Brain Research, 531 (1990) 148-158 EFscvicr
148 BRES 15965
Sustained increases in cerebrospinal fluid quinolinic acid concentrations in rhesus macaques (Macaca mulatta) naturally infected with simian retrovirus type-D M.P. Heyes 1, M. Gravell 2, W.T. London 3, M. Eckhaus 4, J.H. Vickers 5, J.A. Yergey 6, M. April 7, D. B l a c k m o r e 8 a n d S.P. M a r k e y t ~Section on Analytical Biochemistry, Laboratory of Clinical Science, NIMH, Bethesda, MD 20892; eLaboratory of Central Nervous System Studies, N1NDS, Bethesda, MD 20892; 3Georgetown University, Twinbrook II, Rockville, MD, 20852; 4Comparative Pathology Section, Veterinary Resources Branch, Bethesda, MD 20892; 5Pathobiology and Primatology Laboratory, Center for Biological Evaluation and Research, FDA, Bethesda, MD 20892; 6Section of Analytical Chemistry, Laboratory of Clinical Studies, NIAAA, Bethesda, MD 20892; 7NIH Animal Center, Poolesville, MD; and SComparative Medicine Branch, NIEMS, Research Triangle Park, NC 27709 (U.S.A.) (Accepted 24 April 1990)
Key words: Excitotoxin; L-Tryptophan; Kynurenine pathway; Central nervous system; Retrovirus; Simian acquired immune deficiency syndrome; Acquired immune deficiency syndrome
Sustained increases in CSF concentrations of the excitotoxin quinolinic acid (QUIN) occur in patients with AIDS and have been implicated in the pathogenesis of the AIDS dementia complex. Macaques in captivity may also develop immunodeficiency syndromes caused by retrovirus infection, including simian retrovirus type-D. In the present study, CSF QUIN concentrations were moderately increased in retrovirus type-D-positive/antibody-negativemacaques (163.8 + 35.1 nmol/l; P < 0.0001, n = 21) but not virus-negative/antibody-positivemacaques (27.4 + 9.4 nmol/i, n = 8) compared to uninfected control macaques (23.0 + 1.6 nmol/1; n = 22). CSF QUIN concentrations in virus-positive/antibody-negative macaques tended to remain elevated over a 4-20 month period. Post-mortem studies of 9 viruspositive/antibody-negative macaques and 6 virus-negative/antibody-positivemacaques revealed inflammatory responses in the brains of 6 of 9 virus-positive/antibodynegative macaques, including lymphocytic infiltrates of the choroid plexus in 3 macaques, glial nodules in 3 macaques and perivascular infiltrates in 1 macaque. These lesions were not extensive and no evidence of brain atrophy was observed. No lesions were observed in the 6 antibody-positive/virus-negative macaques. Small increases in plasma L-kynurenine in virus-positive/antibody-negative macaques are consistent with activation of indoleamine-2,3-dioxygenase, the first enzyme in the kynurenine pathway, We conclude that sustained moderate increases in CSF QUIN occur in viremic simian retrovirus type-D macaques. The increases in CSF QUIN may reflect inflammatory responses within the brain or synthesis of QUIN precursors in systemic tissues, their entry into brain and subsequent conversion to QUIN. The neuropathologic significance of these increases in CSF QUIN remains to be determined. INTRODUCTION Infection of man with the h u m a n immunodeficiency virus ( H I V - 1 ) is associated with the acquired i m m u n e deficiency s y n d r o m e ( A I D S ) . M a n y patients with A I D S d e v e l o p a profile of neurologic deficits, including m o t o r disturbances, psychiatric abnormalities and dementia, that are collectively referred to as the A I D S d e m e n t i a complex 4°'4a'44. A t autopsy, the brain may exhibit diffuse pallor in the white m a t t e r , perivascular infiltrates of lymphocytes and m a c r o p h a g e s 4°, as well as microglial nodules in cerebral cortex and basal ganglia 4°. HIV-1 m a y e n t e r the central nervous system 27'4° and is p r e d o m inantly localized in m o n o n u c l e a t e d and polynucleated m a c r o p h a g e infiltrates rather than in neurons 32'45. The
severity of the neurologic deficits in A I D S is essentially unrelated to viral load or distribution in the brain 44 and it has been p r o p o s e d that neuronal dysfunction in A I D S may result from toxins p r o d u c e d in the brain 23. Such toxins could induce the s y m p t o m s of the A I D S d e m e n t i a complex by inducing changes in neuronal activity, interfering with synaptic function or by killing neurons. In man, the concentrations of the excitotoxin, quinolinic acid ( Q U I N ) , are increased in the cerebrospinal fluid (CSF) of patients with A I D S 21'z3. Q U I N is an agonist of N - m e t h y l - o - a s p a r t a t e ( N M D A ) type amino acid receptors 43 and when applied to the central nervous system in large quantities, Q U I N is a convulsant 34 and a neurotoxin s'47. Q U I N m a y be considered a candidate neurotoxin in A I D S and o t h e r infectious diseases 22'23.
Correspondence: M.P. Heyes, Section on Analytical Biochemistry, Laboratory of Clinical Science, Building 10, Room 3D40, National Institute of Mental Health, Bethesda, MD 20892, U.S.A., Tel. (301)496-3628, Fax (301)402-0188. Abbreviations: AIDS, acquired immune deficiency syndrome; SAIDS, simian AIDS; CSF, cerebrospinal fluid; HIV-1, human immunodeficiency virus; IDO, indoleamine-2,3-dioxygenase; L-KYN, L-kynurenine; n, number of animals studied; NMDA, N-methyl-D-aspartate; QUIN, quinolinic acid; [180]QUIN, [~80]quinolinic acid; SIV, simian immunodeficiency virus; SRV-D, simian retrovirus type-D; t~-TRP, L-tryptophan. 0006-8993/90/$03.50 (~ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
149 The mechanisms responsible for this increase in brain and CSF QUIN are unclear but may reflect activation of indoleamine-2,3-dioxygenase 19"22"52, the first enzyme of
Poolesville, MD (13 Macaca mulatta and 1 Macaca fascicularis). All of the macaques in Horsfieid units remained clinically well throughout the study and had no overt symptoms of SAIDS.
the k y n u r e n i n e p a t h w a y from r - t r y p t o p h a n (L-TRP) to QUIN. M a c a q u e s in captivity may also develop immunosuppressive s y n d r o m e s analogous to A I D S , caused by retroviral infections including the oncornavirus, simian retrovirus t y p e - D ( S R V - D ) and the lentivirus, simian immunodeficiency virus (SIV) H. S R V - D is related to the Mason Pfizer m o n k e y virus but is not closely related to either SIV or HIV-1. Simian acquired immune deficiency s y n d r o m e ( S A I D S ) caused by S R V - D is associated with reductions in leukocyte populations, neoplasms and opportunistic infections 12"13"14'29'35'36'37. S p o n t a n e o u s
Blood and CSF measures CSF and blood were collected after immobilization by an intramuscular injection of ketamine hydrochloride (10 mg/kg body weight). The skin over the left femoral artery and cisternal space was shaved and the areas thoroughly cleansed and disinfected with 10% betadine diluted in 70% ethanol. A sample of blood was collected using a sterile 23-gauge needle and 6-ml syringe and transferred to sterile heparinized vacutainers and placed on ice. A 0.5-2 ml sample of CSF was colleted from the cisternal space using a sterile 23-gauge needle and 3-ml syringe and the CSF sample was transferred to sterile glass vacutainer tubes or 1.5-ml polypropylene tubes and immediately frozen on dry ice. Antibody titres to SRV-D were determined by enzyme-linked immunosorbent assay53. SRV-D, used as antigen in the assays, was grown in rhesus monkey fibroblasts and purified and concentrated by differential centrifugation and buoyant density centrifugation in neutral sucrose gradients12. The optimum concentrations of virus antigen and enzyme-second antibody conjugate for use'in tests were determined by block titrations. The enzyme-second antibody conjugate was alkaline phosphatase-labelled rabbit anti-human IgG with reactivity to both IgG heavy and light chain epitopes (Kirkegaard and Perry Laboratories, Inc., Gaithersburg, MD). Tests were performed in 96-well, fiat bottomed polystyrene microlitre plates and were read with a MR 580 Dynatech Micro-ELISA spectrometer (Dynatech Laboratories, Inc., Alexandria, VA). Sample dilutions were considered antibody positive if readings were t>0.2 optical density units above that of the negative control.
cases of infection of m a c a q u e s with S R V - D currently occur in the p r i m a t e colony at the N I H . Not all macaques infected with S R V - D develop S A I D S ; some are viruspositive and may be asymptomatic for many months prior to the development of S A I D S . In addition, some macaques may free themselves of virus and become antibody-positive and remain asymptomatic ('recovered'). In the present study, we have quantified C S F Q U I N concentrations in rhesus macaques (Macaca mulatta), either (a) virus-positive plasma and C S F but antibodynegative in p l a s m a or (b) virus-negative in plasma and C S F with antibodies positive to S R V - D in plasma c o m p a r e d to (c) normal virus-negative and antibodynegative. In addition, C S F and plasma Q U I N concentrations were m e a s u r e d in macaques housed in Horsfield units for up to 20 months, in o r d e r to minimize opportunistic infections, and the anatomical a p p e a r a n c e of the brain was d e t e r m i n e d . Plasma r - k y n u r e n i n e (L-KYN), a possible index of I D O activity in vivo, was also m e a s u r e d .
1000-
A ..J 0 100
_z
•&
0
MATERIALS AND METHODS
0 0
Reagents QUIN, hexafluoroisopropanol, sodium octylsulphate, L-TRP, 5-hydroxyindoleacetic acid (5-HIAA) and homovanillic acid (HVA) were obtained from Sigma Chemical Co., (St. Louis, MO) and trifluoroacetylimidazole was obtained from Pierce Chemical Co. (Rockford, IL). Sodium acetate, ammonium acetate, glacial acetic acid, sodium EDTA, and acetonitrile were obtained from Fisher (Fair Lawn, NJ). Macaques The macaques studied were 22 non-infected control macaques (21 Macaca mulatta and 1 Macaca fascicularis) housed separately in standard laboratory primate cages, 21 SRV-D virus-positive macaques (20 Macaca mulatta and 1 Macaca fasciularis) and 8 virusnegative/antibody-positive rhesus macaques. Macaques were fed monkey chow, supplementary fresh fruit and vegetables. Water was provided ad libitum. Fourteen macaques (8 virus-positive and 6 antibody-positive) were housed singly in standard laboratory primate cages within Horsfield units at the NIH Animal Center in
0 10 Control
Ab+ve/ Virus-ve
Vicus+ve/ Ab-ve
Fig. 1. Increased cisternal CSF QUIN concentrations in viremic SRV-D macaques but not antibody-positive macaques plotted on a logarithmic scale. Individual macaque data and group mean + 1 S.E.M. (error bar) are presented. In macaques where more than one sample was collected, the mean value for that particular animal is shown. ©, control; @,virus-negative/antibody-positive; A, viruspositive/antibody-negative. The antibody-positive/virus-negative macaque with an average CSF QUIN concentration of 97.8 nmol/! was found to have inflammatory degenerative arthritis of both knees (M610, see also Fig. 2).
150 TABLE I
TABLE III
Average cisternal CSF concentrations of L-TRP, 5-H1AA and HVA in asymptomatic SRV-D virus-positive macaques (n = 8) and SRV-D antibody-positive macaques (n = 6) housed in Horsfield units to minimize opportunistic infections (see Fig. 2)
Post-mortem features of the 14 macaques housed for up to 20 months in Horsfield units
Control L-TRP (umol/1) 5-HIAA (umol/l) HVA (,umol/l)
2.70 + 0.31 284 + 33 1425 + 109
Viruspositive 2.79 _+0.25 299 + 64 1147 + 112
Antibodypositive 3.08 + 0.26 273 + 42 1264 + 98
The presence of infectious SRV-D in samples of plasma, peripheral blood mononuclear cells and CSF was determined by a syncitia-forming Raji cell assay 5. After inoculation with test samples, Raji cell cultures were subcultured every 3-4 days over a period of at least 21 days before final readings were made to determine the virus status of samples. None of the plasma CSF or peripheral blood mononuclear cells samples tested showed evidence of bacterial or fungal infection in culture.
Biochemical analyses CSF and plasma QUIN. CSF (30/~1 or 50/A), plasma (10/~1 or 20 /~1) and QUIN standards dissolved in water were mixed with 100 ~1 of deionized water containing 12 pmol of [lsO]QUIN as internal standard and freeze dried. QUIN and [lsO]QUIN were converted to dihexafluoroisopropanol esters 17 and determined by electron capture-negative ionization mass spectrometry using a Kratos MS-80 dual sector mass spectrometer as their molecular ions at a mass/charge ratio of 467 and 471, respectively 16'24. Minimum detectable quantity with the assay was 100 amol. Plasma L-TRP and L-KYN. Plasma (10/zl) was mixed with 500/~1 of 0.1 mol/l perchloric acid, centrifuged and a 70-/~1 aliquot collected into a 250-/~1 polypropylene tube for quantification of L-TRP by high-performance liquid chromatography (HPLC) and electrochemical detection. The HPLC system consisted of a 4.6 x 75 mm Altech ultrasphere 3-/~m oetadecylsilane column, an LC-4B amperometric detector (BAS, Lafayette, IN), an electrochemical detector cell and an LKB 2150 pump. An oxidation potential of +0.90 V vs. Ag/AgCI reference electrode was applied a glassy carbon electrode. The mobile phase consisted of 150 mmol/l sodium acetate, 250/zmol/l sodium octylsulphate, 134/~mol/l sodium EDTA, 40 ml/l acetonitrile and the pH adjusted to 3.9 with approximately 90 ml of glacial acetic acid. For the quantification of L-KYN by ultraviolet light absorbance spectrophotometry 28, plasma (100/A) was mixed with 100/~! of 2.5 mol/l perchloric acid, centrifuged and 70/A aliquots collected into 250-/~1 polypropylene tubes. Plasma L-KYN was quantified using a Perris pump, an Altech ultrasphere 5-/~m 4.6 x 150 mm column, a Kratos spectroflow 773 spectrophotometer operated at 365 nm and a mobile phase consisting of 100 mmol/l ammonium acetate and 60
TABLE II
Mean plasma concentrations of L-TRP, L-KYN (*P < 0.05) and QUIN in asympatomatic SRV-D virus-positive macaques (n = 8) and SR V-D antibody-positive macaques (n = 6) housed in Horsfield units (see Fig. 2) Control L-TRP (umol/l) L-KYN (~mol/l) QUIN (umol/I)
48 + 4 1.86 + 0.19 2334 _+325
Viruspositive 42 + 7 2.77 _+0.37* 2495 _+ 379
A ntibodypositive 50 + 8 1.98 + 0.20 2293 + 166
Macaque
Systemic tissues
Brain
32R
Pulmonary acarcasis Spleen and lymphoid hyperplasia Pulmonary acarcasis
21T 256S 249R
Pulmonary acarcasis No significant lesions Pulmonary acarcasis
142X
Lymphoid and spleen hyperplasia Lymphoid and spleen hyperplasia
Lymphocytic choroiditis Two focal granulomas in thalamus Focal granuloma and focal accumulation of mononuclear cells in cerebral cortex No significant lesions No significant lesions Lymphocytic foci in thalamus and lymphocytic choroiditis Lymphocytic choroiditis
Virus-positive 49S 71CY
121X
Antibody-positive M610 Pulmonary acarcasis, degenerative arthritis in knee joints, right eye cataract. 187N No significant lesions 176P Pulmonary acarcasis $22 No significant lesions $24 No significant lesions 123X No significant lesions
No significant lesions
No significant lesions
No significant lesions No significant lesions No significant lesions No significant lesions No significant lesions
ml/l acetonitrile at pH 3.9. A Gilson model 231 autosampler was used to inject 50/~l of sample. Sample peak heights were compared to freshly prepared standards. CSF L- TRP, 5-HIAA and HVA. Undiluted CSF was analyzed for L-TRP and 5-HIAA by the HPLC system used for plasma L-TRP, except that the acetonitrile concentration in the mobile phase was 4 ml/l. Sample peak heights of samples were compared to freshly prepared standards. A WISP 710B autosampler was used to inject 20/ll of sample. Sample peak heights were compared to freshly prepared standards.
Necropsy studies Necropsy studies were done on the 14 macaques housed in Horsfield units. An additional virus-positive/antibody-negative macaque from the main NIH campus, Bethesda (B7965) was also studied. This macaque was euthanized because of aspiration pneumonia and E. coli cultured from the lung. Macaques were immobilized by an intramuscular injection of ketamine (10 mg/kg body weight). A sample of blood and cisternal CSF was collected. Macaques were then euthanized by an overdose of sodium pentobarbital (300 mg/kg body weight). The left cerebral hemisphere of the brain was exposed. A stainless steel cannula was introduced into the aorta, clamped in place and the right atria punctured. Two liters of ice-cold 150 mmol/l saline were flushed through the brain and systemic tissues at 1 l/min using a peristaltic pump. The brain and systemic tissues were then fixed with 5 1 of 10% formaldehyde solution at a flow rate of 0.5 I/min. Brain and samples of liver, lung, spleen, kidney, testis or ovary, lymph glands and skeletal muscle were collected, a gross description made and the tissues fixed in 10% formaldehyde for at least 10 days prior to histologic examination. Fixed samples of brain were serially sectioned, each at 15/lm thickness. All sections were stained with gailocyanin. A microscopic
151 49S
A
$
71CY 32R
10oo
21T
8
256S 249R 142X
A
121X
,L
Control
100
o E vc z
I
B
0 I,i. 0
10'
I
0
,
!
i
i
5
10
15
20
Time
(Months) M610
B
187N
1O1111
[]
176P
8
$22
H
$24
m
123X Control
~ o E
11111
z
5 0
u. 0
10
5,
1 0,
1 5~
Time
210
(Months)
Fig. 2. Cisternal CSF QUIN concentrations throughout the 20-month study period plotted on a logarithmic scale in (A) SRV-D virus-positive macaques (n = 8) and (B) SRV-D antibody-positive macaques (n = 6). The concentrations of QUIN in 15 control macaques housed at the NIH animal center, PoolesviUe, MD was 22.70 _+ 1.90 nmol/I and is presented as O. The error bar represents the 95% confidence limits (+ 2 S.D.) of the control macaques.
examination was made of representative anatomical areas including cerebral cortex, hippocampus, striatum, globus pallidus, thalamus, mesencephaion at the levels of the inferior colliculus, upper and lower medulla and cerebellum. Brain tissue samples were examined and described without knowledge of the viral and antibody status of the macaque.
analyzed by analysis of variance and Dunnett's t-test. CSF QUIN concentration data were analyzed as the logarithm. Regression anaysis was done using the method of least squares. A P-value of
148 BRES 15965
Sustained increases in cerebrospinal fluid quinolinic acid concentrations in rhesus macaques (Macaca mulatta) naturally infected with simian retrovirus type-D M.P. Heyes 1, M. Gravell 2, W.T. London 3, M. Eckhaus 4, J.H. Vickers 5, J.A. Yergey 6, M. April 7, D. B l a c k m o r e 8 a n d S.P. M a r k e y t ~Section on Analytical Biochemistry, Laboratory of Clinical Science, NIMH, Bethesda, MD 20892; eLaboratory of Central Nervous System Studies, N1NDS, Bethesda, MD 20892; 3Georgetown University, Twinbrook II, Rockville, MD, 20852; 4Comparative Pathology Section, Veterinary Resources Branch, Bethesda, MD 20892; 5Pathobiology and Primatology Laboratory, Center for Biological Evaluation and Research, FDA, Bethesda, MD 20892; 6Section of Analytical Chemistry, Laboratory of Clinical Studies, NIAAA, Bethesda, MD 20892; 7NIH Animal Center, Poolesville, MD; and SComparative Medicine Branch, NIEMS, Research Triangle Park, NC 27709 (U.S.A.) (Accepted 24 April 1990)
Key words: Excitotoxin; L-Tryptophan; Kynurenine pathway; Central nervous system; Retrovirus; Simian acquired immune deficiency syndrome; Acquired immune deficiency syndrome
Sustained increases in CSF concentrations of the excitotoxin quinolinic acid (QUIN) occur in patients with AIDS and have been implicated in the pathogenesis of the AIDS dementia complex. Macaques in captivity may also develop immunodeficiency syndromes caused by retrovirus infection, including simian retrovirus type-D. In the present study, CSF QUIN concentrations were moderately increased in retrovirus type-D-positive/antibody-negativemacaques (163.8 + 35.1 nmol/l; P < 0.0001, n = 21) but not virus-negative/antibody-positivemacaques (27.4 + 9.4 nmol/i, n = 8) compared to uninfected control macaques (23.0 + 1.6 nmol/1; n = 22). CSF QUIN concentrations in virus-positive/antibody-negative macaques tended to remain elevated over a 4-20 month period. Post-mortem studies of 9 viruspositive/antibody-negative macaques and 6 virus-negative/antibody-positivemacaques revealed inflammatory responses in the brains of 6 of 9 virus-positive/antibodynegative macaques, including lymphocytic infiltrates of the choroid plexus in 3 macaques, glial nodules in 3 macaques and perivascular infiltrates in 1 macaque. These lesions were not extensive and no evidence of brain atrophy was observed. No lesions were observed in the 6 antibody-positive/virus-negative macaques. Small increases in plasma L-kynurenine in virus-positive/antibody-negative macaques are consistent with activation of indoleamine-2,3-dioxygenase, the first enzyme in the kynurenine pathway, We conclude that sustained moderate increases in CSF QUIN occur in viremic simian retrovirus type-D macaques. The increases in CSF QUIN may reflect inflammatory responses within the brain or synthesis of QUIN precursors in systemic tissues, their entry into brain and subsequent conversion to QUIN. The neuropathologic significance of these increases in CSF QUIN remains to be determined. INTRODUCTION Infection of man with the h u m a n immunodeficiency virus ( H I V - 1 ) is associated with the acquired i m m u n e deficiency s y n d r o m e ( A I D S ) . M a n y patients with A I D S d e v e l o p a profile of neurologic deficits, including m o t o r disturbances, psychiatric abnormalities and dementia, that are collectively referred to as the A I D S d e m e n t i a complex 4°'4a'44. A t autopsy, the brain may exhibit diffuse pallor in the white m a t t e r , perivascular infiltrates of lymphocytes and m a c r o p h a g e s 4°, as well as microglial nodules in cerebral cortex and basal ganglia 4°. HIV-1 m a y e n t e r the central nervous system 27'4° and is p r e d o m inantly localized in m o n o n u c l e a t e d and polynucleated m a c r o p h a g e infiltrates rather than in neurons 32'45. The
severity of the neurologic deficits in A I D S is essentially unrelated to viral load or distribution in the brain 44 and it has been p r o p o s e d that neuronal dysfunction in A I D S may result from toxins p r o d u c e d in the brain 23. Such toxins could induce the s y m p t o m s of the A I D S d e m e n t i a complex by inducing changes in neuronal activity, interfering with synaptic function or by killing neurons. In man, the concentrations of the excitotoxin, quinolinic acid ( Q U I N ) , are increased in the cerebrospinal fluid (CSF) of patients with A I D S 21'z3. Q U I N is an agonist of N - m e t h y l - o - a s p a r t a t e ( N M D A ) type amino acid receptors 43 and when applied to the central nervous system in large quantities, Q U I N is a convulsant 34 and a neurotoxin s'47. Q U I N m a y be considered a candidate neurotoxin in A I D S and o t h e r infectious diseases 22'23.
Correspondence: M.P. Heyes, Section on Analytical Biochemistry, Laboratory of Clinical Science, Building 10, Room 3D40, National Institute of Mental Health, Bethesda, MD 20892, U.S.A., Tel. (301)496-3628, Fax (301)402-0188. Abbreviations: AIDS, acquired immune deficiency syndrome; SAIDS, simian AIDS; CSF, cerebrospinal fluid; HIV-1, human immunodeficiency virus; IDO, indoleamine-2,3-dioxygenase; L-KYN, L-kynurenine; n, number of animals studied; NMDA, N-methyl-D-aspartate; QUIN, quinolinic acid; [180]QUIN, [~80]quinolinic acid; SIV, simian immunodeficiency virus; SRV-D, simian retrovirus type-D; t~-TRP, L-tryptophan. 0006-8993/90/$03.50 (~ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
149 The mechanisms responsible for this increase in brain and CSF QUIN are unclear but may reflect activation of indoleamine-2,3-dioxygenase 19"22"52, the first enzyme of
Poolesville, MD (13 Macaca mulatta and 1 Macaca fascicularis). All of the macaques in Horsfieid units remained clinically well throughout the study and had no overt symptoms of SAIDS.
the k y n u r e n i n e p a t h w a y from r - t r y p t o p h a n (L-TRP) to QUIN. M a c a q u e s in captivity may also develop immunosuppressive s y n d r o m e s analogous to A I D S , caused by retroviral infections including the oncornavirus, simian retrovirus t y p e - D ( S R V - D ) and the lentivirus, simian immunodeficiency virus (SIV) H. S R V - D is related to the Mason Pfizer m o n k e y virus but is not closely related to either SIV or HIV-1. Simian acquired immune deficiency s y n d r o m e ( S A I D S ) caused by S R V - D is associated with reductions in leukocyte populations, neoplasms and opportunistic infections 12"13"14'29'35'36'37. S p o n t a n e o u s
Blood and CSF measures CSF and blood were collected after immobilization by an intramuscular injection of ketamine hydrochloride (10 mg/kg body weight). The skin over the left femoral artery and cisternal space was shaved and the areas thoroughly cleansed and disinfected with 10% betadine diluted in 70% ethanol. A sample of blood was collected using a sterile 23-gauge needle and 6-ml syringe and transferred to sterile heparinized vacutainers and placed on ice. A 0.5-2 ml sample of CSF was colleted from the cisternal space using a sterile 23-gauge needle and 3-ml syringe and the CSF sample was transferred to sterile glass vacutainer tubes or 1.5-ml polypropylene tubes and immediately frozen on dry ice. Antibody titres to SRV-D were determined by enzyme-linked immunosorbent assay53. SRV-D, used as antigen in the assays, was grown in rhesus monkey fibroblasts and purified and concentrated by differential centrifugation and buoyant density centrifugation in neutral sucrose gradients12. The optimum concentrations of virus antigen and enzyme-second antibody conjugate for use'in tests were determined by block titrations. The enzyme-second antibody conjugate was alkaline phosphatase-labelled rabbit anti-human IgG with reactivity to both IgG heavy and light chain epitopes (Kirkegaard and Perry Laboratories, Inc., Gaithersburg, MD). Tests were performed in 96-well, fiat bottomed polystyrene microlitre plates and were read with a MR 580 Dynatech Micro-ELISA spectrometer (Dynatech Laboratories, Inc., Alexandria, VA). Sample dilutions were considered antibody positive if readings were t>0.2 optical density units above that of the negative control.
cases of infection of m a c a q u e s with S R V - D currently occur in the p r i m a t e colony at the N I H . Not all macaques infected with S R V - D develop S A I D S ; some are viruspositive and may be asymptomatic for many months prior to the development of S A I D S . In addition, some macaques may free themselves of virus and become antibody-positive and remain asymptomatic ('recovered'). In the present study, we have quantified C S F Q U I N concentrations in rhesus macaques (Macaca mulatta), either (a) virus-positive plasma and C S F but antibodynegative in p l a s m a or (b) virus-negative in plasma and C S F with antibodies positive to S R V - D in plasma c o m p a r e d to (c) normal virus-negative and antibodynegative. In addition, C S F and plasma Q U I N concentrations were m e a s u r e d in macaques housed in Horsfield units for up to 20 months, in o r d e r to minimize opportunistic infections, and the anatomical a p p e a r a n c e of the brain was d e t e r m i n e d . Plasma r - k y n u r e n i n e (L-KYN), a possible index of I D O activity in vivo, was also m e a s u r e d .
1000-
A ..J 0 100
_z
•&
0
MATERIALS AND METHODS
0 0
Reagents QUIN, hexafluoroisopropanol, sodium octylsulphate, L-TRP, 5-hydroxyindoleacetic acid (5-HIAA) and homovanillic acid (HVA) were obtained from Sigma Chemical Co., (St. Louis, MO) and trifluoroacetylimidazole was obtained from Pierce Chemical Co. (Rockford, IL). Sodium acetate, ammonium acetate, glacial acetic acid, sodium EDTA, and acetonitrile were obtained from Fisher (Fair Lawn, NJ). Macaques The macaques studied were 22 non-infected control macaques (21 Macaca mulatta and 1 Macaca fascicularis) housed separately in standard laboratory primate cages, 21 SRV-D virus-positive macaques (20 Macaca mulatta and 1 Macaca fasciularis) and 8 virusnegative/antibody-positive rhesus macaques. Macaques were fed monkey chow, supplementary fresh fruit and vegetables. Water was provided ad libitum. Fourteen macaques (8 virus-positive and 6 antibody-positive) were housed singly in standard laboratory primate cages within Horsfield units at the NIH Animal Center in
0 10 Control
Ab+ve/ Virus-ve
Vicus+ve/ Ab-ve
Fig. 1. Increased cisternal CSF QUIN concentrations in viremic SRV-D macaques but not antibody-positive macaques plotted on a logarithmic scale. Individual macaque data and group mean + 1 S.E.M. (error bar) are presented. In macaques where more than one sample was collected, the mean value for that particular animal is shown. ©, control; @,virus-negative/antibody-positive; A, viruspositive/antibody-negative. The antibody-positive/virus-negative macaque with an average CSF QUIN concentration of 97.8 nmol/! was found to have inflammatory degenerative arthritis of both knees (M610, see also Fig. 2).
150 TABLE I
TABLE III
Average cisternal CSF concentrations of L-TRP, 5-H1AA and HVA in asymptomatic SRV-D virus-positive macaques (n = 8) and SRV-D antibody-positive macaques (n = 6) housed in Horsfield units to minimize opportunistic infections (see Fig. 2)
Post-mortem features of the 14 macaques housed for up to 20 months in Horsfield units
Control L-TRP (umol/1) 5-HIAA (umol/l) HVA (,umol/l)
2.70 + 0.31 284 + 33 1425 + 109
Viruspositive 2.79 _+0.25 299 + 64 1147 + 112
Antibodypositive 3.08 + 0.26 273 + 42 1264 + 98
The presence of infectious SRV-D in samples of plasma, peripheral blood mononuclear cells and CSF was determined by a syncitia-forming Raji cell assay 5. After inoculation with test samples, Raji cell cultures were subcultured every 3-4 days over a period of at least 21 days before final readings were made to determine the virus status of samples. None of the plasma CSF or peripheral blood mononuclear cells samples tested showed evidence of bacterial or fungal infection in culture.
Biochemical analyses CSF and plasma QUIN. CSF (30/~1 or 50/A), plasma (10/~1 or 20 /~1) and QUIN standards dissolved in water were mixed with 100 ~1 of deionized water containing 12 pmol of [lsO]QUIN as internal standard and freeze dried. QUIN and [lsO]QUIN were converted to dihexafluoroisopropanol esters 17 and determined by electron capture-negative ionization mass spectrometry using a Kratos MS-80 dual sector mass spectrometer as their molecular ions at a mass/charge ratio of 467 and 471, respectively 16'24. Minimum detectable quantity with the assay was 100 amol. Plasma L-TRP and L-KYN. Plasma (10/zl) was mixed with 500/~1 of 0.1 mol/l perchloric acid, centrifuged and a 70-/~1 aliquot collected into a 250-/~1 polypropylene tube for quantification of L-TRP by high-performance liquid chromatography (HPLC) and electrochemical detection. The HPLC system consisted of a 4.6 x 75 mm Altech ultrasphere 3-/~m oetadecylsilane column, an LC-4B amperometric detector (BAS, Lafayette, IN), an electrochemical detector cell and an LKB 2150 pump. An oxidation potential of +0.90 V vs. Ag/AgCI reference electrode was applied a glassy carbon electrode. The mobile phase consisted of 150 mmol/l sodium acetate, 250/zmol/l sodium octylsulphate, 134/~mol/l sodium EDTA, 40 ml/l acetonitrile and the pH adjusted to 3.9 with approximately 90 ml of glacial acetic acid. For the quantification of L-KYN by ultraviolet light absorbance spectrophotometry 28, plasma (100/A) was mixed with 100/~! of 2.5 mol/l perchloric acid, centrifuged and 70/A aliquots collected into 250-/~1 polypropylene tubes. Plasma L-KYN was quantified using a Perris pump, an Altech ultrasphere 5-/~m 4.6 x 150 mm column, a Kratos spectroflow 773 spectrophotometer operated at 365 nm and a mobile phase consisting of 100 mmol/l ammonium acetate and 60
TABLE II
Mean plasma concentrations of L-TRP, L-KYN (*P < 0.05) and QUIN in asympatomatic SRV-D virus-positive macaques (n = 8) and SR V-D antibody-positive macaques (n = 6) housed in Horsfield units (see Fig. 2) Control L-TRP (umol/l) L-KYN (~mol/l) QUIN (umol/I)
48 + 4 1.86 + 0.19 2334 _+325
Viruspositive 42 + 7 2.77 _+0.37* 2495 _+ 379
A ntibodypositive 50 + 8 1.98 + 0.20 2293 + 166
Macaque
Systemic tissues
Brain
32R
Pulmonary acarcasis Spleen and lymphoid hyperplasia Pulmonary acarcasis
21T 256S 249R
Pulmonary acarcasis No significant lesions Pulmonary acarcasis
142X
Lymphoid and spleen hyperplasia Lymphoid and spleen hyperplasia
Lymphocytic choroiditis Two focal granulomas in thalamus Focal granuloma and focal accumulation of mononuclear cells in cerebral cortex No significant lesions No significant lesions Lymphocytic foci in thalamus and lymphocytic choroiditis Lymphocytic choroiditis
Virus-positive 49S 71CY
121X
Antibody-positive M610 Pulmonary acarcasis, degenerative arthritis in knee joints, right eye cataract. 187N No significant lesions 176P Pulmonary acarcasis $22 No significant lesions $24 No significant lesions 123X No significant lesions
No significant lesions
No significant lesions
No significant lesions No significant lesions No significant lesions No significant lesions No significant lesions
ml/l acetonitrile at pH 3.9. A Gilson model 231 autosampler was used to inject 50/~l of sample. Sample peak heights were compared to freshly prepared standards. CSF L- TRP, 5-HIAA and HVA. Undiluted CSF was analyzed for L-TRP and 5-HIAA by the HPLC system used for plasma L-TRP, except that the acetonitrile concentration in the mobile phase was 4 ml/l. Sample peak heights of samples were compared to freshly prepared standards. A WISP 710B autosampler was used to inject 20/ll of sample. Sample peak heights were compared to freshly prepared standards.
Necropsy studies Necropsy studies were done on the 14 macaques housed in Horsfield units. An additional virus-positive/antibody-negative macaque from the main NIH campus, Bethesda (B7965) was also studied. This macaque was euthanized because of aspiration pneumonia and E. coli cultured from the lung. Macaques were immobilized by an intramuscular injection of ketamine (10 mg/kg body weight). A sample of blood and cisternal CSF was collected. Macaques were then euthanized by an overdose of sodium pentobarbital (300 mg/kg body weight). The left cerebral hemisphere of the brain was exposed. A stainless steel cannula was introduced into the aorta, clamped in place and the right atria punctured. Two liters of ice-cold 150 mmol/l saline were flushed through the brain and systemic tissues at 1 l/min using a peristaltic pump. The brain and systemic tissues were then fixed with 5 1 of 10% formaldehyde solution at a flow rate of 0.5 I/min. Brain and samples of liver, lung, spleen, kidney, testis or ovary, lymph glands and skeletal muscle were collected, a gross description made and the tissues fixed in 10% formaldehyde for at least 10 days prior to histologic examination. Fixed samples of brain were serially sectioned, each at 15/lm thickness. All sections were stained with gailocyanin. A microscopic
151 49S
A
$
71CY 32R
10oo
21T
8
256S 249R 142X
A
121X
,L
Control
100
o E vc z
I
B
0 I,i. 0
10'
I
0
,
!
i
i
5
10
15
20
Time
(Months) M610
B
187N
1O1111
[]
176P
8
$22
H
$24
m
123X Control
~ o E
11111
z
5 0
u. 0
10
5,
1 0,
1 5~
Time
210
(Months)
Fig. 2. Cisternal CSF QUIN concentrations throughout the 20-month study period plotted on a logarithmic scale in (A) SRV-D virus-positive macaques (n = 8) and (B) SRV-D antibody-positive macaques (n = 6). The concentrations of QUIN in 15 control macaques housed at the NIH animal center, PoolesviUe, MD was 22.70 _+ 1.90 nmol/I and is presented as O. The error bar represents the 95% confidence limits (+ 2 S.D.) of the control macaques.
examination was made of representative anatomical areas including cerebral cortex, hippocampus, striatum, globus pallidus, thalamus, mesencephaion at the levels of the inferior colliculus, upper and lower medulla and cerebellum. Brain tissue samples were examined and described without knowledge of the viral and antibody status of the macaque.
analyzed by analysis of variance and Dunnett's t-test. CSF QUIN concentration data were analyzed as the logarithm. Regression anaysis was done using the method of least squares. A P-value of
