Appraisal of the PAM Cell Effect as a Diagnostic Test for Multiple Sclerosis Howard Cohn, MD," J6el Oger, MD, Olivier Sabouraud, MD, and Maurice Ardouin, M D
The selective reduction of PAM cell yield reported by Carp and associates in the presence of tissue from patients with multiple sclerosis (MS) has been attributed to replication in culture of a viral agent associated with MS (MSAA). We investigated the diagnostic potential of the PAM cell effect in MS and in optic neuritis (ON). Six serum and 7 CSF samples from 7 patients with MS, 3 serum and 3 CSF samples from 4 patients with idiopathic ON, and 4 sera from 4 patients with ON induced by ethambutal toxicity were tested. Blind counting showed no reduction in PAM cell yield in the MS group nor any significant difference between the two ON groups. Disturbing inconsistencies in PAM celI growth rates over time and between control flasks were demonstrated. Cohn H, Oger J, Sabouraud 0, et al: Appraisal of the PAM cell effect as a diagnostic test for multiple sclerosis. Ann Neurol 3:400-402, 1978 Current theories concerning the cause of multiple sclerosis (MS) are consistent with a double-faceted disease which requires the coming together of a ubiquitous virus and a favorable immunogenetic terrain. Carp et a1 [ 5 ] reported the presence of a transmissible, filtrable factor in tissue from MS patients that causes depression of polymorphonuclear leukocyte counts in mice. This finding has recently been duplicated [91 although other workers have failed to confirm rhe findings of Carp and his colleagues [ 2 , 1 I]. A further claim by the Carp team that yield of a transformed mouse cell line called PAM is reduced in the presence of MS tissue [4,6]has not been corroborated to date. These in vivo and in vitro phenomena have both been presumed attributable to a virus-sized agent associated with MS (MSAA), though the possibility that the agent seemingly detected in vitro is distinct from thac claimed to have been found in vivo cannot be excluded. The PAM technique involves inoculation of cell cultures with MS and non-MS tissues followed by serial subculture of equal numbers of test and control cells at weekly intervals. A persistent reduction in cell yield is expressed as the percentage of the cell yield from an untreated control PAM cell culture carried in parallel. In the initial paper 16180% depression of cell counts was seen in all of 8 cultures treated with MS brain homogenates and in none of 8 cultures treated with non-MS brain homogenates. The difference was detected at the second passage. The reduction of PAM cell yield (PAM effect) persisted for at least 18 passages, was produced by 50 nm but not 2 5 nm filtrates, From the Neurology and Ophthalmology Services, Centre Hospitalier de I'Universiti de Rennes, Rennes, France. *Present address: Department of Ophthalmology, University of Pittsburgh School of Medicine and Eye and Ear Hospital, Pittsburgh, PA 15213.
and was thought to be caused by replication of MSAA in culture. In a second report [4]the effect was positive at the third passage in 80% of 7 1 MS samples from various tissues, including serum and cerebrospinal fluid, and negative in 45 non-MS samples. No false-positive results were reported in the non-MS group, and the effect was said to be positive at all stages of the disease. MS may present as optic neuritis (ON), and MS is recognized as a major cause of ON in adults [12]. However, the relation that MS and isolated idiopathic ON bear to each other is not clear. The reported risk of MS developing in patients with ON has varied from 13 to 78% [8, 101. Evidence favoring the unity of the two conditions includes the findings in ON patients of a significantly increased frequency of CSF measles antibodies [13] and of elevated HLA A3, B7, and DW2 histocompatibility antigens characteristic of MS 1151. O n the other hand, in ON patients who failed to develop MS over several years of follow-up, no increase of A3 or B7 was found [l],and the possibility cannot be excluded that in this group of ON patients there is no link to MS. In addition, in a prospective study of 61 acute cases of ON, the CSF findings were often normal in patients who went on to develop MS 1141. There is therefore need for a test which will predict the risk of MS in ON patients. Acute ON may develop in persons treated with neurotoxic substances such as the antituberculosis drug ethambutol. It is not known whether patients with toxic ON are at risk to develop MS, or for that matter whether any insult to the optic or central nerAccepted for publication Oct 6, 1977. Address reprint requests to Dr Oger, Department of Neurology, University of Chicago Pritzker School of Medicine, 950 E 59th St, Chicago, IL 60637.
400 0364-5134/78/0003-0505$01.25 @ 1 9 7 8 by Howard Cohn
vous system pathways increases risk for MS. Druginduced changes in myelin composition might provoke a n immune response t o altered antigens, though it is generally agreed that most cases o f ethambutolinduced ON reverse when the drug is stopped [7]. In monkeys, ethambutol in extreme doses can produce extensive demyelination in the CNS [ 3 ] although n o inflammatory infiltration was mentioned in the report. If toxic ON does not predispose to the development of MS or if it d o e s so by an autoimmune type of phenomenon, a negative PAM response might be expected. Carp's results with the PAM effect prompted us to investigate its potential as a prognostic test in recent cases of ON.
Materials and Methods Three groups of patients were studied: (1) 7 MS patients fulfilling the Schumacher Committee criteria for definite MS [16]-all were examined and followed by on'e of us 0.);(2) 4 patients who developed O N while being treated with ethambutol, the minimum criterion for diagnosis of toxic O N being the appearance on successive 90-day examinations of a definite new deficiency in color vision with Farnsworth 15-hue testing; and (3) 4 patients considered to have idiopathic ON with an otherwise unremarkable neurological examination. Six serum and 7 CSF samples were obtained from the 7 MS patients. Seven serum and 3 CSF samples were obtained from the 8 O N patients, of which 4 sera were from patients with ethambutol toxicity. All sera and CSF were immediately stored at -40°C and heat inactivated at + 5 6 T for 30 minutes just prior to inoculation of cell cultures. PAM cells were kindly provided by Dr R. Carp and Dr F. Cathala (Lyon, France). The line was maintained in 75 cm2 plastic flasks (Falcon Plastics, Cockeysville, MD) in modified Eagle medium with Earle's balanced salt solution (EBSS) supplemented with nonessential amino acids and 10% heatinactivated fetal calf serum. Penicillin, 100 U per milliliter, streptomycin 100 pg per milliliter, and glutamine were added to the medium. Every two to three weeks the line was trypsinized at 37OC for 30 minutes with 0.25% trypsin in EBSS adjusted to p H 7.6 with sodium bicarbonate and split three for one. Next, 0.3 ml of each serum or CSF sample was mixed with 0.9 ml of EBSS to which was added 1 ml of EBSS containing 3 x 105 PAM cells and 0.25% trypsin. All samples were coded and handled under code from then on. After the cell mixture had stood at room temperature for 20 minutes, 2 ml was seeded into a 25 cm2 plastic flask (Falcon Plastics) with 8 ml of the cell line medium containing 17% fetal calf serum. Because each seeding, changing, and counting of all flasks had to be done at the same sitting for accuracy, two separate series of experiments were done, each with its own controls. Paired serum and CSF samples were run in the same series. Three untreated controls for each series were prepared with 0.3 ml each of heat-inactivated fetal calf serum and handled like the samples. This technique contrasts with Carp's use of only one control flask per series. All flasks were incubated at 37"C, and the medium was changed routinely at day 4.
u.
At day 7 all cultures were trypsinized with 2 ml of 0.25% trypsin in EBSS for 20 minutes at 37"C, and after the mixture was pipeted to dissociate the cells, a 1 : l O dilution of 0.2 ml of the cell suspension was made in 0.255%trypsin in EBSS. A single count for each sample was done with a Nageotte hemocytometer (PrCciss, France). Following the advice of Miss Licursi, any group of more than 4 cells was counted as 4. The possibility that cultures treated with MS tissue might give an apparently lower yield because of increased clumping was excluded by the finding of no significant difference between the number of clumps in MStreated and control cultures. Next, 2 x 105 cells from each culture were seeded into a new flask for a second count one week later, and so o n ad seriatim. Medium was changed in all flasks between days 3 to 5 if growth in any flask was limited. Comparisons were done by the Mann-Whitney nonparametric test. Mycoplasma testingwas kindly carried out by Dr F. Cathala (Lyon, France). Results I n preliminary experiments, considerable difficulty was experienced in counting t h e PAM cells because of their strong tendency to reaggregate in clumps when p u t in suspension. Following vigorous pipetting and the use of trypsin diluting solution, counts became reproducible when repeated o n single flasks, and the same degree of clumping was found in treated and untreated cultures. T h e counts in the three control flasks at each of the third, fourth, and fifth passages were averaged to give control values for each passage. The Table gives t h e yield of the individual control flasks for both series at passages 3 , 4 , and 5. N o t e the substantial variation in counts between flasks. Following Carp, the count for each treated culture is expressed as percentage of the untreated control value (Figure). I n the 13 MS samples, n o significant, persistent reduction in cell yield was observed. T h e single MS serum that averaged below 80% (Carp considers > 20% reduction positive) gave a 91% result at passage 4. W h e n both serum and CSF samples from individual patients were tested, the results were not concordant.
Yield at Passages 3 to 5 of Triplicate 25 cm2 Control Flasks Seeded with 2 x 10' Celh" Passage ExDeriments Series 1 A B C Series 2 D E F
3
4
5
4.1 3.2 2.7
9.2 5.3 6.2
5.0 7.0 6.5
4.1 6.1 4.4
5.1 4.4 5.8
8.6 3.7 4.2
aValues are x 105 cells per flask.
Cohn et al: PAM Effect in Multiple Sclerosis 401
160
8
140 a = serum
120
r 3 =
CSl
and to our knowledge there has been no published confirmation of the effect. The major conclusion of this work is that an in vitro diagnostic procedure based on relative yield of a nonstandardized, heterogeneous cell line contains too many inherent pitfalls to be of much value.
100 x
Supported by a grant from ATF’ 5 1 Inserm, France, and performed in Inserm U-49 (Prof M. Bourel) Rennes, France.
80
We thank Ms F. Latinier and G. Morel for technical assistance and Ms J. Gautier for excellent typing. We thank Barry G. W. Arnason for reviewing and correcting the manuscript.
80 40
MS
(19)
ON
References
(10)
1. Arnason BGW, Fuller TC, Lehrich JR, et d.Histocompatibil-
PAM cell yield rts percentage of control: average of passages 3,4, and 5 . No signzj5cant reduction in yield ofany sample treated culture was observed. MS and ON group results were not sign;ficantly di’jjirent.
In the ON group, no significant decrease in cell yield was observed; the 1serum which averaged below 80% gave 88% at passage 3. No significant difference existed between the results in samples from patients with idiopathic or toxic ON. At the fifth passage, MS and control cultures were infected to a comparable degree with mycoplasma, as was the original line. Discussion We cannot reproduce the PAM effect. The presence of mycoplasma infection is worrisome, but Carp et a1 [6] have reported the effect as positive in their hands even in mycoplasma-infected cultures. The PAM line originated in Carp’s laboratory as a spontaneously transformed line. The line is made up of at least four different cell types [6] and is known to contain the mouse leukemia virus. Work on scrapie and subacute sclerosing panencephalitis was being conducted in Carp’s laboratory at the time the line was developed. Possibly the conditions under which inhibited growth occurs are unique to Carp’s laboratory. Alternatively, the line may have been transformed again so that the MS-triggered inhibiting mechanism no longer operates. Only minor differences in cell yield from control flasks should be found if the growth rate in those flasks is equal and counting is accurate. Our counting is reproducible, yet yield was not constant at different passages or in individual control flasks at the same passage. Expressing the yield of treated cultures as a percentage of the control flask may mask these differences, but in view of our experience it would seem particularly risky to run only one control flask. Our negative results with PAM cultures are not unique (Cathala F: personal communication, 1977),
402
Annals of Neurology Vol 3 No 5
May 1978
ity types and measles antibodies in multiple sclerosis and optic neuritis. J Neurol Sci 22:419-428, 1974 2. Brown P, Gajdusek DC: N o mouse PMN leukocyte depression after inoculation with brain tissue from multiple sclerosis or spongiform encephalopathies. Nature 247:217-218, 1974 3. Buyske DA, Sterling W, Peets E: Pharmacological and biochemical studies on ethambutol in laboratory animals. Ann N Y Acad Sci 135:711-725, 1966 4. Carp RI, Licursi PC, Merz GS: Multiple sclerosis-induced reduction in the yield of a mouse cell line. Infect Immun 11:737-741, 1975 5. Carp RI, Licursi PC, Merz PA, et al: Decreased percentage of polymorphonuclear neutrophils in mouse peripheral blood after inoculation with material from multiple sclerosis patients. J Exp Med 136:618-629, 1972 6. Carp RI, Merz GS, Licursi P C Reduced cell yields of mouse cell line cultures after exposure to homogenates of multiple sclerosis tissues. Infect Immun 9:1011-1015, 1974 7. Deodati F, Bec P, Labro JB, et al: Nivrite optique par Pthambutol. Rev Otoneuroophthalmol46:191-198,1974 8. Hutchinson, WM: Acute optic neuritis and the prognosis for multiple sclerosis. J Neurol Neurosurg Psychiatry 39:283289, 1976 9. Koldovsky U, Koldovsky P, Henle W, e t al: Multiple sclerosis-associated agent: transmission to animals and some properties of the agent. Infect Immun 12:1355-1366, 1975 10. Kurland LT, Auth TL, Beebe GW, et al: Studies on the natural history of multiple sclerosis: I. Objectives, designs, and methods. Trans Am Neurol Assoc 88:231-235, 1963 11. Madden DL, Kreslewicz A, Gravel1 M, et al: Multiple sclerosis associated agent: failure to confirm an association (abstract). Neurology (Minneap) 27:371, 1977 12. McAlpine D:The problem of diagnosis, in McAlpine D, Lumsden CE, Acheson ED (eds): Multiple Sclerosis: A Reappraisal. Second edition. Edinburgh and London, ChurchitV Lvingstone, 1972, pp 148-159 13. Nikoskelainen E, Nikoskelainen J, Salmi AA, et al: Virus antibody levels in the cerebrospinal fluid from patients with optic neuritis. Acta Neurol Scand 51:347-364, 1975 14. Sandberg-Wollheim M: Optic neuritis: studies on the cerebrospinal fluid in relation to clinical course in 61 patients. Acta Neurol Scand 52:167-178, 1975 15. Sandberg-Wollheim M, Platz P, Ryder LP, et al: HL-A histocompatibility antigens in optic neuritis. Acta Neurol Scand 52:161-166, 1975 16. Schumacher GA, Beebe G , Kibler RF, et al: Problems of experimental trials of therapy in multiple sclerosis: report by the Panel on the Evaluation of Experimental Trials of Therapy in Multiple Sclerosis. Ann NY Acad Sci 122:552-567, 1965
The selective reduction of PAM cell yield reported by Carp and associates in the presence of tissue from patients with multiple sclerosis (MS) has been attributed to replication in culture of a viral agent associated with MS (MSAA). We investigated the diagnostic potential of the PAM cell effect in MS and in optic neuritis (ON). Six serum and 7 CSF samples from 7 patients with MS, 3 serum and 3 CSF samples from 4 patients with idiopathic ON, and 4 sera from 4 patients with ON induced by ethambutal toxicity were tested. Blind counting showed no reduction in PAM cell yield in the MS group nor any significant difference between the two ON groups. Disturbing inconsistencies in PAM celI growth rates over time and between control flasks were demonstrated. Cohn H, Oger J, Sabouraud 0, et al: Appraisal of the PAM cell effect as a diagnostic test for multiple sclerosis. Ann Neurol 3:400-402, 1978 Current theories concerning the cause of multiple sclerosis (MS) are consistent with a double-faceted disease which requires the coming together of a ubiquitous virus and a favorable immunogenetic terrain. Carp et a1 [ 5 ] reported the presence of a transmissible, filtrable factor in tissue from MS patients that causes depression of polymorphonuclear leukocyte counts in mice. This finding has recently been duplicated [91 although other workers have failed to confirm rhe findings of Carp and his colleagues [ 2 , 1 I]. A further claim by the Carp team that yield of a transformed mouse cell line called PAM is reduced in the presence of MS tissue [4,6]has not been corroborated to date. These in vivo and in vitro phenomena have both been presumed attributable to a virus-sized agent associated with MS (MSAA), though the possibility that the agent seemingly detected in vitro is distinct from thac claimed to have been found in vivo cannot be excluded. The PAM technique involves inoculation of cell cultures with MS and non-MS tissues followed by serial subculture of equal numbers of test and control cells at weekly intervals. A persistent reduction in cell yield is expressed as the percentage of the cell yield from an untreated control PAM cell culture carried in parallel. In the initial paper 16180% depression of cell counts was seen in all of 8 cultures treated with MS brain homogenates and in none of 8 cultures treated with non-MS brain homogenates. The difference was detected at the second passage. The reduction of PAM cell yield (PAM effect) persisted for at least 18 passages, was produced by 50 nm but not 2 5 nm filtrates, From the Neurology and Ophthalmology Services, Centre Hospitalier de I'Universiti de Rennes, Rennes, France. *Present address: Department of Ophthalmology, University of Pittsburgh School of Medicine and Eye and Ear Hospital, Pittsburgh, PA 15213.
and was thought to be caused by replication of MSAA in culture. In a second report [4]the effect was positive at the third passage in 80% of 7 1 MS samples from various tissues, including serum and cerebrospinal fluid, and negative in 45 non-MS samples. No false-positive results were reported in the non-MS group, and the effect was said to be positive at all stages of the disease. MS may present as optic neuritis (ON), and MS is recognized as a major cause of ON in adults [12]. However, the relation that MS and isolated idiopathic ON bear to each other is not clear. The reported risk of MS developing in patients with ON has varied from 13 to 78% [8, 101. Evidence favoring the unity of the two conditions includes the findings in ON patients of a significantly increased frequency of CSF measles antibodies [13] and of elevated HLA A3, B7, and DW2 histocompatibility antigens characteristic of MS 1151. O n the other hand, in ON patients who failed to develop MS over several years of follow-up, no increase of A3 or B7 was found [l],and the possibility cannot be excluded that in this group of ON patients there is no link to MS. In addition, in a prospective study of 61 acute cases of ON, the CSF findings were often normal in patients who went on to develop MS 1141. There is therefore need for a test which will predict the risk of MS in ON patients. Acute ON may develop in persons treated with neurotoxic substances such as the antituberculosis drug ethambutol. It is not known whether patients with toxic ON are at risk to develop MS, or for that matter whether any insult to the optic or central nerAccepted for publication Oct 6, 1977. Address reprint requests to Dr Oger, Department of Neurology, University of Chicago Pritzker School of Medicine, 950 E 59th St, Chicago, IL 60637.
400 0364-5134/78/0003-0505$01.25 @ 1 9 7 8 by Howard Cohn
vous system pathways increases risk for MS. Druginduced changes in myelin composition might provoke a n immune response t o altered antigens, though it is generally agreed that most cases o f ethambutolinduced ON reverse when the drug is stopped [7]. In monkeys, ethambutol in extreme doses can produce extensive demyelination in the CNS [ 3 ] although n o inflammatory infiltration was mentioned in the report. If toxic ON does not predispose to the development of MS or if it d o e s so by an autoimmune type of phenomenon, a negative PAM response might be expected. Carp's results with the PAM effect prompted us to investigate its potential as a prognostic test in recent cases of ON.
Materials and Methods Three groups of patients were studied: (1) 7 MS patients fulfilling the Schumacher Committee criteria for definite MS [16]-all were examined and followed by on'e of us 0.);(2) 4 patients who developed O N while being treated with ethambutol, the minimum criterion for diagnosis of toxic O N being the appearance on successive 90-day examinations of a definite new deficiency in color vision with Farnsworth 15-hue testing; and (3) 4 patients considered to have idiopathic ON with an otherwise unremarkable neurological examination. Six serum and 7 CSF samples were obtained from the 7 MS patients. Seven serum and 3 CSF samples were obtained from the 8 O N patients, of which 4 sera were from patients with ethambutol toxicity. All sera and CSF were immediately stored at -40°C and heat inactivated at + 5 6 T for 30 minutes just prior to inoculation of cell cultures. PAM cells were kindly provided by Dr R. Carp and Dr F. Cathala (Lyon, France). The line was maintained in 75 cm2 plastic flasks (Falcon Plastics, Cockeysville, MD) in modified Eagle medium with Earle's balanced salt solution (EBSS) supplemented with nonessential amino acids and 10% heatinactivated fetal calf serum. Penicillin, 100 U per milliliter, streptomycin 100 pg per milliliter, and glutamine were added to the medium. Every two to three weeks the line was trypsinized at 37OC for 30 minutes with 0.25% trypsin in EBSS adjusted to p H 7.6 with sodium bicarbonate and split three for one. Next, 0.3 ml of each serum or CSF sample was mixed with 0.9 ml of EBSS to which was added 1 ml of EBSS containing 3 x 105 PAM cells and 0.25% trypsin. All samples were coded and handled under code from then on. After the cell mixture had stood at room temperature for 20 minutes, 2 ml was seeded into a 25 cm2 plastic flask (Falcon Plastics) with 8 ml of the cell line medium containing 17% fetal calf serum. Because each seeding, changing, and counting of all flasks had to be done at the same sitting for accuracy, two separate series of experiments were done, each with its own controls. Paired serum and CSF samples were run in the same series. Three untreated controls for each series were prepared with 0.3 ml each of heat-inactivated fetal calf serum and handled like the samples. This technique contrasts with Carp's use of only one control flask per series. All flasks were incubated at 37"C, and the medium was changed routinely at day 4.
u.
At day 7 all cultures were trypsinized with 2 ml of 0.25% trypsin in EBSS for 20 minutes at 37"C, and after the mixture was pipeted to dissociate the cells, a 1 : l O dilution of 0.2 ml of the cell suspension was made in 0.255%trypsin in EBSS. A single count for each sample was done with a Nageotte hemocytometer (PrCciss, France). Following the advice of Miss Licursi, any group of more than 4 cells was counted as 4. The possibility that cultures treated with MS tissue might give an apparently lower yield because of increased clumping was excluded by the finding of no significant difference between the number of clumps in MStreated and control cultures. Next, 2 x 105 cells from each culture were seeded into a new flask for a second count one week later, and so o n ad seriatim. Medium was changed in all flasks between days 3 to 5 if growth in any flask was limited. Comparisons were done by the Mann-Whitney nonparametric test. Mycoplasma testingwas kindly carried out by Dr F. Cathala (Lyon, France). Results I n preliminary experiments, considerable difficulty was experienced in counting t h e PAM cells because of their strong tendency to reaggregate in clumps when p u t in suspension. Following vigorous pipetting and the use of trypsin diluting solution, counts became reproducible when repeated o n single flasks, and the same degree of clumping was found in treated and untreated cultures. T h e counts in the three control flasks at each of the third, fourth, and fifth passages were averaged to give control values for each passage. The Table gives t h e yield of the individual control flasks for both series at passages 3 , 4 , and 5. N o t e the substantial variation in counts between flasks. Following Carp, the count for each treated culture is expressed as percentage of the untreated control value (Figure). I n the 13 MS samples, n o significant, persistent reduction in cell yield was observed. T h e single MS serum that averaged below 80% (Carp considers > 20% reduction positive) gave a 91% result at passage 4. W h e n both serum and CSF samples from individual patients were tested, the results were not concordant.
Yield at Passages 3 to 5 of Triplicate 25 cm2 Control Flasks Seeded with 2 x 10' Celh" Passage ExDeriments Series 1 A B C Series 2 D E F
3
4
5
4.1 3.2 2.7
9.2 5.3 6.2
5.0 7.0 6.5
4.1 6.1 4.4
5.1 4.4 5.8
8.6 3.7 4.2
aValues are x 105 cells per flask.
Cohn et al: PAM Effect in Multiple Sclerosis 401
160
8
140 a = serum
120
r 3 =
CSl
and to our knowledge there has been no published confirmation of the effect. The major conclusion of this work is that an in vitro diagnostic procedure based on relative yield of a nonstandardized, heterogeneous cell line contains too many inherent pitfalls to be of much value.
100 x
Supported by a grant from ATF’ 5 1 Inserm, France, and performed in Inserm U-49 (Prof M. Bourel) Rennes, France.
80
We thank Ms F. Latinier and G. Morel for technical assistance and Ms J. Gautier for excellent typing. We thank Barry G. W. Arnason for reviewing and correcting the manuscript.
80 40
MS
(19)
ON
References
(10)
1. Arnason BGW, Fuller TC, Lehrich JR, et d.Histocompatibil-
PAM cell yield rts percentage of control: average of passages 3,4, and 5 . No signzj5cant reduction in yield ofany sample treated culture was observed. MS and ON group results were not sign;ficantly di’jjirent.
In the ON group, no significant decrease in cell yield was observed; the 1serum which averaged below 80% gave 88% at passage 3. No significant difference existed between the results in samples from patients with idiopathic or toxic ON. At the fifth passage, MS and control cultures were infected to a comparable degree with mycoplasma, as was the original line. Discussion We cannot reproduce the PAM effect. The presence of mycoplasma infection is worrisome, but Carp et a1 [6] have reported the effect as positive in their hands even in mycoplasma-infected cultures. The PAM line originated in Carp’s laboratory as a spontaneously transformed line. The line is made up of at least four different cell types [6] and is known to contain the mouse leukemia virus. Work on scrapie and subacute sclerosing panencephalitis was being conducted in Carp’s laboratory at the time the line was developed. Possibly the conditions under which inhibited growth occurs are unique to Carp’s laboratory. Alternatively, the line may have been transformed again so that the MS-triggered inhibiting mechanism no longer operates. Only minor differences in cell yield from control flasks should be found if the growth rate in those flasks is equal and counting is accurate. Our counting is reproducible, yet yield was not constant at different passages or in individual control flasks at the same passage. Expressing the yield of treated cultures as a percentage of the control flask may mask these differences, but in view of our experience it would seem particularly risky to run only one control flask. Our negative results with PAM cultures are not unique (Cathala F: personal communication, 1977),
402
Annals of Neurology Vol 3 No 5
May 1978
ity types and measles antibodies in multiple sclerosis and optic neuritis. J Neurol Sci 22:419-428, 1974 2. Brown P, Gajdusek DC: N o mouse PMN leukocyte depression after inoculation with brain tissue from multiple sclerosis or spongiform encephalopathies. Nature 247:217-218, 1974 3. Buyske DA, Sterling W, Peets E: Pharmacological and biochemical studies on ethambutol in laboratory animals. Ann N Y Acad Sci 135:711-725, 1966 4. Carp RI, Licursi PC, Merz GS: Multiple sclerosis-induced reduction in the yield of a mouse cell line. Infect Immun 11:737-741, 1975 5. Carp RI, Licursi PC, Merz PA, et al: Decreased percentage of polymorphonuclear neutrophils in mouse peripheral blood after inoculation with material from multiple sclerosis patients. J Exp Med 136:618-629, 1972 6. Carp RI, Merz GS, Licursi P C Reduced cell yields of mouse cell line cultures after exposure to homogenates of multiple sclerosis tissues. Infect Immun 9:1011-1015, 1974 7. Deodati F, Bec P, Labro JB, et al: Nivrite optique par Pthambutol. Rev Otoneuroophthalmol46:191-198,1974 8. Hutchinson, WM: Acute optic neuritis and the prognosis for multiple sclerosis. J Neurol Neurosurg Psychiatry 39:283289, 1976 9. Koldovsky U, Koldovsky P, Henle W, e t al: Multiple sclerosis-associated agent: transmission to animals and some properties of the agent. Infect Immun 12:1355-1366, 1975 10. Kurland LT, Auth TL, Beebe GW, et al: Studies on the natural history of multiple sclerosis: I. Objectives, designs, and methods. Trans Am Neurol Assoc 88:231-235, 1963 11. Madden DL, Kreslewicz A, Gravel1 M, et al: Multiple sclerosis associated agent: failure to confirm an association (abstract). Neurology (Minneap) 27:371, 1977 12. McAlpine D:The problem of diagnosis, in McAlpine D, Lumsden CE, Acheson ED (eds): Multiple Sclerosis: A Reappraisal. Second edition. Edinburgh and London, ChurchitV Lvingstone, 1972, pp 148-159 13. Nikoskelainen E, Nikoskelainen J, Salmi AA, et al: Virus antibody levels in the cerebrospinal fluid from patients with optic neuritis. Acta Neurol Scand 51:347-364, 1975 14. Sandberg-Wollheim M: Optic neuritis: studies on the cerebrospinal fluid in relation to clinical course in 61 patients. Acta Neurol Scand 52:167-178, 1975 15. Sandberg-Wollheim M, Platz P, Ryder LP, et al: HL-A histocompatibility antigens in optic neuritis. Acta Neurol Scand 52:161-166, 1975 16. Schumacher GA, Beebe G , Kibler RF, et al: Problems of experimental trials of therapy in multiple sclerosis: report by the Panel on the Evaluation of Experimental Trials of Therapy in Multiple Sclerosis. Ann NY Acad Sci 122:552-567, 1965
