Processing of neuropeptide Y and somatostatin in human cerebrospinal fluid as monitored by radioimmunoassay and mass spectrometry

Peptides, Vol. 19, No. 7, pp. 1137–1146, 1998 Copyright © 1998 Elsevier Science Inc. Printed in the USA. All rights reserved 0196-9781/98 $19.00 1 .00

PII S0196-9781(98)00071-0

Processing of Neuropeptide Y and Somatostatin in Human Cerebrospinal Fluid as Monitored by Radioimmunoassay and Mass Spectrometry CAROL NILSSON,1 ANN WESTMAN, KAJ BLENNOW AND ROLF EKMAN Institute of Clinical Neuroscience, Department of Neurochemistry, Go¨teborg University, Sahlgrenska University Hospital/Mo¨lndal, S-431 80 Mo¨lndal, Sweden Received 26 February 1998; Accepted 31 March 1998 NILSSON, C. L., A. WESTMAN, K. BLENNOW AND R. EKMAN. Processing of neuropeptide Y and somatostatin in human cerebrospinal fluid as monitored by radioimmunoassay and mass spectrometry. PEPTIDES 19(7) 1137–1146, 1998.—The processing of four neuropeptides, neuropeptide Y (NPY) 1–36, NPY (18 –36), somatostatin (SOM) 1–28, and SOM (15–28) was studied in human cerebrospinal fluid (CSF) by using a novel combination of methods that included radioimmunoassay (RIA) and mass spectrometry. Untreated CSF samples were chromatographed using reversed-phase high pressure liquid chromatography (RP-HPLC) followed by NPY-RIA or SOM-RIA. These results were compared with those obtained by incubating CSF with exogenous synthetic peptides and directly detecting peptide fragments by matrix-assisted laser desorption/ionization timeof-flight mass spectrometry (MALDI-MS). Using this combination of methods, we were able to determine the probable identities of peptides/peptide fragments recognized in radioimmunoassays. The most important NPYimmunoreactive components in CSF were found to be NPY (1–36) and NPY (3–36). Metabolic products of SOM (15–28) were found to contribute to SOM-like immunoreactivity (SOM-LI) in CSF, but SOM (1–28) only to a lesser degree. Differences in the rate of neuropeptide processing were observed. These differences depended more on the length of the peptide than its sequence. NPY (18 –36) and SOM (15–28) were rapidly and extensively processed, whereas NPV (1–36) and SOM (1–28) were processed much more slowly in CSF. The production of SOM (15–28) from SOM (1–28) by enzymes in CSF was not observed. Also, the presence of a disulfide bond in the somatostatins appeared to stabilize them against enzymatic digestion of the ring structure. The results detailed in this report confirm MALDI-MS important role in studies of neuropeptide processing in CSF. © 1998 Elsevier Science Inc. Cerebrospinal fluid

Neuropeptides

Radioimmunoassay

NEUROPEPTIDE Y (NPY) is a 36 amino acid peptide amide with a wide range of behavioral effects. Five different receptor subtypes have been characterized to date (3,10,33). The recently characterized Y5-receptor subtype is involved in food intake and can be activated by NPY (2–36) and NPY (3–36) (10). The fragment NPY (18 –36) is able to induce hypotension, and thus has the opposite effect of the intact neuropeptide on blood pressure (4). Changes in NPY processing have been postulated in major depression (9) and HIV encephalopathy (17). Given the variable effects of

Mass spectrometry

truncated peptides on neuropeptide receptors, it is possible that changes in neuropeptide processing in cerebrospinal fluid (CSF) could result in altered volume transmission in the brain (31). Somatostatin is a cyclical peptide that exists in the brain in at least two forms, SOM (1–28) and SOM (15–28). The presence of high-molecular weight (HMW) SOM in the brain, probably representing prosomatostatin, has also been established (24,26). SOM (15–28) is primarily secreted by cortical neurons, whereas SOM (1–28) is concentrated to

1 Requests for reprints should be addressed to Carol Nilsson, Institute of Clinical Neuroscience, Department of Neurochemistry, Go¨teborg University, Sahlgrenska University Hospital/Mo¨lndal, S-431 80 Mo¨lndal, Sweden. E-mail: [email protected]

1137

1138

NILSSON ET AL.

FIG. 1. MALDI mass spectra of NPY (1–36) 20 pmol/mL incubated in CSF at time points 0, 1, 2, 3, and 4 days. Intensity of the ion current is given in arbitrary units. m/z 5 mass-to-charge ratio. The numbers above the arrows indicate the amino acid residues retained from the original peptide sequence. Mass assignments are listed in Table 1.

cells in subcortical and hypothalamic levels (24,26). Five different subtypes of somatostatin receptors (SSTR) are known to exist in the human brain and pituitary (5,7,21,22,37). These receptors have different affinities for SOM (1–28) and SOM (15–28), as well as different anatomic localizations. SSTR1– 4 show a higher affinity for SOM (15–28), whereas SSTR5 has a higher affinity for SOM (1–28).

Several neuropeptides have been quantified in human CSF using radioimmunoassay (RIA), and changes in the levels of neuropeptides have been correlated to neurological and psychiatric disease states. Levels of SOM-immunoreactivity are consistently decreased in the CSF of patients with Alzheimer’s disease (6,18,19,25,36) and frontotemporal dementia (8,13,18). Reductions have also been reported in patients with multiple sclerosis (32) or depression (1,2).

PROCESSING OF NPY AND SOMATOSTATIN

1139 TABLE 1

Mass (Da) C-terminal NPY (1–36) in CSF (3–36) (8–36)* (20–36)*

Mass (Da)

Mass (Da)

Observed

Theoretical

N-terminal

Observed

Theoretical

Internal

Observed

Theoretical

4010.9 3471.1 2229.3

4010.0 3470.9 2229.2

(1–8) (1–29)

1902.6 3340.9

1902.8 3341.6

(2–15)* (3–29)* (3–30)* (4–19) (5–22)* (8–26) (10–23)* (15–31) (18–34)

1424.6 3081.6 3193.8 1712.3 1997.4 2147.5 1588.2 2126.7 2136.5

1423.5 3081.3 3192.5 1711.8 1996.8 2148.0 1588.7 2127.1 2137.2

(1–30)

3452.5

3452.6

(10–30) (12–30) (18–34)

2498.2 2254.0 2135.4

2498.8 2253.1 2137.2

NPY (1–36) in 0.9% NaCl

Very few of these studies have been verified by HPLC separation prior to RIA. When this is performed, different patterns of immunoreactivity may be revealed. Immunoreactive peptide fragments, with different retention times than the entire neuropeptide, can be observed in this manner. Some of these peptide fragments possess different receptor affinities and effects, and thus may be important modulators in disease states. The source of these peptide fragments could be from altered intracellular biosynthesis, or processing in CSF after release from the cell. A number of neuropeptide processing enzymes have been characterized in human CSF. Some enzymes that are present in relatively high concentrations are angiotensinconverting enzyme, dynorphin-converting enzyme, neutral endopeptidase EC 3.4.24.11, substance P endopeptidase (SPE), carboxypeptidase E, and aminopeptidase M (27,30,31). Despite the presence of specific peptide epithets in the names of some enzymes, a single specificity does not exist for them (31). Also, all the peptidases in human CSF have not been identified and characterized. Changes in the activity levels of SPE and DCE have been reported in patients with neuropathic pain (11,14,15,16) and in morphine-treated rats (23). However, because the knowledge about peptide-processing enzymes in CSF is incomplete, we chose to study the effects of the entire spectrum of enzymes in human CSF on selected neuropeptides. MALDI-MS is rapidly becoming an important tool in the study of neuropeptides through the method’s ability to directly detect ions at a specific mass-to-charge (m/z) ratio, even in complex biologic mixtures (plasma, urine, CSF, etc.) (20). We used MALDI-MS to determine the identity of peptide fragments produced by neuropeptide-processing, -converting, and -inactivating enzymes in normal human CSF. Neuropeptide Y and somatostatin were chosen as subjects for processing studies, because fragments of these peptides may be detected by RIA, yet have different effects

on the brain. One large form of each peptide (NPY 1–36, SOM 1–28) and one truncated form of each peptide (NPY 18 –36, SOM 15–28) were chosen for this study, in order to determine the effect of peptide sequence and size on in vitro processing in CSF. METHOD CSF The CSF sample was obtained from a patient without major neurological or psychiatric disorders. There was no sign of inflammation or damage to the blood-brain barrier. Lumbar puncture was performed in the lateral decubitus position in the L4-L5 interspace. The 12 mL CSF sample was centrifuged at 2000 3 g and 4°C for 10 min and stored at 270°C pending analysis. Two 800-mL aliquots of untreated CSF were reserved for analysis by HPLC followed by NPY-RIA or SOM-RIA. In order to study enzymatic processing of peptides in CSF, four 800 mL aliquots were spiked with NPY (1–36) (Calbiochem, Lucerne, Switzerland), NPY (18 –36) (Calbiochem), SOM (1–28) (Cambridge Research Biochemicals, Cheshire, UK), or SOM (15–28) (Cambridge Research Biochemicals) to the level of 20 mM, and incubated in a chamber at 137°C. Four 50 ml aliquots of 0.9% NaCl were spiked with these peptides to the same level and were incubated at 37°C, in order to determine if the observed fragments were specific for CSF, or if they arose due to non-specific hydrolysis. Also, a 50-mL untreated CSF sample was incubated to evaluate any production of peptides from proteins in CSF that could be misconstrued as neuropeptide fragments. On Day 0, 1, 2, 3, and 4, aliquots with a volume of 5 mL were removed and analyzed by MALDI-MS without further purification. Fractionation of complex mixtures prior to MALDI-MS analysis can reveal components that might not be detectable in the unseparated mixture, due to the phenomenon of selective desorption.

1140

NILSSON ET AL.

FIG. 2. MALDI mass spectra of NPY (18 –36) 20 pmol/mL incubated in CSF at time points 0, 1, 2, 3, and 4 days. Intensity of the ion current is given in arbitrary units. m/z 5 mass-to-charge ratio. The numbers above the arrows indicate the amino acid residues retained from the original peptide sequence. Mass assignments are listed in Table 2.

Therefore, the remainder of the spiked CSF was frozen on Day 4 to 270°C for later HPLC fractionation, prior to analysis by MALDI-MS.

m-RP-HPLC Chromatographic separation of peptides in CSF was achieved using a mRP-HPLC system (Smart, Pharmacia, Uppsala, Sweden) fitted with a C2/C18 SC 2.1/10 column

(Pharmacia). Different gradients were used for the separation of NPY-related or SOM-related peptides. For NPY, the column was equilibrated with 54% mobile phase A (0.14% TFA in water) and 46% mobile phase B (0.12% TFA in 60% acetonitrile) at a flow rate of 200 mL/min. Elution was achieved by 46% B for 10 min, followed by a linear gradient from 46 –97% during 35 min. In total, 37 300-mL fractions were collected and dried using a Speedvac (Savant, Inc.).

PROCESSING OF NPY AND SOMATOSTATIN

1141 TABLE 2

Mass (Da)

Mass (Da)

Mass (Da)

C-terminal

Observed

Theoretical

N-terminal

Observed

Theoretical

Internal

Observed

Theoretical

NPY (18–36) in CSF (19–36) (20–36) (21–36) (22–36) (23–36) (24–36) (25–36) (26–36) (27–36) (29–36) NPY (18–36) in 0.9% NaCl

2384.2 2228.8 2065.7 1902.6 1815.5 1744.5 1631.3 1475.1 1338.7 1062.9

2385.3 2229.2 2066.1 1903.1 1816.0 1745.0 1631.9 1475.8 1338.7 1062.6

(18–27)* (18–28)*

1299.2 1411.6

1299.5 1411.7

(19–33)* (20–31)* (21–31)* (22–30) (22–31) (22–33) (23–31) (23–34) (24–31) (26–34)

1064.9 1524.5 1361.5 1084.8 1198.9 1455.4 1111.9 1497.2 1040.9 1155.9

1064.6 1524.8 1361.8 1085.6 1199.4 1455.8 1111.7 1496.9 1040.6 1156.6

(19–27)

1227.8

1227.6

For SOM-related peptides, the column was equilibrated with 77% mobile phase A (0.04% TFA in 10% acetonitrile, 90% water) and 23% mobile phase B (0.04% TFA in 70% acetonitrile, 30% water) at a flow rate of 200 mL/min. Elution was achieved by 23% B for 31 min, 23–100% B over 5 min, and 100% B for 5 min. All reagents were of HPLC grade. A total of 41 200-mL fractions were collected and dried. Radioimmunoassay NPY. The levels of NPY-LI were determined using a method previously described in detail (12). In short, a rabbit antiserum raised against synthetic porcine NPY conjugated to bovine serum albumin (BSA) was used at a final dilution of 1:150,000. Porcine [125I]NPY purified by HPLC was used as a tracer. The antiserum cross-reacted with human NPY (1–36) and NPY (2–36) to 100%, and NPY (5–36) to 5%, but not with the C-terminal fragments NPY (18 –36) or NPY (20 –36). The antiserum did not cross-react with peptide YY or pancreatic polypeptide. SOM. SOM-LI was determined using a method described previously (34). The antiserum was used at a final dilution of 1: 875,000. 125I-[Tyr1]-SOM was used as a tracer. The antiserum cross-reacted with SOM (1–28) and SOM (15– 28) to 100%, and linear SOM to 50%. This antiserum is not known to cross-react with any other peptide. MALDI-MS. Samples for MALDI analysis were prepared using the seed layer method (35). A seed layer of matrix crystals was prepared on the metal probe by depositing a 0.5-mL droplet of matrix solution (a-cyano-4-hydroxy-cinnamic acid [CHCA] 1 g/L in acetonitrile). Analyte solution (CSF or dried HPLC fraction reconstituted in 0.1% TFA) and a second matrix solution (CHCA 15 g/L in 0.1% TFA in 50% water, 50% acetonitrile) were mixed and 0.5 mL of the matrix-analyte mixture was deposited on the seed layer. This procedure promotes rapid and homogeneous co-crys-

tallization of matrix and analyte. All MALDI analyses were performed using a MALDI-TOF mass spectrometer (Reflex, Bruker–Franzen Analytik GmbH, Bremen, Germany). The instrument is equipped with a nitrogen laser (337 nm), a two stage electrostatic reflectron and a delayed extraction ion source. A circular gradient neutral density filter permitted continuous attenuation of the laser beam down to 1% of the laser’s output energy. Mass spectra were analyzed using Bruker software on a Sun Sparcstation using external calibration. All spectra were acquired using the same laser intensity, in reflectron mode, at an accelerating voltage of 20 kV, and are the average of 100 laser shots. RESULTS NPY Incubation of synthetic NPY (1–36) in human CSF showed that the dominant metabolic product of this peptide is NPY (3–36). The ratio of NPY (1–36) to NPY (3–36) decreases over time, and after 3 days of incubation NPY (3–36) dominates. Several other C- and N-terminal, and internal fragments were observed (Fig. 1 and Table 1). Only one of the fragments observed in CSF was also found in the salt control, NPY (18 –34). No peptide fragments were detected in untreated CSF during the four days of incubation at 37°C. In contrast to NPY (1–36), NPY (18 –36) appears to be rapidly (within 1 day) processed in CSF, yielding an extensive ladder sequence of C-terminal fragments (Fig. 2 and Table 2). Also, some N-terminal and several internal fragments were observed. Only one fragment was observed in the salt control, NPY (19 –27); this fragment was not detected in CSF. The probable identity of several fragments contributing to the NPY-LI in CSF was established by using mRP-HPLC and RIA (Fig. 5). The detection of NPY-LI in this assay is achieved by using an N-terminally directed antiserum. The elution volumes of potentially immunoreactive fragments

1142

NILSSON ET AL.

FIG. 3. MALDI mass spectra of SOM (1–28) 20 pmol/mL incubated in CSF at time points 0, 1, 2, 3, and 4 days. Intensity of the ion current is given in arbitrary units. m/z 5 mass-to-charge ratio. The numbers above the arrows indicate the amino acid residues retained from the original peptide sequence. Mass assignments are listed in Table 3.

only, as detected using MALDI-MS analysis of the HPLC fractions from CSF incubated with synthetic NPY, are depicted in this diagram. The major component detected in CSF co-elutes with NPY (1–36). Smaller fragments may contribute to the NPY-LI eluting in earlier fractions. SOM SOM (15–28) is rapidly (within 1 day) processed to the fragments SOM (16 –28) and SOM (17–28), indicating that

this peptide could be hydrolyzed by aminopeptidases in CSF (Fig. 4 and Table 4). No further processing of the ring structure was observed. Several peaks are evident surrounding the component identified as SOM (15–28). The four peaks with a higher m/z than this peptide are easily identified as SOM (15–28) with one Na1 (122 Da), one K1 (138 Da), two Na1 (144 Da), and two K1 (176 Da) adducts. Similar salt adducts may be observed in Figs. 1–3 as well, but are especially obvious in Fig. 4.

PROCESSING OF NPY AND SOMATOSTATIN

1143

FIG. 4. MALDI mass spectra of SOM (15–28) 20 pmol/mL incubated in CSF at time points 0, 1, 2, 3, and 4 days. Intensity of the ion current is given in arbitrary units. m/z 5 mass-to-charge ratio. The numbers above the arrows indicate the amino acid residues retained from the original peptide sequence. Mass assignments are listed in Table 4.

The N terminus of SOM (1–28) was highly processed, resulting in several C-terminal fragments, as well as one N-terminal fragment, SOM (1–16) (Fig. 3, Table 3). Only two internal fragments were detected, and these could only be found after HPLC separation of CSF incubated with the synthetic peptide. This indicates that the fragments existed in relatively low abundance. As in the case of SOM (15– 28), the ring structure was spared by enzymes in CSF. Two major peaks are evident in the HPLC-RIA profile of

untreated CSF (Fig. 6). The largest and latest eluting peak may contain high molecular weight forms of SOM, as reported earlier (24). The existence of higher molecular weight forms of peptides was not addressed in this study. The contents of the fractions in the middle peak co-eluted with SOM (16 –28) and SOM (17–28). Neither SOM (15– 28) nor SOM (1–28) were detected in fractions with large amounts of SOM-LI, although these peptides may contribute to a minor amount of SOM-LI detected in the CSF in

1144

NILSSON ET AL. TABLE 3 Mass (Da) C-terminal

SOM (1–28) in CSF (3–28) (4–28) (6–28) (8–28)* (9–28)* (10–28)* (11–28) (12–28) (16–28)* SOM (1–28) in 0.9% NaCl (9–28)

Mass (Da)

Mass (Da)

Observed

Theoretical

N-terminal

Observed

Theoretical

Internal

Observed

Theoretical

2992.2 2878.1 2675.3 2507.2 2376.3 2304.0 2207.1 2051.7 1565.4

2992.9 2879.0 2676.1 2507.9 2375.3 2305.7 2208.6 2052.4 1566.1

(1–16)

1656.7

1657.8

(11–26)* (16–25)*

2019.1 1276.7

2019.3 1277.5

2376.0

2375.7

(1–24)

2711.8

2712.1

this subject, along with SOM (3–28), SOM (4 –28), and SOM (12–28). DISCUSSION We examined enzymatic processing of four neuropeptides in human CSF using MALDI-MS analysis and HPLC-RIA. The manner in which each peptide was processed was different, although certain tendencies toward patterns of peptide processing in human CSF may be observed. The related pairs of peptides, NPY (1–36)/NPY (18 –36) and SOM (1–28)/SOM (15–28) appear to be substrates for dif-

ferent enzymes in CSF. In both cases, the smaller peptides were processed more extensively and more quickly than the larger. Peptide length or conformation appears to be more important than peptide sequence; however, further study is required in order to confirm this observation. NPY (18 –36), a smaller analog of NPY (1–36) appears to be an excellent substrate for both aminopeptidases and carboxypeptidases in CSF. Several fragments of NPY (1–36) are known to have affinity for members of the NPY-receptor family. The major product of NPY (1–36) in CSF is NPY (3–36), which is known to be an agonist for the Y5-receptor subtype (10). The metabolism of this neuropeptide in CSF might affect the modulation of appetite. One further study suggested by our results is to determine whether the processing of NPY is perturbed in individuals suffering from eating disorders, with special attention given to NPY (3–36). Several C-terminal and one N-terminal fragment are evidently produced from SOM (1–28) by enzymes present in CSF. It should be noted that SOM (15–28) was not identified as one of the metabolic products of SOM (1–28). After 4 days of incubation in CSF, much of the original peptide could still be detected. The presence of disulfide bonds in proteins have been reported to stabilize them against enzymatic degradation (29). The paucity of internal fragments produced from SOM (1–28) can be explained by stabilization by the disulfide bond between Cys (17) and Cys (28). Further studies of other cyclical peptides, such as TABLE 4 Mass (Da)

FIG. 5. HPLC-RIA characterization of NPY-LI in untreated CSF. The gradient is indicated by the dotted line. The elution volumes for NPY (1–36) and NPY fragments as detected by MALDI-MS analysis of HPLC fractions from incubated CSF are marked with an arrow. a) NPY (1–18), NPY (4 –19); b) NPY (3–30); c) NPY (2–15); d) NPY (5–22); e) NPY (3–36); f) NPY (1–36).

C-terminal

Observed

Theoretical

SOM (15–28) in CSF (16–28) (17–28)

1565.6 1508.6

1565.7 1508.7

PROCESSING OF NPY AND SOMATOSTATIN

1145

FIG. 6. HPLC-RIA characterization of SOM-LI in untreated CSF. The gradient is indicated by the dotted line. The elution volumes for SOM (1–28), SOM (15–28) and SOM fragments as detected by MALDI-MS analysis of HPLC fractions from incubated CSF are marked with an arrow. a) SOM (1–28), SOM (3–28), SOM (4 –28); b) SOM (12–28); c) SOM (17–28); d) SOM (16 –28); e) SOM (16 –25); f) SOM (15–28).

arginine vasopressin, could also be conducted using MALDI-MS analysis, in order to examine whether this is a general rule for neuropeptides as well as for proteins. We were able to tentatively identify peptide fragments that may affect RIA measurements in CSF. It appears that the major NPY-immunoreactive component detected in CSF is NPY (1–36). However, some C-terminal fragments of NPY may contribute. In contrast to NPY, the major

components contributing to SOM-immunoreactivity in CSF appear to be HMW forms of SOM, SOM (16 –28), and SOM (17–28), the latter two being metabolites of SOM (15–28). This is reasonable in light of our finding that SOM (15–28) is rapidly processed in CSF by peptidases to SOM (16 –28) and SOM (17–28), which are more stable. SOM (1–28) appeared not to contribute significantly to the SOM-LI detected in CSF. SOM (15–28) was not detected as a metabolic product of SOM (1–28) in CSF. However, we found that SOM (15–28) has a short lifetime in CSF. A small amount of one of its metabolites, SOM (16 –28), was detected by MALDI-MS in one HPLC fraction from CSF spiked with SOM (1–28). This implies that SOM (15–28) may be a minor product of SOM (1–28) processing. However, these two peptides are most likely produced intracellularly and secreted separately into CSF, where they undergo different metabolic transformations. Indeed, tissue-specific proteolytic processing of the somatostatin precursor into SOM (1–28) or SOM (15– 28) has been reported (28). Using MALDI-MS, we were able to study the processing of individual neuropeptides in human CSF by the entire spectrum of peptidases present in this fluid. By combining MALDI-MS with HPLC and RIA, we were also able to determine the probable identities of the peptides that contribute to NPY-LI and SOM-LI measured in CSF. The use of mass spectrometry together with more traditional analytical techniques represents a useful combination in the investigation of neuropeptides in biological fluids. ACKNOWLEDGEMENTS The support of the Swedish Medical Research Council (07517, 12103) and Stiftelsen Lars Hiertas Minne is gratefully acknowledged.

REFERENCES 1. Ågren, H; Lundqvist, G. Low levels of somatostatin in human CSF mark depressive episodes. Psychoneuroendocrinology 9:233–248; 1984. 2. Banki, C. M.; Karmacsi, L.; Bissette, G.; Nemeroff, C. B. CSF corticotropin releasing hormone, somatostatin, and thyrotropin releasing hormone in schizophrenia. Psychiatry Res. 43:13–21; 1992. 3. Bard, J. A.; Walker, M. W.; Branchek, T. A.; Weinshank, R. L. Cloning and functional expression of a human Y4 subtype receptor for pancreatic polypeptide, neuropeptide Y, and peptide YY. J. Biol. Chem. 45:26762–26765; 1995. 4. Boublik, J.; Scott, N.; Taulane, J.; Goodman, M.; Brown, M.; Rivier, J. Neuropeptide Y and neuropeptide Y18 –36. Int. J. Peptide Protein Res. 33:11–15; 1989. 5. Corness, J. D.; Demchyshyn, L. L.; Seeman, P.; et al. A human somatostatin receptor (SSTR3), located on chromosome 22, displays preferential affinity for somatostatin-14 like peptides. FEBS Lett. 321:279 –284; 1993. 6. Davis, K. L.; Davidson, M.; Yang, R.-K.; et al. CSF somatostatin in Alzheimer’s disease, depressed patients, and control subjects. Biol. Psychiatry 24:710 –712; 1988.

7. Demchyshyn, L. L.; Srikant, C. B.; Sunahara, R. K.; et al. Cloning and expression of a human somatostatin-14-selective receptor variant (somatostatin receptor 4) located on chromosome 20. Mol. Pharmacol. 43:894 –901; 1993. 8. Edvinsson, L.; Minthon, L.; Ekman, R.; Gustafson, L. Neuropeptides in cerebrospinal fluid of patients with Alzheimer’s disease and dementia with frontotemporal lobe degeneration. Dementia 4:167–171; 1993. 9. Ekman, R.; Juhasz, P.; Heilig, M.; Ågren, H.; Costello, C. E. Novel neuropeptide Y processing in human cerebrospinal fluid from depressed patients. Peptides 17:1107–1111; 1996. 10. Gerald, C.; Walker, M. W.; Criscione, L.; et al. A receptor subtype involved in neuropeptide-Y-induced food intake. Nature 382:168 –171; 1996. 11. Harajiri, S.; Wood, G.; Desiderio, D. M. Analysis of proenkephalin A, proopiomelanocortin and protachykinin neuropeptides in human lumbar cerebrospinal fluid by reversedphase high-performance liquid chromatography (HPLC), radioimmunoassay and enzymolysis. J. Chrom. Biomed. Appl. 575:213–222; 1992. 12. Heilig, M.; Ekman, R. Chronic parenteral antidepressant treat-

1146

13. 14. 15.

16. 17.

18.

19.

20. 21. 22.

23.

24.

25.

ment in rats: unaltered levels and processing of neuropeptide Y (NPY) and corticotropin-releasing hormone (CRH). Neurochem. Int. 26:351–355; 1995. Heilig, M.; Sjo¨gren, M.; Blennow, K.; Ekman, R.; Wallin, A. Cerebrospinal fluid neuropeptides in Alzheimer’s disease and vascular dementia. Biol. Psychiatry 38:210 –216; 1995. Hyyppa¨, M. T.; Alaranta, H.; Lahtela, K.; et al. Neuropeptide converting enzyme activities in CSF of low back pain patients. Pain 43:163–168; 1990. Li, Y.-M.; Brostedt, P.; Hjerte´n, S.; Nyberg, F.; Silberring, J. Capillary liquid chromatography-fast atom bombardment mass spectrometry using a high-resolving cation exchanger, based on a continuous chromatographic matrix: application to studies on neuropeptide peptidases. J. Chromatogr. B 664: 426 – 430; 1995. Lindh, C.; Tho¨rnwall, M.; Hansen, A.-C.; Post, C.; Gordh, T.; Ordeberg; Nyberg, N. Neuropeptide-converting enzymes in cerbrospinal fluid. Acta Orthop. Scand. 67:189 –192; 1996. Malessa, R.; Heimbach, M.; Brockmeyer, N. H.; Hengge, U.; Rascher, W.; Michel, M. C. Increased neuropeptide Y-like immunoreactivity in cerebrospinal fluid and plasma of human immunodeficiency virus-infected patients: relation to HIV encephalopathy. J. Neurol. Sci. 136:154 –158; 1996. Minthon, L.; Edvinsson, L.; Ekman, R.; Gustafson, L. Reduced lumbar cerebrospinal fluid somatostatin levels in Alzheimer’s disease and dementia with frontotemporal degeneration. Dementia 2:273–277; 1991. Molchan, S. E.; Hill, J. L.; Martinez, R. A.; et al. CSF somatostatin in Alzheimer’s disease and major depression: relationship to hypothalamic-pituitary-adrenal axis and clinical measures. Psychoneuroendocrinology 18:509 –519; 1993. Nilsson, C. L.; Karlsson, G.; Bergquist, J.; Westman, A.; Ekman, R. Mass spectrometry of peptides in neuroscience Peptides 19: 781–789; 1998. O’Carroll, A. -M.; Raynor, K.; Lolait, S. J.; Reisine, T. Characterization of cloned human somatostatin receptor SSTR5. Mol. Pharmacol. 46:291–298; 1994. Patel, Y. C.; Greenwood, M.; Kent, G.; Panetta, R.; Srikant, C. B. Multiple gene transcripts of the somatostatin receptor SSTR2: tissue selective distribution and cAMP regulation. Biochem. Biophys. Res. Commun. 192:288 –294; 1993. Persson, S.; Post, C.; Alari, L.; Nyberg, F.; Terenius, L. Increased neuropeptide-converting enzyme activities in cerebrospinal fluid of opiate-tolerant rats. Neurosci. Lett. 107: 318 –322; 1989. Pierotti, A. R.; Harmar, A. J. Multiple forms of somatostatinlike immunoreactivity in the hypothalamus and amygdala of the rat: selective localization of somatostatin-28 in the median eminence. J. Endocr. 105:383–389; 1985. Raskind, M. A.; Peskind, E. R.; Lampe, T. H.; Risse, S. C.;

NILSSON ET AL.

26.

27.

28. 29.

30. 31.

32.

33.

34. 35.

36. 37.

Taborsky, G. J.; Dorsa, D. Cerebrospinal fluid vasopressin, oxytocin, somatostatin, and b-endorphin in Alzheimer’s disease. Arch. Gen. Psychiatry 43:382–388; 1986. Rissler, K.; Jost, S.; Mohadjer, M.; Mundinger, F.; Cramer, H. Molecular size distribution of somatostatin-like immunoreactivity in the cerebroventricular fluid of neurosurgical patients. Neurosci. Res. 4:343–356; 1987. Schwartz, J.-C.; Bouthenet, M.-L.; Giros, B.; Gros, C.; Llorens– Cortes, C.; Pollard, H. Neuropeptidases and neuropeptide inactivation in the brain. In: Fuxe, K.; Agnati, F., Eds. Volume transmission in the brain: Novel mechanisms for neural transmission. New York: Raven Press; 1991:381–394. Shen, L. -P.; Pictet, R. L.; Rutter, W. J. Human somatostatin I: sequence of the cDNA. Proc. Natl. Acad. Sci. USA. 79: 4575– 4579; 1982. Smith, D. L.; Sun, Y. Detection and location of disulfide bonds in proteins by mass spectrometry. In: Desiderio, D. M., Ed. Mass spectrometry of peptides. Boca Raton: CRC Press; 1991:275:287. Terenius, L.; Nyberg, F. Neuropeptide-processing, -converting, and -inactivating enzymes in human cerebrospinal fluid. Int. Rev. Neurobiol. 30:101–121; 1988. Terenius, L.; Nyberg, F. Peptidases and proteases in cerebrospinal fluid, significance in volume transmission. In: Fuxe, K.; Agnati, F., Eds. Volume transmission in the brain: Novel mechanisms for neural transmission. New York: Raven Press; 1991:415– 424. Vecsei, L.; Csala, B.; Widerlo¨v, E.; Ekman, R.; Czopf, J.; Palffy, G. Lumbar cerebrospinal fluid concentrations of somatostatin and neuropeptide Y in multiple sclerosis. Brain Res. Bull. 25:411– 413; 1990. Wahlestedt, C.; Regunathan, S.; Reis, D. J. Identification of cultured cells selectively expressing Y1-, Y2-, or Y3-type receptors for neuropeptide Y/peptide YY. Life Sci 50:7–12; 1991. Wallengren, J.; Ekman, R.; Sundler, F. Occurrence and distribution of neuropeptides in the human skin. Acta Dermatol. Venereol. 67:551–558; 1987. Westman, A.; Karlsson, G; Ekman, R. A comparison of MALDI-MS sample preparation strategies for proteins in complex biological mixtures. Adv. Mass Spectrom., in press, 1998. Wood, P. L.; Etienne, P.; Lal, S.; Gauthier, S.; Cajal, S.; Nair, N. P. V. Reduced lumbar CSF somatostatin levels in Alzheimer’s disease. Life Sci. 31:2037–2079; 1982. Yamada, Y.; Post, S. R.; Wang, K.; Tager, H. S.; Bell, G. I.; Seino, S. Cloning and functional characterization of a family of human and mouse somatostatin receptors expressed in brain, gastrointestinal tract, and kidney. Proc. Natl. Acad. Sci. USA. 89:251–255; 1992.