AJP - Endo Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Endocrinol Metab 290: E750-E756, 2006. First published November 8, 2005; doi:10.1152/ajpendo.00381.2005
0193-1849/06 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
290/4/E750    most recent
00381.2005v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ritzel, R. A.
Right arrow Articles by Butler, P. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ritzel, R. A.
Right arrow Articles by Butler, P. C.

TRANSLATIONAL PHYSIOLOGY

The mass, but not the frequency, of insulin secretory bursts in isolated human islets is entrained by oscillatory glucose exposure

R. A. Ritzel,1,2 J. D. Veldhuis,3 and P. C. Butler1

1Larry Hillblom Islet Research Center, University of California, Los Angeles David Geffen School of Medicine, Los Angeles, California 2Division of Endocrinology and Diabetes, School of Medicine, University of Heidelberg, Heidelberg, Germany 3Endocrine Division, Mayo Medical and Graduate Schools of Medicine, Mayo Clinic, Rochester, Minnesota

Submitted 16 August 2005 ; accepted in final form 29 October 2005


   ABSTRACT
TOP
 ABSTRACT
RESEARCH DESIGN AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Insulin is secreted in discrete insulin secretory bursts. Regulation of insulin release is accomplished almost exclusively by modulation of insulin pulse mass, whereas the insulin pulse interval remains stable at ~4 min. It has been reported that in vivo insulin pulses can be entrained to a pulse interval of ~10 min by infused glucose oscillations. If oscillations in glucose concentration play an important role in the regulation of pulsatile insulin secretion, abnormal or absent glucose oscillations, which have been described in type 2 diabetes, might contribute to the defective insulin secretion. Using perifused human islets exposed to oscillatory vs. constant glucose, we questioned 1) whether the interval of insulin pulses released by human islets is entrained to infused glucose oscillations and 2) whether the exposure of islets to oscillating vs. constant glucose confers an increased signal for insulin secretion. We report that oscillatory glucose exposure does not entrain insulin pulse frequency, but it amplifies the mass of insulin secretory bursts that coincide with glucose oscillations (P < 0.001). Dose-response analyses showed that the mode of glucose drive does not influence total insulin secretion (P = not significant). The apparent entrainment of pulsatile insulin to infused glucose oscillations in nondiabetic humans in vivo might reflect the amplification of underlying insulin secretory bursts that are detected as entrained pulses at the peripheral sampling site, but without changes in the underlying pacemaker activity.

pulsatile insulin secretion; oscillatory glucose; diabetes mellitus; entrainment; islet of Langerhans

INSULIN IS SECRETED in discrete, high-frequency, insulin secretory bursts that constitute ~70% of total insulin secretion (28). The resulting oscillations in insulin concentration are strikingly apparent in the portal circulation (24, 34) but can also be detected in the systemic circulation (5, 14, 28). If highly sensitive and specific insulin assays and 1-min sampling are employed, the detected pulse interval in the systemic circulation is comparable with that in the portal circulation, i.e., 4–6 min (28, 30). The insulin pulse interval is remarkably robust, with regulation of overall insulin secretion being accomplished by modulation of the insulin secretory burst mass in response to stimulation of insulin secretion by glucose, sulfonylureas, or GLP-1 (10, 23, 26, 29, 30, 32).

Patients with type 2 diabetes have impaired glucose-mediated insulin secretion because they secrete smaller pulses (decreased insulin pulse mass and pulse amplitude) compared with nondiabetic humans (12, 13). Several lines of evidence suggest that this deficit in pulse mass is partly due to the loss of available insulin stores. First, pulse mass is restored in patients with type 2 diabetes by prior overnight inhibition of insulin secretion by somatostatin (12). Second, impaired glucose-mediated insulin secretion due to deficient insulin pulse mass can be reproduced in a porcine model with a loss of beta-cell mass comparable with that present in patients with type 2 diabetes (11, 15). Third, isolated human islets that are chronically exposed to glucose concentrations typically present in type 2 diabetes have both decreased insulin content and insulin pulse mass, both of which are avoided if the islets are cultured at the same glucose concentration, with a potassium channel opener to inhibit insulin secretion (31, 35). Although insulin pulse mass is decreased in humans with type 2 diabetes, pigs with a partial loss of beta-cell mass, and human islets cultured at high glucose concentrations, pulse frequency is unchanged in all of these circumstances (11–13, 15, 31, 35).

Given the remarkably robust nature of the insulin pulse interval, it was therefore intriguing that studies in nondiabetic humans (25) reported entrainment of the insulin pulse interval from 4 to 10 min by intermittent infusion (every 10 min) of small glucose pulses. This entrainment could not be reproduced in patients with type 2 diabetes (6, 17). Blood glucose concentrations oscillate in healthy humans, with a frequency similar to that of the insulin pulses (14, 17), but these glucose oscillations are impaired in patients with type 2 diabetes (13). Therefore, a change in the glucose concentration profile might contribute to impaired insulin secretion in type 2 diabetes. In the present experiments, we studied perifused isolated human islets at constant vs. oscillatory (10-min) glucose concentrations to address the following questions. 1) Is the frequency of insulin secretory bursts from isolated islets entrained to slower oscillations of glucose, as reported in humans in vivo? 2) Does exposure of islets to oscillating vs. constant glucose confer an increased signal for insulin secretion?


   RESEARCH DESIGN AND METHODS
TOP
 ABSTRACT
 RESEARCH DESIGN AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Design. Isolated human islets were studied in an islet perifusion system that has previously been validated for detection and quantification of pulsatile insulin release from human islets (33). After an equilibration period of 40 min without sample collection (–40 to 0 min), the islets were studied over a period of 100 min. Superimposed on a baseline glucose concentration of 4 mmol/l, glucose was delivered to the islets in an oscillatory fashion at an interval of 10 min (8 pulses per experiment) and different glucose amplitudes of 0 (control; n = 8 runs), 0.3 ± 0.0 (n = 8 runs), 3.9 ± 0.1 (n = 7 runs), 11.0 ± 0.3 (n = 8 runs), and 22.6 ± 0.9 mmol/l glucose (n = 5 runs). The glucose oscillations were generated by intermittent 2-min infusions of a buffer with higher glucose concentration, alternating with an 8-min period of perifusion at 4 mmol/l glucose. This design was intended to reproduce the previously reported (25) glucose pulse interval (10 min) to entrain insulin pulses in nondiabetic humans in vivo. We reproduced both physiological and supraphysiological glucose amplitudes to reproduce oscillations under conditions of daily living but also to avoid a false-negative outcome. We reasoned that if very large amplitude pulses do not entrain insulin pulses, this negative outcome is safe with smaller pulses present in clinical studies. We define entrainment of pulsatile insulin secretion as a change in pulse frequency so that insulin pulses arise as a consequence of the newly introduced signal (10-min glucose pulse), rather than conforming to the prior intrinsic signal pacemaker generating ~4 min pulses. Islets were recovered from the perifusion chambers for the measurement of islet insulin content. The mean islet insulin content was not significantly different in islets studied at any of the glucose amplitudes: 0, 0.3, 3.9, 11.0, and 22.6 mmol/l [2,440 ± 390, 2,640 ± 400, 3,080 ± 360, 2,120 ± 200, and 2,840 ± 480 fmol/islet, respectively; P = not significant (NS)]. Insulin secretion from the islets is reported as mass units per islet to account for different islet numbers. Islet sizes were measured at the beginning and the end of each perifusion experiment to account for different islet sizes.

The results of the present experiments with oscillatory glucose delivery were compared with islet perifusion experiments that had constant glucose stimulation on the basis of the mean glucose levels of 4, 4.2, 6.3, 9.5, and 14.7 mmol/l (oscillatory) vs. 4, 8, 12, 16, and 24 mmol/l (constant). The comparisons of dose-response relationships are depicted for a glucose range of 4–16 mmol/l. The data of islet perifusion experiments with constant glucose delivery has been published previously (32).

Islet culture. Human islets were isolated from the pancreas of four heart-beating organ donors by the University of Minnesota Diabetes Institute for Immunology and Transplantation (Bernhard J. Hering) and the Northwest Tissue Center, Seattle (R. Paul Robertson). The islets were maintained in RPMI culture medium, with 5 mM glucose and 10% FBS at 37°C in humidified air containing 5% CO2. The experiments were performed in random order after a minimum recovery period of 4 days after the islet isolation process.

Islet perifusion. The details of the islet isolation system and procedure have been described previously (32, 33). Briefly, aliquots of human islets (n = 10–30) were suspended in Bio-Gel P-2 beads (Bio-Rad, Hercules, CA) and placed in perifusion chambers. The islet perifusion system (ACUSYST-S; Cellex Biosciences, Minneapolis, MN) consisted of a multichannel peristaltic pump that delivered perifusate through maximally four parallel tubing sets via a heat exchanger and four perifusion chambers at a constant flow rate of 0.3 ml/min. The perifusion buffers (Krebs-Ringer bicarbonate buffer: NaCl 115 mmol/l, KCl 4.7 mmol/l, CaCl2 2.5 mmol/l, MgCl2 1.2 mmol/l, NaOH 5 mmol/l, pH 7.4) were supplemented with 0.2% human serum albumin, prewarmed to 37°C, and directly oxygenized with 95% O2 and 5% CO2. The effluent was collected in 1-min intervals.

Laboratory determinations. Glucose concentrations in the effluent were measured with the glucose oxidase method (Glucose Analyzer 2; Beckman Instruments, Brea, CA). Insulin was measured in duplicate with a two-site immunospecific ELISA, as previously described (1). There is no cross-reactivity with proinsulin and split 32,33 and 31,32 proinsulins. The lower detection limit is ~4 pmol/l, and the assay range is 4–2,000 pmol/l. The intra-assay coefficient of variation ranged from 1.9 to 3.4%; the interassay coefficient of variation ranged from 3.5 to 4.5%.

Calculations and statistical analysis. Insulin pulse detection and quantification were performed with the multiparameter technique deconvolution analysis (37). The procedure is statistically conditioned on independent exponential insulin kinetics validated in the perifusion system (33). Deconvolution calculates secretion properties represented by 1) the number, duration, mass (time integral of calculated burst), and amplitude (maximum calculated rate of insulin release during burst) of discrete bursts and 2) a time-invariant hormone release. Autocorrelation analysis was performed on insulin concentration series to calculate the correlation coefficients (n) between the time series and its copy at a lag of 0–25 min. The square of n is proportionate to the variance in the mean insulin concentration, which is explained by a periodic signal that recurs at the corresponding time delay (e.g., 2, 4, 8 min). Spectral analysis was performed to further analyze the periodicity of the data. It decomposes the insulin concentration time series into cyclical components with a few underlying sinusoidal (sine and cosine) functions of particular wavelengths. No detrending procedures were used for data analysis.

Data are presented as means ± SE. ANOVA and the unpaired Student’s t-test were used to contrast insulin concentrations and secretion measures. Nonlinear regression analysis {sigmoidal dose reponse, variable slope, y = bottom + [top – bottom]/[1 + 10 exp(log EC50 – x) x Hill slope]} was applied to analyze the relationship between glucose concentration and total insulin secretion, pulse interval, and pulse mass of pulsatile insulin secretion. The assumption of a P value of <0.05 was construed as significant.


   RESULTS
TOP
 ABSTRACT
 RESEARCH DESIGN AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Inspection of a representative data profile from a perifusion experiment with oscillatory glucose stimulation (Fig. 1) shows that insulin pulses were still evident between glucose pulses and appeared to be present with an interval of ~4–5 min, in contrast to the glucose oscillations of 10-min intervals. The size of the insulin pulses that coincided with glucose pulses were markedly larger than those that coincided with troughs in glucose concentration. Deconvolution analysis confirms that the frequency of insulin pulses did not adapt to the frequency of the oscillatory glucose stimulus (Fig. 1). In Fig. 2, additional insulin secretion profiles from islets perifused at oscillatory glucose vs. constant glucose are shown. When islets were perifused at a constant glucose concentration (Fig. 2A), pulsatile insulin release was apparent, as previously reported (31, 35). In perifusion experiments at a similar glucose concentration, but with even subtle (~0.3 mmol/l) glucose oscillations, a more marked pulsatile pattern of secretion was evident (Fig. 2B). However, this more marked pulsatile pattern of insulin secretion was not accompanied by entrainment of insulin pulses to the glucose oscillations, even in experiments with higher glucose amplitudes (Fig. 2, C and D). Deconvolution analysis (Fig. 2, E–H) revealed that the reason for the more distinct pulsatile pattern of insulin secretion with oscillating glucose was a twofold increase in the proportion of insulin secreted by the group of islets in each chamber in coordinate pulses (61.5 ± 3.9 vs. 30.6 ± 1.1%, P < 0.0001). This increased proportion of insulin released by groups of islets in coordinate pulses exposed to oscillatory glucose was also accompanied by a fourfold augmentation of insulin pulse mass (P < 0.001) with oscillatory glucose. However, the frequency of insulin secretory bursts was not different between the islets perifused at oscillating vs. constant glucose across the whole range of glucose concentrations studied (Fig. 3), which affirms the impression from individual insulin concentration profiles that the pacemaker driving insulin pulses was not influenced by oscillations of glucose at 10-min intervals. Cross-correlation analysis indicated that there was a positive relationship between insulin secretion and glucose concentration with zero lag time, affirming that the insulin secretion responded within <1 min to a rise in glucose concentration in the buffer.


Figure 1
View larger version (29K):
[in this window]
[in a new window]
  		Fig. 1. Insulin (solid line, left y-axis) and glucose (dashed line, right y-axis) concentration profiles (A) and insulin secretion rates (derived by deconvolution analysis; B) from 1 representative islet perifusion experiment with oscillatory glucose stimulation at 10-min intervals and amplitudes of ~0.3 mmol/l. *Insulin oscillations corresponding to a detected pulse.

 

Figure 2
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 2. Insulin (left y-axes, solid lines) and glucose (right y-axes, dashed lines) concentration profiles (A–D) and insulin secretion rates (derived by deconvolution analysis; E–H) from 4 representative islet perifusion experiments with constant (A and E) or oscillatory glucose stimulation at 10-min intervals and amplitudes of ~0.3 (B and F), ~3.9 (C and G), and ~11.8 mmol/l (D and H). Note the different y-scale for glucose in B.

 

Figure 3
View larger version (17K):
[in this window]
[in a new window]
  		Fig. 3. Interval of pulsatile insulin secretion from isolated human islets perifused with oscillatory [total n = 36 runs, P = not significant (NS)] vs. constant glucose (total n = 103 runs, P = NS). Because glucose oscillations were administered at 10-min intervals, entrainment of pulsatile insulin secretion to these oscillations would require pulse interval of 10 min. However, the observed insulin pulse interval remained unchanged at ~4–5 min. Data are means ± SE. Statistics by ANOVA.

 
A striking finding was the coexistence of large insulin pulses adjacent to small ones, which caused a marked pulse-to-pulse variance in both pulse mass and pulse amplitude in islets exposed to oscillating glucose. In particular, variability in interburst interval was <35%, whereas that of insulin pulse size was ~150%, and it was greater under intermittent rather than continuous glucose drive (P < 0.05). To examine whether this close proximity of large and small insulin pulses could still be resolved under circumstances that resemble in vivo experiments with sampling from the systemic circulation, we reexamined the raw insulin data from one representative experiment on the basis of the systemic insulin kinetics obtained in humans in vivo (28). In this simulation of systemically sampled data, the insulin concentration profile closely resembled the underlying glucose oscillations (Fig. 4). Therefore, in vivo when sampling from the systemic circulation, small insulin pulses might be missed if they were to occur between larger pulses.


Figure 4
View larger version (33K):
[in this window]
[in a new window]
  		Fig. 4. Dampening of insulin concentration profile as it occurs in systemic circulation in vivo. A: original insulin concentration profile (solid line) from 1 representative islet perifusion experiment with oscillatory glucose stimulation (10-min interval, amplitude ~0.3 mmol/l). B: calculated systemic concentration data derived by convolution of insulin secretion time series corresponding to A. Computations are based on insulin kinetics previously established in humans (see Ref. 28).

 
Autocorrelation analysis, which was used for analysis of insulin concentration time series in vivo (6, 25), showed a dominant periodicity of 10 min in experiments with oscillatory glucose delivery, although it did not detect regular insulin pulses in experiments with constant glucose perifusion (Fig. 5C). Also, the autocorrelograms of individual data sets and the corresponding modeled systemic insulin concentration (by reconvolution of insulin secretion as described above) together revealed oscillations in insulin concentration at the periodicity of the applied glucose oscillations (Fig. 5D). Spectral analysis has also been used for analysis of insulin time series in vivo (6, 25). It detects dominant periodicities at the frequency of infused glucose oscillations but does not distinguish underlying sinusoidal functions of particular wavelengths in experiments with constant glucose (Fig. 5A). In the present studies, when spectral analysis was applied to individual insulin concentration time series, an additional cyclical component with a period of ~5 min was uncovered. However, the ~5-min component was not recognized using the corresponding modeled systemic insulin concentration (Fig. 5B).


Figure 5
View larger version (27K):
[in this window]
[in a new window]
  		Fig. 5. Mean spectral density (A) and mean autocorrelograms (C) of insulin concentration profiles obtained during islet perifusion experiments with oscillatory (interval of 10 min) or constant glucose. Individual spectral analysis (B) and autocorrelograms (D) correspond to insulin concentration profiles shown in Fig. 4B.

 
The dose-response curves for total insulin secretion in response to oscillating or constant glucose were remarkably similar (Fig. 6). Analysis of the time course of insulin secretion revealed that over a period of 1 h, there was no time-dependent difference of insulin secretion in experiments with constant vs. oscillatory glucose stimulation (Fig. 7).


Figure 6
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 6. Glucose dose-response relationship of total insulin secretion from isolated perifused islets in perifusion experiments with oscillatory (total n = 36 runs) and constant glucose stimulation (total n = 103 runs). Solid and dashed lines indicate best-fit values derived by nonlinear regression analysis. Data are means ± SE. Statistics by ANOVA.

 

Figure 7
View larger version (21K):
[in this window]
[in a new window]
  		Fig. 7. Time course of mean insulin secretion from isolated human islets in perifusion experiments with oscillatory glucose stimulation (total n = 36 runs). Solid lines indicate means ± SE of mean insulin secretion in experiments with constant glucose stimulation. Oscillatory vs. constant glucose stimulation does not result in increased rates of insulin secretion. Statistics by ANOVA.

 

   DISCUSSION
TOP
 ABSTRACT
 RESEARCH DESIGN AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
In the present studies, we report that the pacemaker activity for insulin secretion in isolated human islets is insensitive to oscillatory glucose stimulation. Also, an oscillatory pattern of glucose concentrations does not provide a quantitative advantage with respect to total insulin secretion over stimulation with a comparable, but constant, glucose concentration. Therefore, we postulate that entrainment of pulsatile insulin to infused glucose oscillations in vivo likely reflects preferential detection of intermittently enhanced insulin pulses, which occasionally coincide with the momentary glucose elevation.

The concept of entrainment of insulin oscillations was originally introduced to describe the frequency shift of ultradian insulin oscillations (occurring at an ~100–120 min interval) by pulsed glucose infusions (19, 36). The present study is focusing on high-frequency insulin oscillations (~4–6 min interval) that arise as a consequence of discrete secretory bursts. Although the term entrainment was used in several studies of high-frequency insulin oscillations in response to administration of glucose pulses in vivo (17, 25) and in vitro (2), the precise meaning of entrainment has not been clearly defined. In the present analyses, entrainment denotes that the intermittent stimulus (i.e., glucose) would change the frequency of the underlying pacemaker to correspond consistently to that of the experimental stimulus. It has been suggested that entrainment of high-frequency insulin pulses is of potential use as a discriminator between insulin secretion in health and type 2 diabetes and has been examined in perifused rat islets (2) and humans in vivo (7, 17, 25). In rat islets, perifusion with a glucose pulse every 40 min increased insulin pulse periodicity from ~25 min to ~40 min (2). However, in that study, insulin concentrations were only evaluated at ~3 min intervals, a sampling interval that is inadequate to reliably detect high-frequency insulin oscillations that occur every 4–6 min (27). In technical terminology, insufficiently frequent sampling induces aliasing, an artifact that is due to intermittent false-negative estimates. In an in vivo study, glucose boluses were given to diabetic and nondiabetic subjects at an interval of ~29 min. Spectral analysis of smoothed insulin concentration data (3-point moving average) showed a shift of insulin pulse frequency from ~10 to ~15 min in nondiabetic humans (17), which was interpreted as entrainment, although the insulin pulse interval did not adapt to the ~29-min input periodicity that is required by the present definition of entrainment. On the other hand, entrainment of insulin oscillations to a train of 10-min glucose pulses was reported in nondiabetic volunteers (25). The present study illustrates that the latter outcome is consistent with the expected loss of discrimination in systemic blood of underlying high-frequency pulses during intermittent glucose pulsing (18). In the same study, the pulse interval detected with autocorrelation was ~10 min in both type 2 diabetes and controls with or without glucose pulses, but the autocorrelation coefficient was reduced in type 2 diabetes compared with controls during glucose pulses. This important observation could reflect the technical limitation of autocorrelation analysis to identify only those periodicities that approximate multiples of twice the sampling interval and recur with suitably high consistency. Interestingly, spectral analysis of insulin concentration profiles from islets perifused with oscillatory glucose was able to reveal the intrinsic ~4- to 5-min pulse interval. However, when the modeled insulin concentration profile anticipated in the systemic circulation was examined by spectral analysis, like autocorrelation, it failed to reliably recognize the ~4- to 5-min pulse interval next to the dominant ~10-min insulin oscillations, thus incorrectly implying entrainment.

The present data indicate that variability in insulin burst mass reflects occasional concordance between a stable islet pulse frequency and exogenously fixed glucose stimuli. Detection of discrete insulin bursts in the systemic circulation is often a consequence of overlap between small and large insulin pulses, the latter arising as a consequence of the coincident increase in glucose concentration. Stability of the frequency of pulsatile insulin secretion to time-invariant stimulation with various secretagogues has been inferred by others (8, 9, 28–30). The present study extends this notion by direct assessment of pulsatile insulin secretion by isolated human islets with oscillatory glucose. Although the mechanism for generation of insulin pulses is independent of prevailing glucose concentrations, in the present conditions, insulin-secretory burst mass rose severalfold whenever glucose maxima coincided with the expected time of insulin pulses. Cross-correlation analysis using concomitant 1-min glucose and insulin data disclosed that the response lag was <1 min. The secretion pattern in response to oscillatory glucose was therefore characterized by the coexistence of large insulin pulses next to small ones. Given the waveform damping and delay of the insulin secretion profile reaching the systemic circulation in vivo (27, 34), the simplified model prediction (Fig. 4) is that the smaller insulin pulses between the amplified pulses would not be resolved at this sampling site, wherein diffusion and turbulent blood flow introduce further nonuniformities. Although we cannot be certain that the same circumstances prevail in the intact pancreas as observed here in isolated islets, given the robust nature of the pacemaker dictating pulsatile insulin secretion the accompanying data allow the unifying postulate that the apparent entrainment of insulin pulses in vivo reflects the intermittent concordance of pacemaker activity with glucose elevations, undersampling, and false-negative errors in systemic pulse identification in that circumstance. An important question is, why was entrainment not present in patients with type 2 diabetes?

The most likely explanation is that in type 2 diabetes there is an attenuated response of insulin secretion to glucose stimulation (12, 21, 22, 38). Typically, beta-cell function is assessed clinically by supraphysiological glucose stimulation followed by measurement of first-phase insulin secretion (4, 20, 21). The elegant studies of Pørksen et al. (25) established a much more physiological protocol by intermittent infusion of small glucose pulses (amplitude ~0.3 mmol/l), which correspond to glucose oscillations that are present in humans (17). Given the high sensitivity of physiological vs. supraphysiological stimulation tests in endocrinology, the use of physiological glucose pulses might be a more sensitive discriminator, although this has not been formally verified to date. The present data allow the hypothesis that failure to detect entrainment of insulin pulses in patients with type 2 diabetes, given small glucose pulses, reflects lesser stimulation by glucose of concordant insulin pulses compared with nondiabetic subjects.

In the present islet perifusion studies, oscillating vs. constant glucose delivery also had a prominent effect to synchronize pulsatile insulin release by the 10–30 islets in each perifusion chamber. It is well documented that each independent islet has the capacity to serve as a pacemaker (2, 16, 18) and secrete insulin in a pulsatile manner at an interval of ~4 min (32, 33). Single perifused islets secrete ~60% of insulin in pulses (16, 33), a comparable proportion with that present in vivo. However, groups of perifused islets release a much lower proportion of their net insulin release in measurable insulin pulses (30–40%) (32, 33). The likely explanation for this difference is that groups of islets in the perifusion system are only partially synchronized, so although each individual islet is likely releasing most of its insulin in pulses, only those islets that are synchronized secrete insulin in pulses that are detectable in the perifusate. Interestingly, intermittent glucose pulses, even at a pulse interval that exceeded the inherent pulse interval for pulsatile insulin secretion, led to a marked increase (twofold) in the proportion of insulin secreted as detectable insulin pulses from the groups of islets. It is unlikely that this coordinating action is responsible for the reported capacity of infused glucose pulses to discriminate between patients with type 2 diabetes and control subjects (6, 17), because the proportion of insulin secreted in pulses is comparable in these two groups.

In the neuroendocrine system, pulsatile hormone secretion is important for eliciting a sustained and appropriate target tissue response. For example, pulsatile luteinizing hormone-releasing hormone (LHRH) release elicits greater secretion response of luteinizing hormone (LH) than does constant LHRH delivery (3). In healthy humans, blood glucose concentrations oscillate with a frequency (~4 min) that is similar to high-frequency insulin pulses (6, 17). This raises the possibility that one of the contributory factors toward impaired insulin secretion in response to a glucose challenge seen in type 2 diabetes is downregulation of this system due to constant hyperglycemia in type 2 diabetes. However, the present data argue against this notion. We report that there are no time-dependent differences in the insulin secretion rate with oscillatory or constant glucose stimulation (Fig. 7). Therefore, we reject the second hypothesis, that insulin secretion might be enhanced by oscillating vs. constant glucose stimulation. However, one limitation of the present experiments is that the perifusion experiments were performed for a period of only 100 min to avoid functional deterioration of the islets. It is possible that exposure of islets to oscillating vs. constant glucose over much longer periods would result in enhanced insulin release in response to oscillating glucose.

In conclusion, the present experiments demonstrate that the burst interval of pulsatile insulin secretion from isolated human islets is not entrained to glucose oscillations. The entrainment of pulsatile insulin to infused glucose oscillations in nondiabetic humans in vivo likely reflects entrainment of pulse mass without changes in underlying pacemaker activity. Oscillations of stimulatory glucose concentrations provide no quantitative advantage over constant glucose stimulation of insulin secretion but selectively enhance the pulsatile component of insulin secretion. Therefore, it is unlikely that the loss of glucose oscillations contributes to deficient insulin secretion in type 2 diabetes. Given the limitations of high-frequency pulse detection during intermittent glucose oscillations while sampling from the systemic circulation, any future efforts to prove the concept of insulin pulse entrainment in vivo should include portal vein sampling at 1-min intervals and a validated approach for pulse detection. Because entrainment of insulin pulses is absent in the present studies, perhaps it might be more appropriate to consider the phenomenon as the induction of amplified insulin pulses by coincident intermittent glucose pulses.


   GRANTS
TOP
 ABSTRACT
 RESEARCH DESIGN AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
These studies were funded in part by the Deutsche Forschungsgemeinschaft (Ri-1055) and the National Institute of Diabetes and Digestive and Kidney Diseases (DK-61539).


   ACKNOWLEDGMENTS
 
Human islets were obtained from the Human Islet Distribution Program at the University of Minnesota, Diabetes Institute for Immunology and Transplantation (Dr. Bernhard J. Hering) and Northwest Tissue Center, Seattle (Dr. R. Paul Robertson). Institutional Review Board approval for islet isolation was obtained at each institution.


   FOOTNOTES
 

Address for reprint requests and other correspondence: P. C. Butler, Larry Hillblom Islet Research Center, UCLA David Geffen School of Medicine, 24–130 Warren Hall, 900 Veteran Avenue, Los Angeles, CA 90095-7073 (e-mail: pbutler{at}mednet.ucla.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 RESEARCH DESIGN AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Andersen L, Dinesen B, Jorgensen PN, Poulsen F, and Roder ME. Enzyme immunoassay for intact human insulin in serum or plasma. Clin Chem 39: 578–582, 1993.[Abstract/Free Full Text]
  2. Chou HF and Ipp E. Pulsatile insulin secretion in isolated rat islets. Diabetes 39: 112–117, 1990.[Abstract]
  3. Crowley WF Jr and McArthur JW. Simulation of the normal menstrual cycle in Kallman’s syndrome by pulsatile administration of luteinizing hormone-releasing hormone (LHRH). J Clin Endocrinol Metab 51: 173–175, 1980.[Abstract]
  4. Davies MJ, Rayman G, Grenfell A, Gray IP, Day JL, and Hales CN. Loss of the first phase insulin response to intravenous glucose in subjects with persistent impaired glucose tolerance. Diabet Med 11: 432–436, 1994.[ISI][Medline]
  5. Hansen BC, Jen KC, Belbez Pek S, and Wolfe RA. Rapid oscillations in plasma insulin, glucagon, and glucose in obese and normal weight humans. J Clin Endocrinol Metab 54: 785–792, 1982.[Abstract]
  6. Hollingdal M, Juhl CB, Pincus SM, Sturis J, Veldhuis JD, Polonsky KS, Porksen N, and Schmitz O. Failure of physiological plasma glucose excursions to entrain high-frequency pulsatile insulin secretion in type 2 diabetes. Diabetes 49: 1334–1340, 2000.[Abstract]
  7. Juhl CB, Hollingdal M, Porksen N, Prange A, Lonnqvist F, and Schmitz O. Influence of rosiglitazone treatment on beta-cell function in type 2 diabetes: evidence of an increased ability of glucose to entrain high-frequency insulin pulsatility. J Clin Endocrinol Metab 88: 3794–3800, 2003.[Abstract/Free Full Text]
  8. Juhl CB, Porksen N, Hollingdal M, Sturis J, Pincus S, Veldhuis JD, Dejgaard A, and Schmitz O. Repaglinide acutely amplifies pulsatile insulin secretion by augmentation of burst mass with no effect on burst frequency. Diabetes Care 23: 675–681, 2000.[Abstract/Free Full Text]
  9. Juhl CB, Porksen N, Pincus SM, Hansen AP, Veldhuis JD, and Schmitz O. Acute and short-term administration of a sulfonylurea (gliclazide) increases pulsatile insulin secretion in type 2 diabetes. Diabetes 50: 1778–1784, 2001.[Abstract/Free Full Text]
  10. Juhl CB, Schmitz O, Pincus S, Veldhuis J, and Porksen N. Short-term treatment with GLP-1 increases pulsatile insulin secretion in Type II diabetes with no effect on orderliness. Diabetologia 43: 583–588, 2000.[CrossRef][ISI][Medline]
  11. Kjems LL, Kirby BM, Welsh EM, Veldhuis JD, Straume M, McIntyre SS, Yang D, Lefebvre P, and Butler PC. Decrease in beta-cell mass leads to impaired pulsatile insulin secretion, reduced postprandial hepatic insulin clearance, and relative hyperglucagonemia in the minipig. Diabetes 50: 2001–2012, 2001.[Abstract/Free Full Text]
  12. Laedtke T, Kjems L, Porksen N, Schmitz O, Veldhuis J, Kao PC, and Butler PC. Overnight inhibition of insulin secretion restores pulsatility and proinsulin/insulin ratio in type 2 diabetes. Am J Physiol Endocrinol Metab 279: E520–E528, 2000.[Abstract/Free Full Text]
  13. Lang DA, Matthews DR, Burnett M, and Turner RC. Brief, irregular oscillations of basal plasma insulin and glucose concentrations in diabetic man. Diabetes 30: 435–439, 1981.[Abstract]
  14. Lang DA, Matthews DR, Peto J, and Turner RC. Cyclic oscillations of basal plasma glucose and insulin concentrations in human beings. N Engl J Med 301: 1023–1027, 1979.[Abstract]
  15. Larsen MO, Rolin B, Wilken M, Carr RD, and Gotfredsen CF. Measurements of insulin secretory capacity and glucose tolerance to predict pancreatic beta-cell mass in vivo in the nicotinamide/streptozotocin Gottingen minipig, a model of moderate insulin deficiency and diabetes. Diabetes 52: 118–123, 2003.[Abstract/Free Full Text]
  16. Lin JM, Fabregat ME, Gomis R, and Bergsten P. Pulsatile insulin release from islets isolated from three subjects with type 2 diabetes. Diabetes 51: 988–993, 2002.[Abstract/Free Full Text]
  17. Mao CS, Berman N, Roberts K, and Ipp E. Glucose entrainment of high-frequency plasma insulin oscillations in control and type 2 diabetic subjects. Diabetes 48: 714–721, 1999.[Abstract]
  18. Marchetti P, Scharp DW, McLear M, Gingerich R, Finke E, Olack B, Swanson C, Giannarelli R, Navalesi R, and Lacy PE. Pulsatile insulin secretion from isolated human pancreatic islets. Diabetes 43: 827–830, 1994.[Abstract]
  19. O’Meara NM, Sturis J, Van Cauter E, and Polonsky KS. Lack of control by glucose of ultradian insulin secretory oscillations in impaired glucose tolerance and in non-insulin-dependent diabetes mellitus. J Clin Invest 92: 262–271, 1993.[ISI][Medline]
  20. Pfeifer MA, Halter JB, Beard JC, and Porte D Jr. Differential effects of tolbutamide on first and second phase insulin secretion in noninsulin-dependent diabetes mellitus. J Clin Endocrinol Metab 53: 1256–1262, 1981.[Abstract]
  21. Pfeifer MA, Halter JB, and Porte D Jr. Insulin secretion in diabetes mellitus. Am J Med 70: 579–588, 1981.[CrossRef][ISI][Medline]
  22. Polonsky KS, Given BD, Hirsch LJ, Tillil H, Shapiro ET, Beebe C, Frank BH, Galloway JA, and Van Cauter E. Abnormal patterns of insulin secretion in noninsulin-dependent diabetes mellitus. N Engl J Med 318: 1231–1239, 1988.[Abstract]
  23. Porksen N, Grofte B, Nyholm B, Holst JJ, Pincus SM, Veldhuis JD, Schmitz O, and Butler PC. Glucagon-like peptide 1 increases mass but not frequency or orderliness of pulsatile insulin secretion. Diabetes 47: 45–49, 1998.[Abstract]
  24. Pørksen N, Grofte T, Greisen J, Mengel A, Juhl C, Veldhuis JD, Schmitz O, Rossle M, and Vilstrup H. Human insulin release processes measured by intraportal sampling. Am J Physiol Endocrinol Metab 282: E695–E702, 2002.[Abstract/Free Full Text]
  25. Pørksen N, Juhl C, Hollingdal M, Pincus SM, Sturis J, Veldhuis JD, and Schmitz O. Concordant induction of rapid in vivo pulsatile insulin secretion by recurrent punctuated glucose infusions. Am J Physiol Endocrinol Metab 278: E162–E170, 2000.[Abstract/Free Full Text]
  26. Porksen N, Munn S, Steers J, Veldhuis JD, and Butler PC. Effects of glucose ingestion versus infusion on pulsatile insulin secretion. The incretin effect is achieved by amplification of insulin secretory burst mass. Diabetes 45: 1317–1323, 1996.[Abstract]
  27. Pørksen N, Munn S, Steers J, Veldhuis JD, and Butler PC. Impact of sampling technique on appraisal of pulsatile insulin secretion by deconvolution and cluster analysis. Am J Physiol Endocrinol Metab 269: E1106–E1114, 1995.[Abstract/Free Full Text]
  28. Pørksen N, Nyholm B, Veldhuis JD, Butler PC, and Schmitz O. In humans at least 75% of insulin secretion arises from punctuated insulin secretory bursts. Am J Physiol Endocrinol Metab 273: E908–E914, 1997.[Abstract/Free Full Text]
  29. Pørksen NK, Munn SR, Steers JL, Schmitz O, Veldhuis JD, and Butler PC. Mechanisms of sulfonylurea’s stimulation of insulin secretion in vivo: selective amplification of insulin secretory burst mass. Diabetes 45: 1792–1797, 1996.[Abstract]
  30. Ritzel R, Schulte M, Porksen N, Nauck MS, Holst JJ, Juhl C, Marz W, Schmitz O, Schmiegel WH, and Nauck MA. Glucagon-like peptide 1 increases secretory burst mass of pulsatile insulin secretion in patients with type 2 diabetes and impaired glucose tolerance. Diabetes 50: 776–784, 2001.[Abstract/Free Full Text]
  31. Ritzel RA, Hansen JB, Veldhuis JD, and Butler PC. Induction of beta-cell rest by a Kir6.2/SUR1-selective K(ATP)-channel opener preserves beta-cell insulin stores and insulin secretion in human islets cultured at high (11 mM) glucose. J Clin Endocrinol Metab 89: 795–805, 2004.[Abstract/Free Full Text]
  32. Ritzel RA, Veldhuis JD, and Butler PC. Glucose stimulates pulsatile insulin secretion from human pancreatic islets by increasing secretory burst mass: dose-response relationships. J Clin Endocrinol Metab 88: 742–747, 2003.[Abstract/Free Full Text]
  33. Song SH, Kjems L, Ritzel R, McIntyre SM, Johnson ML, Veldhuis JD, and Butler PC. Pulsatile insulin secretion by human pancreatic islets. J Clin Endocrinol Metab 87: 213–221, 2002.[Abstract/Free Full Text]
  34. Song SH, McIntyre SS, Shah H, Veldhuis JD, Hayes PC, and Butler PC. Direct measurement of pulsatile insulin secretion from the portal vein in human subjects. J Clin Endocrinol Metab 85: 4491–4499, 2000.[Abstract/Free Full Text]
  35. Song SH, Rhodes CJ, Veldhuis JD, and Butler PC. Diazoxide attenuates glucose-induced defects in first-phase insulin release and pulsatile insulin secretion in human islets. Endocrinology 144: 3399–3405, 2003.[Abstract/Free Full Text]
  36. Sturis J, Van Cauter E, Blackman JD, and Polonsky KS. Entrainment of pulsatile insulin secretion by oscillatory glucose infusion. J Clin Invest 87: 439–445, 1991.[ISI][Medline]
  37. Veldhuis JD, Carlson ML, and Johnson ML. The pituitary gland secretes in bursts: appraising the nature of glandular secretory impulses by simultaneous multiple-parameter deconvolution of plasma hormone concentrations. Proc Natl Acad Sci USA 84: 7686–7690, 1987.[Abstract/Free Full Text]
  38. Ward WK, Bolgiano DC, McKnight B, Halter JB, and Porte D Jr. Diminished B cell secretory capacity in patients with noninsulin-dependent diabetes mellitus. J Clin Invest 74: 1318–1328, 1984.[ISI][Medline]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Endocrinol. Metab.Home page
R. Bertram, A. Sherman, and L. S. Satin
Metabolic and electrical oscillations: partners in controlling pulsatile insulin secretion
Am J Physiol Endocrinol Metab, October 1, 2007; 293(4): E890 - E900.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
290/4/E750    most recent
00381.2005v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ritzel, R. A.
Right arrow Articles by Butler, P. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ritzel, R. A.
Right arrow Articles by Butler, P. C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2006 by the American Physiological Society.