Immunohistological analysis of T lymphocyte subsets in the central nervous system in chronic progressive multiple sclerosis

Journal of the Neurologieal Sciences, 1983, 62:219-232

219

Elsevier

IMMUNOHISTOLOGICAL ANALYSIS OF T LYMPHOCYTE SUBSETS IN THE CENTRAL NERVOUS SYSTEM IN CHRONIC PROGRESSIVE MULTIPLE SCLEROSIS

J O H N BOOSS <*, M A R G A R E T M. ESIRI 1, W A L L A C E W. T O U R T E L L O T T E 2 and DAVID Y. M A S O N ~

JDepartment of Neuropathology, Radclilfe lnfirmao', OJ(/brd (Great Britain)," 2Neurology Service, VA Medical Center, West Los Angeles, and Department of Neurology, University of California at Los Angeles, Los Angeles. CA (U.S.A.): and "~NzqfieldDepartment of Patholog),, The John Radcliffe Hospital, Headington, O.vford (Greal Britain) (Received 18 May, 1983) (Accepted 28 Jury. 1983)

SUMMARY

In an analysis of pooled data, we have found that cytotoxic-suppressor T cells outnumber the helper-inducer subset and also the population of cells bearing the pan-T cell marker in specimens of CNS from patients dying with MS (Booss et al. 1983). In the present study of individual data, we have reviewed the case histories to determine if these findings were consistent in various clinical settings. Variables examined included disease duration, tempo &evolution, immunosuppressive therapy, and other potentially immunomodulating features such as turnouts. The predominance of the cytotoxicsuppressor subset was not found to be altered by any of these variables. We also present an individual data analysis of cases dying without known CNS disease and of cases with chronic non-inflammatory CNS disease. We found that the low but consistently observed number of T cells was apparently unrelated to the age of the individual or the site of the CNS sampled. Analysis of selected perivascular infiltrates showed, in contrast to the CNS parenchyma, that the pan-T cells and each of the subsets were approximately equal in proportion. Consideration of this observation and of the cytotoxic-suppressor subset preponderance in the parenchyma is compatible with the possibility of antigenic modulation of the T cell differentiation antigens. Finally, the potential contribution of perivascular intiltrates to the CSF pleocytosis is considered.

Fhis work received financial support from the MS Society of Great Britain and Northern Ireland and from the Kroc Foundation * On academic leave from: VA Medical Center, West Haven, CT and Departments of Neurology and l.aborator5 Medicine, Yale University School of Medicine, New Haven, CT, U.S.A. 0022-510X 'S3.${)3.00 © 1983 Elsevier Science Publishers B.V.

220 Key words: Antigenic modulation - Immunohistology o f C N S - Monoclonal antibodies Multiple sclerosis - T lymphocytes

INTRODUCTION

Immune factors have long been suspected to play a part in the pathogenesis of multiple sclerosis (MS). However, the first strong evidence associating temporal changes in immune function with progression of the disease has emerged only recently with the demonstration that relapses in M S are associated with a reduction of Con A-inducible T cell suppression in the blood (Antel et al. 1978) and a fall in the number of circulating cytotoxic-suppressor T cells detectable by use of the monoclonal antibody OKT8 (Bach et al. 1980; Reinherz et al. 1980; Compston 1983). These findings are consistent with suggestions either that MS is an autoimmune disease in which effector cytotoxic mechanisms are suppressed during remission and released from this suppression during relapse, or that cytotoxic lymphocytes themselves are important in pathogenesis and may be sequestered outside the circulation, perhaps in the CNS. during relapses. Some insight into which of these or other possible mechanisms are operating may be expected to be gained from an examination of the lymphocyte subpopulations present within MS lesions (plaques). One qualitative report (Traugott et al. 1983) has described both helper-inducer (OKT4 -r ) and cytotoxic-suppressor (OKT8 + ) T cells at the margin of active chronic plaques and mainly helper-inducer cells in adjacent normal appearing white matter. In contrast, another brief report noted cytotoxic-suppressor cells in the CNS parenchyma in plaques (Nyland et al. 1982). We have studied quantitative aspects o f t lymphocytes in CNS parenchyma in a series of patients with MS and related them to histopathological criteria for disease activity (Booss et al. 1983). We found cytotoxic-suppressor T cells to outnumber helper-inducer T cells by about 3:1 in MS CNS parenchyma. Numbers of cytotoxic-suppressor and helper-inducer cells in plaques did not correlate with plaque activity, but in non plaque areas there were fewer cytotoxic-suppressor cells in areas near active than inactive plaques. A further finding was that the cytotoxicsuppressor subset o f T ceils outnumbered the cells bearing the pan-T cell marker. This was an unexpected finding since in studies of T cells outside the CNS the T cell cytotoxic-suppressor and helper-inducer subsets tog~aer approximately equal the number of pan-T cells (Janossy and Prentice 1982). We here present a quantitative case-by-case analysis of the T lymphocytes present in 6 cases of MS for which detailed clinical information is available, together with findings in two control groups; people dying without known neurological disease and people dying with non-inflammatory neurological disease. Further, qualitative observations drawn from the larger pool of 10 MS patients are presented.

221 MATERIALS AND METHODS

Cases with no known neurological disease

CNS material from 10 cases ranging in age from 49 to 86 years was studied. These cases were people who had died unexpectedly outside hospital with no known neurological disease and were obtained by courtesy of pathologists at the John Radcliffe Hospital, Oxford. Six died of myocardial infarction and/or coronary atherosclerosis and one each of carbon monoxide poisoning, respiratory failure, aspiration, and pulmonary embolus (Table 1). Case 2 had a history of metastatic chemodectoma but died of myocardial infarction with no metastases found in the brain. Case 10 had a history of deep vein thrombosis treated with anticoagulation 2 months before death from a pulmonary embolus. No cases with known immunologic disease occurred in this group. Nineteen sections from various CNS locations were studied immunohistologically from these cases (Table 1). The samples were taken at autopsy from less than 24 h up to 4 days after death. These sections were screened for microscopic lesions with standard

TABLE 1 I M M U N O H I S T O L O G I C A L STUDY OF CASES DYING SUDDENLY IN THE ABSENCE OF KNOWN CNS DISEASE

Case

I

Age Sex

86 M

2

60 M

3

56 M

4

49 M

5

79 M

6

63 M

7

75 M

8 9 10

67 M 55 F 69 M

Section ~'

Monoclonal antibodies b,~ T3

L3A

T8

Lat. angle Lat. angle Cerebrum C.M.J.

0.20 0.23 0.03 0.00 0.03 0.33 0.13 0.03 0.10 0.17 0.10 0.00 0.03 0.37 0.53 0.10 0.13 1.10 (I.83

0.00 0.17 0.00 0.00 0.03 0.07 0.13 0.23 0.00 0.03 0.00 0.07 0.23 0.00 0.03 0.20 0.37 0.63 0.33

0.50 0.20 0.07 0.33 0.00 0.10 0.10 0.23 0.00 (/.07 0.17 0.37 0.20 0.17 0.00 0.33 0.00 0.43 1.60

2 _+ SD

0.23 _+ 0.30

0.13 _+ 0.17

0.26 _+ 0.36

Medulla Lat. angle Cerebrum Lat. angle Cerebrum Lat. angle C8 T4 L3 Pons C.M.J. Cerebrum Lat. angle Eat. angle

Medulla

" Lat. angle = adjacent to lateral angle of lateral ventricle; C8, T4, L3 = cervical, thoracic, and lmnbar spinal cord; C.M.J. = cervical medullary junction. ~' T3 = pan-T, L3A = helper-inducer subset, T8 = cytotoxic-suppressor subset. 4. Positive cells per high-power field ( x 400).

222

techniques. The "lateral angle" samples from cases 2 and 3 revealed sparse perivascutar infiltrates and the "lateral angle" sample from case 8 showed reactive gliosis. Other samples appeared entirely normal.

Cases of chronic non-inflammatory CNS disease Six cases of chronic progressive neurological disease of non-inflammatory type were studied. Material from these cases was obtained from the National Neurologic Research Bank, Los Angeles, U.S.A. Clinical details on these patients were incomplete but 4 of the 6 cases ranged in age from 55 to 74 years. These patients were suffering from Huntington's disease (2), Alzheimer's disease (2), Parkinson's disease(l) and amyotrophic lateral sclerosis (1) (Table 2). In each case the diagnosis had been confirmed by pathological examination at autopsy. Nine sections were examined:

Cases of MS Sufficient clinical information was available on 6 of the 10 patients previously studied (Booss et al. 1983) to evaluate the course of the disease. These clinical summaries are provided below and the quantitative data provided in Table 3. Material from 4 of these cases was obtained from the National Neurologic Research Bank, Los Angeles, U.S.A. The other two cases were obtained from Dr L. Cuzner, London, and Dr P.S. Andrews, Kettering, U.K. The course of the disease in all 6 cases was chronic progressive at the time of death. No case was on immunosuppressive therapy at the time TABLE 2 CASES OF CHRONIC N O N - I N F L A M M A T O R Y CNS D I S E A S E

Disease and Case

Section

Monoclonal antibodies "'b T3

L3A

T8

0.17 0.23 0.03

0.23 1.16 0.04

Huntington's disease 179

Spinal cord Cerebrum

178

Cerebrum

0.43 0.83 0.00

Cerebrum Medulla

0.37 0.53

0.30 0.23

0.07 1.10

Spinal cord Cerebrum

0.10 0.27

0.13 0.10

0.23 0.13

Spinal cord Cerebrum

0.30 0.03

0.07 0.03

0.37 0.00

0.32 + 0.26

0.14 + 0.10

0,37 + 0.45

Alzheimer's disease 215 193

Parkinson's disease 525 ALS 176

+ SD

a Monoclonal antibodies as in Table 1. b Positive cells per high-power field ( x 400).

23 TABLE 3 I M M U N O H I S T O L O G I C A L S T U D Y OF PATIENTS WITH C H R O N I C P R O G R E S S I V E MS

Case

Age Sex

Duration (yr)

7376

51 F

8

449

58 M

30

412

45 M

22

511

42 M

13

372

57 M

21

7457

80 F

40 2 _+ SD x + SD

" ~' ~' "

Site

Monoclonal antibodies ~'-b T3

L3A

1"8

Plaque ~ Non-plaque Plaque Non-plaque Plaque Non-plaq ue Plaque Non-plaque Plaque Non-plaque Plaque ~ Non-plaque

0.66 0.83 0.17 (/.20 0.07 0.10 6.63 0.20 0.10 (I.30 1.47 0.41

0.99 0.82 0.50 0.30 0.20 (}.15 1,03 0.00 0.50 0.10 0.09 0.06

2.05 1.82 1.37 0.13 0.77 0.82 7.37 0.07 0.87 0.63 2.68 2.37

Plaque Non-plaque

1.52 _+ 2.56 0.34 + 0.24

0.55 + 0.39 0.24 _+ 0.30

2.52 _+ 2.49 0.97 _+ 0.93

Monoclonal antibodies as in Table 1. Positive cells per high-power field ( x 400). Pooled data from 6 sites. Pooled data from 3 sites.

of death but all except one (case 7457) had been treated with such therapy at some stage of their illness, and one case (case 412} had completed a course of intramuscular steroid therapy 2 weeks before death. Qualitative observations were drawn from the whole group of 10 patients with MS. M S Case histories Case 7376

This 51-year-old female patient died of a pulmonary embolus after an 8-year history of MS. The patient's son and husband also had MS. The first symptom of gait unsteadiness was followed a year later by right leg weakness. Examination at a year and a half after the start revealed lateral gaze nystagmus, a right upgoing plantar response, right leg weakness, ataxia and evoked potential studies supportive of a left optic neuritis. Weakness in the upper extremities and difficulty of fine finger movement was noted at 4 years. At 7 years from the onset a frame was required for ambulation. In the following years she was no longer ambulating and the left arm was significantly weak. Several trials of therapy including ACTH and azathiaprine were attempted without halting the clinical progression, including a course of ACTH 7 months prior to death. Thus this patient's course was characterised by a relentless progression throughout. Case 449

This 58-year-old male patient died of pneumonia after a 30-year history of MS. The first symptom of blurred vision resolved incompletely. Eight years later impotence developed, followed in succeeding years by urinary incontinence. Thirteen years after the start, sudden paraparesis developed. By 20 years, facial sensation and jaw movement were impaired as was dexterity of the right hand and balance. Twenty-eight years after the start he had developed spastic quadriplegia and optic neuritis approaching amaurosis. Intcrmittcnt confusion noted at that time evolved into a moderate dementia at 29 years. Finally, slurred

224 speech was found after 30 years which was not present 2 years previously. Dexamethasone had been used and discontinued 10 years prior to death. The course of MS had been complicated by seizures, drug overdose and chronic infections. However, this patient's course which was at first of the exacerbating remitting type was relentlessly progressive over the last several years. Case 412

This 45-year-old male patient died of a pulmonary embohis after a 22-year history of MS. Weakness and incoordination of the lower extremities lasting approximately one month was followed by a period of 8 years in which episodes of generalised weakness and fatigue cleared after a few days. At that time progressive and exacerbating weakness of the lower extremities evolved. Use of a walking stick was required 16 years from the onset and a wheelchair at 18 years. Upper extremity weakness, bowel incontinence, urinary urgency, impotence, sensory defects, diplopia, altered visual acuity and slurred speech developed. At 21 years after the start of the illness he enrolled in a therapy programme. In the 9 months prior to death he received independent courses of intravenous, intrathecal, and intramuscular steroid therapy. Two weeks prior to death an IM course had been completed. Thus this patient's course had started in an exacerbatingremitting fashion, but had evolved into a progressive form with intensive steroid therapy in the last 9 months of life. Case 511

This 42-year-old male died of respiratory failure after a 13-year history of MS. Initial symptoms of lightheadedness and unilateral tinnitus cleared in a few weeks. In the following 2 years staggering gait, lower extremities paraesthesiae, left leg paresis, impotence, and diplopia developed. A course of ACTH was associated with improvement of limb strength in the 4th year of the illness. However, over the following 7 years moderate weakness of the lower extremities evolved to require the use of a wheelchair and mild upper extremity weakness was present. Other dysfunction included intranuclear ophthalmoparesis and fluctuating sensory impairment in the lower extremities. During that period, courses of acute and chronic therapy with ACTH, prednisone and dexamethasone were used. In addition a course of intrathecal cytosine arabinoside and hydrocortisone was received. Thirteen years from the onset the patient was paraplegic and bedridden. The last 2 years of life were complicated by repeated infections and their sequelae. Thus, this patient's illness rapidly evolved into the progressive type. Case 372

This 57-year-old man died of pneumonia complicating an untreated large cell carcinoma of the lung after a 2 l-year-history of MS. Ten years earlier he had had a carcinoma of the colon removed, ~ follow-up radiation therapy, and had suffered no metastases or recurrence. The first neurologie dysfunction was paralysis of 6th cranial nerve function with a CSF pleocytosis following a "whip lash" injury in a road accident. In the following years he experienced episodes of loss of equilibrium followed at 5 years after the start by progressive numbness which cleared. Two months later he experienced an episode of retrobulbar neuritis which was treated with steroids. At 10 years after the start he had gait difficulty a n d numbness involving all extremities. After 17 years of the illness he had become quadriparetic, required catherisation, and had extensive sensory changes. Complete nursing care was required for the last 4 years of his life. This patient's course thus started in a remitting form but evolved after several years into a progressive form. Case 7457

This 80-year-old female patient died of respiratory failure after a 40-year course of MS, The patient developed left leg weakness at age 45 after approximately 5 years ofn0tieing her leg Was"jumpy~L Four years later she was found to have a left upgoir~ plantar response and diminished sensation On the left: At the age of 50 she was noted to be improved; however at the age of 59 both lower extremities were noted to be spastic with upgoing plantar responsesl Decreased sensation was noted below TI@ She was ambulating at 28 years after the start of the illness, but on hospitalisation 40 years after the start she was wheelchair bound with a spastic paraplegia. The admission was prompted by the sudden onset of weakness of the left upper extremity. Steroid therapy was recommended. At autopsy 10 months later she was found tohave an old haemorrhage at C6-7 on the left at the site of a MS plaque. Thus this patient appeared to have an extremely gradual course punctuated toward the end by a spontaneous haemorrhage as h ~ been previously reported in MS (Jankovic et al. 1980).

225

MeNo~ Tissue processing Tissues from the UK were processed fresh and tissues from the National Neurological Specimens Bank (Los Angeles) were received frozen. Small tissue sections were attached to chucks in dry ice with OCT mounting compound, cut with a cryostat ( - 20°C) at 10-12/~m thickness, mounted on gelatin-coated slides, fixed in acetone for 10 rain, and stored at - 20°C. Sections to be stained with oil-red-O were not fixed in acetone.

Immunoperoxidase staining for T-lymphocytes. Stored sections were brought to room temperature and all steps were performed at that temperature. The sections were washed in Tris-buffered-saline (TBS), pH 7.4, and treated for 45 rain with a monoclonal antibody in a humidified atmosphere. For total T lymphocytes, monoclonal antibody against the pan-T cell antigen OKT3 and the anti-SRBC receptor antigen OKT11A (both Ortho Diagnostics) were used. The data for the OKT3 antibody is reported here. For the cytotoxic-suppressor subset, the OKT8 antibody (Ortho Diagnostics) was used. Two antibodies against the helper-inducer subset antigen were used, OKT4 (Ortho Diagnostics) and Leu 3A (Becton-Dickinson). The Leu 3A antibody produced a stronger reaction and those data are reported here. After washing for 10 min with TBS, the sections were exposed to peroxidase-conjugated rabbit antibody against mouse immunoglobulin (DAKO) for 30 rain. The antibody was diluted 1:50 in TB S and contained a 1:20 dilution of normal human serum. Following a TB S wash, the sections were reacted with a solution of diaminobenzidine (60 mg in 100 ml TBS) and one drop of H202 for 5 rain. After a wash in water, the sections were lightly counterstained with haematoxylin blue, dehydrated, cleared, and mounted.

Assay controls. Rather than blocking endogenous peroxidase activity, which might damage the relevant membrane antigens, sections were evaluated for the presence of endogenous peroxidase bearing cells in the central nervous system. In this procedure all of the steps outlined above were followed after omission of the primary antibody step. Negligible numbers of positively staining cells were found. In addition, the technique as applied to CNS sections was checked against the known staining patterns for fresh frozen tonsil, spleen and lymph nodes.

Quantitation of lymphocytes in CNS sections Separate sections were evaluated after staining with a modified haematoxylin stain for suitability for further study, location of plaques, and sites of counting. Other sections were stained with oil-red-O to assist in the classification of plaques as active and inactive. The former demonstrated increased cellularity and the presence of oilred-O staining phagocytic cells, whereas inactive plaques did not. Plaques were determined to be active if the majority of the section met criteria for activity. All sections were scanned at low and intermediate ( x 250) power, selected perivascular infiltrates were

226 quantitated, and foci of T lymphocytes noted. Thirty high-power fields ( x 400) were counted in the same relative white matter location in each of the sections stained with the various monoclonal antibodies. The data are expressed as positive cells per highpower field. A similar counting procedure was used for sections containing MS plaques with the additional step of counting plaque and non-plaque areas separately. The sites chosen for counting plaques included the centre, intermediate zone, and the demyelinated portion of the margin. Non-plaque areas chosen for counting varied in location but avoided overlapping the plaque: However, the same locations for any given section were standardised for each of the monoclonal antibody preparations. RESULTS Examination of the spleen sections from 2 cases of MS satisfactorily demonstrated pan-T cells (OKT3 + ), cytotoxic-suppressor cells (OKT8 + ) and helper-inducer cells (Leu 3a + ) in post mortem material. In this material and in the fresh tonsil specimens we obtained staining for the various T cells in the distribution and proportions previously reported by others (Janossy and Prentice 1982). Cases with no neurological disease Small numbers of cells were detected in parenchyma by each of the monoclonal antibodies employed, i.e. pan-T cells, cytotoxic-suppressor and helper-inducer subsets in most of the cases dying without neurological disease (Table 1). Cases with minor incidental histopathologic lesions did not contain particularly high numbers of T cells and 1 case (case 10) with no histopathologic lesions on routine microscopy showed the highest number of T cells. No clear correlation of numbers of cells with age of subject. site of CN S examined nor presence of apparently incidental mild histopathologic change was found. All cell numbers were very low but the cytotoxic-suppressor cells were detected twice as frequently as helper-inducer cells and were equivalent in number to cells bearing the pan-T cell marker. Positively stained ceils were also observed in Virchow-Robin spaces. Cases with chronic non-inflammatory CNS disease In these cases also k was possible to demonstrate small numbers of pan-T. cytotoxic-suppressor and helper-inducer cells. The number of cells in these cases did not differ significantly from the control cases with no neurological disease (Table 2). Again, there were slightly more cytotoxic-suppressor than helper-inducer cells and the cytotoxic-suppressor cells slightly outnumbered cells bearing the pan-T cell marker. Cases of MS A number of cases described (Summaries in Materials and Methods) share important features. Thus cases 7376, 449, and 412 appeared to suffer relentlessly progressive illness up to death. Case 412 had received several types and courses of steroid therapy in the 9 months prior to death. Case 511 had also received several courses and various types of steroid therapy but these occurred 2 years or more before

227 death. Conversely certain features appeared to be unique. Thus case 372 had suffered 2 tumours in addition to MS, including one present at the time of death. Finally, case 7457 had an extremely gradual 40-year course of MS on which was superimposed the rare complication of haemorrhage toward the end. The findings with regard to T cells in the MS cases are summarised in Table 3. In spite of the differences of clinical characteristics, certain features of the studies of T-cell subsets (Table 3) are consistent. Thus, the OKT8 subset is found to outnumber the Leu 3A subset in the parenchyma of all plaques and in all non-plaque counts save one. The OKT8 subset also outnumbered those cells bearing the pan-T-cell marker (OKT3) in all plaques and in all non-plaque counts save two. M S : Variation of T cell content within a single case As described above, case 7376 suffered an 8-year course of MS which appeared to be steadily progressive throughout. It therefore offered a good opportunity to determine if various samples taken from the CNS would have similar content of T cells in an active case. The data presented in Table 3 represent averaged data from 6 sites. Inspection of the data for the individual lesions demonstrates that consistency does not hold. Table 4 demonstrates two divergent sections. First there is a wide variation in the plaque content of each of the three T-cell markers examined. Secondly, the T-cell content does not correlate with the histopathological activity of the demyelinating process. Nor does examination of the whole series support a correlation of the inactive histopathologic appearance and T-cell content. M S : Perivascular infiltrates For comparison with the parenchymal distribution of cells in plaque and non plaque areas in MS we examined the relative proportions of the T-cell categories in perivascular infiltrates in MS plaques and in non plaque areas. Table 5 summarises these observations. We found that in contrast to the parenchyma where cytotoxic-

TABLE 4 VARIATION OF T CELL CONTENT AT DIFFERENT LOCATIONS IN THE CNS OF AN ACTIVE CASE OF CHRONIC PROGRESSIVE MS Location

Activity

Medulla

Inactive

Diencephalon

Active

~' Monoclonal antibodies as in Table 1. t, Positive cells per high-power field ( x 400).

Tissue

Plaque Non-plaque Plaque Non-plaque

Monoclonal antibodies~,b T3

L3A

T8

2.08 2.13 0.40 0.30

1.80 0.97 0.30 0.23

4.97 3.30 0.87 0.73

228 TABLE 5 RATIOS OF T H E H E L P E R - I N D U C E R T-CELL S U B S E T A N D PAN-T CELLS TO T H E C Y T O T O X I C - S U P P R E S S O R T-CELL S U B S E T

Plaque

Non-plaque

Parenchyma H-1 : C-S ~ Pan-T : C-S"

0.22 0.60

0.25 0.35

Perivascular Infiltrate H-I:C-S Pan-T:C-S

0.94 1.03

0.88 1.07

a H-I: C-S = ratio of the helper-inducer T-ceU subset to the cytotoxic,suppressor T-celt subset. b Pan-T: C-S = ratio of pan-T cells to the cytotoxic-suppressor T-cell subset.

suppressor cells outnumber helper-inducer cells by 4-5:1, these 2 subsets of T cells are more or less equally represented within the perivascular spaces This finding suggests that cytotoxic-suppressor cells tend to spread more widely into the C N S parenehyma from perivaseular spaces than do the helper-inducer cells, Figure 1 provides an illustration of this observation. T cells may also reach the parenchyma from vessels lacking a well defined perivascular space such as the one illustrated in Fig. 2. Perivascular infiltrates are sometimes seen in close relationship to the subaraehnoid space (Fig. 3) and might be expected to contribute more than do the parenchymal cells to the population of T cells in the cerebrospinal fluid. DISCUSSION

Peripheral blood has been shown normally to contain about twice as many helper-inducer T cells as cytotoxic-suppressor T cells and the sum of these ceils in the two subsets equals the number of pan-T cells present; each subset carries the pan-T-cell marker antigen together with either the cytotoxic-suppressor or the helper,inducer marker antigen (Janossy and Prentice 1982). In active MS the number of circulating cytotoxic-suppressor T cells is reduced so that the normal excess of helper-inducer cells can become accentuated in chronic progressive disease as well as during acute relapse. In the present series of cases we have not found this preponderance of helper-inducer cells reflected in the CNS. On the contrary, we found the predominant T-cell subset in the CNS to be the cytotoxic-suppressor one. The excess of cytotoxic-suppressor cells found in the cases described here is greater than in our other study. T8 : T4 is 4-5 : 1 whereas in that study it was 3 1. This is due to the fact that the cases included in this study happen to have few histopathologically active lesions. In our other study nonplaque areas around active plaques had significantly fewer cytotoxic-suppressor cells than did non-plaque areas around inactive plaques and plaques themselves (Booss et al. 1983). Hence, exclusion from this study of cases containing many active lesions

229

~e

.J

m

|

ale Q e

m

| ,O e

p4t:

0

.

# P

i

~f

#

4

L

O Q

q J

0

0

tr 60

Fig. 1. Vein situated in base of ports within a demyelinated plaque. A and B are from adjacent sections treated respectively with anti-Leu 3a antibody, recognising helper-inducer cells and anti-OKT8 antibody, recognising cytotoxic-suppressor cells. Many Leu 3a + cells are present in the perivascular space but not within parenchyma (A), while OKT8 + cells are present in both the perivascular space and the surrounding p~renchyma (B). Counterstained with haematoxylin, x 200.

- -?= F;~

~ _

~r

.

.

"o .

Fig. 2. Small venules within p a r e n c h y m a at m a r g i n of a d e m y e l i n a t e d plaque. Section treated with anuO K T 8 antibody, recognising c y t o t o x i c - s u p p r e s s o r cells which are present adjacent to the vessel ~short a r r o w ) and in the p a r c n c h y m a (long arrow). C o u n t e r s t a i n e d with h a e m a t o x y l i n , × 200.

'

ii¸

r,

~

~i

~ .

~

~ ~

i~ I

~

~ ~ ~

~ !i

~ i~ ~

! ~

~

~

~ -

'~

~ ii ~ ~

i ~ i ~ ~ ~i~!i~i¸ iJiii:~:~ii!!~i~!i~!Si,~i 7!¸ ~~I~

~

~

~!

~ ~ =~

~i

~

~ =' i ~

Fig. 3. Section stained with anti-OKT3 antibody, recognising pan-T cells, Many positively stained cetts are present in a perivenous infiltrate in the subaraehnoid space• CNS parenchyma is present at lower left. Counterstained with haematoxylin, x 200.

231 increases overall the relative preponderance of cytotoxic-suppressor cells over helperinducer cells. The present study provides evidence that there is heterogeneity with respect to T-cell infiltrates in different plaques in a given patient, and it is not possible to regard the numbers of T cells present in any one plaque as representative of the numbers in other plaques. Heterogeneity of plaques has been commented on in other respects, for example with respect of the light chain type of IgG present in immunoglobulin-containing cells within plaques (Esiri 1980), The present study provides interesting comparative information about the T cells present in perivascular infiltrates in MS as opposed to the CNS parenchyma. While we recognise that the technique for counting cells in perivascular infiltrates in less satisfactory and more easily biased than the technique used for counting cells in the parenchyma, there seems little doubt that the proportions of cells bearing the different T-cell marker antigens differ in the perivascular spaces from the CNS parenchyma. Firstly, the preponderance of cytotoxic-suppressor cells seen in the parenchyma is greatly reduced in the perivascular spaces where approximately equivalent numbers of both cytotoxic-suppressor and helper-inducer cells are found. Secondly, the finding of more of the cytotoxic-suppressor subset than of the pan-T-cell marker-bearing cells in the parenchyma is abolished in the perivascular spaces. There are several possible explanations for the discrepancy in numbers between pan-T and cytotoxic-suppressor cells. One is that there could be CNS constituent cells which cross-react with the OKT8 antibody and produce an erroneously high figure for cytotoxic-suppressor cells. This possibility is not supported by our failure to find glial cells in normal brain staining with the antibody; cells reacting with the OKT8 antibody morphologically resemble lymphocytes and are found in the perivascular spaces as well as in parenchyma. Secondly, the stronger reaction obtained with the OKT8 antibody might be thought to result in erroneously high levels of cytotoxic-suppressor cells, particularly in post mortem tissue in which autolytic processes might be thought capable &modulating expression of some antigens. This possibility seems unlikely in view of our observations that in freshly obtained tonsil, and in post mortem spleen samples from 2 MS patients, we obtained staining results for T cells consistent with those reported by others for these tissues. Thirdly, we investigated the possibility that the difference might result from the T3 antibody itself. However, use of a second pan-T-cell marker antibody (anti OKT11A) did not alter our findings with respect to pan-T cells. A fourth possibility is that expression of the pan-T-cell marker antigen is modulated by passage within the CNS. Fifthly, factors associated with the disease itself might induce modulation. In vitro modulation of T-cell antigens by monoclonal antibodies has already been reported (Antel et al. 1982). Finally, it may be relevant that cytotoxic-suppressor cells in a different microenvironment (gut epithelium) have been described as lacking another normally expressed marker, OKT1 (Janossy and Prentice 1982). In this situation, however, the pan-T marker OKT3 is still expressed on these cells. Antigenic modulation would be consistent with our finding of less discrepancy in numbers between OKT8 + and OKT3 + cells in the perivascular spaces than within the CNS parenchyma. With regard to the excess of cytotoxic-suppressor cells over helper-inducer cells

232 in CNS parenchyma, antigenic modulation is relevant. However, it is also possible that the lymphocyte subsets migrate differentially from vessels at different sites. We frequently observed T cells adjacent to capillaries and other small vessels, lacking a well defined Virchow-Robin space, as well as in this space surrounding larger vessels. If the small vessels provide easier access for certain cells, e.g. OKT8 cells, into parenchyma, differential migration might result in that subset a p p e ~ g in greater numbers in the parenchyma than in the Virchow-Robin spaces. In conclusion this case by case analysis has confwmed the preponderance of the cytotoxic-suppressor T-cell subset over the population bearing the helper-inducer T-cell marker and that bearing the pan-T-cell marker. It has lent support to the notion of antigenic modulation, either as a function of the CNS or of MS, and it has raised the issue of differential migration into the CNS by T-cell subsets, Further studies not only of MS but also of other inflammatory diseases of the CNS will be needed before the specificity of the findings in the present study with respect to MS can be determined. ACKNOWLEDGEMENTS

Margaret Reading provided skilled technical assistance throughout. Dr Louise Cuzner of the Multiple Sclerosis Society Laboratory at the National Hospital, Queen Square, London, provided material for study. Dr J. Trevor Hughes, head of the Department of Neuropathology of the Radcliffe Inf~mary, Oxford, provided support and encouragement throughout. REFERENCES Antel, J.P., B.G.W. Arnason and M.E. Medof (1979) Suppressor cell function in multiple sclerosis - Correlation with clinical disease activity, Ann. Neurol., 5: 338-342. Antel, J., J.J.-F. Oger, S. Jaekevicius, H.H. Kuo and B.G.W. Arnason (1982] Modulation ofT-lymphocyte differentiation antigens - - Potential relevance for multiple sclerosis, Prec. Nat. Acad. Sci. ¢USA), 79: 3330-3334. Bach, M.-A., E. Tournier, F. Phan-Dinh-Tuy, L. Chatenoud. J.-F. Bach, C. Martin and J.-D. Degos (1980) Deficit of Suppressor T-cells in active multiple sclerosis, Lancet, ii: 1221-1223. Booss, J., M.M. Esiri, W.W. Tourtellotte and D.Y. Mason (1983) Ceils bearing the cytotoxic-suppressor antigen predominate in multiple sclerosis lesions, Submitted for publication. Compston, A. (1983) Lymphocyte subpopulations in patients with multiple sclerosis, J. Neurol. Neurosurg. Psychiat., 46: 105-114. Esiri, M.M. (1980) Multiple sclerosis - - A quantitative and qualitative study of immunoglobutin-containing cells in the central nervous system, Neurapath. Appl. Neurobiol., 6: 9-21. Jankovic, J., H. Derman and D. Armstrong ( 1980] Haemorrhagic complications of multiple sclerosis, J. NeuroL Neurosurg. Psychlat., 43: 76-81. Janossy, G. and H.G. Prentice (1982) T-cell subpopulations, monoclonal antibodies and their therapeutic applications, Clin. HaematoL, 11: 631-660. Nyland, H., R. Matre, S. Mork, J.-R. Bjerke and A. Naess (t982) T-Lymphocyte subpopulations in multiple sclerosis lesions (Letter), N. Engl, J. Med., 307: 1643-1644. Reinherz, E.L.. H.L. Weiner, S.L Hauser, J.A. Cohen, J.A. Distaso and S.F. Schlossman (1980) Loss of suppressor cells in active multiple sclerosis - - Analysis with monoclonal antibodies, N. Engl. J. Med., 303: 125-129. Traugott, U., E.G. Reinherz and C.S. Raine (1983)Multiple sclerosis ~ Distribution ofT-cell subsets within active chronic lesions, Science, 219: 308-310.