESPECT TO T H E TYPE O F SENSORY LOSS IN S T R O K E PATIENTS The numbers represent mean + standard error of the mean (SEM) of individual differences between the affected and non-affected arm with regard to PPD response DTH react~wty is expressed m mm as the sum of two perpendtcular dmmeters of induration Note that all the patients d~splayed mmultaneous paresis
Loss of v~brat~on Loss of pare Loss of pare and v~brat~on
No of patfents stud~ed
No of patients w~th posture DTH test
No of patients w~th enhanced DTH response on the affected rode
Differences m DTH react~wty (affectednon-affected rode) (mm mean + SEM)
Level of significance
6 5 16
4 3 9
4 3 7
5 8 5:1 7 I1 0 + 4 5 1145:39
p = 0 045 NS p=0019
gree o f parests T h e analysis of the D T H reactivity to P P D with respect to the type o f sensory loss dtd not s h o w major dtfferences b e t w e e n the affected arm and the control arm in patients w~th isolated loss of pain, loss of vibration, or c o m bined loss o f pare and vlbratton (Table 2) Patients with a lesion m the cerebral cortex s h o w e d stgnlficantly ( p = 0 035) increased D T H reactton m the paretic arm c o m p a r e d wtth the non-parettc arm Lemons l o c a h z e d m th~ basal gangha or below also gave rtse to signtficant differences (/9 = 0 0 2 5 ) in D T H reactton b e t w e e n the affected arm and the control arm Patients with both right- and left-sided hemtparests s h o w e d a
TABLE 3 DTH REACTIVITY TO PPD IN STROKE PATIENTS WITH RESPECT TO SIDE O F H E M I P A R E S I S The numbers represent mean +standard error of the mean (SEM) of md,v,dual d,fferences between the affected and non-affected arm w,th regard to PPD response DTH reactlv,tv *s expressed m mm as the sum of two perpend,cular dmmeters of mduratton S~de of parems
No of patients stud~ed
D~fferences m DTH reacttv~ty (affectednon-affected rode) (ram mean + SEM)
Level of mgnif~cance
R~ght Left
9 22
88+ 24 q 6-1-2 1
p = 0 005 p < 00001
~*.'f~
,
'"2._
.i
"' .~--~J~++
,
+
+
'
.~.+,,
++++,0 +
+
,°.+
.
+.+
~J~+ " Pdp.
Fig
."" ' +'+
~. "
++. . . .
-
,J
:
,,+++" ,,~,+.+,;+_+~.t,.+.__~+.++ . . . .
~ , , , : ~+,
'.++. ++, ~ - . . . . , + + : , '
~ + + / "
,,
+;+
+.L + " ,',':
,+"+', ' ++" + ~
~
~ ~
~
, ~
""
"
"=
"
" +~.
;
+~m,_+ '+
+/+ + + '
- +
,+-
'~ .
.
++, -,~.+,~'+' "
+.+'++
'
.
.
t
",
+
++
++,+++++
|+. ,
.
1 Mtcrophotographs of skdn punch blopmes from one stroke patient ~mmunized with PPD on the paret,c (left) and control side ( r i g h t ) Note pronounced focal infiltration of mononuclear cells m the paretJc arm Original magmficatlon × lO
65 TABLE 4 A N T I B O D Y LEVELS O F IgG, IgA A N D lgM CLASS T O I N F L U E N Z A V A C C I N E IN S T R O K E P A T I E N T S I M M U N I Z E D s c IN T H E P A R E T I C (n = 17) A N D N O N - P A R E T I C (n = 11) A R M Results are expressed m arbitrary umts (AU) as m e a n s 5: SD Immunization of Parettc arm (n = 17} N o n - p a r e t t c a r m ( n = 11)
IgG (AU)
IgM (AU)
lgA ( A U )
Day 0
Day 14
FI
Day 0
Day 14
FI
Day 0
Day 14
FI
58 5:35 3 8 ± 17
309 + 238 237+278
74 75
18 + 33 3+_ 3
103 5:163 825-110
30 23
18 + 23 14_-I: 15
106 -1- 16 71 5-42
10 3 84
FI = fold mcrease
significantly enhanced D T H reaction on the affected side (Table 3)
Htstopathology Bllareral punch biopsies were obtained 72 h after immunization with PPD Fig 1 demonstrates that mfiltratlon of mononuclear cells was considerably more pronounced on the paretlc s~de as compared to the control side
Htstamme-medmted weal and flare react,on To rule out whether the enhancement of PPD reaction m the paretlc arm was due to non-antigen-specific mechanisms, such as altered vascular responses, mtradermal injectton of histamine to both arms was assayed No significant differences between the paretlc arm (33 6 + 6 5 mm) and the control arm (34 6 + 5 4 mm) were observed after exposure to histamine
Anttgen-spec,fw antzbody response The level of serum antibodies of IgG, IgA and lgM classes to mfluenza was estimated on day 0, l e before the vaccination and 14 days later Seventeen patients received vaccination in the pareuc arm whereas 13 were immunized m the non-paretlc side S~gnfflcant differences wtth regard to antibody levels or fold increases for any of the Ig classes were not detected in the two groups of patients (Table 4)
Discussion Our results show that cerebral stroke with subsequent motor and sensory deficit significantly enhances antigen-specific T-cell reactivity to PPD
m VlVOon the parettc side, assessed by D T H test Stmdar results were obtained when an extract from Candtda albwans, another T-cell antigen, was administered lntradermally in a limited number of stroke patients (data not shown) In contrast, no differences were seen in h~stamme-mduced weal and flare reactions between the paretlc and non-paretlc side m stroke patients This is in agreement with Sonnenschem and Bernstem (1982) who demonstrated that a central paresis due to spinal cord transection did not change histamine flare As the h~stamme response was unaffected by stroke and hemlplegia thls suggests that the vascular mechamsms governing the inflammatory response are left intact We believe that tmmunologlcally mediated production of inflammatory medmtors on the paretlc side of the body may be enhanced by as yet unknown mechanisms during the stroke Recently, Czlonkowska et al (1987) showed that the features of immunological changes during stroke are considerably influenced by the Iocahzatlon of the lesion m the right or left hemisphere The lesion in the right hermsphere changed the T-helper/T-suppressor ratio, whde a decrease of the B-cell subset developed m cases with involvement of the left hemtsphere In contrast, our results show slmdar degrees of enhancement of D T H reactivity irrespective of the side of the brain leston Motor deficit alone - - as well as combined motor and sensory deficit - - led to enhanced D T H reactivity on the affected side of the body Furthermore, neither degree of motor deficit nor quahty of sensory deficit influenced slgmflcantly this enhancement Th~s findings ~mply that paresis of the central type is the only mantfestat~on
60 reqmred for the local enhancement of the D T H reaction on the aftected ann Alternatwely, paresis-reduced immoblhzatlon of the hmb might have contributed to the altered ~mmunolog~cal reactwlty by e g longer t~me movements through lymphatics However, the phenolsulfophthalem clearance time from the skin ~s s~mdar in normal and hemlpleg~c hmbs (Mlzush~ma and Yamamura, 1969) In addition, our recent study demonstrates that peripheral dener~at~on of scmt~c nerve suppresses rather than enhances D T H (Josefsson et a l , 1990) Thus. ~t seems reasonable to assume that the cause of enhanced D T H reaction ~s due to damage of upper motor neurons or other structures m the close V~Clmty to these neurons In stroke, the balance between excitatory and inhibitory descending pathways which influence lower motor neuron m the anterior horn of spinal cord ~s altered (Adams and V~ctor, 1981) The result wdi be a paresis with an increased flexor muscle tonus on the affected s~de leading to spastlclty (Lance. 1980, Lance and de Gad, 1980) All the patients included m our study showed both paresis and s~gns of at least mdd spastlClty as assessed by either ~solated e n h a n c e m e n t of peripheral reflexes a n d / o r an increased tonus of the muscles m the affected hmb If the alpha neurons are revolved m the development of spastlC~ty ~t could result m an increased release of acetylchohne m the periphery Acetylchohne Itself could influence the local T-cell responses (Atweh et a l , 1984) It has been claimed that acetylchohne-producmg neurons contain also other neuropept~des such as substance P, calc~tonln gene-related polypeptlde ( C G R P ) and serotonm, and that they are released at the same ttme as acetylchohne (Cooper and Martin. 1986, Inagak~ and l(ato, 1986) Substance P is known to enhance T-lymphocyte prohferatlon by binding on specific receptors on lymphocytes (Stantsz et a l , 1987) and to enhance actw~ty of macrophages (Peck, 1987) which are important effector cells m D T H reaction Recent cytochem~cal studies revealed that both substance P and C G R P are present m nerves adjacent to mast cells (Stead et a l , 1987) These findings suggest that these pept~des released from axonal vancos~t~es could cause degranulat~on of mast cells Thus, locally mcreased levels of substance P a n d / o r C G R P could
have been one of the mechamsms leading to enhanced D T H reaction m stroke Another possible explanation for the enhancement of D T H reaction m stroke could be downregulatton of the sympathetic mnervatton on the paretlc side Such a downregulatlon could lead to vasoddatatlon facdltatmg the influx of Tlymphocytes and macrophages into the area of D T H reaction However, decrease m sympathetic tone has been shown to downregulate rather than upregulate T-cell-dependent and -independent inflammatory responses (Ramaswamy et al. 1990. Ahto et a i , 1987) In contrast to the clear-cut effects of stroke on the antigen-specific T-cell reactivity, the T-celldependent B-cell reactwtty was not stgnlficantly affected Th~s d~chotomous effect of stroke on tmmune responsiveness might be explained by dffferentml reqmrements of T- and B-cells, respectively More probably, whereas B-cell responses are going on and are detected both locally in regional lymph nodes but also systemically m e g peripheral blood, spleen and bonemarrow, the D T H recall response ts strictly localtzed to the area of antlgemc challenge The latter hmttatlon tmphes that apart from the lmmunoIogtc recognition process also non-specific secondary proflogmttc p h e n o m e n a m~ght contribute m the enhancement of D T H reaction m stroke Studtes are presently going on tn our laboratory. to elucidate the mechamsms influencing the m vlvo enhancement of T-cell reacttvtty m stroke
Acknowledgments We thank Mrs Lena Svensson and I n g - M a n e NiIsson for excellent technical assistance We thank Drs I Milsom and G Walhn for critical reading of the manuscript Thts study was supported by grants from the Goteborg Medlcai Society, the Swedish Medical Society, the Swedish Assocmtton against Rheumatism, the Kang Gustaf V's 80 Years Foundation, Stroke Association. and the Swedish Medical Research Councd
67
References Adams R D and Victor M (1981) Motor paralysis In Principles of Neurology McGraw-Hall New York p 39 Ahto, A B, Romeo, H E Baler, R Chuluyan H E Braun, M and Cardmah D P (1987) Autonomic nervous system regulation of murme Immune responses as assessed by local surgical sympathetic and parasympathetic denervatlon a~cta Physlol Pharmacol Latmoam 37, 305 Atweh, S P Grayhack, M S and Richman, D P (1984) A chohnerg~c receptor s~te on murme Ivmphocvtes with novel binding characteristics Life So 35 2459-2469 Cooper P and Martin J (1986)Neuropepttde functions In A K Asbut3 G M McKhan and W I McDonald (Eds) Diseases of the Nervous System, Saunders Phdadelphm, PA pp 1-73 Czlonkowska, A , Kodak, J and Kuczynska-Zardzewtaly, A (1987) Lymphocyte subsets after stroke J Neurolmmunol 16, 40 (abstract) Dernewk, L (1984) Shrinking Pleuntts with Atelectasls A Clinical Study on Operated patients with Reference to Etiology, Pathogenests Rad,ology and Immunology Thesis Goteborg Umverslty, Goteborg Inagalo, S and IOto S (1986) Peptldes m the peripheral nervous system in P C Emsom M Rossor and M Tohyama (Eds) Peptides and Neurological Disease Progress m Brain Research Vol 66, Elsevier Amsterdam pp 269-316 Josefsson E Carlsten, H , Lange, S and Tarkowskl, A (in press) Peripheral denervatlon suppresses the late phase of delayed-type hypersensmvlty Int Arch Allergy Appl immunol Lance J W (1980)Symposium synopsis In R G Feldman R R Young and W P Koella (Eds) 8pastloty Disordered Motor Control Year Book Publishers Chicago, IL pp 485-494 Lance J W and de Gad, P (1980) The control of muscle tone reflexes and movement Robert Wartenberg Lecture Neurology 30, 1303-1313
Levme J Goetzl E and Basbaum, A (1987) Contribution of the nerxous system to the pathophystology of rheumato=d arthritis and other polyarthnhdes Rheum Dis Chn North Am 13 369-383 Lue, C Tarkov~skl, A and Mestecky, J (1988) S}stemtc immumzat=on ~,lth pneumococcal polysaccharlde vacc=ne induces a predominant IgA2 response of peripheral blood lymphocytes and increases of both serum and secretor) antl-pneumococcal antibodies J Immunol 140, 3793 Mizushlma Y and Yamamura, M (1969) Arthropathy and inflammation reaction in hem~pleg=cpatients Acta Rheum Scand 15,297 Peck, R (1987) Neuropeptldes modulatmg macrophage function Ann N Y Acad So 496, 265-270 Pedersen, F K and Hennchsen, J (1982) Detection of antibodies to pneumococcal capsular polysacchandes by enzyme-hnked lmmunosorbent assay J Chn Microblol 15 372 Ramaswamy K Mathison R , Carter, L, IOrk, D , Green F Da~lson J S and Befus D (1990)Marked antnnflammatory effects of decentralization of the superior cervical ganglia J Exp Med 172 1819-1830 Russel, M W , Brown T A Radl J Haauman, J J and Mesteclcy, J (1986) Assay of human IgA subclass antibodies in serum and secretions by means of monoclonal antibodies J Immunol Methods 87 87 Sonnenschem, R and Bernstem, M (1957) Relation to the central nervous system of neural pathways mediating histamm flare and nicotine sweating J Appl Phys II 481485 Stamsz, A , Socchltano R Pa}an D and B=enenstock, J (1987) In vitro studies of immunoregulation by substance P and somatostatin Ann N Y Acad Scl 496 217-225 Stead R H Bienenstock, J and Stanisz, A M (1987) Neuropepttde regulation of mucosal immunity lmmunol Rev 100, 333-359 Thompson M and Bywaters E G L (1962) Unilateral arthrit,s following hemlplegm Ann Rheum Dis 21 370
61
JNI 02046
Enhancement of antigen-specific T-cell reactivity on the affected side in stroke patients E h s a b e t h T a r k o w s k a 1,2, P e t e r E k e l u n d 2 a n d A n d r e j T a r k o w s l o 1,3 Departments of / Chnwal Immunology, 2 Genamcs and ~ Rheumatolog~, UnwerszO' of Goteborg, Goteborg, Sweden (Received 22 October 1990) (Revised, recewed 18 January 1991 and 29 Aprd 1991) (Accepted 30 April 1991)
Key words Stroke, T lymphocyte, B lymphocyte Delayed-type hypersensmvtty, Motor deficit, Sensory defiot
Summary We have analyzed the unpact of stroke with subsequent hemlparesls and sensory loss on in VlVO mediated unmune functions The delayed-type hypersensitivity ( D T H ) reaction to purified protein derlvate (PPD, tuberculin) was used as a measure of antigen-specific T-cell reactivity, and subcutaneous immunization with influenza vaccine was employed to evaluate T-cell-dependent B-cell function Thirtytwo of the 50 stroke patients tested displayed posttlve D T H reaction to PPD All but two showed equal or stronger D T H reaction on the paretlc arm c o m p a r e d to the contralateral side ( p < 00001) This stroke-reduced enhancement of D T H reactivity was evident in patients with combined motor and sensory deficit as well as in subjects with hemlparesls alone In contrast, immunization of stroke patients with influenza vaccine, a T-cell-dependent B-cell antigen, raised equal antigen-specific serum IgG, lgA and IgM antibody responses irrespective of side (paretlc or not paretic) We conclude that stroke enhances antigen-specific T-cell reactivity on the affected side of the body, and that motor but not sensory deficit seems to be required for this e n h a n c e m e n t Ant'lgenspecific B-cell reactivity was not significantly influenced by the hemiparesls
Introduction Multiple extrinsic and intrinsic factors a p p e a r to influence the physiology of the immune system D u n n g the last decade interest has been focused on interactions between the immune and nervous systems Clinical observations have shown
Address for correspondence Dr Ehsabeth Tarkowslo, Department of Chmcal Immunology, Guldhedsgatan 10, S-413 46 Goteborg, Sweden
that pattents with immunologlcally mediated disease such as rheumatoid arthritis (RA) experience amelioration of the disease activity on the paretlc side of the body during a stroke (Thompson et a l , 1962, Lewne et a l , 1987) This and other observations suggest a role for the central nervous system (CNS) in the regulation of the immune system Stroke caused by hemorrhage or thrombosis m a brain vessel is one of the most prevalent vascular diseases in industrialized countries The outcome of stroke will be a paresis a n d / o r loss of
62 sensitivity for e g pain and vibration m a contralateral s~de of the body However, it IS not known ff the loss of these brain functions will have any consequences on the abdlty of the immune system to respond to foreign s~gnals, such as infectious agents or tumors To study the contribution of the CNS to immune reactivity we have analyzed effects of stroke on in wvo mediated immunological functions m an elderly, but apart from cerebrovascular disease, essentially healthy population Delayed-type hypersensltlWty ( D T H ) reaction to tuberculin was chosen as a s~mple m VlVO test of antigen-specific T-cell function whereas analyses of antibody levels after active immunization with influenza vaccine were apphed to study T-cell-dependent B-cell function
to the manufacturer's recommendations Both arms of each study subject were exposed to 2 transmission umts (TU) of PPD and after 72 h the skin reaction was exammed The test was judged positive if their was an induration of more than 6 mm In diameter In case of absence of the reaction, intradermal administration of 5 T U of PPD was performed into both arms Two perpendicular dmmeters of the induration were measured and the sum of these two diameters was used to express the results (Dernevlk, 1984) To study a non-T-cell-mediated induction of vasod~latatton with weal and flare, 100 tzl of histamine (0 1 m g / m l ) was mjected intradermally into the lateral aspect of both arms in each patient The resulting response was recorded after 20 mm Two perpendicular dmmeters of the induration obtained were measured All the tests were performed and read by the same person
Materials and methods
h~fluenza t'accmatlon and et'aluatton of antibody response Thirteen stroke patients received a subcutaneous rejection with 0 5 ml of influenza vaccine (Varagnp, Rhone-Poulenc, Lyon, France) m the non-parettc arm and another 17 patients were ~mmumzed at the same place on the hemlplegic side Serum antibody levels of lgM. lgG and leA class against influenza were analyzed before and 14 days after the vaccination using enzyme-hnked lmmunosorbent assay (ELISA) Briefly, the ELISA test was performed in 96-well polystyrene mlcrotlter plates (Dynatech, Alexandria, VA, U S A ) Plates were coated overmght at 4 ° C with influenza vaccine diluted 1 250 in phosphate-buffered sahne (PBS), washed with PBS and blocked for 1 h with PBS-bovme serum albumin (BSA) 0 5,% The serum samples, dduted m PBS-BSA 0 5%, were added and after overmght incubation at 4 ° C , affinity purified F(ab') 2 fragments of goat anti-human lgG, IgA or lgM (Tago, Burhngame, CA, U S A ) diluted 1 3000 in PBSTween 20 followed by extravldm-horseradlsh peroxadase (05 ~ g / m l ) (Sigma, St Louis, MO, U S A ) were stepwlse apphed Enzyme-substrate solution included 2 5 m g / m l of 2,2-azmo-bm(3ethylbenzthmzohne sulfomc a o d ) (Sigma) m o trate buffer, pH 42, containing 00075% H 2 0 2 The absorbance was measured in a Titertek Multlscan spectrophotometer (Flow, McLean, VA, U S A ) at 414 nm The optical density (OD)
Patwnts Fifty subjects, 26 men and 24 women, 50-93 years old (mean age 7 7 + 9 years), who were patients at Vasa Hospital m Goteborg, were included in the study All patients had a history of cerebral stroke, which occurred between 4 weeks and 17 years before the start of the study Patients who had a history of multiple strokes with deficits on bot,1 sides of the body were excluded from the study Patients w~th malignant diseases, autolmmune diseases, severe infections or on lmmunosuppresstve drugs were also excluded All the patients were evaluated as to the degree of motor deficit m both arms by tests of the extension in the elbow and dorsofleraon of the hand Also, the strength of opposition between thumb and index and the ability to oppose the thumb to the other fingers of the same hand were examlned SensltWlty for touch, pain and vibration were tested in all except for four patients The study was approved by the ethical committee of the Unwerslty of G o t e b o r g Reagents and procedures Tubercuhn and h~stamm tests Purified protern denvatwe (PPD, tubercuhn) was obtained from the State Serum Institute (Copenhagen, D e n m a r k ) and rejected mtradermally according
b3 values m e a s u r e d m the a n t l g e n - s p e o f i c E L I S A were c o m p a r e d to the O D values o b t a m e d from a pool of reference sera All O D values were converted to antigen-specific E L I S A units using cahbratton curves based on the O D o b t a i n e d from serml dilutions of the reference serum The calibration curves were constructed using a c o m p u t e r program based on weighted Ioglt-log models (Russel et a l , 1986) The fold-increase (FI) values were calculated by dividing p o s t l m m u m z a t m n E L I S A umt values by the prelmmunlzatlon E L I S A unit values (Pedersen and Henrlchsen, 1982, Lue et a l , 1988) Skin biopsy In three stroke patients bdateral skin biopsies were obtained 72 h after the immunization with PPD from the central part of mduration Biopsy punch, 3 mm m diameter, was employed and the spectmens were kept in 4% formalin until paraffin e m b e d d i n g followed by cutting and staining with hematoxyhn-eosm Stausncs Statistical analysis regarding the differences between means was carried out by Student's p a i r e d two-tailed t-test A p-value < 0 05 was considered to be statistically slgmficant
Results
Chmcal findings All the patients included in this study displayed varying degree of hemlplegia of the central type The majority ( 3 5 / 5 0 ) had a complete paralysis Fifteen patients were paretlc but able to perform some voluntary movements Eighteen patients showed an Isolated paresis without loss
of sensory quahtles whereas the other subjects displayed loss of both strength and sensitivity In two cases, the patient had paralysis but could not collaborate in tests of the sensory quahttes and three patients were not tested for wbration A m o n g the 27 patients with d e c r e a s e d sensitivity, 16 showed a decrease of p r o p n o c e p t l o n and vibration indicating lesion in the posterior columns and at the same time, a decrease of pain sensation due to involvement of the lateral splnothalamlc tract Six patients had only loss of vibration and five only loss of pare The control arm of all the patients showed no signs of neurological defiot
T-cell reactu,tty Thirty-two of the 50 patients had a positive PPD reaction upon intradermal injection of 2 TU or 5 T U of tuberculin Thirty patients developed positive PPD reactions in both arms whereas one subject was posmve only on the paretlc side and a n o t h e r one only m the control arm Twenty-six of these 32 PPD-postttve patients showed a difference in D T H reactivity between the paretic and the other arm All but two of these 26 showed increased induration area in the paralyzed arm in comparison with the non-pareUc arm ( p < 0 0001) F u r t h e r statistical analysis revealed that significant increases of D T H reactivity on the p a r e u c side were observed both in the group of patients with combined motor and senspry loss ( p = 0 0 0 1 ) as well as in subjects with motor deficit alone ( p = 0 0 0 1 ) Table 1 illustrates that an e n h a n c e m e n t of antlgen-spec,flc T-cell reactivity occurred trrespecuve of the de-
TABLE 1 ANTIGEN-SPECIFIC T-CELL REACTIVITY TO PPD MEASURED BY DTH IN RELATION TO THE DEGREE OF PARESIS IN STROKE PATIENTS The numbers represent mean ± standard error of the mean (SEM) of individual differences between the affected and non-affected arm with regard to PPD response DTH react~wty is expressed in mm as the sum of two perpendicular diameters of induration No of patients
No of patients
No of patients
Differences m D T H
Level of
stud~ed
with posture DTH test
w~th enhanced DTH response on the affected
reactmty (affectednon-affected side)
significance
(ram, mean + SEM)
side Paresis Paralysis
15 35
8 24
8 19
8 1± 15 100 ± 2 1
p = 0001 p < 00001
64 TABLE 2 A N T I G E N - S P E C I F I C T-CELL REACTIVITY TO PPD, M E A S U R E D BY DTH, WITH RESPECT TO T H E TYPE O F SENSORY LOSS IN S T R O K E PATIENTS The numbers represent mean + standard error of the mean (SEM) of individual differences between the affected and non-affected arm with regard to PPD response DTH react~wty is expressed m mm as the sum of two perpendtcular dmmeters of induration Note that all the patients d~splayed mmultaneous paresis
Loss of v~brat~on Loss of pare Loss of pare and v~brat~on
No of patfents stud~ed
No of patients w~th posture DTH test
No of patients w~th enhanced DTH response on the affected rode
Differences m DTH react~wty (affectednon-affected rode) (mm mean + SEM)
Level of significance
6 5 16
4 3 9
4 3 7
5 8 5:1 7 I1 0 + 4 5 1145:39
p = 0 045 NS p=0019
gree o f parests T h e analysis of the D T H reactivity to P P D with respect to the type o f sensory loss dtd not s h o w major dtfferences b e t w e e n the affected arm and the control arm in patients w~th isolated loss of pain, loss of vibration, or c o m bined loss o f pare and vlbratton (Table 2) Patients with a lesion m the cerebral cortex s h o w e d stgnlficantly ( p = 0 035) increased D T H reactton m the paretic arm c o m p a r e d wtth the non-parettc arm Lemons l o c a h z e d m th~ basal gangha or below also gave rtse to signtficant differences (/9 = 0 0 2 5 ) in D T H reactton b e t w e e n the affected arm and the control arm Patients with both right- and left-sided hemtparests s h o w e d a
TABLE 3 DTH REACTIVITY TO PPD IN STROKE PATIENTS WITH RESPECT TO SIDE O F H E M I P A R E S I S The numbers represent mean +standard error of the mean (SEM) of md,v,dual d,fferences between the affected and non-affected arm w,th regard to PPD response DTH reactlv,tv *s expressed m mm as the sum of two perpend,cular dmmeters of mduratton S~de of parems
No of patients stud~ed
D~fferences m DTH reacttv~ty (affectednon-affected rode) (ram mean + SEM)
Level of mgnif~cance
R~ght Left
9 22
88+ 24 q 6-1-2 1
p = 0 005 p < 00001
~*.'f~
,
'"2._
.i
"' .~--~J~++
,
+
+
'
.~.+,,
++++,0 +
+
,°.+
.
+.+
~J~+ " Pdp.
Fig
."" ' +'+
~. "
++. . . .
-
,J
:
,,+++" ,,~,+.+,;+_+~.t,.+.__~+.++ . . . .
~ , , , : ~+,
'.++. ++, ~ - . . . . , + + : , '
~ + + / "
,,
+;+
+.L + " ,',':
,+"+', ' ++" + ~
~
~ ~
~
, ~
""
"
"=
"
" +~.
;
+~m,_+ '+
+/+ + + '
- +
,+-
'~ .
.
++, -,~.+,~'+' "
+.+'++
'
.
.
t
",
+
++
++,+++++
|+. ,
.
1 Mtcrophotographs of skdn punch blopmes from one stroke patient ~mmunized with PPD on the paret,c (left) and control side ( r i g h t ) Note pronounced focal infiltration of mononuclear cells m the paretJc arm Original magmficatlon × lO
65 TABLE 4 A N T I B O D Y LEVELS O F IgG, IgA A N D lgM CLASS T O I N F L U E N Z A V A C C I N E IN S T R O K E P A T I E N T S I M M U N I Z E D s c IN T H E P A R E T I C (n = 17) A N D N O N - P A R E T I C (n = 11) A R M Results are expressed m arbitrary umts (AU) as m e a n s 5: SD Immunization of Parettc arm (n = 17} N o n - p a r e t t c a r m ( n = 11)
IgG (AU)
IgM (AU)
lgA ( A U )
Day 0
Day 14
FI
Day 0
Day 14
FI
Day 0
Day 14
FI
58 5:35 3 8 ± 17
309 + 238 237+278
74 75
18 + 33 3+_ 3
103 5:163 825-110
30 23
18 + 23 14_-I: 15
106 -1- 16 71 5-42
10 3 84
FI = fold mcrease
significantly enhanced D T H reaction on the affected side (Table 3)
Htstopathology Bllareral punch biopsies were obtained 72 h after immunization with PPD Fig 1 demonstrates that mfiltratlon of mononuclear cells was considerably more pronounced on the paretlc s~de as compared to the control side
Htstamme-medmted weal and flare react,on To rule out whether the enhancement of PPD reaction m the paretlc arm was due to non-antigen-specific mechanisms, such as altered vascular responses, mtradermal injectton of histamine to both arms was assayed No significant differences between the paretlc arm (33 6 + 6 5 mm) and the control arm (34 6 + 5 4 mm) were observed after exposure to histamine
Anttgen-spec,fw antzbody response The level of serum antibodies of IgG, IgA and lgM classes to mfluenza was estimated on day 0, l e before the vaccination and 14 days later Seventeen patients received vaccination in the pareuc arm whereas 13 were immunized m the non-paretlc side S~gnfflcant differences wtth regard to antibody levels or fold increases for any of the Ig classes were not detected in the two groups of patients (Table 4)
Discussion Our results show that cerebral stroke with subsequent motor and sensory deficit significantly enhances antigen-specific T-cell reactivity to PPD
m VlVOon the parettc side, assessed by D T H test Stmdar results were obtained when an extract from Candtda albwans, another T-cell antigen, was administered lntradermally in a limited number of stroke patients (data not shown) In contrast, no differences were seen in h~stamme-mduced weal and flare reactions between the paretlc and non-paretlc side m stroke patients This is in agreement with Sonnenschem and Bernstem (1982) who demonstrated that a central paresis due to spinal cord transection did not change histamine flare As the h~stamme response was unaffected by stroke and hemlplegia thls suggests that the vascular mechamsms governing the inflammatory response are left intact We believe that tmmunologlcally mediated production of inflammatory medmtors on the paretlc side of the body may be enhanced by as yet unknown mechanisms during the stroke Recently, Czlonkowska et al (1987) showed that the features of immunological changes during stroke are considerably influenced by the Iocahzatlon of the lesion m the right or left hemisphere The lesion in the right hermsphere changed the T-helper/T-suppressor ratio, whde a decrease of the B-cell subset developed m cases with involvement of the left hemtsphere In contrast, our results show slmdar degrees of enhancement of D T H reactivity irrespective of the side of the brain leston Motor deficit alone - - as well as combined motor and sensory deficit - - led to enhanced D T H reactivity on the affected side of the body Furthermore, neither degree of motor deficit nor quahty of sensory deficit influenced slgmflcantly this enhancement Th~s findings ~mply that paresis of the central type is the only mantfestat~on
60 reqmred for the local enhancement of the D T H reaction on the aftected ann Alternatwely, paresis-reduced immoblhzatlon of the hmb might have contributed to the altered ~mmunolog~cal reactwlty by e g longer t~me movements through lymphatics However, the phenolsulfophthalem clearance time from the skin ~s s~mdar in normal and hemlpleg~c hmbs (Mlzush~ma and Yamamura, 1969) In addition, our recent study demonstrates that peripheral dener~at~on of scmt~c nerve suppresses rather than enhances D T H (Josefsson et a l , 1990) Thus. ~t seems reasonable to assume that the cause of enhanced D T H reaction ~s due to damage of upper motor neurons or other structures m the close V~Clmty to these neurons In stroke, the balance between excitatory and inhibitory descending pathways which influence lower motor neuron m the anterior horn of spinal cord ~s altered (Adams and V~ctor, 1981) The result wdi be a paresis with an increased flexor muscle tonus on the affected s~de leading to spastlclty (Lance. 1980, Lance and de Gad, 1980) All the patients included m our study showed both paresis and s~gns of at least mdd spastlClty as assessed by either ~solated e n h a n c e m e n t of peripheral reflexes a n d / o r an increased tonus of the muscles m the affected hmb If the alpha neurons are revolved m the development of spastlC~ty ~t could result m an increased release of acetylchohne m the periphery Acetylchohne Itself could influence the local T-cell responses (Atweh et a l , 1984) It has been claimed that acetylchohne-producmg neurons contain also other neuropept~des such as substance P, calc~tonln gene-related polypeptlde ( C G R P ) and serotonm, and that they are released at the same ttme as acetylchohne (Cooper and Martin. 1986, Inagak~ and l(ato, 1986) Substance P is known to enhance T-lymphocyte prohferatlon by binding on specific receptors on lymphocytes (Stantsz et a l , 1987) and to enhance actw~ty of macrophages (Peck, 1987) which are important effector cells m D T H reaction Recent cytochem~cal studies revealed that both substance P and C G R P are present m nerves adjacent to mast cells (Stead et a l , 1987) These findings suggest that these pept~des released from axonal vancos~t~es could cause degranulat~on of mast cells Thus, locally mcreased levels of substance P a n d / o r C G R P could
have been one of the mechamsms leading to enhanced D T H reaction m stroke Another possible explanation for the enhancement of D T H reaction m stroke could be downregulatton of the sympathetic mnervatton on the paretlc side Such a downregulatlon could lead to vasoddatatlon facdltatmg the influx of Tlymphocytes and macrophages into the area of D T H reaction However, decrease m sympathetic tone has been shown to downregulate rather than upregulate T-cell-dependent and -independent inflammatory responses (Ramaswamy et al. 1990. Ahto et a i , 1987) In contrast to the clear-cut effects of stroke on the antigen-specific T-cell reactivity, the T-celldependent B-cell reactwtty was not stgnlficantly affected Th~s d~chotomous effect of stroke on tmmune responsiveness might be explained by dffferentml reqmrements of T- and B-cells, respectively More probably, whereas B-cell responses are going on and are detected both locally in regional lymph nodes but also systemically m e g peripheral blood, spleen and bonemarrow, the D T H recall response ts strictly localtzed to the area of antlgemc challenge The latter hmttatlon tmphes that apart from the lmmunoIogtc recognition process also non-specific secondary proflogmttc p h e n o m e n a m~ght contribute m the enhancement of D T H reaction m stroke Studtes are presently going on tn our laboratory. to elucidate the mechamsms influencing the m vlvo enhancement of T-cell reacttvtty m stroke
Acknowledgments We thank Mrs Lena Svensson and I n g - M a n e NiIsson for excellent technical assistance We thank Drs I Milsom and G Walhn for critical reading of the manuscript Thts study was supported by grants from the Goteborg Medlcai Society, the Swedish Medical Society, the Swedish Assocmtton against Rheumatism, the Kang Gustaf V's 80 Years Foundation, Stroke Association. and the Swedish Medical Research Councd
67
References Adams R D and Victor M (1981) Motor paralysis In Principles of Neurology McGraw-Hall New York p 39 Ahto, A B, Romeo, H E Baler, R Chuluyan H E Braun, M and Cardmah D P (1987) Autonomic nervous system regulation of murme Immune responses as assessed by local surgical sympathetic and parasympathetic denervatlon a~cta Physlol Pharmacol Latmoam 37, 305 Atweh, S P Grayhack, M S and Richman, D P (1984) A chohnerg~c receptor s~te on murme Ivmphocvtes with novel binding characteristics Life So 35 2459-2469 Cooper P and Martin J (1986)Neuropepttde functions In A K Asbut3 G M McKhan and W I McDonald (Eds) Diseases of the Nervous System, Saunders Phdadelphm, PA pp 1-73 Czlonkowska, A , Kodak, J and Kuczynska-Zardzewtaly, A (1987) Lymphocyte subsets after stroke J Neurolmmunol 16, 40 (abstract) Dernewk, L (1984) Shrinking Pleuntts with Atelectasls A Clinical Study on Operated patients with Reference to Etiology, Pathogenests Rad,ology and Immunology Thesis Goteborg Umverslty, Goteborg Inagalo, S and IOto S (1986) Peptldes m the peripheral nervous system in P C Emsom M Rossor and M Tohyama (Eds) Peptides and Neurological Disease Progress m Brain Research Vol 66, Elsevier Amsterdam pp 269-316 Josefsson E Carlsten, H , Lange, S and Tarkowskl, A (in press) Peripheral denervatlon suppresses the late phase of delayed-type hypersensmvlty Int Arch Allergy Appl immunol Lance J W (1980)Symposium synopsis In R G Feldman R R Young and W P Koella (Eds) 8pastloty Disordered Motor Control Year Book Publishers Chicago, IL pp 485-494 Lance J W and de Gad, P (1980) The control of muscle tone reflexes and movement Robert Wartenberg Lecture Neurology 30, 1303-1313
Levme J Goetzl E and Basbaum, A (1987) Contribution of the nerxous system to the pathophystology of rheumato=d arthritis and other polyarthnhdes Rheum Dis Chn North Am 13 369-383 Lue, C Tarkov~skl, A and Mestecky, J (1988) S}stemtc immumzat=on ~,lth pneumococcal polysaccharlde vacc=ne induces a predominant IgA2 response of peripheral blood lymphocytes and increases of both serum and secretor) antl-pneumococcal antibodies J Immunol 140, 3793 Mizushlma Y and Yamamura, M (1969) Arthropathy and inflammation reaction in hem~pleg=cpatients Acta Rheum Scand 15,297 Peck, R (1987) Neuropeptldes modulatmg macrophage function Ann N Y Acad So 496, 265-270 Pedersen, F K and Hennchsen, J (1982) Detection of antibodies to pneumococcal capsular polysacchandes by enzyme-hnked lmmunosorbent assay J Chn Microblol 15 372 Ramaswamy K Mathison R , Carter, L, IOrk, D , Green F Da~lson J S and Befus D (1990)Marked antnnflammatory effects of decentralization of the superior cervical ganglia J Exp Med 172 1819-1830 Russel, M W , Brown T A Radl J Haauman, J J and Mesteclcy, J (1986) Assay of human IgA subclass antibodies in serum and secretions by means of monoclonal antibodies J Immunol Methods 87 87 Sonnenschem, R and Bernstem, M (1957) Relation to the central nervous system of neural pathways mediating histamm flare and nicotine sweating J Appl Phys II 481485 Stamsz, A , Socchltano R Pa}an D and B=enenstock, J (1987) In vitro studies of immunoregulation by substance P and somatostatin Ann N Y Acad Scl 496 217-225 Stead R H Bienenstock, J and Stanisz, A M (1987) Neuropepttde regulation of mucosal immunity lmmunol Rev 100, 333-359 Thompson M and Bywaters E G L (1962) Unilateral arthrit,s following hemlplegm Ann Rheum Dis 21 370
