THE ANATOMICAL RECORD 224~66-78(1989) Regional Distribution of Fiber Types in Developing Baboon Diaphragm Muscles LEO C. MAXWELL, THOMAS J . KUEHL, ROGER J.M. McCARTER, AND JAMES L. ROBOTHAM Departments of Physiology (L.C.M., R.J.M.M.), Pediatrics, and Anesthesiology (J.L.R.), The University of Texas Health Science Center at San Antonio, and Southwest Foundation for Biomedical Research (T.J.K.), San Antonio, Texas 78284 ABSTRACT Fiber type distribution and mean fiber area were determined for seven sites in diaphragm muscles of premature (140 days gestation), full-term (180 days gestation), and adult baboons. Within a group, data did not differ significantly amongst the seven sites. The diaphragm of premature animals had a large proportion [56(f2)%] of type IIc fibers, smaller proportions of type I, 110,and IIh fibers [16(*2), 21(+1), and 7(f2)%, respectively], and no type IIg fibers. Full-term animals had fewer type IIc [2(*1)%] fibers, greater proportions of type I [46(+2)%], IIh [23(f1)%],and IIg [11(+ 1)%]fibers, and a similar proportion of type 110 fibers [17(+1)%]. Diaphragm from adult baboons had similar proportions of type I, 110, IIh, IIg, and IIc fibers in females [39(&4),20(*2), 1(*1), 41(*5), and 1(*1)%] and males [48(*2), 16(*1), O(*O), 36(+2), and 3(*2)%1. Fiber area for premature [143(f9), 210(+15), 231(*15), and 156(f16) pm2 for type I, 110,IIh, and IIc fibers], newborn [317(*32), 374(+36), 468(*42), 498(&43),and 322(*37) pm2 for type I, 110,IIh, IIg, and IIc fibers], and for type I, 110,IIg, and IIc fibers from adult female [1,759(*130), 2,365(*284), 5,026(+742), and 1,843(*111) pm2] and adult male [2,513(+221),3,987(*267), 6,102(*376), and 2,833(*151) pm21baboons indicated growth which correlated with body weight. Our results also show that metabolic and contractile enzymes develop normally, but growth of respiratory muscle fibers is arrested, during 10 days following premature birth. No age group has as high an incidence of respiratory failure as premature human infants. In view of this, considerable attention has been given t o the biochemical, morphological, and mechanical development of the lungs and chest wall (Hallman and Gluck, 1977; Hodson, 1977). In neonates and infants, the highly compliant rib cage and the pathologies associated with hyaline membrane disease, aspiration syndromes, and bronchopulmonary dysplasia all place great loads on the muscles of respiration (Bryan et al., 1977). The role which fatigue of the respiratory muscles can play in the clinical expression of pulmonary disease has gained increasing recognition (Derenne et al., 1978ax). Nevertheless, there remain few studies on the contribution which prenatal respiratory muscle development makes to the respiratory capabilities of premature and term human (Farkus-Bargeton et al., 1977; Keens et al., 1978; Keens and Ianuzzo, 1979; Maxwell et al., 1984) or animal (Maxwell et al., 1983, 1984) infants. We have studied the respiratory muscles of a nonhuman primate (Papio cynocephalus) from late gestation through adulthood. Premature baboon infants develop a disease which is indistinguishable from human respiratory distress syndrome (Coalson et al., 1982; Escobedo et al., 1982). Using this model, we have previously reported histochemical data on selected sites of costal diaphragm and intercostal muscles, and fatigability studies of diaphragm muscle bundles (Maxwell et al., 1983). In agreement with the studies of human in0 1989 ALAN R. LISS, INC. fants (Keens et al., 1978; Keens and Ianuzzo, 19791, we observed a progressive increase in the proportion of type I fibers beginning during the third trimester of gestation and continuing into postnatal life. This increase was a t the expense of immature type IIc fibers and suggested transformation of contractile characteristics during gestation. We found no fibers in premature baboons which exhibited the low oxidative capacity of adult type IIg (type 11, glycolytic) fibers. Furthermore, the muscles of premature baboons were more resistant in vitro to tests of isometric fatigue than the muscles of adults. We therefore concluded that the muscles of premature baboons were not predisposed to fatigue by their intrinsic capabilities. This was in marked contrast t o the conclusions of others on developing human muscles, which were interpreted as being highly fatigable due to low oxidative capacity (Keens et al., 1978; Keens and Ianuzzo, 1979). Possible reasons for this difference include 1) species difference; 2) the use of delayed autopsy samples; 3) variable time of respiratory function; and 4) regional differences in diaphragm characteristics amongst the sample sites chosen for study. We have presented evidence (Maxwell et al., 1984) that muscle fibers of human infant dia- Received June 8, 1988; accepted October 17, 1988. FIBER TYPES IN BABOON DIAPHRAGM phragms have a high oxidative capacity and an appearance very similar to the baboon muscles if obtained shortly after death. However, delay of more than a few hours between death and sampling adversely affected the histochemical demonstration of oxidative enzymes. Thus, the first possibility is eliminated and the second possibility does contribute to the difference in results. The purpose of the current study was to address the third and fourth possibilities. In our previous studies, premature baboons were delivered by hysterotomy and sacrificed immediately (Maxwell et al., 1983, 1984). The current study differs since we studied diaphragm muscles of prematurely delivered baboons maintained by intensive neonatal care for 1-11 days after birth. Although some respiratory movements by fetal baboons can be detected in utero (unpublished observation), the pattern of respiration and the level of respiratory effort change drastically at birth. Therefore, it was necessary to study the development of respiratory muscles of prematurely delivered baboons during the first 2 weeks of extrauterine life in order to detect differences in development of respiratory muscle fibers subjected to the additional work load of breathing. In the current report, we compare and contrast the extrauterine development of diaphragm muscle fibers to the development of diaphragm muscle fibers in utero a t the same postconceptional age. We examined samples from seven standardized sites of the diaphragm muscles of premature, term, and adult baboons to detect any regional differences in the fiber-type proportions during pre- and postnatal development of males and females. MATERIALS AND METHODS Premature baboons were delivered by hysterotomy a t 140'3 days of gestation and maintained by neonatal intensive care procedures. Beginning at birth, all animals received intravenous (IV) fluids containing 5% glucose. After 48 hours, the IV fluid was changed to one containing vitamins, trace elements, amino acids, and an increasing concentration of glucose. For 24-48 hours after delivery, all animals were paralyzed with pancuronium bromide (initial dose 0.2 mglkg IV; supplemental intramuscular [IM] doses as required every 8-12 hours) and received either high-frequency oscillatory ventilation (HFOV) or positive pressure ventilation (PPV). Beginning on the third day, pancuronium administration was discontinued and ventilation was provided as required to maintain P,Oz a t 65k15 torr and P,C02 at 40&10 tom. Most animals breathed spontaneously during some time intervals, and some were able to be extubated for most of their subsequent extrauterine life. The details of ventilation are given in Table 1. Body weight was determined daily, and data are in Table 2. At 1, 4, or 11 days after delivery, the animals were sacrificed by exsanguination following intravenous administration of ketamine (10 mg/kg) and Valium (0.5 mg/kg). Samples were removed from seven standardized sites of diaphragm muscles (Fig. 1). For the study of regional distribution of fiber types during development, diaphragm muscles were obtained from six premature (delivered by hysterotomy at 14023 days gestation and sacrificed at 4-6 days after delivery), ten term (vaginal delivery at 18227 days gestation, sacrificed a t 1-9 days after birth), and 67 TABLE 1. Ventilatory status of prematurely delivered baboons' Ventilatory status 0 day group (N = 5) X4577 Not ventilated; sacrificed at delivery X4578 Not ventilated; sacrificed at delivery R 84 Not ventilated; sacrificed a t delivery AJ 83 Ventilated three hours while paralyzed UB 83 Ventilated three hours while paralyzed 1 day group (N = 3), all paralyzed 0 83 PPV while paralyzed P 83 PPV while paralyzed P 84 PPV while paralyzed 4 to 6 day group (N = 61, all paralyzed days 1 and 2 E 83 HFOV; spontaneous breathing beginning on day 3 F 83 PPV; spontaneous breathing beginning on day 3 H 83 PPV; spontaneous breathing beginning on day 3 I83 HFOV; spontaneous breathing beginning on day 3 J 83 HFOV; spontaneous breathing beginning on day 3 L 83 HFOV; spontaneous breathing beginning on day 3 11 day group (N= 51, all paralyzed days 1 and 2 PPV; occasionally extubated after day 3; Q 83 spontaneous breathing S 83 HFOV; mostly extubated after day 3; spontaneous breathing PPV; mostly extubated after day 4; V 83 spontaneous breathing B 84 HFOV; not extubated; spontaneous breathing D 84 HFOV; not extubated; spontaneous breathing 'PPV = positive pressure ventilation; HFOV = high-frequency oscillatory Ventilation. 13 adult baboons. Of these, the animals were approximately equally distributed between sexes. Seven standardized sample sites were selected. Sampling sites 1 and 2 were unavailable in most premature baboons due to other experimental procedures. In some baboons, additional diaphragm muscle samples were excised, blotted to remove excess surface water, and weighed. The samples were placed in a vacuum oven overnight a t a temperature of 90°C and were dried to constant weight. The ratio of dry to wet muscle weight was calculated. Muscles sampled for histochemical analysis were attached to wooden splints and frozen immediately in isopentane cooled to its freezing point in liquid nitrogen. The frozen samples were packed in dry ice and transported to the histochemistry laboratory. Frozen samples were brought t o a temperature of -20°C and 10 pm cross sections were cut in a cryostat. Sections were dried at room temperature for 30-60 min and then incubated in a battery of histochemical media. Reactions were performed to demonstrate the activities of reduced nicotinamide adenine dinucleotide tetrazolium reductase (NADH-TR)and myofibrillar ATPase activities. The NADH-TR assay was performed according to the method of Dubowitz and Brooke (1973). Myofibrillar ATPase assays were performed as described 68 L.C. MAXWELL ET AL. TABLE 2. Body weight (grams) of prematurely delivered baboons’ Day ID No. 0 1 AJ 83 UB 83 X 4571 4578 R 84 0 83 P 83 P 84 E 83 F 83 H 83 I83 J 83 L 83 Q 83 S 83 Y 83 B 84 D 84 585 536 420 474 584 513 635 549 527 517 646 611 515 587 481 530 611 497 545 510 675 635 625 525 680 644 530 567 495 534 644 469 567 = 2 3 4 5 6 7 8 9 10 11 650 523 705 621 540 630 528 623 489 670 555 539 563 528 473 606 540 565 518 445 565 510 424 541 434 492 530 550 511 460 453 541 47 1 476 500 448 540 458 494 465 463 435 470 483 441 435 545 480 483 455 458 445 462 524 460 483 437 449 500 420 474 682 525 613 478 5 14 no data available. D I A P H R A G M S A M P L I N G SITES (Pleural S u r f a c e ) Sternum Fig. 1. The diaphragm muscles were divided into seven regions from which samples were routinely taken. The sites are numbered as viewed from the pleural surface. Even-numbered samples were from the right diaphragm; odd-numbered samples were from the left diaphragm or substernal region. by Chayen et al. (1972). Serial sections were preincubated at an alkaline pH (10.3) or a t an acid pH (4.3) (Brooke and Kaiser, 1970; Dubowitz and Brooke, 1973) prior to the myofibrillar ATPase assay. Sections were observed with a Zeiss Universal I1 microscope equipped with a Panasonic video camera and a drawing tube which allowed an image drawn on a tracing tablet to be superimposed on the microscopic image. The outlines of 100-200 muscle fibers from the midportion of each sample were drawn from sections incubated for myofibrillar ATPase after alkaline preincubation. Fibers were classified as type I or I1 based on a low or high intensity, respectively, of the myofibrillar ATPase reaction after alkaline preincubation. Then serial sections were viewed and the image of the sampled fibers was superimposed on the sample drawing. We have adopted a subclassification of those type I1 fibers which show acid reversal on the basis of NADHTR activity, while the variable proportion of type I1 fibers which did not demonstrate acid reversal were classified type IIc. Type I1 fibers that did demonstrate “acid reversal” following preincubation at pH 4.3, and that had high NADH-TR activity, were called type 110 69 FIBER TYPES IN BABOON DIAPHRAGM (type I1 oxidative) while those with low NADH-TR activity were designated type IIg (type I1 glycolytic). A similar approach has been used previously for masticatory muscle fibers of rhesus monkeys (Maxwell et al., 19811, regenerating fibers in muscle autografts in cats (Maxwell, 19841, and developing fibers in baboon diaphragm (Maxwell et al., 1983, 1984). In our previous papers, type I1 fibers high in NADH-TR activity were called type IIa and type I1 fibers low in NADH-TR activity were called type IIb. The problem is that some type IIa and IIb fibers classified by using NADH-TR activity might not have been classified as type IIa and IIb fibers by the myosin-based scheme of Brooke and Kaiser (1970). The type IIa and IIb fibers of our previous studies are directly comparable to the 110 and IIg fibers in our current study. Our classification scheme is actually a combination of the metabolic profile classification according to Peter et al. (1972) and the myosin-based scheme of Brooke and Kaiser (1970). The classification according to Peter et al. (1972) is based on the demonstration of myofibrillar ATPase after alkaline preincubation and the demonstration of oxidative enzyme activity. Type I, 110, and IIg fibers are essentially comparable to the slow-twitch oxidative (SO), fast-twitch oxidative-glycolytic (FOG), and fast-twitch glycolytic (FG) fibers identified by Peter et al. (1972) or to the fibers of slow-twitch (S),fasttwitch fatigue-resistant (FR), or fast-twitch fatigable (FF) motor units classified by Burke et al. (19711, but neither of these previous classifications identified type IIc fibers. The classification scheme of Brooke and Kaiser, based on the characteristics of myofibrillar ATPase after preincubation in differing pH solutions, does identify type IIc fibers but makes no assessment of oxidative enzyme activity. Although either the Brooke and Kaiser or the Peter et al. scheme can yield adequate data on adult muscles, we have not been satisfied with strict adherence to either scheme for developing or regenerating muscles. The current nomenclature is a compromise which avoids ambiguity while combining the desirable features of several previous classification schemes. We have taken care to ensure that tissue preparation, section thickness, and incubation procedures were the same for each muscle studied. Whenever possible, samples of fetal and adult muscles were incubated simultaneously, and we also incubated a sample of a limb skeletal muscle. These procedures enable us to control the quality of incubations and make comparisons of reaction intensity in different fibers within a muscle or among muscles. Our classification criteria for oxidative enzyme activity include overall reaction intensity, the size of diformazan granules, and the completeness of the interfibrillar network of activity. Type 110 fibers rank higher than type IIg fibers in each of these criteria. All type I fibers demonstrated acid reversal, and all type IIc fibers had high NADH-TR activity. We have also done a systematic study of a kind of fiber we observed in premature and term muscles, but which was essentially absent from adult muscles. The fiber is characterized by a central region containing a relatively high NADH-TR activity, surrounded by a peripheral zone of very low NADH-TR activity. This gives the fibers an appearance which has led others to call them “halo fibers” (Farkus-Bargeton et al., 1977). We have classified such fibers type IIh in this study. At least three microscopic fields were randomly selected from the midportion of each cross section of diaphragm. Data from these three fields were pooled for analysis. We selected the midportion of cross sections in order to avoid possible differences in the proportions of fiber types on pleural and abdominal surfaces, as has been reported for cat diaphragm (Sieck et al., 1983) and for some regions of rat diaphragm (Metzger et al., 1985). The cross sectional area of each fiber was determined by digital image analysis by using a Bioquant (R&MBiometrics). Mean fiber area and the numbers of fibers of each type in the sample were determined. From these data, the proportion of sampled fibers and the proportion of sample area represented by each type of fiber were calculated. A detailed description of these techniques has been published elsewhere (Maxwell et al., 1983, 1984; Maxwell, 1984). Our new data on diaphragm muscles of premature baboons and of full-term baboons after normal birth and our previously published data (Maxwell et al., 1983) were used for comparison of extrauterine development of diaphragm muscles following premature delivery to the age-matched intrauterine development of diaphragm muscle. For this analysis, we used data only from costal sample sites, since that was the basis for our previous published data, and we combined type IIh fibers with the type IIa fibers of which they are a subset. Mean, standard deviation, and standard error of the mean were calculated on pooled data from each sample site, for each age group, and for each sex. Unless otherwise stated, data are presented as mean ? standard error of the mean. Data were also subjected to an analysis of variance (ANOVA), and when significance was indicated, the groups responsible for the significance were identified by using a Student Neumann Kuels (SNK) test. Significance was accepted a t P<.05. All statistical calculations were performed as described by Zar (1974). RESULTS Data for the sample sites of diaphragm muscles from baboons of premature, term, or adult age are presented in Tables 3 and 4. In diaphragm muscles of premature or term animals, no significant sex-related differences in the proportion of fiber types or in the mean fiber area were found by ANOVA. For adult baboons, however, ANOVA showed a significant difference between males and females in the cross-sectional area of fibers, with no difference in the proportion of fiber types. Therefore, premature, term, adult female, and adult male groups were studied in order t o assess differences amongst sample sites within the diaphragm. Within each of these groups, the proportion of fiber types amongst different sample sites of diaphragm muscles did not differ significantly by ANOVA (Table 3). Nor were there significant differences (ANOVA) within each group, amongst sample sites of diaphragm muscles, in the mean cross-sectional area of the different types of fibers (Table 4) or in the muscle mean fiber area (Fig. 2). Therefore, t o simplify further analysis, the data obtained for the sites within each diaphragm muscle were pooled, and the percentage of fibers and a 70 L.C. MAXWELL ET AL. TABLE 3. Proportion of fiber t es for Sam le sites of premature, term, and adult diaphragm muscles (data are mean f SE!%; three to fPve samples for each site are represented except for sites 1 and 2 of premature diaphragm) % I Site I10 Premature (140 days gestation; 4-6 days extrauterine life) 1 2 13.95 f 3.96 3 7.61 k 2.54 22.14 f 1.86 4 5.23 k 2.49 5 5.94 k 2.99 23.98 k 3.40 30.79 f 3.39 6 7.25 4 1.89 24.59 f 0.57 7 4.85 4 2.54 Term (180 days gestation; 1-9 days postnatal life) 1 48.44 4 2.53 17.47 k 1.75 2 50.88 4 3.67 16.23 f 3.05 15.68 k 1.16 3 50.45 4 2.39 18.20 4 0.87 4 42.72 f 2.32 17.62 2 1.36 5 41.44 f 2.20 16.78 f 1.65 6 44.71 3.54 7 43.82 f 4.10 19.33 k 5.73 Adult female 1 46.70 3.80 16.57 2.17 21.19 2 2.32 2 37.90 f 4.91 14.00 1.23 45.64 4 3.26 3 20.58 4 4.74 4 39.70 4 1.87 20.33 f 2.44 41.46 2.53 5 22.15 k 3.73 38.87 4.01 6 20.98 4 2.61 36.34 f 2.51 7 Adult male 18.87 k 1.67 43.62 4 2.42 1 15.83 f 2.31 2 50.59 4 2.29 11.67 1.23 56.98 3.62 3 20.65 2.55 4 47.07 2 2.68 42.97 f 2.19 19.10 4 1.61 5 15.06 f 2.66 44.47 k 4.22 6 7 44.69 f 3.95 15.48 2 1.79 * * * * * * * * weighted mean fiber area were calculated for each fiber type. These pooled data were taken as representative of the histochemical profile and cross-sectional area of the fibers in that muscle. The proportions of fiber types and mean fiber area for diaphragm muscles of premature, term, adult female, and adult male baboons are plotted in Figures 3 and 4. The diaphragm muscles of premature animals were characterized by a large proportion [56(+2)%] of type IIc fibers, smaller proportions of type I, 110, and IIh fibers [16(&2),21(*1), and 7(*2)%, respectively], and no type IIg fibers. Mean fiber area was small, having mean values of 143(*9), 210(%15), 231(*15), and 156(*16) pm2 for type I, 110,IIh, and IIc fibers. Compared to premature baboons, the diaphragm muscles of term animals delivered after 180 days gestation had few [2( 1)%1 type IIc fibers, greater proportions of type I [46(-+2)%1,IIh [23(*1)%1, and IIg [11(21)%] fibers, and about the same proportion of type 110 fibers [17(21)%1. The fibers had increased in fiber area t o mean values of 317(232), 374(+36), 468(+-42), 498(*43), and 322(*37) pm2 for type I, 110,IIh, IIg, and IIc fibers. Adult baboon diaphragm muscles had similar proportions of fiber types regardless of sex, having 39( +4), 20(*2), 1(*1),41(+5), and 1(*1)%of type I, 110, IIh, * IIh Ik IIC 4.63 4 1.25 4.49 4 0.74 8.93 4 2.45 5.41 f 0.75 8.07 t 1.26 0.00 k 0.00 0.00 f 0.00 0.00 k 0.00 0.00 k 0.00 0.00 k 0.00 63.81 t 3.44 58.14 4 2.45 51.13 2 3.09 46.52 5.32 52.48 f 3.46 23.19 f 1.86 21.24 t 2.76 19.29 f 1.62 23.69 4 2.12 25.62 f 1.90 23.55 f 2.02 24.14 f 2.08 7.25 4 0.95 9.78 2 1.24 11.85 k 1.23 13.19 4 1.32 12.67 f 1.98 12.87 4 1.58 11.66 4 1.28 3.57 1.12 1.87 f 0.66 2.71 4 0.64 2.15 f 0.93 2.64 4 0.70 2.08 f 0.74 1.04 k 0.48 0.00 f 0.00 0.00 f 0.00 0.00 f 0.00 0.00 k 0.00 0.00 f 0.00 0.00 4 0.00 0.00 f 0.00 36.04 f 4.90 40.74 f 5.96 39.82 3.90 39.56 4.01 37.52 k 2.92 38.74 4 3.44 42.46 4 2.72 * 0.67 2 0.45 0.00 4 0.00 0.45 t 0.45 0.17 0.17 0.35 2 0.35 0.21 f 0.21 0.20 t 0.20 0.00 f 0.00 0.00 f 0.00 0.00 4 0.00 0.00 f 0.00 0.00 2 0.00 0.00 f 0.00 36.76 f 1.11 33.23 f 1.84 31.35 4 4.10 32.07 f 2.94 37.70 k 1.80 40.46 f 3.03 39.63 f 2.54 0.59 f 0.39 0.33 f 0.33 0.00 0.00 0.19 f 0.19 0.21 f 0.21 0.00 t 0.00 0.17 t 0.17 0.00 f 0.00 * * * * IIg, and IIc fibers in females and 48(+2), 16(*1), O(+-O), 36(+2), and 3(+2)%of type I, 110,IIh, IIg, and IIc fibers in males. Muscle fiber area in diaphragm muscles from adult females was smaller than for the muscles from adult males. Fiber area in muscles of females was 1,759(* 1301, 2,365(*284), 5,026(*742), and 1,843 (*111) pm2 and in muscles of males was 2,513(+221), 3,987(2267), 6,102(*376), and 2,833(+151) pm2 for type I, 110,IIg, and IIc fibers, respectively. The difference in mean fiber area amongst premature, term, adult female, and adult male baboons correlated very well with body weight (Fig. 5), which averaged 5 2 , .92, 15.66, and 23.84 kg, respectively, in these groups. Figure 6 shows the changes in fiber-type proportions during the third trimester and the first 9 days (solid symbols) of postnatal life after term birth. The data for 120, 138, and 163 days of gestation and for 1,5, and 8 days after term birth are pooled from current results and from our previous report (Maxwell et al., 1983) and plotted as a function of postconceptional age. Since we did not classify the transitional type IIh (“halo”)fibers in our earlier paper on intrauterine development (Farkus-Bargeton et al., 19771, we combined the IIh fibers into the type 110fiber pool, of which the IIh fibers are a subset, for the current analysis. Our earlier data for diaphragm muscle were obtained from a standard- FIBER TYPES IN BABOON DIAPHRAGM 71 TABLE 4. Mean fiber area (MFA) for each type of fiber of sample sites in premature, term, and adult diaphragm muscles (data are mean -C SEM) MFA Ik Site I I10 IIh Premature (140 days gestation; 4-6 days extrauterine life) 1 2 3 4 5 6 7 165 t 2 1 130 t 14 147 f 7 128 f 4 131 f 10 233 t 19 207 f 13 207 f 29 177 f 17 201 f 9 238 215 235 217 219 t 22 t 14 t 31 f 22 f7 IIC MFA 165 f 14 137 f 11 160 I 2 8 121 f 9 130 f 9 177 f 14 156 2 11 174 2 25 144 t 11 154 f 8 247 f 21 313 t 50 348 f 60 311 f 55 282 f 36 407 f 122 274 f 58 347 t 33 366 f 50 395 f 48 363 t 31 360 t 22 431 f 69 389 t 4 Term (180 days gestation; 1-9 days postnatal life) 1 2 3 4 5 6 7 278 299 328 316 303 366 324 Adult female 1 1,898 2 3 4 5 6 7 f 30 f 39 f 41 f 27 f 24 f 66 f 37 445 470 486 419 428 506 474 f 39 t 56 f 62 f 36 t 20 t 82 t 42 f 212 2 113 f 119 3,020 2,520 1,840 2,176 2,096 2,017 1,994 2,329 f 289 2,731 f 382 2,208 f 208 2,113 f 172 2,411 t 122 2,583 t 265 2.442 f 117 4,464 4,566 4,015 3,370 3,794 3,608 3.723 t 715 t 773 t 260 1 2 1,817 1,332 1,583 1,535 1,533 1,579 f 132 366 383 412 348 338 414 355 t 161 t 121 f 124 f 53 t 44 482 507 516 453 474 557 513 f 68 f 55 f 39 f 23 f 76 t 58 t 432 f 143 f 92 t 163 2 175 f 215 f 50 f 83 f 57 f 39 f 26 t 70 f 59 7,069 f 367 5,070 f 808 3,577 4 257 4,206 f 328 3,931 2 217 3,828 2 208 3,900 f 385 2,170 f 50 6,865 f 851 7,295 t 1,168 6,861 f 470 5,184 f 669 6,081 2 710 5,580 f 370 5.955 2 714 2,736 f 508 2,128 1,560 1,840 1,489 t 186 1,678 1,645 3,923 t 345 3,222 t 485 2,256 f 102 2,733 f 205 2,526 f 93 2,542 f 280 2,658 f 280 Adult male 1 2 3 4 5 6 7 209 f 269 f 192 f 375 2 3 4 5 6 1,944 3,982 3.500 4,418 2 580 4,546 2 696 3,792 202 3,344 f 189 4,040 t 307 3,936 f 263 3.946 f 196 * 7 Site Number Fig. 2. The mean fiber area for the sample sites of diaphragm muscles is shown for each group. ized location comparable to site 4 in the current study. Significant increases in the proportions of type I and type 110fibers and a decrease in the proportion of type IIc fibers occurred during the final trimester of gestation. There were no type IIg fibers observed through 163 days of gestation. Following term birth, the proportion of type I fibers stabilized near adult values, and the proportion of type 110 fibers declined as type IIg fibers began to appear. The composition of the diaphragm muscle also L.C. MAXWELL ET AL. 72 Prom Torm Adult Fomak Adult Male Fig. 3. The percentage of each fiber type is shown for the diaphragm muscles of premature (140 days gestation; 4-6 days extrauterine life), term (180 days gestation; 1-9 days postnatal life), adult female, and adult male baboons. a rri a 4 a W m LL ?I a Prom Trm Adult Fomak Adult Malo Fig. 4. The mean fiber area from data pooled for the seven sample sites is shown for each group. Groups are as described for Figure 3. 130212pm2, 283'33 to 209*17pm2, and 188a24 to 141k14pm2 for types I, 110,and IIc, respectively (Fig. 7). This represents a reduction of fiber area for prematurely delivered baboons to 51,63, and 63% for types I, 110,and IIc fibers, respectively, of the values for diaphragm muscle fibers of baboons after intrauterine development alone. Mean fiber cross-sectional area (MFCSA) (Fig. 8) also declined during 11 days following premature birth from 217226 pm to 165215 pm2, reaching 65% of the value for diaphragm muscles at 140 days of gestation. The dry:wet weight ratios of muscle samples obtained from premature (140 days gestation) and fullterm baboons were 0.2120.01 (n=5) and 0.2520.01 (n = 7), respectively. At 4 and 11 days of extrauterine life following premature delivery at 140 days of gestation, ratios were 0.21*0.01 (n=9) and 0.24k0.01 (n = 4), respectively. At 6-9 days following full-term delivery, dry:wet ratio was 0.22+0.00 (n = 8). DISCUSSION Following premature delivery, all animals were parchanged during 11 days following premature delivery (Fig. 6, open symbols). The proportions of type I and alyzed for 2 days, but they performed some spontanetype 110 fibers increased from 6.5'1.31 to 13.8+2.1% ous breathing episodes thereafter. Most breathed sponand from 27.521.8 to 36.8+1.7%, respectively, while taneously for a substantial portion of time after the the proportion of type IIc fibers decreased from third postnatal day, and some long-term animals were 66.222.6 to 49.0*2.9%, and there were no type IIg fi- extubated during the last week of the study. Infants bers. These proportions of fiber types during extrauter- gained weight during the paralysis but then lost ine development after premature birth are in agree- weight as spontaneous respirations were initiated and ment with values predicted by extrapolation from the pulmonary status improved. Body weights were maintained after day 3. Fluid gain during paralysis may data for intrauterine development. The mean fiber area increased throughout the final result since muscle movements were only passive, trimester of gestation for type 110 and type IIc fibers, eliminating muscle activity as a means of eliminating but only during the last month for type I fibers (Fig. 7). extravascular water from muscular tissue at a time The growth of each of these fiber types and of the ap- when fluid support is intensive. The early fluid gain pearing type IIg fibers continued for several days after may be related to the effects of ventilation on the carfull-term birth. However, the growth appears not to be diopulmonary function of a premature animal with sustained, as decreases in mean fiber area were seen patent ductus arteriosus. At the end of paralysis, and for all fiber types during the second week of postnatal coincident with improving pulmonary status, this fluid-saving influence is reduced, and the baboon inlife. During 11days following premature delivery a t 140 fants enter a period when a diuresis of 10-15% of body days of gestational age, the mean fiber area of dia- weight normally occurs in prematurely delivered huto man infants (unpublished observations). For the 11phragm muscle fibers decreased from 244'27 73 FIBER TYPES IN BABOON DIAPHRAGM 10000 z - ri u) E (P g 1000 5if Prem Term Adult Female Adult Male E Q : 100 .1 10 1 100 Body Weight (Kg) Fig. 5. Mean diaphragm muscle fiber area is plotted against body weight. The axes are logarithmic due to the magnitude of the range of the data. day group, weight stabilized a t somewhat lower than initial birth weight during the latter days of support. No significant differences in the proportion of fiber types or mean fiber area were found amongst the seven sampling sites within diaphragm muscles of any of the major groups (Tables 3 and 4). This finding is consistent with data for rat diaphragm muscles (Metzger et al., 1985) in which only one site had proportions of fiber types different from the rest of the muscle, and with data for cat diaphragm muscles (Sieck et al., 1983) which had no differences amongst various regions. Thus, in spite of any differences in recruitment pattern or physiological function of crural and costal portions of the diaphragm (Decramer et al., 19841, the muscle fibers of these portions develop similarly. There are several implications of this finding: different rates of development of specific regions of diaphragm muscles relative to each other cannot account for impaired respiratory function of the diaphragm as a whole; easily accessible sites for diaphragm sampling contain fibers of representative physiological properties; contractile properties and fatigability of bundles from the sample site used in our previous (Maxwell et al., 1983) report (equivalent to the current site 4) are representative of diaphragm function; and, differences in sampling site amongst earlier studies of primate diaphragm muscles (Keens et al., 1978; Keens and Ianuzzo, 1979; Maxwell et al., 1983,1984) do not likely contribute to differences in results. Although we did not systematically study the abdominal and pleural surfaces of the diaphragm muscles, we saw no evidence that major differences in composition occur in baboon diaphragm between these surfaces. This is not surprising for diaphragm muscles of premature and term baboons, which have not completely differentiated to adult fiber-type distributions. Due to the predominance of type IIc fibers in diaphragm muscles of premature animals, a difference in the percentage of type I fibers would be difficult to detect. Sieck et al. (1983) found a higher percentage of type I fibers on the abdominal surface of adult cat diaphragm muscles, whereas Metzger et al. (1985) reported a higher proportion of type I fibers on the thoracic surface of adult rat diaphragm muscles. Apparently, there are species differences in the distribution of muscle fiber types across the diaphragm cross section. The alteration of the proportion of fiber types during the third trimester of gestation continues into postnatal life. The diaphragm muscle of premature baboons contains a high proportion of type IIc fibers. These fibers are highly oxidative and express high myofibrillar ATPase activity after alkaline preincubation as well as after acid preincubation. This lack of “acid reversal” is characteristic of immature, developing muscle fibers which contain fetal isozymes of myosin (Whalen et al., 1981) and of transitional fibers (Kugelberg, 1976) undergoing transformation from type I to type I1 (or vice versa). The low percentage of type I fibers in premature baboon diaphragm is consistent with the results of our earlier study and with the results of Keens et al. (1978) and of Keens and Ianuzzo (1979) on premature human diaphragm. Muscle fibers throughout the costal and crural diaphragm of premature baboons are highly oxidative and are not intrinsically susceptible to fatigue due t o low oxidative enzymatic capacity. The diaphragm of term compared to premature baboons has a greater proportion of type I fibers and type IIh fibers, but a reduced proportion of type IIc fibers. These results are consistent with an ongoing differentiation and transformation of fiber types during gestation. Type IIc fibers likely differentiate into adult type I and 110 fibers as gestation approaches full term. At term, a few fibers demonstrate the histochemical characteristics of the less-oxidative type IIg fibers. This suggests that the appearance of significant proportions 74 L.C. MAXWELL ET AL. Type Ilo 80- 60- wm 40- 200 l i . . 110 wil ldo 160 1 0‘0 Days Post-Conception Fig. 6. The percentages of type I, 110, IIg, and IIc fibers are plotted as a function of postconceptional age. Data obtained after intrauterine development are shown by filled symbols; those following extrauterine life are shown by open symbols. The data for 4 days of extrauterine life are the same as the data used for Figure 3. of type IIh (“halo”)fibers represents transition between type 110 and type IIg characteristics. The halo appearance of fibers destined to become type IIg fibers may result from growth of low oxidative muscle a t the periphery of muscle fibers, from the progressive loss of oxidative capacity toward the central parts of the fibers, or both (Fig. 9). Evidence in support of this interpretation is that type IIh fibers are larger than type 110 fibers in both premature and term baboons. The timing of the appearance of “halo fibers” in diaphragm muscles of human infants (Farkus-Bargeton, 1977) is also consistent with this interpretation. Transformation of type IIa to type IIb fibers has also been reported for diaphragm muscle of young, growing rats (Tamaki, 1985). After full term, the proportion of type I fibers stabilizes a t adult values, but the proportions of type I1 fibers continue to change. Diaphragm muscles of adult baboons have very few type IIc fibers, a reduced proportion of oxidative (types 110 and IIh) fibers, and an increased proportion of type IIg fibers relative to the muscles of premature or term animals. The small proportion of type IIc fibers observed in adult muscles may represent continuing transformation of fibers between type I and I1 myosin isozymes. As respiratory function approaches adult patterns, a large population of type IIg and a smaller population of type 110 fibers characterize adult diaphragm muscles. The presence of a large population of fatigable type IIg fibers in a muscle as vital as the diaphragm is probably a consequence of the large respiratory reserve volumes in adults. The respiratory muscles are capable of generating pressure gradients to rapidly move very large volumes of air if all of their muscle fibers participate in the contraction. However, respiratory muscles are not normally recruited to work at this maximal level, and normal respiration requires the contraction of only a fraction of respiratory muscle fibers a t any instant in time. Less frequently recruited fibers lose the capacity for high levels of oxidative metabolism but remain available for use during periods of increased respiratory drive. This transition from type 110 to type IIg fibers during postnatal life represents an adaptation to lower recruitment levels, as occurs during detraining of diaphragm and other skeletal muscles (Faulkner et al., 1972). There are essentially no type IIh fibers in adult mus- 75 FIBER TYPES IN BABOON DIAPHRAGM Type I Type Ilo r 100 o d . . 120 I . . . . .160 I . . ! . .200, 180 140 Days Post-Conception Fig. 7. The mean fiber areas of type I, 110, IIg, and IIc fibers are plotted as a function of postconceptional age. Data obtained after intrauterine development are shown by filled symbols; those following extrauterine life are shown by open symbols. The data for 4 days of extrauterine life are the same as the data used for Figure 4. cles, further supporting our interpretation that these exist during development when the peripheral growth of the fibers contains relatively few mitochondria. Muscle fibers of all types approximately double in cross-sectional area during the last trimester of gestation. Much of this growth could be a normal consequence of maturation of muscle fibers. Whether spontaneous respiratory movements, which can be detected during gestation in animals and humans (Harding, 19481, influence the growth of muscle fibers is not known. There appears to be no effect of sex on muscle fiber growth during prenatal development. Certainly, the major growth of diaphragm muscle fibers occurs during postnatal life when spontaneous breathing increases the frequency and force for which the diaphragm is recruited. During postnatal life, the diaphragm muscle fibers of male baboons grow larger than the fibers in muscles of female baboons. This sexrelated difference in diaphragm muscle fiber growth correlates with differences in body weight between males and females. This difference could represent greater respiratory load on the diaphragm muscle fi- 400 - 300 - 200 - 100 100 . . I 120 140 . . , . . 160 I, 180 - ,. . . I 200 Days Post-Conception Fig. 8. The overall muscle fiber cross-sectional area (MFCSA) of diaphragm muscle fibers is plotted as a function of postconceptional age. Data obtained after intrauterine development are shown by filled symbols; those following extrauterine life are shown by open symbols. 76 L.C. MAXWELL ET AL. TYPE I I FIBER TRANSFORMATIONS Ilc I lo Ilh -Growth o f l e s s o x i d a t i v e periphery -Loss o f c e n t r a l m i t o c h o n d r i a Fig. 9. A schematic description of the transformation of type I1 fibers during prenatal development is shown. bers of larger animals, different hormonal background for the growth of muscle, or both. The possibility that the activity of breathing may influence the growth and development of respiratory muscles is of concern for researchers investigating the development of respiratory function and for clinicians evaluating the respiratory capabilities of premature infants. Striated muscle responds with a variety of adaptive alterations to meet the demands placed upon it (Baldwin et al., 1972; Faulkner et al., 1972; Maxwell et al., 1973). Following premature birth, the work of breathing represents a large increase in the work load on respiratory muscles, and it is reasonable to question whether adaptive changes in response to that increased work demand alter the normal development of respiratory muscles. The results of the current study address that question. The proportions of fiber types in primate diaphragm muscles change considerably during the third trimester of gestation (Keens et al., 1978; Keens and Ianuzzo, 1979; Maxwell et al., 1983). The proportions of fiber types in baboon diaphragm also change during the first 11days following premature birth, but the proportions of fiber types during extrauterine development after premature delivery do not differ from those during intrauterine development alone a t a comparable postconceptional age (Fig. 5 ) . Possible explanations for this could include the following: 1)the work of extrauterine breathing does not represent an adequate stimulus; 2) there is not time within 11 days for the adaptive response to be demonstrated; or 3) the developmental alterations in the muscles exert control over muscle fiber characteristics which is not interrupted by changes in functional demand. In view of the respiratory distress and respiratory muscle fatigue common amongst premature infants (Derenne et al., 1978a-c), the possibility that premature birth does not represent an increased demand on the respiratory muscles seems unlikely. Of course, the demand of breathing is lessened to the extent that mechanical respiratory assistance is required. For example, all premature baboons delivered after 140 days of gestation received 2 days of ventilatory assistance, and some animals required intermittent ventilatory assistance through the first 4 days of life. During such assistance, there may not be an adequate stimulus for adaptation of respiratory muscle fibers. The rate of change of enzymatic activities during adaptive responses depends upon the intensity of the stimulus and the ability of the muscle fibers to alter protein metabolism. Diaphragm muscles in human respiratory patients requiring ventilation can demonstrate a sufficient response to inspiratory resistive training to permit weaning from the ventilator within as little as 10 days (Aldrich and Karpel, 1985). Furthermore, the respiratory muscles of human and baboon infants undergo rapid changes during the last weeks of gestation (Keens et al., 1978; Keens and Ianuzzo, 1979; Maxwell et al., 1983). Therefore, it seems reasonable that 11 days is an adequate time to permit demonstration of adaptive alterations, but the effective stimulus was only delivered for 7-9 days. There remains the third possibility, that the enzymatic properties of respiratory muscle fibers are unable to demonstrate adaptive responses after birth during the third trimester of gestation. The muscle fibers during this time are undergoing rapid alterations of fiber type. The genetic control of the development of the muscle fibers may be so strong that it cannot be interrupted by altered demand. Whether adaptations of other characteristics occurred which were not measured, or whether the extrauterine development might influence the final muscle composition during adult life, cannot be determined from the current data. However, the proportions of fiber types appear not to be altered during extrauterine compared to intrauterine development. An expected response to an increase in the load against which muscle fibers must work is hypertrophy of muscle fibers (James, 1979). However, muscle fibers in the diaphragms of prematurely delivered baboons were considerably smaller than muscle fibers following only intrauterine development (Fig. 2). Even the muscle fibers of baboons sacrificed during the first few hours after delivery were smaller than muscle fibers at a comparable postconceptional age from our previous study (Maxwell et al., 1983), in which animals were terminated within the first few minutes after delivery. This indicates that the decline in muscle fiber crosssectional area commenced almost immediately at delivery. Much of this decline in fiber area occurred during the period of paralysis, when the diaphragm was not recruited. Thus, for the first 48 hours, an argument could be supported that the muscle fibers undergo a disuse atrophy. There was no evidence of recovery of fiber area in any fiber type during the study period. The initial decline of fiber area during the first 4 days of extrauterine life was not the result of water loss, since the hydration of the muscles at 4 days after premature delivery was unaltered from that immediately after premature delivery. By 11days following premature delivery, the dry:wet ratio increased, indicating a relative loss of 3%of muscle weight due to dehydration. Dehydration cannot account for the decline of fiber area during extrauterine life to 51-63% of the fiber area for age-matched muscles during intrauterine development. Nor can dehydration account for the decline of fiber area a t 6-9 days after full-term delivery, since dry:wet ratio decreased during this time, indicating relatively greater hydration. There is also the possibility that the diaphragm muscle fibers were nutritionally deprived during the intensive neonatal care for 11days after premature delivery. 77 FIBER TYPES I N BABOON DIAPHRAGM In adult hamsters underfed by 33%, body weight declined 25% and diaphragm weight by 29% within 4 weeks (Kelsen et al., 1985). Thus, relative undernutrition could contribute to the failure to maintain diaphragm muscle mean fiber area. This problem may be more severe for the immature muscle fibers of premature baboons when the metabolic requirements for growing, active muscle are increased. Perhaps the rigors of extrauterine life impose demands on muscle fibers after birth which are not adequately met by the nutritional supply. Mean fiber cross-sectional area (MFCSA) of diaphragm muscles from premature baboons only declined to 65% of the fiber area for age-matched muscles after intrauterine development, less than the relative decline for any fiber type. The greater loss of mean fiber area (MFA) for type I fibers than of MFCSA can be accounted for by considering the combined effects of fiber atrophy and fiber-type transformation. As type IIc fibers mature, they grow in cross-sectional area, and they differentiate to other fiber types. Since type IIc fibers are the smallest fibers, transformation of them to type I or type 110fibers would have the effect of reducing the MFA of those types of fibers. This effect can also be seen for type I fibers during intrauterine development when type I MFA remains constant (Fig. 7) during the third trimester, while MFCSA increases (Fig. 8).Furthermore, if it is assumed that the largest of the type IIc fibers are the most likely to transform, then the MFA of the type IIc fibers would not accurately reflect muscle fiber growth. Thus, the reduction of MFA likely results in part from fiber atrophy and in part from transformation of small fibers to the enzymatic characteristics of larger ones. Differentiation could occur either as the IIc fibers grow to a set cross-sectional area or as the IIc fibers reach a set postconceptional age (Fig. 10). The former does not seem likely in view of our current data, since no reduction in type IIc fiber area would be expected in extrauterine compared to intrauterine development if type IIc fibers had to achieve a predetermined fiber area before differentiation could occur. The latter possibility is consistent with our data since the proportions of fiber types appear unaltered by extrauterine development, but the fiber area is reduced. Thus, although fibers atrophy and the growth of fibers may be retarded from normal during extrauterine development, the fiber transformations occur on schedule. Although premature delivery does not appear to interfere with the normal development of fiber types, the failure to maintain fiber growth leaves little doubt that diaphragm muscle of premature baboons becomes less able to meet contractile requirements during the period shortly after birth when the demands are increasing. The diaphragm function may be declining when it is highly needed. This may also occur to some extent even after normal term delivery. How long the relative loss of muscle fiber area continues before recovery and growth of fibers resumes cannot be determined without longer-term studies. ACKNOWLEDGMENTS This work was supported by NIH grants HL 29977 and HL 38427. GESTATIONAL AGE 0 0 .fO .................. ........... ............... ................ > ...... :.:.:.:.:.:.:.: ................. ................. ............. .... t Differentiation occurs at time "t" regardless of size GESTATIONAL AGE Differentiation occurs at size "s" regardless of time Fig. 10. Two hypothetical schemes by which muscle fiber differentiation might be controlled during development are shown. LITERATURE CITED Aldrich, T.K., and J.P. Karpell985 Inspiratory muscle resistive training in respiratory failure. Am. Rev. Respir. Dis., 131:461-462. Baldwin, K.M., G.H. Klinkerfus, R.L. Terjung, P.A. Mole, and J.O. Holloszy 1972 Respiratory capacity of white, red and intermediate muscle: Adaptation to exercise. Am. J . Physiol., 222:373-378. Brooke, M.H., and K.K. Kaiser 1970 Muscle fiber types: How many and what kind? Arch. Neurol., 23:369-379. Bryan, A.C., A.L. Mansell, and H. Levison 1977 Development of the mechanical properties of the respiratory system. In: Development ofthe Lung. W.A. Hodson, ed. Marcel Dekker, Inc., New York, pp. 445-468. Burke, R.E., D.N. Levine, F.E. Zajac 111, P. Tsairis, and W.K. Engle 1971 Mammalian motor units: Physiological-histochemical correlates of three types in cat gastrocnemius. Science, 174:709-712. Chayen, J., L. Bitensky, R.G. Butcher, and L.W. Poulter 1972 A Guide to Practical Histochemistry. Philadelphia: Lippincott, pp. 112-115. Coalson, J.J., T.J. Kuehl, M.B. Escobedo, J.L. Hilliard, F. Smith, K. Meredith, W. Walsh, D. Johnson, and J.L. Robotham 1982 A baboon model of hyaline membrane disease and bronchopulmonary dysplasia: 11. Pathologic features. Exp. Mol. Pathol., 37:335-350. Decramer, M., A. De Troyer, S. Kelly, and P.T. Macklem 1984 Mechanical arrangement of costal and crural diaphragm is dogs. J . Appl. Physiol., 56r1484-1490. Derenne, F.P.H., P.T. Macklem, and C.H. Roussos 1978a The respiratory muscles: Mechanics, control and pathophysiology. Part I. Am. Rev. Respir. Dis., 118:119-133. Derenne, F.P.H., P.T. Macklem, and C.H. Roussos 1978b The respiratory muscles: Mechanics, control and pathophysiology. Part 11. Am. Rev. Respir. Dis., 118:373-390. Derenne, F.P.H., P.T. Macklem, and C.H. Roussos 1978c The respira- 78 L.C. MAXWELL ET AL. tory muscles: Mechanics, control and pathophysiology. Part 111. Am. Rev. Respir. Dis., 118t581-601. Dubowitz, V., and M.H. Brooke 1973 Muscle Biopsy: A Modern Approach. Saunders, Philadelphia. Escobedo, M.B., J.L. Hilliard, F. Smith, K. Meredith, W. Walsh, D. Johnson, J.J. Coalson, T.J. Kuehl, and J.L. Robotham 1982 A baboon model of hyaline membrane disease and bronchopulmonary dysplasia: I. Clinical features. Exp. Mol. Pathol., 37t323-334. Farkus-Bargeton, E., M.F. Diebler, M.L. Asenio-Nunes, M.L. Wehrle, and B. Rosenberg 1977 Etude de la maturation histochemique quantitative et ultrastructural du muscle foetal humain. J. Neurol. Sci., 31~245-259. Faulkner, J.A., L.C. Maxwell, and D.A. Lieberman 1972 Histochemical characteristics of muscle fibers from trained and detrained guinea pigs. Am. J . Physiol., 222t836-840. Hallman, M., and L. Gluck 1977 Development of the fetal lung. J . Perinat. Med., 5:l-31. Harding, R. 1984 Fetal breathing. In: Fetal Physiology and Medicine: The Basis of Perinatology. R.W. Beard and P.W. Nathanielsz, eds. Marcel Dekker, Inc., New York, pp. 255-286. Hodson, W.A. 1977 Development of the Lung. Marcel Dekker, Inc., New York. James, N.T. 1979 Studies on the responses of different types of muscle fibre during surgically induced compensatory hypertrophy. J . Anat., 129:769-776. Keens, T.C., A.C. Bryan, H. Levison, and C.D. Ianuzzo 1978 Developmental patterns of muscle fiber types in human ventilatory muscles. J . Appl. Physiol., 44:909-913. Keens, T.C., and C.D. Ianuzzo 1979 Development of fatigue-resistant muscle fibers in human ventilatory muscles. Am. Rev. Respir. Dis., 119:139-141. Kelsen, S.G., M. Ference, and S. Kapoor 1985 Effects of prolonged undernutrition on structure and function of the diaphragm. J . Appl. Physiol., 58~1354-1359. Kugelberg, E. 1976 Adaptive transformation of rat soleus motor units during growth. J. Neurol. Sci., 27r269-273. Maxwell, L.C., J.A. Faulkner, and D.A. Lieberman 1973 Histochemical manifestations of age and endurance training. Am. J . Physiol., 224:356-361. Maxwell. L.C.. D.S. Carlson. J.A. McNamara. Jr., and J.A. Faulkner 1981 Adaptation of the masseter and temporalis muscles following alteration in length, with or without surgical detachment. Anat. Rec., 2OOt127-137. Maxwell, L.C., R.J.M. McCarter, T.J. Kuehl, and J.L. Robotham 1983 Development of histochemical and functional properties of baboon respiratory muscles. J . Appl. Physiol., 54:551-561. Maxwell, L.C., T.J. Kuehl, J.L. Robotham, andR.J.M. McCarter 1984 Temporal changes after death in primate muscle oxidative enzyme activity. Am. Rev. Respir. Dis., 13Ot1147-1151. Maxwell, L.C. 1984 Muscle fiber regeneration in nerve-intact and free skeletal muscle autografts in cats. Am. J. Physiol., 246(Cell Physid.15):C96-C105. Metzger, J.M., K.B. Scheidt, and R.H. Fitts 1985 Histochemical and physiological characteristics of the rat diaphragm. J . Appl. Physiol., 58t1085-1091. Peter, J.B., V.R. Edgerton, C.A. Gillespie, and K.A. Stemple 1972 Metabolic profiles of three fiber types of skeletal muscles of guinea pigs and rabbits. Biochemistry, 11:2627-2633. Sieck, G.C., R.R. Roy, P. Powell, C. Blanco, V.R. Edgerton, and R.M. Harper 1983 Muscle fiber type distribution and architecture of the cat diaphragm. J . Appl. Physiol., 55t1386-1392. Tamaki, N. 1983 Effect of growth on muscle capillarity and fiber type composition in rat diaphragm. Eur. J. Appl. Physiol., 54t24-29. Whalen, R.G., S.M. Sell, G.S. Butler-Brown, K. Schwarz, P. Bouveret, and I. Pinset-Harstrom 1981 Three myosin heavy chain isozymes appear sequentially in rat muscle development. Nature (Lond.), 2923305-809. Zar, J . 1974 Biostatistical Analysis. Prentice-Hall, Englewood Cliffs, NJ.