Inclusion Criteria:

1. 18-65 yr old HIV-infected men (n=40) and women (n=40).
2. Source documentation of HIV status.
3. Stable on highly active antiretroviral therapy (HAART) that may or may not include HIV-protease inhibitors for at least 3 months prior to enrollment. As of Jan 2005, HIV-infected long-term non-progressors will be included, even though they are not receiving HAART because it is likely that their disease status will not advance during the study period.
4. Impaired glucose tolerance (IGT) or type 2 diabetes and fat redistribution. IGT is defined as: fasting (8hr) plasma glucose 100-126mg/dL or plasma glucose >140 mg/dL 2-hours after a 75g-oral glucose load. This definition (proposed/revised May 2006) includes volunteers with adult onset type 2 diabetes (non-insulin dependent diabetes), because we believe that these volunteers may benefit from the potential glucose-lowering actions of pioglitazone with or without exercise training. Fat redistribution is defined as any 2 of the following: waist-to-hip ratio >0.85 women or >0.95 men, or trunk/appendicular adipose ratio using whole-body DEXA >0.85 women or >1.3 men, or <15% leg fat (DEXA; (leg fat/total body fat) x100), or visceral adiposity VAT >120 cm2 women or >140 cm2 men, or a VAT/TAT ratio >0.30 women or >0.40 men. Overall, the eligibility criteria are designed to include subjects with metabolic syndromes that increase CVD risk, and the potential to gain the greatest benefit from pioglitazone and exercise training. These inclusion criteria will help select/maintain homogeneous groups of study participants.
5. Plasma HIV RNA (Roche Amplicor® assay) <5000 copies/ml OR a CD4 T-cell count >100 cells/µL and stable for previous 3 months.
6. BMI 20-40kg/m2.
7. "Normal" blood chemistries for at least 1 month prior to enrollment; platelet count >30,000/mm3, absolute neutrophil count <750/mm3, transaminases <2.5x the upper limit of normal (ULN), creatinine <3x ULN, fasting triglycerides <500 mg/dL.

Exclusion Criteria:

1. Medications or agents that might alter/impair glucose metabolism (insulin, glucocorticoids, corticosteroids, megace) during the 3 months prior to enrollment or at any time during enrollment. Volunteers who developed type 2 diabetes after HIV-infection or after starting anti-HIV medications, who are receiving insulin sensitizers (metformin, meglitinides, alpha-glucosidase inhibitors) or insulin secretagogues (sulfonylureas), but still do not have their blood sugars in control (defined as IGT above) are eligible.
2. Abnormal or unstable (for 3 months prior to enrollment) endocrine blood chemistries that might otherwise explain insulin resistance. TSH <0.2 or >12µIU/mL, morning cortisol >22µg/dL, IGF-1 <115ng/mL, total testosterone <200ng/dL (men) <15ng/dL (women). Use of testosterone, ACTH, thyroid hormone, or rhGH replacement to normalize low levels is permitted, but must be stable on hormone replacement for 3 months prior to enrollment and will not discontinue this replacement during enrollment. Hormone replacement cannot be started during treatment.
3. Allergy or hypersensitivity to thiazolidinediones. Currently taking a thiazolidinedione.
4. Anti-obesity or anorectic medications during the 3 months prior to enrollment or at any time during enrollment. Lipid-lowering medications are permitted (fibrate or statin), but the subject must be stable on that agent for at least 3 months prior to enrollment. Lipid-lowering agents cannot be started during the treatment period.
5. Chronic hepatitis B infection (HB surface antigen positive). Active hepatitis C infection (detectable Hep C RNA). Those who have cleared hepatitis B or C infection are eligible. This was revised in May 2006 to better clarify some uncertainties about hepatitis and eligibility.
6. History of serious cardiovascular conditions or NY Heart Association (NYHA) cardiac status Class III or IV, (e.g., recent MI, unstable angina, edema, CHF, CAD, CABG, valve disease (murmur), stroke, uncontrolled high blood pressure (resting >140/90 mmHg on 3 occasions), irregular heart rhythm, bundle-branch block, aortic stenosis, resting ST-segment depression >1mm) that would preclude exercise testing/training, or substantially increase risk of a CV-event during exercise, or would limit the subjects ability to participate in exercise training. Treatment with medications for a cardiovascular condition (cardiac glycosides, alpha- or beta-blockers). Some antihypertensive medications (Ca++-channel blocker, diuretic, or ACE inhibitor) will be permitted.
7. Insulin-dependent diabetes mellitus (IDDM) or a history of ketoacidosis, symptomatic diabetic neuropathy or retinopathy, or renal disease (creatinine >3x ULN).
8. Hematocrit <34% in men or <25% in women with symptoms (fatigue, "tired-legs", shortness of breath). Hemoglobin <10 gm/100ml with symptoms.
9. History of eating disorder, significant GI-disease
10. Nausea, vomiting, diarrhea (>4 loose stools/day) that are unresponsive to treatment during 2wks prior to enrollment or that persists for >2wks during enrollment.
11. Active secondary infection. Any significant change in chronic suppressive therapy for an opportunistic infection during the 1-month prior to enrollment.
12. During the 3 months prior to enrollment, regular aerobic or weight lifting exercise training that exceeds the minimum (45min/d, 3d/wk, >75% exercise capacity) required for the metabolic, biochemical, and anthropomorphic benefits of exercise to be attained as described in the American College of Sports Medicine Guidelines. In Apr 2006, this exclusion criterion was modified in order to facilitate enrollment of volunteers who are moderately-highly active during the 3months prior to screening for this study. We will randomize volunteers who exercise regularly into the two treatment groups. This will reduce the potential confounding effects of regular exercise on the metabolic, biochemical, and anthropomorphic benefits of exercise training that is prescribed during the treatment phase. We have excluded several volunteers who are regular exercisers, and we believe we may be unnecessarily excluding them. They can be included, enrollment will be facilitated, and the study design will not be adversely affected, as long as we randomize them into pioglitazone or exercise+pioglitazone (1:1).
13. Unwilling or unable to do supervised exercise 3 sessions/wk at the Medical School exercise facility. Any condition that might be contraindicated for exercise training (disabling joint, cartilage or muscle injury/disorder that prohibits or severely limits participation in regular exercise, disabling arthritis, severe physical disabilities, claudication, pulmonary disease, sinus tachycardia, arrhythmias, premature atrial or ventricular contractions).
14. Pregnancy or nursing mothers. Women cannot be pregnant and must agree to use an acceptable form of birth control during the treatment period. If birth control pills are used, the woman must be stable on these medications for at least 6 months prior to enrollment. All metabolic testing will be done in the early follicular phase. Urine pregnancy tests (hCG) will be done at baseline and every month the woman is enrolled in the study.
15. Pancreatitis within prior 1 year. Serum triglycerides >500mg/dL esp. with history of pancreatitis, or otherwise at high risk for pancreatitis.
16. Medications that inhibit or slow blood coagulation (ie., blood thinners). Prothrombin (clotting) time exceeds 2 sec > control (only in subjects who agree to muscle/fat biopsies).
17. Active malignancy or treatment with chemotherapeutic agents or radiation therapy during the 3 months prior to enrollment.
18. Excessive weight loss (>10% body weight) during the 3 months prior to enrollment. History of hyperlactatemia or lactic acidosis, esp. with rapid weight loss.
19. "Blinded" investigational drugs/medications during the 3 months prior to enrollment that will not be unblinded before enrollment. Open-label investigational drugs are permitted. Must be stable on these for 3 months prior to enrollment, must remain stable on them during the treatment period, and they must be known not to affect glucose, lipid, adipose tissue or liver metabolism.
20. Non-prescription over the counter drugs/supplements that might alter glucose, lipid, or adipose tissue metabolism (eg., creatine monohydrate, chromium picolinate, amino acid/protein supplements, beta-hydroxy-beta-methyl-butyrate, anabolic-androgenic steroids, medium- or long-chain fatty acids, branched chain amino acid supplements, amino acids that potentially increase insulin or GH secretion (eg., arginine)) within 1 month of enrollment. These supplements will not be permitted during the treatment period.
21. Dementia or reduced cognitive function or unable to provide voluntary informed consent. Prisoners will not be enrolled.
22. Active substance abuse (e.g., alcoholism, cocaine, heroin, crack, methamphetamine, phencyclidine). Upon evaluation, if the physician-investigator considers that this substance abuse does not put the volunteer at risk for an adverse (ie. cardiovascular) event, or if the physician-investigator feels that the volunteer will be reliable and compliant with the treatment regimens, and that this substance abuse will not interfere with the study medications, exercise regimens, or data interpretation, then at the physician-investigators discretion the volunteer may be eligible.
23. Any cytokine or anti-cytokine therapy during the 3 months prior to enrollment.
24. Patients felt by the Physician Investigator to have uncontrolled hypertension or other diseases (i.e. vasculitis, pulmonary HTN) that otherwise may be a cause of significant underlying or variable endothelial dysfunction
25. Receiving vasoactive substances (i.e. nitrates, viagra) other than antihypertensives that cannot be discontinued at least 12 hours before endothelial function testing (BART).