Evidence that Systemic Gentamicin Suppresses ... - ATS Journals

Evidence that Systemic Gentamicin Suppresses Premature Stop Mutations in Patients with Cystic Fibrosis J. P. CLANCY, ZSUZSA BEBÖK, FADEL RUIZ, CHRIS KING, JULIE JONES, LYNN WALKER, HEATHER GREER, JEONG HONG, LISA WING, MAURIZIO MACALUSO, RAYMOND LYRENE, ERIC J. SORSCHER, and DAVID M. BEDWELL Departments of Pediatrics, Cell Biology, Medicine, Medical Genetics, Public Health, and Microbiology, University of Alabama at Birmingham; Gregory Fleming James Cystic Fibrosis Research Center, University of Alabama at Birmingham; Children’s Hospital, Birmingham, Alabama; and Department of Pediatrics, University of Mississippi, Jackson, Mississippi

Here we report the effects of gentamicin treatment on cystic fibrosis transmembrane regulator (CFTR) production and function in CF airway cells and patients with CF with premature stop mutations. Using immunocytochemical and functional [6-methoxy-N(3-sulfopropyl) quinolinium (SPQ)-based] techniques, ex vivo exposure of airway cells from stop mutation CF patients led to the identification of surface-localized CFTR in a dose-dependent fashion. Next, five patients with CF with stop mutations and five CF control subjects were treated with parenteral gentamicin for 1 wk, and underwent repeated in vivo measures of CFTR function (nasal potential difference [PD] measurements and sweat chloride [Cl] testing). During the treatment period, the number of nasal PD readings in the direction of Cl secretion was increased approximately 3-fold in the stop mutation patient group compared with controls (p 0.001), and four of five stop mutation patients with CF had at least one reading during gentamicin treatment with a Cl secretory response of more than 5 mV (hyperpolarized). A response of this magnitude was not seen in any of the CF control subjects (p 0.05). In an independent series of experiments designed to test the ability of repeat nasal PDs to detect wild-type CFTR function, evidence of Cl secretion was seen in 88% of control (non-CF) nasal PDs, and 71% were more than 5 mV hyperpolarized. Together, these results suggest that gentamicin treatment can suppress premature stop mutations in airway cells from patients with CF, and produce small increases in CFTR Cl conductance (as measured by the nasal PD) in vivo.

Cystic fibrosis (CF) is a common and lethal autosomal recessive genetic disease, caused by dysfunction of the cystic fibrosis transmembrane conductance regulator (CFTR) protein (15). Defective or absent CFTR function alters Na and Cl transport across multiple epithelia, which in turn contributes to both the morbidity and mortality of the disease (6–8). Since discovery of the cftr gene, more than 800 disease-causing alleles have been identified (6, 9). The genetics of CF are similar to that of other autosomal recessive diseases, in that premature stop mutations within the gene are relatively common (6). Mutations within this class cause defective CFTR biosynthesis, which is characterized by (1) reduced mRNA levels, and (2) production of little or no truncated CFTR protein (10, 11). Cumulatively, premature stop codons are found in 10% of

(Received in original form April 3, 2000 and in revised form October 12, 2000) Supported by Cystic Fibrosis Foundation grants Clancy96LO and Bedwel96P, and by NIH grants DK53090 and DK54781. Correspondence and requests for reprints should be addressed to J.P. Clancy, M.D., 1600 7th Avenue South, Suite 620ACC, Birmingham, AL 35233. E-mail: [email protected] Am J Respir Crit Care Med Vol 163. pp 1683–1692, 2001 Internet address: www.atsjournals.org

all patients with CF, and in much higher numbers of select CF populations. For example, nearly 85% of patients with CF of Ashkenazi Jewish descent possess at least one premature stop allele (6, 9). Identifying clinically useful methods to suppress premature stop mutations within the cftr gene would therefore be of benefit to a significant number of patients with CF worldwide. Our laboratory and others have reported that certain antibiotics in the aminoglycoside family are capable of suppressing clinically significant premature stop mutations within the cftr gene, with subsequent translational readthrough and production of full-length, functional CFTR protein (12, 13). Gentamicin appears to be the most effective agent in vitro with an established human clinical profile. One study reported that systemic gentamicin treatment suppressed a premature stop mutation in the dystrophin gene of the mdx mouse, producing functional dystrophin protein in vivo (14). A subsequent human study also found evidence of stop mutation suppression, as topical gentamicin led to improvements in CFTR-specific Cl secretion across the nasal mucosa of patients with CF predominantly homozygous for premature stop mutations (15). CF patients without stop mutations, however, were not studied. The present study was designed to investigate the effects of standard serum aminoglycoside concentrations on CFTR production and function in primary human tissues ex vivo and in vivo. We studied respiratory cells isolated from patients with CF, and performed a controlled trial of parenteral gentamicin administration to five CF patients heterozygous for premature stop mutations compared with five CF control subjects (no stop mutations). Our ex vivo results indicate that gentamicin exposure leads to dose-dependent CFTR production and function in primary CF nasal cells isolated specifically from patients with CF with stop mutations. Our in vivo results indicate that 1 wk of parenteral gentamicin led to an 3-fold increased incidence of nasal potential difference (PD) readings in the direction of Cl secretion in the stop mutation CF group compared with CF control subjects. Airway Cl secretion was also stronger in the stop mutation group, with four of five subjects exhibiting a level of Cl secretion not seen in any of the control subjects. In an independent series of experiments designed to test the sensitivity of nasal PD measurements to detect wild-type CFTR function, five normal, nonCF control subjects underwent repeat nasal PDs using a protocol resembling our trial (no gentamicin treatment). Of the normal group nasal PDs, 88% revealed evidence of Cl secretion (i.e., in the negative direction), and 71% were greater than or equal to 5 mV (hyperpolarized). Taken together, these results suggest that short-term parenteral gentamicin treatment improves CFTR function as measured by the Cl secretory component of the nasal PD in patients with CF with premature stop mutations.

1684

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE

METHODS This study was approved by the University of Alabama at Birmingham Institutional Review Board for studies involving human subjects. All genotyping was performed through Genzyme Corporation (Framingham, MA).

Primary Human Airway Cell Isolation Primary human nasal cells were isolated from surgical remnant tissue, and propagated as previously described (16, 17). Cells were initially grown in Biocoat medium (Becton Dickinson, Franklin Lakes, NJ) supplemented with gentamicin (100 g/ml) and ceftazidime (100 g/ml). After 24 h, cells were changed to medium supplemented with ceftazidime only. Airway cells were studied 7–10 d post-isolation.

Functional CFTR Assay in Human Airway Cells Halide efflux experiments from airway cells were performed as previously described (12, 13). Briefly, cells were grown on fibronectin-coated glass coverslips in serum-free airway epithelial cell growth medium (Biocoat) and loaded overnight at 37 C in the halide-quenched dye 6-methoxy-N-(3-sulfopropyl) quinolinium (SPQ, 10 mM) in the presence or absence of increasing concentrations of gentamicin (0, 10, and 100 g/ml) for 16 h. Cells were then mounted in a specially designed perfusion chamber for fluorescence measurements. Fluorescence of single cells was measured with a Zeiss (Thornwood, NY) inverted microscope, a PTI (Lawrenceville, NJ) imaging system, and Hamamatsu (Tokyo, Japan) camera. Excitation was at 340 nm, and emission was 410 nm. All functional studies were at 23 C. At the beginning of the experiments, cells were bathed in a quenching buffer (NaI), and after establishment of a stable baseline, switched to a halide-free (NO3) dequenching buffer at 200 s with 20 M forskolin, 200 M isobutylmethylxanthine (IBMX), and 50 M albuterol to elevate cAMP. Fluorescence was normalized to the baseline (quenched) value (average fluorescence from 100–200 s), with increases presented as percent increase in fluorescence (F) over basal. The buffers used in the SPQ assay were (1) NaI buffer: 130 mM NaI, 5 mM KNO3, 2.5 mM Ca(NO3)2, 2.5 mM Mg(NO3)2, 10 mM D-glucose, 10 mM N-(2-hydroxyethyl) piperazine-N-(2-ethanesulfonic) acid (HEPES, pH 7.30), and (2) NaNO3 buffer: identical to NaI buffer except that 130 mM NaNO3 replaced NaI.

VOL 163

2001

Digital Confocal Immunofluorescent Microscopy of Primary Human Airway Cells Primary nasal epithelial cells from patients with CF were obtained and grown on glass coverslips as described above. Cells were placed in serum-free airway epithelial cell growth medium (Biocoat) containing gentamicin at concentrations of 0, 10, and 100 g/ml as indicated for 16 h. After gentamicin exposure, cells were formaldehyde fixed (4% methanol-free formalin diluted in phosphate-buffered saline [PBS], pH 7.4) for 30 min and then washed in PBS. The cells were treated with preimmune swine serum (1:20 dilution) in PBS for 20 min to block nonspecific protein-binding sites. CFTR antigen was detected with the monoclonal mouse MATG-1031 antibody, which recognizes the first extracellular loop of CFTR. MATG-1031 has been shown to be a highly sensitive and specific probe for surface-localized CFTR protein (18, 19). The secondary antibody was a rhodamine-tagged rabbit anti-mouse IgG antibody. Nuclear detection was performed by Hoechst staining. Cells were studied on an Olympus (Norwood, MA) IX70 inverted reflective fluorescence light microscope at 623-nm excitation, using a UplanApo 100 or Uapo/340 40 objective. Digital confocal images were captured with a Photometrics Sensys digital camera and analyzed with IPLab Spectrum software supplemented with Power Microtome extension software (Scanalytics Inc., Fairfax, VA).

Gentamicin Treatment in Patients with CF Five CF subjects with one premature stop mutation and five CF control subjects without a premature stop mutation were admitted to the University of Alabama at Birmingham (UAB) General Clinical Research Center at the Children’s Hospital of Alabama for 7 d of intravenous gentamicin therapy. Patients were instructed to continue all other home medications, and chest physiotherapy was provided twice a day either by a respiratory therapist or by the chest vest method (patient preference). All patients were well at the time of admission. Gentamicin was initially administered at 2.5 mg/kg every 8 h, and dosing was adjusted to achieve peak serum levels between 8 and 10 g/ml and trough values 2 g/ml. Target peak and trough levels were achieved within 48 h of admission for all study subjects (i.e., before measures of CFTR function during gentamicin treatment were initiated). During the study period, in vivo CFTR function was tested before, during, and after gentamicin therapy, including sweat [Cl] measurements

Figure 1. Surface CFTR detection by digital confocal immunofluorescent microscopy in human nasal epithelial cells. CFTR is red, nucleus is blue. Primary nasal epithelial cells from patients with CF were obtained and grown on glass coverslips and treated with gentamicin as described in METHODS. Nasal cells from a G542X/

F508 patient (A, B, C) and a

F508/ F508 patient (E, F ) in increasing concentrations of gentamicin: (A) G542X/ F508 0 gm/ml gentamicin; (B) G542X/

F508 10 gm/ml gentamicin; (C) G542X/ F508 100 gm/ ml gentamicin; (D) G542X/ F508 no gentamicin, no primary antibody (control); (E) F508/ F508, no gentamicin; (F) F508/ F508, 100 gm/ml gentamicin.

1685

Clancy, Bebök, Ruiz, et al.: Effects of Gentamicin on Stop Mutations in CF Sweat [Cl] Testing

Sweat tests were performed through the Children’s Hospital of Alabama laboratory services, using the pilocarpine iontophoresis method. All samples were greater than 100 mg, and tests were performed in duplicate. The values reported are the average from the two daily samples.

Nasal Potential Difference Measurements

Figure 2. Halide permeability in human nasal cells. (A) Cells from a G542X/ F508 patient (open symbols) and a F508/ F508 patient (closed symbols) were isolated, grown, and loaded with SPQ as described in METHODS. Cells were cultured in the presence of no gentamicin (diamonds), 10 gm/ml gentamicin (circles), or 100 gm/ml gentamicin (triangles) for 16 h previous to study. Cells were switched from a quenching NaI buffer to a dequenching NO3 buffer at 200s containing a cAMP elevating cocktail (20 M forskolin, 200 M IBMX, and 50 M albuterol). Positive (responding) cells were defined by a rate of dequench 2 RFU/min. A positive response was obtained for 5/44 G542X/ F508 cells treated with 10 gm/ml gentamicin and 42 of 44 cells treated with 100 gm/ml gentamicin, and the means of these responding cells are plotted ( S.E.M.). No responding cells were identified in the G542X/ F508 cells without gentamicin treatment (n 40), or any of the F508/ F508 cells (n 80 cells tested in each condition). The mean ( S.E.M.) of these nonresponding conditions are plotted to show background. (B) The percent of positive cells in each condition are shown for the F508/ F508 (left) and G542X/

F508 cells (right) in the increasing concentrations of gentamicin.

and nasal PDs (see below) on Days 0, 3, 4, 5, 6, and 7. Spirograms and sputum cultures were performed on Days 0 and 7. Follow-up studies, including nasal PDs, sweat [Cl] tests, and spirograms were performed 1–4 wk after completion of gentamicin treatment.

The nasal PD measurement is a well-established in vivo assay of CFTR function, and has been used extensively to differentiate between a CF and non-CF phenotype (20). For our studies, we used a nasal PD protocol previously validated in human subjects and standardized to the protocol developed by Knowles and colleagues (20, 21). The details of the nasal potential equipment used have been previously described (21). A series of stopcocks was configured to allow perfusion of the following solutions through the port at the tip of the catheter: lactated Ringer’s containing 104 M amiloride (Solution A); a low Cl solution (in mM, 2.4 K2HPO4; 0.4 KH2PO4; 115 sodium gluconate; 25 NaHCO3; 1.24 calcium gluconate2) containing 104 M amiloride (Solution B); and Solution B with added isoproterenol (105 M) (Solution C). All test solutions were perfused at a rate of 5.0 cm3/min. In each nostril, the largest PD readings in lactated Ringer’s at 1, 2, and 3 cm (lumen negative) were averaged and taken as the average baseline PD The catheter tip was then placed at the most negative PD site and maintained for superperfusion measurements with a disposable face shield (Splash Shield, Woburn, MA) adapted to hold the catheter for the duration of the protocol. The use of this face mask improved the quality of the readings compared with our previous experience using a hand-held catheter (data not shown). For each perfusion condition, a steady state recording was obtained. The recording lasted at least 1 min (for Solutions A and B) before proceeding to the next solution within the sequence. Because the 3-min time point following the change to low [Cl]/isoproterenol is taken as a highly sensitive and specific measure of wild-type CFTR activity, all patients and control subjects were monitored for at least 3 min during perfusion with Solution C. To ensure standardization of the assay, measurements in this study were performed by the same three investigators, using the same nasal PD apparatus. Several readings were obtained for data analysis, including (1) average baseline PD as described above, (2) change due to amiloride (the difference between the starting lactated Ringer’s PD and the stable PD after perfusion with Solution A), and (3) change due to low [Cl]/isoproterenol (difference between the lowest stable PD after addition of amiloride before perfusion with Solution C, and the PD after 3 min of perfusion with Solution C). Tracings in subjects and control subjects were not included in data analysis if catheter movement occurred. The final analysis included 80–90% of all tracings. To further validate the ability of repeated nasal PD measurements to detect CFTR-specific ion transport, a separate series of experiments was performed in which we made serial nasal PD measurements in five normal (non-CF) control subjects. The non-CF control subjects underwent a 6-d protocol designed to resemble our CF gentamicin trial, with repeat PDs performed on Days 0, 3, 4, 5, 6, and 7 (no gentamicin treatment). All nasal PD tracings (CF and normal) were analyzed and scored by an investigator blinded to genotype and gentamicin treatment. The

TABLE 1. STUDY PATIENT DEMOGRAPHICS* Premature Stop CF Subjects Age/Sex 18/M 18/M 15/F 28/M 13/F 18.4 yr (5.77)†

Genotype

P.a

G542X/ F508 G542X/ F508 W128X/ F508 G542X/ F508 R553X/621G-T

() () () () ()

Control CF Subjects FEV1 (% pred)

Age/Sex

Genotype

P.a

52 76 105 25 40 59.6 (31.5)†

23/F 18/F 30/M 26/F 28/F 25 yr (4.69)†

F508/ F508

F508/ F508

F508/G55ID

F508/ F508

F508/ F508

() () () () ()

* Age (yr, mean SD), sex (M male, F female), bacterial colonization (P.a Pseudomonas aeruginosa), FEV1 (forced expiratory volume in 1 s, %pred. † Mean ( SD).

FEV1 (% pred ) 38 29 19 53 34 34.6 (11.8)†

1686


VOL 163

2001

TABLE 2. STUDY PATIENT SPIROMETRY* Premature Stop CF Subjects

FVC, % pred FEV1, % pred

Control CF Subjects

Pretreatment

Gentamicin

Follow-Up

Pretreatment

Gentamicin

71.00 (29.61) 59.60 (31.50)

70.60 (26.52) 58.40 (27.40)

71.40 (24.05) 59.00 (25.33)

53.6 (8.02) 34.6 (11.76)

60.6 (7.09) 38.20 (12.19)

Follow-up 57.0 (12.90) 34 (12.06)

* Spirometry was performed in all subjects on Day 0 (pretreatment) and Day 7 (gentamicin) of intravenous gentamicin treatment. Follow-up studies were performed 1–4 wk after gentamicin treatment. Values represent means ( SD).

change following switch to low [Cl]/isoproterenol was analyzed by two blinded investigators, with the average reported. Of the 149 nasal PDs scored, the investigators matched on 125 (84%). The mean of the differing 24 scores was 1 mV ( 0.21 mV [SD]; range, 0.5–3 mV), and 22 of 24 scores differed by 1 mV.

Statistical Analysis Descriptive statistics (mean, standard deviation) and frequency distributions were employed in preliminary analysis of the data and to perform simple comparisons. Generalized linear models for repeated measurements were used to evaluate and compare treatment effects across the three study groups, taking into account the potential correlation among measurements taken in the same individual, and to control for random subject effects (22). Analysis of variance (ANOVA) models were employed to evaluate the effect of group (premature stop CF, control CF, normal control subjects) and day of treatment on the mean PD. To account for clustering effects within the same group (e.g., one patient disproportionately accounting for negative PD changes), the same fixed effects (group and treatment day) were evaluated in a mixed linear model with random subject effects. In some analyses, each PD measurement was reduced to a binary indicator of whether the PD showed a negative (hyperpolarizing) change following switch to low [Cl]/isoproterenol. Logistic regression models for repeated measurements were used to evaluate the effect of group and day of treatment on the average proportion of negative readings (22). In all models, the possibility of substantial differences in treatment effects between groups was evaluated systematically by testing the significance of the appropriate interaction terms. A secondary analysis included a two-tailed Fisher exact test, comparing the number of subjects in each group with Cl secretory readings reaching a threshold value ( 5 mV). A 0.05 level was used to determine significance. Statistical analysis of the Cl secretory results from each blinded investigator and from their combined scores yielded the same results.

and studied for evidence of halide transport, using the fluorescent dye SPQ. Cells were exposed to increasing concentrations of gentamicin (as described above) before study. Figure 2 shows that G542X nasal cells exposed to gentamicin at 10 and 100 g/ml had increased halide efflux in response to stimulation with a cAMP-elevating cocktail, compared with cells with no gentamicin exposure. In contrast, F508 F508 cells were unaffected by gentamicin exposure (10 and 100 g/ml). Similar (immunocytochemical and functional) results have been observed for airway cells derived from other patients with premature stop mutations (G542X/ F508 and R553X/ F508 ([23] and our unpublished observations). Together, these results provide evidence that gentamicin treatment leads to the production of functional CFTR in primary human airway cells specifically containing premature stop mutations. Gentamicin Treatment of CF Subjects with and without Premature Stop Mutations

To investigate the effects of gentamicin treatment on in vivo measures of CFTR activity, five patients with CF possessing premature stop mutations and five control subjects with CF (without stop mutations) underwent serial sweat [Cl] tests,

RESULTS Gentamicin Treatment of Primary Human Airway Cells

Premature stop mutations cause defective protein biosynthesis, leading to reduced CFTR protein production. To investigate the effects of gentamicin on CFTR biosynthesis in human tissues, nasal cells from a G542X/ F508 patient with CF were grown on glass coverslips and exposed to medium with gentamicin at 0, 10, and 100 g/ml for 16 h followed by immunohistochemical staining for surface-localized CFTR. For these experiments, CFTR detection was performed in cells without detergent permeabilization, using the MATG-1031 antibody, a sensitive probe for the first extracellular loop of CFTR (18, 19). Figures 1A–1C identify CFTR antigen at the cell membrane from G542X/ F508 nasal cells with increasing gentamicin concentrations. No surface CFTR was identified without gentamicin (Figure 1A). However, moderate (Figure 1B) and pronounced surface CFTR could be identified with gentamicin at 10 and 100 g/ml, respectively. In contrast, no surface CFTR was identified in nasal cells from a F508/ F508 patient treated with the highest dose of gentamicin (100 g/ml; Figure 1F). In a parallel set of experiments, nasal cells isolated from a G542X/ F508 patient with CF were grown on glass coverslips

Figure 3. Sweat Cl values in (A) control CF and (B) premature stop CF subjects. All subjects had sweat Cl determinations before (“Day 0”) and after initiation of gentamicin treatment. Follow-up measurements were made 1–4 wk following the treatment period. The closed circles are mean values. Bars indicate period of gentamicin treatment. Follow up data was not available for one subject in the control group (subject 5). No statistically significant differences were identified between the two treatment groups.

Clancy, Bebök, Ruiz, et al.: Effects of Gentamicin on Stop Mutations in CF

1687

Figure 4. Initial (“Baseline”) PD measurements in (A) CF control subjects, (B) premature stop CF subjects, and (C) normal (non-CF) controls. The baseline PD was determined as described (see METHODS). Each point is an individual measurement from one nostril. The closed circles are the mean values S.E.M. The bar indicates the period of gentamicin treatment, while the dashed line is 0 mV. No statistically significant differences were identified between the two CF treatment groups, while both CF groups were different than the normal controls (days 3-7, p 0.0001).

spirograms, sputum cultures, and nasal PD measurements before, during, and after intravenous gentamicin treatment. Patient demographics are described in Table 1. The average peak and trough gentamicin levels in the premature stop group were 9.78 g/ml ( 0.39 SD) and 0.98 g/ml ( 0.37 SD), and in the control group they were 8.98 g/ml ( 0.39 SD) and 0.88 g/ ml ( 0.39 SD) respectively. No significant adverse side effects were noted during the study period. Sputum cultures remained Pseudomonas positive in all subjects, and pulmonary functions did not change significantly during the gentamicin treatment period in either CF group (Table 2). Figures 3A–3B present the results of serial sweat [Cl] testing in both treatment groups. Normal (non-CF) sweat [Cl] values are 40 mM, whereas values diagnostic of CF are 60 mM. Sweat [Cl] values between 40 and 60 mM are indeterminate. All sweat [Cl] values in both groups were 75 mM, with no significant change in daily mean values in either group, or trends toward normalization for any individual within either group. Figures 4, 5, and 6 summarize the nasal PD results in the premature stop CF, control CF, and normal (non-CF) groups. The nasal PD is a bioelectric measure of ion transport (predominantly Na and Cl) across the nasal mucosa, and is an in vivo test frequently used to discriminate between a CF and non-CF phenotype. The normal phenotype includes a relatively low (less negative) baseline reading and a small depolarizing change after perfusion with the Na channel blocker

amiloride. These maneuvers measure Na absorption. Cl secretion is then measured by establishing a Cl secretory gradient (perfusion with low [Cl] solution), followed by stimulation with a -adrenergic agonist (isoproterenol). The CF phenotype of excessive Na absorption and poor Cl secretion includes a strongly negative baseline PD, a large depolarizing change after amiloride perfusion, and a depolarizing or minimally polarizing PD following perfusion with low [Cl] solution and -agonist (see Figure 7). Figures 4A–4C and 5A– 5C show the baseline nasal PDs and the change in PDs after perfusion with 100 M amiloride for these groups. No significant differences were noted between the control (Figs. 4A and 5A) and premature stop (Figures 4B and 5B) CF treatment groups, whereas both CF groups were significantly different from the normal (Figures 4C and 5C) control group for both parameters (p 0.0001). Figure 6A–6C compares Cl secretory responses (change in nasal PD [mV]) of the three groups after a change to low [Cl] and low [Cl] plus isoproterenol (change low [Cl]/iso). Because sodium resorption is blocked by amiloride during the low Cl perfusion, changes in the lumen negative direction are conventionally taken as evidence of Cl secretion by the airways. The range of Cl secretory responses measured for each patient and nostril over the study period tended to be greater in the premature stop CF and non-CF control groups compared with the CF controls (mean range of PD change [mV, SD], CF control subjects: left 5.85 [2.35], right 6.90

1688


VOL 163

2001

Figure 5. Change in PD following perfusion with lactated ringers plus 100 M amiloride in (A) CF control subjects, (B) premature stop CF subjects, and (C) normal controls. The amiloride-sensitive component of the nasal PD was determined as described (see METHODS). The data are shown as described in Figure 4. No statistically significant differences were identified between the two CF treatment groups, while both CF groups were different from the normal controls (days 3-7, p 0.0001).

[4.19]; premature stop CF: left 8.10 [3.97], right 9.5 [5.05]; normal control subjects: left 8.85 [3.47], right 12.25 [6.20]). In a logistic regression model for repeated measurements that explicitly accounted for correlation among measurements taken in the same nostril, the average proportion of negative readings during Days 3–7 was significantly larger among normal control subjects (regression-adjusted proportion 81%, 95% confidence interval [CI] of the proportion 66–90%) than among premature stop CF patients (regressionadjusted proportion 48%, CI 37–60%; p 0.0001). The latter (premature stop CF) was also significantly larger than the proportion of negative readings among CF control patients (regression adjusted proportion 7%, CI 2.4–19%; p 0.001). We also analyzed the number of subjects in the premature stop and control CF groups during gentamicin treatment who achieved at least one reading greater than 5-mV (hyperpolarizing). We chose this cutoff based on our prestudy nasal PD experience with CF subjects at our institution (no gentamicin treatment). In nearly 100 nasal PD readings performed on CF subjects with genotypes similar to that of our study groups over the 2 yr preceding this study, only 5% of these readings (5 of 98) reached a 5 mV threshold (of total of 17 CF subjects, 10 were F508 homozygotes, 7 were premature stop mutation mixed heterozygotes, no readings of 5 mV were measured in any of the stop mutation subjects). This

value has therefore been a useful standard for detecting Cl secretory PD changes that are rarely observed in patients with CF. Using this cutoff, we found that four of the five premature stop patients had at least one reading of 5 mV (hyperpolarizing) during the gentamicin treatment period, whereas none of the five CF control subjects had any reading

5 mV (p 0.05). In our normal non-CF group, 41 of 58 readings were 5 mV (71%), and all normal control subjects had at least one reading of 14 mV (hyperpolarizing). Examples of pre-gentamicin tracings (Day 0) and tracings demonstrating Cl secretion at or above 5 mV during gentamicin treatment for the four premature stop patients are shown in Figure 7.

DISCUSSION Many human genetic diseases, including CF, have a relatively high proportion of disease-causing premature stop mutations. Nonsense mutations are found in approximately 10% of patients with CF, and other genetic diseases (i.e., tuberous sclerosis, muscular dystrophies, polycystic kidney disease, phenylketonuria, hemophilias, mucopolysaccharidoses) are characterized by significant percentages of mutations in this class. CF, however, is one of the few genetic diseases (1) in which patients are routinely genotyped, and (2) with identified in vivo measures of


1689

Figure 6. Cl secretion as measured by nasal PD in (A) CF control subjects, (B) premature stop CF subjects, and (C) normal controls. Measurements were performed as described (see METHODS). The data are shown as described in Figures 4 and 5. The dashed line is 0 mV. The average proportion of negative (Cl secretory) readings in the stop mutation CF group (B) was increased above the control CF group (A) (days 3-7, p 0.001, see text), while the proportion of negative readings in the normal controls was greater than in the premature stop group (p 0.0001).

protein function that can be assayed readily and noninvasively. This study was designed to investigate the effects of traditional CF-related gentamicin dosing on measures of CFTR function in CF patients with premature stop mutations. Our ex vivo studies of primary human airway cells (Figures 1 and 2) suggest that relatively low-dose gentamicin exposure led to identification of functional surface CFTR specifically in cells derived from CF patients with premature stop mutations. Our in vivo results suggest that there is a partial restoration of CFTR function (Cl secretion as assayed by nasal PD) specifically in stop mutation CF patients during treatment with gentamicin. These findings therefore suggest that aminoglycoside-based pharmaceutical agents may be able to suppress premature stop codon mutations in vivo. Previous in vitro studies using the IB3-1 human CF bronchial airway cell line (genotype W1282X/ F508) demonstrated that a relatively short incubation (18–24 h) with the aminoglycoside G-418 at concentrations of 100–200 g/ml restored chromosomal W1282X mRNA levels to that of the control F508 allele, induced surface-localized CFTR protein production, and conferred on cells a new, cAMP-activated [Cl] current (13). Similar observations were made with transient expression systems for other CF-associated premature stop mutations, including G542X, R553X, and R1162X mutations, in addition to W1282X (12, 13). These studies helped to define and confirm the effectiveness of aminoglycoside-induced trans-

lational readthrough of cftr mutations, using relatively high levels of aminoglycosides. Follow-up in vitro studies using a reticulocyte lysate model system to assess the intracellular aminoglycoside levels required to suppress premature stop codons indicated that significant translational readthrough and full-length CFTR production could be produced by exposure to G-418 or gentamicin at 1 g/ml (M. Manuvakhova and D. M. Bedwell, personal communication). These findings, coupled with our observation that relatively short exposure of nasal cells derived from premature stop CF patients to gentamicin at 10 g/ml produced functional CFTR protein (Figures 1 and 2), suggest that parenteral gentamicin (with serum levels between 1 and 10 g/ml) may be capable of improving the activity of disease-causing premature stop mutations in vivo. In our human study, a traditional CF gentamicin dosing strategy (i.e., every 8 h to achieve peaks of 8–10 g/ml and troughs 2 g/ml), produced no significant changes in many manifestations of CF in either treatment group (premature stop or control subjects). Specifically, spirograms were unchanged, sputum cultures remained Pseudomonas positive, and sweat [Cl] values showed no consistent trends. End points such as pulmonary function or Pseudomonas colonization would likely not be expected to show differences between the treatment groups during such a short protocol, or in the face of significantly damaged airways. As in other human studies evaluating systemically dosed pharmaceutical agents designed to improve the function

1690


VOL 163

2001

Figure 7. Individual nasal PD tracings from premature stop mutation CF and normal CF control subjects. Nasal PDs were performed as described (see METHODS). Upward deflections are negative or more hyperpolarizing. Tracings 7A-7D are from premature stop mutation CF subjects, and 7E is a normal (non-LF) control. The left side tracing is day 0 (pre-gentamicin), while the right side tracing is during gentamicin treatment (day of gentamicin treatment noted, same nostril).

of mutant CFTRs, we also saw no changes in the sweat [Cl] values during the treatment period (24). Although gentamicin traverses the bronchial epithelium en route to the airway lumen, there are no studies reporting the efficiency of parenteral gentamicin delivery to sweat glands or into sweat itself. Furthermore, there is not a direct correlation between sweat [Cl] concentration and severity of CF disease. Reports indicate that some CF patients with uncommon alleles (i.e., conduction mutants such as R117H or R334W, or the uncommon A455E allele) may have abnormally elevated sweat [Cl] values diagnostic of CF but lower than those of the general CF population (9, 25–27). Despite this, most patients with CF with relatively mild lung disease have abnormal sweat [Cl] values that are not different from the general CF population. Finally, a pathologic study of human tissues suggested that CFTR membrane localization may be particularly reduced in the human sweat gland (28). Thus, whether the sweat [Cl] test is sensitive enough to

detect small changes in CFTR activity that predict a clinical benefit is not clear. Although we obtained and analyzed all the data in a careful and blinded fashion, we acknowledge that the small number of subjects in the study could introduce bias. With this caution in mind, there were several interesting observations of the stop mutation treatment group and also of our normal control subjects, specifically regarding Cl secretion as measured by the nasal PD. First, the premature stop group more commonly demonstrated hyperpolarization than the CF control group. These changes were found throughout the premature stop group, as four of five stop mutation study subjects had hyperpolarization that was of a magnitude not observed in any of the CF control subjects (Figure 7). The magnitude of effect that we observed was smaller than that reported in the recent topical nasal gentamicin trial (15). Whether this represents lower cellular gentamicin concentrations in our trial, stronger


effects in the homozygous premature stop patients studied in the topical trial, or a combination of these factors, is unknown. Second, the response observed in these tracings was unusual, generally showing a stable preisoproterenol baseline during Solution B perfusion (low [Cl] plus amiloride), followed by a gradually increasing hyperpolarization during Solution C perfusion (Solution B plus isoproterenol). Prolonged activation of Cl secretion of this type is taken as strong evidence for CFTR-specific Cl secretion. The pattern in these tracings was different than responses seen in typical, non-CF tracings, in which some hyperpolarization is usually seen immediately after the switch to Solution B (see Figure 7). Whether this difference might reflect low level CFTR function in the gentamicintreated stop mutation group (i.e., so that detection requires maximal activation with isoproterenol in addition to a Cl secretory gradient) is uncertain. Third, although every normal subject had one or more Cl secretory readings during the repeat PD protocol of at least 14 mV (more hyperpolarizing), the range of responses was broad (1 to 26 mV). Twentynine percent of the Cl secretory recordings were 5 mV (less hyperpolarizing) in the normal control group. Furthermore, there were substantial day-to-day differences within each normal subject, with Cl secretory differences as great as 20 mV for the same subject and nostril noted on consecutive days. The Cl secretory response during repeated nasal potential difference measurements has not been thoroughly studied in the past, possibly because repeated frequent PD measurements in the same patient have not generally been analyzed or reported. Some of these differences and the slightly reduced mean Cl secretory response in our normal group (daily mean values from 7.2 to 10.2 mV) may have been due to the short ( 8-mm) distance between the perfusing and recording ports of our double-lumen catheter (20, 26, 29–31). In contrast, the baseline and amiloride-sensitive components of the PD in the normal group (Figures 4C and 5C) were quite consistent and reproducible. This suggests that the values shown in Figure 6C might represent an inherent property of the normal nasal mucosa, and might be an intrinsic aspect of performing the nasal PD. Although we studied a small comparative group, this aspect of the Cl secretory response should be considered when designing human studies that use the nasal PD as a functional end point. Our results show that a single measurement at one time point may not be adequate to detect CFTR function, even in normal individuals. As has been reported in other human studies to improve CFTR function or to transfer cftr to CF-affected tissues, Na transport as measured by the nasal PD was unaffected by gentamicin treatment (Figures 4 and 5). This may reflect the higher sensitivity of measurable Cl secretion in the context of low-level CFTR activity, compared with attenuation of Na hyperabsorption (15, 24, 30, 32–37). In summary, the studies presented here provide evidence that gentamicin is able to improve CFTR production and function in cells from CF patients with premature stop mutations. Ex vivo exposure of airway cells to increasing doses of gentamicin led to increasing surface CFTR localization and function, specifically in cells derived from stop mutation subjects. In our clinical study, evidence of Cl secretion as measured by the nasal PD was found more frequently in CF patients with premature stop mutations compared with CF control subjects during gentamicin treatment. Our results suggest that gentamicin has mild, beneficial effects on CFTR function unique to this subset of patients with CF, and that this strategy may be a viable approach to treat certain genetic defects. Future studies should be directed toward identification of more clinically efficacious and less toxic agents, and/or

1691

alternative drug-dosing regimens that maximize premature stop codon suppression and minimize long-term toxicity. Acknowledgment : The authors thank Sheila Ellison and Jerry Barron for help in preparing this manuscript, and Edward Walthall for technical support. J. P. Clancy is a Leroy Matthews Award recipient.

References 1. Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R, Grzelczak Z, Zielenski J, Lok S, Plavsic N, Chou JL. Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA (published erratum appears in Science 1989;245:1437). Science 1989; 245:1066–1073. 2. Anderson MP, Gregory RJ, Thompson S, Souza DW, Paul S, Mulligan RC, Smith AE, Welsh MJ. Demonstration that CFTR is a chloride channel by alteration of its anion selectivity. Science 1991;253:202–205. 3. Bear CE, Li CH, Kartner N, Bridges RJ, Jensen TJ, Ramjeesingh M, Riordan JR. Purification and functional reconstitution of the cystic fibrosis transmembrane conductance regulator (CFTR). Cell 1992;68:809–818. 4. Cheng SH, Rich DP, Marshall J, Gregory RJ, Welsh MJ, Smith AE. Phosphorylation of the R domain by cAMP-dependent protein kinase regulates the CFTR chloride channel. Cell 1991;66:1027–1036. 5. Pilewski JM, Frizzell RA. Role of CFTR in airway disease. Physiol Rev 1999;79(Suppl 1):S215–S255. 6. Welsh MJ, Tsui L, Boat T, Beauden A. Cystic fibrosis. In: Scriver C, Beaudet A, Valle D, editors. Metabolic and molecular basis of inherited disease. 7th edition. New York: McGraw-Hill; 1995. 7. Schwiebert EM, Egan ME, Hwang TH, Fulmer SB, Allen SS, Cutting GR, Guggino WB. CFTR regulates outwardly rectifying chloride channels through an autocrine mechanism involving ATP. Cell 1995;81:1063–1073. 8. Stutts MJ, Canessa CM, Olsen JC, Hamrick M, Cohn JA, Rossier BC, Boucher RC. CFTR as a cAMP-dependent regulator of sodium channels [see comments]. Science 1995;269:847–850. 9. Kerem E, Kerem B. Genotype-phenotype correlations in cystic fibrosis. Pediatr Pulmonol 1996;22:387–395. 10. Hamosh A, Trapnell BC, Zeitlin PL, Montrose-Rafizadeh C, Rosenstein BJ, Crystal RG, Cutting GR. Severe deficiency of cystic fibrosis transmembrane conductance regulator messenger RNA carrying nonsense mutations R553X and W1316X in respiratory epithelial cells of patients with cystic fibrosis. J Clin Invest 1991;88:1880–1885. 11. Hamosh A, Rosenstein BJ, Cutting GR. CFTR nonsense mutations G542X and W1282X associated with severe reduction of CFTR mRNA in nasal epithelial cells. Hum Mol Genet 1992;1:542–544. 12. Howard M, Frizzell RA, Bedwell DM. Aminoglycoside antibiotics restore CFTR function by overcoming premature stop mutations [see comments]. Nat Med 1996;2:467–469. 13. Bedwell DM, Kaenjak A, Benos DJ, Bebok Z, Bubien JK, Hong J, Tousson A, Clancy JP, Sorscher EJ. Suppression of a CFTR premature stop mutation in a bronchial epithelial cell line. Nat Med 1997;3:1280–1284. 14. Barton-Davis ER, Cordier L, Shoturma DI, Leland SE, Sweeney HL. Aminoglycoside antibiotics restore dystrophin function to skeletal muscles of mdx mice [see comments]. J Clin Invest 1999;104:375–381. 15. Wilschanski M, Famini C, Blau C, Rivlin J, Augarten A, Avital A, Kerem B, Kerem E. A pilot study of the effect of gentamicin on nasal potential difference measurements in cystic fibrosis patients carrying premature stop mutations. Am J Respir Crit Care Med 2000;161:860–865. 16. Wu R, Yankaskas J, Cheng E, Knowles MR, Boucher R. Growth and differentiation of human nasal epithelial cells in culture: serum-free, hormone-supplemented medium and proteoglycan synthesis. Am Rev Respir Dis 1985;132:311–320. 17. Gruenert DC, Basbaum CB, Widdicombe JH. Long-term culture of normal and cystic fibrosis epithelial cells grown under serum-free conditions. In Vitro Cell Dev Biol 1990;26:411–418. 18. Demolombe S, Baro I, Bebok Z, Clancy JP, Sorscher EJ, Thomas-Soumarmon A, Pavirani A, Escande D. A method for the rapid detection of recombinant CFTR during gene therapy in cystic fibrosis. Gene Ther 1996;3:685–694. 19. Clancy JP, Hong JS, Bebok Z, King SA, Demolombe S, Bedwell DM, Sorscher EJ. Cystic fibrosis transmembrane conductance regulator (CFTR) nucleotide-binding domain 1 (NBD-1) and CFTR truncated within NBD-1 target to the epithelial plasma membrane and increase anion permeability. Biochemistry 1998;37:15222–152230. 20. Knowles MR, Paradiso AM, Boucher RC. In vivo nasal potential difference: techniques and protocols for assessing efficacy of gene transfer in cystic fibrosis. Hum Gene Ther 1995;6:445–455. 21. Walker LC, Venglarik CJ, Aubin G, Weatherly MR, McCarty NA,

1692

22. 23. 24.

25.

26.

27.

28.

29.

30.


Lesnick B, Ruiz F, Clancy JP, Sorscher EJ. Relationship between airway ion transport and a mild pulmonary disease mutation in CFTR. Am J Respir Crit Care Med 1997;155:1684–1689. Diggle PJ, Liang K-Y, Zeger SL. Analysis of longitudinal data. Oxford: Oxford University Press; 1994. As presented at the 1998 North American Cystic Fibrosis Conference Plenary Session, Montreal, Québec, Canada; and unpublished observations. Rubenstein RC, Zeitlin PL. A pilot clinical trial of oral sodium 4-phenylbutyrate (Buphenyl) in deltaF508-homozygous cystic fibrosis patients: partial restoration of nasal epithelial CFTR function. Am J Respir Crit Care Med 1998;157:484–490. The Cystic Fibrosis Genotype–Phenotype Consortium. Correlation between genotype and phenotype in patients with cystic fibrosis. N Engl J Med 1993;329:1308–1313. Wilschanski M, Zielenski J, Markiewicz D, Tsui LC, Corey M, Levison H, Durie PR. Correlation of sweat chloride concentration with classes of the cystic fibrosis transmembrane conductance regulator gene mutations. J Pediatr 1995;127:705–710. Gan KH, Veeze HJ, van den Ouweland AM, Halley DJ, Scheffer H, van der Hout A, Overbeek SE, de Jongste JC, Bakker W, Heijerman HG. A cystic fibrosis mutation associated with mild lung disease. N Engl J Med 1995;333:95–99. Kalin N, Claass A, Sommer M, Puchelle E, Tummler B. DeltaF508 CFTR protein expression in tissues from patients with cystic fibrosis [see comments]. J Clin Invest 1999;103:1379–1389. Smith SN, Middleton PG, Chadwick S, Jaffe A, Bush KA, Rolleston S, Farley R, Delaney SJ, Wainwright B, Geddes DM, Alton EW. The in vivo effects of milrinone on the airways of cystic fibrosis mice and human subjects. Am J Respir Cell Mol Biol 1999;20:129–134. Caplen NJ, Alton EW, Middleton PG, Dorin JR, Stevenson BJ, Gao X, Durham SR, Jeffery PK, Hodson ME, Coutelle C. Liposome-mediated CFTR gene transfer to the nasal epithelium of patients with cystic fibrosis [see comments] (published erratum appears in Nat Med 1995;1:272). Nat Med 1995;1:39–46.

VOL 163

2001

31. Middleton PG, Caplen NJ, Gao X, Huang L, Gaya H, Geddes DM, Alton EW. Nasal application of the cationic lipsome DC-Chol:DOPE does not alter ion transport, lung fuction or bacterial growth. Eur Respir J 1994;7:442–445. 32. Zabner J, Couture LA, Gregory RJ, Graham SM, Smith AE, Welsh MJ. Adenovirus-mediated gene transfer transiently corrects the chloride transport defect in nasal epithelia of patients with cystic fibrosis. Cell 1993;75:207–216. 33. Alton EW, Stern M, Farley R, Jaffe A, Chadwick SL, Phillips J, Davies J, Smith SN, Browning J, Davies MG, Hodson ME, Durham SR, Li D, Jeffery PK, Scallan M, Balfour R, Eastman SJ, Cheng SH, Smith AE, Meeker D, Geddes DM. Cationic lipid-mediated CFTR gene transfer to the lungs and nose of patients with cystic fibrosis: a double-blind placebo-controlled trial. Lancet 1999;353:947–954. 34. Gill DR, Southern KW, Mofford KA, Seddon T, Huang L, Sorgi F, Thomson A, MacVinish LJ, Ratcliff R, Bilton D, Lane DJ, Littlewood JM, Webb AK, Middleton PG, Colledge WH, Cuthbert AW, Evans MJ, Higgins CF, Hyde SC. A placebo-controlled study of liposome-mediated gene transfer to the nasal epithelium of patients with cystic fibrosis. Gene Ther 1997;4:199–209. 35. Zabner J, Cheng SH, Meeker D, Launspach J, Balfour R, Perricone MA, Morris JE, Marshall J, Fasbender A, Smith AE, Welsh MJ. Comparison of DNA–lipid complexes and DNA alone for gene transfer to cystic fibrosis airway epithelia in vivo. J Clin Invest 1997;100:1529– 1537. 36. Johnson LG, Olsen JC, Sarkadi B, Moore KL, Swanstrom R, Boucher RC. Efficiency of gene transfer for restoration of normal airway epithelial function in cystic fibrosis. Nat Genet 1992;2:21–25. 37. Johnson LG, Pickles RJ, Boyles SE, Morris JC, Ye H, Zhou Z, Olsen JC, Boucher RC. In vitro assessment of variables affecting the efficiency and efficacy of adenovirus-mediated gene transfer to cystic fibrosis airway epithelia. Hum Gene Ther 1996;7:51–59.