Physiology & Behavior 105 (2012) 1082–1087

Contents lists available at SciVerse ScienceDirect

Physiology & Behavior

j ourna l homepage: www.e lsev ie r .com/ locate /phb

Novel findings in the cephalic phase of digestion: A role for microcirculation?☆

Caroline Buss a, Luiz Guilherme Kraemer-Aguiar b,⁎, Priscila A. Maranhão a, Carolina Marinho a,Maria das Graças C. de Souza a, Nicolas Wiernsperger c, Eliete Bouskela a

a State University of Rio de Janeiro, Biomedical Center, Clinical and Experimental Research Laboratory on Vascular Biology (BioVasc), Rio de Janeiro, RJ, Brazilb State University of Rio de Janeiro, Department of Internal Medicine, Endocrinology, Rio de Janeiro, RJ, Brazilc INSERM U1060, CarMeN Laboratory, INSA Lyon, France

☆ ClinicalTrials ID: NCT01145027. http://www.clinica⁎ Corresponding author at: Rua São Francisco Xavier,

Haroldo Lisboa da Cunha, Sala 104, Térreo, Rio de Janeir+55 21 2334 0703; fax: +55 21 2334 0692.

E-mail address: gkraemer@ig.com.br (L.G. Kraemer-

0031-9384/$ – see front matter © 2011 Elsevier Inc. Alldoi:10.1016/j.physbeh.2011.12.004

a b s t r a c t

a r t i c l e i n f o

Article history:Received 16 August 2011Received in revised form 22 November 2011Accepted 7 December 2011Available online 14 December 2011

Keywords:Cephalic phase of digestionMicrocirculationInsulinPancreatic polypeptideCapillary recruitment

The cephalic phase of digestion (CPD) has been extensively investigated in terms of digestion and metabolism.Nevertheless, microcirculatory changes required to prepare peripheral tissues in order to dispose nutrientshave never been assessed. In this study, microvascular function has been evaluated to determine its behaviorand potential association to hormonal secretions during CPD. Thirty-nine healthy male subjects, 23.4±0.5 years (mean±SD) and BMI of 23.3±2.3 kg/m2, were randomized into receiving cognitive-sensorial stimulito elicit CPD (CPD group, n=20) or not (control group, n=19), after a 12-h overnight fast. Main outcomes weredifferences in resting and peak functional capillary density (FCD, cap/mm2); resting red blood cell velocity(RBCV), peak RBCV (RBCVmax) and time taken to reach it (TRBCVmax); peak flow and vasomotion, before andafter CPD and their associations with insulin and/or pancreatic polypeptide (PP). In the CPD group, basal FCD(24.9±7.6 to 28.3±8.1, p=0.005), peak FCD (27.8±6.3 to 32.6±7.1, p=0.002), RBCV (0.306±0.031 to0.330±0.027 mm/s, p=0.005), RBCVmax (0.336±0.029 to 0.398±0.292 mm/s, p=0.005) and peak flow(23.5±14.3 to 26.9±15.8 PU, pb0.01) increased while TRBCVmax decreased (4.9±1.5 to 3.5±1.2 s, p=0.01).No significant changes could be detected in the control group. Groups have not presented differences for insulin,but PP significantly increased in the CPD group and was positively associated to basal FCD increase (rho=0.527,p=0.03). In conclusion, neurally-mediated anticipatory responses of digestion elicited functional capillary re-cruitment associated to PP in healthy men, suggesting a precocious role for microcirculation in the physiologyof digestion and nutrient homeostasis.

© 2011 Elsevier Inc. All rights reserved.

1. Introduction

The cephalic phase of digestion (CPD) occurs before food consump-tion and its effects can last from 10 to 60 min [1–5]. It results from sen-sory and cognitive stimuli such as sight, smell, taste or thought of food[2]. Cephalic-phase sensory stimuli reach the nucleus of the solitarytract and dorsal motor nucleus of the vagus and they stimulate the re-lease of hormones from the endocrine pancreas. This is consistent witha neural-dependent stimulation rather than a nutrient-induced one[5,6]. Insulin and pancreatic polypeptide (PP) secretions during CPDhave been extensively studied [1,2,7]. Their physiological significanceremains to be elucidated, yet PP secretion during this phase accountsfor 16% of the total PP released during digestion [7].

Literature data about the effects of insulin secretion during the ce-phalic phase point to its important role for the establishment of normalglucose tolerance [3,8]: some authors have shown that it can determine

ltrials.gov.524, Maracanã, UERJ, Pavilhãoo, CEP: 20550–013, Brazil. Tel.:

Aguiar).

rights reserved.

an increment on insulin levels [9] while others have questioned this abil-ity [10]. Nevertheless, neuralmediation of insulin during this periodmayhave multiple effects, ultimately facilitating metabolic homeostasis dur-ing food ingestion [11]. Physiologically, this period is fundamental fornutrient metabolism due to the early supply of insulin to tissues primingits receptors and allowing further glucose metabolization [8,12].

Conceivably, regulation of glucose uptake in peripheral tissues in-volves a vascular component. It implies an increase in nutritive perfu-sion for its maximum delivery and storage in insulin-sensitive tissues,which has been primarily shown in studies performed in skeletalmuscle [13]. However, bulk flow does not necessarily reflect changesin nutritive capillary perfusion achieved either by dilation or vasomo-tion of terminal arterioles. Physiologically, insulin increases mainly inresponse to blood glucose elevation. Investigations have not beenconducted on very low insulin increments and their consequent vaso-active effects, which may occur in humans during CPD in the absenceof concomitant elevations in plasma glucose. Previous studies usingexperimental animals have shown an increase in muscle capillaryperfusion long before the occurrence of any augmentation in bulkblood flow or glucose uptake [14]. In human subjects, physiologicalinsulin secretion and capillary recruitment have been investigatedonly in relation to intestinal glucose absorption [15].

1083C. Buss et al. / Physiology & Behavior 105 (2012) 1082–1087

We have already published that microvascular dysfunction plays arole in glucose homeostasis [16] and impaired capillary recruitmentseems to be involved in insulin resistance due to decreased tissue per-fusion [17,18]. Although hemodynamic effects of insulin have alreadybeen demonstrated and well-tested [13,19,20], there are few studiesinvestigating alterations in microcirculation due to CPD — a periodin which endogenous pancreatic secretions should play an importantrole. Taking into account that several nervous and hormonal signalsare activated during CPD [21,22], we have aimed to evaluate whethermicrocirculatory changes might be related to them and discuss theirrole in CPD in healthy subjects.

2. Material and methods

2.1. Study design, subjects and setting

This was a randomized trial. Healthy subjects aged between 18 and30 years were selected after clinical and laboratorial assessments. Allparticipants provided their written informed consent. The study wasapproved by the institutional ethics committee (Pedro Ernesto Univer-sity Hospital/Project 2145) and was carried out at the Clinical and Ex-perimental Research Laboratory on Vascular Biology (BioVasc) of theState University of Rio de Janeiro. GPower 3.1.10 software was usedfor power analysis and sample size estimation. The statistical powerfor comparisons between two dependent groups was above 0.8 withan α-error probability of 0.05 for functional capillary density (Δ 3.64cap/cm) [23], estimating a sample size of 14 subjects per group.

Subjects were submitted to a screening phase before being eligibleto participate in the study. This phase comprised measurements ofweight, height and percentage of body fat using a skin-fold thicknessprotocol performed by a trained dietitian [24]. Subjects who weresmokers or presented high blood pressure (≥130×85 mm Hg) [25]or any clinical condition were excluded. Subjects with body massindex (BMI) lower or higher than 18.5–24.9 kg/m2 were also exclud-ed. Those with BMI within the overweight range (25.0–29.9 kg/m2),but showing a percentage of body fat below the average (b15%) formale adults [24] were included in the study. Fasting glucose and insu-lin, hemogram, total cholesterol, triglycerides, low and high densitylipoproteins, alanine (ALT) and aspartate (AST) transaminases andcreatinine were assessed after a 12-hour overnight fast. A 75-g oralglucose tolerance test was also performed. Subjects presenting anydegree of glucose intolerance according to the American Diabetes As-sociation criteria [26] or any other altered laboratorial results, wereexcluded.

Forty-eight subjects were recruited. On the day of the trial, volun-teers arrived at the laboratory after a 12-h overnight fast and were ac-commodated in a temperature-controlled room (24±1 °C), on theexamination chair. Their blood pressure was checked using the stan-dard auscultatory method. An intravenous catheter was inserted andkept in place during the exam for blood samples withdrawal to assessglucose, insulin and pancreatic polypeptide basal levels. After a 30-min acclimatization period, the experiment was started. Subjectswere submitted to two microcirculatory evaluations, with a 10 mininterval between each other (Fig. 1). During the interval, accordingto the randomization, volunteers were subjected to a sensorial stimu-lation (CPD group, n=24) or nothing (control group, n=24).

BS

Acclimatization Videocap 1

SM

0’

BS

3’

SM

Laser-Doppler flowmetry an

-45’ -15’

Fig. 1. Experimental design. BS: blood sample collection. SM: salivation measurement. Vpresentation.

2.2. Sensorial stimulation to trigger the cephalic phase of digestion

The aim of the stimulus was to trigger the cephalic phase of diges-tion by placing in front of the volunteer a breakfast tray containingtheir reported favorite food items. We have preferred to ask individ-uals about their favorite choices of breakfast food items in order tobest elicit the sensorial stimulation. The tray was positioned wherethe subject could see and smell it continuously but could not eat it be-fore the end of the exam. Additionally, cognitive stimuli wereattempted by talking about the volunteer's food preferences.

Salivation was measured (Shirmer Test Stripes, Ophtalmos, SãoPaulo, SP, Brazil) during 1 min at baseline and upon presentation ofthe stimulus (or at time 0 for the control group). Afterwards, asugar-free chewing gum (to avoid any absorption bias) was given tosubjects that were presented to the sensorial stimulus, in order toelicit oral stimulation. Blood samples were drawn at 3, 9 and 15 minafter the stimulus presentation (or after the end of the first microcir-culatory evaluation for the control group) to determine levels of insu-lin and pancreatic polypeptide. This time frame was selectedaccording to literature evidence of insulin and pancreatic polypeptidepeaks during the cephalic phase [10,27]. Glucose levels were alsomeasured in both groups for glycemic control at minute 15.

2.3. Microvascular function assessment

Skin microvascular assessment was used as a model of systemicmicrocirculation [28]. For videocapillaroscopy, subjects were seatedcomfortably in a fixed, high-based chair with the left upper armraised at heart level. The forearm and hand (dorsal side up) wererested upon a pedestal and the fourth finger supported by an acrylicbase. This base was mounted on the x–y stage of a three-eyepieceLeica DM/LMmicroscope (Wetzlar, Germany) equipped with an epiil-lumination system (100 W Xenon lamp). A CCD video camera (Sam-sung, Seoul, South Korea) was coupled to the microscope andconnected to a DVD recorder (Samsung, Seoul, South Korea) and aKodo KBM1700E monitor (Seoul, South Korea). The fingertip wasfixed to the acrylic base by a metal ring to minimize the movements.The skin temperature of the finger was monitored throughout the ex-amination with an YSI Precision 4000A digital thermometer (Dayton,Ohio, U.S.A.) with the thermistor probe taped within 1 cm proximalto the nailfold. A drop of mineral oil was spread over the observationsite to improve image quality. A pressure cuff (1 cmwide) was placedaround the proximal phalanx and connected to a mercury manome-ter. The exam was continuously recorded on DVDs for subsequentmeasurements of microcirculatory parameters using the CapImagesoftware [29]. The analyses were performed by the same observerblinded to the stimulus (data entry records were codified, thus, atthe time of the analysis the researcher did not know to which groupthe volunteer belonged).

Two different approaches were applied when recording the nail-fold area. This was a methodological strategy, in order to collect asmany microvascular variables as possible and to test whether or notresults pointed in the same direction. Subjects were randomizedaccording to one of the following methods. The dynamic nailfoldvideocapillaroscopy [23,30] recorded data on red blood cell velocity(RBCV) at rest, after 1 min arterial occlusion (RBCVmax) – which

BS BS

9’ 15’

Iontophoresis

d vasomotion

Videocap 2

24’ 40’

ideocap: videocapillaroscopy exam. The arrow indicates the time point of stimulus

1084 C. Buss et al. / Physiology & Behavior 105 (2012) 1082–1087

allows evaluation of hemodynamic behavior of the microcirculationdue to increased shear stress [31] – and the time taken to reach it(TRBCVmax) with a final magnification of ×680, before and duringthe post-occlusive reactive hyperemia response (PORH). The dorsalfinger videocapillaroscopy validated and described elsewhere [32]assessed only functional capillary density (FCD), i.e. the number ofcapillaries/unit tissue area (mm2) with flowing red blood cells. Brief-ly, it estimates baseline capillary density by counting the number ofcontinuously red blood cell-perfused capillaries during a 15-secondperiod. Post-occlusive reactive hyperemia was also used to assesscapillary recruitment after 4-min ischemia, which is a methodologi-cally appropriate time to assess such variable [32].

2.4. Skin microvascular blood flow and vasomotion

Measurements using laser Doppler flowmetry (LDF) are based ontransmission of low-power laser light (780 nm) to the tissue via afiber-optic probe that penetrates the skin 0.4 to 1.0 mm. The extentof light penetration enables the evaluation of net red blood cell flux[in arbitrary units (PU)] as the product of concentration of movingblood cells and its velocity, in an area of approximately 1 mm2, incapillaries, arterioles and venules, as well as in anastomosis of deeperblood vessels in dermal layers [33]. Blood flow and vasomotion wererecorded (Perimed AB, Stockholm, Sweden) throughout the experi-ment (approximately 30 min) with the probe positioned at the dorsalside of the left wrist.

Fast Fourier transform analysis was performed using Perisoft soft-ware (PSW version 2.50, Perimed AB, Stockholm, Sweden) to deter-mine the contribution of different vasomotion frequencycomponents to the variability of the signal. We have assumed five fre-quency intervals previously defined [34] in the spectrum between0.01 and 1.6 Hz: (1) endothelial (0.01–0.02 Hz); (2) neurogenic(0.02–0.06 Hz); (3) myogenic (0.06–0.15 Hz); (4) respiratory(0.15–0.4 Hz) and (5) heart beat (0.4–1.6 Hz) activities. Absolute am-plitude variations within each frequency band were analyzed, as wellas their normalized values, which are the interval of the mean ampli-tude divided by the total spectrum mean amplitude [34].

Table 1Anthropometrical and clinical characteristics between control and experimental (CPD)groups at baseline.

2.5. Insulin iontophoresis

All of the participants were subjected to transdermal iontophore-sis of insulin [1.8 μl of solution: Humulin-R, Lilly, diluted (1/10) indeionized water] at the end of the study protocol. This was also amethodological strategy to assure their microvascular reactivity tothe hormone exogenously administered, as previously described inhealthy subjects [35]. Insulin was delivered with a cathodal currentduring 12 doses (0.2 mA for 20 s) with 40-s interval between eachdose [35]. Baseline skin blood flow was recorded for 2 min beforethe protocol started. The flow increment was calculated as the per-cent increase of the area under the curve (AUC) subtracting baselineAUC from plateau AUC which was reached in the last two doses.

Control group (n=19) CPD group (n=20)

Age (y) 23.2±2.7 23.7±3.4BMI (kg/m2) 23.3±2.1 23.3±2.5Body fat (%) 14.3±5.1 12.9±5.3Waist circumference (cm) 83±5.3 82.5±6.6SBP (mm Hg) 117±4.2 110±11DBP (mm Hg) 72±5.6 70±7.7Fasting glucose (mg/dl) 84±5.3 84±5.8Glucose post-OGTT (mg/dl) 81±18.6 74±18.8Insulin (μU/dl) 6.9±2.3 5.8±1.9Total cholesterol (mg/dl) 152±29.0 141±26.4HDL-c (mg/dl) 49±7.3 48±7.1LDL-c (mg/dl) 87±24.6 80±25.2TG (mg/dl) 74±29.0 61±17.4

Values are mean±SD. BMI: Body mass index; SBP: systolic blood pressure; DBP:diastolic blood pressure; OGTT: oral glucose tolerance test; TG: triglycerides

2.6. Laboratory analyses

All laboratory measurements were performed in duplicate by anautomated method (Modular Analytics E 170 and P, Roche, Basel,Switzerland) except for pancreatic polypeptide (PP), for which wasused the human enzyme-linked immunosorbent assay Kit (Millipore,Massachusetts, USA). More specifically, serum insulin levels were an-alyzed by electrochemiluminescence, plasma glucose levels by GOD-PAP (enzymatic-colorimetric oxidase–peroxidase method) andserum PP by ELISA and inter-assay coefficients of variation were10.6%, 1.1% and 8.5%, respectively. Analyses were performed by atechnician blinded to randomization.

2.7. Outcomes and statistical analyses

The main outcomes of the trial were differences and correlationsbetween microvascular parameters assessed pre- and post-stimulusand endocrine pancreatic hormonal changes during CPD. Parametersof microvascular function were: basal functional capillary density(FCD) and peak FCD during PORH measured by dorsal finger videoca-pillaroscopy; afferent, efferent and apical diameters of capillary loops,resting red blood cell velocity (RBCV), peak RBCV during PORH(RBCVmax) and time taken to reach it (TRBCVmax) measured by dy-namic nailfold videocapillaroscopy; changes in microflow and vaso-motion frequencies assessed by LDF [36]. Data entry was performedby a researcher blinded to the stimulus and randomization codeswere only opened after the completion of the study.

All of the variables were checked for normality. Clinical baselinevariables normally distributed were compared using Student's t-test. Most microvascular and biochemical outcomes presented non-parametrical distribution and intra-group differences pre- and post-stimulus were analyzed using Wilcoxon matched-pairs test whilerepeated measures used Friedman test followed by post-hoc Dunn'stest. Comparisons between groups were assessed using Mann–Whit-ney U test. Correlations between variables were tested throughSpearman rho test. Results are presented as mean±SD, unless other-wise noted, as median (1st; 3rd quartiles). The level of significanceadopted was 0.05.

3. Results

Thirty-nine subjects aged 23.4±0.5 years with BMI 23.3±2.3 kg/m2 and body fat 13.6±5.2% were included in the analysis. At base-line, no significant differences could be detected between controland experimental (CPD) groups (Table 1). Blood samples of 9(18.7%) subjects [5 in the control group and 4 in the CPD group]were not evaluated due to poor sample quality, but no significant dif-ferences between excluded and included subjects were noticed (datanot shown — Supplementary Table 1).

3.1. Microvascular outcomes

Basal FCD significantly increased in volunteers of the CPD groupwhile it did not change in controls, respectively: 24.9±7.6 to 28.3±8.1 cap/mm2 or 13.7% increment, p=0.005 and 21.2±10.5 to 22.7±9.2 cap/mm2 or 7.1% increment, p=0.2 (Table 2). Peak FCD duringPORH increased in the CPD (17.3% increment, p=0.002) but not inthe control group (1.7% decrement, p=0.38). Additionally, in the CPDgroup we were also able to demonstrate increments in RBCV and in

Table 2Microcirculatory parameters (mean±SD) measured by videocapillaroscopy in thecontrol and experimental (CPD) groups, pre- and post-stimulus presentation.

Control group CPD group

Videocapillaroscopymethod

Pre Post Pre Post

Dorsal finger (n=10) (n=10)Resting FCD(cap/mm2)

21.2±10.5 22.7±9.2 24.9±7.6 28.3±8.1⁎

FCD during PORH(cap/mm2)

23.2±8.4 22.8±9.4 27.8±6.3 32.6±7.1†

Dynamic nailfold (n=9) (n=10)RBCV (mm/s) 0.329

±0.0180.328±0.019

0.306±0.031

0.330±0.027⁎

RBCVmax (mm/s) 0.362±0.021

0.360±0.023

0.336±0.029

0.398±0.030⁎

TRBCVmax (s) 3.7±0.6 3.8±0.8 4.9±1.5 3.5±1.2‡

Values are mean±SD. FCD: functional capillary density; PORH: post-occlusive reactivehyperemia; RBCV: red blood cell velocity; RBCVmax: peak red blood cell velocity;TRBCVmax: time taken to reach RBCVmax.Wilcoxon matched-pairs test difference pre- and post-stimulus.⁎ p=0.005.† p=0.009.‡ p=0.01.

1085C. Buss et al. / Physiology & Behavior 105 (2012) 1082–1087

RBCVmaxwhile TRBCVmax decreased. None of these variables changed inthe controls. No significantmorphological differences were detected foreither group on studied diameters of capillary loops. Skin temperaturedid not change significantly throughout the protocol (CPD group,p=0.6; control group, p=0.5).

Assessment of blood flow by LDF showed a significant increase inpeak flow (perfusion units) in the CPD group (23.5±14.3 to 26.9±15.8, p=0.01), but not in controls (19.5±9.0 to 20.5±11.0,p=0.8) (Table 3). Baseline peak flow was not different betweengroups (p=0.65). CPD and control groups did not show any differ-ences for frequency intervals and total spectrum of spontaneousarteriolar vasomotion (Table 3). These comparisons remained non-significant even after values were normalized. Comparisons of pre-and post-stimulus frequencies within groups were only possible inthe sub-sample (n=10 in each group) subjected to dorsal fingervideocapillaroscopy, due to the minimum time required to analyzevasomotion (15 min). Microvascular responses to transdermal exoge-nous insulin iontophoresis were not different between the CPD andcontrol groups, with both showing increments on microflow respec-tively of 126.5 (38.6; 354.4) % and 90.2 (34.3; 395.5) %, p=0.65.

3.2. Cephalic phase markers

Baseline insulin [6 (5; 9) and 6 (4; 8) μU/ml, p=0.7] and PP [162.4(72.3; 283.3) and 439.6 (111.3; 1149.6) pg/ml, p=0.08] levels were

Table 3Microcirculatory parameters (mean±SD) measured by laser-Doppler flowmetry in thecontrol and experimental (CPD) groups, pre- and post-stimulus presentation.

Control group (n=19) CPD group (n=20)

Pre Post Pre Post

Peak flow (PU) 19.5±9.0 20.5±11.0 23.5±14.3 26.9±15.8a

Vasomotion (PU/Hz)b (n=10) (n=19) (n=10) (n=20)Endothelial 0.70±0.32 0.85±0.35 0.66±0.33 0.97±0.79Neurogenic 0.53±0.23 0.66±0.28 0.56±0.27 0.78±0.53Myogenic 0.43±0.35 0.57±0.45 0.62±0.66 0.83±0.60Respiratory 0.18±0.23 0.27±0.29 0.38±0.46 0.39±0.37Cardiac 0.10±0.08 0.13±0.10 0.15±0.14 0.20±0.15

Values are mean±SD. PU: perfusion units.a Wilcoxon matched-pairs test difference pre- and post-stimulus: p=0.01.b Pre-stimulus intervals could not be analyzed in 9 subjects due to lack of minimum

measurement time.

not significantly different between control and CPD groups. Insulindid not change from minutes 3 to 15 (Table 4) while PP levels signif-icantly increased in the CPD group and remained unaltered in thecontrol group (median percentages: 2.4% and 0%, respectively;p=0.02). After cephalic phase stimulation, absolute insulin levels [6(3.5; 8) and 5 (4.5; 6) μU/ml, p=0.4] were not significantly differentbetween control and CPD groups, respectively, while PP levels were[150.3 (77.8; 345) and 480.6 (119.4; 1198.5) pg/ml, p=0.03]. In ad-dition, salivation, a clinical marker of CPD, similar in both groups atbaseline (p=0.73), significantly increased in the CPD but not in thecontrol group (median: 100% and 0%, respectively; pb0.001). No dif-ferences could be detected in glucose levels at 15 min (p=0.07) fornone of the groups.

3.3. Correlations between endocrine outcomes and microcirculation dur-ing CPD

Unaltered insulin levels limited our correlation analyses to PPchanges from minutes 3 to 15 and their associations to microvascularreactivity during CPD. PP was positively associated with changes inbasal FCD in the CPD group (rho=0.527, p=0.03), but not in thecontrol group (Fig. 2). PP results of four patients in the CPD groupand five in the control group did not reach assay sensitivity(12.3 pg/ml) and therefore were excluded from these analyses with-out compromising the balance between the two groups (data notshown — Supplementary Table 2). The variations in basal FCD post-stimulus in these excluded subjects ranged from −10.0 to 10.5% inthe control group and from 4.0 to 18.2% in the CPD group.

4. Discussion

To the best of our knowledge, this is the first study that demon-strated an association between a hormonal secretion and functionalchanges in both non-nutritive and nutritive microcirculatory param-eters during CPD in healthy men. Our findings support that at pre-ingestive stages of digestion there is not only an activation ofgastro-intestinal endo- and exocrine secretions but also, and concom-itantly, capillary recruitment and an increase on microflow, the for-mer being associated with PP increments. Despite the need foraugmented tissue perfusion in humans to allow nutrient exchange,until now there were no data demonstrating the role of microcircula-tion on such precocious stage of digestion and nutrient metabolism.

Rather than insulin, our data suggest that pancreatic polypeptide(PP) may have a role during this phase. In fact, one recent study inrats has shown that human PP evoked concentration-dependent re-laxation of mesenteric small arteries by activation of pre-junctionalneuropeptide Y1-like receptors followed by the release of calcitoningene-related peptide and of endothelium-derived nitric oxide (NO)[37]. Another experimental study also provided evidence for a directvascular action of acetylcholine secondary to parasympathetic stimu-lation, as well as an indirect vasodilator effect of vasoreactive intesti-nal peptide via NO formation [38]. The hypothesis of the observedfunctional microcirculatory changes being due to both effects, aswell as to some insulin action (although its rise was not detectable

Table 4Changes in markers of cephalic phase of digestion [median (p25;p75)], from minutes 3to 15.

Control group (n=19) CPD group (n=20)

Delta % Delta %

Insulin(μU/ml)

0 (−1; 1) 0 (−17; 20) − 1 (−2; 1) −9 (−33; 20)

PP (pg/ml) 0.12 (−41; 15.3) 0 (−9.6; 4.5) 3.2 (−29; 32.3) 2.4 (−3.4; 6.3)a

Values are median (p25;p75). PP: Pancreatic Polypeptide.a Mann–Whitney U test difference between groups: p=0.02.

Fig. 2. Spearman's rank correlation between changes in pancreatic polypeptide (PP) during cephalic phase stimulation (minutes 3–15) and basal functional capillary density (FCD)in the CPD (n=16) and control (n=14) groups.

1086 C. Buss et al. / Physiology & Behavior 105 (2012) 1082–1087

during CPD), should be investigated further. According to our findings,it is possible that not only insulin [14] but other gastro-intestinalpeptides like PP, could play a role on vascular tissue during CPD.

The lack of increased insulin levels on the experimental groupcannot be defined as unexpected in light of previous studies of CPD,which have shown high-, low-, and negative responses of insulin se-cretion to food sensorial stimulation in healthy subjects [2,39]. Lowincrements of insulin could have occurred under the assay sensibilityand simultaneous sympathetic nervous system activation might havealso hindered the detection of insulin release [11]. Additionally, dueto low insulin increments previously observed in the absence of glu-cose stimulation [2] we could also hypothetically suggest that the ab-sence of detectable insulin increments on CPD group may alsorepresent its transendothelial passage to interstitial space and bind-ing to cellular receptors as a consequence of enhanced microflow[18]. The significant rise in PP in this group gives us evidence of suc-cessful stimulation of the cephalic phase of digestion [40]. In accor-dance with our findings, it was shown that PP could increase even ifinsulin is not incremented during CPD [5]. Its secretion is almost ex-clusively dependent of vagal stimulus, which releases acetylcholinein the pancreatic islet cells, also promoting secretion of insulin andglucagon. Of note, we should emphasize that the observed significantrise of exocrine secretions by salivary glands was the main clinicalfinding demonstrating an effective activation of the parasympatheticsystem by thought, sense and/or anticipation of food intake [9] inthe experimental group.

Cutaneous blood flow is a biological variable that fluctuates dy-namically and has substantial spatial heterogeneity. Nonetheless,our results have pointed in the same direction, demonstrating the ex-istence of functional changes in the microcirculation in healthy menin different sites assessed by distinct techniques, thus suggesting aninvolvement of the microcirculation in the CPD. Despite the fact thatmicrocirculatory assessment techniques in skin do not clearly expressphysiological and pathophysiological alterations displayed in skeletalmuscle, the inference of findings in cutaneous tissue might bring newconcepts on glucose homeostasis and its relation to microcirculation.As wisely pointed out by Holowatz and colleagues [28], cutaneous mi-crovascular responses integrate neural, endothelial and vascularsmooth muscle contributions, and these complex interactions in vas-cular signaling are the nature of human integrative physiology in vivo,regardless of the microvascular bed being studied. In addition, ourfinding of increased resting functional capillary density followingmeal-mediated cognitive–sensorial stimulus is worth noting as a con-sequence of the cephalic phase per se, since it is separated from thecompound stimuli involved in the reactive hyperemia. Further studiesare needed to clarify PP involvement in this situation.

Nutritive skin microcirculation was assessed by two distinctmethods, dynamic nailfold and dorsal finger videocapillaroscopies,in order to supply as many parameters as possible for analyses.

These two methods are well-validated and used by research groupsworldwide [17,32]. Laser-Doppler flowmetry (LDF) measurementswere applied for analyses of net microflow at another site (wrist).While measuring skin microvascular perfusion, LDF probes keep theskin temperature constant, ruling out the possibility of skin tempera-ture variation to act as a misleading factor. This technique confirmedthe occurrence of increased blood flow in predominantly non-nutritive [36] microvessels following cognitive–sensorial stimulus.In addition, skin insulin iontophoresis performed on both groups atthe end of the protocol confirmed that, irrespective of the group,healthy volunteers had similar microvascular responsiveness to exog-enous insulin. Even considering the differences between endogenousand exogenous stimuli on their actions on vascular tissue, the factthat both groups had the same microvascular reactivity during exog-enous insulin iontophoresis reassures that all subjects were able toreact to at least one of the hormones tested.

The absence of significant differences on vasomotion between CPDand control groups strongly suggests that observed increments in FCDand other microcirculatory variables could be due to arteriolar dilata-tion. On the other hand, the fact that vasomotion and insulin secre-tion were not altered during CPD is well in line with findings ofRenaudin and co-workers [41] that point to an increase in vasomo-tion related only to concomitantly infused insulin and glucose.

Data on type 2 diabetes mellitus (T2DM) and obese subjects sug-gest that CPD is impaired [4] and could influence glucose homeosta-sis. Although the upcoming study in our research line is to assesspossible microvascular responsiveness in states of T2DM, obesityand metabolic syndrome during CPD, first we needed to assess it inhealthy individuals in order to confirm microcirculatory involvementin this phase. A phenomenon should be first ascertained before onecan begin to understand its mechanistic features. Therefore, thisstudy was performed as a pilot needed to provide information for fur-ther pathophysiological investigation. Due to the nature of our out-come – microvascular function – we have faced a study designlimitation: any attempt of blockage of mechanisms involved in hor-monal secretion or cholinergic activity during the cephalic phase inhumans (i.e. by somatostatin or its analogs or atropine) would posea bias on circulation functioning as well [42].

Physiologically, our results bring up an important novel finding.The activation of the microcirculation is perhaps involved in theCPD as an anticipatory mechanism to the physiological burden associ-ated with eating, as elegantly reviewed by Woods [43]. Up to thispoint, the role of endogenous insulin and its cause–effect relationshipto microcirculatory activation on pre-ingestive phase of digestionshould be kept as a hypothesis, since the action of exogenous insulinon microcirculation at fasting state [18,20] has already been de-scribed. We must highlight that another possible explanation for theobserved improvement on microvascular reactivity could be the ef-fect of parasympathetic stimulus generated by sensorial stimulation

1087C. Buss et al. / Physiology & Behavior 105 (2012) 1082–1087

acting on the microcirculation while it does so on exocrine secretionsof the gastro-intestinal tract [7]. However, on the basis of our findings,we might suggest that the effect of endocrine secretions secondary tothe same vagal stimulation [7], clinically expressed as PP secretion,was positively correlated to an increase on basal capillary density. Fu-ture protocols should be performed to define the exact clinical andphysiological impact of microcirculatory changes during CPD onstates of disease and its effects on nutrient metabolism and glucosehomeostasis.

Funding sources

National Council for Scientific and Technological Development(CNPq - 141690/2008-9); State of Rio de Janeiro Carlos Chagas FilhoResearch Foundation (FAPERJ); Financing Agency of Studies and Pro-jects (FINEP) and Coordination for the Enhancement of EducationPersonnel (CAPES): these are government funding agencies andthey had no involvement in the study design; in the collection, anal-ysis, and interpretation of data; in the writing of the report; and in thedecision to submit the paper for publication.

Conflicts of interest

The authors declare that they have no conflict of interest.

Acknowledgments

The authors thank Mr. Diogo Silva Francisco, Ms. Yasmine Vieiraand Mrs. Monique Carneiro for their excellent technical assistance.We also thank Mrs. Camila Martins-Bekat and Dr. James Mandibergfor their valuable comments.

Appendix A. Supplementary data

Supplementary data to this article can be found online at doi:10.1016/j.physbeh.2011.12.004.

References

[1] Berthoud HR, Bereiter DA, Trimble ER, Siegel EG, Jeanrenaud B. Cephalic phase, re-flex insulin secretion. Neuroanatomical and physiological characterization. Diabe-tologia 1981;20:393–401 Suppl.

[2] Bellisle F, Louis-Sylvestre J, Demozay F, Blazy D, Le MJ. Cephalic phase of insulinsecretion and food stimulation in humans: a new perspective. Am J Physiol1985;249:E639–45.

[3] Bruce DG, Storlien LH, Furler SM, Chisholm DJ. Cephalic phase metabolic re-sponses in normal weight adults. Metabolism 1987;36:721–5.

[4] Ahren B. Autonomic regulation of islet hormone secretion—implications forhealth and disease. Diabetologia 2000;43:393–410.

[5] Teff K. Nutritional implications of the cephalic-phase reflexes: endocrine re-sponses. Appetite 2000;34:206–13.

[6] Powley TL. The ventromedial hypothalamic syndrome, satiety, and a cephalicphase hypothesis. Psychol Rev 1977;84:89–126.

[7] Zafra MA, Molina F, Puerto A. The neural/cephalic phase reflexes in the physiologyof nutrition. Neurosci Biobehav Rev 2006;30:1032–44.

[8] Ahren B, Holst JJ. The cephalic insulin response to meal ingestion in humans is de-pendent on both cholinergic and noncholinergic mechanisms and is important forpostprandial glycemia. Diabetes 2001;50:1030–8.

[9] Sahakian BJ, Lean ME, Robbins TW, James WP. Salivation and insulin secretion inresponse to food in non-obese men and women. Appetite 1981;2:209–16.

[10] Bellisle F, Louis-Sylvestre J, Demozay F, Blazy D, Le MJ. Reflex insulin response as-sociated to food intake in human subjects. Physiol Behav 1983;31:515–21.

[11] Teff KL. How neural mediation of anticipatory and compensatory insulin releasehelps us tolerate food. Physiol Behav 2011;103:44–50.

[12] Wiernsperger N. Is defective microcirculation responsible for insulin resistance ?Part 2: insulin and microcirculation: physiology/pathophysiology. In: Wiernsper-ger N, Bouskela E, Kraemer-Aguiar LG, editors. Microcirculation and insulin resis-tance. Bentham Books; 2009. p. 140–55.

[13] Baron AD, Steinberg HO, Chaker H, Leaming R, Johnson A, Brechtel G. Insulin-mediated skeletal muscle vasodilation contributes to both insulin sensitivityand responsiveness in lean humans. J Clin Invest 1995;96:786–92.

[14] Vincent MA, Clerk LH, Lindner JR, Klibanov AL, Clark MG, Rattigan S, et al. Micro-vascular recruitment is an early insulin effect that regulates skeletal muscle glu-cose uptake in vivo. Diabetes 2004;53:1418–23.

[15] Vincent MA, Clerk LH, Lindner JR, Price WJ, Jahn LA, Leong-Poi H, et al. Mixed mealand light exercise each recruit muscle capillaries in healthy humans. Am J PhysiolEndocrinol Metab 2006;290:E1191–7.

[16] Kraemer-Aguiar LG, Maranhao PA, Sicuro FL, Bouskela E. Microvascular dysfunction:a direct link among BMI, waist circumference and glucose homeostasis in youngoverweight/obese normoglycemic women? Int J Obes (Lond) 2010;34:111–7.

[17] Kraemer-Aguiar LG, Maranhao PA, Cyrino FZ, Bouskela E. Waist circumferenceleads to prolonged microvascular reactive hyperemia response in young over-weight/obese women. Microvasc Res 2010;80:427–32.

[18] Barrett EJ, Eggleston EM, Inyard AC, Wang H, Li G, Chai W, et al. The vascular ac-tions of insulin control its delivery to muscle and regulate the rate-limiting step inskeletal muscle insulin action. Diabetologia 2009;52:752–64.

[19] Kim JA, Montagnani M, Koh KK, Quon MJ. Reciprocal relationships between insu-lin resistance and endothelial dysfunction: molecular and pathophysiologicalmechanisms. Circulation 2006;113:1888–904.

[20] Serne EH, Ijzerman RG, Gans RO, Nijveldt R, de, V. G., Evertz R, et al. Direct evi-dence for insulin-induced capillary recruitment in skin of healthy subjects duringphysiological hyperinsulinemia. Diabetes 2002;51:1515–22.

[21] Power ML, Schulkin J. Anticipatory physiological regulation in feeding biology: ce-phalic phase responses. Appetite 2008;50:194–206.

[22] Smeets PA, Erkner A, de, G. C.. Cephalic phase responses and appetite. Nutr Rev2010;68:643–55.

[23] Kraemer de Aguiar LG, Laflor CM, Bahia L, Villela NR, Wiernsperger N, Bottino DA,et al. Metformin improves skin capillary reactivity in normoglycaemic subjectswith the metabolic syndrome. Diabet Med 2007;24:272–9.

[24] Jackson AS, Pollock ML. Generalized equations for predicting body density of men.Br J Nutr 1978;40:497–504.

[25] Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo Jr JL, et al. Sev-enth report of the joint national committee on prevention, detection, evaluation,and treatment of high blood pressure. Hypertension 2003;42:1206–52.

[26] ADA. Diagnosis and classification of diabetes mellitus. Diabetes Care 2004;27:S5–S10.

[27] Morricone L, Bombonato M, Cattaneo AG, Enrini R, Lugari R, Zandomenighi R, et al.Food-related sensory stimuli are able to promote pancreatic polypeptide eleva-tion without evident cephalic phase insulin secretion in human obesity. HormMetab Res 2000;32:240–5.

[28] Holowatz LA, Thompson-Torgerson CS, Kenney WL. The human cutaneous circu-lation as a model of generalized microvascular function. J Appl Physiol 2008;105:370–2.

[29] Klyscz T, Junger M, Jung F, Zeintl H. Cap image—a new kind of computer-assistedvideo image analysis system for dynamic capillary microscopy. Biomed Tech(Berl) 1997;42:168–75.

[30] Kraemer-Aguiar LG, Laflor CM, Bouskela E. Skin microcirculatory dysfunction is al-ready present in normoglycemic subjects with metabolic syndrome. Metabolism2008;57:1740–6.

[31] Fagrell B, Intaglietta M. The dynamics of skin microcirculation as a tool for thestudy of systemic diseases. Bibl Anat 1977:231–4.

[32] de Jongh RT, Ijzerman RG, Serne EH, Voordouw JJ, Yudkin JS, de Waal HA, et al.Visceral and truncal subcutaneous adipose tissue are associated with impairedcapillary recruitment in healthy individuals. J Clin Endocrinol Metab 2006;91:5100–6.

[33] Wipke-Tevis DD, Williams DA. Effect of oral hydration on skin microcirculation inhealthy young and midlife and older adults. Wound Repair Regen 2007;15:174–85.

[34] Stefanovska A, Bracic M, Kvernmo HD. Wavelet analysis of oscillations in the pe-ripheral blood circulation measured by laser Doppler technique. IEEE TransBiomed Eng 1999;46:1230–9.

[35] Rossi M, Maurizio S, Carpi A. Skin blood flowmotion response to insulin iontopho-resis in normal subjects. Microvasc Res 2005;70:17–22.

[36] Kraemer-Aguiar LG, Bouskela E. Techniques to measure microcirculatory parame-ters in insulin resistance states in humans. Microcirculation and insulin resis-tance. Bethan Books; 2010. p. 14–30.

[37] Gradin KA, Li JY, Zhu H, Simonsen U. Blunted pancreatic polypeptide-induced va-sodilatation in mesenteric resistance vessels from spontaneously hypertensiverats. Eur J Pharmacol 2008;601:118–23.

[38] Edwards AV, Garrett JR. Nitric oxide-related vasodilator responses to parasympa-thetic stimulation of the submandibular gland in the cat. J Physiol 1993;464:379–92.

[39] Sjostrom L, Garellick G, Krotkiewski M, Luyckx A. Peripheral insulin in response tothe sight and smell of food. Metabolism 1980;29:901–9.

[40] Teff KL. Cephalic phase pancreatic polypeptide responses to liquid and solid stim-uli in humans. Physiol Behav 2010;99:317–23.

[41] Renaudin C, Michoud E, Rapin JR, Lagarde M, Wiernsperger N. Hyperglycaemiamodifies the reaction of microvessels to insulin in rat skeletal muscle. Diabetolo-gia 1998;41:26–33.

[42] Baik SK, Jeong PH, Ji SW, Yoo BS, Kim HS, Lee DK, et al. Acute hemodynamic effectsof octreotide and terlipressin in patients with cirrhosis: a randomized compari-son. Am J Gastroenterol 2005;100:631–5.

[43] Woods SC. The eating paradox: how we tolerate food. Psychol Rev 1991;98:488–505.