29
International Ophthahnology 19: 29-34, 1995. @ 1995 Kluwer Academic Publishers. Printed in the Netherlands.
Shiva-l: in vitro and in vivo tests of the effects of a novel, synthetic, lytic peptide on ocular cells * M o h s e n S h a h s a v a r i , G h o l a m A. P e y m a n * , M i c h a e l R. N i e s m a n , M i c h a e l V. M i c e l i & Jesse J a y n e s l
LSU Eye Center, Louisiana State University Medical Center School of Medicine, New Orleans, LA 70112, USA," I Department of Biochemistry, Louisiana State University, Baton Rouge, LA, USA (* Corresponding author) Accepted 16 January 1995
Key words." in vitro study, retinal toxicity, Shiva-1
Abstract
An investigation was undertaken to determine the toxicity of an intravitreal injection of a novel peptide drug, Shiva-1, in rabbits. The drug, a synthetic peptide modeled after lytic peptides secreted by certain insects, has antiproliferative and antibacterial properties. Initial in vitro experiments showed that the drug, at a concentration of 100 ~M, was toxic to both Y-79 retinoblastoma cells and human retinal pigment epithelial cells. A wide range of doses (6-1200/~g) was injected into the rabbit vitreous in an attempt to determine the maximum tolerated dose. Retinal toxicity was evaluated clinically, by electroretinography, and by light microscopy. Some localized toxicity was evident at 200/~g; all doses of 240 #g and above were toxic. While the drug appears to exhibit a narrow range between effective and toxic doses, the results suggest that this and other peptides of similar design merit further investigation for the treatment of proliferative and infectious diseases of the eye.
Introduction
The intraocular injection of antibacterial, antifungal, and antiproliferative agents has become a viable option for the treatment of severe cases of certain ocular diseases [1]. Intraocular injection of antibacterial agents and antifungal agents are given for the treatment of endophthalmitis [1]. A broad-spectrum agent would be quite useful in the treatment of endophthalmitis, especially when the causative agent is unknown. Proliferative vitreoretinopathy (PVR) is the leading cause of failure of retinal detachment surgery; primary PVR is very difficult to treat [2]. Intraocular injection of antimetabolites is one approach for the treatment of PVR [1, 3]. These agents have the potential to halt cell proliferation and membrane formation. However, many of the agents potentially and currently employed for intraocular injection are toxic to the sensitive structures of the eye. Furthermore, rapid clearance of * Supported in paint by U.S. Public Health Service grants EY07541 and EY02377 fi'om the National Eye Institute, the National Institutes of Health Services, Bethesda, MD, USA
drug from the vitreous limits the usefulness of many agents. Shiva-1 is a lytic peptide (M.W. 4220) that has been shown to inhibit the replication of Plasmodium falciparum and Trypanosoma cruzi [4] and has other antibacterial, antifungal, and antiproliferative properties. The drug is a synthetic peptide originally designed [4] to mimic the charge distribution, amphipathic, and hydrophobic properties of Cecropin B, a peptide secreted by Hyalophora cecropia in response to bacterial insult [5]. It is thought that the peptide acts by causing the formation of membrane defects or 'pores' as is the case with certain Cecropins [6], but at present the exact mechanism of action of Shiva-1 is unknown. Initial experiments have shown that the antibacterial and cytocidal effects of the drug are rapid in onset and complete after as little as one hour of exposure to the drug [7, 8] (J. Jaynes, M. Niesman; unpublished observations). The rapid onset of action of the drug suggests it could be useful as an antibacterial or antiproliferative agent, even if the clearance from the vitreous was rapid. Also, preliminary studies of the antiproliferative
30 effects against certain strains of bacteria indicated that the drug and related peptides may have some future role in the treatment of ocular disease [9]. Reference 4 contains the peptide sequence and a further discussion of its properties. This study was undertaken to evaluate retinal toxicity of Shiva-1 after intravitreal injection. Dose-related toxicity was seen, with the highest doses exhibiting severe toxicity. The higher doses of drug appeared to elicit an inflammatory response as the peptide exited the eye through the trabecular meshwork. Lower doses were not toxic. Though high concentrations of the drug were toxic, the unique nature of the drug suggests that further investigation into the mechanism of action and potential ocular uses of this class of drug may be warranted.
Materials and methods In vitro experiments
Human retinal pigment epithelial (RPE) cells were isolated from donor eyes and grown in Coon's modified Ham's F-12 media (Irvine Scientific, Irvine, CA) as previously described [10]. Cells were i'solated from donor eyes with no known history of ocular disease. The Y-79 retinoblastoma cell line was grown in Dulbecco's modified Eagle's media without L-glutamine (Gibco, Grand Island, NY) with 5% fetal calf serum (Hyclone, Logan, UT). Both media were supplemented with 2.0 mM glutamine, 0.25 mM ascorbate, 100 U/mL penicillin, and 100 #g/mL streptomycin. Cells were free from mycoplasma contamination as determined by hybridization assay (Mycoplasma test kit, Gen-Probe, San Diego, CA). For cell growth experiments, RPE and Y-79 cells were plated in 24 well tissue culture plates. The cells were fed on the day prior to drug exposure. For RPE cells, the drug was dissolved in Coon's modified Ham's F-12 without serum and 1.0 mL was placed in each well of the drug-treated plates. For Y-79 cells, the drug was made at twice the final concentration and 1 mL of media containing the drug was added to 1 mL of media in the plates. Control wells for both cell types received media without serum or drug. After the three-hour exposure to the drug, the media containing the drug was removed, and complete media with 5% fetal calf serum was added to the well. Determination of cell growth was made by counting the cells in the hemocytometer (RPE) or by measuring cellular protein (Y-79). Protein
measurements were made using the Lowry method as modified by Peterson [11]. In vivo experiments Part I
Five New Zealand albino rabbits weighing 2-4 kg were given intravitreal injections of different amount of Shiva-1 diluted in a volume of 0.1 mL phosphate buffered saiine (PBS). The synthesis of the peptide has been previously described [4]. For the initial phase of the study, the following drug amounts were injected: 6/xg, 60 #g, 200/~g, 600 #g, and 1200 #g. This wide range of doses was chosen because we had no information regarding the toxicity of this class of peptides. The right eye of each rabbit was injected with the peptide solution and the left eye served as control and received an intravitreal injection of 0.2 mL of PBS. The injection technique involved anesthetizing the rabbits with intramuscular ketamine and xylazine (50 mg/kg and 5 mg/kg, respectively) and anterior chamber paracentesis of 0.2 mL of aqueous with a 30-gauge needle followed by trans pars plana injection of 0.1 mL of drug or PBS using a 30-gauge needle. The animals were examined one week after injection by indirect ophthalmoscopy through a fully dilated pupil. An electroretinogram (ERG) was performed before injection in all rabbits and was repeated three weeks after injection under the same conditions. The details of the ERG procedure have been previously reported [11]. Briefly, the ERG was recorded and filtered (130 Hz) with contact lens electrodes (jet) and stainless steel needle electrodes placed behind the ear and also under the forehead skin. The ERG was simultaneously performed from both eyes under scotopic conditions after 30 minutes of dark adaptation. Wave amplitudes and implicit times were gathered and analyzed by computer and the differences between values before and after injection were calculated [12]. Three weeks after injection, the ERG was repeated and the animals were sacrificed. The eyes were enucleated for histology. All 10 eyes were fixed in a 2% paraformaldehyde/3% glutaraldehyde solution. They were later dehydrated in a graded series of alcohol and chloroform and embedded in paraffin. Sections were cut at 5 mm and stained in hematoxylin and eosin. Part H
For the second phase of the study, 10 New Zealand albino rabbits were used for intravitreal injections. Results
31
Results 10
9
In vitro results
9 8 7
4-"
9
6 5
-$
o
9
9
2 ] ...... r~m~
0 0
1
2
3
4
5
6
7
8
9
Days After Plating
Fig,. ].
Growth curves for RPE cells that were untreated ( I ) or treated with 100/~M Shiva (o) for three hours on day 4 (arrow). The points plotted represent the mean of three wells counted on each day of the experiment.
500 400 300 200
o n
100
.,..,,_llJ" /
0
o
0
"",
m~
1
,
I
i
n
o
I
n
2
3
4
5
6
7
I, Days After Plating
Fig. 2. Growth curves for Y-79 retinoblastoma cells untreated ( ' ) or treated with 100 t~M Shiva- 1 (o) for three hours on day 3 (arrow). The values on the graph represent the mean protein concentration for 3 wells harvested on each day of the experiment.
from Part I showed injection of > 200 #g of Shiva- 1 to be toxic; therefore, nine rabbits were injected intravitreally with the following amounts of Shiva-l: 60/zg (4 eyes), 100 #g (4 eyes), 140/zg (2 eyes), 180 #g (4 eyes), and 240/~g (4 eyes). One rabbit was used as control and was injected intravitreally with PBS in both eyes. The method of injection, ERGs before and after injection, follow-up examination, enucleation; tissue fixation, and processing were exactly the same as used for the first part of the study.
Preliminary experiments (data not shown) suggested that 10 #M Shiva-1 had little effect on the proliferation of RPE or Y-79 cells. However, 100 #M Shiva-1 produced a nearly complete kill of both cell types (Figs 1 and 2). Both cell types were experiencing log phase growth when the drug was applied, and regrowth was not seen in the cell populations exposed to the drug at any time during the remainder of the experiment. The growth curves for control and drug-treated RPE are shown in Fig. 1 and the corresponding curves for Y-79 cells are shown in Fig. 2. In vivo toxicity
In the first set of the experiments, some anterior segment complications including conjunctival injection, photophobia, mild corneal edema and slight lens opacification, which were of short duration, were noted. Only the rabbit with the highest intravitreal concentration (1200/tg) showed definite macroscopic, posterior segment involvement in the form of scattered intraretihal hemorrhages. The vitreous remained clear in all rabbits after injection. In Part II of the experiment, the same minor anterior segment manifestations were seen, as well as some vitreous opacities, which were noted in four individual eyes injected with the following amounts of peptide: 60 #g (one eye), 180/zg (one eye), 240/.tg (two eyes). Indirect ophthalmoscopy was impossible in these eyes because of hazy media, but was unremarkable in other eyes with clear media. Histologic evaluation of slides from Part I of the study revealed evidence of toxicity at 1200 ~g, 600 #g, and 200 #g drug concentrations. Specimens from animals given 60/~g and 6 l~g were normal by histologic evaluation. Slides from control eyes were all unremarkable. Careful examination of all of the retinal specimens using light microscopy revealed doserelated toxicity. The changes, which are summarized in Table 1, merit further description. Specimens taken from the retina of the animals receiving the highest doses of the peptide show marked toxic changes. For example, in an eye that receive an injection of 1200 ,ug of Shiva-1, the retina was totally disorganized with no remaining photoreceptors visible. Similar changes were seen in the eye that received doses between 200/~g to 600 #g of peptide, although the changes were less severe in the eyes that received
32 Table 1. Histopathologicalfindings in the retina of Shiva and PBS injected rabbit eyes.
Drug concentration
Number of eyes injected
Number of eyes with histopathologic changes
Histopathological findings
Pro1 1 1200/~g 600 >g 200/%g
1 1 1
1 1 l
60 #g 6 ~g PBS (control)
1 l 5
cell infiltration + photoreceptordisruption same changes as 1200 ~zg some photoreceptordisruption, but less than 600, 1200/zg/dose normal normal normal
Part II 240/l.g
4
180 #g
4
140 ~g 100 ~g 60/~g PBS* (control)
2 4 4 2
cell infiltration localized cell infiltration localized cell infiltration localized cell infiltration localized cell infiltration + localized RD localized cell infiltration
* phosphate buffered saline.
200-240 #g. Figure 3A shows a control retina, while a section from an eye that received 240 #g is shown in Fig. 3B. In the eye in this dose range, the inner and outer nuclear layers were still discernable, but in certain localized areas, all the outer segments were absent (Fig. 3B). Other areas of the retina were normal and resembled the control eye pictured in Fig. 3A. On the other hand, the eyes that received injections of 140 #g or less of peptide show no peptide-related toxicity. One control eye and one eye that received an injection of 60/.tg of peptide did exhibit some inflammation at the injection site, but normal retinal architecture was present elsewhere throughout the retina. The eyes that received between 180 and 240 #g of peptide all showed regions of toxicity localized to a particular region of the eye, while other areas where normal. In the eyes in this dose range, the anterior chamber and retinal toxicity appeared to be related in an inversely proportional manner; that is, the more severe the reaction in the retina, the less reaction in the anterior chamber, and vice versa.
Discussion
Early work with Shiva- 1 indicated that is seemed to be more toxic to rapidly dividing cells than to quiescent,
confluent cells [8]. One of the goals of this work was to determine if the peptide was toxic to cells that were not proliferating, but maintained a high level of metabolic activity. Photoreceptor cells fall into this category and are often the first cells to show signs of retinal toxicity following the intravitreal injection of pharmaceuticals. In this work, damage to the photoreceptor cells was seen following an injection of 240 #g or greater. If one assumes a volume of 1.5 mL for the rabbit vitreous [ 13], then this corresponds to an initial concentration in the vitreous of 38/.tM, assuming an equal distribution throughout the vitreous. Thus, it appears that damage occurs in vitro at relatively moderate dose levels. A dose of 140 #g, which was found to be nontoxic, when diluted into rabbit vitreous of 1.5 mL, is equivalent to a concentration of 22/~M. Exposure of Trypanosom a cruzi to this concentration of peptide resulted in a greater than 60% reduction in the number of viable trypomastigotes [4]. The eyes that received between 180 #g and 240 #g of peptide all showed signs of toxicity localized to a particular region of the eye, while other areas were normal. The anterior chamber and retinal toxicity appeared to be related in an inversely proportional manner in the eyes in this dose range; that is, the more severe the reaction in the retina, the less reaction in the anterior chamber, and vice versa.
33
a
b
Fig. 3. Paraffin-embeddedsection from an eye that received a sham injection of vehicle (a) and 240 #g Shiva-I (b). The outer segments have been entirely lost in 3B and some disorganizationis apparent. Other area of the same retina had an appearance similar to the control.
A possible explanation for this behavior is that Shiva-1 is toxic in areas where it is in high concentration and the variability o f the toxicity results from differences in the injection technique. Peptides with this type o f structure are thought to be preferentially taken into membranes in response to a membrane potential [5, 6, 14]. If the drug is injected into the midvitreous, it may exit the eye through the trabecular meshwork before it affects the retina, and it becomes concentrated in the cells along the efferent pathway, in the ciliary body. However, if the injection site is close to the retina, the drug will diffuse into the retina and be concentrated locally by photoreceptor cells before it diffuses out of the vitreous cavity. This may also explain the observed toxicity in the retina and optic nerve at high doses. The ERG data confirmed the neural damage at the high doses, with flat ERGs being exhibited by the rabbits with severe optic nerve damage and/or retinal degeneration. The peptide drug Shiva- 1 was not toxic to the rabbit eye at doses of 140 ~g or less. Moderate toxicity was seen at doses o f up to 240 # g and moderate to severe toxicity seen at higher doses. More conclusive studies on the effectiveness of the drug against proliferating RPE cells will need to be completed before the potential o f this and related peptides against ocular disease can be fully assessed.
Acknowledgements The authors gratefully acknowledge the expert technical assistance provided by JoAnn Canale and Martha Juban. This work was supported by grants EY02377 and EY07541 from the National Eye Institute.
References 1. Peyman GA, Schulman JA. lntravitreal drug therapy. In lntravitreal Surgery: Principles and Practice. Norwalk, CT: Appleton-Century-Crofts1986: 407-55. 2. YoshinoY, Ideta H, NagasakiH, Uemura A. Comparative study of clinical factors predisposing patientsto proliferative vitreoretinopathy.Retina 1989; 9: 97-100. 3. Blumenkranz MS, Hartzer MK. The mechanism of action of drugs for the treatment of vitreoretinal scarring. In: Ryan SJ (ed) Retina, Vol 3: Surgical Retina. St Louis: CV Mosby 1989: 401-11. 4. Jaynes JM, Burton CA, Barr SB, Gale JW, Julian GR, White KL et al. In vitro cytocidal effect of novel lytic peptides on PlasmodiumJalciparum and Trypanosoma cruzi. FASEB J 1988; 2: 2878-83. 5. BomanHG, Hultmark D. Cell-free immunity in insects. Ann Rev Microbio11987;41: 103-26. 6. Christensen B, Fink J, Merrifield RB, Mauzerall D. Channelforming properties of cecropins and related model compounds incorporated into planar lipid membranes. Proc Natl Acad Sci USA 1988; 85: 5072-6. 7. ArrowoodMJ, Jaynes JM, Healey MC. In vitro activities of lytic peptides against the sporozoitesof Cryptosporidiumparvum. Antimicrob Agents Chemother 1991; 35: 224-7.
34 Jaynes JM, Julian GR, Jeffers GW, White KL, Enright FM. In vitro cytocidal effect of lytic peptides on several transformed mammalian cell lines. Pept Res 1989; 2: 157-60. 9. Gunshefski L, Mannis MJ, Cullor JS, Schwab IR, Jaynes J, Smith WL et al. In vitro antimicrobial activity of Shiva-I 1 against ocular pathogens. Cornea 1994; 13: 237-42. 10. Miceli MV, Newsome DA, Novak LC, Beuerman RW. Cytomegalovirus replication in cultured human retinal pigment epithelial cells. Curr Eye Res 1989; 8: 835-9. 11. Peterson GL. A simplification of the protein assay method of Lowry et al. which is more generally applicable. Anal Biochem 1977; 83: 346-56. 8.
12.
Shahsavari M, Peyman G, Niesman MR. Retinal toxicity and in vitro efficacy study of cilofungin (LY 121019). Ophthalmic Surg 1990; 21: 726-8. 13. Green H, Sawyer JL, Leopold 1H. Elaboration of bicarbonate ion in intraocular fluids. II. Vitreous humor, normal values. Arch Ophthalmo11957; 57: 85-9. 14. Miller MA, Garry RE Jaynes JM, Montelaro RC. A structural correlation between lentivirus transmembrane proteins and natural cytolytic peptides. AIDS Res Hum Retroviruses 1991 ; 7: 511-9.
Address.[or correspondence: Gholam A. Peyman, LSU Eye Center, 2020 Gravier St., Suite B, New Orleans, LA 70112, USA