Carboxypeptidase E activity is deficient in mice with the fat mutation. Effect on peptide processing.

Carboxypeptidase E (CPE) is involved in the biosynthesis of many peptide hormones and neurotransmitters. Mice with the fat mutation have previously been found to have a point mutation in the cpe gene, and to have greatly reduced levels of CPE-like enzyme activity in the pituitary and pancreatic islets (Naggert, J. K., Fricker, L. D., Varlamov, O., Nishina, P. M., Rouille, Y., Steiner, D. F., Carroll, R. J., Paigen, B. J., and Leiter, E. H. (1995) Nat. Genet. 10, 135-142). In the present report, we examined CPE-like activity and peptide processing in several tissues of C57BLKS/LtJ-Cpefat/Cpefat mutant (Cpefat/Cpefat) mice. Whereas CPE-like activity is detected in homogenates of Cpefat/Cpefat mouse tissues, the majority of this activity is not due to CPE based on the sensitivity to p-chloromercuriphenyl sulfonate. In addition, the Cpefat/Cpefat activity does not bind to a substrate affinity column under conditions that bind CPE. Furthermore, the enzyme activity and immunoreactive properties of the activity purified from Cpefat/Cpefat brain are distinct from those of CPE. Taken together, these data suggest that CPE is completely inactive in the Cpefat/Cpefat mice, and that all of the CPE-like activity is due to other carboxypeptidases such as carboxypeptidase D. Levels of Leu-enkephalin in Cpefat/Cpefat mouse brain are approximately 5-fold lower than those in control brain. Treatment of the Cpefat/Cpefat brain extract with carboxypeptidase B restores the level of Leu-enkephalin to the level in control brain. Interestingly, the large molecular weight enkephalin-containing peptides are elevated 2-3-fold in Cpefat/Cpefat mouse brain. These data indicate that CPE plays an important role in the processing of peptide hormones in various tissues, but that other carboxypeptidases also contribute to peptide processing. Furthermore, the increase in levels of high molecular weight enkephalin peptides in the Cpefat/Cpefat mouse suggests that CPE is required for efficient peptide processing by the endopeptidases.

Peptide hormones and neurotransmitters are usually produced as larger pro-peptides, requiring a series of enzymes to generate the bioactive peptide (1)(2)(3). Most of these cleavages occur at specific basic residue sites, and enzymes that initially cleave the precursor have been identified (4 -8). Following this initial cleavage, a carboxypeptidase is then usually required to remove the C-terminal basic residues from the peptide to produce the bioactive moiety (9,10). For many years, a single carboxypeptidase was thought to be involved with the processing of most secreted peptides (9,10). This enzyme is alternatively known as carboxypeptidase E (CPE), 1 carboxypeptidase H, and enkephalin convertase, and has been designated EC 3.4.17.10 (11). CPE was initially discovered associated with the production of enkephalin in the adrenal medulla (12), and has been found in all neuroendocrine tissues (9,10,13,14). However, the view that CPE is the only intracellular peptide-processing carboxypeptidase has been challenged by the finding that mice with the fat mutation are still capable of producing insulin, albeit at lower levels (15).
The fat mutation has been mapped to the CPE locus on chromosome 8, and a point mutation has been found in the coding region (15). This point mutation converts Ser 202 into a Pro residue. When this mutation is created in the homologous rat CPE and the protein expressed in Sf9 cells using the baculovirus system, the enzyme is inactive and is not secreted into the medium (15). Comparison of the enzyme activity measured between mutant and wild type CPE in the baculovirus system showed that the mutation resulted in less than 0.1% of the activity of the wild type CPE. However, the levels of CPE-like activity in Cpe fat /Cpe fat mouse pituitary and pancreatic islets were found to be 5-10% of the levels in tissues from control mice (15). Furthermore, the C-terminal processing of insulin is not completely eliminated in the Cpe fat /Cpe fat mice (15), suggesting that an active carboxypeptidase is present in the insulin secretory pathway of the Cpe fat /Cpe fat mouse.
A newly reported enzyme, carboxypeptidase D (CPD), may be involved in the processing of secretory pathway peptides and partially compensate for the defective CPE in Cpe fat /Cpe fat mice (16). CPD is present in bovine pituitary and adrenal (16) and in many rat tissues. 2 In contrast, CPE is present mainly in neuroendocrine tissues, with high levels in pituitary and pancreatic islets, lower levels in other neuroendocrine tissues, and undetectable levels in liver (14,17). The major difference between the two enzymes is the size: CPE is approximately 50 -56 kDa, whereas CPD is 180 kDa in bovine pituitary (16) and 100 -180 kDa in various rat tissues. 2 CPD is not recognized by antisera raised against the N-or C-terminal regions of CPE (16).
The major purpose of the present study was to investigate whether CPE activity is present in the Cpe fat /Cpe fat mouse, or whether all of the detected activity is due to other enzymes (such as CPD). A related goal was to examine whether peptide processing was affected in Cpe fat /Cpe fat mouse tissues. For this, Leu-enkephalin (enkephalin) was chosen as a representative peptide that requires endopeptidase and carboxypeptidase cleavages. The finding that the level of enkephalin with a correctly processed C terminus is greatly reduced in the Cpe fat / Cpe fat mouse brain indicates that CPE is physiologically important for the processing of this peptide. However, CPE is not essential since some correctly processed enkephalin is detected in the Cpe fat /Cpe fat mouse brain. Furthermore, the finding that the peptide precursors are greatly elevated in the Cpe fat /Cpe fat mouse tissues suggests that CPE is required for an efficient endopeptidase processing reaction.

MATERIALS AND METHODS
Animals-Mice were bred at The Jackson Laboratory as described previously (15). In each litter, the Cpe fat /Cpe fat animals were identified by genetic markers and by the development of obesity. Non-obese littermates were used as controls in the various analyses. In the studies on CPE activity, no differences were found between animals heterozygous for the fat mutation and with wild-type animals, so these groups were pooled as indicated. The age of the animals ranged from 22 to 105 weeks.
Assay for CPE-like Enzymes in Mouse Tissues-Tissues were dissected and frozen on dry ice, and then stored at Ϫ20°C until analysis. Each tissue was homogenized in 0.2 to 2 ml, depending on tissue mass, of 0.1 M NaAc, pH 5.5, containing 1 mM phenylmethylsulfonyl fluoride. Protein concentration was determined using the Bradford assay, with bovine serum albumin as a standard. For the carboxypeptidase assay, 25 l of the homogenate, or 25 l of a 1:10 dilution, were combined with 50 mM NaAc, pH 5.0, 0.01% Triton X-100, and 200 M dansyl-Phe-Ala-Arg substrate in a final volume of 250 l. The substrate was prepared as described (18). In addition, tubes contained either 1 mM CoCl 2 or 1 M guanidinoethylmercaptosuccinic acid (GEMSA), each assayed in duplicate. Following incubation at 37°C for 30 -60 min, 100 l of 0.5 M HCl and 2 ml of chloroform were added, the tubes mixed, and then centrifuged at 500 ϫ g for 2 min. The amount of product was determined by measuring the fluorescence (excitation 350 nm, emission 500 nm) in the chloroform layer. Metallocarboxypeptidase activity is defined as the difference between activity measured in the presence of Co 2ϩ (an activator of CPE) and in the presence of GEMSA (an inhibitor of CPE). These conditions will also detect CPD (16), and may also detect carboxypeptidase N and carboxypeptidase M (19).
The enzymatic properties of affinity purified material were determined using a similar assay to that described above, except that 50 mM NaAc, pH 5.5, or Tris-Cl, pH 7.4, was used, and the tubes contained various ions or inhibitors in place of the Co 2ϩ or GEMSA. The samples were preincubated with inhibitors for 15 min at 4°C, and then substrate was added and the tubes incubated for 1 h at 37°C. For these experiments, carboxypeptidase activity was defined as the difference in fluorescence between the tubes containing enzyme and those with only buffer and substrate, and is expressed as the % of the control tube containing enzyme, buffer, and substrate but no divalent ions or inhibitors.
Purification of CPE-like Enzymes by Substrate Affinity Chromatography-Tissue homogenates were combined with Triton X-100 to a final concentration of 1%, and NaCl to a final concentration of 1 M in a volume of 10 ml. The samples were mixed, centrifuged for 30 min at 50,000 ϫ g at 4°C, and the supernatants applied to a p-aminobenzoyl-Arg Sepharose 6B column, as described (20,21). The columns (0.5-ml bed volume) were washed with 30 ml of 1 M NaCl and 1% Triton X-100 in 50 mM NaAc, pH 5.5, and rinsed with 5 ml of 10 mM NaAc, pH 5.5. CPE was eluted with 3.6 ml of 50 mM Tris-Cl, pH 8.0, containing 100 mM NaCl and 0.01% Triton X-100 (ϭelute 1). Then, the column was eluted with 3.6 ml of 5 mM Arg in the same buffer (ϭelute 2). Carboxypeptidase activity of each fraction was determined as described above. The final concentration of Arg in the carboxypeptidase assay was 0.5 mM, which does not interfere with enzymatic activity.
Western Blot Analysis of Purified Proteins-Aliquots of the affinity column elute fractions were dried in a vacuum centrifuge, resuspended in polyacrylamide sample buffer (containing 1% SDS), heated to 95°C for 5 min, and then loaded onto a 10% denaturing polyacrylamide gel. Following electrophoresis, the proteins were transferred to nitrocellulose overnight at 50 mA. The blot was probed with antisera raised against peptides corresponding to regions of CPE that are not conserved in other carboxypeptidases. The N-terminally directed antiserum was raised against a 15-residue peptide corresponding to the N terminus of the mature form of bovine CPE (21). This antiserum also recognizes rat and mouse CPE. The C-terminally directed antiserum was raised against the 9-residue peptide KMMSETLNF corresponding to the C terminus of the full-length mouse CPE. The peptide was coupled to keyhole limpet hemocyanin using the glutaraldehyde procedure, as described (22). Antisera were used at final dilutions of 1:1000. The enhanced chemiluminescence method (Amersham) was used as described by the manufacturer to detect the primary antisera. For the study examining the levels of CPD immunoreactive protein in control and Cpe fat /Cpe fat mouse tissues, the tissues were extracted directly with 1 M NaCl and 1% Triton X-100 and purified on 0.5 ml of affinity resin, as described above. Brain extracts were pooled from four control mice and from four Cpe fat /Cpe fat mice, and approximately 2 mg of protein used for the affinity column purification. Testis and duodenum extracts were pooled from two control mice and two Cpe fat /Cpe fat mice, and approximately 10 mg of protein used for the affinity purification. Heart extracts from two control mice and two Cpe fat /Cpe fat mice were pooled, and approximately 30 mg used for the affinity purification. The affinity columns were eluted first with 1.8 ml of Tris-Cl, pH 8.0, containing 100 mM NaCl and 0.01% Triton X-100 and then with 1.8 ml of 5 mM Arg in the same buffer. One-hundred l of the affinity column elute was analyzed on a Western blot, as described above, using an antiserum raised against rat brain CPD.
Analysis of Enkephalin and Prodynorphin-Brains were homogenized in 10 volumes of boiling acetic acid (0.1 M) and incubated for 30 min in a boiling water bath. Following centrifugation at 10,000 ϫ g for 30 min, the supernatants were lyophilized and stored at Ϫ70°C until analysis. An aliquot (30% of the total) of extract from two brains was rehydrated with 500 l of 0.15 M acetic acid containing 0.1% Triton X-100 and 0.15 M NaCl and centrifuged again at 10,000 ϫ g for 10 min. The supernatant was applied to a Sephadex G-50 column (1 ϫ 57 cm) pre-equilibrated with 0.15 M acetic acid containing 0.15 M NaCl and 0.1% Triton X-100. The column flow rate was 9 ml/h, and 3.5-min fractions were collected. Fractions were lyophilized and resuspended in 200 l of 0.15 M sodium phosphate buffer. One aliquot (60 l) of each fraction was treated with 5 g/ml tosylphenylalanylchloromethyl ketone-treated trypsin (Sigma) for 16 h followed by treatment with 5 ng/ml carboxypeptidase B (CPB, Sigma) for 120 min. The reaction was terminated by boiling for 20 min. Another aliquot of 60 l was treated only with CPB for 120 min. A third aliquot ("untreated") of 60 l was subjected to the same incubation and boiling conditions as above except that enzymes were omitted.
Immunoreactive Leu-enkephalin was measured in the untreated and enzyme-treated fractions by radioimmunoassay as described previously (23). The antiserum used to measure Leu-enkephalin in the radioimmunoassay does not recognize peptides with C-or N-terminal extensions of Leu-enkephalin (24). To measure prodynorphin (pro-enkephalin B), aliquots of the gel filtration fractions eluting at 30 kDa (18 ml) were pooled, dried in a vacuum centrifuge, resuspended in gel loading buffer, and subjected to electrophoresis and Western blot analysis as described above. The 13 S antiserum that recognizes the midportion of prodynorphin (25) was used at 1:500 dilution. Recombinant rat prodynorphin (a gift from Drs. Robert Day and Iris Lindberg) was used as a standard.

RESULTS
When assayed with dansyl-Phe-Ala-Arg at pH 5, many tissues contain a Co 2ϩ -activated GEMSA-inhibited carboxypeptidase activity (Table I). This assay will detect both CPE and CPD (16,18), and could also detect the neutral pH optimal carboxypeptidase N and carboxypeptidase M (19). High levels of carboxypeptidase activity are detected in the pancreatic islets and pituitary of the control mice, moderate levels are detected in the brain, adrenal, and duodenum, and low levels are detected in the testis and heart (Table I). In the Cpe fat / Cpe fat mice, the level of carboxypeptidase activity is only 6 -7% of the control level in pancreatic islets and pituitary (Table I). In brain, adrenal, and testis, the levels of carboxypeptidase activity in the Cpe fat /Cpe fat mice are approximately 50 -57% of the levels in control tissue. In heart and duodenum, the levels of carboxypeptidase activity in Cpe fat /Cpe fat mice are not statistically different from the levels in control mice ( Table I).
The finding that carboxypeptidase activity is not uniformly lower in the Cpe fat /Cpe fat mouse tissues, compared to the corresponding control mouse tissues, suggests that either the point mutation affects the enzyme differently in the various tissues, or that other enzymes are responsible for the carboxypeptidase activity detected at pH 5.0. To directly investigate these possibilities, brain extracts were fractionated on a substrate affinity column. The majority of the carboxypeptidase activity in control brain extracts elutes from the column when the pH is raised from 5.5 to 8 (elute 1, Table II); this condition has been previously shown to elute CPE from the affinity column (21). A smaller amount of enzyme activity is retained on the column in the high pH buffer, and subsequently elutes when 5 mM Arg is included in the buffer (elute 2, Table  II); this condition has been previously shown to elute CPD from the affinity column (16). In contrast to the results with control mouse brain, the Cpe fat /Cpe fat brain extracts contains very little carboxypeptidase activity that elutes with the high pH treatment (Table II). However, the amount of carboxypeptidase that elutes with the Arg is similar in Cpe fat /Cpe fat and control brain extracts (Table II).
The enzymatic properties of the carboxypeptidase in the elute 1 fraction from control brain (Table III) are similar to those previously reported for CPE (16,20). The activity is greatly stimulated by CoCl 2 , and inhibited by HgCl 2 , p-chloromercuriphenyl sulfonate (PCMPS), GEMSA, aminopropylmercaptosuccinic acid, and 1,10-phenanthroline (Table III). In addition, the enzyme is virtually inactive at pH 7.4 (Table III). The carboxypeptidase activity present in the second affinity column elute from control brain has properties similar to those of CPD (16). This activity is stimulated by CoCl 2 to a smaller extent than CPE, and is less sensitive to inhibition by HgCl 2 and PCMPS (Table III). In contrast, GEMSA and the peptide Hipp-Arg are better inhibitors of the elute 2 carboxypeptidase activity than the elute 1 activity (Table III). The properties of the carboxypeptidase activity in the second elute of the Cpe fat / Cpe fat mouse brain are similar to those of the corresponding fraction of control mouse brain (Table III), and to CPD (16). In contrast, the properties of the activity in the first elute from Cpe fat /Cpe fat brain are distinct from those of the control brain (Table III) or from CPE (16,20).
Western blot analyses of the material in the first and second elute fractions were performed to determine if any CPE protein was present in the Cpe fat /Cpe fat mouse extracts. Antisera to either the N-or C-terminal regions of CPE recognize proteins of 50 -56 kDa in the first elute fraction of control mouse brains (Fig. 1). The size of these proteins correspond to the range previously found for CPE (20,21). Immunoreactive CPE is not detected in any of the other fractions (Fig. 1).
Western blot analysis of affinity purified material from several tissues showed results similar to those found with brain (Fig. 2). Immunoreactive CPE was present in the affinity column eluates only from control mouse tissues and not from Cpe fat /Cpe fat mouse tissues. In the control tissues, the amount of immunoreactive CPE is highest in the brain (Fig. 2). Levels of immunoreactive CPE are lower in the testis, very low in heart, and not detectable in the duodenum.
To investigate whether CPD is up-regulated in the Cpe fat / Cpe fat mice to compensate for the absence of CPE activity, mouse tissues from 2-4 animals were extracted, purified on the affinity resin, and analyzed on Western blots using the antiserum raised against purified rat brain CPD (Fig. 3). No CPD is detected in the elute 1 fraction (not shown), consistent with previous studies (16). The elute 2 fractions of brain, testis, and heart show a major band of 180 kDa. Minor bands of 120 and 100 kDa are detected in brain and testis, and of 125 and 100 kDa in heart. Duodenum shows a major band of approximately 125 kDa. In all tissues, the relative amount of CPD immunoreactive protein is comparable between the control and Cpe fat / Cpe fat mouse tissues. This result suggests that CPD protein is not up-regulated to compensate for the defect in CPE activity in the Cpe fat /Cpe fat mouse. However, it is possible that the recovery of CPD during the purification process was not identical for the two samples. To investigate whether the CPD-like activity in whole brain extracts is altered in Cpe fat /Cpe fat mice, the sensitivity to PCMPS was examined. In this experiment, only the Co 2ϩ stimulated activity was measured, and the assay was performed at pH 5.5. Control brain extract shows a biphasic response to PCMPS, with approximately 60 -70% of the activity inhibited by 10 -100 M PCMPS (Fig. 4). In contrast, only 20 -30% of the activity in Cpe fat /Cpe fat mouse brain is inhibited by these low concentrations of PCMPS (Fig. 4). Whereas the activity measured in the absence of PCMPS is approximately 3-fold higher in the control brain compared to the Cpe fat /Cpe fat brain, there is no difference in carboxypeptidase activity measured in the presence of 30 M PCMPS (Fig. 4). This finding shows that only carboxypeptidase activities that are sensitive  to low concentrations of PCMPS are altered in the Cpe fat /Cpe fat brain, and the CPD-like activity is present in comparable levels.
To examine the processing of a representative peptide in the Cpe fat /Cpe fat mouse brain, two brains were pooled, extracted with 0.1 M acetic acid, and fractionated on gel filtration columns. The peak of enkephalin immunoreactivity in control brain extracts is detected in the low molecular mass fraction, consistent with a molecular mass of 555 Da (Fig. 5A). The antiserum used for this analysis does not react to a significant amount with enkephalin precursor peptides. To measure the amount of enkephalin with a C-terminally extended basic residue (i.e. Lys and/or Arg), the fractions were treated with CPB after separation on the gel filtration column. This treatment did not substantially alter the amount of enkephalin immunoreactivity (Fig. 5A). To measure the total amount of enkepha-  1. Western blot analysis of CPE in substrate affinity column eluates: comparison of control and Cpe fat /Cpe fat mice. Two brains from female Cpe fat /Cpe fat mice or from ϩ/ϩ wild type mice were pooled, extracted, purified on a p-aminobenzoyl-Arg affinity resin, and 100 l analyzed on a Western blot as described under "Materials and Methods." The blots were probed with antisera to the N-terminal region of CPE (left panel) or the C-terminal region of CPE (right panel). E1, affinity column elute 1 (high pH buffer); E2, affinity column elute 2 (5 mM Arg in high pH buffer).

FIG. 2. Western blot analysis of CPE in control (c) and Cpe fat /
Cpe fat (f) mouse tissues after purification on a substrate affinity column. Tissue was extracted, purified on the affinity resin, and analyzed on a Western blot as described under "Materials and Methods." Heart, testis, and duodenum were pooled from two male animals per group (either ϩ/ϩ or Cpe fat /Cpe fat ). For the control brain group, the samples were pooled from one male ϩ/ϩ and 3 female ϩ/fat mice. For the Cpe fat /Cpe fat group, two male and two female brains were pooled. Aliquots (100 l) of each affinity column elute were analyzed on the Western blot, and the blot was probed with an antisera directed against the C-terminal region of CPE, as described under "Materials and Methods."

FIG. 3. Western blot analysis of CPD in control and Cpefat/
Cpefat mouse tissues after purification on a substrate affinity resin. Tissues from two to four animals were pooled, purified on the affinity resin, and analyzed on a Western blot using an antiserum raised against purified rat brain CPD. Different amounts of protein were used for the affinity column procedure: brain (2 mg), testis (10 mg), heart (30 mg), and duodenum (10 mg). Approximately 5% of the affinity column eluate was used for the Western blot. The positions and molecular weights (in kDa) of prestained protein standards (Bio-Rad) are indicated.
FIG. 4. Effect of PCMPS on carboxypeptidase activity in control and Cpe fat /Cpe fat mouse brain. Homogenates were treated with the indicated concentration of PCMPS for 15 min at 4°C and then substrate was added and carboxypeptidase activity determined (at pH 5.5) as described under "Materials and Methods." lin-containing peptides in each of the gel filtration fractions, these fractions were treated (after the gel filtration) with a combination of trypsin and CPB. This analysis reveals a moderate amount of 1-4 kDa sized enkephalin-containing peptides in the control brain, and little material larger than 4 kDa (Fig.  5A). In contrast to the results with control mouse brains, the low molecular weight peak of immunoreactive-enkephalin is approximately 5-fold smaller in the Cpe fat /Cpe fat mouse brains (Fig. 5B). Treatment of these fractions with CPB causes a 5-fold increase in the level of immunoreactive enkephalin, and restores the level of immunoreactive enkephalin in the Cpe fat / Cpe fat mouse brain to that of the control brain (Fig. 5B). Treatment of the gel filtration fractions from the Cpe fat /Cpe fat mouse brains with trypsin/CPB reveals a relatively large amount of high-molecular weight enkephalin-containing peptides (Fig.  5B). The amount of this high molecular weight material is 2-3-fold higher in the Cpe fat /Cpe fat mouse brain extracts compared to the control brain extracts.
To directly compare the level of prodynorphin in the control and Cpe fat /Cpe fat mice, gel filtration fractions representing 25-30-kDa sized proteins were analyzed on a Western blot using an antiserum directed against the midportion of prodynorphin (Fig. 6). A 29-kDa protein is detected in both control and Cpe fat /Cpe fat mouse brain extracts; the size of this protein is similar to that of recombinant prodynorphin. The amount of immunoreactive prodynorphin is approximately 3-fold higher in the Cpe fat /Cpe fat brain extracts compared to control brain extracts (Fig. 6). DISCUSSION The major finding of the present study is that there is no active CPE in the Cpe fat /Cpe fat mouse. This conclusion is based on the absence of immunoreactive CPE in the material purified from the Cpe fat /Cpe fat mouse tissues using a substrate affinity column. This material does show a small amount of carboxypeptidase activity in the affinity column fractions corresponding to CPE, but the enzymatic properties are not identical to those of CPE. Thus, it is likely that the Ser 202 to Pro mutation of CPE found in the Cpe fat /Cpe fat mice causes the enzyme to be completely inactive. This result supports the previous finding that mutation of Ser 202 of rat CPE into a Pro results in an inactive protein when expressed in either insect cells using the baculovirus system, or expressed in mouse AtT-20 cells (26). The previous report that CPE activity was not completely eliminated, but was reduced by 90 -95% in the Cpe fat /Cpe fat mouse pituitary and pancreatic islets (15), did not take into account the possibility that other carboxypeptidases were responsible for the low levels of CPE-like activity in these tissues.
Until the recent discovery of CPD, CPE was the only known metallocarboxypeptidase with an acidic pH optimum and so this was the primary criteria previously used to distinguish CPE from carboxypeptidases M, N, B, and A (18). In addition to CPD, other members of the metallocarboxypeptidase gene family have recently been discovered, including a protein designated AEBP1 (27), and a novel cDNA isolated by homology cloning and tentatively designated CPZ. 3 Nothing is presently known regarding the enzymatic properties or the subcellular distribution of CPZ protein. AEBP1 was identified as a tran-3 L. Song and L. Fricker, submitted for publication.

FIG. 5. Gel filtration analysis of female control (A) and
Cpefat/ Cpefat (B) mouse brain extracts; analysis of Leu-enkephalin immunoreactivity. Mouse brains were extracted and subjected to gel-filtration chromatography on Sephadex G-50 as described under "Materials and Methods." Following the fractionation on a gel filtration column, one aliquot was directly assayed for immunoreactive Leuenkephalin (solid lines) as described under "Materials and Methods." Additional aliquots of the fractions after the gel filtration column were treated with CPB alone (dotted lines) or a combination of trypsin and CPB (dashed lines) prior to the assay for immunoreactive Leu-enkephalin. Molecular weight calibration standards: a, chymotrypsinogen (M r ϭ 25,000); b, cytochrome c (M r ϭ 12,400); c, aprotinin (M r ϭ 6, 500); d, dynorphin B (M r ϭ 1570); e, 125 I-Leu-enkephalin (M r ϭ 680).

FIG. 6. Western blot analysis of prodynorphin in control and
Cpe fat /Cpe fat mouse brain. Aliquots of the gel filtration-purified material were pooled and analyzed on a Western blot using an antiserum directed against the midportion of prodynorphin, as described under "Materials and Methods." The positions and molecular weights of prestained protein standards are indicated. scription repressor present in the nucleus and based on the lack of an N-terminal signal peptide, this protein is not expected to enter the secretory pathway where peptide processing occurs (27). Based on the immunoreactivity (Fig. 3), as well as previous studies on CPD (16), it is likely that CPD accounts the CPE-like activity in the second elute fraction of the affinity column. However, a small amount of CPE-like activity is detected in the high pH eluate from the affinity columns of the Cpe fat /Cpe fat mouse brain, and since this activity does not appear to be CPE based on the immunoreactivity and enzyme properties, it is possible that this activity is a novel carboxypeptidase. Still, this activity represents a minor amount of CPElike activity compared to either the levels of CPE in control mouse tissues, or to the amount of CPD in the control or Cpe fat /Cpe fat tissues.
The absence of active CPE in the Cpe fat /Cpe fat mice causes a large decrease in the levels of the fully processed peptides that were examined in the present study. As expected, the levels of the peptide precursors which contain C-terminal basic amino acids are markedly elevated in the Cpe fat /Cpe fat mouse tissues examined. In the brain extracts of Cpe fat /Cpe fat mice, CPB treatment produces a large increase in immunoreactive enkephalin (Fig. 5B) indicating a large amount of C-terminally extended peptide. In contrast, there is very little of the enkephalin precursor with C-terminal basic residues in control mouse brain since treatment with CPB fails to produce a measurable increase in enkephalin immunoreactivity (Fig. 5A). These results are similar to those found previously for proinsulin and proneurotensin in the Cpe fat /Cpe fat mouse (15,28). In the islets, the C-terminally extended processing intermediate form of insulin was undetectable in the control mice, but was abundant in the Cpe fat /Cpe fat mouse (15). In the brain, the majority of the neurotensin was found to be present as neurotensin-Lys-Arg (28). Taken together, these results indicate that the deficiency of CPE in the Cpe fat /Cpe fat mice leads to a dramatic accumulation of peptides with C-terminal basic residues, and a decrease in the levels of correctly processed peptides.
The finding that some C-terminal processing of the peptides occurs in the Cpe fat /Cpe fat mice implies that an alternative pathway exists. It is possible that CPD contributes to the processing of peptides. Previously, CPD has been found in vesicles of the bovine pituitary (16), and 5-10% of the total cellular pool of the duck homologue (gp180) has been detected on the surface of hepatocytes (29), indicating transit through a secretory pathway. The lack of up-regulation of CPD protein in Cpe fat /Cpe fat mouse tissues (Fig. 3) implies that this enzyme is not induced to compensate for the absence of CPE activity. In addition to CPD, it is possible that other carboxypeptidases are present in the secretory pathway and contribute to peptide processing. Alternatively, some of the endopeptidases may initially cleave the peptide precursors to the N-terminal side of the basic residues, thereby generating the C-terminally processed peptide without the need for a carboxypeptidase. Although the well studied prohormone convertases 1 and 2 do not appear to cleave to the N-terminal side of the basic residues within the cleavage site (30,31), other enzyme activities have been reported which can perform this cleavage (32,33).
The increase in prodynorphin and other high molecular weight enkephalin-containing peptides in the Cpe fat /Cpe fat mouse brain is not predicted from the known function of CPE. The absence of CPE activity was expected to cause only the accumulation of peptides with C-terminal basic residues attached. The high molecular weight enkephalin-containing pep-tides result from incomplete endopeptidase cleavage of the precursors prodynorphin, which contains 3 copies of Leu-enkephalin, and proenkephalin, which contains a single copy of Leu-enkephalin and 6 copies of Met-enkephalin. Peptides larger than 2 kDa are a small fraction of the total enkephalincontaining peptides in the control brains, but are roughly comparable to the level of fully processed enkephalin in the Cpe fat / Cpe fat mouse brain (Fig. 5). A similar phenomenon was found previously in the processing of proinsulin in the Cpe fat /Cpe fat mice; proinsulin represents only 5% of the total peptide (insulin, proinsulin, and processing intermediates) in control mouse islets, and 40% of the total in the Cpe fat /Cpe fat mouse islets (15). These results suggest that the defect in CPE inhibits the endopeptidase processing reaction. It is possible that CPE is required to activate the endopeptidases, or to remove an inhibitor. Evidence in favor of the latter possibility has been reported for the endopeptidase PC2 and the endogenous inhibitor 7B2 (34). Further studies are needed to further explore the reason for the accumulation of high molecular weight peptides in the Cpe fat /Cpe fat mouse, and to examine the effect of the CPE mutation on the processing of other peptides and proteins.