A Novel Intracellular Peptide Derived from G1/S Cyclin D2 Induces Cell Death*

Background: Intracellular peptides probably regulate several biological processes. Results: pep5 derived from G1/S cyclin D2 specifically increases during the S phase of the cell cycle and, reintroduced into the cell, induces apoptosis and necrosis. Conclusion: pep5 has potential therapeutic applications and could have biological functions. Significance: pep5 discovery advances our understanding of limited proteolysis. Intracellular peptides are constantly produced by the ubiquitin-proteasome system, and many are probably functional. Here, the peptide WELVVLGKL (pep5) from G1/S-specific cyclin D2 showed a 2-fold increase during the S phase of HeLa cell cycle. pep5 (25–100 μm) induced cell death in several tumor cells only when it was fused to a cell-penetrating peptide (pep5-cpp), suggesting its intracellular function. In vivo, pep5-cpp reduced the volume of the rat C6 glioblastoma by almost 50%. The tryptophan at the N terminus of pep5 is essential for its cell death activity, and N terminus acetylation reduced the potency of pep5-cpp. WELVVL is the minimal active sequence of pep5, whereas Leu-Ala substitutions totally abolished pep5 cell death activity. Findings from the initial characterization of the cell death/signaling mechanism of pep5 include caspase 3/7 and 9 activation, inhibition of Akt2 phosphorylation, activation of p38α and -γ, and inhibition of proteasome activity. Further pharmacological analyses suggest that pep5 can trigger cell death by distinctive pathways, which can be blocked by IM-54 or a combination of necrostatin-1 and q-VD-OPh. These data further support the biological and pharmacological potential of intracellular peptides.

Intracellular peptides are constantly produced by the ubiquitin-proteasome system, and many are probably functional. Here, the peptide WELVVLGKL (pep5) from G 1 /S-specific cyclin D2 showed a 2-fold increase during the S phase of HeLa cell cycle. pep5 (25-100 M) induced cell death in several tumor cells only when it was fused to a cell-penetrating peptide (pep5-cpp), suggesting its intracellular function. In vivo, pep5cpp reduced the volume of the rat C6 glioblastoma by almost 50%. The tryptophan at the N terminus of pep5 is essential for its cell death activity, and N terminus acetylation reduced the potency of pep5-cpp. WELVVL is the minimal active sequence of pep5, whereas Leu-Ala substitutions totally abolished pep5 cell death activity. Findings from the initial characterization of the cell death/signaling mechanism of pep5 include caspase 3/7 and 9 activation, inhibition of Akt2 phosphorylation, activation of p38␣ and -␥, and inhibition of proteasome activity. Further pharmacological analyses suggest that pep5 can trigger cell death by distinctive pathways, which can be blocked by IM-54 or a combination of necrostatin-1 and q-VD-OPh. These data further support the biological and pharmacological potential of intracellular peptides.
Cell signaling induces modification of the protein interactome network. The ubiquitin-proteasome system is responsible for maintaining protein homeostasis in cells, and among its essential functions is its involvement with cancer biology and the immune system. The proteasome is responsible for the initial cleavage of antigenic proteins and the generation of the major histocompatibility complex class I (MHC-I)-bound antigens, which can later be trimmed within the endoplasmic reticulum by endoplasmic reticulum aminopeptidases to produce the correct N terminus to interact with the MHC-I (1,2). In cancer biology, the ubiquitin-proteasome system is important for cell cycle progression because it targets cyclins for degradation (3). Proteasome inhibition is used clinically for the treatment of certain types of cancer and is the treatment of choice for multiple myeloma (4). Protein degradation by the proteasome generates intermediate peptides shown to contain from 2 to 21 amino acids (5,6), which in theory suggests that cells are continually producing large quantities of peptides containing 2-21 amino acids. However, according to current knowledge, only one peptide from each cellular protein would escape postproteasome proteolytic degradation to be presented at the cell surface bound to the MHC-I, suggesting that ϳ10,000 peptides cover the mammalian cell surface to allow self-recognition by the immune system (7,8). Thus, almost the entire protein is believed to be recycled to amino acids within cells. Indeed, in yeast, mammalian cells, and flies, the injurious consequences of proteasome inhibition are rescued by amino acid supplementation, revealing the importance of the proteasome system in amino acid recycling for de novo protein synthesis (9). Further reports have suggested that peptides produced by the proteasome have a half-life of seconds (10 -12). In specific circumstances, the ubiquitin-proteasome system can perform limited proteolysis, such as in the generation of the active dimeric NFB transcriptional complexes (13).
Therefore, it is clear that the proteasome can perform both extensive and limited proteolysis, depending on the protein substrate. In our laboratory, we have shown that the proteasome could play a more extensive role in limited proteolysis than previously anticipated, generating intracellular peptides (14 -16). These findings were based on a well established con-cept that rationally designed peptides, structurally similar to the ones produced by the proteasome, can regulate protein spatial localization within cells and control cell signal transduction (17,18). As such, naturally occurring intracellular peptides generated by the proteasome would constitute an as yet poorly understood mechanism by which cells increase their protein network complexity and function (16). Hemopressin, the first intracellular peptide identified using this rationale (19), was shown to have cannabinoid inverse agonist action regulating food intake (20,21), whereas the natural brain hemopressins are secreted and suggested to play an important role as novel endocannabinoids (14,22).
Later it was shown that intracellular peptides can function in modulating signal transduction from inside the cells because peptides structurally related to proteasome products were identified by mass spectrometry, chemically synthesized, and reintroduced into cells, where they modulated both angiotensin II and ␤-adrenergic signal transduction (23). These peptides were used for affinity chromatography and were suggested to bind to a specific set of proteins, many involved in protein and vesicular traffic (23). In addition to the proteasome, thimet oligopeptidase (EC 3.4.24.15; EP24.15), which is an intracellular peptidase that only degrades small peptides (ϳ5-17 amino acids), was also shown to participate in intracellular peptide metabolism (24). By manipulating intracellular EP24.15 activity either by overexpressing the enzyme or inhibiting its activity by means of siRNA, it was possible to modulate G-protein-coupled receptor signal transduction in HEK293 and CHO-S cells (23,25). These data suggest a previously unknown connection between intracellular peptide metabolism and signal transduction. Other signal transduction pathways could also be related to intracellular peptides because two similar peptides identified in the Wistar rat adipose tissue where shown to bind specific proteins and facilitate insulin-induced glucose uptake in 3T3-L1 adipocyte cells (26). Although the intracellular peptides have not yet been shown to directly modulate protein-protein interactions in vivo, the in vitro use of surface plasmon resonance demonstrates that at concentrations of 1-50 M, several intracellular peptides can modulate the interactions of calmodulin and 14-3-3⑀ with proteins from the mouse brain cytoplasm or with recombinant EP24.15. One of these peptides (VFDVELL; VFD-7), shown to be a proteasome product (24), increases the free cytosolic Ca 2ϩ concentration in a dose-dependent manner but only if introduced into HEK293 cells (27).
In the present report, we aim to obtain further information on the cell biology and therapeutic potential of intracellular peptides by investigating their possible participation in the cell cycle. To that end, we identified in extracts of HeLa cells a novel peptide fragment (WELVVLGKL; pep5) that specifically increases during the S phase of the cell cycle and is derived from the G 1 /S-specific cyclin D2 protein. The peptide pep5 induces cell death in HeLa and several other tumor cells and in vivo reduces by 50% the volume of the rat C6 glioblastoma. Collectively, the above results suggest that peptides generated by the proteasome and additional intracellular peptidases need further attention as novel natural modulators of cell function. These data suggest the therapeutic potential of intracellular peptides.
Cell Cycle Synchronization by Double-thymidine Block-HeLa cells were synchronized using the double-thymidine block procedure (31,32). Thymidine (Sigma-Aldrich) was diluted in serum-free DMEM and stored at 4°C before use. For cell cycle synchronization, HeLa cells were treated with 2 mM thymidine for 18 h and washed in phosphate-buffered saline (PBS), cell medium was replaced with complete medium, and cells were cultured for an additional 9 h at 37°C and 5% of CO 2 . The above procedure was repeated once more to arrest cells in S phase. At different times (4, 10, and 16 h, at which ϳ84% of cells were in S, G 1 , and G 2 /M phases, respectively) cells were harvested, analyzed by flow cytometry, and submitted to additional assays. Control cells were asynchronous.
Flow Cytometry-Cell cycle was analyzed by flow cytometry using a Guava Easy City Mini Flow Cytometry (Millipore) instrument. Cells were fixed in 70% ethanol for 1 h and incubated with propidium iodide (1 mg/ml) and RNase (10 mg/ml) for 30 min on ice, and a total of 10,000 events were analyzed in each sample.
Apoptosis Assay-Cells were treated with pep5-cpp, cpp alone, and ⌬N or ⌬C-pep5-cpp for 10 or 30 min. After incubation, cells were harvested and washed in cold phosphate-buffered saline (PBS) and resuspended in 100 l of 1ϫ annexin-binding buffer (Molecular Probes). Alexa Fluor 488 annexin V (5 l) was added with 1 l of 100 g/ml PI working solution to each 100 l of cell suspension. The samples were incubated at room temperature for 15 min, and after the incubation period, 400 l of 1ϫ annexinbinding buffer were added, mixed gently, and kept on ice. As soon as possible, the samples were analyzed by flow cytometry, measuring the fluorescence emission at 530 and Ͼ575 nm.
Surgery-C6 rat glioma (Rattus norvegicus) (33) cells were grown in DMEM containing 10% fetal bovine serum and antibiotics (penicillin/streptomycin). Cells in the exponential phase of growth were used, and a suspension was prepared in sterile saline at a concentration of 5 ϫ 10 5 cells/4 -5 l. Adult male

Summary of protein precursors and sequence of the peptides identified in HeLa cells along the progression of the cell cycle
Protein name, precursor protein of the respective peptide sequence; pep#, peptide identity, according to the order in which it was sequenced; Sequence, the respective peptide sequence identified by MS/MS analyses (all of the MS and MS/MS spectra are presented in the supplemental material); z, the peptide charge state; #T, the number of tags in the identified peptide; Obs M, observed mass; Theor M, theoretical mass of the peptide; ppm, mass accuracy; Ratio, relative level of each peptide in the specific cell cycle phase indicated (G 1 , S, or G 2 )/control asynchronous cells. Except for peptides 16  Wistar rats of 250 -350 g (n ϭ 5) were anesthetized with an intramuscular injection of ketamine/xylazine (10 mg/1.5 mg/100 g of body weight) to provide deep anesthesia and analgesia. The rats were placed on a stereotaxic surgical table, a midline incision was made, and a burr hole was drilled 0.48 mm anterior and 3 mm lateral to the bregma. The C6 cell suspension was slowly injected into the striatum using a Hamilton syringe at a depth of 5.4 mm to the bone surface, and the needle was left in situ for 3 min before its removal. After 14 days, Alzet osmotic pumps containing either pep5-cpp or ⌬N-pep5-cpp (100 M) diluted in artificial cerebrospinal fluid were surgically implanted and attached to Alzet brain infusion kits. Artificial cerebrospinal fluid was chosen as a vehicle solution in order to mimic more closely the composition of the interstitial fluid within the brain. The peptide concentrations were chosen based on our previous results using cell lines. The pump infusion rate was 0.5 l/h with a duration of 2 weeks. After a further 14 days, the rats were killed by transcardiac perfusion with 4% formaldehyde in 0.1 M phosphate buffer, pH 7.4. After perfusion, the brains were removed, fixed in 4% formaldehyde, and cryoprotected with 30% sucrose in PBS. The frozen tissue sections (30 m) were obtained on a freezing microtome. The brain slices were stained with H&E, and tumor area was analyzed under a microscope. The largest areas of tumor were measured using image analysis software (Image ProPlus). The volume was calculated using the formula, V ϭ FIGURE 3. Effect of intracellular peptides in HeLa cells. HeLa cells were treated with peptide 3 or 5 free or fused to a cpp, cpp alone, or SCB for 24 h. After that, the cells were fixed in 70% ethanol and analyzed by propidium iodide staining and flow cytometry. Results were considered significant when p was Յ0.001 (***). The data shown here are representative of three independent experiments performed in triplicate. NT, not treated; pep-cpp, cpp fused to the C terminus of the peptides; cpp-pep, cpp fused to the N terminus of the peptides. The histograms show the difference among the groups treated with pep5 bound or not to the cpp in the N or C terminus. Error bars, S.E. FIGURE 2. MS spectra representing the semiquantitative analysis of peptides from synchronized HeLa cell extracts (S, G 2 /M, and G 1 ) or controls. HeLa cells were synchronized with 2 mM thymidine, and the peptides were labeled with isotopes TMAB (D0, D3, D9, and D12). A and B, representative MS spectra of two peptides (pep5 and pep3) that reproducibly changed during the HeLa cell cycle. The data shown here are representative of two independent experiments. /6 ϫ L ϫ W ϫ H. This procedure was approved by the Ethical Commission for Animal Experimentation of the Biomedical Sciences Institute (University of São Paulo), protocol number 116/08.
Peptide Extraction and Quantification-HeLa cells (1 ϫ 10 6 cells/plate) were grown in 10-cm cell culture plates (Corning Inc.) in complete medium. Fifteen plates of cells were used per group, and each group was characterized by flow cytometry to be in specific phases of the cell cycle, meaning asynchronous, G 1 , S, or G 2 /M. Complete medium was removed, and cells were washed three times with PBS, scraped from the plates, and centrifuged at 830 ϫ g for 5 min. The pellet was resuspended in 10 ml of 80°C water and incubated in an 80°C water bath for 20 min. The cell lysates were stored at Ϫ80°C overnight. To extract peptides, samples were acidified with 10 l of ice-cold 0.1 M HCl to a final concentration of 10 mM HCl and sonicated three times with 20 pulses (4 Hz) on ice. The samples were centrifuged at 2,500 rpm in a microcentrifuge for 40 min at 4°C, and the supernatant was collected for peptide extraction. The supernatant was ultracentrifuged at 34,000 rpm for 60 min at 4°C, transferred to a centrifugal filter device for the separation of molecules of less than 5,000 Da (Millipore), and centrifuged again at 2,500 rpm for 1 h at 4°C. The peptide quantification reaction was performed at pH 6.8 to ensure that only the amino groups of peptides and not free amino acids reacted with fluorescamine, as described previously (23,34). Briefly, 2.5 l of peptide samples were mixed with 25 l of 0.2 M phosphate buffer (pH 6.8) and 12.5 l of a 0.3 mg/ml acetone fluorescamine solution. After vortexing for 1 min, 110 l of water was added, and fluorescence was measured with a SpectraMax M2e plate reader (Molecular Devices) at an excitation wavelength of 370 nm and an emission wavelength of 480 nm. A peptide mixture of known composition and concentration was used as the standard reference for determining the peptide concentration in the experimental samples.
Isotopic Labeling and Mass Spectrometry for Peptide Identification-The labeling procedure has been previously described in detail (35,36). In brief, each group of cells within an experiment was labeled with one of the isotopic TMAB-Nhydroxysuccinimide labels. The S, G 2 /M, or G 1 groups of characterized cells were labeled with D3-, D9-, or D12-TMAB, respectively (Run 1), whereas the control asynchronous cells were labeled with D0-TMAB. This labeling was altered (reverse labeling) between experiments (Run 2). These two experiments above were performed independently. A total of 50 g of the peptide extract was combined with 250 l of 0.4 M phosphate buffer, pH 9.5, and 15 mg of TMAB label per group. After the addition of TMAB, the mixture was incubated at room temperature for an additional 60 min. To quench any remaining labeling reagent, 30 l of 2.5 M glycine was added to the reaction. The sample was dried in a vacuum centrifuge and analyzed by liquid chromatography/mass spectrometry (LC-MS/MS) on a Synapt G1 mass spectrometer with a nanoACQUITY UltraPerformance LC System (Waters Co.). The MS spectra were analyzed using the MassLynx software (Waters) to identify groups of peaks representing peptides labeled with the different isotope tags. Quantification was performed by determining the relative intensity of each isotopic peak (37). To identify the peptides, the MS/MS data were analyzed using the Mascot search engine (Matrix Science Ltd.).
Peptide Synthesis-Peptides 3 and 5, whose concentrations were altered in different phases of the cell cycle, and a control scrambled peptide were synthesized with cpp (YGRKKR-RQRRR) covalently bound to either their C or N terminus (38 -40). In all assays, peptide purity was greater than 95%. All synthesized peptides were provided by Proteimax Biotechnology LTDA.
Caspase Assays-Caspase 3/7, 8, and 9 activities were evaluated in extracts of HeLa cells previously treated for 10 min with either the proapoptotic pep5-cpp or the control peptide (100 M), using a commercial luminescent assay kit (Caspase-Glo assay, Promega). The relative luminescence was measured in a SpectraMax luminometer (Molecular Devices).
Phospho-MAPK Array-A human phospho-MAPK array kit (Proteome Profiler) was used to simultaneously test pep5-cpp effects on the phosphorylation of diverse MAPK signaling pathways, according to instructions provided by the manufacturer. Briefly, all arrays were incubated with 200 g of HeLa cell extracts treated previously for 10 min with either pep5-cpp or the control peptide (100 M).
Proteasome Activity Assays-HeLa cells treated or not with pep5-cpp (50 -200 M) were washed twice with PBS and lysed on ice for 30 min. Cells extracts were centrifuged at 13,000 rpm for 10 min, and total protein concentration was determined by the Bradford assay (41) using bovine serum albumin as a standard. The proteasome substrate Glo TM Chymotrypsin-like (Suc-LLVY-Glo TM , Promega) was used to determine the proteasome activity, as recommended by the manufacturer.
Statistics-Values are expressed as means Ϯ S.E. Statistical analyses were conducted by Student's unpaired t test for independent samples or analysis of variance followed by Tukey's or Bonferroni's test to compare more than two groups, using GraphPad Prism version 5.0. p values of Ͻ0.05 were considered significant.

RESULTS
The initial hypothesis of our work was that the relative concentration of specific intracellular peptides, similarly to that of certain specific proteins, could vary throughout the cell cycle, and perhaps even contribute to its control. Therefore, we began by synchronizing HeLa cells using the double-thymidine block procedure (31,32) to further isolate and characterize intracellular peptides from each individual phase of the cell cycle. As a result of the double-thymidine block, analyses by flow cytometry suggested that ϳ84% of HeLa cells were arrested in each phase of the cell cycle (G 1 , S, or G 2 /M) when compared with the asynchronous cell group (Fig. 1). The intracellular peptide content was then extracted from asynchronous, G 1 , S, or G 2 /M HeLa cells, labeled with specific TMAB isotopes, and analyzed by mass spectrometry. In these analyses, 19 peptides were sequenced and identified in order of appearance (Table 1; MS and MS/MS data are shown in the supplemental material). pep3, a fragment from the 40 S ribosomal protein S21, and pep5, a fragment of G 1 /S-specific cyclin D2 (Table 1), were observed to clearly fluctuate along the cell cycle (Fig. 2) and were further investigated with respect to their possible biological activity on regulating the progression of the cell cycle and/or cell survival. pep3, pep5, and SCB were chemically synthesized either free or covalently bound to a cpp at their C or N terminus. The cpp was necessary to allow cell penetration and consequent investigation of the possible intracellular function of these peptides. Moreover, the cpp used herein has been extensively described to transport its cargo into the cells (37,38).
First, HeLa cells were treated with each of these peptides at a specific concentration (100 M), and after 24 h, they were analyzed by flow cytometry (Fig. 3). The control peptides SCB and pep3, either free or bound to the cpp, or the cpp alone had no effect on cell cycle progression or cell death (Fig. 3). pep5 had no effect on cell cycle progression, whereas it killed ϳ90% of cells when cpp was linked to its C terminus and ϳ40% when cpp was linked to its N terminus (Fig. 3). A dose-response curve of pep5cpp (1-100 M) was performed in HeLa cells, showing cell death activity at concentrations above 50 M (Fig. 4, A and B). A time course was also performed, indicating that pep5-cpp (50 -100 M) starts to induce significant cell death in HeLa cells after 2 h of incubation (Fig. 4C). HeLa cell morphology was observed following the treatments with SCB, pep5, cpp, SCB-cpp, or pep5-cpp (50 or 100 M; 24 h). These results corroborate the suggestion that pep5-cpp at either 50 or 100 M is inducing large cell death (Fig. 5).
Next we performed several modifications in the pep5 sequence to investigate the relationship between structure and activity. In order to identify the minimal active sequence, we started by deleting 1, 2, or 3 amino acids from either the N or C terminus. Removal of a single amino acid from the N terminus (⌬N-pep5-cpp) totally abolished the induction of cell death by pep5-cpp (Fig. 6). The potency of pep5-cpp was only slightly reduced by the removal of two or three amino acids from its C terminus, whereas it was totally abolished after the removal of four amino acids from the C terminus (Fig. 6A). Additional experiments were performed, substituting specific hydrophobic amino acids (Leu or Val) from the original sequence by Ala, which inactivated the cell death activity of pep5-cpp (Fig. 6B). pep5-cpp was also acetylated at the N terminus, and the results suggested that the cell death activity is retained, although with reduced potency compared with the original sequence (Fig. 6C). Table 2 summarizes the pep5-cpp sequences investigated herein. Taken together, these data suggest a strong structural specificity of pep5-cpp in inducing cell death.
Next, we attempted to investigate the possible cell deathinducing mechanism of pep5-cpp. HeLa cells were treated with pep5-cpp or the cpp alone (50 M) for 30 min and analyzed by flow cytometry in the presence of annexin V (Molecular Probes). Results suggest that pep5-cpp, but not cpp alone, significantly induces apoptosis and necrosis in HeLa cells (Fig. 10). After a 30-min treatment with pep5-cpp, apoptotic cells corresponded to ϳ30% of the total population, and necrotic cells corresponded to ϳ60%, whereas ϳ10% of the cells remained viable (Fig. 10A). Similar results were obtained using either pep5-cpp or its minimal active sequence (⌬C-pep5-cpp), which has three amino acids deleted from the C terminus (Fig. 10B). HeLa cells treated for 10 min with a lower dose of pep5-cpp (50 M) showed no difference between apoptosis versus necrosis (Fig. 10C). Those cells treated with pep5-cpp (100 M, 10 min) showed a significant increase of caspase 3/7 and 9 activities, whereas the activity of caspase 8 remained unaltered (Fig. 11). Treatment of HeLa cells with the control SCB peptide caused no cell death or caspase activation (Fig. 11).
Next, the mitogen-activated protein kinase (MAPK) pathway was evaluated in a standard array assay, which simultaneously investigates the phosphorylation of several kinases, including extracellular signal-regulated kinases (ERKs), c-Jun N-terminal or stress-activated protein kinases (JNK/SAPK), ERK/big MAP kinase 1 (BMK1), and the p38 kinases (42). HeLa cells were treated with pep5-cpp (100 M, 10 min), the cell extracts were analyzed in duplicates according to the manufacturer's instructions, and the results were compared with untreated HeLa cells. Specific kinases or their substrates, such as ERK1/2, HSP27, p38 ␣/␥, and p70 S6 kinase had increased phosphorylation upon treatment with pep5-cpp, when compared with the untreated control group. However, substrates, such as Akt2 and GSK-3␤, had decreased phosphorylation after treatment with pep5-cpp (Fig. 12A). Additional experiments were similarly conducted to  ⌬N, one or more amino acids deleted from N terminus; ⌬C, one or more amino acids deleted from C terminus; A, B, and C, amino acids substitutions for the minimal active sequence (⌬C 3 -pep5-cpp*) of pep5-cpp. Ac-pep5-cpp, ⌬C 3 -pep5-cpp acetylated in the N terminus.  compare the effect of cell death inducer pep5-cpp with the cell death-inactive ⌬N-pep5-cpp; the latter peptide only lacks the original N-terminal amino acid residue from pep5-cpp. These results further suggest the structural specificity of pep5-cpp to inhibit Akt2 and to activate ERK2, HSP27, and p38␥ phosphorylation (Fig. 12B). The ability of pep5 to affect the chymotrypsin activity of the proteasome was evaluated both in whole cells and in cell extracts FIGURE 9. In vivo effects of ⌬N-pep5-cpp or pep5-cpp in the rat C6 glioma. Male Wistar rats (n ϭ 5) with 14-day pre-established tumors were treated with ⌬N-pep5-cpp or pep5-cpp (100 M) for an additional 14 days, and after infusion, the animals were killed with an overdose of ketamine-xylazine (Parke-Davis). After perfusion, the brains were removed, fixed in 4% formaldehyde, and cryoprotected with 30% sucrose in PBS. The frozen tissue sections (30 m) were obtained on a freezing microtome. The brain slices were stained with H&E, and tumor area was analyzed under a microscope. The largest areas of tumor were measured using image analysis software (Image ProPlus). The volume was calculated using the formula, V ϭ /6 ϫ L ϫ W ϫ  H. A, tumor volume. B-D, H, and I, ⌬N-pep5-cpp. E-G, J, and K, pep5-cpp  using a standard commercial assay. In HeLa cells treated with pep5-cpp for 24 h, the proteasome activity was significantly reduced (Fig. 13A). pep5 also showed inhibitory activity of the chymotrypsin activity of the proteasome in cell extracts (Fig. 13B).

Amino acid sequences from pep5-cpp Abbreviation
Furthermore, pharmacological treatments using distinctive inhibitors of cell death were conducted (Fig. 14). Necrostatin-1, an inhibitor of necroptosis (43); qVD, a potent caspase inhibitor that protects cells from capase-dependent apoptosis (44); SB203580, a selective inhibitor of p38 mitogen-activated protein kinase, which inhibits also the phosphorylation and activation of protein kinase B (PKB, also known as Akt1) (45,46); and/or IM-54, a cell-permeable, potent, and selective inhibitor of oxidative stress-induced necrosis (47) were evaluated (30 M, 1 h of pretreatment). Used alone, only the IM-54 significantly reduced the cell death induction caused by pep5-cpp. On the other hand, only in combination were nec-1 and qVD able to reduce the cell death induced by pep5-cpp in HeLa cells. Nec-1, qVD, or SB203580 treatment alone was not effective in blocking pep5-cpp-induced cell death (Fig. 14).

DISCUSSION
In the present study, several novel intracellular peptides were identified in HeLa cells. The relative levels of some of these intracellular peptides appear to fluctuate throughout the cell cycle, which resembles that which is already well established for proteins controlling cell cycle progression. Whereas the broad biological significance of these findings remains elusive, to our knowledge this is the first report to describe fluctuation in the relative levels of intracellular peptides during the cell cycle progression of HeLa cells.
The relative level of the peptide WELVVLGKL (pep5) increases 2-fold in the S phase of HeLa cell cycle, compared with asynchronous cells. Reintroduction of pep5 in HeLa or several additional tumor cell lines induces cell death, suggesting its functionality. In a previous study in HEK293T cells, the concentration of the intracellular peptide VFD-7 was determined to be ϳ16 M (27). It is possible that the minimal effective pharmacological dose of pep5 that induces cell death in MDA-MB-231 cells (ϳ25 M) could be occurring within cells. Differences in the kinetics of administration and/or coupling to a transport mechanism could be related to the distinctive efficacy of pep5 seen in the tumor cell lines investigated in this study. A high structural specificity was observed for pep5, based on the finding that a single amino acid removal, modification, or substitution in the minimal active sequence (WELVVL) reduced or abolished pep5 cell death activity. Therefore, it is also possible that differences in the efficacy of pep5 observed between cell lines are due to variation in the activity of proteases and/or peptidases. Indeed, it has been shown previously that oligopeptidases (24,25) as well as the proteasome (15,48) can metabolize a number of intracellular peptides. FIGURE 11. Caspase assay after treatment with pep5-cpp. Cells were treated with peptide 5 and SCB (scrambled) for 10 min (100 M). After treatment, the substrates were added for the detection of different caspase activities. The relative luminescence was read using a luminometer (Spectramax, Molecular Devices). BL, blank; NT, not treated; SCB, control; Pep5, pep5-cpp. The data shown here are representative of three experiments performed in triplicate, and the results were considered significant when p was Ͻ0.001 (***). RLU, relative light units. Error bars, S.E. Pharmacological examination conducted in HeLa cells suggests that IM-54, a potent and selective inhibitor of oxidative stress-induced necrosis (47), efficiently blocked the cell death activity of pep5. IM-54 selectively inhibits oxidative stress-induced necrosis, as observed in ischemia-reperfusion injury caused by heart attack, and does not inhibit caspase-dependent apoptosis (47). IM-54 is suggested to directly interact with mitochondrial protein(s), which may therefore play a critical role in induction of necrosis. Some studies have demonstrated that necrosis induced by H 2 O 2 treatment in HL60 cells also is inhibited by IM-54 (47,49). Together, these data suggest that pep5 can trigger cell death through the oxidative stress pathway.
Other pathways may also be involved in the induction of cell death by pep5. The simultaneous inhibition of necroptosis and caspase activity by a combination of nec-1 and qVD inhibitors efficiently blocked pep5-cpp cell death activity. Neither of these inhibitors individually reduced cell death induced by pep5-cpp. Mixed type cell death modes containing features of both forms of cell death have been reported and named "aponecrosis." This form of cell death may represent aborted or partially executed apoptotic programs, which occur in the context of a final necrotic outcome (50,51). The existence of necrotic cell death pathways regulated by an intrinsic death program distinct from that of apoptosis has also been proposed and named necroptosis (52). In apoptosis, caspases 3/7 are effector caspases involved in both the intrinsic and extrinsic pathways. Caspase 8 is involved in the extrinsic apoptotic pathway, and it is activated by ligand binding to membrane-associated death receptors, whereas caspase 9 participates in the intrinsic apoptotic pathway, and it is activated by mitochondrial perturbation, which causes cytochrome c release (53)(54)(55)(56).
The phosphorylation levels of only six MAPKs or their substrates were affected by pep5 among the 26 proteins evaluated, including Akt2 inhibition and p38␥ and p38␣ activation. It is known that Akt2-deficient mice have an increase in apoptotic cell death in the aorta, whereas the knockdown of all three isoforms of Akt (Akt1/Akt2/Akt3) induces apoptosis in several human tumor cell lines (57). On the other hand, Akt2 overexpression causes an increase in adhesion, invasion, and metastasis in ovarian and human breast cancer cells (58,59). Other reports show that Akt2 knockdown in non-small cell lung cancer simultaneously causes cleavage of one antiapoptotic protein, cytochrome c release, and caspase activation (60). In asynchronous HeLa cells, pep5 was not able to induce cytochrome c release (data not shown). pep5 has a similar effect in HeLa cells, inhibiting Akt2 and activating caspase 9. Moreover, mitogenactivated protein kinases such as p38␥ and p38␣ were activated by treatment of HeLa cells with pep5, suggesting that pep5 induces a cross-talk in proapoptotic pathways. Some studies suggest that p38␥ is regulated by hypoxia, decreasing the levels of cyclin D1 in PC12 cells, and by ionizing radiation in NIH-3T3 cells, inducing cell cycle arrest (61,62). Furthermore, when p38␣ is inhibited, it blocks a checkpoint in G 2 /M after ultraviolet radiation in murine and human cells (63). Here, when HeLa cells were treated with the specific p38 inhibitor SB203580, no inhibition of pep5-cpp cell death activity was observed. Therefore, the present results do not suggest that Akt2 inhibition or p38␥/p38␣ activation are key effectors of pep5-cpp cell death activity in HeLa cells. One exciting possibility is that endogenous pep5 alters cell-specific signaling pathways without causing cell death.
pep5-cpp induced cell death in both C6 rat glioma cells and in the rat glioblastoma in vivo. Glioblastoma is the most malignant astrocytoma in humans. Several drugs have been developed to treat gliomas, such as inhibitors of key oncogenic signaling pathways, apoptosis-inducing drugs, and DNA-damaging drugs (64 -67). The group of animals treated with pep5cpp showed a decrease of ϳ50% of the tumor volume when compared with the control group. It was interesting to note that pep5-cpp induced an increase in macrophage-like cells in the tumor area, probably due to induction of cell death. These data suggest that pep5-cpp also could be a new pharmacological tool in glioblastoma treatment.
pep5 was shown to inhibit the chymotrypsin activity of the proteasome, suggesting the possibility of an inhibitory feedback mechanism over the proteasome during cyclin D2 degradation. Cell cycle, gene expression, and apoptosis are events regulated by the cleavage of several proteins by the ubiquitin-proteasome system, including transcription factors, cyclins, and cyclin-dependent kinase inhibitors (68,69). Several transcription factors, such as NF-B, are involved in the control of the immune response, cell proliferation, and programmed cell death. In the development of malignancy, these proteins can alter the regulation of other factors, contributing to tumorigenesis (70 -72). Cyclins are degraded by the proteasome in order to permit the exit of cells from mitosis and their entry into another round of the cell cycle (73). Five main classes of mammalian cyclins have been described. Cyclins C, D (1, 2, and 3), and E act during the G 1 phase and regulate the transition from G 1 to S phase. In contrast, cyclins A and B (1 and 2) have activity during the S and G 2 phases and are regulators of entry into mitosis (74 -76). The expression levels of cyclin-dependent kinase inhibitors, including p21, p27, and p57, also are degraded by the ubiquitin-proteasome system, and this control is an important event for cell regulation (77). In tumor cells, there is uncontrolled cell growth and nearly constant proliferation and division, which requires extensive protein degradation. For this reason, proteasome inhibitors have been used as anti-tumor agents, helping to regulate the uncontrolled cell growth and inducing apoptosis in several tumor cells (78 -81).
The sequence of pep5 is present only in cyclin D2 and is not found in the other cyclins (Fig. 15). There is total homology of the pep5 sequence within cyclin D2 from a wide range of organisms, including humans, rats, bovines, mice, and chickens, suggesting its biological significance. Cyclin D2 must be degraded by the proteasome during G 1 /S transition to allow cell cycle progression (82), which suggests that pep5 can be endogenously generated by the proteasome. The proteasome has been shown to regulate the levels of most intracellular peptides (14,15), and pep5 was seen by bioinformatics analysis to be compatible with generation by the proteasome, followed by aminopeptidase activity (data not shown). Therefore, it is possible that proteasome inhibitors could alter levels of pep5, which could contribute to cell death.
In the maintenance of mitochondrial protein stability, proteases degrade proteins into peptides that are exported from the mitochondrial matrix into the intermembrane space by ABC transporters and reach the cytosol by passive diffusion. Haynes et al. (83) demonstrated by genetic analysis of Caenorhabditis elegans that under stress of mitochondrial protein misfolding, signal peptides generated by the mitochondrial ATPdependent proteolytic complex can activate the cell genome, providing a mechanism of intracellular communication among mitochondria, cytosol, and nucleus. As a consequence, the organelle can react against the loss of thermodynamic stability and the propensity of proteins to aggregate, inducing the expression of a nuclear encoded protein gene termed ubiquitinlike ClpXP, which is activated by a homeobox containing the transcription factor bZIP. bZIP is believed to be activated by a pathway that involves different peptides produced by ATP-dependent Clp proteases. The combined expression of these proteins using intracellular signal peptides represents the cell response toward the damage produced by irreversible aggregates and protein misfolding affecting cell functions and survival (83). These findings corroborate previous evidence that endogenous intracellular peptides are probably functional and of biological significance (84).
In summary, these findings suggest that pep5 is a novel bioactive intracellular peptide. There is an exciting possibility that pep5 could serve as a novel prototype molecule in pharmacology and therapeutics.